Cargando…

MDB-12. TRANS-SPECIES ANALYSIS OF REPLICATION-REPAIR DEFICIENT (RRD) MEDULLOBLASTOMA AND RESPONSE TO IMMUNE-CHECKPOINT INHIBITION: AN IRRDC REPORT

BACKGROUND: Replication-repair deficiency (RRD) stemming from mismatch repair and polymerase-proofreading gene defects (MMRD/PPD) lead to hypermutant childhood cancers, most frequently brain tumors. While medulloblastoma is reported, the clinical, genomic and immune landscape remains unknown. METHOD...

Descripción completa

Detalles Bibliográficos
Autores principales: Das, Anirban, Fernandez, Nicholas R, Levine, Adrian, Smith, Kyle, Wang, Evan, Galati, Melissa, Aamir, Zoya, Chung, Jill, Negm, Logine, Crump, Owen, Trinh, Quang, Nunes, Nuno M, Bianchi, Vanessa, Stengs, Lucie, Edwards, Melissa, Stein, Lincoln, Bouffet, Eric, Taylor, Michael, Northcott, Paul, Ramaswamy, Vijay, Hawkins, Cynthia, Tabori, Uri
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Oxford University Press 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10260105/
http://dx.doi.org/10.1093/neuonc/noad073.245
Descripción
Sumario:BACKGROUND: Replication-repair deficiency (RRD) stemming from mismatch repair and polymerase-proofreading gene defects (MMRD/PPD) lead to hypermutant childhood cancers, most frequently brain tumors. While medulloblastoma is reported, the clinical, genomic and immune landscape remains unknown. METHODS: We analysed the genome, methylome, transcriptome (bulk, single-nuclei) and immune-microenvironment of the largest cohort of RRD-medulloblastoma patients enrolled by the International RRD Consortium, and correlated these to clinical outcomes. RRD mice-models were used to preclinically assess response to immunotherapy. RESULTS: RRD-medulloblastoma (n=40) were enriched for anaplasia (55%) and localised disease (85%). Methylation/nano-string-based subgrouping failed to classify 40%, while the remaining clustered with the SHH-subgroup. Copy-number changes were notably infrequent (<20%). All tumors harboured hypermutation and microsatellite instability in contrast to non-RRD controls (p<0.0001). Pathogenic variants were frequent in POLE/POLD1 (80%), TP53 (48%) and SHH-pathway genes (PTCH1, SUFU, SMO: 56%). Interestingly, some tumors additionally had glioma-driver alterations (ATRX, NF1: ~50%), similar to RRD cerebellar glioblastoma, but distinct from both non-RRD medulloblastoma (n=791) and hemispheric glioblastoma (n=733) controls. Moreover, although bulk-transcriptome matched non-RRD SHH-medulloblastoma, single-nuclei analyses suggested an earlier cell-of-origin potentially explaining the heterogenous phenotypes resulting from the driver mutations. Uniquely, immune analyses (gene expression and immunohistochemistry) demonstrated high intra-tumoral CD8 T-cell infiltration. Three-year progression-free survival was 60%. Outcome was worse for RRD-medulloblastoma harbouring TP53-mutation (p=0.04) and failing subgroup-classification (p=0.004). Nestin/Cre MSH2-/- POLE;p.S459F mice developing medulloblastoma recapitulated human disease and demonstrated response to anti-PD1 monotherapy. Recurrent/progressive human tumors including those harbouring TP53-mutations exhibited radiological responses to anti-PD1 monotherapy, which was associated with prolonged ongoing survival in comparison to those not treated with checkpoint-inhibitors (p=0.02). CONCLUSIONS: Hypermutant RRD-medulloblastoma reveal heterogenous phenotypes due to unique spectrum of driver mutations and an early cell of origin. Their immune-hot microenvironment allows successful salvage treatment with checkpoint-inhibitors for those failing chemo-radiation, including the high-risk SHH/TP53-mutant and unclassified subgroups.