High-Resolution Whole-Genome Copy Number Profiling Identifies High-Risk and Low-Risk Patients with Mantle Cell Lymphoma, a Result of the Lyma-Genomic Project Conducted on Behalf the Lysa Group

Abstract

Abstract Introduction: Prognosis markers available at diagnosis are needed to discriminate high-risk (HR) from low-risk (LR) mantle cell lymphoma (MCL) patients (Delfau-Larue et al. Blood 2005; Balasubramanian et al. ASH 2014 abstract 78). In the present work, we report a whole-genome copy number analysis performed with a new technical approach. Samples from ninety-six young MCL patients treated in the phase III LyMa trial (Le Gouill et al. ASH 2014) have been investigated. Methods: Samples were selected according to material availability and patient's outcome. The cohort included 9 HR patients with primary refractory disease or relapse within one year post-diagnosis and 87 patients still in response one year after diagnosis, including 64 LR patients who were still in complete remission more than 30 months after diagnosis. Lymph node biopsies collected at diagnosis, formalin-fixed and paraffin-embedded were used to extract DNA, even when highly degraded. Both whole-genome copy number profiling and the most frequent somatic mutations of TP53 were analyzed with 50 ng of genomic DNA using the OncoScan® FFPE Assay, a new robust and validated single nucleotide polymorphism (SNP) array (Foster et al. BMC Med Genomics 2015). This assay uses the Molecular Inversion Probe technology (MIP) optimized for highly degraded FFPE samples. The ~200000 probes allowed for the detection of genome-wide copy number alterations (CNAs) with a higher concentration in cancer-related genes. The frequency and prognosis impact of CNAs were evaluated. Results and discussion: Overall, 68 recurrently altered regions were observed in 98% of patients. Deletions were more frequent than amplifications, at 9 vs 3 by patient respectively. Recurrent CNAs included losses at 1p21 (43%), 11q22 (ATM) (40%), 13q14 (24%), 9q22-31 (CDKN2A/CDKN2B) (25%), 13q33-34 (RB1) (21%), 8p11 (18%), 17p13 (TP53) (17%) and gains at 3q26-27 (35%), 3q21 (27%), 10q11 (13%) 15q11 (13%), 11q13 (CCND1) (12%), 13q31 (mir-17-92) (11%), 7p22 (CARD11) (10%), 10p12 (BMI1) (9%), 8q24 (MYC) (8%) and 12q13 (CDK4) (7%). TP53 mutations were detected in 5 patients including two with 17p13 deletion and showed a trend to be more frequent in the HR group vs LR (22% vs 3%; p=0.07). Deletions of TP53 (44% vs 14%; p=0.04), CDKN2A (67% vs 29%; p=0.054) and 8p11 (89% vs 24%; p=0.0002) were more frequent in the HR. The CDK4 (33% vs 6%; p=0.03) and mir-17-92 (44% vs 9%; p=0.01) loci were more frequently amplified in HR patients. Amplification of the miR-17-92 locus could explain why miR-17-92 overexpression, a PI3K/AKT pathway regulator, was associated with a worse prognosis in MCL (Roisman et al. Genes Chromosomes Cancer 2016). In contrast, amplification of the CARD11 locus was associated with LR (16% vs 0%; p=0.03). Conclusion: This study confirms the poor prognostic impact of TP53 alterations and reveals new CNAs associated with HR MCL such as 8p11 deletion and mir-17-92 locus amplification. Conversely CARD11 amplification appears to be associated with LR and absent from HR patients. These findings provide important clues for future theranostic-driven therapies in MCL. Disclosures Hermine: Alexion: Research Funding; Novartis: Research Funding; Celgene: Research Funding; AB science: Consultancy, Equity Ownership, Membership on an entity's Board of Directors or advisory committees, Patents & Royalties, Research Funding, Speakers Bureau.