Genetic Profiling and Novel Recurrent Chromosomal Alterations in Patients with Light Chain Amyloidosis

Introduction

High-throughput sequencing studies have rendered seminal knowledge in monoclonal gammopathies such as multiple myeloma (MM) and Waldenström's macroglobulinemia (WM). Unfortunately, the low incidence of AL amyloidosis and its typically low tumor burden, often masked by a polyclonal plasma cell (PC) background, account for the limited information on its tumor cell biology. Thus, it remains unknown if AL amyloidosis harbors a unifying mutation as occurs in WM or if, in its absence, there are recurrent mutations and if these overlap with those observed in MM. With this background , the aim of this study is to perform a whole exome sequencing (WES) in a series of patients with AL amyloidosis and to compare mutational profiles in AL amyloidosis vs MM and analyze the copy number variation in this series of patients.

Methods

A total of 27 patients with confirmed diagnosis of AL were included. WES was performed in 56 paired samples of FACSorted bone marrow tumor plasma cells and peripheral blood mononucleated cells. Each tumor sample was captured in triplicate using Agilent's SureSelect Human All Exon V6 + UTR kit and sequenced on the Illumina NextSeq 500 platform. Data was analyzed with Strelka software to discard germinal mutations, ANNOVAR for functional annotation, and a data reduction strategy to identify candidate variants. The mutational signature was analyzed with Mutational Signatures in Cancer (MuSiCa) software. We used the MMRF CoMMpass dataset (895 patients) to compare the mutational landscape of MM vs AL. We also determined immunoglobulin gene rearrangements in AL by next generation sequencing. Besides, we analyzed the copy number variation (CNV) with CNVkit program.

Results

The mean depth coverage for control and tumor samples was 64x and 186x, respectively. A total of 1983 somatic SNV and 133 INDEL were identified, with an average of 71 (20-281) SNV and 5 (0-25) INDEL per patient. Overall, the most frequently mutated genes in this series were IGLL5 and MUC16 (recurrence of 17% each). When compared to MM (average of 66 SNV and 2,5 INDEL), we observed a similar mutational load. However, none of the most frequently mutated genes in MM (i.e. KRAS, NRAS, FAM46C, BRAF, TP53, DIS3, PRDM1, SP140, RGR1, TRAF3, ATM,CCND1, HISTH1E, LTB, IRF4, FGFR3,RB1, ACTG1, CYLD, MAX, ATR) were recurrently mutated in patients with AL. The only genes commonly mutated in AL amyloidosis and MM were MUC16 (recurrence of 17% and 8%, respectively) and IGLL5 (recurrence of 17% each).Most patients with AL harbored between 1 and 8 mutational signatures, implying that multiple mutational processes are operative. The most frequent mutational signature were (signatures 6, 15 and 20) associated with mismatch repair protein deficiency (MMR) and high microsatellite instability (93%), mutational signature 2 (89%), related with the aberrant activity of APOBECs, a family of proteins that enzymatically modify single-stranded DNA and mutational signature 1 (81%), profile that appear in all types of cancers and has been correlated with the age of cancer diagnosis. The signature 2 is also representative of MM. Regarding the immunoglobulin gene repertoire, we noted that 26% of patients with AL harbored more than one clone; this extent in clonal heterogeneity being similar to that found in MM (23%).The most frequent IGH gene involved was IGHV3-30 in both AL (recurrence of 10%) and MM (recurrence of 12%).Regarding CNV, recurrent gains included chromosomes 1q (29%), 5 (38%), 6p (14%), 7 (43%), 9 (43%), 15 (24%), 18 (14%) and 19 (43%). Recurrent losses affected chromosome 13 (33%), 6q (14%) and 16q (19%).

Conclusions

This is the first WES study performed in a series of patients with AL. We demonstrated the lack of a common driver mutation in this disease and unveiled that recurrently mutated genes in AL amyloidosis do not overlap with those observed in MM. We also confirm the existence of numerous chromosomal alterations in patients with AL. The frequencies of aberrations and alterations detected by NGS are comparable with those describe in previous studies by copy number array analysis, but here we show some novel recurrent chromosomal aberrations as gain of chromosome 7 (43%) and losses of chromosome 18 (14%). Overall, these results may have significant impact in our understanding of the pathogenesis of AL amyloidosis and its differential diagnosis vs other monoclonal gammopathies.