Integrative Analysis of Genetic and Tumor Microenvironment Topology to Delineate Prognostic Subtypes in Mantle Cell Lymphoma

Le résumé fourni par la source

BACKGROUND Mantle Cell Lymphoma (MCL) is a B-cell lymphoma characterized by t(11;14) leading to CyclinD1 overexpression with highly variable clinical response. Although the tumor intrinsic mechanisms have been well studied and the role of the tumor microenvironment (TME) in survival and therapeutic responses are marginally examined, understanding how distinct tumors regulate their respective immune landscapes has been constrained, by a lack of appropriate experimental techniques that recapitulate the heterogeneity of the TME in MCL. Herein, we integrated high-throughput genomics and characterized TME using a highly multiplexed, spatially resolved, single-cell digital pathology approach termed imaging mass cytometry (IMC) to delineate the tumor-immune landscape. Thus, comprehensive whole exome sequencing and TME imaging analysis demonstrated prognostic subtypes elucidating the interplay between tumor intrinsic and extrinsic factors in MCL. METHODS MCL cohort with complete clinicopathological information and treatment protocol (n=580) was utilized from the North America Mantle Cell Lymphoma Consortium. These cases have been reviewed by at least two hematopathologists to confirm the MCL diagnosis. Whole exome sequencing (WES) was performed on 179 formalin-fixed tissue embedded (FFPE) samples. 44 samples have matched microarray data. Genetic driver lesions were identified using VarScan 2.4, MutSigCV 1.3.01, and GISTIC 2.0. Formalin-fixed paraffin-embedded (FFPE) tissue tumor microarrays (TMA) consisting of 55 MCL cases with duplicate or triplicate cores were generated. To perform the TME analysis, we designed a 42 marker IMC panel which includes canonical markers for immune cell subsets and markers for two additional highly mutated/aberrated genes in MCL (ATM and TP53). TMA was stained with antibodies tagged with heavy metal isotopes, laser ablated to generate spatial digital maps of the tissue and unsupervised clustering was performed to identify immune cell types using the Phenograph algorithm. Survival analysis was performed using the Kaplan-Meier method (logrank test). RESULTS Initial genomic analysis (n=179) suggested at least two prognostic subtypes, dependent on ATM or TP53 mutation status, and the latter associated with inferior-clinical outcome. Transcriptional profiling in a subset identified enrichment of gene signatures associated with B-cell receptor (BCR), MYC Targets, apoptosis, and CCND1 upregulation in the TP53-mutant subgroup. TME analysis showed that the TP53 positive tumor cell populations exhibited a receded T cell population (CD4+ and CD8+). Interestingly, the TP53-positive tumor cells expressed high protein levels of p-STAT-3, p-NFKB, and HLA expression compared to ATM-positive tumor cells. Additionally, unsupervised analysis using NMF consensus clustering showed four patient subgroups wherein the worst survival subgroup (OS: 45 months vs 62 months, p < 0.05) was independent of TP53 mutations and highly enriched in ATM mutations, 17p11.1-13.3 deletions, and depleted in CD8 naïve T cell and Macrophages M1. Whereas the good survival group (OS: 75 months) was highly enriched in ATM and TP53 mutations but deficient in 17p deletions. CONCLUSION While most MCL studies looked at the MCL heterogeneity at the genomic level, this study shows the integration of genomic and TME factors to elucidate prognostic MCL subtypes from paraffin samples at a single cell resolution. The current analysis suggests that the TP53 mutated MCL subgroup has reduced T-cell enrichment in the TME and high protein expression of oncogenes. A patient subgroup highly enriched in ATM mutations, independent of the TP53 mutations and depleted immune inflitration showed the worst prognosis. Thus, this study suggests a new angle of ATM mutations in MCL pathogenesis which could improve risk stratification. Overall, this study may unravel distinct mechanisms of tumorigenesis with relevant clinical and biological implications.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.