Patient-derived organoids and orthotopic xenografts of primary and recurrent gliomas represent relevant patient avatars for precision oncology

Patient-based cancer models are essential tools for studying tumor biology and for the assessment of drug responses in a translational context. We report the establishment a large cohort of unique organoids and patient-derived orthotopic xenografts (PDOX) of various glioma subtypes, including gliomas with mutations in IDH1, and paired longitudinal PDOX from primary and recurrent tumors of the same patient. We show that glioma PDOXs enable long-term propagation of patient tumors and represent clinically relevant patient avatars that retain histopathological, genetic, epigenetic, and transcriptomic features of parental tumors. We find no evidence of mouse-specific clonal evolution in glioma PDOXs. Our cohort captures individual molecular genotypes for precision medicine including mutations in IDH1, ATRX, TP53, MDM2/4, amplification of EGFR, PDGFRA, MET, CDK4/6, MDM2/4, and deletion of CDKN2A/B, PTCH, and PTEN. Matched longitudinal PDOX recapitulate the limited genetic evolution of gliomas observed in patients following treatment. At the histological level, we observe increased vascularization in the rat host as compared to mice. PDOX-derived standardized glioma organoids are amenable to high-throughput drug screens that can be validated in mice. We show clinically relevant responses to temozolomide (TMZ) and to targeted treatments, such as EGFR and CDK4/6 inhibitors in (epi)genetically defined subgroups, according to MGMT promoter and EGFR/CDK status, respectively. Dianhydrogalactitol (VAL-083), a promising bifunctional alkylating agent in the current clinical trial, displayed high therapeutic efficacy, and was able to overcome TMZ resistance in glioblastoma. Our work underscores the clinical relevance of glioma organoids and PDOX models for translational research and personalized treatment studies and represents a unique publicly available resource for precision oncology. Electronic supplementary material The online version of this article (10.1007/s00401-020-02226-7) contains supplementary material, which is available to authorized users.

. Note strong aneuploidization of P3 and P13 GSC lines upon in vitro cultures (P3NS, P13NS). b Statistical analysis of array-CGH data. Chi 2 test for independence reveals limited impact of sample type, treatment and array type on genetic profiles. Individual tumor genetic aberrations and IDH1 mutation status are main sources of variation. c Examples of array-CGH profiles of GBM patient tumors and corresponding PDOX models are shown for T101 and T476. Genetic aberrations were recapitulated over serial transplantations. See Table  S2 for detailed description. d Array-CGH profiles of longitudinal samples (T347-T470) of GBM patient LIH0347 showing the same genetic aberrations upon recurrence. These profiles are largely recapitulated in PDOX models, only an additional 1p31.1-p11.2 loss was detected in PDOX T470. e Array-CGH profiles of GBM T341 patient tumor and corresponding PDOX. PDOX model was derived from an additional MDM4/CDK6-amplified clone with different chromosomal breakpoints. Right panels show presence of different amplicons in the patient tumor and corresponding PDOX. f Array-CGH profiles of GBM P8 patient tumor fragments and corresponding PDOX. Analysis of 2 tumor fragments revealed intra-tumoral genetic heterogeneity and different EGFR amplicon. g-i Array-CGH profiles of GBM patient tumors, corresponding PDOXs and in vitro GSC lines for T158 (g), P3 (h), P13 (i). Additional aberrations occurred upon in vitro passaging. Note that PDOX T158 arose from an additional MDM2-amplified, EGFR-non amplified clone, not detected in the patient sample. j In vitro passaging of T16 tumor cells as GSC line (T16NS) led to loss of EGFR amplicon (array-CGH, left panel) and decreased EGFR expression (flow cytometry, right panel).

SUPPLEMENTARY FIGURE 3
Supplementary Figure 3. Recapitulation of genetic heterogeneity in glioma PDOXs. a Recapitulation of overall variants in PDOX-derived cell lines detected by targeted sequencing. Cell lines were compared to respective PDOX models. The number of total variants detected for each cell line and PDOXs is displayed. b Digital PCR-based analysis of IDH1 wild-type and R132H fractions in IDH1mut gliomas and corresponding preclinical models. c Western blots against EGFR (antibody cocktail recognizing wild-type and structural variants) and EGFRvIII proteins in PDOX-derived organoids T347 and T470 (patient LIH0347) show wildtype EGFR protein as well as the structural variant EGFRΔ2-15, which is recognized by EGFRvIII antibody, despite decreased molecular weight. d Cellular prevalence estimates from PyClone representing clonal subpopulations detected in patient tumors and respective PDOXs. Examples shown for longitudinal samples (T347, T470) of patient LIH0347. Each cluster of mutations was computationally inferred to reflect a subclone. Number of genetic variants contributing to each clone is depicted. e Cellular prevalence estimates from PyClone representing clonal subpopulations detected in longitudinal samples of patient LIH0347 and the respective PDOXs. Each line represents a cluster of mutations computationally inferred to reflect a subclone. Only genetic variants detected in all samples were considered for analysis. f Evolutionary dynamics of EGFR genetic variants. Targeted DNA sequencing revealed longitudinal evolution of EGFR genetic variants in LIH0192 patient tumors and PDOXs derived thereof. A specific subclonal variant was present only in T192 patient tumor and was enriched in the respective PDOX (55168634 G->C). Another genetic variant present in T192 patient tumor and PDOXs at the subclonal level was selected out during disease progression (chr7:55154017 C>T). Comparison of overall read depths suggests that these variants are discordantly inherited upon tumor recurrence in patients and PDOX derivation, supposedly via unequal distribution of extrachromosomal DNA.
ANOVA based on global beta-value distributions reveals IDH mutation status as a main source of variation in the DNA methylation cohort. c Heatmap representing 1000 most variable features in DNA methylation shows genomic loci differentially methylated between IDH1mut versus IDHwt patients and preclinical models. d beta-value distributions are very similar between IDH1mut and IDHwt tumors (patient samples and PDOX models), in accordance with the G-CIMP low status of the IDH1mut gliomas. IDHwt and IDH1mut GSC lines increase DNA methylation at numerous sites corresponding to open seas, shelfs, and shores.

SUPPLEMENTARY FIGURE 5
Supplementary Figure 5. Gene expression profiles in glioma preclinical models. a Principal component analysis indicating similarity of genome-wide gene expression profiles between normal human brain, glioma patient samples, PDOXs, GSC lines (NCH421k, NCH644) and classical glioma lines (U87, U251) grown in vitro or as corresponding xenograft ('X'). Human specific arrays were applied for transcriptome analysis. b Correlation of gene expression profiles to TCGA GBM patients shows close resemblance of PDOXs to patient tumors at the transcriptomic level. c Stem-cell associated marker expression profiles were interrogated by flow cytometry in tumor cells of PDOXs over serial transplantation. Example shown for PDOX T101. d Single cell RNA-Seq of mouse brain showing overall gene expression relationship between cells of normal brain (red) and upon GBM implantation (shown for PDOX P8) (black). Identified TME subpopulations are depicted. MGMT promoter unmethylated and methylated tumors are shown in black and red respectively. c Mean AUC upon exposure to TMZ and VAL-083 in PDOX models derived from treatment-naïve and treated tumors. ns = not significant (unpaired t-test). d Blood vessels in vivo were visualized with mouse specific anti-CD31 staining (in green). Tumor was defined as nuclei dense area (nuclei in blue = DAPI). Representative pictures are shown for each experimental group (Scale bar = 100µm). Quantification of vessel number per mm 2 upon treatment and area covered by vessels confirmed normalization of the tumor vasculature in Bevacizumab (Bev) treated mice (Mean±SD, *pvalue<0.05, ***pvalue <0.01, ****pvalue <0.001, n = 5-6 for Control and Bev groups, n = 2 for VAL-083 treated groups, 3-5 pictures were taken per each tumor, unpaired t-test). The statistical analysis between Control and VAL-083 treated groups is not shown due to major differences in tumor volumes, leading to lower aberrations in tumor vasculature of the small VAL-083 treated tumors. e Representative sections of PDOX stained against human-specific Nestin (n = 3). f IHC for H2AX-P in PDOX sections (brown nuclei = H2AX-P). Counterstaining for nuclei with hematoxyline. Induction of H2AX-P was observed in VAL-083 treated tumors. Minor induction of H2AX-P was also observed in a subpopulation of normal brain cells. Scale bar = 100 µm. g Quantification of AUC upon exposure to EGFR inhibitors: Gefitinib, AG490, AZD3759 and Daphtenin (*pvalue < 0.05, unpaired t-test); wt = wildtype, mut = mutated, Amp = amplified, SV = structural variant, exp = protein expression. Experiments were performed twice with 3 technical replicates each. See Supplementary Table 8 for mean AUC +/-SEM. h Response curves (non-linear fit, n = 2) of PDOX T434 to EGFR and CDK4/6 inhibitors displayed as %-Viability +/-SEM. Similar treatment responses were observed with the two protocols applied. Chromosomal aberrations of glioma patient samples and corresponding PDOX models and GSC lines.

SUPPLEMENTARY REFERENCES
Supplementray Table 4 List of genetic variants private to patient tumors and respective PDOXs.  Table 5 Glioma specific mutations in patient tumors and preclinical models.
Mean AUC values +/-SEM are shown for each PDOX. Experiments were performed twice with 3 technical replicates each. SEM represents variation between biological replicates.