CO67 | Secondary genetic events and their relationship to TP53 mutation in mantle cell lymphoma. A sub-study from the Fondazione Italiana Linfomi Mantle First-BIO study

Abstract

Mantle Cell Lymphoma (MCL) is an aggressive malignancy with variable clinical behavior, largely reflecting the underlying molecular heterogeneity. The genomic landscape of MCL encompasses gene mutations with strong prognostic implications (such as TP53, NOTCH1, KTDM2D, etc) and secondary genetic events, such as copy number variations (CNVs), also implicated in the pathogenesis and prognosis of MCL.
We extracted and analyzed, by targeted Next Generation Sequencing (tNGS) with a customized panel of 37 genes, the genomic DNA (gDNA) of 73 diagnostic samples of patients with relapsed/refractory MCL that were enrolled in the Fondazione Italiana Linfomi (FIL) Mantle First-BIO study (NCT04882475). CNVs were annotated with CNVkit, while the TP53 mutation was annotated with MuTec2 and validated through Integrative Genomics Viewer (IGV). The statistical analysis was conducted on R v4.1.2, adopting the time to first relapse or progression of disease (POD) as the primary outcome measure, which was previously defined as a threshold of 24 months since MCL diagnosis, enabling us to divide the cohort into early-POD24 and late-POD24. All patients also had available data for correlating CNVs with the presence of TP53 mutation.
We detected 18 different alterations (11 amplifications and 7 deletions) associated with MCL. The most recurrent CNVs were Amp 3q26-q28 (BCL6, 49%), followed by Amp 5q35.3 (HNRNPH1, 45%), Del13q14 (RB1, 34%), and Del 11q22.3-q23.2 (32%).
When comparing early and late POD groups, Fisher’s exact test identified Del 13q14 (RB1, p= 0.02, 52% early vs 24% late), Del 6q (p= 0.01, 37% early vs 11% late), and Del 9p21.3 (CDKN2A, p= 0.02, 33% early vs 11% late) more prevalently associated with early POD. Among these, the Del 9p21.3 (CDKN2A) resulted as the stronger predictor of shorter time to POD (p=0.01), independently of TP53 mutation in multivariable analysis. Unsupervised clustering identified molecularly defined clusters associated with significantly different time to POD (p= 0.01). Pairwise log-rank tests confirmed TP53 (mutated vs wild-type) as the strongest prognostic factor, with cluster assessment that improved the prognostic predictivity among patients: clusters TP53-mut vs TP53-WT, p= 0.001, HR=3.92; and p= 0.014, HR= 2.23, respectively (Figure 1).
In conclusion, CNV-based molecular clusters might represent a novel approach to identify patients at higher risk of treatment failure, contributing to the prognostic predictivity of TP53 mutation.

Acknowledgments. We thank FIL for research funding (Premio Giovani Ricercatori, Ed. 2019) and for supporting our work, the European Union - Next Generation EU - and Ministero della Salute - PNRR-TR1-2023-12378287 - Rafforzamento e potenziamento della ricerca biomedica del SSN (CUP: E13C24000610007).