MONITORING MINIMAL RESIDUAL DISEASE IN MULTIPLE MYELOMA PATIENTS BY NGS APPROACHES

Novel treatments for multiple myeloma (MM) have increased rates of complete response (CR) raising interest in more accurate methods to evaluate residual disease. Cell-free tumor DNA (cfDNA) analysis could represent a minimally invasive approach complementary to multiparameter flow cytometry (MFC) and molecular methods on bone marrow (BM) aspirates. A sequencing approach using the Ion Torrent Personal Genome Machine was applied to identify clonal immunoglobulin heavy chain (IGH) gene rearrangements in tumor plasma cells (PCs) and in serial plasma samples of 25 MM patients receiving second-line therapy. The same clonal IGH rearrangement identified in tumor PCs was detected in paired plasma samples and levels of IGH cfDNA correlated with outcome and mirrored tumor dynamics evaluated using conventional laboratory parameters. In addition, IGH cfDNA levels reflected the number of PCs enumerated by MFC immunophenotyping even in the CR context. Minimal residual disease (MRD)-negative patients by MFC were characterized by low frequencies of tumor clonotypes in cfDNA and longer survival. Despite the limited sample size, results showed that our developed deep-sequencing workflow is feasible for the identification and the monitoring of IGH gene rearrangements in tumor samples and in serial plasma samples of patients affected by MM. Moreover, since our method is based on a straightforward pipeline tailored for the analysis of the monoclonal B-cell expansion, we decided to evaluate the possibility of applying our developed NGS workflow in the non-invasive monitoring of other mature B-cell neoplasms such as diffuse large B-cell lymphomas (DLBCLs) and Hodgkin lymphomas (HLs) that have very rarely circulating tumor cells, but release tumor DNA in the bloodstream. Profiling the IGH repertoire of 26 newly diagnosed DLBCL patients we demonstrated that our NGS workflow allows the identification of clonal IGH rearrangements in archival formalin-fixed paraffin-embedded (FFPE) tissue biopsies and paired cfDNA samples. IGH cfDNA frequencies reflected levels of DLBCL burden detected by imaging modalities at diagnosis and during the course of treatment. In HL cases (n=20), data obtained applying a reverse approach and starting with the analysis of cfDNA to overcome the limitations imposed by the paucity of tumor cells in tissue biopsies, suggested that methods to select pure HL tumor cells are mandatory to exclude inflammatory contamination and to identify clonal IGH rearrangements in HL. Overall, results I have obtained during my PhD project confirm the clinical utility of IGH cfDNA in the evaluation of tumor burden and response to treatment in mature B-cell malignancies, and support the application of the developed NGS workflow in large prospective studies that will validate the feasibility in the clinical setting and the prognostic value of disease monitoring through the quantitative assessment of plasma clonotypic IGH gene rearrangements.