Search

We present a unique database of tumor samples from 399 multiple myeloma (MM) patients, whose functional genomic landscape was assessed by integrating ex vivo drug sensitivity to 138 drugs, clinical variables, cytogenetics, mutational profiles and transcriptomes. These analyses revealed a novel MM transcriptomic topology that, when associated with ex vivo drug sensitivity, generates “footprints” having both mechanistic and predictive applications. We validated the transcriptomic footprint for the nuclear export inhibitor selinexor in two clinical trials, BOSTON and MCC17814, and showed such a gene signature can accurately classify clinical response and suggest mechanism of resistance. Finally, we propose a novel evolutionary-based therapeutic strategy involving sequential therapy of monoclonal antibody daratumumab and selinexor, based on anti-correlative transcriptomic footprints, which was validated in two clinical trials (STOMP and XPORT-MM-028). Collectively, this database and computational framework can be leveraged to inform underlying biology, and to identify novel therapeutic strategies to improve treatment of MM.

Multiple myeloma is an incurable hematological malignancy that impacts tens of thousands of people every year in the United States. Treatment for eligible patients involves induction, consolidation with stem cell rescue, and maintenance. High-dose therapy with a DNA alkylating agent, melphalan, remains the primary drug for consolidation therapy in conjunction with autologous stem-cell transplantation; as such, melphalan resistance remains a relevant clinical challenge. Here, we describe a proteometabolomic approach to examine mechanisms of acquired melphalan resistance in two cell line models. Drug metabolism, steady-state metabolomics, activity-based protein profiling (ABPP, data available at PRIDE: PXD019725), acute-treatment metabolomics, and western blot analyses have allowed us to further elucidate metabolic processes associated with melphalan resistance. Proteometabolomic data indicate that drug-resistant cells have higher levels of pentose phosphate pathway metabolites. Purine, pyrimidine, and glutathione metabolisms were commonly altered, and cell-line-specific changes in metabolite levels were observed, which could be linked to the differences in steady-state metabolism of naïve cells. Inhibition of selected enzymes in purine synthesis and pentose phosphate pathways was evaluated to determine their potential to improve melphalan's efficacy. The clinical relevance of these proteometabolomic leads was confirmed by comparison of tumor cell transcriptomes from newly diagnosed MM patients and patients with relapsed disease after treatment with high-dose melphalan and autologous stem-cell transplantation. The observation of common and cell-line-specific changes in metabolite levels suggests that omic approaches will be needed to fully examine melphalan resistance in patient specimens and define personalized strategies to optimize the use of high-dose melphalan.

The clinical utility of histone/protein deacetylase (HDAC) inhibitors in combinatorial regimens with proteasome inhibitors for patients with relapsed and refractory multiple myeloma (MM) is often limited by excessive toxicity due to HDAC inhibitor promiscuity with multiple HDACs. Therefore, more selective inhibition minimizing off-target toxicity may increase the clinical effectiveness of HDAC inhibitors. We demonstrated that plasma cell development and survival are dependent upon HDAC11, suggesting this enzyme is a promising therapeutic target in MM. Mice lacking HDAC11 exhibited markedly decreased plasma cell numbers. Accordingly, in vitro plasma cell differentiation was arrested in B cells lacking functional HDAC11. Mechanistically, we showed that HDAC11 is involved in the deacetylation of IRF4 at lysine103. Further, targeting HDAC11 led to IRF4 hyperacetylation, resulting in impaired IRF4 nuclear localization and target promoter binding. Importantly, transient HDAC11 knockdown or treatment with elevenostat, an HDAC11-selective inhibitor, induced cell death in MM cell lines. Elevenostat produced similar anti-MM activity in vivo, improving survival among mice inoculated with 5TGM1 MM cells. Elevenostat demonstrated nanomolar ex vivo activity in 34 MM patient specimens and synergistic activity when combined with bortezomib. Collectively, our data indicated that HDAC11 regulates an essential pathway in plasma cell biology establishing its potential as an emerging theraputic vulnerability in MM.

The plasma cell malignancy multiple myeloma (MM) evolves from a pre-malignant state and remains all but incurable due to emergence of therapy resistance. Despite intensive analyses, mechanisms driving MM progression and refractory disease are poorly understood. Integrating topologic, expression and epigenetic analyses of 1,016 patient specimens, we report super-enhancer core regulatory circuits (SECRCs) that drive and sustain MM epigenetic states. Reprogramming of cell identity and tumor microenvironment genes drive malignant conversion, while alterations in cell cycle and metabolic control genes cause refractory disease. Thus, select epigenetic states drive progression and therapy resistance, providing strategies to prevent and effectively treat MM.

Background: Multiagent therapies, due to their ability to delay or overcome resistance, are a hallmark of treatment in multiple myeloma (MM). The growing number of therapeutic options in MM requires high-throughput combination screening tools to better allocate treatment, and facilitate personalized therapy. Methods: A second-order drug response model was employed to fit patient-specific ex vivo responses of 203 MM patients to single-agent models. A novel pharmacodynamic model, developed to account for two-way combination effects, was tested with 130 two-drug combinations. We have demonstrated that this model is sufficiently parameterized by single-agent and fixed-ratio combination responses, by validating model estimates with ex vivo combination responses for different concentration ratios, using a checkerboard assay. This new model reconciles ex vivo observations from both Loewe and BLISS synergy models, by accounting for the dimension of time, as opposed to focusing on arbitrary time-points or drug effect. Clinical outcomes of patients were simulated by coupling patient-specific drug combination models with pharmacokinetic data. Findings: Combination screening showed 1 in 5 combinations (21.43% by LD50, 18.42% by AUC) were synergistic ex vivo with statistical significance (P < 0.05), but clinical synergy was predicted for only 1 in 10 combinations (8.69%), which was attributed to the role of pharmacokinetics and dosing schedules. Interpretation: The proposed framework can inform clinical decisions from ex vivo observations, thus providing a path toward personalized therapy using combination regimens. Funding: This research was funded by the H. Lee Moffitt Cancer Center Physical Sciences in Oncology (PSOC) Grant (1U54CA193489-01A1) and by H. Lee Moffitt Cancer Center's Team Science Grant. This work has been supported in part by the PSOC Pilot Project Award (5U54CA193489-04), the Translational Research Core Facility at the H. Lee Moffitt Cancer Center & Research Institute, an NCI-designated Comprehensive Cancer Center (P30-CA076292), the Pentecost Family Foundation, and Miles for Moffitt Foundation. Keywords: Clinical synergy, Multiple myeloma, Ex vivo, Pharmacodynamic model, High-throughput combination screening

Multiple Myeloma (MM) remains an incurable malignancy, despite the advent of several new therapeutic agents, including immunomodulatory drugs (IMiDs, e.g., Lenalidomide (Len)) and proteasome inhibitors (PIs, e.g., Bortezomib (Btz)). Accordingly, there is an urgent need to identify new targetable vulnerabilities for MM patients. We developed an ex vivo 384-well platform that allows one to define drug sensitivities of primary patient CD138+ MM cells in the context of a reconstructed tumor microenvironment (TME), including allogeneic bone marrow stromal cells, extracellular matrix and MM patient serum. Using this platform and activity-based proteomic profiling (ABPP), we identified shared signaling pathways induced by the interactions of MM with stromal cells and integrated these data with screens performed using a bank of protein kinase inhibitors (PKI) and current anti-MM therapeutics. These analyses revealed that the serine/threonine kinases casein kinase-1δ (CK1δ) and CK1ε as high priority targets for MM. Indeed, a highly selective and potent dual inhibitor of CK1δ/CK1ε coined SR-3029 is the most potent PKI versus MM. Further, our studies revealed SR-3029 has potent activity in 138/153 primary patient MM specimens tested thus far, including quad and penta-refractory MM. Analysis of RNAseq data of over 600 Moffitt Cancer Center (MCC) MM patients revealed that patients with high expression of CK1ε had worse survival outcomes while no survival difference was seen with CK1δ expression. Importantly, using the established 5TGM1/Kal-Ridge (C57B6/KaLwRijHsd) syngeneic mouse model of multiple myeloma, we show that tumors derived from 5TGM1 MM cells, which rapidly die following exposure to SR-3029 ex vivo, are also sensitive to CK1δ/CK1ε inhibition in vivo, where SR-3029 treatment reduced tumor burden and significantly improved survival. Similar results were observed using NSG immune compromised animals inoculated with human MM1.S multiple myeloma cells (both flank and tail vein models), where SR-3029 treated animals had reduced tumor burden and extended survival. Analysis of RNAseq on patients' samples (on stroma) treated ex vivo with SR-3029 revealed CK1δ/CK1ε inhibition suppressed multiple metabolic pathways (oxidative phosphorylation, glycolysis, xenobiotic metabolism). Interestingly, analyses of MCC MM patient RNAseq data revealed upregulation of the genes identified in these metabolic pathways as patients progress from pre-treatment to relapse, and that patient MM samples that were resistant to CK1δ/CK1ε inhibition had an upregulation of some of these metabolic genes. Functional studies are being performed to define the mechanism(s) by which CK1δ/CK1ε inhibition disables MM metabolism. Collectively, these findings establish CK1ε and/or CK1δ as attractive targets for anti-myeloma therapy that are required to sustain MM metabolism. Disclosures Dai: M2Gen: Employment. Shain:Bristol-Myers Squibb: Membership on an entity's Board of Directors or advisory committees; Takeda: Membership on an entity's Board of Directors or advisory committees; Sanofi Genzyme: Membership on an entity's Board of Directors or advisory committees; AbbVie: Research Funding; Janssen: Membership on an entity's Board of Directors or advisory committees; Adaptive Biotechnologies: Consultancy; Amgen: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees.

Problem: Multiple myeloma (MM) is a treatable yet incurable hematologic cancer that lacks predictive biomarkers. Approach: Here we apply a systems biology approach to determine patient-specific mechanisms, as well as signatures of drug resistance in MM. To achieve this goal, we have combined ex vivo drug sensitivity data from 307 MM fresh primary samples tested with 162 drugs and combinations, with paired molecular data (RNAseq and mutational profiling) from a larger overlapping cohort of 606 MM samples from Moffitt's Multiple Myeloma Working Group (MMWG) repository in collaboration with M2Gen/Oncology Research Information Exchange Network (ORIEN). With the purpose of decoupling biological function from intracellular control mechanisms, we have re-constructed a MM-specific transcriptional regulatory network composed of clusters of co-expressing genes. We demonstrate how this gene cluster network regulates biology, and how different biological functions (e.g. Proteasome, Ribosome, Oxidative Phosphorylation) share common regulatory circuits. We have used gene set enrichment analysis (GSEA) to identify gene clusters with transcriptional profiles, and investigated mutations associated with drug resistance. Results: As a preliminary validation of this approach, we have confirmed established mechanisms of resistance (MOR) to targeted therapies, as well as proposed novel MOR to clinically relevant and experimental drugs in MM, as well as putative synergistic drug combinations. In addition, we have identified a list of low frequency mutations ( We have also explored evolution of drug resistance in sequential samples. Consistent with altered transcriptional programming in therapeutic escape, single sample GSEA demonstrated cumulative dysregulation of cancer-related genes with increasing lines of therapy. We have identified 60 MM-specific transcriptional core auto-regulatory circuits (CRC) correlated with ex vivo drug resistance, suggesting that characterization of transcriptional regulatory circuits is a critical approach to infer mechanisms of MM resistance, and providing a novel rationale for combination therapy. We hypothesized that identifying and targeting these transcriptional CRCs could facilitate patient-specific rational combination therapies, with the goal to overcome therapy resistance in the clinic. As proof-of-principle, we have identified a novel transcriptional network consisting of 3 of these CRCs (FOXP1, JUNB and JUN) associated with BCL2 inhibitor (BCL2i) response in MM. Our preliminary data suggests that this transcriptional regulatory circuit is associated to t(11;14) MM through CCND1 up-regulation, but is also present in non-t(11;14) BCL2i-sensitive primary samples, and can be modulated to induce BCL2i sensitivity in non-t(11;14) MM through HDAC inhibitors. Conclusion and next steps: Preliminary results confirm the potential of this combination of unsupervised and supervised, yet functionally testable approach, to infer novel, and patient-specific MOR for MM drugs. Disclosures Dai: M2Gen: Employment. Dalton:MILLENNIUM PHARMACEUTICALS, INC.: Honoraria. Shain:Bristol-Myers Squibb: Membership on an entity's Board of Directors or advisory committees; Amgen: Membership on an entity's Board of Directors or advisory committees; Takeda: Membership on an entity's Board of Directors or advisory committees; Sanofi Genzyme: Membership on an entity's Board of Directors or advisory committees; AbbVie: Research Funding; Adaptive Biotechnologies: Consultancy; Janssen: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees.

Over the last decade we have witnessed an explosion in the number of therapeutic options available to patients with multiple myeloma (MM). In spite of the marked improvements in patient outcomes paralleling these approvals, MM remains an incurable malignancy for the vast majority of patients following a course of therapeutic successes and failures. As such, there remains a dire need to develop new tools to improve the management of MM patients. A number of groups are leading efforts to combine big data and artificial intelligence to better inform patient care via precision medicine. At Moffitt, in collaboration with the M2Gen/ORIEN (Oncology Research Information Exchange Network), we have begun to accumulate big data in MM. Patients opt in to (consent) for collection of rich clinical data (demographics, staging, risk, complete disease course treatment data) and in the setting of bone marrow biopsy the allocation of CD138-selected cells for molecular analysis (whole exome sequencing (WES) and RNA sequencing as well as peripheral blood mononuclear cells for WES). To date, we have collected over 1000 samples for over 800 individual patients with plasma cell disorders. In the setting of oncology, the ultimate goal of model will be selection of ideal treatments. We expect that AI analysis may validate of patient response to treatments and enable cohort selection, as real patient cohorts can be selected from those predicted by the model. One approach is to utilize reinforcement learning (RL). In RL, the algorithm attempts to learn actions to optimize a type action a defined state and weight any tradeoffs for maximal reward. Our initial utilization of RL involved a relatively small cohort of 402 patients with treatment medication data. This encompassed 1692 lines of treatment with a mean of 4.21 lines of therapy per patient (Median of 4 lines per patient). This included 132 combinations of 22 myeloma therapeutics. The heterogeneity in treatment is highlighted by the fact that no pathways overlap after line 4. Each Q-value in Q-table is the current reward for an action in a state plus the discounted anticipated future reward for taking that action. Iteration helps you converge on the actual values for the future reward (can be model-free). The end result is a policy, P(s), that tells you what the ideal action is at state. There are a near infinite number of possible states, considering treatment history, age, GEP, cytogenetics, comorbidities, staging and others. We presume that action makes intuitive sense as medication (treatment) only and that reward should be some form of treatment response. We have begun the iterative process of trying different state and reward functions. Median imputation shows 5% improvement in response accuracy over listwise, but median imputation throws off practical accuracy in a binary reward case. While we found that the exercise has great potential. We found that there are possible improvements (multiple imputation). We will need to expand covariate analysis. Combinatorics need to be considered in machine learning in medium-sized data sets. Model-free machine learning is limited on medium-sized data. As such, combined resources and/or utilization of large networks such as ORIEN will be critical for the successful integration of RL or other AI tools in MM. We also learned that adding variables to the model doesn't necessarily increase accuracy. Future work will involve continued application of alternate state/reward functions. Loosen iQ-learning framework to allow for better covariate selection for state/reward functions. Improve imputation techniques to include more covariates and have more certainty in model accuracy. We may also refine accuracy metric to allow for prediction of bucketed response and temporal disease burden (M-spike vs. time). Updated data on a larger cohort will be presented at the annual meeting. Disclosures Shain: Adaptive Biotechnologies: Consultancy; Celgene: Membership on an entity's Board of Directors or advisory committees; Bristol-Myers Squibb: Membership on an entity's Board of Directors or advisory committees; Amgen: Membership on an entity's Board of Directors or advisory committees; Takeda: Membership on an entity's Board of Directors or advisory committees; Sanofi Genzyme: Membership on an entity's Board of Directors or advisory committees; AbbVie: Research Funding; Janssen: Membership on an entity's Board of Directors or advisory committees. Dai:M2Gen: Employment. Nishihori:Novartis: Research Funding; Karyopharm: Research Funding. Brayer:Janssen: Consultancy, Speakers Bureau; BMS: Consultancy, Speakers Bureau. Alsina:Bristol-Myers Squibb: Research Funding; Janssen: Speakers Bureau; Amgen: Speakers Bureau. Baz:Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding; Karyopharm: Membership on an entity's Board of Directors or advisory committees, Research Funding; AbbVie: Research Funding; Merck: Research Funding; Sanofi: Research Funding; Bristol-Myers Squibb: Research Funding. Dalton:MILLENNIUM PHARMACEUTICALS, INC.: Honoraria.

e19513 Background: Although there is much to be optimistic about in the multiple myeloma community as the approval of new therapies and regimen-combinations for relapsed refractory disease continues to grow, determining the best option for a patient can be complicated. Both carfilzomib- (C) and daratumumab- (D) based regimens have demonstrated superior efficacy in this setting, but there is a paucity of data supporting which should be selected first, and if regimen sequence influences outcomes. The aim of this study is to describe sequencing patterns in the era of these newer agents and to determine if there is a difference in outcomes for patients with RRMM who received one of the following treatment sequences: C-regimen with a D-regimen given immediately prior (DC); C-regimen without any prior D (C only); D-regimen with a C-regimen given immediately prior (CD); or D-regimen without any prior C (D only). Methods: This is a retrospective analysis of patients with RRMM consecutively treated at Moffitt Cancer Center between 1/1/2015 and 6/25/18. Response to therapy was assessed using the International Myeloma Working Group (IMWG) criteria. Progression-free survival (PFS) was measured in days from the start of therapy to progression. Time to response (TTR) was measured in days from the start of therapy to first response. Results: 132 patients with RRMM who received 1-3 prior lines of therapy with at least one line of therapy containing either C or D were identified. Overall, the majority of patients were treated with C only (n = 101), 10 received DC, 31 received D only, and 35 received CD. In patients that received C only, partial response (PR) was achieved in 38%, very good partial response (VGPR) was 20%, and stringent complete response (sCR) was 2%. In patients that received DC, PR was 20% and VGPR was 10%; no patient achieved a sCR. Of the patients that received D only, PR was 29%, VGPR was 10%, and sCR was 3%. In patients that received CD, PR was 31% and VGPR was 26%; no patient achieved sCR. Median PFS in patients who received C only, DC, D only, and CD was 117 days, 126 days, 104 days, and 190 days, respectively. TTR in patients who received C only, DC, D only, and CD was 82 days, 39 days, 98 days, and 88 days, respectively. Conclusions: The data suggests that RRMM patients who receive either CD or DC appear to have a PFS advantage over those patients who did not. Notably, an early TTR was found in patients that received DC. Further analysis is ongoing.