ETV4-Dependent Transcriptional Plasticity Maintains MYC Expression and Results in IMiD Resistance in Multiple Myeloma

Immunomodulatory drugs (IMiD) are a backbone therapy for multiple myeloma (MM). Despite their efficacy, most patients develop resistance, and the mechanisms are not fully defined. Here, we show that IMiD responses are directed by IMiD-dependent degradation of IKZF1 and IKZF3 that bind to enhancers necessary to sustain the expression of MYC and other myeloma oncogenes. IMiD treatment universally depleted chromatin-bound IKZF1, but eviction of P300 and BRD4 coactivators only occurred in IMiD-sensitive cells. IKZF1-bound enhancers overlapped other transcription factor binding motifs, including ETV4. Chromatin immunoprecipitation sequencing showed that ETV4 bound to the same enhancers as IKZF1, and ETV4 CRISPR/Cas9-mediated ablation resulted in sensitization of IMiD-resistant MM. ETV4 expression is associated with IMiD resistance in cell lines, poor prognosis in patients, and is upregulated at relapse. These data indicate that ETV4 alleviates IKZF1 and IKZF3 dependency in MM by maintaining oncogenic enhancer activity and identify transcriptional plasticity as a previously unrecognized mechanism of IMiD resistance.

SIGNIFICANCE

We show that IKZF1-bound enhancers are critical for IMiD efficacy and that the factor ETV4 can bind the same enhancers and substitute for IKZF1 and mediate IMiD resistance by maintaining MYC and other oncogenes. These data implicate transcription factor redundancy as a previously unrecognized mode of IMiD resistance in MM. See related article by Welsh, Barwick, et al., p. 34. See related commentary by Yun and Cleveland, p. 5. This article is featured in Selected Articles from This Issue, p. 4.

Clinical Outcomes and Evolution of Clonal Hematopoiesis in Patients with Newly Diagnosed Multiple Myeloma

Clonal hematopoiesis (CH) at time of autologous stem cell transplant (ASCT) has been shown to be associated with decreased overall survival (OS) and progression-free survival (PFS) in patients with multiple myeloma not receiving immunomodulatory drugs (IMiD). However, the significance of CH in newly diagnosed patients, including transplant ineligible patients, and its effect on clonal evolution during multiple myeloma therapy in the era of novel agents, has not been well studied. Using our new algorithm to differentiate tumor and germline mutations from CH, we detected CH in approximately 10% of 986 patients with multiple myeloma from the Clinical Outcomes in MM to Personal Assessment of Genetic Profile (CoMMpass) cohort (40/529 transplanted and 59/457 non-transplanted patients). CH was associated with increased age, risk of recurrent bacterial infections and cardiovascular disease. CH at time of multiple myeloma diagnosis was not associated with inferior OS or PFS regardless of undergoing ASCT, and all patients benefited from IMiD-based therapies, irrespective of the presence of CH. Serial sampling of 52 patients revealed the emergence of CH over a median of 3 years of treatment, increasing its prevalence to 25%, mostly with DNMT3A mutations.

SIGNIFICANCE

Using our algorithm to differentiate tumor and germline mutations from CH mutations, we detected CH in approximately 10% of patients with newly diagnosed myeloma, including both transplant eligible and ineligible patients. Receiving IMiDs improved outcomes irrespective of CH status, but the prevalence of CH significantly rose throughout myeloma-directed therapy.

Mechanisms of antigen escape from BCMA- or GPRC5D-targeted immunotherapies in multiple myeloma

B cell maturation antigen (BCMA) target loss is considered to be a rare event that mediates multiple myeloma (MM) resistance to anti-BCMA chimeric antigen receptor T cell (CAR T) or bispecific T cell engager (TCE) therapies. Emerging data report that downregulation of G-protein-coupled receptor family C group 5 member D (GPRC5D) protein often occurs at relapse after anti-GPRC5D CAR T therapy. To examine the tumor-intrinsic factors that promote MM antigen escape, we performed combined bulk and single-cell whole-genome sequencing and copy number variation analysis of 30 patients treated with anti-BCMA and/or anti-GPRC5D CAR T/TCE therapy. In two cases, MM relapse post-TCE/CAR T therapy was driven by BCMA-negative clones harboring focal biallelic deletions at the TNFRSF17 locus at relapse or by selective expansion of pre-existing subclones with biallelic TNFRSF17 loss. In another five cases of relapse, newly detected, nontruncating, missense mutations or in-frame deletions in the extracellular domain of BCMA negated the efficacies of anti-BCMA TCE therapies, despite detectable surface BCMA protein expression. In the present study, we also report four cases of MM relapse with biallelic mutations of GPRC5D after anti-GPRC5D TCE therapy, including two cases with convergent evolution where multiple subclones lost GPRC5D through somatic events. Immunoselection of BCMA- or GPRC5D-negative or mutant clones is an important tumor-intrinsic driver of relapse post-targeted therapies. Mutational events on BCMA confer distinct sensitivities toward different anti-BCMA therapies, underscoring the importance of considering the tumor antigen landscape for optimal design and selection of targeted immunotherapies in MM.

Genome-scale functional genomics identify genes preferentially essential for multiple myeloma cells compared to other neoplasias

Clinical progress in multiple myeloma (MM), an incurable plasma cell (PC) neoplasia, has been driven by therapies that have limited applications beyond MM/PC neoplasias and do not target specific oncogenic mutations in MM. Instead, these agents target pathways critical for PC biology yet largely dispensable for malignant or normal cells of most other lineages. Here we systematically characterized the lineage-preferential molecular dependencies of MM through genome-scale clustered regularly interspaced short palindromic repeats (CRISPR) studies in 19 MM versus hundreds of non-MM lines and identified 116 genes whose disruption more significantly affects MM cell fitness compared with other malignancies. These genes, some known, others not previously linked to MM, encode transcription factors, chromatin modifiers, endoplasmic reticulum components, metabolic regulators or signaling molecules. Most of these genes are not among the top amplified, overexpressed or mutated in MM. Functional genomics approaches thus define new therapeutic targets in MM not readily identifiable by standard genomic, transcriptional or epigenetic profiling analyses.

Abstract 2061: Protein network analysis uncovers a poor-survival subtype in multiple myeloma

Multiple myeloma (MM) prognosis incorporates a variety of metrics including treatment received, clinical factors, and genomic characteristics, with several specific genomic features predicting for shorter progression free survival (PFS). No study to date has integrated genomic data from a systems view, incorporating interactions between biomarkers in a network. We use a geometric network analysis that integrates complex interactions to characterize patterns of biological behavior not captured by individual genomic events. The methodology is mathematically well-defined and has no fitting parameters. We hypothesized that such a systems mathematical approach applied to gene interaction networks may delineate biologically relevant MM subtypes and potential new therapeutic targets. We overlaid RNA-Seq and copy number alteration data from the MMRF CoMMpass study (IA19) on a gene interactome derived from the Human Protein Reference Database using a novel graph metric of network robustness — Ollivier-Ricci curvature (ORC). Results were clustered, with the optimal number determined via silhouette score. Survival analysis for PFS was performed employing Kaplan-Meier and log-rank tests. A differential gene expression analysis between high and low risk groups was conducted. Differences in scalar ORC between the low-risk and high-risk groups were examined and contextualized using a pathway analysis. Pathway analysis was performed using the Broad Institute’s Gene Set Enrichment Analysis tool and the pathways used are from the hallmark gene set from the Human MSigDB collection. The dataset included 659 patients and the incorporated protein-protein interactions resulted in a network with 8,468 nodes and 33,695 edges. The ORC analysis discovered 6 clusters, with specific genomic features being associated with clusters predicting for long [hyperdiploidy, t(11:14)], and short [t(4;14), MAF/MAFB translocations] PFS. A differential gene expression analysis comparing the high risk and low risk groups identified 118 key genes. These genes were associated with various pathways both known and unknown to be associated with multiple myeloma, including mitotic spindle, DNA repair, inflammatory response, and the P53 pathways. Further scalar curvature analysis showed differences in the apoptosis, TGF beta signaling, and other signaling pathways. In summary, we applied the geometric network analysis tool ORC to multi-omics data in MM represented as biological networks to identify individuals at high risk of short PFS and relevant biological correlates. Decreased robustness of signaling near immune-related genes was associated with shorter survival, highlighting the plausible utility of using these methods to uncover new biological insights.

Citation Format: Anish K. Simhal, Kylee H. Maclachlan, Rena Elkin, Jiening Zhu, Saad Z. Usmani, Jonathan J. Keats, Larry Norton, Joseph O. Deasy, Jung Hun Oh, Allen Tannenbaum. Protein network analysis uncovers a poor-survival subtype in multiple myeloma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 2061.

Perspectives on the Risk-Stratified Treatment of Multiple Myeloma

The multiple myeloma treatment landscape has changed dramatically. This change, paralleled by an increase in scientific knowledge, has resulted in significant improvement in survival. However, heterogeneity remains in clinical outcomes, with a proportion of patients not benefiting from current approaches and continuing to have a poor prognosis. A significant proportion of the variability in outcome can be predicted on the basis of clinical and biochemical parameters and tumor-acquired genetic variants, allowing for risk stratification and a more personalized approach to therapy. This article discusses the principles that can enable the rational and effective development of therapeutic approaches for high-risk multiple myeloma.

Unbiased cell surface proteomics identifies SEMA4A as an effective immunotherapy target for myeloma

The accessibility of cell surface proteins makes them tractable for targeting by cancer immunotherapy, but identifying suitable targets remains challenging. Here we describe plasma membrane profiling of primary human myeloma cells to identify an unprecedented number of cell surface proteins of a primary cancer. We used a novel approach to prioritize immunotherapy targets and identified a cell surface protein not previously implicated in myeloma, semaphorin-4A (SEMA4A). Using knock-down by short-hairpin RNA and CRISPR/nuclease-dead Cas9 (dCas9), we show that expression of SEMA4A is essential for normal myeloma cell growth in vitro, indicating that myeloma cells cannot downregulate the protein to avoid detection. We further show that SEMA4A would not be identified as a myeloma therapeutic target by standard CRISPR/Cas9 knockout screens because of exon skipping. Finally, we potently and selectively targeted SEMA4A with a novel antibody-drug conjugate in vitro and in vivo.

Genomic and Transcriptomic Analysis of Relapsed and Refractory Childhood Solid Tumors Reveals a Diverse Molecular Landscape and Mechanisms of Immune Evasion

Children with treatment-refractory or relapsed (R/R) tumors face poor prognoses. As the genomic underpinnings driving R/R disease are not well defined, we describe here the genomic and transcriptomic landscapes of R/R solid tumors from 202 patients enrolled in Beat Childhood Cancer Consortium clinical trials. Tumor mutational burden (TMB) was elevated relative to untreated tumors at diagnosis, with one-third of tumors classified as having a pediatric high TMB. Prior chemotherapy exposure influenced the mutational landscape of these R/R tumors, with more than 40% of tumors demonstrating mutational signatures associated with platinum or temozolomide chemotherapy and two tumors showing treatment-associated hypermutation. Immunogenomic profiling found a heterogenous pattern of neoantigen and MHC class I expression and a general absence of immune infiltration. Transcriptional analysis and functional gene set enrichment analysis identified cross-pathology clusters associated with development, immune signaling, and cellular signaling pathways. While the landscapes of these R/R tumors reflected those of their corresponding untreated tumors at diagnosis, important exceptions were observed suggestive of tumor evolution, treatment resistance mechanisms, and mutagenic etiologies of treatment.

Plasma cell leukemia: A review of the molecular classification, diagnosis, and evidenced-based treatment

Plasma cell leukemia is a rare and aggressive plasma cell dyscrasia associated with dismal outcomes. It may arise de novo, primary plasma cell leukemia, or evolve from an antecedent diagnosis of multiple myeloma, secondary plasma cell leukemia. Despite highly effective therapeutics, survival for plasma cell leukemia patients remains poor. Molecular knowledge of plasma cell leukemia has recently expanded with use of gene expression profiling and whole exome sequencing, lending new insights into prognosis and therapeutic development. In this review, we describe the molecular knowledge, clinical characteristics, evidenced-based therapeutic approaches and treatment outcomes of plasma cell leukemia.

Venetoclax sensitivity in multiple myeloma is associated with B-cell gene expression

Venetoclax is a highly potent, selective BCL2 inhibitor capable of inducing apoptosis in cells dependent on BCL2 for survival. Most myeloma is MCL1-dependent; however, a subset of myeloma enriched for translocation t(11;14) is codependent on BCL2 and thus sensitive to venetoclax. The biology underlying this heterogeneity remains poorly understood. We show that knockdown of cyclin D1 does not induce resistance to venetoclax, arguing against a direct role for cyclin D1 in venetoclax sensitivity. To identify other factors contributing to venetoclax response, we studied a panel of 31 myeloma cell lines and 25 patient samples tested for venetoclax sensitivity. In cell lines, we corroborated our previous observation that BIM binding to BCL2 correlates with venetoclax response and further showed that knockout of BIM results in decreased venetoclax sensitivity. RNA-sequencing analysis identified expression of B-cell genes as enriched in venetoclax-sensitive myeloma, although no single gene consistently delineated sensitive and resistant cells. However, a panel of cell surface makers correlated well with ex vivo prediction of venetoclax response in 21 patient samples and may serve as a biomarker independent of t(11;14). Assay for transposase-accessible chromatin sequencing of myeloma cell lines also identified an epigenetic program in venetoclax-sensitive cells that was more similar to B cells than that of venetoclax-resistant cells, as well as enrichment for basic leucine zipper domain-binding motifs such as BATF. Together, these data indicate that remnants of B-cell biology are associated with BCL2 dependency and point to novel biomarkers of venetoclax-sensitive myeloma independent of t(11;14).

Chromatin Accessibility Identifies Regulatory Elements Predictive of Gene Expression and Disease Outcome in Multiple Myeloma

PURPOSE

Multiple myeloma is a malignancy of plasma cells. Extensive genetic and transcriptional characterization of myeloma has identified subtypes with prognostic and therapeutic implications. In contrast, relatively little is known about the myeloma epigenome.

EXPERIMENTAL DESIGN

CD138+CD38+ myeloma cells were isolated from fresh bone marrow aspirate or the same aspirate after freezing for 1-6 months. Gene expression and chromatin accessibility were compared between fresh and frozen samples by RNA sequencing (RNA-seq) and assay for transpose accessible chromatin sequencing (ATAC-seq). Chromatin accessible regions were used to identify regulatory RNA expression in more than 700 samples from newly diagnosed patients in the Multiple Myeloma Research Foundation CoMMpass trial (NCT01454297).

RESULTS

Gene expression and chromatin accessibility of cryopreserved myeloma recapitulated that of freshly isolated samples. ATAC-seq performed on a series of biobanked specimens identified thousands of chromatin accessible regions with hundreds being highly coordinated with gene expression. More than 4,700 of these chromatin accessible regions were transcribed in newly diagnosed myelomas from the CoMMpass trial. Regulatory element activity alone recapitulated myeloma gene expression subtypes, and in particular myeloma subtypes with immunoglobulin heavy chain translocations were defined by transcription of distal regulatory elements. Moreover, enhancer activity predicted oncogene expression implicating gene regulatory mechanisms in aggressive myeloma.

CONCLUSIONS

These data demonstrate the feasibility of using biobanked specimens for retrospective studies of the myeloma epigenome and illustrate the unique enhancer landscapes of myeloma subtypes that are coupled to gene expression and disease progression.

Integrated whole-exome and transcriptome analysis of 250 treatment-refractory or relapsed (R/R) childhood solid tumors

10006

Background: The major genomic profiling studies that have helped define the molecular landscapes of pediatric cancers have typically focused on untreated pediatric cancers at diagnosis. Despite improvements in overall survival for childhood cancers, patients with treatment-refractory or relapsed (R/R) solid tumors face a poor prognosis. The genomic underpinnings of R/R disease are less well-characterized. Here, we describe the integrated genomic and transcriptomic analysis of 250 R/R solid tumors from 202 children profiled within precision medicine studies (NCT01355679, NCT01802567, NCT02162732) conducted by the Beat Childhood Cancer Consortium. Methods: Tumor-normal whole-exome and tumor mRNA sequencing was performed by Ashion Analytics (Phoenix, Arizona), a CAP-accredited, CLIA-certified laboratory, or within the research setting at TGen. Longitudinal tumor samples were sequenced for 20 patients. Variant calling included single nucleotide variants, indels, copy number alterations, and fusions. Integrated genomic and transcriptomic research analysis included microsatellite instability assessment, immunogenomic profiling, and functional gene set enrichment analysis. Results: Forty-six tumor types were represented, grouped into four general categories: sarcomas (36.1%; n = 73), neuroblastomas (29.2%; n = 59), CNS tumors (23.3%; n = 47), and other rare tumors (11.4%; n = 23). For patients with whole exome sequencing data, 78.3% (n = 144/184) of tumors bore a somatic alteration in at least one known cancer gene. Over one-third (39.1%; 72/184) of the cohort bore oncogenic fusions and/or oncogenic/likely-oncogenic hotspot mutations in a known cancer gene. Pathognomonic fusions were identified in 25% (46/184) of tumors, occurring most frequently in sarcomas. Pathogenic or likely pathogenic germline variants were identified in 8.7% (16/184) of patients. Microsatellite instability was detected in five different tumor types. Despite nearly all tumors (94%, 173/184) having at least one predicted strong binding neoantigen, over a quarter of tumors lacked transcript expression of these neoantigens or exhibited low MHC class I expression. Further, a subset of tumors showed elevated expression of the co-inhibitory immune checkpoint molecule PDL1. Transcriptional analysis and functional gene set enrichment analysis identified cross-pathology tumor clusters associated with immune signaling, development, and cellular signaling pathways. Longitudinal analysis revealed temporal heterogeneity pointing to the importance of re-biopsy at relapse for targeted treatment planning. Conclusions: Together, these data suggest R/R childhood solid tumors exhibit shared molecular features that are reflective of underlying biology, demonstrating the importance of comprehensive profiling to inform molecularly-guided treatment of R/R disease.

Functional Genomics Identify Distinct and Overlapping Genes Mediating Resistance to Different Classes of Heterobifunctional Degraders of Oncoproteins

Heterobifunctional proteolysis-targeting chimeric compounds leverage the activity of E3 ligases to induce degradation of target oncoproteins and exhibit potent preclinical antitumor activity. To dissect the mechanisms regulating tumor cell sensitivity to different classes of pharmacological "degraders" of oncoproteins, we performed genome-scale CRISPR-Cas9-based gene editing studies. We observed that myeloma cell resistance to degraders of different targets (BET bromodomain proteins, CDK9) and operating through CRBN (degronimids) or VHL is primarily mediated by prevention of, rather than adaptation to, breakdown of the target oncoprotein; and this involves loss of function of the cognate E3 ligase or interactors/regulators of the respective cullin-RING ligase (CRL) complex. The substantial gene-level differences for resistance mechanisms to CRBN- versus VHL-based degraders explains mechanistically the lack of cross-resistance with sequential administration of these two degrader classes. Development of degraders leveraging more diverse E3 ligases/CRLs may facilitate sequential/alternating versus combined uses of these agents toward potentially delaying or preventing resistance.

Revealing the impact of structural variants in multiple myeloma

The landscape of structural variants (SVs) in multiple myeloma remains poorly understood. Here, we performed comprehensive analysis of SVs in a large cohort of 752 multiple myeloma patients by low coverage long-insert whole genome sequencing. We identified 68 SV hotspots involving 17 new candidate driver genes, including the therapeutic targets BCMA (TNFRSF17), SLAMF and MCL1. Catastrophic complex rearrangements termed chromothripsis were present in 24% of patients and independently associated with poor clinical outcomes. Templated insertions were the second most frequent complex event (19%), mostly involved in super-enhancer hijacking and activation of oncogenes such as CCND1 and MYC. Importantly, in 31% of patients two or more seemingly independent putative driver events were caused by a single structural event, demonstrating that the complex genomic landscape of multiple myeloma can be acquired through few key events during tumor evolutionary history. Overall, this study reveals the critical role of SVs in multiple myeloma pathogenesis.

Glucocorticoid receptor expression in multiple myeloma patients is a predictor of survival

Multiple myeloma (MM) is a blood neoplasia characterized by abnormal proliferation of plasma cells. Various treatments such as stem cell transplant (SCT), proteasome inhibitors, immune-modulating drugs, monoclonal antibodies and selective inhibitors of nuclear export have been routinely used to treat MM. However, relapse and treatment resistance are common problems in MM patients. Treatments are enhanced by Dexamethasone (Dex), a synthetic steroid that activates the glucocorticoid receptor (GR) which leads to apoptosis. To evaluate the potential impact of GR expression on overall survival, MM patient data from the CoMMpass study of 650 patients were analyzed. Multivariate modeling results show that increased GR expression at diagnosis is associated with a decreased risk of dying relative to those with lower levels of expression.

c-MYC expression and maturity phenotypes are associated with outcome benefit from addition of ixazomib to lenalidomide-dexamethasone in myeloma

OBJECTIVES

In the TOURMALINE-MM1 phase 3 trial in relapsed/refractory multiple myeloma, ixazomib-lenalidomide-dexamethasone (IRd) showed different magnitudes of progression-free survival (PFS) benefit vs placebo-Rd according to number and type of prior therapies, with greater benefit seen in patients with >1 prior line of therapy or 1 prior line of therapy without stem cell transplantation (SCT).

METHODS

RNA sequencing data were used to investigate the basis of these differences.

RESULTS

The PFS benefit of IRd vs placebo-Rd was greater in patients with tumors expressing high c-MYC levels (median not reached vs 11.3 months; hazard ratio [HR] 0.42; 95% CI, 0.26, 0.66; P < .001) compared with in those expressing low c-MYC levels (median 20.6 vs 16.6 months; HR 0.75; 95% CI, 0.42, 1.2). Expression of c-MYC in tumors varied based on the number and type of prior therapy received, with the lowest levels observed in tumors of patients who had received 1 prior line of therapy including SCT. These tumors also had higher expression levels of CD19 and CD81.

CONCLUSIONS

PFS analyses suggest that lenalidomide and ixazomib target tumors with different levels of c-MYC, CD19, and CD81 expression, thus providing a potential rationale for the differential benefits observed in the TOURMALINE-MM1 study. This trial was registered at www.clinicaltrials.gov as: NCT01564537.

Stability and uniqueness of clonal immunoglobulin CDR3 sequences for MRD tracking in multiple myeloma

Minimal residual disease (MRD) tracking, by next generation sequencing of immunoglobulin sequences, is moving towards clinical implementation in multiple myeloma. However, there is only sparse information available to address whether clonal sequences remain stable for tracking over time, and to what extent light chain sequences are sufficiently unique for tracking. Here, we analyzed immunoglobulin repertoires from 905 plasma cell myeloma and healthy control samples, focusing on the third complementarity determining region (CDR3). Clonal heavy and/or light chain expression was identified in all patients at baseline, with one or more subclones related to the main clone in 3.2%. In 45 patients with 101 sequential samples, the dominant clonal CDR3 sequences remained identical over time, despite differential clonal evolution by whole exome sequencing in 49% of patients. The low frequency of subclonal CDR3 variants, and absence of evolution over time in active multiple myeloma, indicates that tumor cells at this stage are not under selective pressure to undergo antibody affinity maturation. Next, we establish somatic hypermutation and non-templated insertions as the most important determinants of light chain clonal uniqueness, identifying a potentially trackable sequence in the majority of patients. Taken together, we show that dominant clonal sequences identified at baseline are reliable biomarkers for long-term tracking of the malignant clone, including both IGH and the majority of light chain clones.

MiR-16 regulates crosstalk in NF-κB tolerogenic inflammatory signaling between myeloma cells and bone marrow macrophages

High levels of circulating miR-16 in the serum of multiple myeloma (MM) patients are independently associated with longer survival. Although the tumor suppressor function of intracellular miR-16 in MM plasma cells (PCs) has been elucidated, its extracellular role in maintaining a nonsupportive cancer microenvironment has not been fully explored. Here, we show that miR-16 is abundantly released by MM cells through extracellular vesicles (EVs) and that differences in its intracellular expression as associated with chromosome 13 deletion (Del13) are correlated to extracellular miR-16 levels. We also demonstrate that EVs isolated from MM patients and from the conditioned media of MM-PCs carrying Del13 more strongly differentiate circulating monocytes to M2-tumor supportive macrophages (TAMs), compared with MM-PCs without this chromosomal aberration. Mechanistically, our data show that miR-16 directly targets the IKKα/β complex of the NF-κB canonical pathway, which is critical not only in supporting MM cell growth, but also in polarizing macrophages toward an M2 phenotype. By using a miR-15a-16-1-KO mouse model, we found that loss of the miR-16 cluster supports polarization to M2 macrophages. Finally, we demonstrate the therapeutic benefit of miR-16 overexpression in potentiating the anti-MM activity by a proteasome inhibitor in the presence of MM-resident bone marrow TAM.

Abstract 5113: A flexible pipeline for precision medicine biomarker detection and prediction of treatment response

Recent years have seen an explosion in the availability of paired molecular profiling and drug screen data, providing an unprecedented opportunity for the development of targeted therapies based on an individual’s genetic background. Despite a number of recent successes in diseases ranging from cystic fibrosis to cancer, significant hurdles remain in our ability to accurately predict treatments based on molecular profiling data. In particular, few such tools exist that allow the integration of heterogeneous data types (e.g. genomic, transcriptomic, and somatic mutations), along with high-throughput drug screen data to make predictions about treatment efficacy. Here, we describe a generalized open-source pipeline developed for the analysis of precision medicine data, Pharmacogenomics Prediction Pipeline, or “P3”. The modular design of P3 enables the inclusion of arbitrary input data types and the selection from multiple alternative machine learning algorithms, while automated statistical and visualization reporting steps incorporated throughout the pipeline assist in parameter tuning and early detection of problematic data components. Molecular profiling data is further enriched by the incorporation of external biological information in the form of pathway and gene set annotations such and Gene Ontology (GO) and The Molecular Signatures Database (MSigDB). To demonstrate the use of P3 for preclinical biomarker prediction, we applied P3 to an unpublished multiple myeloma dataset consisting of exome, RNA-Seq, CNV, and drug screen data for 1,912 compounds across 47 tumor cell lines. Specifically, P3 was used to predict molecular features associated with response to treatment for all drugs where a differential response to treatment was observed across patients. Furthermore, molecular profiling and drug screen data for 267 drugs and over a thousand cell lines spanning multiple cancer types from the Genomics of Drug Sensitivity in Cancer (GDSC) project were analyzed using P3, providing insights into shared mechanisms of drug sensitivity and resistance across different cancer and treatment types.

Citation Format: V Keith Hughitt, Sayeh Gorjifard, Aleksandra M. Michalowski, John K. Simmons, Ryan Dale, Eric C. Polley, Jonathan J. Keats, Beverly A. Mock. A flexible pipeline for precision medicine biomarker detection and prediction of treatment response [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 5113.

Abstract 839: Whole genome DNA methylation analysis of multiple myeloma identifies pervasive hypomethylation and biomarkers of survival

Multiple myeloma is a malignancy of terminally differentiated, antibody secreting B cells known as plasma cells. Normal plasma cell differentiation and cell fate are coupled to epigenetic and transcriptional reprogramming, including a proliferation-dependent global loss of DNA methylation. However, relatively little is known about the epigenetic changes that underlie myelomagenesis and how these contribute to disease etiology. To this end, we have analyzed the DNA methylome of 119 myeloma specimens from the CoMMpass study (NCT01454297) by whole genome bisulfite sequencing (WGBS) and more than 90% of these same specimens were also characterized for structural variants (long-insert whole genome sequencing) and gene expression (RNA-seq). Unsupervised hierarchical clustering grouped together specimens with t(4;14) translocations that upregulate the H3K36 dimethyltransferase NSD2 (also known as MMSET and WHSC1), which likely impacts the DNA methylation state through epigenetic cross-talk. These data also revealed a dramatic genome-wide hypomethylation where myeloma samples had a median global CpG methylation level of 41% as compared to 71% and 89% in normal plasma cells and B cells, respectively. Demethylation of the myeloma methylome occurred in large megabase domains encompassing genes that were devoid of gene expression. Conversely, DNA methylation remained mostly unchanged in the bodies of genes that were highly expressed. Although the majority of these hypomethylated domains were common across myelomas, many regions of variable methylation exist and these differences corresponded with proximal gene expression differences. These variably methylated regions were compared to PFS and OS and this identified 6,314 CpG loci where the level of DNA methylation was prognostic of outcome (P≤0.00001). These loci were clustered into discrete regions and in the majority of cases (79%), reduced DNA methylation at these loci corresponded with poor outcome. For example, several loci in the gene bodies of PRKCE, MGMT, FHIT, WWOX were prognostic of poor survival. Interestingly, myeloma t(14;16) translocations disrupt the tumor suppressor WWOX and induce the oncogene MAF. These data suggest that WWOX expression may also be lost by epigenetic mechanisms. Finally, we analyzed the DNA methylome of primary and relapsed samples for 22 patients, including multiple relapsed samples for 2 patients. These data identified genome-wide DNA methylation remodeling with a median of 1.9 million differential methylated CpGs between the newly diagnosed and relapsed specimens. These relapse differentially methylated loci coincided at the same regions in several patients and significantly overlapped loci where the DNA methylation level was prognostic of outcome. Integrative genetic, epigenetic, and transcriptional analyses for 120 myeloma samples will be presented.

Citation Format: Benjamin G. Barwick, Doris R. Powell, Daniel Penaherrera, Sheri Skerget, Jonathan J. Keats, Daniel Auclair, Sagar Lonial, Lawrence H. Boise, Paula M. Vertino. Whole genome DNA methylation analysis of multiple myeloma identifies pervasive hypomethylation and biomarkers of survival [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 839.

Multiple myeloma immunoglobulin lambda translocations portend poor prognosis

Multiple myeloma is a malignancy of antibody-secreting plasma cells. Most patients benefit from current therapies, however, 20% of patients relapse or die within two years and are deemed high risk. Here we analyze structural variants from 795 newly-diagnosed patients as part of the CoMMpass study. We report translocations involving the immunoglobulin lambda (IgL) locus are present in 10% of patients, and indicative of poor prognosis. This is particularly true for IgL-MYC translocations, which coincide with focal amplifications of enhancers at both loci. Importantly, 78% of IgL-MYC translocations co-occur with hyperdiploid disease, a marker of standard risk, suggesting that IgL-MYC-translocated myeloma is being misclassified. Patients with IgL-translocations fail to benefit from IMiDs, which target IKZF1, a transcription factor that binds the IgL enhancer at some of the highest levels in the myeloma epigenome. These data implicate IgL translocation as a driver of poor prognosis which may be due to IMiD resistance.

Multiple myeloma is frequently characterised by translocation of genes next to the immunoglobulin heavy chain locus. In this study, the authors sequence a large cohort of high risk myeloma samples and find translocations of cMyc to the immunoglobulin heavy chain locus and this is associated with poor prognosis.

RNA-seq of newly diagnosed patients in the PADIMAC study leads to a bortezomib/lenalidomide decision signature

Improving outcomes in multiple myeloma will involve not only development of new therapies but also better use of existing treatments. We performed RNA sequencing on samples from newly diagnosed patients enrolled in the phase 2 PADIMAC (Bortezomib, Adriamycin, and Dexamethasone Therapy for Previously Untreated Patients with Multiple Myeloma: Impact of Minimal Residual Disease in Patients with Deferred ASCT) study. Using synthetic annealing and the large margin nearest neighbor algorithm, we developed and trained a 7-gene signature to predict treatment outcome. We tested the signature in independent cohorts treated with bortezomib- and lenalidomide-based therapies. The signature was capable of distinguishing which patients would respond better to which regimen. In the CoMMpass data set, patients who were treated correctly according to the signature had a better progression-free survival (median, 20.1 months vs not reached; hazard ratio [HR], 0.40; confidence interval [CI], 0.23-0.72; P = .0012) and overall survival (median, 30.7 months vs not reached; HR, 0.41; CI, 0.21-0.80; P = .0049) than those who were not. Indeed, the outcome for these correctly treated patients was noninferior to that for those treated with combined bortezomib, lenalidomide, and dexamethasone, arguably the standard of care in the United States but not widely available elsewhere. The small size of the signature will facilitate clinical translation, thus enabling more targeted drug regimens to be delivered in myeloma.

Abstract 3006: Molecular characterization of baseline and serial multiple myeloma patients from the MMRF CoMMpass study

Multiple myeloma (MM) is a hematological malignancy of plasma cells accounting for ~2% of new cancer cases each year in the United States. Our understanding of MM pathogenesis has improved dramatically with the development of whole genome analysis technologies, however, to date no study has comprehensively analyzed a large cohort of MM patients. The Multiple Myeloma Research Foundation CoMMpass Study (NCT01454297) is a fully accrued observational clinical trial with 1143 newly diagnosed MM patients from sites in the United States, Canada, Spain, and Italy. Clinical parameters are collected at baseline and every three months through the eight-year observation period. Tumor samples are collected and characterized using whole genome, exome, and RNA sequencing at diagnosis and each progression event. This unique prospective study design differentiates CoMMpass from other large cancer genomics studies performed to date.

This represents the first analysis of the CoMMpass interim analysis 12 dataset including 982 of whom are molecularly characterized at baseline. Median follow-up of the cohort exceeds 2 years, and while the median OS has still not been reached, median PFS of the cohort is 36 months. We identified a median of 153 non-immunoglobulin mutations, 29 structural events, and 133 copy number (CN) events per tumor at baseline. In order to identify a set of significantly mutated genes, we applied a consensus-based approach identifying 60 genes mutated in at least 1% of the baseline cohort. Consensus clustering of the CN and gene expression profiles identified 14 and 12 distinct MM subtypes, respectively. Integration of WGS and RNA sequencing data identified 1163 cross-validated fusion transcripts. An integrated analysis of all data sources identified a series of potential gain-of-function and loss-of-function genes, from which a pathway analysis highlighted alterations in the NF-kB, Ras, DNA repair, and cell-cycle pathways.

This dataset includes serial data for 121 patients (171 specimens), 21 of whom had multiple progression events. Mutational analyses revealed that ~ 24% of mutations identified at progression were not identified at the previous timepoint. Although rarely mutated at baseline, RRBP1 was frequently mutated at relapse and may represent a novel driver of disease progression or treatment resistance in MM. In progression samples we observed a greater proportion of patients with NRAS mutations, attributable to four patients who acquire NRAS mutations and five patients with baseline KRAS mutations that shift to NRAS mutations at relapse. All patients who exhibit a KRAS to NRAS shift were bortezomib treated, suggesting this shift may represent a mechanism of resistance. This comprehensive study has identified distinct genetic subgroups with variable clinical outcome and demonstrates the value of prospective collections to identify mechanisms of progression and resistance.

Citation Format: Sheri Skerget, Austin Christofferson, Sara Nasser, Jessica Aldrich, Daniel Penaherrera, Christophe Legendre, Martin Boateng, Lori Cuyugan, Jonathan Adkins, Erica Tassone, The MMRF CoMMpass Network, Jen Yesil, Daniel Auclair, Winnie Liang, Jonathan J. Keats. Molecular characterization of baseline and serial multiple myeloma patients from the MMRF CoMMpass study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 3006.

HSF1 Is Essential for Myeloma Cell Survival and A Promising Therapeutic Target

Purpose: Myeloma is a plasma cell malignancy characterized by the overproduction of immunoglobulin, and is therefore susceptible to therapies targeting protein homeostasis. We hypothesized that heat shock factor 1 (HSF1) was an attractive therapeutic target for myeloma due to its direct regulation of transcriptional programs implicated in both protein homeostasis and the oncogenic phenotype. Here, we interrogate HSF1 as a therapeutic target in myeloma using bioinformatic, genetic, and pharmacologic means.Experimental Design: To assess the clinical relevance of HSF1, we analyzed publicly available patient myeloma gene expression datasets. Validation of this novel target was conducted in in vitro experiments using shRNA or inhibitors of the HSF1 pathway in human myeloma cell lines and primary cells as well as in in vivo human myeloma xenograft models.Results: Expression of HSF1 and its target genes were associated with poorer myeloma patient survival. ShRNA-mediated knockdown or pharmacologic inhibition of the HSF1 pathway with a novel chemical probe, CCT251236, or with KRIBB11, led to caspase-mediated cell death that was associated with an increase in EIF2α phosphorylation, CHOP expression and a decrease in overall protein synthesis. Importantly, both CCT251236 and KRIBB11 induced cytotoxicity in human myeloma cell lines and patient-derived primary myeloma cells with a therapeutic window over normal cells. Pharmacologic inhibition induced tumor growth inhibition and was well-tolerated in a human myeloma xenograft murine model with evidence of pharmacodynamic biomarker modulation.Conclusions: Taken together, our studies demonstrate the dependence of myeloma cells on HSF1 for survival and support the clinical evaluation of pharmacologic inhibitors of the HSF1 pathway in myeloma. Clin Cancer Res; 24(10); 2395-407. ©2018 AACRSee related commentary by Parekh, p. 2237.

Whole Exome Library Construction for Next Generation Sequencing

Whole exome sequencing (WES) is a DNA sequencing strategy that provides a survey of base substitutions across coding genomic locations and other regions of interest. As the coding portion of the genome encompasses only 1-2% of the entire genome, this approach represents a more cost-effective strategy to detect DNA alterations that may alter protein function, compared to whole genome sequencing. Although the research community has and is currently delineating the functional implications of sequence changes in noncoding regions of the genome, WES is a currently available assay that provides valuable information for both discovery research and precision medicine applications. In this chapter, we present a WES library preparation protocol using the KAPA Hyper Prep Kit with Agilent SureSelect Human All Exon V5+UTR probes that demonstrates high DNA-to-library conversion efficiency for sequencing on the Illumina HiSeq platform.

Whole Genome Library Construction for Next Generation Sequencing

With the rapid evolution of genomics technologies over the past decade, whole genome sequencing (WGS) has become an increasingly accessible tool in biomedical research. WGS applications include analysis of genomic DNA from single individuals, multiple related family members, and tumor/normal samples from the same patient in the context of oncology. A number of different modalities are available for performing WGS; this chapter focuses on wet lab library construction procedures for complex short insert WGS libraries using the KAPA Hyper Prep Kit (Kapa Biosystems), and includes a discussion of appropriate quality control measures for sequencing on the Illumina HiSeq2000 platform. Additional modifications to the protocol for long insert WGS library construction, to assess structural alterations and copy number changes, are also described.

Baseline mutational patterns and sustained MRD negativity in patients with high-risk smoldering myeloma

Early results of a prospective phase 2 clinical trial of carfilzomib, lenalidomide, and dexamethasone followed by lenalidomide maintenance in high-risk smoldering myeloma showed promising results that were previously published. Here, we provide novel insights into the genetic landscape of high-risk smoldering myeloma and information on sustained minimal residual disease (MRD) negativity with an expanded cohort of patients. Eighteen patients with high-risk smoldering myeloma were enrolled between 29 May 2012, and 14 January 2014. We included patients with newly diagnosed multiple myeloma enrolled in a parallel trial who received the same therapy (reference group). The overall response rate was 100%. With median potential follow-up of 43.3 months, 10 (63%) remain in MRD negativity, and the estimated 4-year progression-free and overall survival rates are 71% and 100%, respectively. Importantly, we report differences in mutational patterns in patients with high-risk smoldering myeloma and newly diagnosed multiple myeloma, reflected in a lower frequency of mutations in significant myeloma genes (6.6% vs 45%) and NFKB pathway genes (6.6% vs 25%). Treatment with carfilzomib, lenalidomide, and dexamethasone followed by lenalidomide maintenance was associated with a 100% response rate and 63% MRD negativity with a safety profile consistent with previous reports for this regimen. This study had a small numbers of participants, but there seemed to be important differences in the genetic landscape of patients with high-risk smoldering myeloma and those with newly diagnosed multiple myeloma, suggestive of a more treatment-responsive biology in early disease.

Abstract 5271: Optimization and detection of focal somatic copy number variants in whole genome, whole exome and panel sequencing for tumor/normal matched pairs and tumor only analysis

Often in the clinical setting, tumor samples may be derived from historical tissue or from fresh frozen tissue and it may not be feasible to obtain normal tissue from the individual. Comparing unmatched tissues from different individuals possess a unique challenge of addressing common copy number variations not faced when comparing matched samples. In addition to the challenges of comparing unmatched samples, DNA extracted from samples preserved as formalin-fixed, paraffin-embedded (FFPE) tissue is often highly degraded and when compared to more intact DNA extracted from different source (e.g. blood) can increase the inherent noise in the system. Here we analytically characterize the accuracy and performance of algorithmic approaches for separating germline inherited copy number variation from somatic copy number changes, focusing on both filtering approaches and use of pooled reference samples sequenced under similar conditions. Example approaches include filtering known common copy number variation within 1000 Genomes Phase 3 and Database of Genomic Variants (DGV) Gold Standard. Additionally, we utilized a tumor/reference pool based analysis where a reference pool was constructed by equimolar pooling of multiple individuals. Determination of fold changes between tumor and reference was calculated by determining physical coverage of read pair fragments in 100 bases increments. Next, to address differences in sequencing performance between tumor and reference, the read depth data for each sample is collapsed/averaged into a lower resolution according to user-selected parameters (e.g. distance between points and read depth). Normalized log2 fold-changes between tumor and reference samples are then calculated and an adjustable smoothing window is applied. In addition, we utilize tumor allele frequencies of known heterozygous germline SNPs identified within the normal to both evaluate potential false positives and correct biases. Lastly, a segmentation algorithm is applied to summarize the individual log2 fold-changes into intervals with a constant copy number state. We will present the advantages and limitations of these approaches both when a germline normal is available and when tumor only analysis is necessary.

Citation Format: Jessica Aldrich, Jonathan J. Keats, Austin Christofferson, Winnie S. Liang, John D. Carpten, Lisa Baumbach-Reardon, David W. Craig. Optimization and detection of focal somatic copy number variants in whole genome, whole exome and panel sequencing for tumor/normal matched pairs and tumor only analysis. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 5271.

Abstract 45: Detection of focal somatic copy number variants in whole genome, whole exome, and targeted next-generation sequencing data of tumor/normal pairs

We describe the development and validation of an analysis pipeline for detection of focal somatic copy number aberrations from next-generation sequencing (NGS) data. In the clinical setting, tumor samples may be derived from historical tissue or from recently collected fresh frozen biopsies. DNA extracted from samples preserved as formalin-fixed, paraffin-embedded (FFPE) tissue are often highly degraded. We describe challenges inherent to sequencing different tissue sources and algorithmic approaches for optimizing detection of focal alterations in these samples with a high degree of accuracy. The first area of optimization takes into account clonal or fragment coverage based on user-defined intervals by accounting for coverage between paired-end reads. Normalized log2 fold-changes between tumor and normal samples are then calculated and a smoothing window is applied. In addition, we utilize tumor allele frequencies of known heterozygous germline SNPs identified within the normal to both evaluate potential false positives and correct biases. Lastly, a segmentation algorithm (circular binary segmentation as implemented in DNAcopy package1) is applied to summarize the individual log2 fold-changes into intervals with a constant copy number state. The results from the segmentation algorithm are then annotated and reformatted into the VCF format. We have analytically optimized and validated our algorithm across a series of samples with known alterations for implementation on both clinical and research samples.

1. Seshan VE and Olshen A. DNAcopy: DNA copy number data analysis. R package version 1.40.0.

Citation Format: Jessica Aldrich, Jonathan J. Keats, Winnie S. Liang, John D. Carpten, David W. Craig. Detection of focal somatic copy number variants in whole genome, whole exome, and targeted next-generation sequencing data of tumor/normal pairs. [abstract]. In: Proceedings of the AACR Precision Medicine Series: Integrating Clinical Genomics and Cancer Therapy; Jun 13-16, 2015; Salt Lake City, UT. Philadelphia (PA): AACR; Clin Cancer Res 2016;22(1_Suppl):Abstract nr 45.

An integrated framework for reporting clinically relevant biomarkers from paired tumor/normal genomic and transcriptomic sequencing data in support of clinical trials in personalized medicine

The ability to rapidly sequence the tumor and germline DNA of an individual holds the eventual promise of revolutionizing our ability to match targeted therapies to tumors harboring the associated genetic biomarkers. Analyzing high throughput genomic data consisting of millions of base pairs and discovering alterations in clinically actionable genes in a structured and real time manner is at the crux of personalized testing. This requires a computational architecture that can monitor and track a system within a regulated environment as terabytes of data are reduced to a small number of therapeutically relevant variants, delivered as a diagnostic laboratory developed test. These high complexity assays require data structures that enable real-time and retrospective ad-hoc analysis, with a capability of updating to keep up with the rapidly changing genomic and therapeutic options, all under a regulated environment that is relevant under both CMS and FDA depending on application. We describe a flexible computational framework that uses a paired tumor/normal sample allowing for complete analysis and reporting in approximately 24 hours, providing identification of single nucleotide changes, small insertions and deletions, chromosomal rearrangements, gene fusions and gene expression with positive predictive values over 90%. In this paper we present the challenges in integrating clinical, genomic and annotation databases to provide interpreted draft reports which we utilize within ongoing clinical research protocols. We demonstrate the need to retire from existing performance measurements of accuracy and specificity and measure metrics that are meaningful to a genomic diagnostic environment. This paper presents a three-tier infrastructure that is currently being used to analyze an individual genome and provide available therapeutic options via a clinical report. Our framework utilizes a non-relational variant-centric database that is scaleable to a large amount of data and addresses the challenges and limitations of a relational database system. Our system is continuously monitored via multiple trackers each catering differently to the diversity of users involved in this process. These trackers designed in analytics web-app framework provide status updates for an individual sample accurate to a few minutes. In this paper, we also present our outcome delivery process that is designed and delivered adhering to the standards defined by various regulation agencies involved in clinical genomic testing.

Abstract 2375: A method to identify copy number aberrations (CNAs) from whole exome sequence (WES) data and its application to multiple myeloma cell lines and patient samples

Duplications and deletions of large-scale genomic regions, including whole chromosomes, are a hallmark of malignant tumors and represent an important predictor of outcome for many tumor types. In multiple myeloma, a core set of structural variants are regularly used for disease prognosis (e.g., del17p, t(4;14), t(14,16)). The presence of one or more of these lesions is predictive of a more aggressive disease and poorer outcomes for patients. The application of next-generation sequencing (NGS) to cancer samples has enhanced our ability to detect prognostic and predictive copy number aberrations (CNAs). However, most NGS-based methods for the identification of CNAs in tumors require whole genome sequencing (WGS). Relatively fewer methods exist for the detection of CNAs from exome sequencing and, to our knowledge, these methods all rely on having matched normal data. However, WES is still far more common than WGS and matched normal samples are frequently not available. Therefore, we have developed a novel method for the identification of CNAs from single-sample WES data. Our method combines LOWESS smoothing and discrete-wavelet-transformation to normalize the exome coverage data with an HMM for coverage segmentation. We have applied our new method to WES data from 40 myeloma cell-lines for which aCGH copy number calls were also available for training and validation. Our method shows a high correlation with the aCGH calls (mean = 95%). In addition we have currently analyzed 42 (of 102) tumor samples that were collected during screening of relapsed/refractory multiple myeloma patients on three Onyx-sponsored Phase 2 clinical trials of the proteasome inhibitor carfilzomib (PX-171-003,PX-171-004, PX-171-005). Based on analysis of the completed subset we find that the frequency of known and novel lesions within these samples is similar to what has previously been observed in other publically available myeloma CNA datasets. The most frequent event that we find is loss of chr13 followed by gains on chr1q and chr15 all at near 40% frequency. We also find gains of chr3, chr5, chr9 and chr11 in ∼30% of patients. These results indicate that we can apply this new method to our larger set of clinical samples in order to discover new prognostic markers for patient outcomes and response to therapy in MM.

Citation Format: Jeremiah D. Degenhardt, Kenneth B. Hoehn, Kevin A. Kwei, Kristi Stephenson, Jonathan J. Keats, Chris J. Kirk, Brian B. Tuch. A method to identify copy number aberrations (CNAs) from whole exome sequence (WES) data and its application to multiple myeloma cell lines and patient samples. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 2375. doi:10.1158/1538-7445.AM2014-2375

Abstract 898: Expression of immunoglobulin and its receptor are major determinants of multiple myeloma patient sensitivity to proteasome inhibitors

Several therapies are now approved for multiple myeloma treatment, creating a pressing need to predict which drug or combination of drugs will be most effective and safe for each patient.  Proteasome inhibitors (PrIs), including carfilzomib and bortezomib, are one such class of drugs that has become a standard therapy across all lines of myeloma treatment. Despite extensive study, the mechanism of selective tumor cell death following proteasome inhibition is poorly understood. However, the uniquely high sensitivity of myeloma cells to proteasome inhibition, the high burden of protein (immunoglobulin) secretion these cells experience, and the key role of the proteasome in maintaining protein homeostasis together point toward a unifying model in which protein load drives tumor sensitivity to PrIs. This simple model is supported by published studies of myeloma cell lines. As part of company-sponsored Phase II & III clinical trials of single-agent PrIs, CD138+ tumor cells collected during patient screening were banked for comprehensive genomic analyses.  Here, examining our carfilzomib RNA-Seq data along with publically-available bortezomib microarray-based RNA data, we find a strong association between higher immunoglobulin (Ig) expression and sensitivity to both carfilzomib (N = 60; P = 3x10-3) and bortezomib (N = 167; P = 2x10-4). In fact, using IGH expression alone, we are able to classify carfilzomib response with 56% sensitivity and 91% specificity. Median time to progression for IGH-High carfilzomib patients was 5.4-fold longer than for IGH-Low patients (7.3 vs. 1.3 months; P = 0.005). As expected for a bona fide predictive biomarker of proteasome inhibition, no association was found between IGH expression and response to single-agent, high-dose dexamethasone (P = 0.82). Looking more comprehensively, we identified a large set of associations between gene expression and carfilzomib response (N = 361 at FDR &lt; 15%). The set is strongly enriched for genes encoding structural folds of the Ig superfamily (P &lt; 10-13), implying that high expression of this class of proteins sensitizes cells to PrIs. One such gene is Fc gamma receptor 2B (FCGR2B), which binds Ig and down-modulates its production in B cells. High FCGR2B expression, therefore, may demarcate a tumor that is experiencing particularly high levels of proteotoxic stress from Ig production. Combining FCGR2B and IGH expression, we are able to classify carfilzomib response with 70% sensitivity and 94% specificity. Median time to progression for IGH/FCGR2B+ patients was 7.3-fold longer than for IGH/FCGR2B- patients (8.9 vs. 1.2 months; P = 3x10-5). IGH expression is, to our knowledge, the first retrospectively validated biomarker for this important class of anti-tumor agents and therefore represents a promising basis for the development of a PrI companion diagnostic.

Citation Format: Brian B. Tuch, Andrea Loehr, Jeremiah D. Degenhardt, Kevin A. Kwei, Eric Lowe, Kristi Stephenson, Jonathan J. Keats, Christopher J. Kirk. Expression of immunoglobulin and its receptor are major determinants of multiple myeloma patient sensitivity to proteasome inhibitors. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 898. doi:10.1158/1538-7445.AM2014-898

Widespread genetic heterogeneity in multiple myeloma: implications for targeted therapy

We performed massively parallel sequencing of paired tumor/normal samples from 203 multiple myeloma (MM) patients and identified significantly mutated genes and copy number alterations and discovered putative tumor suppressor genes by determining homozygous deletions and loss of heterozygosity. We observed frequent mutations in KRAS (particularly in previously treated patients), NRAS, BRAF, FAM46C, TP53, and DIS3 (particularly in nonhyperdiploid MM). Mutations were often present in subclonal populations, and multiple mutations within the same pathway (e.g., KRAS, NRAS, and BRAF) were observed in the same patient. In vitro modeling predicts only partial treatment efficacy of targeting subclonal mutations, and even growth promotion of nonmutated subclones in some cases. These results emphasize the importance of heterogeneity analysis for treatment decisions.

Xbp1s-negative tumor B cells and pre-plasmablasts mediate therapeutic proteasome inhibitor resistance in multiple myeloma

Proteasome inhibitor (PI) resistance mechanisms in multiple myeloma (MM) remain controversial. We report the existence of a progenitor organization in primary MM that recapitulates maturation stages between B cells and plasma cells and that contributes to clinical PI resistance. Xbp1s(-) tumor B cells and pre-plasmablasts survive therapeutic PI, preventing cure, while maturation arrest of MM before the plasmablast stage enables progressive disease on PI treatment. Mechanistically, suppression of Xbp1s in MM is shown to induce bortezomib resistance via de-commitment to plasma cell maturation and immunoglobulin production, diminishing endoplasmic reticulum (ER) front-loading and cytotoxic susceptibility to PI-induced inhibition of ER-associated degradation. These results reveal the tumor progenitor structure in MM and highlight its role in therapeutic failure.

Genomic imbalance of HMMR/RHAMM regulates the sensitivity and response of malignant peripheral nerve sheath tumour cells to aurora kinase inhibition

Malignant peripheral nerve sheath tumours (MPNST) are rare, hereditary cancers associated with neurofibromatosis type I. MPNSTs lack effective treatment options as they often resist chemotherapies and have high rates of disease recurrence. Aurora kinase A (AURKA) is an emerging target in cancer and an aurora kinase inhibitor (AKI), termed MLN8237, shows promise against MPNST cell lines in vitro and in vivo. Here, we test MLN8237 against two primary human MPNST grown in vivo as xenotransplants and find that treatment results in tumour cells exiting the cell cycle and undergoing endoreduplication, which cumulates in stabilized disease. Targeted therapies can often fail in the clinic due to insufficient knowledge about factors that determine tumour susceptibilities, so we turned to three MPNST cell-lines to further study and modulate the cellular responses to AKI. We find that the sensitivity of cell-lines with amplification of AURKA depends upon the activity of the kinase, which correlates with the expression of the regulatory gene products TPX2 and HMMR/RHAMM. Silencing of HMMR/RHAMM, but not TPX2, augments AURKA activity and sensitizes MPNST cells to AKI. Furthermore, we find that AURKA activity is critical to the propagation and self-renewal of sphere-enriched MPNST cancer stem-like cells. AKI treatment significantly reduces the formation of spheroids, attenuates the self-renewal of spheroid forming cells, and promotes their differentiation. Moreover, silencing of HMMR/RHAMM is sufficient to endow MPNST cells with an ability to form and maintain sphere culture. Collectively, our data indicate that AURKA is a rationale therapeutic target for MPNST and tumour cell responses to AKI, which include differentiation, are modulated by the abundance of HMMR/RHAMM.

Genomic analysis of marginal zone and lymphoplasmacytic lymphomas identified common and disease-specific abnormalities

Lymphoplasmacytic lymphomas and marginal zone lymphomas of nodal, extra-nodal and splenic types account for 10% of non-Hodgkin lymphomas. They are similar at the cell differentiation level, sometimes making difficult to distinguish them from other indolent non-Hodgkin lymphomas. To better characterize their genetic basis, we performed array-based comparative genomic hybridization in 101 marginal zone lymphomas (46 MALT, 35 splenic and 20 nodal marginal zone lymphomas) and 13 lymphoplasmacytic lymphomas. Overall, 90% exhibited copy-number abnormalities. Lymphoplasmacytic lymphomas demonstrated the most complex karyotype (median=7 copy-number abnormalities), followed by MALT (4), nodal (3.5) and splenic marginal zone lymphomas (3). A comparative analysis exposed a group of copy-number abnormalities shared by several or all the entities with few disease-specific abnormalities. Gain of chromosomes 3, 12 and 18 and loss of 6q23-q24 (TNFAIP3) were identified in all entities. Losses of 13q14.3 (MIRN15A-MIRN16-1) and 17p13.3-p12 (TP53) were found in lymphoplasmacytic and splenic marginal zone lymphomas; loss of 11q21-q22 (ATM) was found in nodal, splenic marginal zone and lymphoplasmacytic lymphomas and loss of 7q32.1-q33 was found in MALT, splenic and lymphoplasmacytic lymphomas. Abnormalities affecting the nuclear factor kappa B pathway were observed in 70% of MALT and lymphoplasmacytic lymphomas and 30% of splenic and nodal marginal zone lymphomas, suggesting distinct roles of this pathway in the pathogenesis/progression of these subtypes. Elucidation of the genetic alterations contributing to the pathogenesis of these lymphomas may guide to design-specific therapeutic approaches.

Abstract 219: TP53 loss: An overriding marker of disease progression in Multiple Myeloma

Multiple Myeloma (MM) is an incurable plasma-cell disorder progressing from indolent monoclonal gammopathy of undetermined significance (MGUS) and smoldering myeloma (SMM) to relapse MM and finally plasma cell leukemia (PCL). Alterations affecting TP53 are among the principal genetic progression events associated with disease progression.

We analyzed results from gene expression profiling (GEP), array-based comparative genomic hybridization (aCGH), fluorescent in situ hybridization (FISH) and mutational data on the largest cohort published to date of 806 patients from all MM stages and 48 human myeloma cell lines (HMCLs).

The prevalence of monoallelic TP53 deletions by aCGH is 1.4%, 3.7%, 8.9% and 25%, in MGUS, SMM, newly diagnosed and relapsed MM respectively. It reaches 53% in HMCLs and 75% in secondary PCL. No biallelic TP53 deletion were observed in patients and only 8% in HMCLs. In relapsed MM, the prevalence of TP53 deletions increases from 12/75 patients (16%) at first, to 9/31 patients (29%) in second and 5/7 patients (71%) in third relapse or more. We studied by FISH the status of TP53 in 113 patients from early stages progressing to relapse. Overall 15/113 patients had TP53 deletion, we confirmed that 10/15 patients acquired TP53 deletion at late stages. Next, we investigated the mutational status of TP53 in 59 relapsed MM patients, 5/59 samples (8.5%) had TP53 mutations, 4/5 mutations were located in the DNA binding domain. Presence of TP53 deletion conferred poor overall survival in relapsed MM patients (4.2 vs. 37.8 months, p = 0.015).

A cohort of 239 patients had GEP-aCGH paired analysis we used a validated 70-gene model to stratify patients based on calculations obtained by GEP. The cohort was divided into quartiles and we observed the prevalence of TP53 in the low and high-risk groups. TP53 deletion was observed in 33/239 patients (14%) by aCGH, 2/33 (6%) had GEP values in the low-risk and 13/33 (39%) in the high-risk group. Next, we used a previously proposed cut-off value predicting TP53 deletion when GEP values are less than 733. We tried to validate this value as a surrogate for direct detection of TP53 deletion, and used the positive predictive value (PPV) of such level as a potential clinically useful marker. The fact that aCGH is the gold standard approach to define copy-number alterations, the proposed cut-off value failed to predict TP53 deletion as the PPV was low at 18% and the sensitivity of 69%.

Finally, we studied by aCGH the status of MDM2 and CDKN2A, two key regulators of TP53. MDM2 gain and CDKN2A loss were infrequent events in patients even in advanced disease. In 48 HMCLs, only one had MDM2 copy gain (2%) while 14 cell lines had CDKN2A loss (29%).

In MM, emergence of TP53 deletion/mutations in MM and its increase in aggressive stages of disease remain the overriding genetic factor for poor prognosis determination and a marker of progressive genetic events. Based on GEP, the 733 cut-off value did not accurately predict TP53 deletion.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 219. doi:10.1158/1538-7445.AM2011-219

Classical and/or alternative NF-kappaB pathway activation in multiple myeloma

Mutations involving the nuclear factor-kappaB (NF-kappaB) pathway are present in at least 17% of multiple myeloma (MM) tumors and 40% of MM cell lines (MMCLs). These mutations, which are apparent progression events, enable MM tumors to become less dependent on bone marrow signals that activate NF-kappaB. Studies on a panel of 51 MMCLs provide some clarification of the mechanisms through which these mutations act and the significance of classical versus alternative activation of NF-kappaB. First, only one mutation (NFKB2) selectively activates the alternative pathway, whereas several mutations (CYLD, NFKB1, and TACI) selectively activate the classical pathway. However, most mutations affecting NF-kappaB-inducing kinase (NIK) levels (NIK, TRAF2, TRAF3, cIAP1&2, and CD40) activate the alternative but often both pathways. Second, we confirm the critical role of TRAF2 in regulating NIK degradation, whereas TRAF3 enhances but is not essential for cIAP1/2-mediated proteasomal degradation of NIK in MM. Third, using transfection to selectively activate the classical or alternative NF-kappaB pathways, we show virtually identical changes in gene expression in one MMCL, whereas the changes are similar albeit nonidentical in a second MMCL. Our results suggest that MM tumors can achieve increased autonomy from the bone marrow microenvironment by mutations that activate either NF-kappaB pathway.

Abstract 4957: Comparison of commercial and custom microarrays: which is the best choice for copy-number analysis in tumor samples

Microarray technologies are powerful assays, which can identify copy-number abnormalities (CNA) underlying the pathogenesis of cancer. Several commercial arrays are available but it is unclear which is the best platform.

We did a side-by-side comparison between two leading arrays; the Affymetrix SNP6 and Agilent 1M, to compare data compression, noise, gene and exon coverage, and CNA detection. Five samples were run on both platforms and analyzed as recommended by the manufacturer. We first performed an in silico comparison to define the number of probes per serial 5k interval, number of probes per gene and exon, and the coefficient of variation of different smoothing windows. Comparing the CV of the two arrays showed a similar level of variation between a 3 and 5 probe averaging on the 1M and SNP6, respectively. Using these smoothing levels the 1M interrogates 81.2% of the genes in the genome compared to 69.3% on the SNP6. At the exon level the 1M has at least one probe in 64.8% compared to 41.2% on the SNP6. We observed a similar level of data compression (∼10%) on the two arrays but the SNP6 had nearly twice the noise (SD 0.38 versus 0.21). Finally, there was a marked increment in the number of CNA identified by 1M (mean of 78.5) than SNP6.0 (59.5). In this comparison the Agilent 1M out performed the Affymetrix SNP6. Even with half the number of probes, the 1M was superior in CNA detection, gene/exon coverage and showed less signal variation.

A comparison with previous platforms showed that these two arrays did not identify new gene amplifications but did detect new deletions, however, neither platform detected several known deletions detected by other methods. Therefore, we designed a custom, exon-centered, 2×400K array (Agilent) with the goal of identifying additional deletions. The design outline was: a) all the probes from the Agilent CGH 44K array; b) 3 probes per exon across the human exome; c) high resolution tiling of 1450 cancer implicated genes; and d) tiling the 4kb flanking all MIRs. To improve performance, we only included probes with performance scores higher than 0.8 and 0.85 in queries b) and c), respectively. Lastly, 15000 probes from the 1M array that map to exons missed by our design parameters were also included. Our custom array has half the probes (415K) of the 1M (960K) and 4.5x less than the SNP6 (1.85M) but shows the best gene and exon coverage. When the minimum number of probes needed for calculating a CNA is considered, 86% of genes are interrogated compared to 81.2% (1M) and 69.3% (SNP6). The difference is even more marked at the exon level with 84% covered by at least one probe and at least 63.2% interrogated by two probes compared to 5.3% (1M) and 18.0% (SNP6). The promising in silico comparison of this exon-centered array, its more stringent design parameters in combination with significantly reducing the cost per run should make this a very promising array design for CNA analysis of tumor samples in the future

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 4957.

Correlation between array-comparative genomic hybridization-defined genomic gains and losses and survival: identification of 1p31-32 deletion as a prognostic factor in myeloma

In this study, we correlated array-comparative genomic hybridization-defined abnormalities with survival in two different cohorts of patients treated with therapy based on high-dose melphalan with autologous stem-cell transplantation (64 from the Mayo Clinic and 67 from the University of Arkansas Medical School) and identified that several regions of genomic gains and losses were significantly associated with poorer survival. Three noncontiguous survival relevant regions covering 1p31-33 and two noncontiguous regions covering 20p12.3-12.1 were common between the two datasets. The prognostic relevance of these hotspots was validated in an independent cohort using fluorescent in situ hybridization, which showed that 1p31-32 loss is significantly associated with shorter survival (24.5 months versus 40 months, log-rank P-value=0.01), whereas 20p12 loss has a trend toward shorter survival (26.3 months versus 40 months, log-rank P-value=0.06). On multivariate analysis, 1p31-32 loss is an independent prognostic factor. On further analysis, the prognostic impact of 1p31-32 loss is due to shortening of post-relapse survival as there is no impact on complete response rates and progression-free survival.

High-resolution genomic analysis in Waldenström's macroglobulinemia identifies disease-specific and common abnormalities with marginal zone lymphomas

Cytogenetic analyses have been historically limited in Waldenström's macroglobulinemia (WM) by the difficulty to obtain tumor metaphases. Thus, few recurrent karyotypic abnormalities have been reported and the molecular consequences of these imbalances are largely unknown. We used an array-based comparative genomic hybridization approach to better characterize the recurrent chromosome abnormalities associated with WM pathogenesis and to compare them with the publicly available findings in other B-cell neoplasias. The majority of the recurrent chromosome abnormalities identified in WM were shared with marginal zone lymphomas (MZL), as deletions of 6q23 and 13q14 and gains of 3q13-q28, 6p and 18q. On the other hand, gains of 4q and 8q were recurrently identified in WM but have not been described as being common abnormalities in MZL. The genetic consequences of these specific abnormalities remain elusive and further studies are critical to refine the search and to precise the molecular pathways affected by these abnormalities.

Identification of copy number abnormalities and inactivating mutations in two negative regulators of nuclear factor-kappaB signaling pathways in Waldenstrom's macroglobulinemia

Waldenström's macroglobulinemia (WM) is a distinct clinicobiological entity defined as a B-cell neoplasm characterized by a lymphoplasmacytic infiltrate in bone marrow (BM) and IgM paraprotein production. Cytogenetic analyses were historically limited by difficulty in obtaining tumor metaphases, and the genetic basis of the disease remains poorly defined. Here, we performed a comprehensive analysis in 42 WM patients by using a high-resolution, array-based comparative genomic hybridization approach to unravel the genetic mechanisms associated with WM pathogenesis. Overall, 83% of cases have chromosomal abnormalities, with a median of three abnormalities per patient. Gain of 6p was the second most common abnormality (17%), and its presence was always concomitant with 6q loss. A minimal deleted region, including MIRN15A and MIRN16-1, was delineated on 13q14 in 10% of patients. Of interest, we reported biallelic deletions and/or inactivating mutations with uniparental disomy in tumor necrosis factor (TNF) receptor-associated factor 3 and TNFalpha-induced protein 3, two negative regulators of the nuclear factor-kappaB (NF-kappaB) signaling pathway. Furthermore, we confirmed the association between TRAF3 inactivation and increased transcriptional activity of NF-kappaB target genes. Mutational activation of the NF-kappaB pathway, which is normally activated by ligand receptor interactions within the BM microenvironment, highlights its biological importance, and suggests a therapeutic role for inhibitors of NF-kappaB pathway activation in the treatment of WM.

Nonredundant and complementary functions of TRAF2 and TRAF3 in a ubiquitination cascade that activates NIK-dependent alternative NF-kappaB signaling

The adaptor and signaling proteins TRAF2, TRAF3, cIAP1 and cIAP2 may inhibit alternative nuclear factor-kappaB (NF-kappaB) signaling in resting cells by targeting NF-kappaB-inducing kinase (NIK) for ubiquitin-dependent degradation, thus preventing processing of the NF-kappaB2 precursor protein p100 to release p52. However, the respective functions of TRAF2 and TRAF3 in NIK degradation and activation of alternative NF-kappaB signaling have remained elusive. We now show that CD40 or BAFF receptor activation result in TRAF3 degradation in a cIAP1-cIAP2- and TRAF2-dependent way owing to enhanced cIAP1, cIAP2 TRAF3-directed ubiquitin ligase activity. Receptor-induced activation of cIAP1 and cIAP2 correlated with their K63-linked ubiquitination by TRAF2. Degradation of TRAF3 prevented association of NIK with the cIAP1-cIAP2-TRAF2 ubiquitin ligase complex, which resulted in NIK stabilization and NF-kappaB2-p100 processing. Constitutive activation of this pathway causes perinatal lethality and lymphoid defects.

Promiscuous mutations activate the noncanonical NF-kappaB pathway in multiple myeloma

Activation of NF-kappaB has been noted in many tumor types, however only rarely has this been linked to an underlying genetic mutation. An integrated analysis of high-density oligonucleotide array CGH and gene expression profiling data from 155 multiple myeloma samples identified a promiscuous array of abnormalities contributing to the dysregulation of NF-kappaB in approximately 20% of patients. We report mutations in ten genes causing the inactivation of TRAF2, TRAF3, CYLD, cIAP1/cIAP2 and activation of NFKB1, NFKB2, CD40, LTBR, TACI, and NIK that result primarily in constitutive activation of the noncanonical NF-kappaB pathway, with the single most common abnormality being inactivation of TRAF3. These results highlight the critical importance of the NF-kappaB pathway in the pathogenesis of multiple myeloma.

Ten years and counting: so what do we know about t(4;14)(p16;q32) multiple myeloma

Multiple myeloma is a genetically heterogenous disease with a wide variety of characterized genetic aberrations. Until recently, the impact of these aberrations on patient outcome was not known. However, in the last 5-10 years, several genetic markers have been linked to patient outcome. One of the strongest predictors of outcome identified to date is t(4;14)(p16;q32). Although this translocation is tightly linked to chromosome 13 deletions, another poor prognosis marker, it is becoming apparent that the translocation and not the deletion of 13 is the important factor. Unfortunately, despite the known association with outcome, an understanding of the mechanism(s) whereby the translocation contributes to developing and maintaining this aggressive form of myeloma remains elusive.