Genetic Characterization of Waldenstrom Macroglobulinemia By Next Generation Sequencing: An Analysis of Fouteen Genes in a Series of 61 Patients

Background: Waldenström's macroglobulinemia (WM) is a rare immunoproliferative neoplasia with indolent characteristics that shows important variability, involving three different stages of presentation: IgM Monoclonal Gammopathy of Undetermined Significance (IgM-MGUS), asymptomatic WM (AWM), and Symptomatic WM (SWM). Whole-genome sequencing and some specific approaches have identified MYD88 L265P (90%) and CXCR4 (29%) mutations as the most recurrent somatic mutations in WM. However, other genetic abnormalities under as well as the mechanisms responsible for this clinical heterogeneity still remain to be clarified. Therefore, our aim was to analyze the genomic landscape of WM, distinguishing between the three stages of the disease, by using a targeted next generation sequencing (NGS) strategy.

Methods: In this study, we performed a comprehensive mutation analysis of genes previously described as frequently involved in Waldenstrom Macroglobulinemia in a large and well characterized cohort of WM patients with the aim to dissect relationships between genotype and clinical and biological characteristics to integrate somatic mutations into a clinical/molecular prognostic model. Twelve genes of interest (ARID1A, CD79A, CD79B, TP53, MYBBP1A, TRAF2, TRAF3, RAG2, HIST1H1B, HIST1H1C, HIST1H1D, and HIST1H1E) were analyzed by high throughput sequencing (Illumina MiSeq, San Diego, CA) with a novel custom amplicon-based panel in a cohort of 61 patients (pts) diagnosed according to WHO classification as follows: 14 MGUS, 23 AWM and 24 SWM. DNA was extracted from bone marrow separated CD19+ B-cells and sequenced in a MiSeq (Illumina) using 150-bp paired-end reads and a mean depth of 2000X. Bioinformatics analysis was carried out with Illumina VariantStudio 2.2. Results were correlated with biological and clinical data of the patients. MYD88 and CXCR4 mutation status, available in all cases, was assessed by ASO-PCR and Sanger Sequencing, respectively.

Results: Apart from MYD88 L265P mutations (present in 90% of cases) and CXCR4^WHIM (21% of cases), 23 non-synonymous mutations were found, corresponding to 18/61 (30%) patients. Only one patient with MGUS demonstrated one additional mutation (7%), while seven of the AWM (30%), and 10 of the SWM (42%) demonstrated additional mutations (p<0.05 for linear association), suggesting an association between the clinical behavior of the disease and a higher number of mutations. Interestingly, patients with a wild MYD88 gene (n=6), showed no additional mutations in any of these studied genes. Genes most frequently mutated were CD79B (n=5, 8%), HIST1H1E (n=4, 7%), MYBBP1A (n=3, 5%), ARID1A and HIST1H1B (n=2, 3% for both). There were three patients who presented more than one gene mutated (TP53/CD79B; RAG2/ARID1A; HIST1H1B/HIST1H1E).

Apart from the clinical diagnosis and the requirement of therapy, no relevant correlations between the presence of mutations and the final clinical behavior was found in any patient, although the patient with a mutated TP53 corresponded to a very high resistant form of the disease. Finally, no relevant differences in progression free and overall survival were seen in this series based on the presence or absence of somatic mutations.

Conclusion: Our data reveal an increased incidence of mutations along the different steps of evolution in WM: IgM MGUS, asymptomatic WM and symptomatic WM. Thus, this would mean that in contrast to MYD88 L265P, present from the beginning of the pathogenesis, most of these mutations would be acquired during the evolution of the disease, and before therapy initiation. Finally, CD79B, which is part of the B-cell receptor pathway, was frequently mutated gene in our series, emerging as an interesting therapeutic target. This confirms the relevance of the BCR signaling pathway, reinforcing the use of biological agents blocking this pathway in the treatment of these patients.