NL2037929B1

NL2037929B1 - Brain Organoid

Info

Publication number: NL2037929B1
Application number: NL2037929A
Authority: NL
Inventors: Bessler Nils; Caroline Rios Anne
Original assignee: Prinses Maxima Centrum Voor Kinderoncologie B V
Priority date: 2024-06-11
Filing date: 2024-06-11
Publication date: 2026-01-12
Also published as: WO2025259108A1

Abstract

The invention provides hindbrain organoids and methods of producing such organoids. Said hindbrain organoids may include healthy or healthy and diseased tissue. Also provided are methods of testing therapeutic agents using said hindbrain organoids. The invention also provides culture mediums used for producing (patterning) said hindbrain organoids and kits including the culture medias.

Description

Brain Organoid

The present invention provides methods of producing organoids that are mimetic of the hindbrain, in particular, the pons. The methods and organoids produced thereby may include healthy and diseased tissue. For example, the organoids may include cancerous tissue, such as diffuse midline glioma tissue. Therefore, also provided are organoids produced by said methods. Also provided are methods of testing therapeutic agents using the organoids, uses of the organoids and culture mediums and kits thereof for use in said methods.

Background

Human brain organoids have become valuable in vitro tools for investigating brain development and understanding the onset, progression, and potential therapeutical targeting of nervous system disorders, including cancer?%25. There is however a lack of organoids which accurately recapitulate the hindbrain or regions thereof, such as the pons.

Diffuse midline gliomas (DMGs) are rare and aggressive paediatric brain tumours often caused by somatic mutations in histone 3 (H3) genes, commonly a lysine27-to-methionine (K27M) substitution’ and occurring at a high prevalence in the pons, as compared to thalamic or cortical locations?. Primarily, affecting children under 10 years old, they present the highest mortality rate of any cancer, with a median overall survival of only 9-15 months*®. This detrimental prognosis underscores a critical need to gain more insight into the unique biology of the disease to develop more effective treatments.

Efforts to unravel the cellular composition of H3K27M-altered DMG through single-cell analysis have revealed intratumoral heterogeneity, with a spectrum of tumour cell profiles ranging from a stalled stem-like oligodendrocyte progenitor cell (OPC-like) to astrocyte (AC-like) and oligodendrocyte (OC-like) phenotypes, along with a recently identified mesenchymal (MES-like) state’™®. Moreover, location-dependent profiles of these cell types have recently come to light, introducing another layer of complexity’-'°. Most notably, OPC-like cancer cells, considered pivotal in K27M-mediated tumorigenesis”®'!, exhibit varying levels of maturation dependent on the anatomical location of the tumours®, underscoring a spatial specificity of DMG gliomagenesis. In addition, insights from both animal studies*2*5 and human pluripotent stem cell-derived research’ suggest an early developmental window of tumour initiation. Thus, dysregulated mechanisms during hindbrain development13131° particularly within the region responsible for the formation of the pons™, likely play a central role in driving H3K27M-altered gliomagenesis and capturing this region-specific embryonic patterning appears essential for accurate pontine DMG modelling.

Given the rarity and inoperable nature of DMG, which limits the availability of patient material? organoids could offer scalable models for in vitro experimental testing of emerging therapies against DMG. This includes the latest advances in immunotherapy for DMG; GD2 Chimeric

Antigen Receptor (CAR) T cells82827 that in a recent first in-patient clinical trial showed promising initial relief of neurological symptoms yet offered only limited survival benefits.

There is a need for improved systems for modelling the hindbrain.

There is a need for improved systems for modelling the pons.

There is a need for improved systems for modelling diseases of the hindbrain.

There is a need for improved systems for modelling diseases of the pons.

There is a need for improved systems for testing therapeutics in diseases and conditions associated with the hindbrain and/or pons.

Brief summary of the disclosure

The invention is based on a novel human cerebral guided organoid model with pontine identity.

By genetically modelling H3.3K27M-altered DMG in these organoids, the inventors are able to reflect the regional and developmental features of diseases such as DMG, recapitulating the tumour transcriptomic landscape found in patient tumours and uncovering unique cancer cell subpopulations with pontine DMG specificity. The inventors also demonstrate the utility and scalability of this new accessible human DMG model for modelling therapeutic treatments such as CAR T cell functional heterogeneity during prolonged treatment (up to 1 month).

In a first aspect of the invention there is provided a method for producing a hindbrain organoid, the method comprising: a. culturing pluripotent stem cells in an initial culture medium for an initial time period to produce an embryoid body comprising the pluripotent stem cells; b. after the initial time period, culturing the embryoid body under conditions for proliferation and at least partial differentiation of the pluripotent stem cells in a first culture medium for a first time period to produce an organoid comprising neuroectodermal cells, wherein the first culture medium comprises:

FGF2; a bone morphogenetic protein (BMP) pathway inhibitor; a TGF-6 inhibitor; and a WNT activator; c. after the first time period, culturing the neuroectodermal cells in a second culture medium for a second time period to produce committed neuroectodermal cells, wherein the second culture medium comprises:

FGF4 at a concentration of about 10 ng/pl;

Retinoic acid at a concentration of about 10 uM;

Purmorphamine at a concentration of about 1 WM;

Dorsomorphin at a concentration of about 1 WM;

SB431542 at a concentration of about 10 uM; and

CHIR99021 at a concentration of about 3 uM;

d. after the second time period, culturing the committed neuroectodermal cells in a third culture medium for a third time period, wherein the third culture medium comprises:

FGF4;

Retinoic acid; and a sonic hedgehog activator; e. after the third time period, maturing the organoid in a fourth culture medium for providing a hindbrain organoid, wherein the fourth culture medium comprises a maturation medium.

In a first aspect of the invention there is provided method for producing a hindbrain organoid comprising healthy and diseased tissue for modelling brain tissue, the method comprising a. culturing pluripotent stem cells in an initial culture medium (DAY O medium) for an initial time period to produce an embryoid body comprising the pluripotent stem cells; b. after the initial time period, culturing the embryoid body under conditions for proliferation and at least partial differentiation of the pluripotent stem cells in a first culture medium for a first time period to produce neuroectodermal cells to form an organoid comprising neuroectodermal cells, wherein the first culture medium (WEEK 1 medium) comprises:

FGF2; a one bone morphogenetic protein (BMP) pathway inhibitor; a TGF-B inhibitor; and a WNT activator; c. after the first time period, culturing the neuroectodermal cells in a second culture medium for a second time period produce committed neuroectodermal cells, wherein the second culture medium (WEEK 2 medium) comprises:

FGF4 at a concentration of about 10 ng/pl;

Retinoic acid at a concentration of about 10 uM;

Purmorphamine at a concentration of about 1 WM;

Dorsomorphin at a concentration of about 1 HM;

SB431542 at a concentration of about 10 HM; and

CHIR99021 at a concentration of about 3 HM; d. during the second time period, inducing a disease state in one or more of the committed neuroectodermal cells to produce an organoid comprising healthy and diseased tissue; e. culturing the committed neuroectodermal cells in a third culture medium for a third time period, wherein the third culture medium (WEEK 3 medium) comprises:

FGF4;

Retinoic acid; and a sonic hedgehog activator; f. after the third time period, maturing the organoid in a fourth culture medium for providing a hindbrain organoid comprising healthy and diseased tissue, wherein the fourth culture medium comprises a maturation medium.

In certain embodiments, step (e) of the first aspect or step (f) the second aspect comprises maturing the hindbrain organoid in the maturation medium from day 21 from step (a) of claim 1 or 2 onwards.

In certain embodiments of the first or second aspect, the method further comprises maintaining the matured organoid for at least 30 days in the maturation medium. In certain embodiments of the first or second aspect, the method further comprises maintaining for about 1 year or more.

In certain embodiments of the second aspect, the diseased tissue comprises cancer tissue, neurodegenerative tissue and/or malformed tissue; optionally wherein the cancer tissue is diffuse midline glioma (DMG) tissue.

In certain embodiments of the first or second aspect, the hindbrain organoid comprises a pontine organoid.

In certain embodiments of the first or second aspect, the initial culture medium comprises:

FGF2 at a concentration of about 4 ng/ul; and Y-27632 at a concentration of about 10 HM.

In certain embodiments of the first or second aspect, the first culture medium comprises:

FGF2 at a concentration of at most about 50 ng/ul;

Dorsomorphin at a concentration of about 1 WM;

SB431542 at a concentration of about 10 uM; and

CHIR99021 at a concentration of about 3 HM.

In certain embodiments of the first or second aspect, the third culture medium comprises:

FGF4 at a concentration of about 10 ng/pl; retinoic acid at a concentration of about 10 uM; and purmorphamine at a concentration of about 1 UM.

In certain embodiments of the first or second aspect, the first, second and third media further comprise: neurobasal medium;

Advanced DMEM/F-12 medium; ; 1xN2 supplement; and

Heparin solution at a concentration of at least 2 ug/ml.

In certain embodiments of the first or second aspect, the initial medium further comprise: neurobasal medium, advanced DMEM/F-12 medium and an L-glutamine supplement.

In certain embodiments of the first or second aspect, the initial time period comprises 2 days or 48 hours.

In certain embodiments of the first or second aspect, the first time period starts on day 2 from step (a) and is up to day 7 from step (a) of the first or second aspect. 5 In certain embodiments of the first or second aspect, the second time period starts on day 7 from step (a) and is up to day 14 from step (a) of the first or second aspect.

In certain embodiments of the first or second aspect, the third time period starts on day 14 from step (a) and is up to day 21 from step (a) of the first or second aspect.

In certain embodiments of the first or second aspect, culturing in step (d) of the first aspect or step (e) of the second aspect comprises culturing the organoid with agitation at about 16 days from step (a).

In certain embodiments of the second aspect, inducing a diseased state comprises mutating one or more disease associated genes of the committed neuroectodermal cells.

In certain embodiments of the second aspect, inducing a diseased state comprises providing the committed neuroectodermal cells with one or more disease associated genes.

In certain embodiments of the second aspect, inducing a diseased state comprises providing the committed neuroectodermal cells with one or more disease associated proteins.

In certain embodiments of the second aspect, inducing a diseased state comprises providing the committed neuroectodermal cells with one or gene editing systems for mutating one or more disease associated genes.

In certain embodiments of the second aspect, inducing a diseased state comprises providing the committed neuroectodermal cells with one or more interfering nucleic acid molecules.

In certain embodiments of the second aspect, the one or more genes are selected from p53,

PDGFRA, and histone H3.

In certain embodiments of the second aspect, the mutations comprise:

PDGFRA-D842V and H3K27M

In certain embodiments of the second aspect, the disease associated proteins comprise:

DNp53, PDGFRA-D842V and H3K27M.

In certain embodiments of the second aspect, inducing a diseased state comprises: a. introducing one or more nucleic acid vectors encoding the disease associated genes, disease associated proteins, gene editing systems for mutating one or more disease associated genes, and/or interfering nucleic acid molecules into one or more of the committed neuroectodermal cells at a specified time point.

In certain embodiments of the second aspect, the specified time points is on day 11 from step (a).

In certain embodiments of the first or second aspect, the neuroectodermal cells and/or committed neuroectodermal cells comprises cells that express one or more of HOXB1, GBX2,

MEIS1, MEIS2 and/or MEIS3 from about 7 days from step (a) of the first or second aspect.

In certain embodiments of the first or second aspect, the neuroectodermal cells and/or committed neuroectodermal comprises cells that do not express one or more of OTX2 and/or spinal cord-specific CDX genes about 7 days from step (a) of the first or second aspect.

In certain embodiments of the first or second aspect, the hindbrain organoid comprises cells that express one or more of TPH2, GFAP, AQP1, AQP4, OLIG1, PDGFRA, OLIG2, NRG3,

NRXN1, GRIA2, RBFOX1, MAP2, ERB4, PLCG2, VIM, SOX2, SOX10, SOX9, SOX4, CLU,

BCAN, NCKAP5, PPP2R2B, NTN1, RMST, STMN2 and/or SLIT2; optionally from about at least 7 to about at least 30 days from step (a) of the first or second aspect.

In certain embodiments of the first or second aspect, hindbrain organoid comprises one or more of: astrocytes; oligodendrocytes; glioblasts; radial glial cells; axon-guiding neuroepithelium cells; optionally comprising choroid plexus cells and/or ependymal cells; stromal cells; and/or neurons; optionally comprising one or more of hindbrain-specific serotonergic neurons, excitatory neurons, inhibitory neurons, and/or dopaminergic neurons.

In certain embodiments of the second aspect, the hindbrain organoid comprises one or more cancer cells. In certain embodiments of the first or second aspect, the cancer cells comprise one or more of: astrocyte like-cells (AC-like cells), mesenchymal like-cells (MES-like cells), oligodendrocyte-like cells, neural stem cell-like cell, oligodendrocyte precursor like-cells (OPC-like cells), and/or cycling cells.

In certain embodiments of the second aspect, the OPC-like cells express CRAPB1, OLIG2 and/or OLIG1, the MES-like cells express VIM and/or TIMP1, the cycling cells express TOP2A and/or MKI67, the neural stem cell -like cells express STMN2 and/or AC-like cells express

AQP1 and/or AQP4.

In certain embodiments of the first or second aspect, the method further comprises co-culturing one or more immunological components with the embryoid body, neuroectodermal cells, committed neuroectodermal cells and/or hindbrain organoid for providing a matured organoid comprising an immune microenvironment.

In certain embodiments of the first or second aspect, the one or more immunological components comprise endothelial cells, myeloid progenitor cells and/or microglia cells.

In certain embodiments of the first or second aspect, the one or more immunological components further comprises induced pluripotent stem cell derived T-cells; optionally wherein the induced pluripotent stem cell derived T-cells are derived from a patient.

In a third aspect there is provided an organoid produced by the methods described herein. Thus in one aspect there is provided a healthy hindbrain organoid as described herein. Thus in one aspect there is provided a healthy pontine organoid as described herein. Thus in one aspect there is provided a hindbrain organoid comprising healthy and diseased tissue as described herein. Thus in one aspect there is provided a pontine organoid comprising healthy and diseased tissue as described herein.

In another aspect there is provided a method testing one or more therapeutic agents, the method comprising: a. providing a hindbrain or pontine organoid according to any of the aspects described herein, b. contacting the organoid with at least one therapeutic agent after maturing the hindbrain organoid in the maturation medium for at least about 30 days; c. detecting one or more changes in the organoid; d. determining the effects of the therapeutic agent based on the absence or presences of the one or more changes.

In certain embodiments, contacting the organoid comprises contacting after about 30 days.

In certain embodiments, the therapeutic agent comprises an anti-cancer agent.

In certain embodiments, the anti-cancer agent comprises a T cell therapy; optionally selected from tumour infiltrating lymphocyte (TIL) or a chimeric antigen receptor T-cell (CAR T-cell).

In certain embodiments, the one or more changes comprises: death or survival of cells of the organoid; transcriptional changes; epigenetic changes; protein changes; metabolic changes; genomic changes; post-translational protein changes; and/or phenotypic changes.

In certain embodiments, the organoid comprises cancer tissue and the changes comprise the amount of cancer tissue.

In certain embodiments, the method further comprises analysing the T cells after contacting the organoid.

In certain embodiments, analysing comprises determining: an exhaustion profile; behavioural changes;

transcriptional changes; epigenetic changes; protein changes; metabolic changes; genomic changes; post-translational protein changes; and/or phenotypic changes.

In certain embodiments, the method of testing further comprises determining one or more markers for cytotoxic T cells of the T cells based on the exhaustion profile.

In a further aspect there is provided use of a hindbrain or pontine organoid according to any of the aspects described herein for drug discovery, efficacy and/or toxicity studies.

In a further aspect there is provided a culture medium for patterning a hindbrain organoid comprising: i FGF2 at a concentration of about 50 ng/pl; i. Dorsomorphin at a concentration of about 1 uM; ii. SB431542 at a concentration of about 10 uM; and iv. CHIR99021 at a concentration of about 3 HM.

In a further aspect there is provided a culture medium for patterning a hindbrain organoid comprising: i. FGF4 at a concentration of about 10 ng/pl; ii. Retinoic acid at a concentration of about 10 pM; iii. Purmorphamine at a concentration of about 1 WM; iv. Dorsomorphin at a concentration of about 1 uM; v. SB431542 at a concentration of about 10 uM; and vi. CHIR99021 at a concentration of about 3 HM.

In a further aspect there is provided a culture medium for patterning a hindbrain organoid comprising: vii. FGF4 at a concentration of about 10 ng/pl; viii. Retinoic acid at a concentration of about 10 uM and ix. purmorphamine at a concentration of about 1 WM.

In a further aspect there is provided a kit of parts for producing a brainstem organoid comprising: a first, second and third culture medium as described herein.

In certain embodiments, the kit further comprises one or more pluripotent stem cells.

In certain embodiments, the kit further comprises an initial culture medium comprising:

FGF2 at a concentration of about 4 ng/ul; and

Y-27632 at a concentration of about 10 UM.

In certain embodiments, the kit further comprises one or more nucleic acid vectors encoding one or more disease associated genes, disease associated proteins, one or more interfering nucleic acid molecules and/or gene editing systems for mutating disease associated genes.

In certain embodiments, the kit further comprises a maturation media.

In certain embodiments, the kit further comprises instructions for use of the kit.

In certain embodiments, the first, second and/or third culture medium as described herein further comprise: neurobasal medium and Advanced DMEM/F-12 medium at a ratio of 1:1; an L-glutamine supplement; 1xN2 supplement; and

Heparin solution at a concentration of about 2 ug/ml.

Throughout the description and claims of this specification, the words “comprise” and “contain” and variations of them mean “including but not limited to”, and they are not intended to (and do not) exclude other moieties, additives, components, integers or steps.

Throughout the description and claims of this specification, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise.

Features, integers, characteristics, compounds, chemical moieties or groups described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein unless incompatible therewith.

Various aspects of the invention are described in further detail below.

Brief description of the Fiqures

Embodiments of the invention are further described hereinafter with reference to the accompanying drawings, in which:

Figure 1 shows the concept, timings and co-culture of the claimed invention. From left to right:

Healthy hindbrain/pontine organoids are established using the claimed morphogen cocktail to induce and maintain hindbrain/pontine identity, with varying combinations and concentrations in

WEEK1, 2 and 3. After initial patterning, the healthy organoid mature (MATURATION1, week 4- 20) to develop all relevant neuroectodermal celltypes, e.g. Neurons (serotonergic, excitatory, inhibitory, dopaminergic etc.) and Macroglial (Astrocytes and Oligodendrocytes). After

MATURATION1 phase, the hindbrain organoid is in its mature state (MATURE2, Week20+) and can be kept up to years. Mutations and any other form of alteration can be introduced already during the patterning phase, e.g. mutations for modelling Diffuse Midline Glioma at d11 in

WEEK2. Treatment can be started as early as the beginning of MATURATION (d30+), but preferably is added towards the end of MATURATION1 (d100+), to investigate effects on all relevant cell types.

Figure 2 shows generation and validation of pontine cerebral organoids. a, Schematic representation of timely morphogen stimulated patterning of hESCs/hIPSCs towards pontine organoids and their subsequent application for DMG tumour modelling, genetic lineage tracing and CAR T cell treatment. b, Heatmap showing relative bulk RNA expression of homeobox (HOX) genes at week 02 and 03 depicted as log2 fold change normalized to week 01 (blue-to- red colour gradient). Data from 3 independent batches with 3 pooled organoids per batch. ¢,

Immunofluorescent 3D images of a 200 um thick organoid slice at week 03 labelled for F-ACTIN (white), SOX2 (yellow) and HOXB1 (red). White insert indicates zoom area displayed on the right. Overview image and zoom scale bars 250 um and 25 pm respectively. d, Voxhunt spatial similarity map of week 04 and week 12 pontine organoids with E13.5 and E18.5 mouse brain. e,

Normalized combined gene counts in pontine organoids over time for gene signatures related to human foetal pontine tissue from gestation week (GW) 9 — 25 (top) and GW 21-28 (bottom). *p = 0.0321, **p = 1.07e-03, ***p = 2.23e-04, one-way ANOVA with post-hoc Tukey honest significance distance. Boxplots showing min/max values, including median and lower/upper quartile with shaded area representing 95% confidence interval. b, d, e, n = 3 pooled organoids from each independent batch of total 3 batches per timepoint. f, Immunofluorescent 3D image of a 200 pum thick organoid slice at week 16 labelled for TUBB3 (orange) and TPH2 (blue).

White insert indicates zoom area displayed on the right. Overview image and zoom scale bars 250 um and 25 um respectively.

Figure 3 shows FGF4 patterning comparison and organoid specification and reproducibility. a, Schematic representation of a human foetal brain in gestational week (GW) 05 with indicated morphogens influencing the differentiation of hindbrain rhombomeres (r) and their HOX gene code respectively. b, Heatmap of z-score measuring relative brain region identity based on Voxhunt similarity mapping for various supplemented concentrations of FGF2 or 4. ¢, Bulk RNA seq derived normalized mean gene expression of pontine organoids at week 03 compared to mouse rhombomeres. Normalized mean expression (+S.D.) from three independent experiments. Colour code refers to colour codes used in panel a for the sub- compartment of the hindbrain region. d, Normalized RNA counts of organoids for hindbrain- specific GBX2 and fore-/midbrain-specific O7X2 at week 01, 02 and 03, one-way ANOVA with post-hoc Tukey honest significance distance. e, Normalized RNA counts for the hindbrain specific genes Meis homeobox (ME/S) 1, 2 and 3 and spinal cord specific Caudal Type homeobox (CDX) 1, 2 and 4 at week 01, 02 and 03, one-way ANOVA with post-hoc Tukey honest significance distance f, g, Relative marker gene expression for neural, serotonergic and macroglia progenitor cells representing the neurogenic phase based on Fan et a/.* (f) and gliogenic phase based on Fu ef al.”® in pontine organoids over time (g). A smoothed line trend between the averaged values at different timepoints between week 01 and 12 was plotted using the LOESS algorithm and shaded area reflects the 95% confidence interval. h, Representative immunofluorescent 3D image of a 200 um thick organoid slice at week 16 labelled for neurofilament (NF, white), GFAP (red-to-white gradient) and AQP4 (green). White square indicates the zoomed in area of the optical section displayed in the top middle panel. i,

Representative immunofluorescent 3D image of a 200 um thick organoid slice at week 16 labelled for DAPI (white) and OLIG2 (red). White square indicates the zoomed in area zoomed of the optical section displayed in the bottom middle panel. h and i, Scale bars: overview image, 250 um; zoomed area, 50. j, Normalized RNA counts for organoids at week 04, 08 and 12 compared to a foetal cortical single cell reference®. *p = 0.0155 (week 04 compared to week 08) and *p = 0.0212 (week 04 compared to week 12), one-way ANOVA with post-hoc Tukey honest significance distance. k, Boxplot representation of Spearman’s rank coefficient between different organoid batches at week 02, 03, 04, 08 and 12. I, PCA of organoids at different timepoints between week 01 to 12 (grey scale) and derived from hESCs (circles) or iPSCs (square). j-l, n=9 organoids per 3 independent batches were analysed for each timepoint, except for 2 independent batches analysed at week 01. All boxplots depict min/max values, including median and lower/upper quartile with shaded area representing 95% confidence interval, if shown.

Figure 4 shows tumour induction efficiency, histopathological characterization, and comparison to unguided cerebral organoids. a, Stacked bar plot quantifying electroporation efficiency (grey columns) and tumour induction (dark grey columns) in pontine organoids tested at various timepoints ranging from day 11 to day 28. ns = not significant, **p < 0.01, two tailed independent t-test. Mean (+S.E.M) from n=23-35 individual organoids per timepoint from 9 independent batches. b, Representative image of GFP expression (left; tumour-inducing mix) or control (right; PiggyBac backbone including CAG-mVenus) as a measure of tumour outgrowth at week 6. Scale bars = 1 mm. n = 139 organoids from 9 independent batches. c,

Representative images of tumorigenic outgrowth of the same organoid at week 4, 6 and 8.

Scale bars = 500 um. Red arrowheads depict invasive and diffuse patterns. d, e,

Pathohistological staining for H3K27M and H3K27me3 on consecutive slices of a week 8 organoid. Black squares indicate displayed zoom area, H3K27M+/H3K27me3- cells annotated by red arrows. Scale bars = 100 um. e, Haematoxylin and eosin (HE), glial fibrillic acidic protein (GFAP) and neurofilament (NF). f, Stacked bar plot quantifying electroporation efficacy (grey columns) and tumour induction (dark grey columns) for guided pontine organoids as compared to unguided cerebral organoids at day 11. ns = not significant, *p < 0.05, two-tailed independent t-test. Mean (+ S.E.M) from n=3 independent experiments with a total of >35 individual organoids for each condition. g, Representative images of tumorigenic outgrowth (GFP; green) at week 4 and 8 for unguided cerebral organoids. Scale bars = 500 um. n = 36 unguided organoids and 35 pontine organoids from 3 independent batches. Stacked bar plots represent summarized data, see Methods for details.

Figure 5 shows DMG tumour induction and characterization in pontine-fated organoids. a,

Immunofluorescent 3D image of an intact DMGO at week 18. GFP signal colour-coded for z- depth on a rainbow scale, grey outline created by masking of Propidium lodide fluorescence.

Scale bar = 500 um. b, UMAP representation of tumour (red) and microenvironment (grey) cell fractions retrieved from single-cell and single-nuclei sequencing. ¢, UMAP representation of the tumour states (OPC-, MES-, AC-like and cycling) and the local microenvironment (neurons, glial intermediates, axon-guiding neuroepithelium and mesenchyme) identified in DMGOs, two months post-electroporation. n = 11 organoids from 5 independent batches. d, Heatmap representation of average transcriptomic similarity between DMGO tumour cells and in vitro models (Cell lines and Patient-derived Xenografts, PDX), H3K27M-altered DMG, high-grade adult Glioblastoma (GBM) and H3K27M-altered Posterior Fossa Group A Ependymoma (PFA,

H3K27M/EZHIP-mutants) patient samples. Average similarity (colour intensity) represents an averaged prediction score of all DMGO subsetted tumour cells mapped into a merged dataset consisting of transcriptomic in vitro and patient datasets’ (see Methods). e, Barplots showing proportional distribution of tumour cell subsets and OPC-like malignant cells predicted in detail to decipher OPC-like-1/-2/-3 states®. f, Boxplot of normalized expression of CRABP1 marker for

OPC-like tumour states across the 3 OPC-like subsets. g, Dotplot representation of normalized expression of CRABP1 in location-specific (Thalamic, Pontine and Cortical) DMG single cell data®. h, Representative immunofluorescent 3D images of a 200 um thick DMGO slice and

FFPE patient sample (far right) stained for DAPI (grey), H3K27M (green) and CRABP1 (red).

Square indicates displayed zoom area for the DMGO optical section (middle panel), white arrows annotate overlayed signals for H3.3K27M and CRABP1. n = 3 organoids and 2 patient

FFPE samples, overview image DMGO, scale bar = 250 um; optical section zoom, scale bar = 50 um; patient sample, scale bar = 25 um. i, Dotplot representation of normalized expression of

AQP1 in AC-, MES-, OPC-like tumour cells (see Table 2) in DMGOs and patient single-cell data®. j, Dotplot representation of normalized expression of AQP? and AQP4 in location-specific (Thalamic, Pontine and Cortical) DMG single cell data®. g, i, j, Dot size is proportional to the percentage of cells expressing a gene and color intensity to the average normalized gene expression. k, Representative immunofluorescent 3D images of a 100 um thick DMGO slice stained with DAPI (white), AQP1 (yellow) and GFAP (red) (left panel), square indicates optical section (middle top panel). FFPE section of H3K27M-altered DMG patient sample (right panel) stained with DAPI (white), AQP1(yellow), GFAP (red) and H3K27M (green), square indicates optical zoom (middle bottom panel). Scale bars: overview DMGO and Patient, 250 um; Optical

Section and Zoom, 25 um.

Figure 6 shows embedding of single nuclei/cell datasets and validation of malignant cell identification. a, b, Embedded datasets derived from either single nuclei (green) or single cell

(red) sequencing represented in the latent space using scVI-tools (a), or represented as UMAP and annotated for individual organoids (b). ¢, 30 clusters were detected using Louvain shared nearest neighbour computation, represented in embedded UMAP. d, Infer CNV computed copy number variation of subsetted DMGO tumour cells compared to a healthy brain organoid reference at day 11

Figure 7 shows annotation of subsetted healthy clusters. a, Heatmap representation of scaled expression (colour gradient) of top 25 DEGs scaled and normalized based on Louvain clustering and grouped by transcriptomic similarities of the healthy subsets. Canonical marker genes for each healthy subset (neurons, glioblasts, axon-guiding neuroepithelial and mesenchyme) are annotated. b, Barplot representation of neural subsetted healthy clusters mapped with cell identities {colour legend) form the latest foetal brain datasets by Ramos et al.44 grouped by transcriptomic similarities. ¢c, GO and KEGG pathway enrichment of neuronal identities for the neuronal clusters 5, 15 and 19.

Figure 8 shows annotation of subsetted tumour clusters. a, Barplot representation of tumour clusters mapped to the latest DMG cell states (OPC-, AC-, MES-like and cycling), based on Liu et al.° grouped by the cancer states majorly contributing to each cluster (>30%). b, Summarized prediction (dark) and mapping (grey) score per cell per cancerous state represented as a double violin plot. ¢, Heatmap representation of scaled expression (colour gradient) of top 25 DEGs scaled and normalized based on Louvain clustering grouped by transcriptomic similarities of the tumour subsets. Canonical marker genes for each cancerous DMG-state (OPC-, AC-, MES-like and cycling) are annotated.

Figure 9 shows barcode tracing of hindbrain-specific DMG tumorigenesis. a, UMAP of single cell RNAseq and Tracker-seq data. Cells are coloured according to unique lineage barcodes. Cells without barcodes are presented in grey. b, c‚ Larger versus smaller clones (<20% clone size per sample) are compared via DEGs. b, METASCAPE results showing selected GO terms form the highest scoring summary GO terms for small and large clones. Cc,

Volcano plot showing top genes expressed in larger and smaller clones. d, Module scores of gene programs as derived from cNMF projected onto the UMAP of a. e, UpSet plot displaying clonal intersection events. Only clonal families found in more than 1 cNMF module are depicted and filtered with at least 3 cells present per unique barcode. Bar plots depicts the frequency of each lineage combination (top} and the number of clones that contains each program (left). f,

Heatmap based on Jaccard Index gauging the similarity of the cNMF derived programs to the previous patient-derived annotation from Fig. 8a. g, SCENIC plot showing selected top10 regulons among program 3, 2 and 1. h, UMAP presenting normalized expression of STMN2 expression in DMGOs among the gene programs 1-8 (right UMAP). i, Heatmap presenting the mean cellular enrichment scores of gene programs 1 and 2 for midbrain-, forebrain- and pons oligo lineage signatures in first trimester brain cell atlas from Braun et af”.

Figure 10 shows barcode representation after quality control and filtering. a, Schematic representation of the applied approach for simultaneously recovering transcriptomic information,

HTO hashtags and lineage barcodes from DMGOs on a single cell level. A nested PCR strategy was applied for the TrackerSeq barcode b-f, Quality control assessment and filters used to select barcodes from experimental replicate 1 (top) and experimental replicate 2 (bottom). b,

Histogram depicting the total number of UMI counts prior to any filtering. ¢, Histogram displaying read counts and the cut-off (red dashed line) set as a minimum total read count of log10? and log103 for experimental replicate 1 and 2, respectively. d, Scatter plot depicting read counts plotted against UMI counts and the applied threshold for minimum total read counts indicated (red dashed line). e, Histogram depicting the mean oversequence per barcode and thresholding applied (dashed blue line). f, Scatter plot depicting read counts plotted against max mean oversequence showing both thresholds applied. g, UMAP embedding and cells coloured based on the total number of cells recovered from each of the different samples. h, Bargraph depicting the total number of cells for each clonal barcode after applying the filtering of >3 cells per clonal family. i, Pie charts of the relative size of each recovered clonal family among all barcoded cells per sample used for the large versus small clone comparison. Percentage is depicted for clonal families that are equal to, or above 20%, which are defined as large clones.

Figure 11 shows DMGO GD2 expression and CAR T cell mediated tumour control. a,

DMGO tumour cell GD2 expression (orange) analysed by flow cytometry compared to an unstained control (black). b, GD2 CAR T cell treatment outcome measured as tumour GFP intensity relative to the start of treatment (day 0, 100%). DMGOs were either left untreated (grey line, n=1), treated with mock transduced T cells (black lines, n=2), or GD2 CAR T cells (orange lines, n=4) and for each DMGO a smoothed line trend was plotted between the values at different timepoints using the LOESS algorithm. ¢, Images of tumour GFP signal on day 0, 7, 10 and 14 for an untreated DMGO, DMGOs treated with mock transduced T cells, or GD2 GAR T cells. GD2 CAR T cells and mock transduced T cells were administrated at day 0 and 7.

Figure 12 shows DMGOs model CAR T cell functional heterogeneity. a, GD2 CAR T cell treatment outcome measured as a relative change in tumour GFP intensity quantified by imaging compared to the start of treatment (100%). DMGOs were either left untreated (grey line, n=1), treated with mock transduced T cells (black line, n=2), or GD2 CAR T cells (orange line, n=4) and for each treatment condition a smoothed line trend between the averaged values at different timepoints was plotted using the LOESS algorithm. Shaded area reflects the 95% confidence interval. b, Representative images of the tumour GFP signal at the indicated timepoints for a DMGO subjected to prolonged GD2 CAR T cell treatment administrated at day 0, day 8 and day 15. ¢, UMAP visualization of GD2 CAR T cell clusters. d, Cytotoxic effector molecule and cytokine gene expression across the GD2 CAR T cell clusters. e, Gene expression of selected exhaustion associated receptors, ligands, and transcription factors across the GD2 CAR T cell clusters. d, e, Dot plot representing the percentage of cells expressing selected genes. Colour intensity represents the average scaled gene expression.

Figure 13 shows key marker genes, GO terms and reference data projection of GD2 CAR

T cell clusters. a, Dot plot showing key marker gene expression (selected from the top 20

DEGs) across the GD2 CAR T cell clusters. Dot size is proportional to the percentage of cells expressing a gene and colour intensity to the average scaled gene expression. Grid colours highlight genes that are closely related in function; HLA genes (green), metabolic stress-related genes (red) and ISGs (blue). b-f, Selected significant GO terms associated with the DEGs of the

Tuno (b), Ti2 (€), Tm (d), Ter (€) and Tus (Ff) GD2 CAR T cell clusters. g, UMAP visualization of the CD8* TIL clusters from the Chu et al. pan-cancer atlas®® used as a reference dataset.

Annotated clusters are highlighted because of their overlap with, or use in defining the GD2

CAR T cell clusters. h-j, Curated marker gene signatures (DEG analysis adjusted p-value < 0.00001) of the Tuno (h), Ti2 (i) and Tiss (j) GD2 CAR T cell clusters projected onto the CD8*

TIL dataset from g.

Figure 14 shows violin plots of lineage- and stage specific markers for OC-like (A) and AC-like (B) cancer cells, based on single cell sequencing performed at week 8 (T1) and week 16 (T2).

Significance calculated using t-test, with 95% confidence, p-value "***"=0.001, "**"=0.01, ""=0.05.

The patent, scientific and technical literature referred to herein establish knowledge that was available to those skilled in the art at the time of filing. The entire disclosures of the issued patents, published and pending patent applications, and other publications that are cited herein are hereby incorporated by reference to the same extent as if each was specifically and individually indicated to be incorporated by reference. In the case of any inconsistencies, the present disclosure will prevail.

Various aspects of the invention are described in further detail below.

Detailed Description

Provided herein are method for producing hindbrain organoids. The methods utilise various culture medias in order to pattern cells in order to provide organoids that are mimetic of the hindbrain. In particular, the organoids produced may be mimetic of the pons. Such organoids may be referred to herein as “pontine” or “pontine organoids”.

As used herein, "organoid" refers to a heterogeneous 3D agglomeration of cells that recapitulates aspects of cellular self-organization, architecture and signalling interactions present in a native organ. The term "organoid" includes spheroids or cell clusters formed from suspension cell cultures. Such organoids may be derived from pluripotent stem cells. For example, organoids may be derived from embryonic stem cells or induced pluripotent stem cells.

In some examples, the organoids are derived from induced pluripotent stem cells. In some examples, the organoids are derived from human induced pluripotent stem cells. Induced pluripotent (iPS) cells are somatic cells which are re-programmed to ESC-like cells capable of differentiation into representative tissues of the three embryonic germ layers both in vitro and in vivo. As used herein “induced pluripotent stem (iPS) cell” (or embryonic-like stem cell) refers to a proliferative and pluripotent stem cell which is obtained by de-differentiation of a somatic cell (e.g., an adult somatic cell).

Methods of producing induced pluripotent stem cells are well known. For example, IPS cells can be endowed with pluripotency by genetic manipulation which re-programs the cells to acquire embryonic stem cells characteristics. For example, iPS cells can be generated from somatic cells by induction of expression of Oct-4, Sox2, Kfl4 and c-Myc in a somatic cell as described in Takahashi and Yamanaka, 2006, Takahashi et al, 2007, Meissner et al, 2007, and

QOkita K., et al, 2007, Nature 448: 313-318). Additionally or alternatively, iPS cells can be generated from somatic cells by induction of expression of Oct4, Sox2, Nanog and Lin28 as described in Yu et al, 2007, and Nakagawa et al, 2008. It should be noted that the genetic manipulation (reprogramming) of the somatic cells can be performed using any known method such as using plasmids or viral vectors, or by derivation without any integration to the genome (Yu J, et al, Science. 2009, 324: 797-801).

IPS cells can be obtained by inducing de-differentiation of embryonic fibroblasts (Takahashi and

Yamanaka, 2008; Meissner et al, 2007), fibroblasts formed from hESCs (Park et al, 2008), Fetal fibroblasts (Yu et al, 2007; Park et al, 2008), foreskin fibroblast (Yu et al, 2007; Park et al, 2008), adult dermal and skin tissues (Hanna et al, 2007; Lowry et al, 2008), b-lymphocytes (Hanna et al 2007) and adult liver and stomach cells (Aoi et al, 2008).

IPS cell lines are also available via cell banks such as the WiCell bank. In some examples, the induced pluripotent stem cells are human induced pluripotent stem cells.

In some examples, the organoids are derived from embryonic stem cells (ESCs). ESCs can be obtained using well-known cell-culture methods. For example, human embryonic stem cells can be isolated from human blastocysts. Human blastocysts are typically obtained from human in vivo preimplantation embryos or from in vitro fertilized (IVF) embryos. Alternatively, a single cell human embryo can be expanded to the blastocyst stage. For further details on methods of preparation of human ESC see Thomson et al., (U.S. Pat. No. 5,843,780; Science 282: 1145, 1998; Curr. Top. Dev. Biol. 38: 133, 1988; Proc. Natl. Acad. Sci. USA 92: 7844, 1995); Bongso etal, (Hum Reprod 4: 706, 1989); and Gardner et al., (Fertil. Steril. 69: 84, 19980).

In some examples, the embryonic stem cells are human embryonic stem cells. Embryonic stem cells may be produced or obtained by methods that do not damage or destroy an embryo, such as: blastomere extraction - a procedure that involves removing one or two cells from an early embryo, which is often used in preimplantation genetic testing (PGD). The remaining embryo can still be implanted in a uterus, and the blastomeres can be used to generate stem cells; single-cell embryo biopsy - a technique similar to PGD that involves taking a single cell from an early stage IVF embryo. This cell can then be used to develop a new line of ESC. For other examples of non-destructive methods for providing ESCs see DE102004082184B, Dittrich, R et al. “Non-embryo-destructive Extraction of Pluripotent Embryonic Stem Cells: Implications for

Regenerative Medicine and Reproductive Medicine.” Geburtshilfe und Frauenheilkunde vol. 75,12 (2015): 1239-1242. doi:10.1055/5-0035-1558183, Mertes, Heidi, Guido Pennings, and

André Van Steirteghem. "An ethical analysis of alternative methods to obtain pluripotent stem cells without destroying embryos." Human Reproduction 21.11 (2006): 2749-2755 and

Suaudeau, Jacques. "From embryonic stem cells to iPS—an ethical perspective." Cell proliferation 44.Suppl 1 (2011): 70 and the references cited therein.

As used herein “hindbrain organoid” refers to an organoid that recapitulates or is mimetic of the hindbrain. The hindbrain (developmentally derived from the rhombencephalon) is one of the three major regions of the mammalian (e.g. human) brain, located at the lower back part of the brain. It includes most of the brainstem and a dense coral-shaped structure called the cerebellum. The brainstem connects the brain to the spinal cord and coordinates many vital functions, such as breathing and heartbeat. There are three main parts of the hindbrain - pons, cerebellum, and medulla oblongata. Most of the 12 cranial nerves are found in the hindbrain.

As used herein “pontine organoid” refers to an organoid that recapitulates or is mimetic of the pons. The pons is the largest part of the brainstem, located above the medulla and below the midbrain. The pons develops from the embryonic metencephalon {part of the hindbrain, developed from the rhombencephalon), alongside the cerebellum. The pons is a group of nerves that function as a connection between the cerebrum and cerebellum. The anterior or ventral surface of the pons is marked by a bulging formed by the transverse pontocerebellar fibres. These fibres wrap around the otherwise vertically oriented brainstem. It measures around 2.5 cm in adults. The basilar groove demarcates the midline of the ventral surface and is where the basilar artery is located. The pontomedullary junction is an important anatomical landmark defined by the angle between the lower border of the pons and the superior border of the medulla. Several cranial nerves originate from the ventral surface of the pons: Cranial nerve V: trigeminal — originates from the lateral aspect of mid pons; Cranial nerve VI: abducens — originates from the pontomedullary junction, close to the midline; Cranial nerve VII: facial — originates from the cerebellopontine angle, the more lateral aspect of the pontomedullary junction; and Cranial nerve VIII: vestibulocochlear — originates laterally to the facial nerve. The pons is intimately related to the cerebellum and is connected to it by the middle cerebellar peduncles. Removal of the cerebellum will reveal the underlying fourth ventricle. The floor of the fourth ventricle is composed of the dorsal surface of the pons and the medulla. The pons is comprised of two major components — the ventral pons and the tegmentum. The ventral pons contains the pontine nuclei, which are responsible for coordinating movement. Fibres from the pontine nuclei cross the midline and form the middle cerebellar peduncles on their way to the cerebellum. The tegmentum is the evolutionarily older part of the pons which forms part of the reticular formation — a set of nuclei found throughout the brainstem that are responsible for arousal and attentiveness. The blood supply of the pons is formed by branches of the vertebrobasilar system. Most of the pons is supplied by the pontine arteries, branches of the basilar artery. A smaller part of the blood supply comes from the anterior inferior cerebellar artery and the superior cerebellar artery (AICA and SCA). The venous drainage of the pons consists of the anterior pontomesencephalic vein, which drains superiorly into the basal vein, that in turn drains into the cerebral veins. Inferiorly, the pons drains into the inferior petrosal sinus, which drains into the internal jugular veins.

The methods described herein include the use of certain mediums which are described in detail below.

Culture Media

The culture media described herein may share the same or similar base medium composition (also referred to herein as BASE medium). The base medium composition may include a basal medium, a neurobasal medium and one or more supplement compositions. “Basal medium” refers to any culture medium capable of supporting cell growth, typically comprising inorganic salt, vitamin, glucose, buffer system and essential amino acids, and typically having an osmolarity of about 280 to 330 mOsmol. Examples of basal medium include

Dulbecco's medium (e.g. IMDM), Eagle's medium (e.g. DMEM, EMEM, BME, MEM, aMEM),

Ham medium (e.g. F10 medium, F12 medium}, RPMI medium (e.g. RPMI-1640). Medium,

RPMI-1630 medium), MCDB medium (e.g. MCDB104, 107, 131, 151, 153 medium), Fisher medium, 199 medium, culture medium for primate ES cells (culture medium for primate ES/iPS cells, Reprocell), Mouse ES cell medium (TX-WES culture medium, Thromb-X), serum-free medium (mTeSR, Stemcell Technologyies), ReproFF, StemSpan (registered trademark) SFEM,

StemSpan (registered trademark) H3000, Stemlinell, ESF-B medium, ESF-C medium, CSTI-7 medium, Neurobasal medium (Life Technologies), StemPro-34 medium, StemFit (registered trademark) (e.g., StemFit AKO3N, StemFit AKO2N), and the like.

In some examples, the basal medium is Dulbecco's Modified Eagle Medium/Ham's F-12 (Advanced DMEM/F-12 — available from Gibco, Cat. #12634010). “Neurobasal media” refers to cell or organoid growing basal medium that is designed for long- term maintenance and maturation of substantially pure or pure pre-natal and embryonic neuronal cell populations without the need for an astrocyte feeder layer when supplemented.

Neurobasal media is commercially available from a variety of vendors, including

ThermoFisher™ Scientific, VWR™ Sigma Aldrich™, US Bio™, and STEMCELL™

Technologies.

In some examples, the neurobasal medium is neurobasal medium available from Gibco, Cat. #10888022.

In some examples, the base culture medium includes a basal medium and a neurobasal medium at a ratio of 1:1

The one or more supplements may include factors and growth agents that promote the survival, growth and differentiation of neuronal cells. For example, supplements for the base medium can be purchased from a variety of vendors, and include B-27™ Plus, N-2 and GlutaMAX™ supplements from ThermoFisher Scientific, NeuroCult™ and STMdiff™ supplements from

STEMCELL Technologies, GEM21 NeuroPlex™ and N2 NeuroPlex™ from Gemini Bio-

Products, and NDiff™ supplements from Sigma Aldrich.

In some examples, the base media includes GlutaMAX™ as a supplement. In some examples, the base media includes GlutaMAX™ at a 1x concentration.

The methods described herein utilise an initial culture media in which an embryoid body comprising pluripotent stem cells as described herein is initially cultured. The initial culture medium is suitable for the formation of an embryoid body from pluripotent stem cells. The term "embryoid body" refers to a three-dimensional aggregate of pluripotent stem cells. The initial culture medium may also be referred to as “day 0” culture medium herein. The initial culture medium may include a base medium as described above (for example a basal medium, a neurobasal medium and one or more supplements). In some examples, the initial culture medium includes a basal medium and a neurobasal medium at a ratio of 1:1.

The initial culture medium also includes a ROCK inhibitor. A ROCK inhibitor is a type of protein kinase inhibitor that inhibits rho-associated protein kinase (ROCK), a kinase of the serine- threonine protein kinase family, and prevents apoptosis of cells after dissociation or thawing. In some examples, the ROCK inhibitor is Y-27632. In some examples, the ROCK inhibitor is included in the initial culture medium at a concentration of about 10uM.

The initial culture medium also includes one or more fibroblast growth factors (FGFs). The term “fibroblast growth factor” refers to a family of growth factors, with members involved in angiogenesis, wound healing, embryonic development and various endocrine signaling pathways. The fibroblast growth factors are heparin-binding proteins and interactions with cell- surface-associated heparan sulfate proteoglycans have been shown to be essential for fibroblast growth factor signal transduction. Fibroblast growth factors are key players in the processes of proliferation and differentiation of a wide variety of cells and tissues. The functions of FGFs in developmental processes include mesoderm induction, antero-posterior patterning, limb development, neural induction and neural development, and in mature tissues/systems angiogenesis, keratinocyte organization, and wound healing processes.

In some examples, the initial culture medium includes FGF2. In some examples, the initial culture medium includes FGF2 at a concentration of about 4ng/ml.

As such, provided herein is an initial culture media comprising or essentially consisting of:

neurobasal medium and Advanced DMEM/F-12 medium at a ratio of 1:1; 1xGlutaMAXTM supplement;

FGF2 at a concentration of about 4 ng/ul; and

Y-27632 at a concentration of about 10 HM.

Also provided herein are culture media for patterning a hindbrain organoid as described herein.

These may be referred to as patterning media and may all include a patterning base medium composition (also referred to herein as PATTERNING medium).

The patterning base media composition includes the base media as described above, N2 supplement and heparin solution. In some examples, the N2 supplement is at a concentration of about 1xN2 supplement.

In some examples, the heparin solution is at a concentration of at least 1 ug/ml. In some examples, the heparin solution is at a concentration of about 2 ug/ml.

For example, patterning base medium composition comprises: neurobasal medium and Advanced DMEM/F-12 medium at a ratio of 1:1; 1xGlutaMAXTM supplement; 1xN2 supplement; and 2 pg/ml heparin solution.

In one aspect there is provided a first culture medium for proliferation and at least partial differentiation of pluripotent stem cells of an embryoid body. This may be referred to herein as “Week 1 medium”.

The first culture medium may include FGF2, a bone morphogenetic protein (BMP) pathway inhibitor, a TGF-B inhibitor and a WNT activator.

In some examples, the first culture media comprises FGF2 at a concentration of at least 10 ng/ml. In some examples, the first culture media comprises FGF2 at a concentration of at most 50 ng/ml. For example, the first culture media comprises FGF2 at a concentration of 10, 15, 20, 25, 30, 35, 40, 45 or 50 ng/ml. In some examples, the first culture media comprises FGF2 at a concentration of up to 50 ng/ml. In some examples, the first culture media comprises FGF2 at a concentration of about 50 ng/ml.

Bone morphogenetic protein (BMP) pathway inhibitor refers to compounds capable of inhibiting

BMP signalling pathway, including but not limited to dorsomorphin 2HCI, dorsomorphin, and

LDN-193189. In some examples, the BMP pathway inhibitor is dorsomorphin. In some examples, the first culture medium includes a BMP pathway inhibitor at a concentration of about 10 pM.

The first culture medium may include at least one TGF-B inhibitor. A TGF-B inhibitor as used herein includes an agent that reduces the activity of the TGF-B signalling pathway. There are many different ways of disrupting the TGF-B signalling pathway known in the art, any of which may be used in conjunction with the methods and culture mediums described herein. For example, TGF-B signalling may be disrupted by: inhibition of TGF-B expression by a small- interfering RNA strategy; inhibition of furin (a TGF-B activating protease); inhibition of the pathway by physiological inhibitors, such as inhibition of BMP by Noggin, dorsomorphin and

LDN-193189 (e.g. as described above), DAN or DAN-like proteins; neutralization of TGF-B with a monoclonal antibody; inhibition with small-molecule inhibitors of TGF-B receptor kinase 1 (also known as activin receptor-like kinase, ALK5}, ALK4, ALK6, ALK7 or other TGF-B-related receptor kinases; inhibition of Smad 2 and Smad 3 signalling by overexpression of their physiological inhibitor, Smad 7, or by using thioredoxin as an Smad anchor disabling Smad from activation (Fuchs, Inhibition of TGF-B Signalling for the Treatment of Tumor Metastasis and

Fibrotic Diseases. Current Signal Transduction Therapy 6(1):29-43(15), 2011). For example, a

TGF-B inhibitor may target a serine/threonine protein kinase selected from: TGF- receptor kinase 1, ALK4, ALKS, ALK7, or p38. ALK4, ALK5 and ALK? are all closely related receptors of the TGF-B superfamily. ALK4 has GI number 91; ALKS (also known as TGF-B receptor kinase 1) has GI number 7046; and ALK7 has Gl number 658. An inhibitor of any one of these kinases is one that effects a reduction in the enzymatic activity of any one (or more) of these kinases.

As such, in some examples, the TGF-B inhibitor acts as a SMAD inhibitor. In some examples, the BMP pathway inhibitor (e.g. dorsomorphin) may be referred to as a TGF- inhibitor or in some cases a SMAD inhibitor.

The term “SMAD inhibitor” refers to a compound of molecule capable of inhibiting (that is preventing or downregulating) the activity of a SMAD protein. “SMAD” refers to intracellular proteins that transduce extracellular signals from transforming growth factor beta (TGF-B) ligands to the nucleus where they activate downstream gene transcription. The SMADs, which form a trimer of two receptor-regulated SMADs and one co-SMAD, act as transcription factors that regulate the expression of certain genes. Other SMAD proteins are, but are not limited to,

SMAD1, SMAD2 (also known as Mothers against decapentaplegic homolog 2, JV18, JV18-1,

MADH2, MADR2, hMAD-2, or SMAD family member 2), SMAD3 (also known as Mothers against decapentaplegic homolog 3, HSPC193, HsT17436, JV15-2, LDS1C, LDS3, MADH3, or

SMAD family member 3), SMAD4 (a common-mediator SMAD (co-SMAD), also known as

SMAD family member no 4, Mothers against decapentaplegic homolog 4, JIP, MADH4,

MYHRS, or DPC4 (Deleted in Pancreatic Cancer-4)), SMADS5 (also known as Mothers against decapentaplegic homolog 5, DWFC, JV5-1, MADHS5, or SMAD family member 5), SMAD6 (an antagonistic or inhibitory SMAD, which blocks activation of R-SMADs and co-SMADs; also known as AOVD2, HsT17432, MADH8, MADH7, SMAD family member 6), SMAD7 (an antagonistic or inhibitory SMAD, which blocks activation of R-SMADs and co-SMADs; also known as CRCS3, Mothers against decapentaplegic homolog 7 (MADH7), MADH8, SMAD family member 7), and SMAD8/9 (also known as Mothers against decapentaplegic homolog 9,

SMAD9, SMAD8, MADH9, PPH2, SMAD8, SMAD8A, SMAD8B, SMAD family member 9 or

MADHB6). SMAD inhibitors include A-83-01 (3-(6-Methylpyridin-2-yl)-1-phenylthiocarbamoyl-4-

quinolin-4-ylpyrazole; Alk-5 inhibitor, Masayoshi et al, 2005), GW6604 (2-phenyl-4-(3-pyridin-2- yl-1H-pyrazol-4-yl)pyridine; Alk-5 inhibitor. Sawyer et al, 2003), and SB-431542 (4-(5- benzo[1,3]dioxol-5-yl-4-pyridin-2-yl-1H-imidazol-2-yl)-benzamide). Also suitable for use is a pyrimidine derivative as described in WO 2008/006583

In some examples, the TGF-B inhibitor is SB431542. In some examples, the first culture medium comprises a TGF-B inhibitor at a concentration of about 1 uM.

In some examples, the use of dorsomorphin and SB431542 may be referred to as “dual SMAD inhibitors”. It will be recognised by those skilled in the art these may also be referred to as a first and second TGF-6 inhibitor or more specifically referred to as a SMAD inhibitor and BMP pathway inhibitor. “Wnt- activator” refers to a molecule or compound which activates or upregulates genes involved in the Wnt signalling pathway. Wnt signalling pathway refers to a group of signal transduction pathways made of proteins that pass signals into a cell through cell surface receptors. Three Wnt signalling pathways have been characterized: the canonical Wnt pathway, the non-canonical planar cell polarity pathway, and the non-canonical Wnt/calcium pathway. All three pathways are activated by binding a Wnt-protein ligand to a Frizzled family receptor, which passes the biological signal to the Dishevelled protein inside the cell.

Examples of Wnt activators include, , 2-Amino-4-[3,4-(methylenedioxy)benzylamino]-6-(3- methoxyphenyl)pyrimidine (CAS no. 853220-52-7), (1-(4-(Naphthalen-2-yl)pyrimidin-2- ylhpiperidin-4-ylymethanamine (WAY 262611 or DKK1 inhibitor), WAY-316606 (5- (Phenylsulfonyl)-N-4-piperidinyl-2-{trifluoromethyl)benzene sulfonamide hydrochloride), heteroarylpyrimidines, arylpyrimidines, IQ1 (2-[2-(4-Acetylphenyl)diazenyl]-2-(3,4-dihydro-3,3- dimethyl-1(2H)-isoquinolinylidene)acetamide; CAS no. 331001-62-8), QS11 ((2S)-2-[2-(Indan-5- yloxy)-9-(1,1'-biphenyl-4-ylymethyl)-SH-purin-6-ylamino]-3-phenyl-propan-1-ol; CAS no. 944328- 88-5), SB-216763 (3-(2,4-dichlorophenyl)-4-(1-methylindol-3-yl)pyrrole-2,5-dione), BIO(6- bromoindirubin-3'-oxime), deoxycholic acid (DCA), 2-amino-4-[3,4-(methylenedioxy)benzyl- amino]-8-(3-methoxyphenyl)pyrimidine, or derivatives thereof.

Wht activators also include GSK3 inhibitors such as CHIR-99021 (6-[2-[[4-(2,4-dichlorophenyl)- 5-(5-methyl-1H-imidazol-2-yl)pyrimidin-2-ylJamino]ethylamino]pyridine-3-carbonitrile), BIO(6- bromoindirubin-3'-oxime), SB 216763 (3-(2,4-dichlorophenyl)-4-(1-methylindol-3-yl)pyrrole-2,5- dione), CHIR-98014 (6-N-[2-[[4-(2,4-dichlorophenyl)-5-imidazol-1-ylpyrimidin-2-ylJamino]ethyl]- 3-nitropyridine-2,6-diamine), TWS119 (3-[[6-(3-aminophenyl)-7H-pyrrolo[2,3-d]pyrimidin-4- ylloxylphenol), IM-12 (3-[2-{4-fluorophenyhethylamino]-1-methyl-4-(2-methyl-1H-indol-3- ylpyrrole-2,5-dione), 1-azakenpaullone 9-bromo-7,12-dihydropyrido[3',2".2,3]azepino[4,5- b]indol-6{(5H)-one, AR-A014418 1-[{4-methoxyphenyl)methyl]-3-(5-nitro-1,3-thiazol-2-yl)urea,

SB415286 3-(3-chloro-4-hydroxyanilino)-4-(2-nitrophenyl)pyrrole-2,5-dione, AZD1080 (3E)-3-[5- (morpholin-4-ylmethyl)- 1H-pyridin-2-ylidene]-2-oxo- 1H-indole-5-carbonitrile, AZD2858 3-amino-

6-[4-{4-methylpiperazin-1-yl) sulfonylphenyl]-N-pyridin-3-ylpyrazine-2-carboxamide, indirubin (3E)-3-{3-0x0-1H-indol-2-ylidene)-1H-indol-2-0ne or derivatives thereof.

In some examples, the Wnt activator is CHIR99021. In some examples, the first culture medium comprises a Wnt activator at a concentration of about 2.5 to about 3.5 HM. For example, about 2.5, 2.6, 2.7, 2.8, 2.9, 3, 3.1, 3.2, 3.3, 3.4, or 3.5 uM. In some examples, the first culture medium comprises a Wnt activator at a concentration of about 3 WM.

As such, in one aspect there is provided a first culture medium comprising:

FGF2 at a concentration of at most about 50 ng/l;

Dorsomorphin at a concentration of about 1 uM;

SB431542 at a concentration of about 10 uM; and

CHIR99021 at a concentration of about 3 HM.

As such, in one aspect there is provided a first culture medium comprising or essentially consisting of: neurobasal medium and Advanced DMEM/F-12 medium at a ratio of 1:1; 1xGlutaMAXTM supplement; 1xN2 supplement; 2 pg/ml heparin solution;

FGF2 at a concentration of at most about 50 ng/yl;

Dorsomorphin at a concentration of about 1 WM;

SB431542 at a concentration of about 10 uM; and

CHIR98021 at a concentration of about 3 HM.

Also provided herein is a second culture medium for producing committed neuroectodermal cells in an organoid as described herein. The second culture medium may be referred to herein as “week 2” medium.

The second culture medium may include FGF4, retinoic acid, a sonic hedgehog activator, a bone morphogenetic protein (BMP) pathway inhibitor, a TGF-B inhibitor, and a WNT activator.

Sonic hedgehog (Shh)” refers to a protein constituting a mammalian signalling pathway called hedgehog, which is the most studied ligand in the hedgehog signalling pathway, and it is known to play an important role in regulating organ formation in vertebrates. Examples of sonic hedgehog activators include proteins belonging to the hedgehog family (e.g. Shh and Shh-N),

Shh receptors, and Shh receptor agonists such as purmorphamine.

In some examples, the sonic hedgehog activator is purmorphamine. In some examples, the second culture medium comprises a sonic hedgehog activator at a concentration of about 1 WM.

Retinoic acid refers to an active form (synthetic or natural) of vitamin A, capable of inducing neural cell differentiation. In some examples, the second culture medium comprises retinoic acid at a concentration of at least 1 uM. For example, at least about 1, 2, 3,4,5,6,7, 8,9, 10

MM. In some examples, the second culture medium comprises retinoic acid at a concentration of about 10 uM. In some examples, the retinoic acid is all-trans retinoic acid.

In some examples, the second culture medium comprises FGF4 at a concentration of about 10 ng/ml.

In some examples, the bone morphogenetic protein (BMP) pathway inhibitor, the TGF-B inhibitor, and the WNT activator of the second culture media are as described above. For example, BMP pathway inhibitor is dorsomorphin. For example, the Wht activator is

CHIR99021. For example, the TGF-B inhibitor is SB431542.

As such, in one aspect there is provided a second culture medium comprising:

FGF4 at a concentration of about 10 ng/pl;

Retinoic acid at a concentration of about 10 HM;

Purmorphamine at a concentration of about 1 HM;

Dorsomorphin at a concentration of about 1 HM;

SB431542 at a concentration of about 10 uM; and

CHIR99021 at a concentration of about 3 uM.

As such, in one aspect there is provided a second culture medium comprising or essentially consisting of: neurobasal medium and Advanced DMEM/F-12 medium at a ratio of 1:1; 1xGlutaMAXTM supplement; 1xN2 supplement; 2 pg/ml heparin solution;

FGF4 at a concentration of about 10 ng/pl;

Retinoic acid at a concentration of about 10 uM;

Purmorphamine at a concentration of about 1 WM;

Dorsomorphin at a concentration of about 1 WM;

SB431542 at a concentration of about 10 uM; and

CHIR99021 at a concentration of about 3 pM.

In some examples, the second culture medium may include an extracellular matrix component.

In some examples, the extracellular matrix component is added to the second culture medium during the second time period described herein. In some examples, cells cultured in the second culture medium may be added to an extracellular matrix component in the second culture medium. Examples of extracellular components include matrigel, gelatine, methylcellulose, collagen, alginate, alginate beads, agarose, fibrin, fibrin glue, fibrinogen, blood plasma fibrin beads, whole plasma or components thereof, laminins, fibronectins, proteoglycans, HSP, chitosan, heparin, other synthetic polymer or polymer scaffolds and solid support materials. In one example, the extracellular matrix component is made of matrigel.

Also provided herein is a third culture medium for producing an organoid as described herein.

The third culture medium may be referred to herein as “week 3” medium.

The third culture medium may include FGF4, retinoic acid, and a sonic hedgehog activator.

In some examples, the third culture medium includes FGF4 at a concentration of about 10 ng/ml.

In some examples, the third culture medium includes retinoic acid at a concentration of about 10

WM.

In some examples, the third culture medium includes a sonic hedgehog activator at a concentration of about 1 HM. In some examples, sonic hedgehog activator is as described herein.

As such, in one aspect there is provided a third culture medium comprising:

FGF4 at a concentration of about 10 ng/pl;

Retinoic acid at a concentration of about 10 HM; and

Purmorphamine at a concentration of about 1 uM;

As such, in one aspect there is provided a third culture medium comprising or essentially consisting of: neurobasal medium and Advanced DMEM/F-12 medium at a ratio of 1:1; 1xGlutaMAXTM supplement; 1xN2 supplement; 2 pg/ml heparin solution;

FGF4 at a concentration of about 10 ng/pl;

Retinoic acid at a concentration of about 10 HM; and

Purmorphamine at a concentration of about 1 WM.

In some examples, the third culture medium includes the extracellular matrix component added to the second culture medium.

Without being bound by theory, the relatively high concentration of retinoic acid and use of

FGF4 is believed to help form organoids having hindbrain identity and more specifically pontine identity (e.g. including HOXB1 expressing cells).

Also provided herein is a maturation medium (also referred to herein as a fourth culture medium). “Maturation medium” refers to a chemically defined medium useful for development of hindbrain organoids as described herein. Examples of such medium are well known in the art.

In some examples, the maturation medium includes a neurobasal medium, a basal medium,

GlutaMax supplement, N2 supplement, B27 supplement, and an antibiotic.

In some examples, the B27 supplement does not include vitamin A. In some examples, the 27 supplement is 1x B27 without vitamin A available from Gibco, Cat. #12587010.

In some examples, the neurobasal medium is as described herein.

In some examples, the basal medium is as described herein.

In some examples, the GlutaMax supplement is as described herein.

In some examples, the antibiotic is penicillin and/or streptomycin. In some examples, the antibiotic is penicillin-streptomycin. For example, Pen-Strep available from Gibco, Cat. #15140122.

In some examples, the maturation medium comprises or essentially consists of neurobasal medium and Advanced DMEM/F-12 medium at a ratio of 1:1, 1xGlutaMax supplement, and 0.5xN2 supplement, 0.5xB27 supplement and 1x Penicillin-Streptomycin.

In some examples, the maturation medium includes the extracellular matrix component added to the second culture medium.

In some examples, the pluripotent stem cells may be cultured in a proliferation medium prior to culturing in the initial culture medium. Proliferation medium may be a medium comprising a source of nutrients, such as vitamins, minerals, carbon and energy sources, and other beneficial compounds that facilitate the biochemical and physiological processes occurring during expansion or proliferation of cells. The proliferation medium may comprise one or more carbon sources, vitamins, amino acids, and inorganic nutrients. Representative carbon sources include monosaccharides, disaccharides, and/or starches. For example, the proliferation medium may contain one or more carbohydrates such as sucrose, fructose, maltose, galactose, mannose, and lactose. The proliferation medium may also comprise amino acids. Suitable amino acids may include amino acids commonly found incorporated into proteins as well as amino acids not commonly found incorporated into proteins, such as argininosuccinate, citrulline, canavanine, ornithine, and D-steroisomers. The proliferation medium may also comprise proteins such as foetal bovine serum albumin. The proliferation medium may also comprise antibiotics.

For example, the proliferation medium may be mTeSR Plus medium available from Stem Cell

Technologies, Cat. #100-0276.

Method of culturing

The methods of culturing provided herein include a number of steps which include a step of embryoid body formation, organoid patterning steps and a maturation step. In some examples, the methods may also include a step of proliferating pluripotent stem cells prior to embryoid body formation. In some examples, the methods may also include a step of maintaining the hindbrain organoids after maturation.

In some examples, the methods described herein include culturing pluripotent stem cells in a proliferation medium as described herein. The pluripotent stem cells may be cultured until a desired confluence of cells is achieved. For example, the pluripotent stem cells may be cultured to a confluence of about 70% or more. For example, 70%. For example, 80%.

Once the pluripotent stem cells have reached a desired confluence, they may be transferred from the proliferation medium to a base medium as described herein.

In some examples, a number of pluripotent stem cells are transferred from the base medium to the initial culture medium as described herein. In some examples, the concentration of cells transferred to the initial culture medium is at least 50,000 cells/ml. In some examples, the concentration of cells transferred to the initial culture medium is at most 90,000 cells/ml. For example, the concentration of cells is 50,000, 60,000, 70,000, 80,000, or 20,000 cells/ml. In some examples, the concentration of cells transferred to the initial culture medium is about 70,000 cells/ml.

After transfer of the pluripotent stem cells to the initial medium, a number of cells are then deposited in a culture vessel. For example, a well of a culture plate. For example, the vessel may be a well of an ultra-low attachment (ULA) treated U-bottom 96-well plate. Other suitable culture vessels will be known. In some examples, around 7000 cells are deposited into the vessel. It will be understood that the volume of cells used will be dependent on concentration of cells. In some examples, the number of cells deposited is less than 9000 cells. In some examples, the number of cells deposited is 7000 cells. Without being bound by theory, due to the effects of WNT activation in WEEK1 and 2 medium, the cells may undergo enhanced proliferation behaviour, meaning the organoids grow in size quickly. Due to the rapid proliferation, the organoid may develop a necrotic core, due to a lack of oxygen diffusion. When starting with 9000 cells/organoids, the necrotic core and organoids may become too big, and this may lead to negative effects on cellular behaviour and/or differentiation.

Deposition of the pluripotent stem cells to the vessel may be referred to as “day 0°. This means that time periods described herein are counted as hours, days or weeks from day O unless otherwise specified. For example, if the pluripotent stem cells are added to the vessel and cultured in the initial culture medium for 48 hours, this would be culturing to a time of 2 days.

After deposition of the pluripotent stem cells to the vessel, the cells are cultured for an initial time period suitable for producing an embryoid body. In some examples, the initial time period is about 48 hours or 2 days.

Culturing in the initial time period may be carried under conditions suitable to maintain the pluripotent stem cells and for formation of the embryoid body.

Standard ambient growth conditions for cells in cell culture are usually a temperature of 37°C., a

CO: content of 5% and a humidity of 95%. All these conditions can be achieved and maintained by using, for example, an incubator.

In some examples, pluripotent stem cells are cultured in the initial culture medium at a temperature of 37°C and a CO: content of 5% for about 48 hours.

After the initial time period (e.g. up to 2 days), the embryoid body is cultured in a first culture medium as described herein.

As used herein the term refresh refers to removal of a portion of culture medium in a vessel and addition of a volume of fresh culture medium (e.g. the same type of culture medium removed or a different culture medium). The term replace is used to refer to the removal of the majority {substantially all) of the culture medium in a vessel and then the addition of a volume of replacement culture medium (e.g. the same type of culture medium removed or a different culture medium). The term topped-up refers to the addition of culture medium to a sample without removal of culture medium already in the vessel.

It will be understood that when culture medium is refreshed or topped up with a different culture medium, in order to provide a desired concentration of components in the different culture medium dilution needs to be taken into account. For example, if 200pl of initial media is refreshed with 100 of first culture medium, 100ul of initial culture medium is removed and 100pl of double concentrated first culture medium is added.

In some examples, the initial culture medium is replaced with the first culture medium. For example, the initial culture media may be removed from the vessel and the first culture medium added to the vessel. In some examples, the initial culture medium may be topped-up with first culture medium {2x week 1 culture medium) comprising double concentration of the components thereof to reach a final concentration as described herein. In some examples, the initial culture medium may be refreshed with a volume of week 1 culture media. For example, 50% of the volume of initial culture media is refreshed with the same volume of 2x concentrated first culture medium.

In some examples, the first culture medium may be added to the vessel which includes the initial culture medium and embryoid body. The volume of first culture medium added to the vessel may be equal to the volume of cells and initial culture medium added to the vessel. For example, if 100 pl of initial culture medium including 7000 cells (i.e. 100 HI of cells at a concentration of 70,000 cells/ml) then 100 pl of the first culture medium may be added to the vessel after initial embryoid body formation.

The first culture medium is added to the vessel at day 2. The embryoid body and pluripotent stem cells thereof are then cultured in the first culture medium for a first time period. In some examples, the first time period starts at day 2 and proceeds to about day 7. For example, the first time period is up to about 5 days. For example, the first time period starts on day 2 from day O and is up to day 7 from day 0.

The first culture medium may be refreshed, replaced or topped up with additional first culture medium during the first time period. In some examples, the first culture media is replaced (e.g. first culture media is removed from the vessel and fresh first culture medium is added to the vessel). For example, the method may include adding one, two, three or four volumes of first culture medium to the vessel during the first time period. For example, at day 5 from day 0, a volume of first culture medium may be added to the vessel. For example, the volume may be equal to the volume of first culture medium added at day 2.

In some examples, the embryoid body is cultured in the first culture medium at a temperature of 37°C and a CO: content of 5% for about up to 5 days (i.e. up to day 7 from day 0).

During the first time period, the pluripotent stem cells at least partially differentiate. For example, the pluripotent stem cells of the embryoid body differentiate to neuroectodermal cells.

At this point, the embryoid body may be referred to as an organoid.

After the first time period the organoid including neuroectodermal cells is cultured in a second culture medium as described herein. In some examples, the second culture medium may be added to the vessel which includes the first culture medium and organoid (e.g. topped up or refreshed with second culture medium). The volume of second culture medium added to the vessel may be equal to the volume of cells and initial culture medium added to vessel. In some examples, volume of second culture medium added to the vessel may be more than the volume of the first culture medium in the vessel. In some examples, volume of second culture medium added to the vessel may be equal to or substantially equal to the volume of the first culture medium in the vessel. In some examples, volume of second culture medium added to the vessel may be equal to the volume of the first culture medium removed from the vessel if replacing the first culture medium with the second culture medium. In some examples, the first culture medium may be replaced with the second culture medium. For example, 190 pl of the second culture medium may be added to the vessel when 190ul of first culture medium is removed (i.e. replaced). The volume of second culture medium added to the vessel may be dependent on the number of cells initially added to the vessel. The volume of second culture medium added to the vessel may be dependent on the working volume of the vessel.

The second culture medium is added to the vessel at day 7. The organoid and neuroectodermal cells thereof are then cultured in the second culture medium for a second time period. In some examples, the second time period starts at day 7 and proceeds to about day 14.

For example, the second time period is up to about 7 days. For example, the second time period starts on day 7 from day O and is up to day 14 from day 0.

The second culture medium may be refreshed, replaced or topped up with additional second culture medium during the second time period. In some examples, the second culture medium is replaced (e.g. second culture media is removed from the vessel and fresh second culture medium is added to the vessel). For example, the method may include adding one, two, three or four volumes of second culture medium to the vessel during the second time period. For example, at day 9 from day 0, a volume of second culture medium may be added to the vessel.

For example, the volume may be equal to the volume of second culture medium added at day 7.

For example, the volume may be less than the volume of second culture medium added at day 7.

During the second time period an extracellular component may be added to the culture or the culture added to an extracellular matrix component as described herein. For example, during the second time period, the organoid may be embedded in an extracellular matrix scaffold such as Matrigel. After addition of the extracellular component the organoid embedded in the extracellular matrix component may be transferred to another (second) vessel.

In some examples, the method comprises embedding the organoid in Matrigel droplet. For example, the Matrigel droplet may have a volume of about 12 to 15 pl. For example, about 12, 13, 14, or 15 pl.

The second vessel may be a vessel suitable for 3D culturing of cells. For example, the second vessel may be suspension culture plate. A suspension culture plate is a type of cell culture plate used to grow cells in suspension, which is a liquid culture where cells are free-floating in a culture medium. As such, the organoid may be cultured in suspension during the second time period.

In some examples, the organoid is cultured in suspension for at least part of the second time period and optionally is maintained in suspension for the rest of the method. In some examples, the organoid is embedded in an extracellular matrix scaffold on day 11 from day 0. In some examples, the organoid is cultured in suspension from day 11 from day 0 onwards. For example, after day 11 the organoid may be cultured or maintained in suspension.

In some examples, the organoid is cultured in suspension with about 1ml of the second culture medium.

In some examples, the organoid is cultured in the second culture medium in the first vessel for about up to 4 days (e.g. up to day 11 from day 0) and cultured in suspension in the second culture media and embedded in Matrigel for about up to 3 days (e.g. from day 11 from day 0 to day 14 from day 0).

In some examples, the organoid is cultured in the second culture medium at a temperature of 37°C and a CO: content of 5% for about up to 7 days (i.e. up to day 14 from day 0).

During the second time period, the neuroectodermal cells of the organoid may further differentiate and at least a portion of the cells may become committed neuroectodermal cells.

Committed neuroectodermal cells refers to cells that have become committed to a particular pathway of differentiation are no longer pluripotent.

After the second time period the organoid may be cultured in a third culture medium as described herein. In some examples, the third culture medium may be added to the vessel or second vessel which includes the second culture medium and organoid. The volume of third culture medium added to the vessel or second vessel may be equal to the volume of cells and initial culture medium added to the vessel or second vessel. In some examples, the volume of third culture medium added to the vessel or second vessel may be more than or less than the volume of the second culture medium in the vessel or second vessel. The volume of third culture medium added to the vessel or second vessel may be dependent on the number of cells initially added to the vessel or second vessel.

The third culture medium is added to the vessel at day 14. The organoid and committed neuroectodermal cells thereof are then cultured in the third culture medium for a third time period. In some examples, the third time period starts at day 14 and proceeds to about day 21.

For example, the third time period is up to about 7 days. For example, the third time period starts on day 14 from day 0 and is up to day 21 from day 0.

The third culture medium may be refreshed, replaced or topped up with additional third culture medium during the third time period. In some examples, the third culture medium is replaced

(e.g. third culture media is removed from the vessel or second vessel and fresh third culture medium is added to the vessel or second vessel). For example, the method may include replacing the third culture media once, twice, three times or more during the third time period. In some examples, the method includes replacing the third culture medium every two days during the third time period.

During the third time period the method may include culturing the organoid with agitation. For example, shaking or rotating the vessel or second vessel and the organoid therein. For example, using an orbital shaker. Agitation may be started at about 16 days from day 0.

Agitation may then be maintained.

After the third time period the method includes maturing the organoid to obtain a hindbrain organoid as described herein. To mature the organoid, a maturation medium as described (fourth culture medium) is added to the vessel or second vessel. The maturation medium may be added at about 21 days from day 0.

The maturation medium may be added to the vessel or second vessel with the third culture medium. For example, the third culture medium may be refreshed with the maturation medium.

That is to say that the third culture medium may not be removed before adding a volume of maturation medium. In some examples, from day 21 from day 0 onwards, the maturation medium is replaced or refreshed. For example, every 2 or 3 days from day 21 from day 0 onwards additional volumes of maturation medium are added to the vessel or second vessel.

In some examples, the method includes maturing the organoid for at least about 10, 20, 30, 40, 50, 60 or more days. In some examples, the organoid is matured for at least 30 days from day 0. In some examples, the organoid is matured for at least 60 days from day 0. In some examples, the organoid is matured for at least 100 days from day 0. In some examples, the organoid is matured for at least 120 days from day 0.

In some examples, organoid is matured in the maturation medium at a temperature of 37°C and a CO: content of 5%.

In some examples, the organoid may be maintained in the maturation medium. In some examples, the organoid may be maintained for 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 14, 16, 18, 20, 22, 24 months or more in the maturation medium.

In some examples, organoid is maintained in the maturation medium at a temperature of 37°C and a CO: content of 5%.

In one aspect the methods described herein provide a healthy hindbrain organoid. For example, a pontine organoid that does not include diseased cells or tissue.

In another aspect, the methods described herein may be used to provide a hindbrain organoid that includes healthy and diseased cells or tissue. In methods of producing a hindbrain organoid comprising healthy and diseased tissue the methods described above include an additional step of inducing a disease state in one or more of the committed neuroectodermal cells.

The step of inducing a diseased state may be carried out after the second time period. In some examples, inducing the disease state in one or more of the committed neuroectodermal cells may be carried out during the second time period. For example, from day 7 from day O and up to day 14 from day 0. In some examples, inducing the disease state is carried out at day 11 from day 0.

In some examples, inducing a diseased state is carried out prior to adding an extracellular component or prior to adding the organoid to an extracellular matrix component (e.g. prior to embedding the organoid in an extracellular matrix component such as Matrigel).

In some examples, inducing a diseased state may include mutating one or more disease associated genes of the committed neuroectodermal cells. In some examples, inducing a diseased state may include providing the committed neuroectodermal cells one or more mutated genes. In some examples, inducing a diseased state may include providing the committed neuroectodermal cells with one or more disease associated proteins. In some examples, inducing a diseased state may include providing the committed neuroectodermal cells with one or more gene editing systems for mutating disease associated genes. In some examples, inducing a diseased state may include providing the committed neuroectodermal cells with one or more interfering nucleic acid molecules.

In some examples, inducing a diseased state may include a combination of mutating one or more diseases associated genes, providing one or more mutated genes, providing one or more disease associated proteins, providing one or gene editing systems for mutating disease associated genes and/or providing one or more interfering nucleic acid molecules.

As used herein “disease associated protein” refers to any protein that may be expressed in the cells of an individual suffering from the associated disease. In some cases, the disease associated protein may be a cause of the pathology of the disease and/or a marker for the disease. Disease associated proteins may be mutant proteins or proteins modified from the wild-type version of the protein (i.e. by deletions, substitutions, truncations, missense mutations, as well as by changes in post-translational modifications and/or localisation of the protein). In some examples, disease associated protein is a mutant protein.

As used herein “disease associated gene” refers to a gene that is expressed in an individual suffering from the associated disease. In some examples, disease associated genes may be mutant genes. In some examples, disease associated genes may be genes that are overexpressed or underexpressed in an expressed in an individual suffering from the associated disease compared to an expressed in an individual not suffering from the associated disease. In some examples, disease associated genes may be genes that are expressed in cells that do not express the protein in an individual not suffering from the associated disease.

In some examples, the disease associated gene is a mutant gene.

In some examples, mutating one or more disease associated genes may include mutagenesis by any known means. For example, modifications may be achieved by any suitable method known to those skilled in the art. Merely by way of example, suitable methods include using gene editing techniques (such as CRISPR), transposon based mutagenesis (see Levitan, Anton et al. “Comparing the utility of in vivo transposon mutagenesis approaches in yeast species to infer gene essentiality.” Current genetics vol. 66,6 (2020): 1117-1134. doi:10.1007/s00294-020- 01096-6), homologous recombination, site-directed mutagenesis (see D. Court et. al., “Genetic

Engineering Using Homologous Recombination”, Annual Review of Genetics, Vol. 36, p. 361 (2002)) or viral over-expression.

Methods for mutating or modifying genes are well known in the art. For example, genes may edited in situ by way of gene editing techniques in order to provide a mutated gene as described herein. Such genome editing and/or mutagenesis technologies are well known in the art.

Particularly, the modification to a nucleic acid sequence is introduced by way of site-directed nuclease (SDN). The SDN may be selected from: meganuclease, zinc finger, transcription activator- like effector nucleases system (TALEN) or Clustered Regularly Interspaced Short

Palindromic Repeats system (CRISPR) system. SDN is also referred to as “genome editing”, or genome editing with engineered nucleases (GEEN). This is a type of genetic engineering in which DNA is inserted, deleted or replaced in the genome of an organism using engineered nucleases that create site-specific double-strand breaks (DSBs) at desired locations in the genome. The induced double-strand breaks are repaired through nonhomologous end-joining (NHEJ)} or homologous recombination (HR), resulting in targeted mutations (‘edits’). Particularly

SDN may comprises techniques such as: Meganucleases, Zinc finger nucleases (ZFNs),

Transcription Activator-Like Effector-based Nucleases (TALEN) (Feng et al. 2013 Cell Res. 23, 1229-1232, Sander & Joung Nat. Biotechnol. 32, 347-355 2014), and the Clustered Regularly

Interspaced Short Palindromic Repeats (CRISPR-Cas) system. Gene editing may also be achieved by SDN-2. SDN-2 is similar to SDN, but also provides a small nucleotide template complementary to the area of the break. The template contains one or more sequence modifications to the genomic DNA which are incorporated to create the modification to the target gene. Preferably, the gene editing system may include a CRISPR-Cas system.

In some examples, mutating may be carried out by a transposon based system such as by using PiggyBac transposon vectors. The term "piggyBac transposon’ refers to a mobile genetic element that is transposed between a vector and a chromosome by a "cut-and-paste" mechanism. During transposition, the PB transposase recognizes transposon-specific Inverted

Terminal Repeats (ITRs) located at both ends of the transposon vector, and effectively transfers and integrates the contents from the original site to the TTAA chromosomal site. The resulting transformed cell or cell group is a stable transformant. In addition to transposable activity, ITRs can also be used as enhancers to stimulate expression of endogenous genes near the insertion site. As such, in some examples, mutating comprises introducing one or more mutated genes in piggyBac transposon vectors into the pluripotent cells, embryoid body, neuroectodermal cells,

committed neuroectodermal cells and/or hindbrain organoids. The mutated gene may be inserted into a specific insertion site that leads to expression of the mutated gene.

In some examples, mutations may be introduced using a Cre-loxP system. For example, an inducible Cre-loxP system (CreERT). Cre-loxP system is a widely used technology for mammalian gene editing. This system has advantages which are simple manipulation and no requirement for additional factors for efficient recombination. Concerning the mechanism of Cre- loxP system, a single Cre recombinase recognizes two directly repeated loxP site, then the Cre excises the loxP flanked (floxed) DNA, thus creating two types of DNA with circular, excised and inactivated gene. While the Cre-loxP system is predominantly used in genetic excision, it also induces the inversion and translocation of DNA between two loxP sites depending on the orientation and location of loxP sites (see Kim, Hyeonhui et al. “Mouse Cre-LoxP system: general principles to determine tissue-specific roles of target genes.” Laboratory animal research vol. 34,4 (2018): 147-159. doi:10.58625/lar.2018.34.4. 147).

Interfering nucleic acid molecules refers to any nucleic acid molecule capable of inhibiting or down regulating gene expression. For example, by mediating RNA interference “RNAI” or gene silencing in a sequence-specific manner; see for example WO2009024599A1.

Providing disease associated genes, disease associated proteins, gene editing systems, and/or interfering nucleic acid molecules may be carried using methods well known in the art. For example, the disease associated genes, disease associated proteins, gene editing systems, and interfering nucleic acid molecules may be encoded by one or more nucleic acid vectors that can be introduced into one or more cells of the organoid.

By way of example only, the vector may be a plasmid, a cosmid, or a viral vector, such as a retroviral vector or a lentiviral vector. Adenovirus, adeno-associated virus, vaccinia virus, canary poxvirus, herpes virus, minicircle vectors and naked (synthetic) DNA/RNA may also be used (for details on minicircle vectors, see for example non-viral Sleeping Beauty transposition from minicircle vectors as published by R Monjezi et al., Leukemia 2017).

As used herein, the term “vector” refers to a nucleic acid molecule capable of transporting another nucleic acid molecule to which it has been operably linked. The vector can be capable of autonomous replication or it can integrate into a host DNA. The vector may include restriction enzyme sites for insertion of recombinant DNA and may include one or more selectable markers or suicide genes. The vector can be a nucleic acid molecule in the form of a plasmid, a bacteriophage or a cosmid. Preferably the vector is suitable for expression in a cell (i.e. the vector is an “expression vector”). Preferably, the vector is suitable for expression in neuroectodermal cells and/or committed neuroectodermal cells as described herein.

The vector may comprise regulatory sequences. "Regulatory sequences" as used herein, refers to, DNA or RNA elements that are capable of controlling gene expression. Examples of expression control sequences include promoters, enhancers, silencers, TATA- boxes, internal ribosomal entry sites (IRES), attachment sites for transcription factors, transcriptional terminators, polyadenylation sites etc. Optionally, the vector includes one or more regulatory sequences operatively linked to the nucleic acid sequence to be expressed. Regulatory sequences include those which direct constitutive expression, as well as tissue-specific regulatory and/or inducible sequences.

In some examples, the vector comprises the nucleic acid sequence of interest operably linked to a promoter. "Promoter", as used herein, refers to the nucleotide sequences in DNA to which

RNA polymerase binds to start transcription. The promoter may be inducible or constitutively expressed. Alternatively, the promoter is under the control of a repressor or stimulatory protein.

The promoter may be one that is not naturally found in the host cell (e.g. it may be an exogenous promoter). The skilled person in the art is well aware of appropriate promoters for use in the expression of target proteins, wherein the selected promoter will depend on the host cell.

The vector may comprise a transcriptional terminator. “Transcriptional terminator” as used herein, refers to a DNA element, which terminates the function of RNA polymerases responsible for transcribing DNA into RNA. Preferred transcriptional terminators are characterized by a run of T residues preceded by a GC rich dyad symmetrical region.

The vector may comprise a translational control element. “Translational control element”, as used herein, refers to DNA or RNA elements that control the translation of MRNA. Preferred translational control elements are ribosome binding sites. Preferably, the translational control element is from a homologous system as the promoter, for example, a promoter and its associated ribozyme binding site. Preferred ribosome binding sites are known, and will depend on the chosen host cell.

The vector may comprise reporter sequences. Reporter sequences include nucleic acid sequences encoding reporter proteins such as fluorescent proteins such as GFP. Reporter proteins may be used to analyse expression and/or levels of expression of genes encodes by a vector either expressed from the vector or after insertion into a cell's genome (i.e. by transposase systems or gene editing systems).

Preferably the vector comprises those genetic elements which are necessary for expression of the disease associated genes and/or disease associated proteins described herein by a neuroectodermal cell as described herein. The elements required for transcription and translation in the host cell include a promoter, a coding region for the protein(s) of interest, and a transcriptional terminator.

A person of skill in the art will be well aware of the molecular techniques available for the preparation of (expression) vectors and how the (expression) vectors may be transduced or transfected into an appropriate cell (thereby indicating a diseased state as described further below). The (expression) vectors described herein can be introduced into cells by conventional techniques such as transformation, transfection or transduction. “Transformation”, “transfection” and “transduction” refer generally to techniques for introducing foreign (exogenous) nucleic acid molecules into a host cell, and therefore encompass methods such as electroporation, microinjection, gene gun delivery, transduction with retroviral, lentiviral or adeno-associated vectors, lipofection, superfection etc. The specific method used typically depends on both the type of vector and the cell. Appropriate methods for introducing nucleic acid molecules and vectors into cells such as human cells are well known in the art; see for example Sambrook et al (1989) Molecular Cloning, A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring

Harbor, N.Y; Ausubel et al (1987) Current Protocols in Molecular Biology, John Wiley and Sons,

Inc., NY; Cohen et al (1972) Proc. Natl. Acad. Sci. USA 69, 2110; Luchansky et al (1988) Mol.

Microbiol. 2, 637-646.

In some examples, the nucleic acid vectors are introduced into the committed neuroectodermal cells using electroporation.

In some examples, after introducing the nucleic acid vector or vectors the method comprises incubating the organoid for a recovery time period. For example, after introducing the nucleic acid vector or vectors the method comprises incubating the organoid for at least 1, 2, or 3, hours before continuing with patterning of the organoid. For example, after introducing the nucleic acid vector or vectors the method comprises incubating the organoid for at least 2 hours prior to adding an extracellular component or prior to adding the organoid to an extracellular matrix component (e.g. prior to embedding the organoid in an extracellular matrix component such as Matrigel). In some examples, incubating the organoid during the recovery time period comprises incubating at 37°C with 5% CO:.

In some examples, the pluripotent stem cells may include one or more inducible mutated genes.

In such examples, inducing a disease state may include providing an inducer which leads to expression of the mutated gene or genes. For example, the mutated gene or genes may be introduced in the pluripotent stem cells using a Cre-LoxP system and inducing a disease state may include providing the committed neuroectodermal cells with a Cre recombinase to induce the mutation or expression thereof. Cre recombinase may be introduced using nucleic acid vectors as described herein.

As mentioned above induction of a diseased state may be carried out at a specific timepoint.

For example, at day 11 from day 0. Without being bound by theory, induction of a disease state at day 11 from day 0 may provide for more efficient induction of a diseased state.

In methods that produce hindbrain organoids that include healthy and diseased cells or tissue, some time periods described herein may be measured from the timepoint at which the induction of a disease state is carried out (e.g. from day 11 from day 0). This timepoint may be referred to as “TO”. Further timepoints described in reference to TO may be sequentially named “T” timepoints. For example, T1, T2, T3 etc.

In some examples, maturing hindbrain organoids that include healthy and diseased cells or tissue is carried out for at least 30 days from TO. In some examples, maturing hindbrain organoids that include healthy and diseased cells or tissue is carried out for at least 60 days from TO (referred to herein as T1). In some examples, maturing hindbrain organoids that include healthy and diseased cells or tissue is carried out for at least 120 days from TO (referred to herein as T2). In some examples, the hindbrain organoids that include healthy and diseased cells or tissue may be maintained for at least about 6 months (referred to herein as Tx).

In some examples, the methods described herein further comprise co-culturing the embryoid body, neuroectodermal cells, committed neuroectodermal cells and/or hindbrain organoids (with or without diseased cells or tissue) with other cell types that do not occur during the methods described herein. As used herein the term “co-culture” or “co-culturing” refers to growing or culturing two or more (e.g., three or more) distinct cell types, tissues or organoids (e.g. culturing a hindbrain organoid as described herein with another organoid of the same type of a different organoid together or culturing a hindbrain organized as described herein with at least one cell type not present in the organoid) within a single recipient or environment (e.g., a single cell culture vessel).

In some examples, the methods include co-culturing the embryoid body, neuroectodermal cells, committed neuroectodermal cells and/or hindbrain organoids {with or without diseased cells or tissue) with one or more immunological components. Immunological components is used herein to refer to any part of the immune system, including, immune cells, proteins and molecules derived from immune cells, proteins and molecules that interact with immune cells, and proteins and molecules produced by immune. Creating an immune microenvironment may allow for investigating trafficking behaviour, including, homing, migration and persistence of T- cells in a combined healthy and diseased hindbrain organoid model. Trafficking behaviour refers to the engraftment of T-cells in between healthy and diseased tissue. With healthy tissue acting as a barrier function for T-cells to overcome in order to disseminate and physically touch the tumour tissue ( also referred to as homing). Subsequent recognition and killing of tumour tissue and/or migration to adjacent tumour tissue and cells after killing (referred to as migration). Due to repeated contact and recognition activities, T-cells may persist in their activity of Killing tumour, being stimulated to enhance proliferative and/or migratory behaviour (referred to as persistence).

In some examples, the embryoid body, neuroectodermal cells, committed neuroectodermal cells and/or hindbrain organoids (with or without diseased cells or tissue) are co-cultured with endothelial cells. Endothelial cells form a single cell layer that lines all blood vessels and regulates exchanges between the bloodstream and the surrounding tissues. Signals from endothelial cells organize the growth and development of connective tissue cells that form the surrounding layers of the blood-vessel wall. The vascularization of organoids, may be a critical aspect of creating physiologically relevant miniature organs. As such, by co-culturing with endothelial cells, the hindbrain organoids (with or without diseased cells or tissue) as described herein may comprise vasculature structures which are mimetic of the neurovascular system (for example, see Rizzuti, Mafalda, et al. "Shaping the Neurovascular Unit Exploiting Human Brain

Organoids." Molecular Neurobiology (2024): 1-186).

In some examples, the embryoid body, neuroectodermal cells, committed neuroectodermal cells and/or hindbrain organoids (with or without diseased cells or tissue) are co-cultured with microglia cells. Microglia account for approximately 10% of cells and are the most abundant mononuclear phagocytes in the central nervous system (CNS). During development, microglia help shape neural circuits by modulating the strength of synaptic transmissions and sculpting neuronal synapses. During CNS injury, microglia are responsible for phagocytosis and elimination of microbes, dead cells, and protein aggregates, as well as other particulate and soluble antigens that may endanger the CNS. Moreover, microglia secrete many soluble factors, such as chemoattractants, cytokines, and neurotropic factors that contribute to various aspects of immune responses and tissue repair in the CNS (for example see Colonna, Marco, and Oleg Butovsky. "Microglia function in the central nervous system during health and neurodegeneration.” Annual review of immunology 35 (2017): 441-468.). Therefore, by co- culturing with microglia cells the hindbrain organoids (with or without diseased cells or tissue) described herein may be provided with an immune microenvironment. In general microenvironment refers to cells, molecules and structures that surround and support other cells and tissue.

In some examples, the embryoid body, neuroectodermal cells, committed neuroectodermal cells and/or hindbrain organoids (with or without diseased cells or tissue) are co-cultured with myeloid progenitor cells. Myeloid progenitor cells are a heterogeneous population of cells that can give rise to all major myelo-erythroid cell lineages in the brain. These cells are strategically located in the brain's parenchyma and meninges, near blood vessels, to perform surveillance and homeostatic tasks (for example, see Herz, Jasmin, et al. "Myeloid cells in the central nervous system." Immunity 46.6 (2017): 943-956.).

Methods of co-culturing hindbrain organoids (with or without diseased cells or tissue) as described herein may include co-culturing with terminally differentiated cells. For example, endothelial cells, microglia and/or myeloid progenitor cells can be obtained through differentiation from human pluripotent stem cells, or by isolating the cells from the tissue of interest. Endothelial cells may be isolated from the umbilical vein (human umbilical vein endothelial cells or HUVEC), while microglia may be obtained by isolating the cells directly from the brain (primary microglia). Co-culturing these differentiated cells with organoids may provide a vascularized organoid (when using endothelial cells) or immunized (having and immune microenvironment) organoids (when using microglia).

Another method of co-culturing may include the use of genetic engineering. Genetic engineering can be used to induce the working vessels and immune cells within the hindbrain organoids (with or without diseased cells or tissue). To do this, pluripotent stem cells as described herein may be mixed with pluripotent stem cells that have been genetically engineered to over-express a specific transcription factor when an inducer is applied. For example, cells carrying ETV2 will differentiate into endothelial cells when th inducer is applied, while cells carrying PU.1 will differentiate into microglia. The wild-type cells may then be mixed with the genetically modified cells to form embryoid bodies as described herein. The embryoid bodies with cells that overexpress ETV2 develop into vascularized organoids while the embryoid bodies with cells that overexpress PU.1 develop into immunized organoids.

Another method of co-culturing may include fusion of hindbrain organoids (with or without diseased cells or tissue) as described herein. For example, by fusion with endothelial cell spheroids.

For example of such methods see Cakir, Bilal, and In-Hyun Park. "Getting the right cells." Elife 11 (2022): e80373.

Other methods of co-culturing brain organoids that may be used in conjunction with the methods described herein include those described in Park, Dong Shin, et al. "iPS-cell-derived microglia promote brain organoid maturation via cholesterol transfer." Nature 623.7986 (2023): 397-405 and WO2020204827A1 both of which are incorporated herein in their entirety.

In some examples, the embryoid body, neuroectodermal cells, committed neuroectodermal cells and/or hindbrain organoids (with or without diseased cells or tissue) described herein may be co-cultured with induced pluripotent stem cell derived T-cells. Induced pluripotent stem cell derived T-cells may be produced using any known methods such as those described in Gutbier,

Simon, et al. "Large-scale production of human iPSC-derived macrophages for drug screening."

International journal of molecular sciences 21.13 (2020): 4808, Netsrithong, Ratchapong, Laura

Garcia-Perez, and Maria Themeli. "Engineered T cells from induced pluripotent stem cells: from research towards clinical implementation." Frontiers in Immunology 14 (2024): 1325209 and

Iriguchi, Shoichi, et al. "A clinically applicable and scalable method to regenerate T-cells from iPSCs for off-the-shelf T-cell immunotherapy." Nature communications 12.1 (2021): 430.

Co-culturing with induced pluripotent stem cell derived T-cells may be carried out using the methods described above.

In some examples, the pluripotent stem cell derived T-cells may derived from a patient. Use of patient derived T cells may allow the assessment of clinical applicability of T cells and other therapeutics in a personalized context.

Organoid structure

Healthy Tissue

The hindbrain organoids produced by the methods described herein may comprise one or more cells and/or tissue which is specific to the hindbrain. The cells that are comprised with the organoids is dynamic and the markers expressed may change over time.

For example, during the second time period onwards (e.g. from 7 days from day 0), the organoids comprising neurcectodermal cells and/or committed neuroectodermal cells may express one or more of HOXB1, GBX2, MEIS1, MEIS2 and/or MEIS3.

For the purposes of the present disclosure, references to “positive” (or “+”) cells, or to cells “expressing” a protein or other marker, should be interpreted as encompassing both cases in which an expressed protein is detectable in respect of a cell, and cases in which increased expression of a recited gene (or gene encoding a recited protein) is detectable in respect of a cell.

In the case of a detectable protein, this may be confirmed by a suitable approach such as antibody labelling. In the case of increased expression of a gene, this may be confirmed by the presence of elevated levels of MRNA, which can be detected by any appropriate RNA analysis approach. Elevation of mRNA levels can be determined by normalization with respect to an appropriate control, such as a housekeeping gene.

References to cells being “negative” (or “-*) should be construed with the same considerations in mind.

HOXB1 belongs to the homeobox family of genes. The homeobox genes encode a highly conserved family of transcription factors that play an important role in morphogenesis in all multicellular organisms. Mammals possess four similar homeobox gene clusters, HOXA, HOXB,

HOXC and HOXD, located on different chromosomes, consisting of 9 to 11 genes arranged in tandem. HOXB1 encodes a sequence-specific transcription factor which is part of a developmental regulatory system that provides cells with specific positional identities on the anterior-posterior axis. Expression of HOXB1 is a marker for pontine precursor cells (see

Philippidou, P. & Dasen, J. S. Hox Genes: Choreographers in Neural Development, Architects of Circuit Organization. Neuron 80, 12-34 (2013)).

GBX2 encodes gastrulation And Brain-Specific Homeobox Protein 2 which enables sequence- specific double-stranded DNA binding activity. GBX2 is predicted to be involved in regulation of nervous system development and regulation of transcription by RNA polymerase Il, and predicted to act upstream of or within several processes, including branching involved in blood vessel morphogenesis; nervous system development; and neural crest cell migration. GBX2 is an early hindbrain marker.

In some examples, GBX2 may be expressed from at least day 2 from day 0. For example, from day 2, 3, 4, 5, 6 or 7 from day 0 onwards.

MEIS1, MEIS2 and MEIS3 encode homeobox proteins belonging to the TALE (‘three amino acid loop extension’) family of homeodomain-containing proteins. The MEIS genes are considered hindbrain specific markers. In some examples, the expression of ME/S genes may increase from day 7 onwards.

In some examples, the neuroectodermal cells and/or committed neuroectodermal cells of the organoids comprise cells that do not express one or more of OTX2 and/or spinal cord-specific

CDX genes about 7 days from day O onwards.

OTX2 encodes a member of the bicoid subfamily of homeodomain-containing transcription factors. The encoded protein acts as a transcription factor and plays a role in brain, craniofacial, and sensory organ development. OTX2 expression is a maker for the midbrain.

CDX genes refers to the genes CDX1, CDX2 and CDX4. These genes are members of the caudal-related homeobox transcription factor gene family and are involved in transcriptional regulation. In particular, CDX1 and CDX4 are involved in spinal chord development and may be referred to as “spinal chord specific CDX genes”. As such, CDX1 and/or CDX4 may not be expressed. In some examples, the expression of CDX1 and/or CDX4 in neuroectodermal cells and/or committed neuroectodermal cells of the organoids may decrease from day 7 from day 0 onwards.

In some examples, the hindbrain organoids described herein may include cells expressing one or more of TPH2, GFAP, AQP1, AQP4, OLIG1, PDGFRA, OLIG2, NRG3, NRXN1, GRIA2,

RBFOX1, MAP2, ERB4, PLCG2, VIM, SOX2, SOX10, SOX9, SOX4, CLU, BCAN, NCKAP5,

PPP2R2B, NTN1, RMST, and/or SLIT2.

RBFOX1, MAP2, ERB4, PLCG2, VIM, SOX2, SOX10, SOX9, SOX4, CLU, BCAN, NCKAPS,

PPP2R2B, NTN1, RMST, and/or SLIT2 from at least about 30 to at least about 60 days from day 0. For example, from about 30 days from day 0. For example, about 60 days from day 0. In some examples, the hindbrain arganoids described herein may include cells expressing one or more of TPH2, GFAP, AQP1, AQP4, OLIG1, PDGFRA, OLIG2, NRG3, NRXN1, GRIA2,

RBFOX1, MAP2, ERB4, PLCG2, VIM, SOX2, SOX10, SOX9, SOX4, CLU, BCAN, NCKAPS,

PPP2R2B, NTN1, RMST, and/or SL/T2 from about 100 days from day 0. In some examples, the hindbrain organoids described herein may include cells expressing one or more of TPH2,

GFAP, AQP1, AQP4, OLIG1, PDGFRA, OLIG2, NRG3, NRXN1, GRIA2, RBFOX1, MAP2,

ERB4, PLCG2, VIM, SOX2, SOX10, SOX9, SOX4, CLU, BCAN, NCKAP5, PPP2R2B, NTN1,

RMST, and/or SLIT2 from about 120 days from day 0.

TPH2 encodes tryptophan hydroxylase 2 and is a marker for pons-specific serotonergic neurons.

GFAP encodes glial fibrillary acidic protein which is a marker for committed astrocytes.

AQP4 encodes aquaporin 4 which is a marker for committed astrocytes.

AQP1 encodes aquaporin 1 which is a marker for astrocytes and its progenitors exclusively arising in human hindbrain tissue.

PDGFRA, OLIG2 and OLIG1 are markers for oligodendrocyte precursor cells.

NRG3, NRXN1, GRIA2, RBFOX1, MAP2, ERB4 and PLCG2 are markers for committed neurons.

VIM, SOX2, CLU, BCAN, NCKAP5, PPP2R2B, and GFAP are markers for glioblasts.

NTN1, RMST, and SLIT2 are markers for axon-guiding neuroepithelium cells.

SOX170 and PDGFRA are markers for oligodendrocyte precursor cells.

Additional markers that may be expressed include:

OLIG1, HOPX, and SPARCL 1 which are markers of glial precursor cells (OAPCs) and astrocyte precursor cells (APCs);

PAX6, and HMGA2 which are markers of early neural progenitor cells (NPCs);

GATA2, and GATA3 which are markers of early serotonin (5-HT) neurons;

DCX, and STMN2 which are markers of late neural progenitor cells;

HTRA2, and SLC17A8 which are markers of late serotonin (5-HT) neurons; and/or

DLL1, DLL3, and PPP1R14B which are markers of pre-oligodendrocyte precursor cells.

The hindbrain organoids described herein may include one or more cell types selected from astrocytes; oligodendrocytes; glioblasts; radial glial cells; axon-guiding neuroepithelium cells; stromal cells; and/or neurons.

In some examples, the axon-guiding neuroepithelium cells include choroid plexus cells and/or ependymal cells. The choroid plexus has a lining of specialized epithelial tissue known as ependyma. Ependymal cells are glial cells with a ciliated simple columnar form that line the ventricles and central canal of the spinal cord. Apical surfaces have a covering of hair-like projections known as cilia (which circulate CSF) and microvilli (which help in CSF absorption).

Microvilli perform this function via their brush border, which significantly increases the surface area of the choroid plexus, permitting increased CSF absorption. Ependymal cells are essential in the production of CSF.

In some examples, neurons include one or more of hindbrain-specific serotonergic neurons, pons-specific serotonergic neurons, excitatory neurons, inhibitory neurons, and/or dopaminergic neurons.

Excitatory neuron refers to a neuron that increases the frequency of action potentials in postsynaptic neurons when it releases a neurotransmitter in the synaptic cleft as a presynaptic neuron.

Inhibitory neuron refers to a neuron that reduces the probability of an action potential in a post- synaptic neuron when it releases a neurotransmitter, e.g., gamma-aminobutyric acid (GABA), at the synaptic cleft as a pre-synaptic neuron.

Serotonergic neuron refers to a neuron that secretes serotonin or that is capable of recapturing serotonin (i.e., serotonin transporters expressed on their cell surfaces).

Astrocyte generally refer to characteristic star-shaped glial cells in the brain and spinal cord, that is characterized by one or more of: star shape; expression of markers like glial fibrillary acidic protein (GFAP), aldehyde dehydrogenase 1 family member LI (ALDH1L1), excitatory amino acid transporter 1 /glutamate aspartate transporter (EAAT1/GLAST), glutamine synthetase, S100 beta, or excitatory amino acid transporter 1 /glutamate transporter 1 (EAAT2/GLT-1); participation of blood-brain barrier together with endothelial cells; transmitter uptake and release; regulation of ionic concentration in extracellular space; reaction to neuronal injury and participation in nervous system repair; and metabolic support of surrounding neurons.

Oligodendrocytes are a subtype of glial cells in the central nervous system that originate from oligodendrocyte progenitor cells (OPCs). OPCs account for about 5% of cells in the central nervous system. Oligodendrocytes help support and insulate axons by producing myelin. Myelin sheaths in the central nervous system are made of extended oligodendrocyte plasma membranes. While mature oligodendrocytes cannot self-renew, OPCs can repopulate oligodendrocytes following injury to the central nervous system in healthy individuals. Myelin

Basic Protein (MBP) and Myelin Oligodendrocyte Glycoprotein (MOG) are expressed in terminal differentiation of OPCs to oligodendrocytes. They both are oligodendrocyte- specific genes and may be used as markers of mature oligodendrocyte formation.

Radial glial cells, or radial glial progenitor cells (RGPs), are bipolar-shaped progenitor cells that are responsible for producing all of the neurons in the cerebral cortex. RGPs also produce certain lineages of glia, including astrocytes and oligodendrocytes.

Glioblasts (also referred to as glial precursor cells) are progenitor cells and cells in a state which are committed to the gliogenic lineage, meaning glioblasts able to give rise to astrocytes and oligodendrocytes but not neurons. Glioblasts are proliferative cells which are in the process of differentiation, wherein their fate is a hybrid between astrocytes and oligodendrocytes and/or their progenitors.

Stromal cells refers to precursors of skeletal tissue components such as bone, cartilage, hematopoietic supporting stroma, and adipocytes.

In some examples, the hindbrain organoids described herein may have a size larger than other adult-tissue-derived and/or epithelial organoids. For example, the hindbrain organoids described herein may have one or more dimensions in the millimetre range. For example, the organoids may have an average diameter of at least 1 mm during patterning and while maturing.

Diseased Tissue

The hindbrain organoids described herein may include healthy and diseased tissue. As such, hindbrain organoids comprising diseased tissue described herein may include one or more of the cells expressing the markers described above and include cell types as described above.

In addition, in hindbrain organoids comprising diseased tissue as described herein may include additional cell types or markers which are associated with a diseases or diseased state. For example, in the hindbrain organoids comprising diseased tissue may include detectable levels of disease associated proteins or genes which are associated with a specific disease or disease state. In some examples, the hindbrain organoids comprising diseased tissue may have physical structures that differ from a healthy hindbrain organoid as a result of the disease or diseased state of at least a portion of the cells of the hindbrain organoid.

For example, in some examples, the disease is a cancer such as diffuse midline glioma (DMG) and the hindbrain organoids comprise DMG tumour tissue.

In some examples, the hindbrain organoids comprise DMG tissue made up of cancer cells. In some examples, the cancer cells include one or more of astrocyte like-cells, mesenchymal like- cells, oligodendrocyte-like cells, neural stem cell-like cell, oligodendrocyte precursor like-cells, and/or cycling cells.

In some examples, the diseased tissue (e.g. DMG tissue) comprises cancer cells expressing one or more of HOXAS, HOXBS, SOX9, OLIG2, PDGFRA, SOX10, OLIG1, GFAP, CRAPB1,

VIM, TIMP1, TOP2A, MKI&7, AQP1 and/or AQP4.

In some examples, the diseased tissue (e.g. DMG tissue) comprises oligodendrocyte precursor -like cells expressing CRAPB1. Without being bound by theory CRAPB1 has been found to be expressed to a greater level in pontine, as compared to thalamic and cortical DMG tumours. In some examples, the oligodendrocyte precursor -like cells express or further express OL/G1. In some examples, the oligodendrocyte precursor -like cells express or further express OL/G2.

In some examples, the diseased tissue (e.g. DMG tissue) comprises astrocyte-like cells expressing AQP1. AQP1 expression has been shown to be exclusive to astrocytes arising in the human brainstem.

In addition, AQP1 has been observed only in tumours found in the pons and not those arising from the cortex and thalamic region.

In some examples, the diseased tissue (e.g. DMG tissue) comprises astrocyte-like cells expressing AQP4. AQP4 has been identified as a canonical AC-like marker present in DMG tumours at all locations.

In some examples, the diseased tissue (e.g. DMG tissue) comprises neural stem cell -like cells expressing STMN2. Without being bound by theory STMN2 has been found to be expressed to a greater level in hindbrain, as compared to thalamic and cortical DMG tumours.

In some examples, the diseased tissue (e.g. DMG tissue) comprises cycling cells expressing

TOP2A and/or MKI67.

In some examples, the diseased tissue (e.g. DMG tissue) comprises mesenchymal like-cells expressing T/IMP1 and/or VIM.

It will be understood that the presence of cell types and markers may differ depending on the disease induced in the hindbrain organoids as described herein. Detection of specific disease markers and cell types associated with a specific disease will be known by those skilled in the art and may be detected or determined using well known methods such as the analysis methods described below in relation to uses of the hindbrain organoids described herein.

Diseases

As described above, the methods described herein may be used to produce hindbrain organoids comprising diseased tissue. While the inventors have exemplified the production of hindbrain organoids comprising DMG tissue it will be understood that the methods described herein may be applied to numerous diseases and conditions which may be induced in the hindbrain organoids using the methods described herein.

In some examples, the diseased tissue comprises cancer tissue, neurodegenerative tissue and/or malformed tissue.

Neurodegenerative tissue refers to tissue that may produced due to Neurodegenerative disease such as Alzheimer's disease and other memory disorders, ataxia, Huntington's disease,

Parkinson's disease, motor neuron disease, multiple system atrophy, and progressive supranuclear palsy. Tissue effected by such disease may have a specific phenotype or structure that may be detected.

Malformed tissue refers to tissue that may be formed due to a brain malformation (such as due to damage or abnormal development of the brain) and may include tissue having a structure and/or cell types that differ from the expected tissue.

In some examples, the disease and diseased tissue is DMG. Diffuse intrinsic pontine glioma (DIPG; recently re-classified as diffuse midline glioma, H3K27M mutant (Louis et al. (2018) Acta

Neuropathol. 131, 803-820) is the most common high grade glioma of childhood and the leading cause of paediatric brain tumour-related death, with a median survival of only 9 months and a 5- year survival of less than 1% (Donaldson et al. (2006) J. Clin. Oncol. 24(8), 1266-1272). DMG tends to not only infiltrate the brainstem, where it originates, but also the forebrain, with a particular propensity for spread to the SVZ, which occurs in 65% of cases (Caretti et al. (2014)

Acta Neuropathol. 128, 605-607). The World Health Organization's (WHO) 5th Classification of

Central Nervous System (CNS) Tumours, designates DMG as “diffuse midline glioma, H3 K27- altered” representing the majority of DIPGs. This classification encompasses molecular subtypes categorized according to alterations to lysine 27 in histone H3 (H3 K27-altered), as well as patients harbouring wildtype H3 and concomitant overexpression of the EZH inhibitory protein (EZHIP).

DMG cases harbouring wildtype H3 are seen in approximately 10-15% of cases, with a median

OS of 15 months, similar to H3.1K27M DMG. Characterized by the overexpression of the

CXorf67 gene which encodes EZHIP Histone 3 lysine 27 to methionine (H3K27M) mutations occur in both H3.1 and H3.3 histone variants and are mutually exclusive. H3.1K27M is identified in 12-19% of DMG cases, with a median OS rate of 15 months, while H3.3K27M is identified in 65% of cases, with a median OS of 9 months.

Mutations that have been associated with DMG are shown in Table 1.

Cellular tumour antigen p53 (TP53) is the second most recurring lesion in H3.3K27M DMG (60- 80%). TP53 mutations are also seen in H3.1K27M and EZHIP DMG, however, considerably less frequently (13.3% and 11.1% respectively). Activation of PDGFRA accelerates DMG formation in mice, with recurring mutations seen in 14.4% of DMG patients and gene amplification in 30% of DMG, primarily H3.3K27M tumours. Activin receptor type | (ACVR1) is mutated in approximately 32% of all DMG 87% of H3.1K27M and 72% (13/18 cases) of EZHIP.

Mutations in the components of the Phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K) signalling axis are recognized drivers of gliomagenesis in DMG. MYC alterations are common in human cancers including DMG, reported in 20% of the H3.3K27M subtypes. G1/S-specific cyclin-D2 (CCND2) functions as a regulator of Cyclindependent kinase 4 and 6 (CDK4/CDK6) which contributes to the temporal coordination of the cell cycle and typically altered in

H3.3K27M DMG.

See Findlay, Izac J., et al. "Pharmaco-proteogenomic profiling of paediatric diffuse midline glioma to inform future treatment strategies.” Oncogene 41.4 (2022): 461-475 which is expressly incorporated herein by reference in its entirety, for details of DMG and associated mutations.

In some examples, the methods include inducing DMG in a hindbrain organoid described herein. This may be achieved by providing pluripotent stems, embryoid bodies, neuroectodermal cells and/or committed neuroectodermal cells with a mutant H3 histone comprising lysine 27 substituted with a methionine (i.e. K27M - H3K27M mutant). As described above, this may be provided by introducing a H3K27M mutant protein, nucleic acid vectors encoding and suitable for expression of a H3K27M mutant protein and/or nucleic acid vectors encoding a gene editing system or parts thereof to mutate the endogenous H3 histone genes.

In some examples, the pluripotent cells may be provided with an inducible (e.g. Cre inducible

H3K27M mutant protein) prior to the methods described herein and the methods include inducing expression of the H3K27M mutant protein.

In some examples, the H3K27M mutant is a H3.3K27M mutant protein. In some examples, the

H3K27M mutant gene is encoded in a piggyBac transposon vector such as a PBCAG-eGFP plasmid (see Chen, Fuyi, and Joseph LoTurco. "A method for stable transgenesis of radial glia lineage in rat neocortex by piggyBac mediated transposition.” Journal of neuroscience methods 207.2 (2012): 172-180.).

In some examples, the disease associated genes or proteins includes a H3K27M mutant and at least one additional DMG associated protein or gene selected from Table 1.

In some examples, the disease associated genes or proteins include a mutant PDGFRA gene or protein. For example, a gene encoding a PDGFRA-D842V protein or PDGFRA-D842V protein (i.e. a PDGFRA gene encoding a protein or protein comprising a substitution of D842 with a valine). In some examples, the PDGFRA mutant protein may be selected from any one or more of a PDGFRA protein (or gene encoding) comprising one or more mutations selected from Y288C and C235Y. In some examples, the PDGFRA-D842V mutant gene is encoded in a piggyBac transposon vector such as a PBCAG-eGFP plasmid.

In some examples, the disease associated genes or proteins include a DNp53 gene or protein.

In some examples, the DNp53 mutant gene is encoded in a piggyBac transposon vector such as a PBCAG-eGFP plasmid. DNp53 refers to dominant negative Tp53 (DNp53) form of p53 and provides a means of mimicking missing genetic transactivation, interfering in protein-complex formation with p21, MDM2, and PIG3 as well as supressing wild-type p53 induced cell-cycle arrest activity.

In some examples, the disease associated genes or proteins include a mutant p53 gene or protein. In some examples, the p53 mutant gene is encoded in a piggyBac transposon vector such as a PBCAG-eGFP plasmid. For example, a p53 mutant gene or protein thereof may be selected from any one or more of a p53 gene or protein thereof comprising one or more mutations selected from G245S, R175H, R248Q, R248W, R273C, R273H, S241F, and V157.

In some examples, the disease associated genes include a mutant Kinase Insert Domain

Receptor gene (KDR) or protein thereof. In some examples, the mutant KDR gene or protein thereof comprises the mutation S1154P. In some examples, the KDR mutant gene is encoded in a piggyBac transposon vector such as a PBCAG-eGFP plasmid.

In some examples, the disease associated genes include a mutant KIT or protein thereof. In some examples, the mutant KIT gene or protein thereof comprises the mutation T96P. In some examples, the KIT mutant gene is encoded in a piggyBac transposon vector such as a PBCAG- eGFP plasmid.

In some examples, the disease associated genes include a mutant TOP3A or protein thereof. In some examples, the mutant TOP3A gene or protein thereof comprises the mutation C633Y. In some examples, the TOP3A mutant gene is encoded in a piggyBac transposon vector such as a PBCAG-eGFP plasmid.

In some examples, the disease associated genes include a mutant NTRK1/2/3 or protein thereof. In some examples, the NTRK1/2/3 mutant gene is encoded in a piggyBac transposon vector such as a PBCAG-eGFP plasmid.

In some examples, the disease associated genes include a mutant FGFR1 or protein thereof. In some examples, the mutant FGFR1 gene or protein thereof comprises one or more mutations selected from K697E, N98S, N546K, and KB56E. In some examples, the FGFR1 mutant gene is encoded in a piggyBac transposon vector such as a PBCAG-eGFP plasmid.

In some examples, the disease associated genes include a mutant ACVR1 or protein thereof. In some examples, the mutant ACVR1 gene or protein thereof comprises one or more mutations selected from R206H, G328V, G328W. In some examples, the ACVR1 mutant gene is encoded in a piggyBac transposon vector such as a PBCAG-eGFP plasmid.

In some examples, the disease associated genes include a mutant PPM1D or protein thereof. In some examples, the mutant PPM1D gene or protein thereof comprises one or more mutations selected from W427 deletion, E525X, Q404X, E405X, and 428 frame shift. In some examples,

the W427* E525X, Q404X, E405X, 428 fs mutant gene is encoded in a piggyBac transposon vector such as a PBCAG-eGFP plasmid.

Other mutated genes and proteins thereof that may be included when inducing a DMG diseased state include PTEN, PIK3R1, RB1, CCND1/2/3, CDK4/6, MYC, PIK3CA, ATRX, MET, NF1,

BCOR, IGF2R, MYCN, BCORL1, ATM, DDX11, GNAQ, KDM6A, EGFR, TERT, TSC2, KMT5B,

MTOR and/or RPTOR.

In some examples, the disease state included may be one or more of midbrain and hindbrain malformations, a PontoCerebellar Hypoplasia (PCH), Pontine Tegmental Cap Dysplasia (PTCD), Congenital Cranial Dysinnervation Disorders (CCDD) (such as Athabaskan brainstem dysgenesis, Bosley-Salih-Alorainy syndromes, Horizontal Gaze Palsy with Progressive

Scoliosis), Congenital Disorders of Glycosylation (such as CDGS Type 1a), midbrain and hindbrain Lissencephaly, Moebius syndrome, Joubert syndrome and nuerodegnerative diseases (such as Alzheimer, Parkinson, Frontotemporal dementias).

PontoCerebellar Hypoplasias a clinically and genetically heterogeneous group of autosomal recessive developmental defects. In some examples, the PCH may be selected from PCH Type 1,2, 3,4, 5 and 6. PCH Type 1 may include mutations to genes or proteins thereof selected from EXOSC3, and VRK1. PCH Type 1 may include mutations of the RARS2 gene or proteins thereof. PCH Type 2 may include mutations to genes or proteins thereof selected from TSEN54,

TSEN2, and TSEN34. PCH Type 3 may include mutations of the PCLO gene or proteins thereof. PCH Type 4 may include mutations of the TSEN54 gene or proteins thereof. PCH Type 5 may include mutations of the TSEN54 gene or proteins thereof.

Any of the mutations associated with PCHs may be encoded in a nucleic acid vector as described herein and introduced into the organoids as described herein or introduced using gene editing systems described herein into the organoids as described herein.

Pontine Tegmental Cap Dysplasia (PTCD), Pontine tegmental cap dysplasia (PTCD) is a disorder characterized by substantial developmental delay, cranial nerve dysfunction, and a distinctive hindbrain malformation including hypoplasia of the pons, hypoplasia of the cerebellar peduncles, cerebellar vermis hypoplasia and an ectopic band of white matter on the dorsal side of the pons within the 4th ventricle. While PTCD can be classified within the broad category pontocerebellar hypoplasias, it can be distinguished from the autosomal recessive pontocerebellar hypoplasias by the ectopic white matter band on the dorsal pons, and it differs in its inheritance and prognosis. Genes and proteins thereof associated with PTCD include DCC and NTN1.

Any of the mutations associated with PTCD may be encoded in a nucleic acid vector as described herein and introduced into the organoids as described herein or introduced using gene editing systems described herein into the organoids as described herein.

Congenital Cranial Dysinnervation Disorders (CCDDs) are a spectrum of congenital non- progressive diseases with a similar underlying pathophysiology. The CCDD spectrum includes the Marcus Gunn jaw-winking phenomenon, Mébius syndrome, Duane Syndrome, Congenital

Fibrosis of the muscles, and other congenital facial palsies.

Athabaskan brainstem dysgenesis is associated with HOXA1 deficiency. Bosley-Salih-Alorainy syndrome (BSAS) is caused by homozygous mutations in the HOXA1 gene (7p15. 2) and is transmitted in an autosomal recessive manner. The syndrome overlaps clinically and genetically with Athabaskan brain dysfunction syndrome (ABDS). However, unlike ABDS, BSAS does not manifest central hypoventilation. Horizontal gaze palsy with progressive scoliosis (HGPPS) is a rare autosomal recessive disorder characterized by congenital absence of conjugate horizontal eye movements, preservation of vertical gaze and convergence and progressive scoliosis developing in childhood and adolescence. HGPPS is associated with mutations in the ROBO3 gene.

Any of the mutations associated with CCDDs may be encoded in a nucleic acid vector as described herein and introduced into the organoids as described herein or introduced using gene editing systems described herein into the organoids as described herein.

Congenital disorders of glycosylation (CDG) are a large group of rare genetic disorders that affect the addition of sugar building blocks, called glycans, to proteins in cells throughout the body. CDGs are associated with mutations in the EXT7 gene. CDGS Type 1a is associated with mutations in the PMM2 gene.

Any of the mutations associated with CDG may be encoded in a nucleic acid vector as described herein and introduced into the organoids as described herein or introduced using gene editing systems described herein into the organoids as described herein.

Midbrain and hindbrain Lissencephaly is a rare brain disorder that causes the brain's surface to appear smooth. It's caused by defective neuronal migration during the 12th to 24th weeks of gestation, which prevents the development of brain folds (gyri) and grooves (sulci). Midbrain and hindbrain Lissencephaly is associated with mutations in the RELN, ARX, and TUBA1A genes.

Any of the mutations associated with Midbrain and hindbrain Lissencephaly may be encoded in a nucleic acid vector as described herein and introduced into the organoids as described herein or introduced using gene editing systems described herein into the organoids as described herein.

Moebius syndrome is a rare birth defect that mainly affects the muscles that control facial expression and eye mavement. It is caused by the absence or underdevelopment of the sixth and seventh cranial nerves, which control eye movements and facial expression. Moebius syndrome is associated with mutations in the HOXA1, HOXB1, TUBB3, PLXND1 and REV3L genes.

Any of the mutations associated with Moebius syndrome may be encoded in a nucleic acid vector as described herein and introduced into the organoids as described herein or introduced using gene editing systems described herein into the organoids as described herein.

Joubert syndrome is a rare genetic condition characterized by abnormal brain development that includes the absence or underdevelopment of the cerebellar vermis (an area of the brain that controls balance and coordination) and a malformed brain stem. Joubert syndrome is associated with mutations in the AH!1, ARL13B, C5ORF42, CC2D2A, CEP41, CEP290,

INPPSE, KIF7, MKS1, NPHP1, OFD1, RPGRIP1L, TCTN1, TCTN2, TCTN3, TMEM67,

TMEM138, TMEM216, and TMEM237 genes.

Any of the mutations associated with Joubert syndrome may be encoded in a nucleic acid vector as described herein and introduced into the organoids as described herein or introduced using gene editing systems described herein into the organoids as described herein.

For other diseases and associated genes and mutations thereof see Doherty D, Millen KJ,

Barkovich AJ. Midbrain and hindbrain malformations: advances in clinical diagnosis, imaging, and genetics. Lancet Neurol. 2013 Apr;12(4):381-93. doi: 10.1016/51474-4422(13)70024-3.

Epub 2013 Mar 18. PMID: 23518331; PMCID: PMC4158743.

Neurodegenerative diseases such as Alzheimer’s, Parkinson's, frontotemporal dementias may be incorporated into the organoids as described herein by co-culturing the organoids with induced pluripotent stem cell derived neurodegenerative tissue or organoids. In some examples, neurodegenerative disease associated gene mutations or proteins include AB plaques, Amyloid precursor protein (APP), Presenilin 1 (PSEN1T), and Presenilin 2 (PSEN2) for

Alzheimer’s disease; mutations in LRRK2, PARK7, PINK1, PRKN, and SNCA gene for

Parkinson's disease; and mutations in MAPT, C9ORF72 and PGRN genes for frontotemporal dementias.

In some examples, the disease state is an infected state. For example, the methods may include infecting the organoid with one or more pathogens such as viruses or bacteria or proteins or genes thereof.

Uses

The hindbrain organoids provided herein may be used for studying development of the hindbrain and regions thereof (such as the pons). For example, using healthy hindbrain organoids as described herein. Healthy hindbrain organoids as described herein may also be used to study the interactions and effects of drugs on healthy hindbrain.

In addition, hindbrain organoids including diseased tissue as described herein may be used to model and study diseases as described herein. For example, study the progression and effects of diseases. In addition, hindbrain organoids including diseased tissue as described herein may be used for drug discovery, efficacy and/or toxicity studies.

For example, the hindbrain organoids including DMG tissue may be used to study DMG progression and the effectiveness of therapeutic agents or compounds for treatment thereof.

As such, in one aspect there is provided a method of testing one or more therapeutic agents, the method comprising: providing a hindbrain organoid as described herein,

contacting the organoid with at least one therapeutic agent after at least 30 days after maturing the hindbrain organoid in maturation medium for at least about 30 days; detecting one or more changes in the organoid; determining the effects of the therapeutic agent based on the absence or presences of the one or more changes.

Allowing the organoid to mature for at least 30 to about at least 60 days may allow for the diseased tissue to develop into diseased tissue which is more reflective of in vivo diseased tissue. For example, in the case of DMG tissue, allowing at least 30 days may allow for an immature tumour to form which provides a better resemblance to early stage DMG in vivo (i.e. in subjects suffering from DMG). For example, in the case of DMG tissue, allowing at least 60 days may allow for an immature tumour to form which provides a better resemblance to early stage DMG in vivo (i.e. in subjects suffering from DMG)

In some examples, contacting may be at timepoint T1 or after as described herein.

In some examples, contacting may be after at least 100 days after maturing the hindbrain arganoid in maturation medium as described herein. In some examples, contacting may be at timepoint T2 or after as described herein.

Allowing the organoid to mature for at least 100 days may allow for the diseased tissue to develop into diseased tissue which is more reflective of in vivo diseased tissue. For example, in the case of DMG tissue, allowing at least 100 days may allow for a mature tumour to form which provides a better resemblance to DMG in vivo (i.e. in subjects suffering from DMG).

Mature tumour refers to a tumour that includes at least some of the cells of a cancer (such as

DMG) to form.

Contacting may be achieved by any suitable method such as adding one or more therapeutics agents into the culture medium the hindbrain organoid is situated.

Suitable therapeutics may be selected depending on the diseased tissue present in the hindbrain organoids. Examples of therapeutics that may be of interest for studying effects on specific disease associated genes and proteins of DMG are provided in Table 1.

For example, the therapeutic compound may be one or more of APR-246, GSK-J4,

Mebendazole, PIP-199, Larotrectinib, AZ4547, dovatinib, PD173074, ponatinib, LDN212854,

Crenolanib, dasatinib, CCT007093, GSK2830371, Olaparib, Fimepinostat, Paxalisib, everolimus, Palbociclib, ribociclib, abemaciclib, Omomyc, Pyridostatin, Cabozantinib,

Binimetinib, trametinib, GSK1838705A, Bromodomain inhibitors, AZD1390, Irinotecan, Tris DBA palladium, Gefitinib, erlotinib, Imetelstat, Rapamycin, talazoparib, Everolimus, and/or AZD2014.

Therapeutics for Congenital Disorders of Glycosylation (such as CDGS Type 1a), include rare sugar therapy (with mannose or galactose)

Treatment of Moebius syndrome, includes bilateral selective neurolysis.

Therapeutics for neurodegenerative diseases include levodopa (L-DOPA, LARODOPA), pramipexole (MIRAPEX), and selegiline (ELDEPRYL, EMSAM, ZELAPAR), donepezil

(ARICEPT), entacapone (COMTAN, STAVELO), galantamine (NIVALIN), memantine (NAMENDA), rivastigmine (EXELON), and ropinirole (REQUIP).

In some examples, the therapeutic agent is an anti-cancer agent.

Examples of ant-cancer agents include, but are not limited to, antibodies, antibody fragments, conjugates, drugs, cytotoxic agents, proapoptotic agents, toxins, nucleases (including DNAses and RNAses), hormones, immunomadulators, chelators, boron compounds, photoactive agents or dyes, radioisotopes or radionuclides, oligonucleotides, interference RNA, peptides, anti- angiogenic agents, chemotherapeutic agents, cytokines, chemokines, prodrugs, enzymes, binding proteins or peptides or combinations thereof.

For example, chemotherapeutic drugs include vinca alkaloids, anthracyclines, epidophyllotoxins, taxanes, antimetabolites, tyrosine kinase inhibitors, alkylating agents, antibiotics, Cox-2 inhibitors, antimitotics, antiangiogenic and proapoptotic agents, doxorubicin, methotrexate, taxol, other camptothecins, and others from these and other classes of anticancer agents, and the like. Other cancer chemotherapeutic drugs include nitrogen mustards, alkyl sulfonates, nitrosoureas, triazenes, folic acid analogs, pyrimidine analogs, purine analogs, platinum coordination complexes, hormones, and the like. Suitable chemotherapeutic agents are described in REMINGTON'S PHARMACEUTICAL SCIENCES, 19th Ed. (Mack Publishing

Co. 1995), and in GOODMAN AND GILMAN'S THE PHARMACOLOGICAL BASIS OF

THERAPEUTICS, 7th Ed. (MacMillan Publishing Co. 1985), as well as revised editions of these publications. Other suitable chemotherapeutic agents, such as experimental drugs, are known to those of skill in the art.

Exemplary drugs include, but are not limited to, 5-fluorouracil, afatinib, aplidin, azaribine, anastrozole, anthracyclines, axitinib, AVL-101, AVL-291, bendamustine, bleomycin, bortezomib, bosutinib, bryostatin-1, busulfan, calicheamycin, camptothecin, carboplatin, 10- hydroxycamptothecin, carmustine, Celebrex, chlorambucil, cisplatin (CDDP), Cox-2 inhibitors, irinotecan (CPT-1 1), SN-38, carboplatin, cladribine, camptothecans, crizotinib, cyclophosphamide, cytarabine, dacarbazine, dasatinib, dinaciclib, docetaxel, dactinomycin, daunorubicin, doxorubicin, 2-pyrrolinodoxorubicine (2P-DOX), cyano- morpholino doxorubicin, doxorubicin glucuronide, epirubicin glucuronide, erlotinib, estramustine, epidophyllotoxin, erlotinib, entinostat, estrogen receptor binding agents, etoposide (VP 16), etoposide glucuronide, etoposide phosphate, exemestane, fingolimod, floxuridine (FUdR}), 3',5'-0-dioleoyl-

FudR (FUdR-dO), fludarabine, flutamide, farnesyl- protein transferase inhibitors, flavopiridol, fostamatinib, ganetespib, GDC-0834, GS-1101, gefitinib, gemcitabine, hydroxyurea, ibrutinib, idarubicin, idelalisib, ifosfamide, imatinib, L- asparaginase, lapatinib, lenolidamide, leucovorin,

LFM-A13, lomustine, mechlorethamine, melphalan, mercaptopurine, 6-mercaptopurine, methotrexate, mitoxantrone, mithramycin, mitomycin, mitotane, navelbine, neratinib, nilotinib, nitrosurea, olaparib, plicomycin, procarbazine, paclitaxel, PCI-32765, pentostatin, PSI-341, raloxifene, semustine, sorafenib, streptozocin, SU 11248, sunitinib, tamoxifen, temazolomide

(an aqueous form of DTIC), transplatinum, thalidomide, thioguanine, thiotepa, teniposide, topotecan, uracil mustard, vatalanib, vinorelbine, vinblastine, vincristine, vinca alkaloids and ZD 1839.

In some examples, the therapeutic agent is a chimeric antigen receptor therapeutic. Chimeric antigen receptor therapeutics typically include chimeric antigen receptor cells, which may be chimeric antigen receptor T cells, chimeric antigen receptor NK cells, and the like. The term "chimeric antigen receptor" (CAR), as used herein, refers to a fused protein comprising an extracellular domain capable of binding to an antigen, a transmembrane domain derived from a polypeptide different from a polypeptide from which the extracellular domain is derived, and at least one intracellular domain. The "chimeric antigen receptor (CAR)" is sometimes called a "chimeric receptor”, a "T-body", or a "chimeric immune receptor (CIR) " The "extracellular domain capable of binding to an antigen" means any oligopeptide or polypeptide that can bind to a certain antigen. The "intracellular domain" or "intracellular signalling domain” means any oligopeptide or polypeptide known to function as a domain that transmits a signal to cause activation or inhibition of a biological process in a cell. In certain embodiments, the intracellular domain may comprise, alternatively consist essentially of, or yet further comprise one or more costimulatory signalling domains in addition to the primary signalling domain. The "transmembrane domain" means any oligopeptide or polypeptide known to span the cell membrane and that can function to link the extracellular and signalling domains. A chimeric antigen receptor may optionally comprise a "hinge domain" which serves as a linker between the extracellular and transmembrane domains.

Examples of CAR therapeutics Abecma®, Breyanzi ®, Kymriah ®, Tecartus ®, Yescarta ®, and

Carvykti ®. Other examples of CAR therapeutics can be found in, for example,

WO2019220109A1, US11034750B2, WO2013123081A1, US20130287748A1, _WO2014055668A1, WO2014138704A1, WO2015075468A1, and WO2017216561A1.

In some examples, the therapeutic agent is a tumour infiltrating Iymphocyte (TIL) or Tumour specific TIL. As used herein the term “tumour infiltrating lymphocytes (TILs) refers to mononuclear white blood cells that have left the bloodstream and migrated into a tumour. TILs may be selected from the group consisting of T cells, B cells, NK cells and monocytes. Methods of obtaining TILs are well known in the art, such as obtaining tumour samples from a subject by e.g. biopsy or necropsy and preparing a single cell suspension thereof. The single cell suspension can be obtained in any suitable manner, e.g., mechanically (disaggregating the tumor using, e.g., a GentleMACS™ Dissociator, Miltenyi Biotec, Auburn, Calif.) or enzymatically (e.g., collagenase or DNase). Following, the TILs can be purified from the cell suspension.

There are several methods and reagents known to those skilled in the art for purifying the desired type of TILs, such as selection of specific cell types using cell surface markers (using e.g. FACS sorter or magnetic cell separation techniques such as are commercially available e.g. from Invitrogen, Stemcell Technologies, Cellpro, Advanced Magnetics, or Miltenyi Biotec.),

and depletion of specific cell types by methods such as eradication (e.g. killing} with specific antibodies or by affinity based purification based on negative selection (using e.g. magnetic cell separation techniques, FACS sorter and/or capture ELISA labeling). Such methods are described for example in the handbook of experimental immunology, volumes 1 to 4, (d.n. weir, editor) and flow cytometry and cell sorting (a. radbruch, editor, springer verlag, 2000).

In some examples, the therapeutic agent is an anti-infectious pathogen agent. For example, an anti-viral agent or antibiotic.

In some examples, detection of changes may be carried out using any suitable methods. The methods used will depend on the change being detected.

In some examples, changes detected may include phenotypic changes, changes in post- translational modification of proteins, changes in structure of the organoids, changes in the viability (e.g. whether cells are living or dead) of cells, transcriptional changes, genomic changes, epigenetic changes, protein changes, secretomic changes, and metabolic changes.

Detection may include analysis of changes in one or more of the transcriptome, genome, epigenome, proteome, epigenome, secretome, and metabolome, of organoid cells either by themselves or in combinations (multi-omics).

In some examples, the detected change is a phenotypic change. Phenotype may include one or more of protein expression, RNA expression, protein activity, or RNA activity, cell death, cell growth, cell motility, cell metabolism, drug resistance, drug sensitivity, and response to a stimulus.

In some examples, the detected change is death or survival of cells of the organoid. Detection of death of cells may be determined visually or by using microscopic or histological methods.

For example, by staining cells of the hindbrain organoids using compounds specific for dead or living cells. For example, using ethidium homodimer dye, calcein, CalceinAM, TO-PRO-3,

SYTOX, DIOC18 and/or BOBO-3 lodide. In some examples, dead or living cells may be detected by analysis of markers expressed by the cells. In some examples, dead or living cells may be detected by cell sorting methods such as FACs.

In some examples, the detected change is a transcriptional change. Transcriptional changes may be detected by any suitable method. Methods for determining changes in expression of genes are known. Non-limiting examples for methods of determining expression include, but are not limited to, RT-PCR, real time RT-PCR, next generation sequencing, western blot, dot blot, enzyme linked immunosorbent assay (ELISA). Differential expression of genes may be determined by performing RNA expression analysis. RNA may be extracted from samples of a hindbrain organoid and the level of RNA may be quantified by hybridisation of probes to provide a gene count. The level of expression or gene count, of each gene may then be normalised based on the expression levels of a number of housekeeping genes.

In some examples, the detected change is a change in the transcriptome of one or more cells of the hindbrain organoid. “Transcriptome analysis" refers to the analysis of all mRNAs (or primary transcripts, transcripts) present in one or proliferating cells of organisms in the same differentiation state under a specific cell biological condition. Since mRNA undergoes various changes due to the accumulation of extracellular influences received during the production of the cell, it is possible to analyse the properties of the current cell in detail. Specifically, analysis is performed using a microarray or the like.

In some examples differential expression is evaluated by determining a magnitude of change in nucleic acid molecule or protein expression, to determine if gene or protein expression is up- or down-regulated. For example, a relative value of expression can be determined. In some examples, a decrease in the relative value of expression indicates that the gene or protein is downregulated, while an increase in the relative value of expression indicates that the gene or protein is upregulated.

Differential expression of genes, and the expression levels of genes may be determined by any known methods. For example, using RNAseq based methods such as DESeq , edgeR ,

NBPSeq, TSPM, baySeq, EBSeq, NOISeq, SAMseq and ShrinkSeq. Other non-limiting examples for methods of determining expression include, but are not limited to, RT-PCR, real time RT-PCR, next generation sequencing, western blot, dot blot, enzyme linked immunosorbent assay (ELISA).

In some examples, the detected change is an epigenetic change. In some examples, analysis may be of single genes or an entire genome, for example epigenome analysis. "Epigenome" refers to the state or pattern of alteration of genomic DNA due to covalent modifications of the

DNA or proteins attached to the DNA. Examples of such alterations include methylation at position 5 of cytosine in CpG dinucleotides, acetylation of histone lysine residues, and other genetic or non-hereditary causes not due to alterations in the underlying DNA sequence.

Methods of analysing epigenetics may be similar to methods of gene or genome analysis but may include using epigenetic sensitive amplification and/or sequencing methods. For example, using western blot analysis; Chromatin Immunoprecipitation; Chromatin Immunoprecipitation followed by quantitative PCR (ChIP-gPCR); chromatin immunocleavage (ChlC) methods, cleavage under targets and tagmentation (CUT&Tag) methods, cleavage Under Targets and

Release Using Nuclease (CUT&RUN) methods, Directed Methylation with Long-read sequencing (DiMeLo), DNA adenine methylase identification (DamlD) methods, chromatin endogenous cleavage (ChEC) methods and/or nanopore-sequencing-based Histone- modification and Methylome joint-profiling methods; and/or Biotin-ChlP.

In some examples, the detected change is a protein change. Protein changes as used herein may be used to refer to changes in all proteins expressed by a cell of the hindbrain organoids (proteome) as well as changes in expression of specific proteins of a cell or cells. In addition, protein changes may include detecting the presence or absence of specific proteins and mutants thereof. Methods of protein analysis and detection are well known. Methods of proteome (all proteins expressed) analysis refers to the analysis of the relationship between genetic information and various proteins that perform complex interactions in cells. It is a large- scale analysis method that targets the structure and function of proteins, and can comprehensively analyse protein. Proteome refers to all proteins produced in specific cells, organs and viscera. For example, two-dimensional electrophoresis, a technique for separating proteins, can be used as a protein because of its high resolution and the ability to detect thousands of proteins at a time.

In some examples, the detected change is a metabolic change. In some examples, detection includes analysis of one or more metabolites of one or more cells. In some examples, detection includes analysis of all metabolites of one or more cells (e.g. metabolome analysis). “Metabolome” as used herein refers to the complete set of small-molecule metabolites to be found within an organism or cell. Metabolites analysed may include small molecule compounds, such as substrates for enzymes of metabolic pathways, intermediates of such pathways or the products obtained by a metabolic pathway. Metabolic pathways are well known in the art and may vary between species. Siad pathways include citric acid cycle, respiratory chain, photosynthesis, photorespiration, glycolysis, gluconeogenesis, hexose monophosphate pathway, oxidative pentose phosphate pathway, production and B-oxidation of fatty acids, urea cycle, amino acid biosynthesis pathways, protein degradation pathways such as proteasomal degradation, amino acid degrading pathways, biosynthesis or degradation of: lipids, polyketides (including e.g. flavonoids and isoflavonoids), isoprenoids {including e.g. terpenes, sterols, steroids, carotenoids, xanthophylls), carbohydrates, phenylpropanoids and derivatives, al- caloids, benzenoids, indoles, indole-sulfur compounds, porphyrines, anthocyans, hormones, vitamins, cofactors such as prosthetic groups or electron carriers, lignin, glu- cosinolates, purines, pyrimidines, nucleosides, nucleotides and related molecules such as tRNAs, microRNAs (miRNA) or mRNAs. Metaboilic analysis may be carried out using a number of different methods such as mass spectrometry, nuclear magnetic resonance spectroscopy, and chromatography.

In some examples, the detected change is a genomic change. For example, changes to individual genes, two or more genes or changes to the entire genome of one or more cells of the hindbrain organoids. Methods of detecting and analysing changes to genetic material are well known and include next generation sequencing methods. For example, sequencing by synthesis, semiconductor sequencing (lon Torrent), Sequencing by hybridisation (SOLID), 454 pyrosequencing, hanopore sequencing and/or single molecule real time sequencing, such as techniques available from Pacific BioSciences®.

In some examples, the detected change is one or more post-translational protein changes.

Post-translational protein changes refers to changes in post-translational modifications of proteins. Post-translational modification refers to modifications that occur on a protein after its translation by ribosomes is complete. A post-translational modification may be a covalent chemical modification or enzymatic modification. Examples of post-translation modifications include, acylation, acetylation, alkylation (including methylation), biotinylation, butyrylation, carbamylation, carbonylation, deamidation, deiminiation, diphthamide formation, disulfide bridge formation, eliminylation, flavin attachment, formylation, gamma-carboxylation, glutamylation, glycylation, glycosylation, glypiation, heme C attachment, hydroxylation, hypusine formation, iodination, isoprenylation, lipidation, lipoylation, malonylation, methylation, myristolylation, oxidation, palmitoylation, pegylation, phosphopantetheinylation, phosphorylation, prenylation, propionylation, retinylidene Schiff base formation, S-glutathionylation, S-nitrosylation, S- sulfenylation, selenation, succinylation, sulfmation, ubiquitination, and C-terminal amidation.

Methods of detecting post-translational modifications are well known and include mass spectroscopy.

In some examples, when the disease is a cancer (such as DMG) the change detected may be a change in the amount of cancer tissue present in the hindbrain organoid including cancer tissue as described herein.

In some examples, wherein the therapeutic includes T cells, the method may further include analysis of the T cells. For example, the method may include determining one or more of an exhaustion profile; behavioural changes; transcriptional changes; epigenetic changes, protein changes; metabolic changes; genomic changes; post-translational protein changes; and/or phenotypic changes of the T cells. Such changes may be determined and analysed by the methods described above respectively.

Behavioural changes of T cells includes, individual and collective changes in suppression, activation, localisation, proliferation ,cell motility (including change of direction and/or speed), retention time, resting time (e.g. in contact with diseased and/or healthy tissue), and morphological changes of cell body like polarization, elongation, and formation of protrusions.

T-cell exhaustion is a state where T cells gradually lose their functions over time, which can eventually lead to their physical deletion. It's characterized by an altered transcriptional program, decreased effector cytokines, and increased inhibitory receptors. Exhausted T cells can also produce fewer proteins that stimulate the immune response, and they may become less able to kill tumour cells or virus-infected cells.

Exhaustion is a complex phenotype that can be induced by several factors, most notably by chronic exposure to stimulatory antigen as occurs in chronic conditions such as cancer.

Substantial development is ongoing to develop CAR systems that can minimize exhaustion or counteract its effects. For both development purposes, and the profiling of exhaustion phenotype at the point of administration to patients, reporting the exhaustion status of CAR-T cells has substantial value.

Current methods of exhaustion profiling use flow cytometry or bulk analyses. Exhaustion is a complex phenotype that cannot be detected simply by the presence or absence of one marker, instead requiring the detection of multiple markers in combination. The most informative such markers may be inhibitory receptors such as PD1, CTLA-4, TIM3, LAGS, and TIGIT, in part because these are also effectors of exhaustion since signalling from these receptors strongly inhibits cell activation in response to antigen. Increased expression of just one such marker does not in itself denote exhaustion, however the concurrent increase of multiple markers within a population indicates a shift towards an exhausted phenotype. Since the major makers of exhaustion are also effectors of T cell inhibition, profiling their spatial organization is also highly informative. The extent of clustering of these molecules may be correlated with their extent of phosphorylation (since phosphorylation of tyrosine-based signalling motifs typically leads to receptor clustering), and so report the strength of underlying inhibitor signalling. The relative organization of such molecules to antigen receptors such as CARs or TCRs is also informative.

Methods of exhaustion profiling include those described in WO2023218071A1, Schillebeeckx, lan, et al. "T cell subtype profiling measures exhaustion and predicts anti-PD-1 response.”

Scientific reports 12.1 (2022): 1342 and Chow, Andrew, et al. "Clinical implications of T cell exhaustion for cancer immunotherapy.” Nature reviews Clinical oncology 19.12 (2022): 775- 790.

In some examples, the exhaustion profile may be used to identify markers that are indicative of a specific T cell group or type based. For example, markers for cytotoxic T cells may be identified.

Kit of Parts

Also provided herein is a kit of parts that includes at least one of a first, second or third culture medium as described herein.

For example, the kit includes at least one of: a first culture medium comprising:

FGF2 at a concentration of at most about 50 ng/pl;

Dorsomorphin at a concentration of about 1 HM;

SB431542 at a concentration of about 10 uM; and

CHIR99021 at a concentration of about 3 uM; or a first culture medium comprising or essentially consisting of: neurobasal medium and Advanced DMEM/F-12 medium at a ratio of 1:1; 1xGlutaMAXTM supplement; 1xN2 supplement; 2 ug/ml heparin solution;

FGF2 at a concentration of at most about 50 ng/pl;

Dorsomorphin at a concentration of about 1 HM;

SB431542 at a concentration of about 10 uM; and

CHIR99021 at a concentration of about 3 uM; a second culture medium comprising:

FGF4 at a concentration of about 10 ng/pl;

Retinoic acid at a concentration of about 10 HM;

Purmorphamine at a concentration of about 1 uM;

Dorsomorphin at a concentration of about 1 HM;

SB431542 at a concentration of about 10 uM; and

CHIR99021 at a concentration of about 3 uM; or a second culture medium comprising or essentially consisting of: neurobasal medium and Advanced DMEM/F-12 medium at a ratio of 1:1; 1xGlutaMAXTM supplement; 1xN2 supplement; 2 ug/ml heparin solution;

FGF4 at a concentration of about 10 ng/pl;

Retinoic acid at a concentration of about 10 HM;

Purmorphamine at a concentration of about 1 HM;

Dorsomorphin at a concentration of about 1 WM;

SB431542 at a concentration of about 10 uM; and

CHIR99021 at a concentration of about 3 uM; and/or a third culture medium comprising:

FGF4 at a concentration of about 10 ng/pl;

Retinoic acid at a concentration of about 10 HM; and

Purmorphamine at a concentration of about 1 HM; or a third culture medium comprising or essentially consisting of: neurobasal medium and Advanced DMEM/F-12 medium at a ratio of 1:1; 1xGlutaMAXTM supplement; 1xN2 supplement; 2 4g/ml heparin solution;

FGF4 at a concentration of about 10 ng/ul;

Retinoic acid at a concentration of about 10 HM; and

Purmorphamine at a concentration of about 1 HM.

In some examples, the kit may include a first, second and third culture medium as described herein.

In some examples, the kit further includes an initial culture medium as described herein. For example, initial culture medium comprising FGF2 at a concentration of about 4 ng/ul; Y-27632 ata concentration of about 10 uM; and Purmorphamine at a concentration of about 1 HM.

In some examples, the kit further includes one or more nucleic acid vectors encoding one or more disease associated genes, disease associated proteins, one or more interfering nucleic acid molecules and/or gene editing systems for mutating disease associated genes as described herein. For example, the kit may include one or more nucleic acid vectors encoding

PDGFRA-D842V, a H3K27M mutant and DNp53. In some examples, the kit may include one or more nucleic acid vectors encoding PDGFRA-D842V, a Cre recombinase and DNp53. In some examples, the one more vectors are piggyBac transposon vector such as a PBCAG-eGFP plasmid.

For example, the kit may include ane or more nucleic acid vectors gene editing systems for mutating disease associated genes. For example, one or more vectors encoding a CRISPR-

Cas9 system.

In some examples, the kit may include a maturation medium as described herein. In some examples, the maturation medium includes neurobasal medium and Advanced DMEM/F-12 medium at a ratio of 1:1, 1xGlutaMax supplement, and 0.5xN2 supplement, 0.5xB27 supplement and 1x Penicillin-Streptomycin.

In some examples, the kit of parts includes pluripotent stem cells. For example, human pluripotent stem cells. For example, induced pluripotent stem cells. For example, embryonic stem cells. In some examples, the pluripotent stem cells may include one or mare modifications for providing a hindbrain organoid comprising diseased and healthy tissue as described herein.

For example, the pluripotent stem cells may include one or more disease associated mutations which may be inducible.

In some examples, the kit of parts includes instructions for producing a healthy hindbrain organoid as described herein. For example, a healthy pontine organoid. In some examples, the kit of parts includes or further includes instructions for producing a hindbrain organoid comprising diseased and healthy tissue as described herein. For example, a healthy pontine organoid comprising diseased and healthy tissue as described herein.

Each part of the kit of parts may be provided in a separate container. In some examples, each culture medium, vector and/or distinct target binding region may be provided in a separate container.

While the making and using of various embodiments of the present invention are discussed in detail below, it should be appreciated that the present invention provides many applicable inventive concepts that can be embodied in a wide variety of specific contexts. The specific embodiments discussed herein are merely illustrative of specific ways to make and use the invention and do not delimit the scope of the invention.

The practice of the present invention will employ, unless otherwise indicated, conventional techniques of cell biology, cell culture, molecular biology, transgenic biology, microbiology, recombinant DNA, and immunology, which are within the skill of the art. Such techniques are explained fully in the literature. See, for example, Current Protocols in Molecular Biology (Ausubel, 2000, Wiley and son Inc, Library of Congress, USA); Molecular Cloning: A Laboratory

Manual, Third Edition, (Sambrook et al, 2001, Cold Spring Harbor, New York: Cold Spring

Harbor Laboratory Press); Oligonucleotide Synthesis (M. J. Gait ed., 1984); U.S. Pat. No.

4,683,195; Nucleic Acid Hybridization (Harries and Higgins eds. 1984); Transcription and

Translation (Hames and Higgins eds. 1984); Culture of Animal Cells (Freshney, Alan R. Liss,

Inc., 1987); Immobilized Cells and Enzymes (IRL Press, 1986); Perbal, A Practical Guide to

Molecular Cloning (1984); the series, Methods in Enzymology (Abelson and Simon, eds. -in- chief, Academic Press, Inc., New York), specifically, Vols.154 and 155 (Wu et al. eds.) and Vol. 185, "Gene Expression Technology" (Goeddel, ed.); Gene Transfer Vectors For Mammalian

Cells (Miller and Calos eds., 1987, Cold Spring Harbor Laboratory); Immunochemical Methods in Cell and Molecular Biology (Mayer and Walker, eds., Academic Press, London, 1987);

Handbook of Experimental Immunology, Vols. I-IV (Weir and Blackwell, eds., 1986); and

Manipulating the Mouse Embryo, (Cold Spring Harbor Laboratory Press, Cold Spring Harbor,

N.Y., 1986).

Terms such as "a", "an" and "the" are not intended to refer to only a singular entity, but include the general class of which a specific example may be used for illustration. The terminology herein is used to describe specific embodiments of the invention, but their usage does not delimit the invention, except as outlined in the claims.

Unless otherwise indicated, nucleic acids are written left to right in 5' to 3' orientation; amino acid sequences are written left to right in amino to carboxy orientation, respectively. It is to be understood that this invention is not limited to the particular methodology, protocols, and reagents described, as these may vary, depending upon the context they are used by those of skill in the art

Aspects of the invention are demonstrated by the following non-limiting examples.

EXAMPLES

Materials and Methods

Ethics

For the use of all DMG patient samples, patients and/or parents or guardians provided written informed consent according to national laws and in agreement with the declaration of Helsinki (2013). This study is Institutional Review Board (IVB) approved and registered under national registry number 2020. 142.

Stem cell culture

Brain organoids were generated from 3 different cell lines encompassing human Embryonic

Stem Cells (hESCs) H9 (WA09, Wicell) and H1 (WA01, Wicell) and induced Pluripotent Stem

Cells (iPSCs) C7-a (RUID 06C52463). The iPSC line C7-a was obtained from Rutgers

University Cell and DNA Repository (RUCDR) and contained a Cre-inducible H3.3K27M reading frame in the endogenous H3F3A locus”. Cell lines were cultured in mTeSR Plus medium (Stem Cell Technologies, Cat. #100-0278) and incubated at 37°C with 5% CO:. The cells were grown on Matrigel-coated (Corning, Cat. #354277) 6-well plates and passaged when 70-80% confluent by non-enzymatic detachment of colonies using Gentle Cell Dissociation

Reagent (GCDR, Stem Cell Technologies, Cat. #100-0485). All cell cultures were routinely tested for the presence of Mycoplasma species.

Embryoid body formation

The different stem cell sources were washed with 1X Dubecco’s Phosphate-Buffered Saline (DPBS, Gibco, Cat. #14190144) and detached with GCDR, before spinning down at 300 RCF for 5 minutes. Next, the cells were resuspended in BASE medium (1:1 Advanced DMEM/F-12 medium (Gibco, Cat. #12634010) and Neurobasal medium (Gibco, Cat. #10888022), 1X

GlutaMax (Gibco, Cat. #35050061)) and counted. 70,000 cells/ml were added to DAY O medium (BASE medium, 10 pM Y-27632 (ROCKi, AbMole BioScience, Cat. #M1817), 4 ng/ml Fibroblast

Growth Factor 2 (FGF2, PeproTech, Cat. #100-18C)). For Embryoid Body (EB) formation, 7,000 cells in 100 pl medium are seeded per well of an ultra-low attachment (ULA) treated U-bottom 96-well plate (Nexcelom, Cat. #ULA96U020/PHC Europe B.V., #MS-9096UZ) and incubated at 37°C with 5% CO. From day 2 — day 21, PATTERNING medium (BASE medium, 1X N2 (Gibco, Cat. #17502048), 1 mg/ml Heparin Solution (Stem Cell Technologies, Cat. #07980) was used.

Organoid patterning

To induce hindbrain and, more specifically, pontine identity, organoids were patterned using timely additions and replacement of morphogen supplemented media. In week 1, WEEK 1 medium (PATTERNING medium, 50 ng/ml FGF2, 1 uM Dorsomorphin (DM, Stem Cell

Technologies, Cat. #72102), 10 uM SB431542 (SB43, Stem Cell Technologies, Cat. #72232), 3

MM CHIR99021 (CHIR, Stem Cell Technologies, Cat. #72052)) was used. On day 2, 100 pl

WEEK 1 medium was added per well. On day 5, 100 pl medium per well was replaced with fresh WEEK 1 medium. In the second week, WEEK 2 medium (PATTERNING medium, 1 uM

DM, 10 uM SB43, 3 uM CHIR, 10 ng/ml Fibroblast Growth Factor 4 (FGF4, Stem Cell

Technologies, Cat. #78103.1), 10 pM All-Trans Retinoic Acid (RA, Stem Cell Technologies, Cat. #72262), 1 uM Purmorphamine (PMA, Stem Cell Technologies, Cat. #72202) was used. On day 7, 190 pl medium was replaced with fresh WEEK 2 medium and on day 9 100 yl medium was replaced. On day 11, the EBs were embedded in 12 pl Matrigel droplets and 5 droplets were transferred to each well of a 12-well suspension plate (Greiner Bio-One, Cat. #665102) with 1 ml WEEK 2 medium and incubated at 37°C with 5% CO:. In week 3, WEEK 3 medium (PATTERNING medium, 10 ng/ml FGF4, 10 uM RA, and 1 uM PMA) was used. Until day 21, every 2 days, the medium was refreshed with WEEK 3 medium. On day 1%, the plates were placed on an orbital shaker inside a 5% CO2 incubator at 37°C. From day 21 onwards, every 2-

3 days, the medium was refreshed with MATURATION medium (1:1 Advanced DMEM/F-12 medium and Neurobasal medium, 1X GlutaMax, 0.5X N-2, 0.5X B27 without vitamin A (Gibco,

Cat. #12587010), and 1X Penicillin-Streptomycin (Pen-Strep, Gibco, Cat. #15140122)).

DMG driver mutation-expressing and genetic lineage tracing plasmids

To induce DMG tumour growth in hESC-derived pontine organoids, the following plasmids were used: pCAGPbase, PBCAG DNp53_IRES luciferase, PBCAG_PDGFRA-D842V_IRES_eGFP, and PBCAG_H3K27M_eGFP. Alternatively, to induce DMG tumour growth in iPSC-derived pontine organoids, the H3K27M-expressing plasmid was replaced with 1.00 pg/ul Ssi-Cre to induce with an inducible H3.3-K27M mutation targeted to the endogenous histone locus. As a control, the following plasmids were used: 1.50 pg/ul pCAGPbase and 1.50 ug/ul PB_Venus. All plasmids were kindly provided by the Pheonix laboratory™. For genetic lineage tracing, 1.50 ug/ul

TrackerSeq*® was added to the tumour and control plasmid mix.

In situ electroporation and monitoring of tumour growth

On day 11, unless stated otherwise, pontine organoids were injected with a mixture of plasmid

DNA (1.50 pg/ul per plasmid) and 0.1% (w/v) FastaGreen (Merck, Cat. #F7252-5G) using a

FemtoJet 4i (Eppendorf, Cat. #5252000013) with the following parameters: Injection pressure (Pi) = 15 hPa and compensation pressure (Pc) = 5 hPa. Subsequently, the organoids were electroporated using a NEPA21 Super Electroporator (Nepagene) and CUY650P1 (Nepagene) tweezers with the following parameters: Voltage = 50 V, pulse length = 10 ms, pulse interval = 50 ms, number of pulses = 4, and decay rate = 10%. Transfer Pulse; Voltage = 20 V, pulse length = 50 ms, pulse interval = 50 ms, number of pulses = 5 and decay rate = 40%. Using the impedance (kQ) measurement of the NEPA21 Super Electroporator, voltage was automatically re-adjusted to optimize cell perforation and viability per individual organoid. Electroporation was performed by applying a shock twice in orthogonal direction. After electroporation, the organoids were incubated at 37°C with 5% CO2 for at least two hours to recover before Matrigel embedding. To monitor tumour growth over time, organoids were imaged on a Leica DM IL LED microscope with an N PLAN 5x/0,12 PHO objective and compared to the mVenus-positive control.

Multi-spectral large-scale single-cell resolution 3D (mLSR-3D) imaging

A comprehensive list of buffers, products, and clearing agents used for sample preparation can be found in the protocol by van Ineveld et al’. In short, organoids were fixed in 4% paraformaldehyde (PFA, Sigma-Aldrich, Cat. #441244) for 30 minutes at 4°C, washed 3X in PBT {1:1000 Tween-20 in 1X PBS) for 15 minutes at 4°C, embedded in 4% low melting point (LMP) agarose (Invitrogen, Cat. #16520-050), and sliced into 100-250 ym sections using a Leica VT 1200 S Vibratome. Sliced organoids were permeabilized in washing buffer 1 (WB1) on a shaker at 4°C for 3 hours and, subsequently, stained with primary antibodies diluted in washing buffer 2

(WB2) on a shaker overnight at 4°C. After primary antibody staining, the slices were washed with

WB2 at 4°C for 5 hours and stained with secondary antibodies diluted in WB2.. Additionally, the cell nuclei and membranes were stained with DAPI (1:2000, Invitrogen, Cat. #D1306) and

Phalloidin-Atto 488 (1:400, Invitrogen, Cat. #A12379), respectively. After secondary antibody staining, the organoid slices were washed with WB2 for 5 hours at 4°C and cleared with the fructose-based clearing agent FUnGI according to the protocol by Rios et al.’? and mounted on coverslips with silicone as spacer. The slices were imaged using a Zeiss LSM 880 Confocal microscope with a 25X (NA 0.8) objective and Leica Stellaris with 20X (NA 0.75) and 40X (NA 1.3) objectives. Alternatively, intact organoids were fixed and cleared using the organic solvent- based vDISCO method according to the protocol by Cai et a/.73 and imaged using a Leica SP8 microscope with a 16X (NA 0.6) BABB-compatible objective. 2D FFPE imaging

DMG patient material, as well as tumour-bearing pontine organoids, were fixed in formalin and embedded in paraffin at the histopathology department of Princess Maxima Center for Pediatric

Oncology to obtain formalin-fixed paraffin-embedded (FFPE) tissue for WHO-standardized tumour classification. Patient and organoid material was sliced into 3 um sections prior to hematoxylin and eosin (H&E) and subsequent stainings. Immunohistochemical staining was performed on the Leica BOND RX Fully Automated Research Stainer using the Bond Polymere

Refine Detection kit (Leica, Cat. #DS9800). The following antibodies were used: GFAP (RTU,

Leica, Cat. #PA0026), NF (RTU, Leica, Cat. #PA0371), H3K27M (1:400, Abcam, Cat. #AB190631), and H3K27me3 (1:200, Cell Signaling, Cat. #3733S). Stained tissue sections were analysed by an experienced neuropathologist. Sections for fluorescent-labelled imaging were subjected to antigen retrieval using Target Retrieval Solution pH 9 (Agilent Dako, Cat. #52367) with 60 minutes boiling time and stained using the mLSR-3D protocol with reduced incubation times (1h incubation at RT). Additionally, cell nuclei were stained with DAPI (1:2000) and the slices were imaged using a Leica Stellaris confocal microscope with a 20X (NA 0.75) and 40X {NA 1.3) objective.

Bulk RNA sequencing

Pontine organoids were used for bulk RNA sequencing at different patterning and maturation time points (week 1, 2, 3, 4, 8 and 12) from at least 3 different batches and generated from different stem cell sources, either H9 or C7-a. Bulk RNA sequencing was performed on pooled organoids, which were collected in 2 ml DNA low-binding tubes (Eppendorf, Cat. #0030108078). In addition, pontine organoids that were patterned with low (10 ng/ml) and higher (20 ng/ml) concentrations of FGF2 or FGF4 from week 2 to 3 were collected at the end of week 3. Organoids were mechanically dissociated in Hank's Balanced Salt Solution (HBSS, Gibco, Cat. #14025092), before spinning down at 800 RCF for 5 minutes at 4°C. Subsequently, supernatant was removed,

and the cell pellet snap-frozen on dry ice and stored at -80°C. Total RNA was extracted from cell pellets using the RNeasy Mini Kit (Qiagen, Cat. #74104) according to manufacturer’s instructions.

RNA concentration and integrity were evaluated by running 1 pl of the total RNA sample on an

RNA 6000 pico gel (Agilent, Cat. #5067-1513) using a 2100 Bioanalyzer (Agilent, Cat. #(G2939BA). Samples with an RNA Integrity Number (RIN) above 8 were sequenced on the lumina NextSeq500 platform and subsequently mapped and aligned by the Utrecht Sequencing

Facility (Used).

Bulk RNA data analysis

Raw counts of bulk sequencing data were combined into a single matrix and normalized by

DESeq2 median-of-ratios and variance-stabilized transformation (vst) (v1.40.2}"*. Gene counts below 10 were filtered out. To determine organoid batch variability, Spearman’s Rank

Coefficient was computed using built-in R functions (package “stats”). For principal component analysis (PCA) single organoid samples of the same batch were merged. Analysis of the regional identity of pontine organoids was done using VoxHunt (v1.0. 1)33, allowing spatial similarity mapping on in situ hybridization (ISH) data in the Allen Developing Mouse Brain Atlas.

Barplots were generated using top 15 genes per region on custom level 2 and 3, while spatial maps were generated using top 15 genes on custom level 1. Expression of neurogenic and gliogenic phase markers were plotted using normalized cell- and phase-specific genes combined per progenitor cell type: early NPC (PAX6, HMGA2), late NPC (DCX, STMN2), 5HT- progenitors (GATA2, GATA3), 5HT-neurons (HTRA2, SLC17A8); preOPC (DLL1, DLL3,

PPP1R14B), OPC (SOX10, PDGFRA) and OAPC/APC (OLIG1, HOPX, SPARCL 1). Expression values per gene and cell type were summarized per sample using the "stat_ summary" function from ggplot2 and rescaled between 0 and 1 using the minimum and maximum values to plot relative expression between samples. Evaluation of markers distinctive to human foetal brain regions was conducted to compare bulk RNA sequencing data from the pontine organoids with region-specific gene expression in the human foetal brain, as previously described’. Briefly, processed data from human foetal brain region specific genes were downloaded from Fan ef al.*® (https://www.science.org/doi/suppl/10.1126/sciadv.aaz2978/suppl_file/aaz2978_table_s1.xIsx).

Markers were selected that exhibited an average log2 fold change surpassing 0.35 and expression in over 85% of cells from a respective region. Specific markers for different regions within the organoid RNA sequencing dataset were determined by aggregating the consistent changes in respective marker genes derived from the human foetal brain dataset. This aggregation was accomplished through loadings of the first PC of their expression. Only region- specific markers with a minimum of 70% concordant change were considered in the analysis.

Subsequently, the average normalized expression of these region-specific gene markers was graphically represented for different time points in the pontine organoid dataset.

Single nuclei sequencing

DMGOs and organoid controls were micro-dissected under a Leica M205 FA fluorescence stereomicroscope at day 60 post electroporation. Micro-dissected tissues containing GFP* tumours were snap-frozen in liquid nitrogen for 30 seconds and stored at -80°C. For nuclei isolation, the tumour tissue was resuspended in 500 pl of Nonidet P-40 with salts and Tris (NST) lysis buffer and homogenized on ice using a glass-on-glass Dounce homogenizer with 10 strokes of the loose pestle, followed by 15 strokes of the tight pestle. Nuclear homogenates were applied to a 70 um Flowmi cell strainer (Merck, Cat. #BAH136800070) and centrifuged at 500 RCF for 5 minutes at 4 °C. The pellet was resuspended in WB1 and applied to a 40 um Flowmi cell strainer (Merck, Cat. #BAH136800040). The cell nuclei were stained with DAPI (1:200) for 5 minutes at room temperature before sorting. FACS was performed on a Sony SH800 cell sorter (Sony

Biotechnology) using a 100 um nozzle. Single Cell Gene Expression 3’ v3.1 (10x Genomics) was used for single nuclei capturing and library construction, as described in the Genomics Single Cell

RNA Reagent Kits User Guide. Briefly, 15,000 single sorted nuclei were loaded into a channel of a Chromium Single Cell Gene Expression 3’ Chip. Single nuclei were partitioned into droplets with gel beads in the Chromium followed by barcoded reverse transcription of RNA, cDNA amplification, fragmentation, and sample index ligation. The quality of the libraries was assessed on a 2100 Bioanalyzer (Agilent) and sequenced on a NovaSeq (Illumina).

Single cell RNA of tumour and GD2-CAR T cells and Trackerseg libraries

Single cell dissociation was performed using the Neural Tissue Dissociation Kit (P) (Miltenyi

Biotec, Cat. #130-092-628), adjusted for DMGOs. After cutting the organoids into smaller pieces and adding the enzyme mixes, the samples were incubated on an orbital shaker at 37°C and resuspended in regular intervals with a P1000 until a single cell suspension was reached, which was verified under the microscope using trypan blue to check for cell death. After dissociation, the single cell suspension was washed twice with PBS +/+ (magnesium/calcium+ 3% FBS). Next,

DAPI was used as a viability dye (Dapi 1:5000). Stained single cells suspensions were filtered using a 40 um Flowmi cell strainer and sorted by FACS on DAPI exclusion and enriched for GFP expression for tumour cells. For T cells, see section ‘FACS of GD2 CAR T cell treated DMGOs’.

FACS experiments were performed using the CytoFLEX SRT Benchtop Cell Sorter (Beckman

Coulter).

All samples analysed simultaneously using Hash tag oligo (HTO) demultiplexing (described in subsequent section) were pooled together, and single-cell encapsulation was performed according to the manufacturer's protocol 10X Genomics (Cell Preparation for Single Cell,

Demonstrated Protocol, CG000053). Pooled cDNA amplification was generated using the

Chromium Single Cell 3' V3 Library & Gel Bead Kit using the TruSegR1, TruSeqR2 and partial

TSO (template switch oligo) standard primers. Total cDNA of this reaction was then used for

Library preparation for mRNA and hashtag oligos as described per the manufacturer's protocol.

For the lineage tracing strategy, single-cell gene-expression, hashing, and TrackerSeq lineage barcode 10x genomics barcoded libraries were constructed for 60-day post-electroporation

DMGOs. nested PCR were additionally used to further amplify the lineage barcodes (primers in first PRC (10 cycles): sequences: 5-CTACACGACGCTCTTCCGATCT-3 (SEQ ID NO: 1) (Read1-Forward primers from 10X, 5- CTTCTCGTTGGGGTCTTT 3 (SEQ ID NO: 2) (eGFP primer-Reversed) Annealing Temperature was 60 and elongation time 30 sec; primers in the second PCR (10 cycles) Annealing Temperature was 60 and elongation time 30 sec: standard

P5-Read 1_Forward and Fun series 70x primer-Reversed from 10x). Sequencing of the prepared libraries was performed on an Illumina NovaSeg6000 in PE150 mode, and raw fastq files were processed and mapped with CellRanger v3.1.0. using a custom reference hg38 genome including sequences used in the electroporation methods, namely EGFP, H3.3K27M, Dnp53, Luciferase and Pdgfra.D824V.

Hash tag oligo (HTO) demultiplexing

Each organoid was then stained with a different TotalSeqg-A anti-human hashtag (Hashtag A0251-

A0265, Biolegend), allowing us to tag each cell with a sample-specific artificial oligonucleotide that can be recovered by sequencing. The hashing oligos were designed to recognize most human cells using a combination of two clones against CD298 and 2 microglobulin. In sum, the cell pellet was resuspended in staining buffer (50 ul for 500,000 cells}. Unspecific binding was reduced by adding 5ul of human Fc blocking reagent to the sample (FcX Human truStain,

Biolegend Cat. #422301). After 10 min incubation at 4 °C, 1 pl of a unique cell hashing antibody was added to each sample and incubated for 20 min at 4 °C and then washed 3x using PBS +0.04% of BSA. Demultiplexing of HTO-hashed single cell RNA sequenced samples was done using Seurat (v4.4.0). Briefly, CellRanger derived count matrices for ‘Gene Expression’ and "Antibody Capture’ were loaded and the Unique Molecular Identifiers (UMI) intersected to filter cells that are detected in both matrices. The ‘Gene Expression’ count matrix was used to generate a Seurat object and the “Antibody Capture’ was added as an assay (“HTO”). The “HTO” assay was normalized using Centered Log-Ratio (CLR). The Seurat object was subsequently demultiplexed with the embedded HTODemux() function using default parameters. Cells which were assigned one HTO-barcode (Singlets) were subsetted and used for further analysis.

Pre-processing and curation of single cell and single nuclei datasets

Downstream processing was done on individual samples using the Seurat workflow. Cells/nuclei were filtered on mitochondrial content (indicative of dying cells}, number of detected genes and number of UMIs. Threshold for these parameters were defined per sample based on their distribution. One sample with less than 100 cells after filtering was excluded from subsequent analysis. A gender score was assigned to each cell based on the expression of XIST using the

AddModuleScore() function in Seurat. Cells after filtering were scaled to 10.000 UMIs per cell and log-normalized. Mitochondrial content, gender and UMI counts were regressed out from normalized gene counts, the genes scaled and centred. Dimensionality reduction was applied to the top 2,000 highly variable genes using PCA. The first 30 principal-components were used for projection in Uniform Manifold Approximation and Projection (UMAP) space, for construction of a shared nearest neighbour (SNN) graph and clustering based on the Louvain algorithm. Potential doublets were detected using DoubletFinder3 (v2.0.3) and excluded. Cell cycle phase was determined as implemented in Seurat. Samples were visually inspected for expression of cell type marker genes and expression of tumorigenic plasmid genes. One sample contained high mesenchymal markers (PAX7, MYOG, MYOD1) and was therefore removed from the dataset.

The remaining cells were used for integration and analysis.

Integration of single cell and single nuclei data

After pre-processing, all datasets were merged into one single Seurat object. The top 2,000 highly variable genes were recalculated, and the Seurat object subsetted to only contain these genes.

Mitchondrial and ribosomal content, gender, cell cycle phase and UMI counts were regressed out from the merged normalized counts and the genes scaled and centred. The Python package scVI (v1.0.4) was employed using reticulate (v1.34.0) to balance confounding factors driving differences between the multiple datasets. To do this, the Seurat object was converted to AnnData format (sceasy, v0.0.7) and loaded into a scVI model with “RunlID” as batch variable. The model was trained for 400 epochs and after training the 10 latent embeddings were extracted and added to the Seurat object as a DimReducObject. These embeddings were used to project the cells in

UMAP space using default parameters. A SNN graph was constructed, and clusters detected using the Louvain algorithm on different resolutions. Clusters were inspected by building a clustertree using clustertree (v0.5.1) and representation in UMAP space. Gene signatures for each cluster were derived by comparing differentially expressed genes (DEGs) from one cluster to all other clusters based on MAST, using logfc.threshold = 0.5 and min.pct = 0.25.

Tumour and healthy classification

To identify tumorigenic clusters in the merged dataset, all cells expressing one or multiple of the tumour-inducing plasmids were binned per cluster (Louvain, res=0.8). Clusters that contain > 25% tumorigenic cells were classified as malignant, remaining clusters were annotated as non- malignant. To confirm the malignant cells' identity, the copy number of variation (iCNV) was inferred from single-cell gene expression data using inferCNV (v 1.18.1). InferCNV utilizes a sliding window approach across the genome, comparing the expression levels in test cells to reference cells. On a sample-by-sample basis, a non-transfected organoid was used as a healthy reference to estimate iCNVs in the malignant cells. The iCNV result is displayed in a heatmap representing CNV inference across the genome. Elevated or reduced expression levels indicative of CNVs are visually represented, facilitating easy identification of malignant clusters.

Tumorigenic clusters were subsetted, scaled and projected in UMAP space for further inspection.

GO terms and KEGG analysis of healthy neurons was performed using gprofiler2 (v0.2.2). DEGs from previous analysis per cluster were combined and selected with a p-value threshold of 0.05, highlighted terms were subsequently exported.

Reference comparisons and gene set enrichment

Raw expression matrices from reference datasets were obtained and processed using Seurat, as described in the corresponding papers’ 945. For annotation of tumour clusters, datasets were subsetted to their respective tumour annotations, scaled and centred. Anchors were obtained by

FindTransferAnchors(), using the first 30 PCs of the reference PCA and all genes of the reference dataset as features, and used to transfer labels from the reference to the query. A mapping score was calculated to identify cells that are poorly represented in the reference dataset, while the max prediction score was used to assess the confidence of the label transfer. This process was repeated on the cells annotated as OPC-like to further annotate the OPC-cluster as defined in Liu etal® For comparisons of the tumour-labelled clusters against other tumours or models, a similar approach was used. Multiple references were merged, and annotation was unified and condensed. Anchors were defined as described above, labels were transferred to the query

Seurat as a separate assay and used as input for visualization. The AddModuleScore() function in the Seurat package was used with default settings to compute gene set enrichments from curated lists of marker genes which were subsequently visualized. To compare gene expression between DMGO and reference datasets, the raw counts from the objects were merged, normalized, and scaled as one object before visualizing the gene of interest. All sequencing analysis was performed on R (v4.3.1) in RStudio (2023.090+463) or Python (v3.10.4). Molecular signatures for spatially restricted oligodendrocyte precursor lineage retrieved from Braun et al.’ were used for similarity comparison of CNMF generated program 1 and 2.

TrackerSeq barcode recovery

Lineage barcode recovery and downstream analysis was performed with custom made bash and python codes. Reads containing a perfect match with tracker-seq barcodes were extracted and count tables of tracker-seq barcode occurrence were built for each cell barcode and for each UMI.

This was done for both the library resulting from the nested PCR strategy, and the gene- expression library (Fig. 10a). Downstream analysis was performed only using the data obtained from the nested PCR approach. First, the total number of sequencing reads was quantified and the total number of unique UMIs for each cell barcode. Additionally, the mean oversequencing value for each cell barcode was computed, defined as the average number of reads detected for each UMI and for each tracker-seq barcode. Then, cell barcodes with more than 100 and 1000 total reads in experiment 1 and 2 were selected, respectively (represented by dashed red lines in

Fig. 10b). This criterion automatically selects cells in which at least one tracker-seq barcode was found to be oversequenced with an average maximum of 25 or 10 times {see dashed blue lines in Fig. 10e,f). Next, for each remaining cell barcodes the percentage of read counts (normalized to total reads) and observed UMIs (normalized to total observed UMIs) per tracker-seq barcodes was calculated, and only tracker-seq barcodes that are present at least 10% for both read count and UMI count fractions in at least one cell barcode were kept. For each cell barcode, the maximum oversequencing value for each tracker-seq barcode extracted and the fraction of counts using only these values was calculated. Tracker-seq barcodes with a fraction value below 0.2 were removed. If more than one tracker-seq barcode still remained per cell, tracker-seq barcodes that are less than 4 edits away from each other were pooled. If it was still the case that more than one tracker-seq barcode remained per cell, it was assumed that these should be detected according to a multinomial distribution. Therefore, if the fraction of counts for each resulting tracker-seq barcode was lower than 1/N-(1/N)**2 (mean-variance), it was deleted from the pool in that cell. For each cell, the final clonal barcode was defined as the union of the organoid ID from where it was derived from, and the tracker-seq barcode.

Consensus non-negative matrix factorization (cNMF) and referencing

Gene expression programs in tumour scRNA-seq data were inferred using cNMF (v1.4.1)%' as described in the vignette. In short, CNMF was performed using 100 iterations of NMF with different random seeds for each value of k, the number of components, from 5 to 9. For each value of k, metrics indicating stability, Silhouette score and Frobenius error, were calculated and the k maximizing the Silhouette score and minimizing the Frobenius error was selected for further processing. Outlier components from the selected k were filtered by removing components with a higher mean distance to most similar component of 0.02, which resulted in a program activity matrix and a gene scores matrix. For each malignant cell, program activity values for each component of the selected k were added as metadata to the Seurat object and classified based on highest program activity. Subsequently this annotation was used for further analysis. To compare the meta programs to previous annotations “scclusteval™s {v.1.0}, was used for cluster stability and similarity evaluation. The heatmap was generated via “PairWiseJaccardSetsHeatmap” function. Accordingly, fraction of louvian clustering per cNMF programs was compared by using “dittoBarPlot” function using shared nearest neighbor resolution grouped by meta programs. pYSCENIC

Putative regulatory networks in each cNMF acquired gene program were identified using the python implantation of SCENIC (pySCENIC, v0.12.1) (ref doi:10.1038/s41596-020-0336-2).

Processed malignant scRNA-seq data was converted from Seurat to loom format and pySCENIC ran using the CLI-mode with default options. Shortly, a gene regulatory network was created with

GRNBoost, using a list of transcription factors (TFs) provided by the Aerts lab

(https://resources.aertslab.org/cistarget/tf_lists/) and used to find enriched motifs. Activity of gene signatures across single cells was quantified using AUCell, defining a threshold of 0.05.

Upset plot generation

Upset plots were used to represent the intersections between the different clonal families and cNMF modules. Only clonal families found in more than one cNMF module are shown. Upset plots were depicted using the function plot from the upsetplot python package.

Comparison of large versus small clonal families

The fraction of each clonal family was computed for each sample. Big clones were defined as the clones that are detected in more than 20% of all the cells from each sample. Sample DMG0155 was removed from this analysis as it presented as only one large clone (>99% of the cells had the same tracker-seq barcode). DEG analysis was performed comparing the large versus small clones using the function rank_genes_groups from the scanpy python package and selected

DEGs (log2fold change >0.7) to perform METASCAPE analysis.

GD2 CAR T cell expansion and selection

CD8 GD2 CAR T cells (14G2a GD2-4-1BBz CAR) and donor-matched mock-transduced CD8 T cells, were produced as previously described’®. CAR T cells and mock transduced T cells were expanded using a rapid expansion protocol”. T cells were cultured in RPMI 1840 + GlutaMax (Thermo Fisher, Cat. #61870036), supplemented with 2.5-10% human serum (Sanquin), 1% Pen-

Strep, and 0.5M beta-2-mercaptoethanol (Thermo Fisher, Cat. #21985023), on a feeder cell mixture comprising of sub-lethally irradiated allogenic PBMCs, Daudi, and LCL-TM cells, in the presence of 50 U/ml IL-2 (R&D Systems, Cat. #P60568), 5 ng/ml IL-15 (R&D Systems, Cat. #P40933), and 1 pl/ml PHA-L (Sigma-Aldrich, Cat. #11249738001) and cryopreserved after 14 days of expansion. Prior to experiments, T cells were thawed and rested in RPMI 1640 +

GlutaMax, with 10% Fetal Bovine Serum (FBS, Thermo Fisher, Cat. #10500064) and 1% Pen-

Strep, supplemented with 50 U/ml IL-2 (Miltenyi, Cat. #130-097-743), 2000 U/ml IL-7 (Miltenyi,

Cat. #130-095-367), and 50 U/ml IL-15 (Miltenyi, Cat. #130-095-760) for 3 days at 37°C with 5%

CO...

Treatment of DMGOs with GD2 CAR T cells

Four months after tumour induction, DMGOs were transferred to 12-well suspension plates and untreated or treated with 500,000 CD8* GD2 CAR T cells, or mock transduced CD8* T cells per

DMGO. 7 days after the start of treatment, 500,000 T cells were added per DMGO for a second round of treatment. Tumour size during treatment was monitored by imaging on day 0, 3, 7, 10, and 14, using a Leica Thunder DMi8 microscope with a 10X objective. In addition, one DMGO was treated on day 0, 8 and 15 with GD2 CAR T cells and imaged on day 8, 15, 28 and 35. After

THUNDER software-mediated computational clearing of the imaging data, tumour size for each time point was quantified using Fiji. In short, background signal, defined as GFP-negative areas within the organoid, was subtracted. The organoid surface was set as region-of-interest (ROI) and mean gray values of the GFP channel for the ROI were calculated.

FACS of GD2 CAR T cells and DMGO GD2 expression

DMGOs treated with GD2 CAR T cells were dissociated 14 or 35 days after initial T cell addition with the Neural Tissue Dissociation Kit (P) (Miltenyi Biotec, Cat. #130-092-628), as described above for preparation of single cell RNA and tracker seq libraries. Dissociated cells were washed and stained in FC buffer with CD3-APC (1:80; BD Biosciences, clone SK7) and LIVE/DEAD

Fixable Near-IR Dead Cell Stain (1:1000; Thermo Fisher) for 30 min at 4°C. CD3* T cells and

GFP* tumour cells were sorted on a CytoFLEX SRT Benchtop Cell Sorter (Beckman Coulter) and immediately processed for scRNA-seq. To confirm DMGO GD2 expression for GD2 CAR T cell treatment evaluation, a day 60 post-electroporation DMGO sample was dissected for the tumour region to enrich for tumour material, mechanically dissociated, and cultured for 2 additional weeks to expand tumour cells. Cells were retrieved from the culture plate using StemPro Accutase (Gibco, Cat. #A1110501) and passed through a 70 pm Flowmi cell strainer (Merck, Cat. #BAH136800070) to create a single cell suspension. Dissociated cells were centrifuged at 500

RCF for 5 minutes at 4 °C and resuspended and washed in FC buffer (2% fetal bovine serum (FBS), 1x PBS). Cells were either left unstained, or stained with LIVE/DEAD Fixable Near-IR

Dead Cell Stain (1:1000; Thermo Fisher) and GD2-PE (1:200, clone 14.G2a, BD Biosciences,

Cat. #562100) for 30 min at 4°C. After staining, cells were washed twice in FC buffer, acquired on a Sony SH800s (Sony Biotechnology), and analysed using FlowJo Software (v10.9.0).

Pre-processing and analysis of GD2 CAR-T cell scRNA-seq datasets

As the first quality control step, doublets (two, or more, cells captured in the same droplet) for each sample were identified and removed using the scDblFinder package’, with default settings.

Low quality cells with high mitochondrial content (> 15%), or cells with extremely high or low reads (< 200 genes or > 6500 genes), or cells with extremely high reads (> 35000 reads) were removed.

Normal Seurat 47° workflow was used to normalize and scale reads, and the 3000 most variable features determined using “FindVariableFeatures” in the Seurat package. Cell cycle confounding effect was eliminated from the dataset via the removal of cell cycle-related genes from the variable features of the dataset. PCA was performed using “RunPCA” function. First 30 PCs were used for non-linear dimensionality reduction utilizing UMAP® method, implemented via “RunUMAP” function of the Seurat package. Clustering analysis was performed on the first 10 PCs using the

Seurat package's ‘FindNeighbors’ and ‘FindClusters’ functions. A resolution parameter of 0.45 was applied, and the original Louvain algorithm was used. To identify sub-populations, marker genes for each cluster were determined through the "FindAllMarkers™ function. Markers obtained from this analysis were then examined to profile genes associated with known CD8 T cell subsets, as well as to project previously published signatures (see T cell signature projection below). Only markers with adjusted p values below 0.05 were taken into consideration. In addition, DEGs were used as input for gene ontology (GO) enrichment analysis using the GO resource (https://geneontology.org).

T cell signature projection

To evaluate the expression of established T cell signatures in GD2 CAR T cell scRNA-seq datasets, a gene sighature specific to serial killer engineered T cells that was previously obtained (Dekkers et al’, see Supplementary Table 4 therein) was used. Utilizing the VISION R package®’, the overall enrichment of the identified gene set atop UMAP cell embeddings of the dataset was computed and visualized. In addition, GD2 CAR T cell signature profiles were projected onto a pan-cancer CD8 tumour infiltrating lymphocyte (TIL) atlas from Chu et al.%%, which encompasses T cells infiltrating brain tumours. For each GD2 CAR T cell subset, markers obtained through DEG analysis were meticulously curated to obtain the most relevant markers, ensuring an adjusted p-value below 0.00001. Accessing a publicly available and interactive online data portal (https://singlecell.mdanderson.org/TCM/), a rds file containing the Seurat object pertinent to scRNA seq data of CD8 TILs was acquired. Subsequently, the VISION package was employed to perform the projection of the GD2 CAR T cell signatures onto this dataset.

Statistical and heatmap analysis

Statistics on bulk sequencing data was computed by built-in functions of R (“stats”, v4.3.1) using one-way ANOVA with post-hoc Tukey Honest significance difference. PCA, Spearman’s Rank and gene expressions were plotted using ggplot2(v3.4.2), heatmaps were generated using pheatmap package (v1.0.12). Statistics on electroporation efficiency and tumour induction was calculated using the two-tailed independent t-test (function: t.test). For each batch and timepoint the mean and standard deviation was calculated, individual values were imported and plotted in

GraphPad Prism (v.8.0.2) using summarizing stacked bar plots. All statistical tests have been performed with the assumption of a normal distribution, equal variance per sample and a confidence interval of at least 95% (alpha = 0.05).

Data availability

All used R and Python scripts are available in the laboratory GitHub. All sequencing datasets (bulk, single nuclei and single cell) will be deposited on NCBI Gene Expression Omnibus (GEO) before publication. Sequencing metadata is provided in Table 2.

Results

Human pontine organoids

To create a human organoid with pontine identity for subsequent DMG tumour induction and treatment response modelling, morphogen guidance based on a timely sequence of Wnt, dual

SMAD inhibitors, retinoic acid (RA), fibroblast growth factors (FGFs) and sonic hedgehog (SHH) was implemented to specify hindbrain identity (Fig. 2a and Fig. 3a).

While FGF2 and FGF8 can be used in combination with RA and Wnt to pattern midbrain?°, cerebellum? or spinal cord?! in growing organoids, FGF4 was evaluated because of its role in specifying rostral hindbrain, particularly in prepontine and pontine areas?233. Additionally, FGF4 has been shown pivotal for generating serotonergic neurons from human pluripotent stem cells, which are predominantly found in the brainstem region. A direct comparison of replacing common FGF2 supplementation with FGF4 after one week of patterning, demonstrated that 10 ng/ml of FGF4 specifically gives rise to developing pontine, including prepontine to retropontine areas based on bulk RNA sequencing data (Figs. 3b,c, Table 2a). Expression of HOXB1, a marker of pontine precursor cells35, emerged from week 2 onwards (Fig. 1b) and 3D imaging reveals HOXB1 expressing cells within early neurodevelopmental SOX2+ neural rosette structures (Fig. 1¢). Similar to human neurodevelopment, the hindbrain-specific early marker

GBX2 and fore-/midbrain-specific OTX2% are only expressed at week 1 and sharply decline afterwards (Extended Data Fig. 1d). In contrast, the expression of hindbrain-associated ME/S genes are increasing over time, whereas spinal cord-specific CDXs® are either not expressed or decline (Fig. 3e). Together, this strongly indicates anatomical restriction to the hindbrain area. To further investigate this regional profile, organoid development beyond the initial 3-week patterning period was transcriptomically tracked until 12 weeks of maturation. This demonstrates that the established pontine-dominated identity remains stable, as shown by spatial similarity mapping to foetal mus musculus transcriptomic data (VoxHunt33) (Fig. 2d).

Furthermore, referencing the RNA profiles with human foetal datasets® (Gestation Week 9-25, 21-28) shows established pontine identity over time, while foetal cortical similarities decline (Fig. 2e, Fig. 3j). Reflection of naturally occurring segregation of neuro- and gliogenesis phases was also observed (Fig. 3f,g). At 16 weeks of age, pons-specific serotonergic neurons (Tryptophan hydroxylase 2, TPH2+) (Fig. 2f) emerge, as well as committed astrocytes (Glial fibrillary acidic protein (GFAP)+; Aquaporin 4 (AQP4)+), and sparse committed oligodendrocytes (OLIG2+) (Fig. 3h,i) could be observed. Lastly, the patterning remained consistent and reproducible across and within multiple batches, as well as between human embryonic stem cell (hESC) and induced pluripotent stem cell (IPSC) sources (Fig. 3k,l). Taken together, a guided hindbrain brain organoid protocol, implementing FGF4 to reliably give rise to human pontine identity has been developed.

De novo H3.3K27M-altered DMG organoids

It was investigated whether the pontine identity of the newly developed human organoid model could be exploited to model DMG tumours. The most common H3.3K27M-defining DMG mutation? alongside typical accompanying and pons-specific tumour suppressor TP53 and platelet-derived growth factor A (PDGFRA) alterations3#-*3, expressing plasmids were introduced using in situ electroporation of developing pontine organoids (Fig. 2a). This mutation cocktail has been shown to be time-sensitive in in utero electroporation mouse models121415 hence different timepoints of electroporation were tested between day 11 and 28. Day 11 was identified as the timepoint most efficiently inducing tumorigenic growth (Fig. 4a,b), reinforcing the concept of a restricted early developmental time window for DMG transformation'2,

Tracking tumour growth over two months showed that the resulting tumours display infiltrative growth, until reaching a diffuse growth pattern specific to DMG (Fig. 4c), whereas the use of empty control plasmids resulted in only a few localized electroporated cells (Fig. 4b). Analysis of histopathological appearance shows that H3.3K27M cells (H3K27M+) display loss in H3K27 trimethylation (H3K27me3) (Fig. 4d), a hallmark of H3K27-altered DMG**° and again confirms invasive diffuse growth (Neurofilament (NF)+/GFAP+ tumour cells) into surrounding healthy tissue (Fig. 4e). Whole-organoid 3D imaging on week 16 (4 months post-electroporation) showed that tumours remain stable in their diffuse growth (Fig. 5a). While unguided cerebral organoids have been shown to not give rise to H3K27M+ tumorigenic growth before?! they could potentially contain hindbrain cells. Therefore, this approach was tested on unguided cerebral organoids, revealing a significant reduction in tumour induction (Fig. 4f), as well as a non-diffuse outgrowth (Fig. 4g). Together, these data demonstrate that the correct anatomical cellular identity modelled with the pontine organoids gives rise to the same diffuse and invasive growth pattern and trimethylation loss as DMG tumours found in patients.

Developing and region-specific microenvironment

To gain more insight into tumour heterogeneity within its developing microenvironment, single cell and single nuclei transcriptomic profiling was performed and a total of approximately 13,000 cells from 11 pontine DMG organoids (DMGOs) were analysed, as well as two control pontine organoids (empty vectors) (Fig. 2a, Table 2b) two months post-electroporation. Integration of these datasets resulted in 30 clusters (Fig. 6a-c), which were first divided using a plasmid- detection approach {see Methods), profiling 6,634 tumour and 4,731 healthy cells collected from the microenvironment after quality control filtering (Fig. 5b). The tumorigenic profile of the cancerous cells was further validated through inferred copy number variation, showing chromosomal gains and losses in several genomic regions, as compared to controls cells from healthy brain organoid (Fig. 6d). Differentially expressed gene (DEG) analysis and reference mapping of a recent single-cell human healthy foetal brain dataset** was performed to assign cell subsets among non-malignant clusters (Fig. 7a,b). Committed neurons (NRG3, NRXN1,

GRIA2, RBFOX1, MAP2, ERB4 and PLCG?2), together with glioblasts (VIM, SOX2, CLU, BCAN,

NCKAP5, PPP2R2B, GFAP) and axon-guiding neuroepithelium (NTN1, RMST, SLIT2) were identified (Fig. 7a). Furthermore, gene set analysis of DEGs from the neuron clusters (clusters 5, 15 and 19) uncovered enrichment for excitatory glutaminergic, inhibitory GABAergic, as well as dopaminergic and serotonergic neuron signatures (Fig. 7¢). Together, this data shows a diverse developing and committed neuronal environment, as well as macroglia differentiation, in line with the developing and regional specification of the pontine organoids, providing a representative environment for DMG progression.

Patient-representative tumour cell subsets

Tumorigenic populations were classified using published DMG references”? and major tumour cell types present in patients were identified; OPC -, MES-, AC-like and cycling cells (Fig. 5c,

Fig. 8a-c). In line with the early developmental window of the model, only few cells with a more mature OC-like phenotype were identified. For each of the main tumour cell subsets the resemblance of DMGOs to current widely applied in vitro and in vivo models were quantified, as well as transcriptomic data from human patients. It was found the highest similarity score between DMGOSs and primary DMG patient material’, as opposed to cell lines, patient-derived xenografts (PDXs), adult glioblastoma (GBM)’ and posterior fossa ependymoma (PFA 1 and 2) patient material*®, the latter presenting with a similar loss of H3K27M trimethylation caused by

EZHIP overexpression or in a few cases also an H3K27M mutation*® (Fig. 5d). Importantly, a major proportion of OPC-like tumour cells resembled a recently defined paediatric and pons- specific OPC-like-2 state (also described as a pre-OPC state®) (Fig. 5e). Cellular retinoic acid binding protein 1 (CRABP1T), previously shown predominantly expressed in hindbrain-specific

OPC lineages, as compared to midbrain or forebrain’ and strongly correlating with the OPC- like-2 cell population in DMG99, was accordingly enriched in this OPC-like-2 cluster (Fig. 5f).

Furthermore, in patient samples? it was found that CRAPB1 specifically expressed in pontine, as compared to thalamic and cortical DMG tumours (Fig. 5g), highlighting the right regional specificity of DMGOs for the development of a cancer cell population specific to the pons. This finding was validated at the protein level, by showing a consistent CRABP1 expression pattern in DMGOs as in primary DMG patient samples (Fig. 5h). Furthermore, an aquaporin 1 (AQPT) expressing AC-like cancer cell population (cluster 9) was identified (Fig. 8c). AQP1 expression has been shown exclusive to astrocytes arising in the human brainstem*®. It was demonstrated that AQP7 was present at the transcriptomic level in both patients? and DMGOs (Fig. 5i).

Furthermore, AQP1 was observed only in tumours found in the pons? and not those arising from the cortex and thalamic region, as opposed to AQP4, a canonical AC-like marker present in

DMG tumours at all locations (Fig. 5j). Using 3D imaging, the presence AQP 1+ tumour cells in

DMGOs and patient samples at the protein level was validated (Fig. 5k), confirming AQP1 as a marker specific to pontine DMG tumours. These findings underscore the human regional relevance of the model leading to patient-representative cell types. Altogether this highlights the potential of pontine DMG organoids to model heterogeneity in DMG tumour cell subsets and, importantly, human pontine-specific disease.

Hindbrain glial specification essential for DMG tumorigenesis

Exploring the possibilities of DMGOs to investigate the mechanisms underlying DMG tumorigenesis, the clonal relationship and transcriptomic signatures of de novo DMG tumours was traced (Fig. 2a). TrackerSeq*®, a PiggyBac-based barcoding plasmid was employed, and in total retrieved 167 unique barcodes from six different DMGOs and two healthy pontine organoids (Fig. 9a, Fig. 10a-g. The TrackerSeq approach detected individual clones spanning up to approximately 800 cells per barcode, indicating cancerous transformation for larger clones (Fig. 10h,i). This provides an opportunity to dive deeper into the mechanisms that drive clonal expansion within the DMG model, by comparing large versus small barcoded clones (Fig. 9b,c).

DEG and METASCAPE analysis identified response to growth factor (Fig. 9b) and Early Growth

Factor 1 (EGR1)* (Fig. 9c) to be upregulated in larger clones, aligning with their proliferative capabilities. This analysis also revealed glial specification as a pivotal feature associated with cancer clone expansion, as opposed to neuronal fate (enriched in small clones), indicative that the trajectory of glial cell differentiation is essential for DMG tumour development (Fig. 9b).

Since DMG tumorigenesis was studied within the neural developmental context, where cells transition continuously among different states rather than fitting discrete categories, consensus non-negative matrix factorization (cNMF)*'-% was applied and delineated eight meta gene programs with sets of genes that are coordinately regulated across the analysed cell population (Fig. 9d). After excluding clones with fewer than three cells per barcode, 34 clones remained for analysis. Reconstruction of lineage relationships revealed that the transcriptional profile of cells arising from single clones span across multiple programs (Fig. 9e), suggesting a divergence from early common gene programs into distinct cellular identities. Therefore, programs that were present in most clones were sought out and Programs 1, 2, and 3 were identified to be present in the highest number of clones, in 30, 26, and 24 clones, respectively (Fig. 9e). As this implies a central role for Programs 1, 2, and 3 in tumorigenesis, all identified programs were cross-referenced with DMG tumour cell subsets to interrogate their cell type dominance. This demonstrated Program 1 and 2 to be highly specific to the OPC-like lineage, again emphasizing the central role of this lineage in H3K27M DMG tumorigenesis®, whereas Program 3 presented with a MES-like cell dominance (Fig. 9f). To gain more insight into the features that unify

Program 1, 2 and 3 in their clonal dominance, despite having either an OPC-like (Program 1 and 2) or MES-like profile (Program 3), SCENIC regulon analysis was employed to identify key transcription factors (TFs) linked to these gene programs (Fig. 9g,). All three programs were characterized by early neural molecular features and ventricular zone (VZ) characteristics; a transient embryonic layer primarily housing radial glial cells, which serve as neural stem cells

(NSCs) and undergo gradual differentiation into neuronal and glial transient progenitors as they transition from the (sub)ventricular zone (SVZ) into the mantle zone?25+. More specifically,

Program 3 was defined by the earliest NSC-specific controlled regulons; SOX2 and HESS, as well as the SVZ marker; EOMES®®. Importantly, the top Program 3-specific TF; SOX9, plays a crucial role in specifying the glial lineage in NSCs during early embryonic development, aligning with the critical role of glial trajectory specification in DMG. Program 1 and 2 also exhibited early neural molecular features, with Program 2 displaying Nescient Helix-Loop-Helix 1 (NHLHT) (Fig. 9g) and both programs showing elevated expression of Stathamin 2 (STMN2) (Fig. 9h). Both these genes have been identified as specific to the VZ in early embryonic brain development, with STMN2 particularly enriched to the hindbrain mantle zone*'. Furthermore, other hindbrain-specific TFs; GBX2 and HOX genes, such as HOXAS and HOXBS, also known to be associated with pontine H3.3K27M DMG tumours?, were identified as top TFs, supporting the strong hindbrain specificity of these OPC-like programs. In the context of human early gestation, regionally distinct gene signatures for the glial lineage have been observed, which have been suggested to underly the strong region-specific pattern of occurrence of glial-related diseases, such as DMG*. In line with this, both Program 1 and 2 specifically enrich for the hindbrain-pons oligodendrocyte precursor lineage (referred as oligo), as opposed to midbrain and forebrain (Fig. Si). Altogether, the lineage tracing approach reveals a crucial role for hindbrain glial specification in DMG tumour progression, further underscoring the importance of accurately modelling DMGs within their appropriate spatial and developmental context.

CAR T cell heterogeneity

If DMGOs could serve as a human in vitro platform for preclinical evaluation of CAR T cell therapy was investigated (Fig. 2a). The recent first clinical outcome of T cell therapy for DMG was achieved using GD2 CAR T cells? and GD2 target expression in DMGOs was confirmed (Fig. 11a).

Further, it was assessed if DMGO's will continue to give rise to more differentiated states of

DMG, mimicking naturally occurring differentiation processes as observed in foetal brain development. Indeed, while at week 8 post-electroporation (71, Week 8, ~d60) only a few committed oligodendrocyte-like (OC-like) cancer cells were observed (Fig. 5¢), with low expression of lineage- and stage-specific markers (SOX10+, PDGFRA+, OLIG1+, OLIG2+), the expression significantly increased with ongoing organoid maturation (72, Week 16, ~d120) (Fig. 14A). Similar to ongoing OC-like maturation, AC-like cancer cells followed the same trajectory and increased significantly in expression of lineage- and stage-specific markers (SOX9+,

AQP4+, AQP1+, GFAP+) (Fig. 14B). Together this data suggests that with increasing age of

DMGO's, the cancer improves in patient representativeness and thus later stages of DMGO development enhance the predictability and evaluation of CAR T cell therapy.

Hence DMGOs were treated four months after tumour induction with CD8* GD2 CAR T cells and monitored tumour control over time. Similar to incomplete clinical outcomes reported in patients?®, GD2 CAR T cells were able to partially reduce tumour burden in DMGOs (Fig. 12a).

However, heterogenous response rates over time were observed, with sometimes a delayed response to treatment that for one DMGO resulted in overall no tumour reduction compared to the start of treatment (Fig. 11b,c). Importantly, therapy effects could be detected even after >1 month of treatment (Fig. 12b), offering advantages for modelling CAR T cell functionality in vitro in a manner that is representative of T cell states at the tumour site in vivo. This includes potential exhaustion profiles associated with prolonged tumour exposure. For in vitro model systems this has not yet been achieved in the context of naturally expressed tumour-antigen, only through persistent anti-CD3 and anti-CD28 antibody stimulation®’, or using repeated rounds of stimulation with antigen- pulsed, or overexpressing® tumour cell lines. To test the potential of the new model for this purpose, over 30,000 GD2 CAR T cells retrieved from

DMGOs were sequenced and unbiased clustering and UMAP projection of the obtained gene expression dataset was performed. This revealed a substantial level of heterogeneity; identifying 9 transcriptional states (Fig. 12c) that, based on combined interrogation of curated gene signatures (Fig. 13a), DEGs , DEG-associated GO terms (Fig. 13b-f), expression of canonical immune effector (Fig. 12d) and exhaustion markers (Fig. 12e) and comparison to a recently published pan-cancer infiltrating T cell (TIL) dataset that includes brain malignancies (Fig. 139), reflected different T cell activation, differentiation and effector states. For instance, a

GD2 CAR T cell population that, although activated (based on HLA gene expression (Fig. 13a), is not fully differentiating towards effector function (undifferentiated; Tunp) was identified (Fig. 13b,h). In addition, an /L-2 responsive population (Ti) (Fig. 13¢,i}, probably differentiating into effector T cells was identified (Fig. 13i), as well as an interferon-stimulated gene (ISG) expressing population (Fig. 13a) (Tisc) strongly corresponding to /SG expressing TILs®? (Fig. 13j) and considered an interferon-induced activation state®! 2. Other clusters included a CAR T cell population with migrating properties and interconnectivity (Tw) that appears to be predominantly shaped by the brain tissue environment and might even acquire direct cellular communication with neurons (Fig. 13d), as well as proliferating (Ter) (Fig. 13e) and metabolically stressed T cells (Tus) (Fig. 13a,f). Importantly, potential DMG-targeting effector T cell populations based on their cytotoxic profile (Fig. 12d) and putative level of exhaustion were distinguished (Fig. 12e). While one of these clusters predominantly expressed GZMK (Tezk), cytotoxic T cells (Tort) expressed GZMB, PRF1 and IFNG (Fig. 12d). In contrast, exhausted T cells (Tex) displayed reduced /FNG and concomitant expression of immune checkpoint genes;

LAGS3, HAVCR2, TIGIT®3 and SELPLG®, as well as the transcriptional repressor PRDM1 associated with exhaustion® (Fig. 12e), demonstrating the potential advantage of prolonged treatment in DMGO for uncovering T cell functional exhaustion, considered an actionable axis to enhance treatment outcomes® and, therefore, critical to recognize during pre-clinical evaluation.

Discussion

By featuring a unique combination of human-, developmental-, location- and cell state- specificity, DMGOs accurately capture pontine DMG, phenotypically in its diffuse nature, as well as its transcriptomic heterogeneity. This distinctive profile enabled comprehensive investigation into DMG tumorigenesis, further supporting the intricate interplay between developmental processes and tumour progression’3, with hindbrain development influencing DMG growth trajectory and biological characteristics. This data and others 1%" indicate that NSCs may serve as a state of transformation, potentially being overgrown in fully transformed tumours, as previously proposed by Liu et al.®. Furthermore, these findings suggest that DMG, like other childhood cancers originating from embryonic structures, could emerge as early as the first trimester, possibly within NSCs located within the transient developmental structure of the hindbrain ventricular zone (VZ). However, by leveraging advanced barcode-tracing methodology, it was also uncovered that glial specification is a pivotal feature associated with cancer clone expansion. This highlights that the critical trajectory leading towards glial fate, as opposed to neuronal, is essential for the emergence of DMG tumours. Future studies focusing on glial specification within the hindbrain region throughout development might help elucidate the molecular mechanisms reflective of its developmental origin and potentially unmask therapeutic targets necessary for sustaining this paediatric brain tumours, while being dispensable in postnatal tissues. This approach could facilitate the development of effective, relatively non-toxic treatments.

Towards clinical application, these models prove to be compatible with prolonged engineered T- cell treatment, accurately reflecting CAR T cell heterogeneity and treatment outcomes as observed in patients. The identification of potent CAR (Tcyr) T cells is promising and aligns with the ‘super-engager engineered T cell profile that was previously described as the most potent cancer-targeting engineered T cell state in a short organoid co-culture assay®’. However, despite the presence of these cells, only a partial reduction in tumour burden was observed, consistent with observations in patients?®. Given the wide functional heterogeneity of CAR T cells uncovered, one possibility is that there are insufficient numbers of these potent Teyr CAR T cells to eradicate the entire tumour burden. There is increasing recognition that the tumour microenvironment (TME) can significantly impact tumour progression and treatment response.

By inducing disease in a human pontine-fated organoid, the TME mirrors the developmental stage and regional identity of DMG. Establishment from human pluripotent stem cells enables integration of additional environmental cell compartments, including immune cell lineages, such as critically important tumour-associated macrophages®®. This could allow investigating their role in DMG progression and assessing their impact on the phenotype and performance of T cell therapy, especially for the most potent Tcyr CAR T cells identified. Moreover, validation of

DMGOs using iPSCs as a cell source makes patient-specific modelling a future possibility, creating a scalable in vitro system for personalized drug screening. Thus, a bona fide human organoid model for DMG with critical applications towards understanding disease progression and response to treatment that has been developed here. Given the general fatality of DMG, this DMGO model can provide a critical chain towards uncovering new knowledge for improved therapy development for this detrimental disease.

The reader's attention is directed to all papers and documents which are filed concurrently with or previous to this specification in connection with this application and which are open to public inspection with this specification, and the contents of all such papers and documents are incorporated herein by reference.

All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and/or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and/or steps are mutually exclusive.

Each feature disclosed in this specification (including any accompanying claims, abstract and drawings), may be replaced by alternative features serving the same, equivalent, or similar purpose, unless expressly stated otherwise. Thus, unless expressly stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features.

The invention is not restricted to the details of any foregoing embodiments. The invention extends to any novel one, or any novel combination, of the features disclosed in this specification (including any accompanying claims, abstract and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed.

Tables

Table 1 — Mutations associated with DMG

Genetic Mutated {| Molecular | Preval Location Treatment

Sie sma

TP53 G245S, | H3.1K27M, 60- | Pons, APR-246, GSK-J4

R175H, | H3.3K27M, 80% | thalamus,

R248Q, | EZHIP midbrain

R248W,

R273C,

R273H,

S241F,

V157F

KDR S1154P, | H3.3K27M 32% | Pons Mebendazole

Amplifica tion,

CNG

KIT T96P, H3.3K27M 30% | Pons Mebendazole

Amplifica tion,

CNG

NTRK1/2/3 | TPM3_N | H3.3K27M 19% | Pons, midbrain | Larotrectinib

TRK1

VCL_NT

RK2

ETV6_N

TRK3

FGFR1 K697E, H3.3K27M, 18% | Pons, AZ4547, dovatinib,

N98S, H3.1K27M, thalamus, PD173074, ponatinib

N546K, | EZHIP midbrain

K656E

ACVR1 R206H, | H3.1K27M, 16% | Pons, thalamus | LDN212854

G328V, | EZHIP

G328W

PDGFRA Y288C, H3.3K27M 15% | Pons Crenolanib, dasatinib

C235Y,

Amplifica tion

PPM1D WA427*, | H3.1K27M, 15% | Pons, CCT007093,

E525X, H3.3K27M, thalamus, GSK2830371, olaparib

Q404X, | EZHIP midbrain

E405X, 428 fs

PTEN A1268S, H3.1K27M, 13% | Pons, thalamus | Fimepinostat

R130X, | H3.3K27M,

Deletion | EZHIP

PIK3R1 K567E, H3.3K27M, 12% | Pons, Paxalisib, everolimus,

G376R EZHIP thalamus, fimepinostat midbrain

RB1 Amplifica | H3.1K27M, 12% | Pons, thalamus tion, H3.3K27M

Deletion

CCND1/2/3 | Amplifica | H3.3K27M 10% | Pons, thalamus | Palbociclib, ribociclib,

EL a”

CDK4/6 Amplifica {| H3.3K27M 10% | Pons Palbociclib, ribociclib, tion, abemaciclib

L185V

MYC R33C, H3.3K27M 10% | Pons Omomyc

Amplifica tion

PIK3CA E545K, H3.3K27M, 8% | Pons, Paxalisib, fimepinostat 391M, H3.1K27M, thalamus,

H1047R | EZHIP midbrain

ATRX H2254R, | H3.3K27M, 8% | Pons, thalamus | Pyridostatin

R2197L, | EZHIP

L1357fs

MET P664P, H3.3K27M 8% | Pons, midbrain | Cabozantinib

Amplifica tion

NF1 G295R, H3.3K27M 7% | Pons Binimetinib, trametinib

R1204W

Deletion

BCOR C1363fs, H3.1K27M, 6% | Pons

A535V, EZHIP

G101fs

IGF2R K162R, | H3.1K27M 6% | Pons GSK1838705A

MYCN | Amplifica | EZHIP 6% | Pons Bromodomain inhibitors tion,

CNG

BCORL1 S425], H3.1K27M, 6% | Pons on woe

ATM G2342V, | H3.3K27M, 5% | Pons AZD1390 oe ae | TT

DDX11 R186W, | H3.3K27M 5% | Pons Irinotecan 0 a

KDM6A Deletion, | H3.3K27M 4% | Pons

EGFR | R108K, | H3.1K27M, 4% | Pons, thalamus | Gefitinib, erlotinib

Amplifica | H3.3K27M, tion, EZHIP

CNG

TERT C228T, H3.3K27M 2% | Pons Imetelstat an al

TSC2 D1587V, | EZHIP 2% | Pons Rapamycin ae

KMT5B R187%, H3.3K27M, 1% | Pons Olaparib, talazoparib

Tl we | TT

MTOR A1971V | H3.3K27M 1% | Pons Everolimus, fimepinostat,

RPTOR D857N 1%

Table 2 - Metadata information for bulk (a) and single cell and nuclei sequencing performed on organoids {b) and GD2 CAR T cells (c).

Table 2A

Sample Cell Timep | Concentr | Treatme | Bat | RURID Sample n on goe an mene en

BRO12-7 | H9 d7 10 uM Retinoic | V12 | ARI5927 | Pooled 3 er ers

BRO13-7- | H9 d7 10 uM Retinoic | V13 | ARI5927 | Single 1

El a = cl a

BRO13-7- | H9 d7 10 uM Retinoic | V13 | ARI5927 | Single 1

EA a = a

BRO13-7- | H9 d7 10 uM Retinoic | V13 | ARI5927 | Single 1

SP fe

BRO12- H9 d14 10 uM Retinoic | V12 | ARI5927 | Pooled 3

Ea a = a

BRO13- HO d14 10 uM Retinoic | V13 | ARI5927 | Single 1 wo em

BRO13- HO d14 10 uM Retinoic | V13 | ARI5927 | Single 1

El A a

BRO13- HO d14 10 uM Retinoic | V13 | ARI5927 | Single 1

El > a d14-v18- | HO d14 10 uM Retinoic | V18 | ARI6293 | Pooled 3 wo CT

BRO12- HO d21 10 uM Retinoic | V12 | ARI5927 | Pooled 3 "

BRO13- HO d21 10 uM Retinoic | V13 | ARI5927 | Single 1 wo TTT Le Ee

BRO13- HO d21 10 uM Retinoic | V13 | ARI5927 | Single 1

A I a =

BRO13- HO d21 10 pM Retinoic | V13 | ARI5927 | Single 1 ne TTT La i d21-v18- | HO d21 10 pM Retinoic | V18 | ARI6293 | Pooled 3 a

BRO12- HO d28 10 pM Retinoic | V12 | ARI5927 | Pooled 3 a wT

BRO13- HO d28 10 pM Retinoic | V13 | ARI5927 | Single 1

Ei =

BRO13- HO d28 10 pM Retinoic | V13 | ARI5927 | Single 1

Er =

BRO13- HO d28 10 pM Retinoic | V13 | ARI5927 | Single 1

El EE d28-v18- | HO d28 10 pM Retinoic | V18 | ARI6293 | Pooled 3 wt EE iPSC- IPSC d28 10 HM Retinoic | V1 | ARI6293 | Pooled 3 a i d50-v12 HO d50 10 HM Retinoic | V12 | ARI6293 | Pooled 3 ee d50-v13 H9 d50 10 HM Retinoic | V13 | ARI6293 | Pooled 3 hee d50-v14 H9 d50 10 uM Retinoic | V14 | ARI6293 | Pooled 3

Or ae d85-v12 HO dss 10 uM Retinoic | V12 | ARI6293 | Pooled 3 ee a ld ll d85-v13 HO dss 10 uM Retinoic | V13 | ARI6293 | Pooled 3 ill ll ddl d85-v14 HO d85 10 uM Retinoic | V14 | ARI8293 | Pooled 3 ee [ae [aa [ee es

FGF2-10- | H9 d21 10 ng FGF2 V44 | HON777 | Pooled 3 pee [ew |r Jo

FGF2-10- | H9 d21 10 ng FGF2 V45 | HON777 | Pooled 3

See [ew jee

FGF2-10- | H9 d21 10 ng FGF2 V46 | HON777 | Pooled 3 eee [er ed nn

FGF2-20- | H9 d21 20 ng FGF2 V44 | HON777 | Pooled 3 eee [er oe os

FGF2-20- | H9 d21 20ng FGF2 V45 | HON777 | Pooled 3

Sem jen Jr :

FGF2-20- | H9 d21 20ng FGF2 V46 | HON777 | Pooled 3 ep je jn Jr

FGF4-10- | H9 d21 10ng FGF4 V44 | HON777 | Pooled 3 ee [ee oe

FGF4-10- | H9 d21 10ng FGF4 V45 | HON777 | Pooled 3 tll ld al

FGF4-10- | H9 d21 10ng FGF4 V46 | HON777 | Pooled 3

Few [ewe Joe

FGF4-20- | H9 d21 20 ng FGF4 V44 | HON777 | Pooled 3 etl ll l=

FGF4-20- | H9 d21 20 ng FGF4 V45 | HON777 | Pooled 3 tll ld al

FGF4-20- | H9 d21 20 ng FGF4 V46 | HON777 | Pooled 3 tl wl ln idl

Lc cE «= wd OQ Cc

Q A

ILG

Q

Ez Ez

Q

= IR o o

L158 5 2 2 bl © E a oo E c © 2 £ sxe © E o oo E 3 LG E = > TT £ £ 2 SL © < wn Ec 5 ZZ 5 = e= 5 5 rar -— t cd at N N © = 0 gq < oo 9

O 7

FE oo 3 < <

E

Ss © 5 a w oO Ww oO © 5 O £2 5 © = ~ cn cn L6& 8 x xX @ E B® N © Q I SOx w um 8 EG J © = 8 & « = SNN dt 5 0 — 0 — © & CS E 9 5 Oo gE 48 35 0

HEE oO) + <I — ee BE

LO © u) —

FR BT

£ + ©

N I~

NN N

Oo oO °

FFF

— — oN

Oo oO ~ ö oO ©

FB RB

0 0

EH << O Oo O OQ O < HOE < zZz Feoodz= =z OO OF £2 VF FF OOk

Ex kenen kenen ees k2= pes

EE

— =

SE 228, 228, 5s 8488-55555 56 So 556 2 3S 8 38 83L 3|wvw BE 3a do or 2 8&1 _ 0 Qd ~ w O — w Q _

En vo Sn To Ls co LS ee = 3 Oo > Zo « > Zo eN

Fs’s BERN BESS

O + OO = OO + 0 = OO 4 OQ =O On

N ON N

VOOr 0 Oo QO = O vO OO 0

OF <o d= EO Oo = 00 «@ =

EO << pa ~EQQICpa 1X0 Aa ~

Te _ _—

EDT ES DT EDT

ST 556 So 556 So 556 dd a2 3 na » oF 8 8a do oa 2 3 a w OQ _ _ Ww QL _ wn OO _ cm LS En oo LC cao LL en

SE 90 3 EO 3 £90 9 nw DO oO wy DO OO nw DD oO << << “om na ZOO a zo 0 on no zoog

EB BE

N A A —

Oo 0npg = OFF Og= FEO <@z OFF

E64 oo ES ¢@ oo ES Po ES

BERN BERN BERN BE

28 9e LZ

B

NN

N

I~ <~ u)

N

O RO O

OSU 50z0d= =

O LE iF OF OO LZ 2 «

Oo — < Oo Non mw = ce ewes _—

Be o co 9

Do ES @ oo 56 So 556 3 9x » sr 3 21m

T =

EI DEN

)

Oo 1 — oc 8 i oN

A 2 £ & zo © 0

IES u) x I 2 =~

T © 3

E OO 1

LH o “ES 3 £ c 3 E 9 E O 7 >= = 5 u 9 Eg 0 © x © Oo ©) c tT O = <f © E 0 E ~ 2 5 a € © c 9% £ § EE > E 5 Cc a 4 ] c 2 5 0 LI 2 < oN c xX ® 2 u“) oN = C 0 on t E UL 1Ì © h LT} ~ c 9 == lS < 0 c 2 = = ®

E oo x

U 9 i © 2 > # SL << Om 3 oo Laoag" oO oe d Ex = © > 2 Cc 3 = c 2 F ® © OT © TI 2 c © oo [ex _ & x = CH ow oN 3 U go © 5 SN 2 of e= 5 © ¢ © 8 2 3 o € oo 8 ~ 2 gp 0

LL Oo OQ oo «on O mo 0

SSD Tos 588 Sc o 3 o oc 2 oo 5 oo co I) oO

A ~~ 0 Oo oO oN oO <

ED

N - 0 00 = 0 0 0 vo x <~ vo 0 o o 0 oO ' ° co

N oO Oo <

POF FFL

Q = OO

DO << 8 I< OQ «coogi 9 Q |< F

QO Op OF EEA@MOIK EQ SOOKE gol @Q

O OF LZ 00 «<< ZI DO Zig OF Zi OOO co © Oo LO oa > «© > cn > oa > «© > > > > > > <t © Jz 0 < © [Te © Jd © oo 0 © oo © o © 0 © ow N ow NN ow YN « ow N ow NN

GN | © ™ | © T+ | © 0 | © © ~ OO I ~~ OO 1 ~ OO 1 ~ OO I ~~ OO 1 z GG EE > Eo g z GG EE > o , 3 e SH $ 222 gl0 vB eo B2agizL 8 a oO 0s Loe _ LE Os

MN ©)

BB eee age 9 >= 9) > ee

EE

Sequences wmewE [WWE | maemo ii | wed |e ommcToTIGes | wor | 8

References 1. Louis, D. N. et al. The 2021 WHO Classification of Tumors of the Central Nervous System: a summary. Neuro-Oncol. 23, 1231-1251 (2021). 2. Findlay, |. J. ef al. Pharmaco-proteogenomic profiling of pediatric diffuse midline glioma to inform future treatment strategies. Oncogene 41, 461-475 (2022). 3. Schwartzentruber, J. et al. Driver mutations in histone H3.3 and chromatin remodelling genes in paediatric glioblastoma. Nature 482, 226-231 (2012). 4. Tilburg, C. M. van et al. The Pediatric Precision Oncology INFORM Registry: Clinical

Outcome and Benefit for Patients with Very High-Evidence TargetsThe Pediatric Precision

Oncology INFORM Registry. Cancer Discov. 11, 2764-2779 (2021). 5. Vuong, H. G., Ngo, T. N. M., Le, H. T. & Dunn, |. F. The prognostic significance of

HIST1H3B/C and H3F3A K27M mutations in diffuse midline gliomas is influenced by patient age. J. Neuro-Oncol. 158, 405-412 (2022). 8. Thomas, B. C. et al. CAR T cell therapies for diffuse midline glioma. Trends Cancer 9, 791- 804 (2023). 7. Filbin, M. G. et al. Developmental and oncogenic programs in H3K27M gliomas dissected by single-cell RNA-seq. Science 360, 331-335 (2018).

8. Jessa, S. et al. K27M in canonical and noncanonical H3 variants occurs in distinct oligodendroglial cell lineages in brain midline gliomas. Nat. Genet. 54, 1865-1880 (2022). 9. Liu, |. et al. The landscape of tumor cell states and spatial organization in H3-K27M mutant diffuse midline glioma across age and location. Nat. Genet. 54, 1881-1894 (2022). 10. Jessa, S. ef al. Stalled developmental programs at the root of pediatric brain tumors. Nat.

Genet. 51, 1702-1713 (2019). 11. Monje, M. et al. Hedgehog-responsive candidate cell of origin for diffuse intrinsic pontine glioma. Proc. Natl. Acad. Sci. 108, 4453-4458 (2011). 12. Pathania, M. ef al. H3.3K27M Cooperates with Trp53 Loss and PDGFRA Gain in Mouse

Embryonic Neural Progenitor Cells to Induce Invasive High-Grade Gliomas. Cancer Cell 32, 684-700.e9 (2017). 13. Larson, J. D. et al. Histone H3.3 K27M Accelerates Spontaneous Brainstem Glioma and

Drives Restricted Changes in Bivalent Gene Expression. Cancer Cell 35, 140-155.e7 (2019). 14. Patel, S. K. ef al. Generation of diffuse intrinsic pontine glioma mouse models by brainstem targeted in utero electroporation. Neuro-Oncol. 22, 381-392 (2019). 15. Siddaway, R. & Hawkins, C. Modeling DIPG in the mouse brainstem. Neuro-Oncol. 22, 307-308 (2020). 16. Funato, K., Major, T., Lewis, P. W., Allis, C. D. & Tabar, V. Use of human embryonic stem cells to model pediatric gliomas with H3.3K27M histone mutation. Science 346, 1529-1533 (2014). 17. Haag, D. et al. H3.3-K27M drives neural stem cell-specific gliomagenesis in a human iPSC- derived model. Cancer Cell 39, 407-422.e13 (2021). 18. Bressan, R. B. et al. Regional identity of human neural stem cells determines oncogenic responses to histone H3.3 mutants. Cell Stem Cell 28, 877-893.e9 (2021). 19. Pun, M. et al. Common molecular features of H3K27M DMGs and PFA ependymomas map to hindbrain developmental pathways. Acta Neuropathol. Commun. 11, 25 (2023). 20. Lancaster, M. A. et al. Cerebral organoids model human brain development and microcephaly. Nature 501, 373-379 (2013). 21. Bian, S. et al. Genetically engineered cerebral organoids model brain tumor formation. Nat

Methods 15, 631-639 (2018). 22. Ogawa, J., Pao, G. M., Shokhirev, M. N. & Verma, |. M. Glioblastoma Model Using Human

Cerebral Organoids. Cell Reports 23, 1220-1229 (2018). 23. Lago, C. et al. Medulloblastoma and high-grade glioma organoids for drug screening, lineage tracing, co-culture and in vivo assay. Nat. Protoc. 18, 2143-2180 (2023). 24. Hendriks, D. et al. Human fetal brain self-organizes into long-term expanding organoids.

Cell (2024) doi:10.1016/j.cell.2023.12.012. 25. Ballabio, C. et al. Modeling medulloblastoma in vivo and with human cerebellar organoids.

Nat. Commun. 11, 583 (2020).

26. Persson, M. L. et al. The intrinsic and microenvironmental features of diffuse midline glioma:

Implications for the development of effective immunotherapeutic treatment strategies. Neuro-

Oncol. 24, 1408-1422 (2022). 27. Chen, Y., Zhao, C., Li, S., Wang, J. & Zhang, H. Immune Microenvironment and

Immunotherapies for Diffuse Intrinsic Pontine Glioma. Cancers 15, 602 (2023). 28. Majzner, R. G. et al. GD2-CAR T cell therapy for H3K27M-mutated diffuse midline gliomas.

Nature 603, 934-941 (2022). 29. Qian, X. et al. Brain-Region-Specific Organoids Using Mini-bioreactors for Modeling ZIKV

Exposure. Cell 165, 1238-1254 (2016). 30. Muguruma, K., Nishiyama, A., Kawakami, H., Hashimoto, K. & Sasai, Y. Self-Organization of Polarized Cerebellar Tissue in 3D Culture of Human Pluripotent Stem Cells. Cell Rep. 10, 537-550 (2015). 31. Andersen, J. et al. Generation of Functional Human 3D Cortico-Motor Assembloids. Celf 183, 1913-1929.e26 (2020). 32. Ye, W., Shimamura, K., Rubenstein, J. L. R., Hynes, M. A. & Rosenthal, A. FGF and Shh

Signals Control Dopaminergic and Serotonergic Cell Fate in the Anterior Neural Plate. Cell 93, 755-766 (1998). 33. Wurst, W. & Bally-Cuif, L. Neural plate patterning: Upstream and downstream of the isthmic organizer. Nat. Rev. Neurosci. 2, 99-108 (2001). 34 Lu, J. et al. Generation of serotonin neurons from human pluripotent stem cells. Nat.

Biotechnol. 34, 89-94 (2016). 35. Philippidou, P. & Dasen, J. S. Hox Genes: Choreographers in Neural Development,

Architects of Circuit Organization. Neuron 80, 12-34 (2013). 36. Schulte, D. & Geerts, D. MEIS transcription factors in development and disease.

Development 146, dev174706 (2019). 37. Skromne, |, Thorsen, D., Hale, M., Prince, V. E. & Ho, R. K. Repression of the hindbrain developmental program by Cdx factors is required for the specification of the vertebrate spinal cord. Development 134, 2147-2158 (2007). 38. Fleck, J. S. et al. Resolving organoid brain region identities by mapping single-cell genomic data to reference atlases. Cell Stem Cell 28, 1148-1159.e8 (2021). 39. Fan, X. et al. Single-cell transcriptome analysis reveals cell lineage specification in temporal-spatial patterns in human cortical development. Sci Adv 6, eaaz2978 (2020). 40. Harutyunyan, A. S. et al. H3K27M induces defective chromatin spread of PRC2-mediated repressive H3K27me2/me3 and is essential for glioma tumorigenesis. Nat. Commun. 10, 1262 (2019). 41. Mackay, A. et al. Integrated Molecular Meta-Analysis of 1,000 Pediatric High-Grade and

Diffuse Intrinsic Pontine Glioma. Cancer Cell 32, 520-537.e5 (2017).

42. Schulte, J. D. ef al. Clinical, radiologic, and genetic characteristics of histone H3 K27M- mutant diffuse midline gliomas in adults. Neuro-Oncol. Adv. 2, vdaa142 (2020). 43. Khuong-Quang, D.-A. et al. K27M mutation in histone H3.3 defines clinically and biologically distinct subgroups of pediatric diffuse intrinsic pontine gliomas. Acta Neuropathol. 124, 439-447 (2012). 44. Ramos, S. |. ef al. An atlas of late prenatal human neurodevelopment resolved by single- nucleus transcriptomics. Nat. Commun. 13, 7671 (2022). 45. Gillen, A. E. et al. Single-Cell RNA Sequencing of Childhood Ependymoma Reveals

Neoplastic Cell Subpopulations That Impact Molecular Classification and Etiology. Cell Rep. 32, 108023 (2020). 46. Mariet, C. et al. Posterior fossa ependymoma H3 K27-mutant: an integrated radiological and histomolecular tumor analysis. Acta Neuropathol. Commun. 10, 137 (2022). 47. Braun, E. et al. Comprehensive cell atlas of the first-trimester developing human brain.

Science 382, eadf1226 (2023). 48. Gao, J. et al. Cellular Localization of Aquaporin-1 in the Human and Mouse Trigeminal

Systems. PLoS ONE 7, e46379 (2012). 49. Bandler, R. C. et al. Single-cell delineation of lineage and genetic identity in the mouse brain. Nature 601, 404409 (2022). 50. Wang, B. et al. The Role of the Transcription Factor EGR1 in Cancer. Front. Oncol. 11, 642547 (2021). 51. Kotliar, D. et al. Identifying gene expression programs of cell-type identity and cellular activity with single-cell RNA-Seq. eLife 8, 43803 (2019). 52. Chan, J. M. et al. Signatures of plasticity, metastasis, and immunosuppression in an atlas of human small cell lung cancer. Cancer Cell 39, 1479-1496.e18 (2021). 53. Stein-O’Brien, G. L. et al. Enter the Matrix: Factorization Uncovers Knowledge from Omics.

Trends Genet. 34, 790-805 (2018). 54. Kriegstein, A. & Alvarez-Buylla, A. The Glial Nature of Embryonic and Adult Neural Stem

Cells. Annu. Rev. Neurosci. 32, 149-184 (2009). 55. Arnold, S. J., Hofmann, U. K., Bikoff, E. K. & Robertson, E. J. Pivotal roles for eomesodermin during axis formation, epithelium-to-mesenchyme transition and endoderm specification in the mouse. Development 135, 501-511 (2008). 56. Vong, K. |, Leung, C. K. Y., Behringer, R. R. & Kwan, K. M. Sox9 is critical for suppression of neurogenesis but not initiation of gliogenesis in the cerebellum. Mol. Brain 8, 25 (2015). 57. Jenkins, E., Whitehead, T., Fellermeyer, M., Davis, S. J. & Sharma, S. The current state and future of T-cell exhaustion research. Oxf. Open Immunol. 4, iqad006 (2023). 58. Trefny, M. P. et al. Deletion of SNX9 alleviates CD8 T cell exhaustion for effective cellular cancer immunotherapy. Nat. Commun. 14, 86 (2023).

59. Vardhana, S. A. et al. Impaired mitochondrial oxidative phosphorylation limits the self- renewal of T cells exposed to persistent antigen. Nat Immunol 21, 1022-1033 (2020). 60. Chu, Y. et al. Pan-cancer T cell atlas links a cellular stress response state to immunotherapy resistance. Nat. Med. 29, 1550-1562 (2023). 61. Zheng, L. et al. Pan-cancer single-cell landscape of tumor-infiltrating T cells. Science 374, abeb474 (2021). 62. Szabo, P. A. et al. Single-cell transcriptomics of human T cells reveals tissue and activation signatures in health and disease. Nat. Commun. 10, 4706 (2019). 63. Anderson, A. C., Joller, N. & Kuchroo, V. K. Lag-3, Tim-3, and TIGIT: Co-inhibitory

Receptors with Specialized Functions in Immune Regulation. Immunity 44, 989-1004 (2016). 84. Tinoco, R. et al. PSGL-1 Is an Immune Checkpoint Regulator that Promotes T Cell

Exhaustion. Immunity 44, 1190-1203 (2016). 65. Shin, H. et al. A Role for the Transcriptional Repressor Blimp-1 in CD8+ T Cell Exhaustion during Chronic Viral Infection. Immunity 31, 309-320 (2009). 66. Chow, A., Perica, K., Klebanoff, C. A. & Wolchok, J. D. Clinical implications of T cell exhaustion for cancer immunotherapy. Nat. Rev. Clin. Oncol. 19, 775-790 (2022). 67. Dekkers, J. F. ef al. Uncovering the mode of action of engineered T cells in patient cancer organoids. Nat Biotechnol 1-10 (2022) doi:10.1038/s41587-022-01397-w. 68. Albelda, S. M. CAR T cell therapy for patients with solid tumours: key lessons to learn and unlearn. Nat. Rev. Clin. Oncol. 21, 47-66 (2024). 69. Park, D. S. ef al. iPS-cell-derived microglia promote brain organoid maturation via cholesterol transfer. Nature 1-9 (2023) doi:10.1038/s41586-023-06713-1. 70. Fu, Y. et al. Heterogeneity of glial progenitor cells during the neurogenesis-to-gliogenesis switch in the developing human cerebral cortex. Cell Rep. 34, 108788 (2021). 71. Ineveld, R. L. van et al. Multispectral confocal 3D imaging of intact healthy and tumor tissue using mLSR-3D. Nat. Protoc. 17, 3028-3055 (2022). 72. Rios, A. C. et al. Intraclonal Plasticity in Mammary Tumors Revealed through Large-Scale

Single-Cell Resolution 3D Imaging. Cancer Cell 35, 618-632.e6 (2019). 73. Cai, R. et al. Whole-mouse clearing and imaging at the cellular level with vDISCO. Nat.

Protoc. 18, 1197-1242 (2023). 74. Love, M. |, Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for

RNA-seq data with DESeq2. Genome Biol. 15, 550 (2014). 75. Nido, G. S. et al. Common gene expression signatures in Parkinson's disease are driven by changes in cell composition. Acta Neuropathol. Commun. 8, 55 (2020). 76. Andersch, L. et al. CD171- and GD2-specific CAR-T cells potently target retinoblastoma cells in preclinical in vitro testing. BMC Cancer 19, 895 (2019). 77. Marcu-Malina, V. et al. Redirecting aBT cells against cancer cells by transfer of a broadly tumor-reactive yòT-cell receptor. Blood 118, 50-59 (2011).

78. Germain, P.-L., Lun, A., Meixide, C.

G., Macnair, W. & Robinson, M.

D.

Doublet identification in single-cell sequencing data using scDblFinder.

F7000Research 10, 979 (2021). 79. Hao, Y. et al.

Integrated analysis of multimodal single-cell data.

Cell 184, 3573-3587.e29 (2021).

80. Mclnnes, L., Healy, J. & Melville, J.

UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction. arXiv (2018). 81. DeTomaso, D. et al.

Functional interpretation of single cell similarity maps.

Nat.

Commun. 10, 4376 (2019).

NL P382144NL Prinses Maxima Centrum voor Kinderoncologie B.V Prinses Maxima

Centrum voor Kinderoncologie B.V Brain Orgnoid 13 18 DNA PAT source 1..18 mol_type other DNA organism synthetic construct cttctcgttggggtcttt 22 DNA PAT source 1..22 mol_type other DNA organism synthetic construct ctacacgacgctcttccgatct 15 DNA PAT source 1..15 mol_type other DNA organism synthetic construct ggttgccagatgtca 15 DNA PAT source 1..15 mol_type other DNA organism synthetic construct tgtctttcctgccag 15 DNA PAT source 1..15 mol_type other DNA organism synthetic construct ctcctctgcaattac 15 DNA PAT source 1..15 mol_type other DNA organism synthetic construct attgacccgcgttag 15 DNA PAT source 1..15 mol_type other DNA organism synthetic construct taacgaccagccata 15

DNA PAT source 1..15 mol_type other DNA organism synthetic construct ctgtatgtccgattg 15 DNA PAT source 1..15 mol_type other DNA organism synthetic construct gtcaactctttagcg 15 DNA PAT source 1..15 mol_type other DNA organism synthetic construct agtaagttcagcgta 15 DNA PAT source 1..15 mol_type other DNA organism synthetic construct tgatggcctattggg 15 DNA PAT source 1..15 mol_type other DNA organism synthetic construct cagtagtcacggtca 15 DNA PAT source 1..15 mol_type other DNA organism synthetic construct aagtatcgtttcgca

Claims

Conclusions

1. A method for producing a hindbrain organoid, the method comprising:

a. culturing pluripotent stem cells in an initial culture medium for an initial period of time so as to produce an embryoid body comprising the pluripotent stem cells;

b. culturing the embryoid body, after the initial period of time, under conditions positive for proliferation and at least partial differentiation of the pluripotent stem cells in a first culture medium for a first period of time to produce an organoid comprising neuroectodermal cells, the first culture medium comprising: FGF2; a bone morphogenetic protein (BMP) pathway inhibitor; a TGF-β inhibitor; and a WNT activator;

c. culturing the neuroectodermal cells in a second culture medium and for a second time period, after the first period of time, to produce dedicated neuroectodermal cells, wherein the second culture medium comprises: FGF4 at a concentration of about 10 ng/µl; retinoic acid at a concentration of about 10 µM; purmorphamine at a concentration of about 1 µM; dorsomorphin at a concentration of about 1 µM; SB431542 at a concentration of about 10 µM; and CHIR99021 at a concentration of about 3 µM;

d. after the second time period, culturing the reserved neuroectodermal cells in a third culture medium and for a third time period, the third culture medium comprising: FGF4; retinoic acid; and a sonic hedgehog activator;

e. maturing the organoid in a fourth culture medium after the third time period to provide a hindbrain organoid, wherein the fourth culture medium comprises a maturation medium.

2. A method for producing a hindbrain organoid comprising healthy and dead tissue for modeling brain tissue, the method comprising:

a. culturing pluripotent stem cells in an initial culture medium (day 0 medium) and for an initial period of time, so as to produce an embryoid body comprising the pluripotent stem cells;

b. culturing the embryoid body, after the initial period of time, under conditions conducive to proliferation and at least partial differentiation of the pluripotent stem cells in a first culture medium and for a first period of time to produce neuroectodermal cells to form an organoid comprising the neuroectodermal cells, wherein the first culture medium (WEEK 1 medium) comprises: FGF2; a bone morphogenetic protein (BMP) pathway inhibitor; a TGF-β inhibitor; and a WNT activator;

c. culturing the neuroectodermal cells, after the first time period, in a second culture medium and for a second time period to produce dedicated neuroectodermal cells, wherein the second culture medium (WEEK 2 medium) comprises: FGF4 at a concentration of about 10 ng/µl; retinoic acid at a concentration of about 10 µM; purmorphamine at a concentration of about 1 µM; dorsomorphin at a concentration of about 1 µM; SB431542 at a concentration of about 10 µM; and CHIR99021 at a concentration of about 3 µM;

d. during the second time period, inducing a disease state in one or more of the reserved neuroectodermal cells, so as to produce an organoid comprising healthy and deceased tissue;

e. culturing the reserved neuroectodermal cells in a third culture medium for a third period of time, wherein the third culture medium (WEEK 3 medium) comprises: FGF4; retinoic acid; and a sonic “hedgehog” activator;

f. maturing the organoid in a fourth culture medium after the third time period to provide a hindbrain organoid comprising healthy and deceased tissue, wherein the fourth culture medium comprises a maturation medium.

The method of claim 1 or 2, wherein step (e) of claim 1 or step (f) of claim 2 comprises maturing the hindbrain organoid in the maturation medium from day 21 of step (a) of claim 1 or 2.

A method according to any preceding claim, wherein the method further comprises maintaining the mature organoid in the maturation medium for at least 30 days, optionally maintaining it therein for about 1 year or more.

The method of any one of claims 2 to 4, wherein the dead tissue comprises cancerous tissue, neurodegenerative tissue, and/or malformed tissue, optionally wherein the cancerous tissue is diffuse midline glioma (DMG) tissue.

The method of any preceding claim, wherein the hindbrain organoid comprises a pontine organoid.

The method of any preceding claim, wherein the initial culture medium comprises: FGF2 at a concentration of about 4 ng/µl; and Y-27632 at a concentration of about 10 HM.

The method of any preceding claim, wherein the first culture medium comprises: FGF2 at a concentration of up to about 50 ng/pul; dorsomorphin at a concentration of about 1 µM; SB431542 at a concentration of about 10 µM; and CHIR 99021 at a concentration of about 3 µM.

The method of any preceding claim, wherein the third culture medium comprises: FGF4 at a concentration of about 10 ng/µl; retinoic acid at a concentration of about 10 µM; and purmorphamine at a concentration of about 1 µM.

The method of any preceding claim, wherein the first, second, and third media each further comprise: neurobasal medium; advanced DMEM/F-12 medium; 1xN2 supplement; and heparin solution at a concentration of at least 2 pg/ml.

The method of any preceding claim, wherein the initial culture medium further comprises: neurobasal medium, advanced DMEM/F-12 medium, and an L-glutamine supplement.

12. A method according to any preceding claim, wherein:

i. the initial time period is 2 days or 48 hours;

il. the first time period starts on day 2 of step (a) and runs until day 7 of step (a) according to claim 1 or 2;

ii. the second time period starts on day 7 of step (a) and runs until day 14 of step (a) according to claim 1 or 2; and/or

IV. the third time period starts on day 14 of step (a) and runs until day 21 of step (a) according to claim 1 or 2.

The method of any preceding claim, wherein the culturing in step (d) of claim 1, or in step (e) of claim 2, further comprises culturing the organoid with agitation at about 16 days from step (a).

The method of any one of claims 12 to 13, wherein inducing a disease state comprises: mutating one or more disease-associated genes of reserved neuroectodermal cells; and/or providing the reserved neuroectodermal cells with one or more of: one or more disease-associated genes; one or more disease-associated proteins; one or more gene editing systems for mutating one or more disease-associated genes; and/or one or more interfering nucleic acid molecules.

The method of claim 14, wherein the one or more genes are selected from p53, PDGFRA, and histone H3.

The method of claim 13, wherein the mutations comprise PDGFRA-D842V and H3K27M; and/or the disease-associated proteins comprise DNp53, PDGFRA-D842V, and H3K27M.

The method of any one of claims 14 to 16, wherein inducing a disease state comprises:

a. introducing one or more nucleic acid vectors encoding the disease-associated genes, disease-associated proteins, gene editing systems for mutating one or more disease-associated genes, and/or interfering nucleic acid molecules into one or more of the reserved neuroectodermal cells at a specified time point; optionally, wherein the specified time points fall on day 11 from step (a).

18. A method according to any preceding claim, wherein:

a. the neuroectodermal cells and/or the reserved neuroectodermal cells comprise cells expressing one or more of HOXB1, GBX2, MEIS1, MEIS2, and/or MEIS3, and this from about 7 days from step (a) according to claim 1 or 2;

b. the neuroectodermal cells and/or the reserved neuroectodermal cells comprise cells that do not express any of OTX2 and/or spinal cord-specific CDX genes from about 7 days after step (a) of claim 1 or 2; and/or c. the hindbrain organoid cells comprise cells that express one or more of TPH2, GFAP, AQP1, AQP4, OLIG1, PDGFRA, OLIG2, NRG3, NRXN1, GRIA2, RBFOX1, MAP2, ERB4, PLCG2, VIM, SOX2, SOX10, SOX9, SOX4, CLU, BCAN, NCKAP5, PPP2R2B, NTN1, RMST, STMN2 and/or SLIT2; optionally from about at least 7 to about at least 30 days after step (a) of claim 1 or 2.

The method of any preceding claim, wherein the hindbrain organoid comprises one or more of: astrocytes; oligodendrocytes; glioblasts; radial glial cells; axon-guiding neuroepithelial cells; optionally choroid plexus cells and/or ependymal cells; stromal cells; and/or neurons, optionally comprising one or more of hindbrain-specific serotonergic neurons, excitatory neurons, inhibitory neurons, and/or dopaminergic neurons.

20. The method of any one of claims 2 to 18, wherein the hindbrain organoid comprises one or more cancer cells; optionally wherein the cancer cells comprise one or more of: astrocyte-like cells (AC-like cells), mesenchymal-like cells (MES-like cells), oligodendrocyte-like cells, neural stem cell-like cells, oligodendrocyte precursor-like cells (OPC-like cells), and/or cycling cells.

The method of claim 20, wherein the OPC-like cells express CRAPB1, OLIG2, and/or OLIG1, the MES-like cells express VIM and/or TIMP1, the cycling cells express TOP2A and/or MKI67, the neural stem cell-like cells express STMN2, and/or the AC-like cells express AQP1 and/or AQP4.

22. The method of any preceding claim, wherein the method further comprises culturing one or more immunological components together with the embryoid body, the neuroectodermal cells, the reserved neuroectodermal cells, and/or the hindbrain organoid, so as to provide a mature organoid with an immune microenvironment.

The method of claim 22, wherein the one or more immunological components comprise endothelial cells, myeloid progenitor cells and/or microglia cells.

The method of claim 22 or 23, wherein the one or more immunological components further comprise(s) induced pluripotent stem cell-derived T cells, optionally wherein the induced pluripotent stem cell-derived T cells are derived from a patient.

25. Organoid produced by a method according to any one of the preceding claims.

The organoid of claim 25, wherein the organoid comprises cells that:

The organoid of claims 25 or 26, wherein the organoid is produced by a method according to any one of claims 2 to 24, wherein the organoid comprises one or more cancer cells; optionally wherein the cancer cells comprise one or more of: astrocyte-like cells, mesenchymal-like cells, oligodendrocyte-like cells, neural stem cell-like cells, oligodendrocyte precursor-like cells, and/or cycling cells.

28. A method for testing one or more therapeutic agents, the method comprising:

a. providing an organoid according to any one of claims 25 to 27,

b. contacting the organoid with at least one therapeutic agent after maturing the hindbrain organoid in the maturation medium and for at least about 30 days;

c. detecting one or more changes in the organoid;

d. determining the effects of the therapeutic agent, based on the absence or presence of one or more changes.

The method of claim 28, wherein contacting the organoid comprises contacting after about 30 days.

The method of claim 29, wherein the therapeutic agent comprises an anticancer agent.

The method of claim 30, wherein the anticancer agent comprises a T cell therapy; optionally selected from tumor infiltrating lymphocytes (TIL) or a chimeric antigen receptor T cell (CAR T cell).

32. The method of any one of claims 28 to 31, wherein the one or more alterations comprise: death or survival of cells of the organoid; transcriptional alterations; epigenetic alterations; protein alterations; metabolic alterations; genomic alterations; post-translational protein alterations; and/or phenotypic alterations.

The method of any one of claims 28 to 32, wherein the organoid comprises cancer tissue, and wherein the alterations comprise the amount of cancer tissue.

The method of claims 31 to 33, wherein the method further comprises analyzing the T cells after contacting the organoid.

The method of claim 34, wherein the analyzing comprises determining: an exhaustion profile; behavioral changes; transcriptional changes; epigenetic changes; metabolic changes; genomic changes; post-translational protein changes; and/or phenotypic changes.

36. The method of claim 35, wherein the method further comprises determining one or more cytotoxic T cell markers of the T cells based on the exhaustion profile.

37. Use of an organoid according to any one of claims 25 to 27 for developing drugs, for determining their efficacy, and/or for toxicity studies.

38. Culture medium for modeling a hindbrain organoid, comprising:

i. FGF2 at a concentration of approximately 50 ng/µl;

il. dorsomorphine at a concentration of approximately 1 μM;

il. SB431542 at a concentration of approximately 10 uM; and

IV. CHIRS9021 at a concentration of approximately 3 pM.

39. Culture medium for modeling a hindbrain organoid, comprising:

i. FGF4 at a concentration of approximately 10 ng/ul;

il. retinoic acid at a concentration of approximately 10 μM;

ii. purmorphamine at a concentration of approximately 1 μM;

IV. dorsomorphine at a concentration of approximately 1 μM;

Vv. SB431542 at a concentration of approximately 10 uM; and

Vi. CHIR99021 at a concentration of approximately 3 HM.

40. Culture medium for modeling a hindbrain organoid, comprising:

i. FGF4 at a concentration of approximately 10 ng/ul;

11. retinoic acid at a concentration of approximately 10 μM; and 12. purmorphamine at a concentration of approximately 1 μM.

41. Kit of components for producing a brainstem organoid, comprising:

a. a first culture medium according to claim 38;

b. a second culture medium according to claim 39; and c. a third culture medium according to claim 40.

42. A kit of parts as claimed in claim 41, wherein the kit further comprises one or more of:

a. one or more pluripotent stem cells;

b. an initial culture medium comprising:

i. FGF2 at a concentration of approximately 4 ng/l; and ii. Y-27632 at a concentration of approximately 10 μM;

c. one or more nucleic acid vectors encoding one or more disease-associated genes, disease-associated proteins, one or more interfering nucleic acid molecules, and/or gene editing systems for mutating disease-associated genes;

d. a maturation medium; and/or e. instructions for use of the kit.

The culture medium of any one of claims 38 to 40, or the kit of parts of claim 41 or 42, wherein the first, second, and third media further comprise: neurobasal medium and Advanced DMEM/F-12 medium in a 1:1 ratio; an L-glutamine supplement; 1xN2 supplement; and a heparin solution at a concentration of about 2 µg/ml.