Drug response prediction in high-risk multiple myeloma

Identifying clinical response to daratumumab therapy in relapsed/refractory multiple myeloma using a patient-derived in vitro model

Response to daratumumab in patients with relapsed/refractory multiple myeloma is heterogeneous, and a reliable biomarker of response is lacking. We aimed to develop a method that identifies response to daratumumab therapy. Patient-derived MM cells were collected before start of daratumumab treatment and were cultured in a hydrogel-based culture system. The extent of antibody-dependent cellular cytotoxicity and complement-dependent cytotoxicity in vitro was associated with both clinical response and progression-free survival in corresponding patients. Together, our results demonstrate that in vitro sensitivity to daratumumab therapy in a hydrogel culture with primary MM cells might be used to identify patients most likely to benefit from treatment.

A novel drug specific mRNA biomarker predictor for selection of patients responding to dovitinib treatment of advanced renal cell carcinoma and other solid tumors

PURPOSE

Dovitinib is a receptor tyrosine kinase inhibitor of VEGFR1-3, PDGFR, FGFR1/3, c-KIT, FLT3 and topoisomerase 1 and 2. The drug response predictor (DRP) biomarker algorithm or DRP-Dovitinib is being developed as a companion diagnostic to dovitinib and was applied retrospectively.

PATIENTS AND METHODS

Archival tumor samples were obtained from consenting patients in a phase 3 trial comparing dovitinib to sorafenib in renal cell carcinoma patients and the DRP-Dovitinib was applied. The biomarker algorithm combines the expression of 58 messenger RNAs relevant to the in vitro sensitivity or resistance to dovitinib, including genes associated with FGFR, PDGF, VEGF, PI3K/Akt/mTOR and topoisomerase pathways as well as ABC drug transport, and provides a likelihood score between 0-100%.

RESULTS

The DRP-Dovitinib divided the dovitinib treated RCC patients into two groups, sensitive (n = 49, DRP score >50%) or resistant (n = 86, DRP score ≤ 50%) to dovitinib. The DRP sensitive population was compared to the unselected sorafenib arm (n = 286). Median progression-free survival (PFS) was 3.8 months in the DRP sensitive dovitinib arm and 3.6 months in the sorafenib arm (hazard ratio 0.71, 95% CI 0.51-1.01). Median overall survival (OS) was 15.0 months in the DRP sensitive dovitinib arm and 11.2 months in the sorafenib arm (hazard ratio 0.69, 95% CI 0.48-0.99). The observed clinical benefit increased with increasing DRP score. At a cutoff of 67% the median OS was 20.6 months and the median PFS was 5.7 months in the dovitinib arm. The results were confirmed in five smaller phase II trials of dovitinib which showed a similar trend.

CONCLUSION

The DRP-Dovitinib shows promise as a potential biomarker for identifying advanced RCC patients most likely to experience clinical benefit from dovitinib treatment, subject to confirmation in an independent prospective trial of dovitinib in RCC patients.

Radiomics Models Based on Magnetic Resonance Imaging for Prediction of the Response to Bortezomib-Based Therapy in Patients with Multiple Myeloma

Purpose. To identify significant radiomics features based on MRI and establish effective models for predicting the response to bortezomib-based regimens. Materials and Methods. In total, 95 MM patients treated with bortezomib-based therapy were enrolled, including 77 with bortezomib, cyclophosphamide, and dexamethasone (BCD) and 18 with bortezomib, lenalidomide, and dexamethasone (VRD). Based on T1-weighted imaging (T1WI) and T2-weighted imaging with fat suppression (T2WI-fs), radiomics features were extracted and then selected. The random forest (RF), $k$ -nearest neighbor, support vector machine, logistic regression, decision tree, and Bayes models were built using the selected features. The predictive power of six models for response to BCD and VRD regimens were evaluated. The correlation between the selected features and progression-free survival (PFS) was also analyzed. Results. Four wavelet features were correlated with BCD treatment response. The six models all showed predictive power for BCD regimen (AUC: 0.84-0.896 in the training set, 0.801-0.885 in the validation set), and RF performed relatively better than others. Nevertheless, all the BCD-based models were incapable of predicting the VRD treatment response. The wavelet-HLH_firstorder_kurtosis was also associated with PFS (log-rank $P=0.019$ ). Conclusion. The four wavelet features were valuable biomarkers for predicting the response to BCD regimen. The six models based on these features showed predictive power, and RF was the best. One wavelet feature was also a survival-related biomarker. MRI-based radiomics had the potential to guide clinicians in MM management.

Survival prediction and treatment optimization of multiple myeloma patients using machine-learning models based on clinical and gene expression data

Multiple myeloma (MM) remains mostly an incurable disease with a heterogeneous clinical evolution. Despite the availability of several prognostic scores, substantial room for improvement still exists. Promising results have been obtained by integrating clinical and biochemical data with gene expression profiling (GEP). In this report, we applied machine learning algorithms to MM clinical and RNAseq data collected by the CoMMpass consortium. We created a 50-variable random forests model (IAC-50) that could predict overall survival with high concordance between both training and validation sets (c-indexes, 0.818 and 0.780). This model included the following covariates: patient age, ISS stage, serum B2-microglobulin, first-line treatment, and the expression of 46 genes. Survival predictions for each patient considering the first line of treatment evidenced that those individuals treated with the best-predicted drug combination were significantly less likely to die than patients treated with other schemes. This was particularly important among patients treated with a triplet combination including bortezomib, an immunomodulatory drug (ImiD), and dexamethasone. Finally, the model showed a trend to retain its predictive value in patients with high-risk cytogenetics. In conclusion, we report a predictive model for MM survival based on the integration of clinical, biochemical, and gene expression data with machine learning tools.

Multi-omics tumor profiling technologies to develop precision medicine in multiple myeloma

Multiple myeloma (MM), the second most common hematologic cancer, is caused by accumulation of aberrant plasma cells in the bone marrow. Its molecular causes are not fully understood and its great heterogeneity among patients complicates therapeutic decision-making. In the past decades, development of new therapies and drugs have significantly improved survival of MM patients. However, resistance to drugs and relapse remain the most common causes of mortality and are the major challenges to overcome. The advent of high throughput omics technologies capable of analyzing big amount of clinical and biological data has changed the way to diagnose and treat MM. Integration of omics data (gene mutations, gene expression, epigenetic information, and protein and metabolite levels) with clinical histories of thousands of patients allows to build scores to stratify the risk at diagnosis and predict the response to treatment, helping clinicians to make better educated decisions for each particular case. There is no doubt that the future of MM treatment relies on personalized therapies based on predictive models built from omics studies. This review summarizes the current treatments and the use of omics technologies in MM, and their importance in the implementation of personalized medicine.

Multi-omics tumor profiling technologies to develop precision medicine in multiple myeloma

Multiple myeloma (MM), the second most common hematologic cancer, is caused by accumulation of aberrant plasma cells in the bone marrow. Its molecular causes are not fully understood and its great heterogeneity among patients complicates therapeutic decision-making. In the past decades, development of new therapies and drugs have significantly improved survival of MM patients. However, resistance to drugs and relapse remain the most common causes of mortality and are the major challenges to overcome. The advent of high throughput omics technologies capable of analyzing big amount of clinical and biological data has changed the way to diagnose and treat MM. Integration of omics data (gene mutations, gene expression, epigenetic information, and protein and metabolite levels) with clinical histories of thousands of patients allows to build scores to stratify the risk at diagnosis and predict the response to treatment, helping clinicians to make better educated decisions for each particular case. There is no doubt that the future of MM treatment relies on personalized therapies based on predictive models built from omics studies. This review summarizes the current treatments and the use of omics technologies in MM, and their importance in the implementation of personalized medicine.

Ensembled machine learning framework for drug sensitivity prediction

Drug sensitivity prediction is one of the critical tasks involved in drug designing and discovery. Recently several online databases and consortiums have contributed to providing open access to pharmacogenomic data. These databases have helped in developing computational approaches for drug sensitivity prediction. Cancer is a complex disease involving the heterogeneous behaviour of same tumour-type patients towards the same kind of drug therapy. Several methods have been proposed in the literature to predict drug sensitivity. However, these methods are not efficient enough to predict drug sensitivity. The present study has proposed an ensemble learning framework for drug-response prediction using a modified rotation forest. The proposed framework is further compared with three state-of-the-art algorithms and two baseline methods using Genomics of Drug Sensitivity in Cancer (GDSC) and Cancer Cell Line Encyclopedia (CCLE) drug screens. The authors have also predicted missing drug response values in the data set using the proposed approach. The proposed approach outperforms other counterparts even though gene mutation data is not incorporated while designing the approach. An average mean square error of 3.14 and 0.404 is achieved using GDSC and CCLE drug screens, respectively. The obtained results show that the proposed framework has considerable potential to improve anti-cancer drug response prediction.

Overexpression of HIF-1α contributes to melphalan resistance in multiple myeloma cells by activation of ERK1/2, Akt, and NF-κB

Multiple myeloma (MM) commonly displays multidrug resistance and is associated with poor prognosis. Therefore, it is important to identify the mechanisms by which MM cells develop multidrug resistance. Our previous study showed that multidrug resistance is correlated with overexpression of multidrug resistance protein 1 (MDR1) and Survivin, and downregulation of Bim expression in melphalan-resistant RPMI8226/L-PAM cells; however, the underlying mechanism of multidrug resistance remains unclear. In the present study, we investigated the mechanism of multidrug resistance in melphalan-resistant cells. We found that RPMI8226/L-PAM and ARH-77/L-PAM cells showed increased phosphorylation of extracellular signal-regulated protein kinase 1/2 (ERK1/2) and Akt, and nuclear localization of nuclear factor κB (NF-κB). The combination of ERK1/2, Akt, and NF-κB inhibitors with melphalan reversed melphalan resistance via suppression of Survivin expression and enhanced Bim expression in melphalan-resistant cells. In addition, RPMI8226/L-PAM and ARH-77/L-PAM cells overexpressed hypoxia-inducible factor 1α (HIF-1α) via activation of ERK1/2, Akt, and NF-κB. Moreover, suppression of HIF-1α by echinomycin or HIF-1α siRNA resensitized RPMI8226/L-PAM cells to melphalan through downregulation of Survivin expression and upregulation of Bim expression. These results indicate that enhanced Survivin expression and decreased Bim expression by HIF-1α via activation of ERK1/2, Akt, and NF-κB play a critical role in melphalan resistance. Our findings suggest that HIF-1α, ERK1/2, Akt, and NF-κB inhibitors are potentially useful as anti-MDR agents for the treatment of melphalan-resistant MM.

Osteoblast suppression in multiple myeloma bone disease

Multiple myeloma (MM) is the most frequent cancer to involve the skeleton with patients developing osteolytic bone lesions due to hyperactivation of osteoclasts and suppression of BMSCs differentiation into functional osteoblasts. Although new therapies for MM have greatly improved survival, MM remains incurable for most patients. Despite the major advances in current anti-MM and anti-resorptive treatments that can significantly improve osteolytic bone lysis, many bone lesions can persist even after therapeutic remission of active disease. Bone marrow mesenchymal stem cells (BMSCs) from MM patients are phenotypically distinct from their healthy counterparts and the mechanisms associated with the long-term osteogenic suppression are largely unknown. In this review we will highlight recent results of transcriptomic profiling studies that provide new insights into the establishment and maintenance of the persistent pathological alterations in MM-BMSCs that occur in MM. We will we discuss the role of genomic instabilities and senescence in propagating the chronically suppressed state and pro-inflammatory phenotype associated with MM-BMSCs. Lastly we describe the role of epigenetic-based mechanisms in regulating osteogenic gene expression to establish and maintain the pro-longed suppression of MM-BMSC differentiation into functional OBs.

Predicting treatment benefit in multiple myeloma through simulation of alternative treatment effects

Many cancer treatments are associated with serious side effects, while they often only benefit a subset of the patients. Therefore, there is an urgent clinical need for tools that can aid in selecting the right treatment at diagnosis. Here we introduce simulated treatment learning (STL), which enables prediction of a patient’s treatment benefit. STL uses the idea that patients who received different treatments, but have similar genetic tumor profiles, can be used to model their response to the alternative treatment. We apply STL to two multiple myeloma gene expression datasets, containing different treatments (bortezomib and lenalidomide). We find that STL can predict treatment benefit for both; a twofold progression free survival (PFS) benefit is observed for bortezomib for 19.8% and a threefold PFS benefit for lenalidomide for 31.1% of the patients. This demonstrates that STL can derive clinically actionable gene expression signatures that enable a more personalized approach to treatment.

Selection of the right cancer treatment is still a challenge. Here, the authors introduce a framework to analyze treatment benefits, using the idea that patients with similar genetic tumor profiles receiving different treatments can be used to model their responses to the alternative treatment.

Phase 1 trial of ibrutinib and carfilzomib combination therapy for relapsed or relapsed and refractory multiple myeloma

This phase 1, dose-finding study investigated ibrutinib and carfilzomib ± dexamethasone in patients with relapsed or relapsed/refractory multiple myeloma (≥2 lines of therapy including bortezomib and an immunomodulatory agent). Of 43 patients enrolled, 74% were refractory to bortezomib and 23% had high-risk cytogenetics. No dose-limiting toxicities were observed. The recommended phase 2 dose was ibrutinib 840 mg and carfilzomib 36 mg/m² with dexamethasone. The most common ≥ grade 3 (>10%) treatment-emergent adverse events were hypertension, anemia, pneumonia, fatigue, diarrhea, and thrombocytopenia. Overall response rate was 67% (very good partial response, 21%; stringent complete response, 2%), with an additional 9% minimal response. Median progression-free survival was 7.2 months and was not inferior in refractory nor high-risk patients. Median overall survival was not reached. Ibrutinib plus carfilzomib demonstrated encouraging responses with a manageable safety profile in this advanced population.