Genomic Proximity Mapping for Identification of Chromosomal Aberrations in Multiple Myeloma

Multiple myeloma (MM) is a hematological malignancy with heterogeneous clinical courses despite morphological similarity, a phenomenon driven primarily by the genetic abnormalities seen in the disease. A plethora of primary genetic abnormalities, such as translocations involving immunoglobulin loci and trisomies, along with secondary chromosomal aberrations (mainly 17p, 1p, 13q deletions, 1q gains, and/or MYC translocations) are present in MM, with the latter underlying the serial progression of MGUS to MM [1]. Cytogenetic analysis with fluorescence in situ hybridization (FISH) is currently the gold-standard method for identifying these genetic abnormalities. However, owing to the poor proliferative capacity of plasma cells in frequently hemodiluted bone marrow aspirates, it is often difficult to get adequate yields of plasma cell metaphases, preventing karyotyping analysis and necessitating interphase FISH [2]. This limitation has largely been overcome by using a variety of FISH probes, which allowed detection of various genetic alterations including numerical amplifications or deletions, structural rearrangements, and translocations [3]. However, there are still several limitations of FISH-based analyses including detection sensitivity and inter-laboratory variation in analysis and scoring [4]. More importantly, genomic variations can also result in false positive/negative findings—for instance, 50% of MYC rearrangements are missed by the break apart probe due to variable MYC genomic rearrangements [5]. The establishment of individualized and mutation-guided targeted therapies is of great importance to further improve clinical outcomes [1]. The Revised International Staging System (R-ISS) [6] and the Mayo Clinic Risk Stratification for Multiple Myeloma (mSMART) [7] are the two most widely accepted prognostic classification systems that have adopted these methodologies. Overall, classifying the patients into high-risk or standard risk categories is mainly driven by the genetic abnormalities. Patients with del(17p), t(4;14), t(14;16), t(14;20), and specific chromosome 1 abnormalities are considered high-risk in whom survival remains poorer in spite of novel drugs and maintenance strategies. More recently, the International Myeloma Society (IMS) and the International Myeloma Working Group (IMWG) released the Consensus Genomic Staging (CGS) of high-risk MM (HRMM) to identify poor prognosis patients [8]. This CGS system for HRMM relies on the presence of one of these abnormalities derived from genomic methods: (1) Deletion of 17p (with a cancer clonal fraction of 20% or more) and/or the presence of a TP53 mutation; (2) IGH translocations—including t(4;14), t(14;16), or t(14;20)—in combination with either 1q gain/amplification and/or deletion of 1p32; (3) Monoallelic deletion of 1p32 accompanied by 1q gain/amplification, or the presence of a biallelic deletion of 1p32. This seminal report emphasizes the need for a scalable and accurate identification of these genomic abnormalities. To address this, we evaluated the application of Genomic Proximity Mapping (GPM) [9-11] (Figure 1A), a novel method based on Hi-C technology designed to capture chromatin conformation [12], on five newly diagnosed MM patients to identify large genomic aberrations including structural variants (SVs), copy number variations (CNVs), and ploidy changes. GPM can be performed on a wide variety of substrates including as few as 2 00 000 fresh or frozen cells, formalin-fixed paraffin-embedded (FFPE) tissue or clot preparations, and methanol-acetic acid fixed cells. In this study, clonal sorted CD138+ cells from bone marrow of these patients (Figure 1B) were used as input for library preparation. Cells were crosslinked prior to lysis followed by chromatin digestion with restriction enzymes and proximity ligation. The proximity ligated product was used to generate a paired-end sequencing library that was then sequenced to a depth of at least 200 M read pairs. Sequence data was analyzed on the CytoTerra cloud analysis platform. Variant callers are used in an ensemble to detection deviation from expected patterns of long-range sequence interactions, read-depth coverage, and B-allele frequency (BAF) to identify SVs and CNVs. Variant calls were reviewed by an analyst blinded to previous findings. GPM confirmed all the abnormalities previously identified using FISH on these 5 patient samples (Figure 1B) and the full list of high-confidence genomic aberrations as well as the concordance with FISH results are listed in Table S1. Additionally, GPM enabled identification of multiple important large genomic alterations critical in the context of MM (Figure 1C–G). For patient MM439 (Figure 1C), GPM identified recurrent rearrangement including an IGH::CCND3 fusion resulting from t(6;14)(p21.1;q32.33) (Figure 2A), loss of chromosome 13, and deletion of TP53 (Figure 2B). Not revealed by FISH analysis, this patient also demonstrated a 10 Mbp copy-neutral loss of heterozygosity encompassing the known myeloma driver FUBP1 [13] (Figure 2C) and a del(1) (p22.1p12). While the del(1p) in MM439 does not overlap with the critical region of chromosome 1p associated with myeloma risk, a deletion in this critical region was identified in patient MM526 (Figure 1D), which was not tested by FISH. In agreement with previous findings, this patient demonstrated hyperdiploidy and an unspecified IGH rearrangement. GPM demonstrates that this unspecified rearrangement is likely to be the result of two large deletions and a complex series of inversions all at some distance from the IGH locus. This rearrangement is also accompanied by a chromothripsis event between chromosomes 13 and 16, a high-risk biomarker [14-16] that cannot be assessed by FISH (Figure 2D). Gain or amplification of 1q and hyperdiploidy were also observed in patients MM355 (Figure 1E) and MM655 (Figure 1F). MM655 additionally showed the canonical t(4;14) rearrangement (Figure 2A) and −13. The findings for these two patients were concordant with FISH findings. MM365 (Figure 1G) is a 76-year-old woman presented in 2012 with progressive bone pain and headaches. Imaging revealed multiple lytic bone lesions and a left skull-base mass. Laboratory evaluation demonstrated an IgA lambda monoclonal gammopathy, and bone marrow examination showed 80% plasma cell infiltration consistent with multiple myeloma. Conventional FISH studies performed externally did not initially identify any high-risk cytogenetic abnormalities, and the patient was therefore started on lenalidomide plus dexamethasone (Rd). The patient achieved a complete hematologic remission within 9 months of therapy, which was sustained for approximately 1 year before discontinuation of maintenance therapy. Six months later after discontinuation of maintenance therapy, the patient developed a symptomatic relapse with a new right clavicular plasmacytoma that was treated with localized radiation. Three months later following radiation, the patient experienced biochemical progression and was re-treated with lenalidomide/dexamethasone without a response. Subsequent therapy with CyBorD (cyclophosphamide, bortezomib, and dexamethasone) led to a partial response lasting 6 months before another rapid and aggressive relapse occurred. Given the patient's declining performance status, the patient declined further treatment and passed away shortly thereafter. Retrospective review by GPM revealed that this patient's myeloma harbored both t(8;14) resulting in a MYC::IGH rearrangement (Figure 2E) and 1q gain (Figure 2F). FISH identified hyperdiploidy, monosomy 13, and an IGH rearrangement without a partner, and MYC FISH was not performed. Given the complex clone “53,X,-X,+add(1)(p13)x2,+3,del(4)(q31),+5,+6,+7,add(8)(q24.1),+9,-13, add(14)(q32),+15,add(18)(q21),+19 [7]/52,sl,-9,+der(9)t(1;9)(q12;q32),-15 [4]/46,XX[9]with a add(8)(q24.1)” and add(14)(q32), the identification of the t(8;14) was consistent with the GPM result but not appreciated by chromosome analysis. The absence of this information may have influenced the initial treatment decision. At the time, our institutional approach for standard-risk disease utilized lenalidomide-dexamethasone therapy (Rd), whereas patients with t(8;14) and 1q gain/amplification cytogenetics would have received triplet induction therapy with bortezomib, lenalidomide, and dexamethasone (VRd). Knowledge of the patient’s true cytogenetic risk profile might also have prompted consideration of autologous stem cell transplantation (ASCT) for consolidation, which - though used selectively in patients aged 76–79—remains a potential option for fit elderly patients with high-risk disease. This case underscores the importance of comprehensive and accurate cytogenetic assessment at diagnosis to guide optimal therapy selection. Overall, this study demonstrates the potential of GPM to increase the diagnostic yield of testing in multiple myeloma. In this cohort, 4/5 patients had additional findings that have known clinical impact or are recurrent features of multiple myeloma. The relatively modest sequencing depth requirements, genome-wide scope, and ability to detect a wide range of clinically relevant biomarkers suggest that GPM could efficiently address current shortcomings of standard-of-care myeloma diagnostics. R.K.K. and A.P. designed the study. J.-S.H. and S.E. performed the experiments. R.K.K. and S.E. interpreted the results. H.M. and Z.D.S. processed the sequencing data. A.D., E.B., P.L.B., R.F., S.V.R., S.K., L.B.B., and W.I.G. gave critical feedback on the interpretation of the results. R.K.K., W.I.G., and A.P. wrote the manuscript. All the authors reviewed and approved the manuscript. This work was supported in part by funding from the Scholarly Clinician Award by the Mayo Clinic Department of Laboratory Medicine and Pathology; NCI U01CA271410, R01CA254961, P30CA15083, and CA186781 grants; and the Paula and Rodger Riney Family Foundation. This study was conducted with approval from the Mayo Clinic Institutional Review Board. All patients included in the analysis provided written informed consent allowing the use of their medical records for research purposes. Competing interests: R.F. reports providing consulting services for AbbVie, Adaptive Biotechnologies, Amgen, Apple, BMS (Celgene), GSK, Janssen, Karyopharm, Pfizer, RA Capital, Regeneron, and Sanofi; serving on the scientific advisory board of Caris Life Sciences; serving on the board of directors of Antengene (for profit); and having a patent for fluorescence in situ hybridization in multiple myeloma (∼$2000 per year). S.K.K. reports providing consulting services (with no personal payments) for AbbVie, Amgen, ArcellX, Beigene, BMS, Carsgen, GSK, Janssen, K36, Moderna, Pfizer, Regeneron, Roche-Genentech, Sanofi, Takeda and (with personal payments) for CVS Caremark and BD Biosciences; and received clinical trial support to the institution from AbbVie, Amgen, Astra Zeneca, BMS, Carsgen, GSK, Gracell Bio, Janssen, Oricell, Roche-Genentech, Sanofi, Takeda and Telogenomics. S.E. and H.M. are employees of Phase Genomics Inc., the developer of the CytoTerra platform. L.B.B. reports consulting services for Genentech. The remaining authors declare no competing financial interests. The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. Table S1: ajh70219-sup-0001-TableS1.xlsx. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.