Introduction:
Treatment strategies incorporating proteasome inhibitors, immunomodulators, and autologous transplantation induce durable remissions in most newly diagnosed multiple myeloma (NDMM) patients. However, for 20% of patients even the most intensive therapies have not resulted in satisfactory outcomes. Currently available risk scores do not fully appreciate the complex biology of MM and have limited sensitivity and/or specificity for identification of high risk (HR) disease. We therefore aimed to characterize the mutational landscape of transplant-eligible NDMM patients who relapsed within 2 years after treatment initiation, thereby defining true clinical HRMM. To elucidate the clonal structure and evolution in these patients, we performed deep whole genome sequencing (WGS, ~80x) and RNAseq of samples collected at baseline and first relapse.
Methods:
We included 34 transplanted NDMM patients who experienced early relapse during maintenance within 2 years after treatment initiation. Tumor samples were collected from 20 and 31 patients at baseline and first relapse, respectively. Paired samples taken at both time points were available from 17 patients. WGS and RNAseq data were pre-processed using in-house pipelines. Single nucleotide variants (SNVs), indels, translocations, and copy number variants (CNVs) were called using Platypus, SOPHIA and ACESeq. Subclones were identified using SciClone. RNAseq data was aligned using STAR. Fusion genes were called by Arriba. Differential gene expression was assessed using DESeq2.
Results:
At baseline, only 12/20 patients would have been classified as HR according to conventional markers, including presence of t(4;14), t(14;16), amp(1q), clonal del(17p) or ISS3. In 5 patients del(17p) was solely observed in a minor sublone, which was selected during treatment and became dominant at relapse in 3 of them. Selection of amp(1q)-positive subclones was seen in 2 patients, illustrating that subclonal amp(1q) or del(17p) are frequent events in HR patients, and - in contrast to recent results - could contribute to early relapse. Translocations involving MYC have also been reported to be of prognostic impact. At baseline 9 of 20 patients were positive for this event, with BMP6 and the lambda locus being the translocation partner in 2 patients each. At relapse we found an additional MYC-lambda, and two MYC-kappa translocations, supporting recent observations that MYC-light chain translocations are associated with aggressive disease. We identified a median of 40 (range: 17-233) nonsilent somatic SNVs per patient at baseline and 61 (range: 14-322) at relapse. Yet, comparing paired samples there was no significant increase in SNVs. In our HRMM set, 21 of 64 recently identified driver genes were mutated at baseline with KRAS (n = 6), TP53 (n = 6), NRAS (n = 2), and DIS3 (n = 2) being the most frequently affected genes. 6 of them - ACTG1, DIS3, FAM46C, NFKB2, RB1, and TRAF3 - were involved in fusion genes. At relapse the number of mutated driver genes increased to 29, and 10 of 31 patients presented with a clonal TP53 mutation. All patients with a TP53 mutation also showed deletion of the second allele or LOH. Including other tumor suppressor genes, such as RB1, CDKN2C, or TRAF3, 12/20 NDMM and 20/31 relapsed patients had at least one bi-allelic aberration, doubling the number of HR patients with such events compared to considering TP53 alone. Longitudinally, we observed all patterns of clonal evolution that were recently described for unselected patients. Stable evolution was primarily seen in patients achieving partial remission, supporting a model where some tumor cells survive in a protective microenvironment (ME). In deep responders, however, branching evolution was the dominant patterns. This observation rather supports strong cell-intrinsic mechanisms and rapid selection of aggressive minor subclones in clinically defined HRMM.
Conclusions:
Understanding the mutational landscape in HRMM and drivers of early relapse is crucial in order to improve treatment options. Our study highlights the importance of bi-allelic events in HR and suggests that focusing on TP53 is not sufficient, if all HR cases are to be identified. Of note, 3 patients in the entire cohort would have been classified as low risk by conventional risk scores and even at relapse did not carry any bi-allelic event, indicating the existence of unknown somatic HR aberrations or a protective ME.
Müller-Tidow:MSD: Membership on an entity's Board of Directors or advisory committees. Goldschmidt:MSD: Research Funding; Sanofi: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Chugai: Honoraria, Research Funding; Mundipharma: Research Funding; Takeda: Membership on an entity's Board of Directors or advisory committees, Research Funding; Novartis: Membership on an entity's Board of Directors or advisory committees, Research Funding; Janssen: Consultancy, Research Funding; Amgen: Consultancy, Research Funding; Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; Molecular Partners: Research Funding; Dietmar-Hopp-Stiftung: Research Funding; John-Hopkins University: Research Funding; Adaptive Biotechnology: Membership on an entity's Board of Directors or advisory committees; Bristol-Myers Squibb: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding; John-Hopkins University: Research Funding.
Author notes
Asterisk with author names denotes non-ASH members.
