The Extent of Intra-Clonal Genetic Diversity within the Myeloma Clone Is a Predictive Biomarker of Progression and Outcome after Treatment

Multiple myeloma (MM) is a disease characterized by the abnormal proliferation of plasma cells in the bone marrow. We and others have recently demonstrated the existence of different myeloma subclones phylogenetically related to the founding clone. This intra-clonal heterogeneity is the basis for disease progression, treatment resistance, and relapse. However, the clinical and biological relevance of the presence and diversity of different myeloma subclones has not been fully established.

In this study, we used whole exome sequencing (WES) plus a pull down of the MYC, IGH, IGL and IGK loci as a tool to analyze the largest series of presenting cases of myeloma (463 patients) to date, which were entered into the Myeloma XI trial (NCT01554852). DNA from both tumor and peripheral blood samples were used in the exome capture protocol following the SureSelect Target Enrichment System for Illumina Paired-End Sequencing Library v1.5. Exome reads were used to call single nucleotide variants (SNVs), indels, translocations, and copy number aberrations. The proportion of tumor cells containing an SNV was inferred. The presence and proportion of subclones were defined in a subset of 437 patients using a genetic algorithm based-tool (GAUCHO), which also calculated different indices of clonal diversity: number of clones, mean pairwise genetic divergence, Shannon and Inverse Simpson diversity index and Berger-Parker dominance index. Based on these results, we aimed to determine the clinical implications of the number of mutations and the subclonal diversity of MM at presentation in progression free (PFS) and overall survival (OS).

We found that MM patients with t(14;16) and t(14;20) had more exonic mutations (not including Ig variants) than the rest of samples (median 87 versus 43, p<0.001). Additionally, we found that MM patients with an APOBEC signature or with mutations in ATM/ATR had significantly more mutations than patients without these genetic lesions with a median number of 137 mutations (range 20-569) and 84.5 (range 33-319) respectively (p<0.001). Subsequently, we identified patients with high number of mutations (>59 mutations) that had a worse outcome in terms of OS (2-year OS rate of 71% (95% CI, 63-80%) vs. 82% (95% CI, 78-87%), p=0.02), but not progression free survival (median 22.5 (95% CI 18.7-30.2) vs. 27.5 (95% CI, 25.8-30.5) months, p=0.1)

We reported recurrent mutated genes in myeloma with mutations being present at both clonal and subclonal levels (IRF4, RB1, DIS3, BRAF, KRAS, and NRAS), whereas other genes were mutated only at clonal (HIST1H1E, LTB, TP53 or EGR1), or subclonal levels (CYLD, TRAF3, MAX). These results give insights about the differences in mutation acquisition times and potential subclonal fitness.

We inferred that the median number of clones present in this myeloma series was 5, and determined the prognostic value of the number and diversity of subclones in MM patients. The prognostic impact of having high number of clones was unclear as no significant differences were found. On the contrary, there was a significant difference in terms of outcome when calculating distinct measurements of subclonal diversity. Briefly, MM patients with high values of inverse Simpson diversity index had a significantly poorer PFS (median 13.2 (95% CI, 9.4-∞) vs. 26.9 months (95% CI, 24-30.2) months, p=0.02) and OS (66% (95% CI, 52-82%) vs. 81% (95% CI, 77-85%) alive at 2-years, p=0.01); and, alternatively, MM patients who did not have a dominant subclone accounting for >25% of MM cells (low values of Berger-Parker Dominance index, n=56) had a significantly shorter PFS than those with a dominant clone accounting for more than 25% of cells with a median of 22 (95% CI, 12.3-26.3) vs. 27.5 months (95% CI, 23.9-30.9) respectively, p=0.02.

Our results show that mutational load and subclonal diversity are poor prognostic factors in myeloma. Previous studies from massive-parallel sequencing and single cell analyses of myeloma plasma cells already revealed that myeloma had the features of an evolutionary ecosystem, where different tumour subclones coexist and have differential response to treatment. We have demonstrated in this study that measures of tumor diversity have important clinical consequences. To our knowledge, this is the first time that the use of clonal diversity indices as predictive biomarkers of progression is proposed in haematological malignancies, and more specifically, myeloma.