P829: genomic profiling in cfdna complements molecular diagnosis in patients with multiple myeloma

Topic: 13. Myeloma and other monoclonal gammopathies - Biology & Translational Research Background: Imbalanced spatial distribution of tumor cells throughout the body is well known in multiple myeloma (MM). At least 75% of patients have been shown to harbour a spatial molecular heterogeneity related to the genomic aberrations acquired during disease development and further progression. The profile of such genomic alterations impacts the prognosis of patients. Observing the kinetics and clonal evolution in myeloma progression independently from the bone marrow (BM) sampling site would therefore allow for a comprehensive identification of molecular risk profiles. Analysis of cell-free DNA (cfDNA) is an emerging non-invasive approach to assess drug resistance, and to profile apoptotic tumor cells or tumor load, potentially overcoming limitations due to spatial distribution of the disease. Aims: Paired pre-treatment samples from BM and cfDNA were used to evaluate the complementary impact of cfDNA analysis to detect molecular genomic signatures in MM. Methods: Immunoglobuline (IG) rearrangements, structural variants, and mutations in 72 genes in paired BM aspirates and plasma samples were analyzed in 21 patients enrolled in the randomized phase 3 DSMMXVII trial (NCT03948035). BM plasma cells were enriched using CD138 magnetic bead technology. No cut-off was applied for BM plasma cell infiltration for patient selection. BM and cfDNA were analyzed using the Euroclonality-NDC assay and results were compared to cytogenetic assessments. Results: The median quantity of cfDNA before treatment did not correlate with the plasma cell infiltration of the corresponding BM sample (R² = 0.19). cfDNA levels in plasma differed significantly between patients staged to R-ISS I and R-ISS II/ III (7.8 vs 14.9 ng/ml; p=7.3E-07). Genomic profiles detected in BM and cfDNA were partially verified by data from FISH and IG-NGS, depending on the detection limits and sampling for the comparator methods. In BM, Del(13q) (52%) was the most frequent copy number alteration. Trisomy 19, Del(1p), and Del(17p) were observed in 24%, 19% and 10% of patients, respectively. Eight translocations with known translocation partners were identified with t(11;14), t(14;16), and t(4;14) in 19.0%, 14% and 5% of patients, respectively. 33 somatic mutations were detected in BM samples, with KRAS (23.8%), NRAS (24%), TRAF3 (19%), and TP53 (14%) as the most frequently mutated genes. On average two mutations were detected for each patient (range 0 (n=3) to 5). In 43% of patients (9/21) the genomic profiles were concordant in BM and cfDNA for either mutations, translocations or both. 33% (11/33) of the mutations and 63% (5/8) of the translocations were detected in the concomitant cfDNA. MYC-translocations were identified in five cfDNAs, and individual mutations of NRAS, TP53, DNMT3A, CREBBP, KMT2D and STAT5B were uniquely detected in six plasma samples. Thereby, in 43% of patients (9/21) additional, potentially myeloma-associated genomic aberrations were detected only in cfDNA. Interestingly, there was a highly significant correlation for concomitant detection of genomic aberrations to the R-ISS staging of patients, as 73% (8/11) patients in stage II/III but only 13% (1/8) patients in stage I showed tumor-specific variants in cfDNA. No obvious differences in the range of BM infiltration or tested cfDNA amount were observed among patient groups. Conclusion: Concomitant analysis of cfDNA partially mirrors the genomic profiles of enriched BM plasma cells in this limited cohort of MM patients at diagnosis. Mutation profiles detected in cfDNA reflect spatial heterogeneity of the disease, and detectability of tumor cfDNA correlates highly with advanced stage of the disease. Keywords: Multiple myeloma, Genotype, Plasma, liquid biopsy