Assessment of urine colour using a wallet card: a randomised study of a novel patient self-care tool during chemoradiation for oesophageal cancer.

BACKGROUND: Patients undergoing neoadjuvant chemoradiation for oesophageal cancer often experience dehydration from decreased fluid intake and increased losses. Despite frequent clinical visits during treatment, patients can still present with dehydration, suggesting the need for increased patient awareness and engagement around adverse event management at home. Evidence for benefits of self-monitoring may help motivate patients to engage proactively in their own care to improve their treatment experience. METHODS: We performed a randomised single-centre study of a urine colour self-monitoring card (UCC) during chemoradiation therapy for oesophageal cancer, compared with standard dietitian counselling. Primary outcome was self-efficacy as determined by the Self-Management Resource Centre Self-Efficacy for Managing Chronic Disease Scale (SMCD). Secondary outcomes included Burge thirst scores, Edmonton Symptom Assessment System scores (ESAS), patient-initiated hydrations, creatinine rise and satisfaction with the UCC. RESULTS: Thirty-five patients were randomised. UCC use was not associated with improved SMCD or ESAS scores compared with standard counselling. The card was highly rated by patients as a welcome tool for self-monitoring. CONCLUSIONS: No beneficial effect on self-efficacy or dehydration markers with UCC use was demonstrated. The study nonetheless drew attention to several factors potentially hindering its use for effective self-care: the unexpected severity of other symptoms consuming patients' attention, reduced sensitivity of urine colour due to chemotherapy, absence of active inquiry by the healthcare team and the inconvenient location of the UCC in wallet/purse. Urine colour monitoring in patients with oesophageal cancer to improve the patient experience during treatment warrants further study but supported by active healthcare provider inquiry, more accessible format of the card, and possibly home vital checks to increase its sensitivity in the clinical context.

The JAK1/2 inhibitor ruxolitinib downregulates the immune checkpoint protein B7H3 in multiple myeloma.

We hypothesized that ruxolitinib may inhibit the immune checkpoint protein, B7H3; and, thus, investigated its effects on this immune inhibitor using multiple myeloma (MM) cell lines, bone marrow (BM) mononuclear cells from MM patients and human MM LAGλ -1A xenografts. Ruxolitinib reduced B7H3 gene and protein expression and increased IL-2 and CD8 gene expression. These results suggest that ruxolitinib inhibition of B7H3 may restore exhausted T-cell activity in the MM BM tumor microenvironment.

Multiscale profiling of protease activity in cancer.

Diverse processes in cancer are mediated by enzymes, which most proximally exert their function through their activity. High-fidelity methods to profile enzyme activity are therefore critical to understanding and targeting the pathological roles of enzymes in cancer. Here, we present an integrated set of methods for measuring specific protease activities across scales, and deploy these methods to study treatment response in an autochthonous model of Alk-mutant lung cancer. We leverage multiplexed nanosensors and machine learning to analyze in vivo protease activity dynamics in lung cancer, identifying significant dysregulation that includes enhanced cleavage of a peptide, S1, which rapidly returns to healthy levels with targeted therapy. Through direct on-tissue localization of protease activity, we pinpoint S1 cleavage to the tumor vasculature. To link protease activity to cellular function, we design a high-throughput method to isolate and characterize proteolytically active cells, uncovering a pro-angiogenic phenotype in S1-cleaving cells. These methods provide a framework for functional, multiscale characterization of protease dysregulation in cancer.

Protease Activity Analysis: A Toolkit for Analyzing Enzyme Activity Data.

Analyzing the activity of proteases and their substrates is critical to defining the biological functions of these enzymes and to designing new diagnostics and therapeutics that target protease dysregulation in disease. While a wide range of databases and algorithms have been created to better predict protease cleavage sites, there is a dearth of computational tools to automate analysis of in vitro and in vivo protease assays. This necessitates individual researchers to develop their own analytical pipelines, resulting in a lack of standardization across the field. To facilitate protease research, here we present Protease Activity Analysis (PAA), a toolkit for the preprocessing, visualization, machine learning analysis, and querying of protease activity data sets. PAA leverages a Python-based object-oriented implementation that provides a modular framework for streamlined analysis across three major components. First, PAA provides a facile framework to query data sets of synthetic peptide substrates and their cleavage susceptibilities across a diverse set of proteases. To complement the database functionality, PAA also includes tools for the automated analysis and visualization of user-input enzyme-substrate activity measurements generated through in vitro screens against synthetic peptide substrates. Finally, PAA supports a set of modular machine learning functions to analyze in vivo protease activity signatures that are generated by activity-based sensors. Overall, PAA offers the protease community a breadth of computational tools to streamline research, taking a step toward standardizing data analysis across the field and in chemical biology and biochemistry at large.

Testing psychosocial mediators of a mind-body resiliency intervention for cancer survivors.

PURPOSE: Group-based mind-body interventions such as the Stress Management and Resiliency Training-Relaxation Response Resiliency Program (SMART-3RP) hold promise for enhancing resiliency among cancer survivors. Mechanisms underlying improvements in psychological outcomes are theoretically established but remain unexamined empirically. METHODS: Adult cancer survivors (n = 105) participating in the SMART-3RP completed surveys of resiliency and five hypothesized mediators: coping (ability to relax physical tension and assertive social support-seeking), mindfulness, positive affect, and worry. Pre-post intervention changes were assessed using repeated-measures t-tests. Bivariate correlations between change scores and a more conservative within-person parallel mediation model tested covariance between resiliency and mediators. RESULTS: Participants experienced moderate to large improvements in all patient-reported outcomes (ds = 1.01-0.46). Increased resiliency was significantly associated with increases in mindfulness, positive affect, and assertive social support-seeking (rs = 0.36-0.50); smaller associations with increased relaxation and decreased worry were not significant. Mindfulness and positive affect explained the largest proportion of variance in resiliency increase in the full multivariate model. CONCLUSIONS: Cancer survivors completing the SMART-3RP had increased resiliency, which was associated with improvements in mindfulness, positive affect, and the ability to assertively seek social support. Enhancing mindfulness and positive affect were critical components for enhancing resiliency. Implications for resiliency interventions with cancer survivors are discussed.

Aging and matrix viscoelasticity affect multiscale tendon properties and tendon derived cell behavior.

Aging is the largest risk factor for Achilles tendon associated disorders and rupture. Although Achilles tendon macroscale elastic properties are suggested to decline with aging, less is known about the effect of maturity and aging on multiscale viscoelastic properties and their effect on tendon cell behavior. Here, we show dose dependent changes in native multiscale tendon mechanical and structural properties and uncover several nanoindentation properties predicted by tensile mechanics and echogenicity. Alginate hydrogel systems designed to mimic juvenile tendon microscale mechanics revealed that stiffness and viscoelasticity affected Achilles tendon cell aspect ratio and proliferation during aging. This knowledge provides further evidence for the negative impact of maturity and aging on tendon and begins to elucidate how viscoelasticity can control tendon derived cell morphology and expansion. STATEMENT OF SIGNIFICANCE: Aging is the largest risk factor for Achilles tendon associated disorders and rupture. Although Achilles tendon macroscale elastic properties are suggested to decline with aging, less is known about the effect of maturity and aging on multiscale viscoelastic properties and their effect on tendon cell behavior. Here, we show dose dependent changes in native multiscale tendon mechanical and structural properties and uncover several nanoindentation properties predicted by tensile mechanics and echogenicity. Alginate hydrogel systems designed to mimic juvenile tendon microscale mechanics revealed that stiffness and viscoelasticity affected Achilles tendon cell spreading and proliferation during aging. This knowledge provides further evidence for the negative impact of maturity and aging on tendon and begins to elucidate how viscoelasticity can control tendon derived cell morphology and expansion.

Penetrance of male breast cancer susceptibility genes: a systematic review.

PURPOSE: Several male breast cancer (MBC) susceptibility genes have been identified, but the MBC risk for individuals with a pathogenic variant in each of these genes (i.e., penetrance) remains unclear. We conducted a systematic review of studies reporting the penetrance of MBC susceptibility genes to better summarize current estimates of penetrance. METHODS: A search query was developed to identify MBC-related papers indexed in PubMed/MEDLINE. A validated natural language processing method was applied to identify papers reporting penetrance estimates. These penetrance studies' bibliographies were reviewed to ensure comprehensiveness. We accessed the potential ascertainment bias for each enrolled study. RESULTS: Fifteen penetrance studies were identified from 12,182 abstracts, covering five purported MBC susceptibility genes: ATM, BRCA1, BRCA2, CHEK2, and PALB2. Cohort (n = 6, 40%) and case-control (n = 5, 33%) studies were the two most common study designs, followed by family-based (n = 3, 20%), and a kin-cohort study (n = 1, 7%). Seven of the 15 studies (47%) adjusted for ascertainment adequately and therefore the MBC risks reported by these seven studies can be considered applicable to the general population. Based on these seven studies, we found pathogenic variants in ATM, BRCA2, CHEK2 c.1100delC, and PALB2 show an increased risk for MBC. The association between BRCA1 and MBC was not statistically significant. CONCLUSION: This work supports the conclusion that pathogenic variants in ATM, BRCA2, CHEK2 c.1100delC, and PALB2 increase the risk of MBC, whereas pathogenic variants in BRCA1 may not be associated with increased MBC risk.

Baseline and Changes in Serum B-Cell Maturation Antigen Levels Rapidly Indicate Changes in Clinical Status Among Patients with Relapsed/Refractory Multiple Myeloma Starting New Therapy.

BACKGROUND: B-cell maturation antigen (BCMA) is expressed on malignant plasma cells from patients with multiple myeloma (MM). These patients have higher levels of serum (s)BCMA than healthy subjects, and levels correlate with disease status. The half-life of sBCMA is only 24-36 h, and levels are independent of renal function. OBJECTIVE: We determined whether baseline sBCMA values, a ≥ 25% increase, and a ≥ 50% decrease during treatment predicted progression-free survival (PFS) and overall survival (OS) among 81 patients with relapsed/refractory MM (RRMM) starting new treatments. METHODS: Serum was obtained on day 22 of each patient's 28-day cycle of new therapy. Kaplan-Meier survival analysis and log-rank comparison tests were used to determine the effect of baseline sBCMA. The effect of percentage change in sBCMA was investigated using time-dependent Cox proportional hazard models. RESULTS: Patients with baseline sBCMA levels above the median had a shorter PFS (p = 0.0077), and those in the highest quartile had a shorter PFS (p = 0.0012) and OS (p = 0.0022). A ≥ 25% increase at week 4, week 8, and anytime through week 12 predicted a shorter PFS (p = 0.0011, p = 0.0005, and p < 0.0001, respectively). A ≥ 50% decrease at week 4, week 8, and anytime through week 12 predicted a longer PFS (p = 0.0045, p = 0.029, p = 0.0055, respectively). A ≥ 25% increase in sBCMA occurred before progression according to International Myeloma Working Group criteria in 67.5% of patients. CONCLUSIONS: Our results indicate the potential for the use of sBCMA as a new biomarker for monitoring patients with RRMM.

A ganglionic blocker and adrenoceptor ligands modify clozapine-induced insulin resistance.

Clozapine is a second generation antipsychotic drug that has proven to be helpful in the management of patients with psychotic disorders that are resistant to other medications. Unfortunately, the majority of patients treated with clozapine develop metabolic dysregulation, including weight gain and insulin resistance. There are few treatments available to effectively counter these side-effects. The goal of the present study was to use an established animal model to better understand the nature of these metabolic side-effects and determine whether existing drugs could be used to alleviate metabolic changes. Adult female rats were treated with a range of doses of clozapine (2, 10 and 20 mg/kg) and subjected to the hyperinsulinemic-euglycemic clamp, to measure whole-body insulin resistance. Clozapine dose-dependently decreased the glucose infusion rate, reflecting pronounced insulin resistance. To reverse the insulin resistance, rats were co-treated with the ganglionic blocker mecamylamine (0.1, 1.0 and 5.0 mg/kg) which dose-dependently reversed the effects of 10 mg/kg clozapine. A 1.0 mg/kg dose of mecamylamine independently reversed the large increase in peripheral epinephrine caused by treatment with clozapine. To study the influence of specific adrenoceptors, rats were treated with multiple doses of α1 (prazosin), α2 (idazoxan), ß1 (atenolol) and ß2 (butoxamine) adrenoceptor antagonists after the onset of clozapine-induced insulin resistance. Both beta blockers were effective in attenuating the effects of clozapine, while idazoxan had a smaller effect; no change was seen with prazosin. The current results indicate that peripheral catecholamines may play a role in clozapine's metabolic effects and be a target for future treatments.

Baseline serum B-cell maturation antigen levels predict time to disease progression for patients with smoldering multiple myeloma.

Multiple myeloma (MM) patients with smoldering (S) disease are defined by a lack of CRAB/SLiM criteria but may transform into disease requiring treatment. The International Myeloma Working Group risk stratification model for SMM uses serum M-protein, serum-free light chain ratio, and bone marrow plasma cell percentage. We investigated whether baseline serum B-cell maturation antigen (sBCMA) levels are predictive of disease progression among 65 patients with SMM. A receiver operating characteristic curve was used to establish a definition for high-risk baseline sBCMA. Mantel Byar analysis was used to examine whether high-risk sBCMA was correlated with shorter time to transformation, and a time-dependent cox proportional hazard was used to determine whether it is independent of other risk factors. A z test for proportions was used to compare the percentage of patients that progressed among high-risk versus low-risk sBCMA patients. A baseline sBCMA level ≥137.5 mg/ml was found to be the optimal cutoff between high- and low-risk SMM patients. Patients with high-risk sBCMA levels had a shorter time to transformation (P = .000332). sBCMA was also higher at the time of transformation than baseline levels (P = .0116). sBCMA was the only variable found to be significantly predictive of time to transformation and additionally was found to be independent of other risk factors. In this study, we have shown for the first time that sBCMA levels predict transformation of SMM to active disease and that these levels increase at the time of transformation. These results are consistent with other studies showing that active MM patients undergoing therapy with higher baseline sBCMA levels are more likely to progress early and its levels increase at the time of disease progression.

Ruxolitinib reverses checkpoint inhibition by reducing programmed cell death ligand-1 (PD-L1) expression and increases anti-tumour effects of T cells in multiple myeloma.

Multiple myeloma (MM) tumour cells evade host immunity through a variety of mechanisms, which may potentially include the programmed cell death ligand-1 (PD-L1):programmed cell death protein-1 (PD-1) axis. This interaction contributes to the immunosuppressive bone marrow (BM) microenvironment, ultimately leading to reduced effector cell function. PD-L1 is overexpressed in MMBM and is associated with the resistance to immune-based approaches for treating MM. Ruxolitinib (RUX), an inhibitor of the Janus kinase (JAK) family of protein tyrosine kinases, is approved for myeloproliferative diseases. We investigated the effects of RUX alone or in combination with anti-MM agents on the expression of PD-L1 and T-cell cytotoxicity in MM. We showed that the expression of the PD-L1 gene was markedly increased in BM mononuclear cells from patients with MM with progressive disease versus those in complete remission. Furthermore, RUX treatment resulted in a concentration-dependent reduction of PD-L1 gene expression in the MM tumour cells cultured alone or co-cultured with stromal cells compared with untreated cells. The results also demonstrated that RUX increased MM cell apoptosis in the presence of interleukin-2-stimulated T cells to a similar degree as the treatment with anti-PD-1 or anti-PD-L1 antibodies. In summary, these results indicate that RUX can block PD-L1 expression resulting in augmentation of anti-MM effects of T cells.

Normalization of serum B-cell maturation antigen levels predicts overall survival among multiple myeloma patients starting treatment.

Serum B-cell maturation antigen (sBCMA) is a novel biomarker for B-cell malignancies. A normal reference range (<82·59 ng/ml) has been recently established but the impact of achieving normal levels to outcomes for patients receiving treatment for B-cell malignancies has not been studied. We first found that among multiple myeloma (MM) patients starting a new treatment, those who begin treatment within normal sBCMA limits (<82·59 ng/ml) have improved progression-free survival (PFS; P = 0·0398) and overall survival (OS; P = 0·0217) than those who do not. Furthermore, among patients who begin treatment with elevated (≥82·59 ng/ml) sBCMA levels, we assessed the relationship of a decrease in sBCMA to the normal range to OS and found that those who normalize sBCMA demonstrated improved OS (P = 0·0078). Normalizing patients also experienced a markedly improved overall response rate (P < 0·0001). Moreover, all patients who achieved complete remission (CR) showed normalization of sBCMA, and time to normalization (median 0·9 months) was faster than time to CR (5·0 months; P = 0·0036) for these patients. These results suggest that normalization of sBCMA may be an accurate predictor of OS for MM patients during treatment and predict for a higher likelihood of response.

Establishing and Evaluating an ASCO Learning Cohort: a Longitudinal Project Assessing the Learning Needs and Behaviors of Oncology Professionals.

BACKGROUND: In 2013, the American Society of Clinical Oncology (ASCO)'s Continuing Education Committee recommended establishing an interprofessional, longitudinal cohort pilot project. The main goals of the cohort were to gain feedback from oncology providers on how they use resources to address their learning needs and gain insights into the utility of different ASCO educational activities. METHODS: The ASCO Learning Cohort Pilot Project included 49 ASCO members that were representative of the overall Society membership demographics and ran from November 2015 through August 2016. Participants documented monthly learning needs and completed monthly feedback activities focused on specific ASCO educational resources. RESULTS: The Learning Cohort Pilot Project proved a viable and innovative cohort model for analyzing the learning process for oncology healthcare professionals. The development, operations, and compliance required unique infrastructure to accomplish this project. Relationships between participant demographic variables and learning preferences are reported elsewhere. CONCLUSION: The ASCO Learning Cohort Project is a unique educational project that demonstrated feasibility and has met its goals. This paper outlines the processes of establishing a learning cohort, including participant selection, project design, and participant feedback. We evaluate the project model as a means to better understand the learning needs and behaviors of oncology healthcare professionals.

Quantitative Assessment of Learning Behaviors for Oncology Providers.

How health care providers select topics and activities for learning is key to meeting their needs. The goal of this study was to investigate how oncology providers identify knowledge gaps and choose learning activities. An online focus group within a larger longitudinal study was conducted between November 2015 and August 2016. Participants were chosen by convenience and stratified random sampling of diverse types of oncology providers. Providers were asked monthly to identify learning needs, explain how they identified those needs, and describe the learning activity they chose to meet those needs. Thirty-two oncology providers recorded 201 learning needs via online journal entries (mean 6 entries per person). Needs were associated with practice setting and professional role (p < .05). Colleague recommendation predicted learning needs for advanced practice providers (APPs) (p = .003). Patient cases drove > 50% of identified learning needs across groups. Learning activity preferences were associated with practice setting (p < .05). Choice of learning activity was associated with practice setting, professional role, and geographic location. Colleague recommendation was important for APPs (p = .025). Over 75% of learner responses identify convenience and content quality as important factors in choosing an activity. This study represents a quantitative assessment of learning behaviors for oncology providers and shows that identification of learning needs and activity selection differ by provider demographics. Limitations include small size and underrepresentation of some groups. Our findings should be confirmed with larger samples. Future research should focus on assessment of cohort versus individual needs and learning priorities.

Estimating a normal reference range for serum B-cell maturation antigen levels for multiple myeloma patients.

The serum B-cell maturation antigen (sBCMA) has been identified as a novel serum biomarker for patients with multiple myeloma. However, no study has yet established a reference range for sBCMA levels. Its levels were determined in 196 healthy subjects and showed a right-tailed distribution with a median value of 37·51 ng/ml with a standard deviation of 22·54 ng/ml (range 18·78-180·39 ng/ml). Partitioning of subgroup reference ranges was considered but determined to be irrelevant. A non-parametric method using the median ± 2 standard deviations suggests using a universal reference interval of <82·59 ng/ml.

Penetrance of Colorectal Cancer Among Mismatch Repair Gene Mutation Carriers: A Meta-Analysis.

BACKGROUND: Lynch syndrome, the most common colorectal cancer (CRC) syndrome, is caused by germline mismatch repair (MMR) genes. Precise estimates of age-specific risks are crucial for sound counseling of individuals managing a genetic predisposition to cancer, but published risk estimates vary. The objective of this work is to provide gene-, sex-, and age-specific risk estimates of CRC for MMR mutation carriers that comprehensively reflect the best available data. METHODS: We conducted a meta-analysis to combine risk information from multiple studies on Lynch syndrome-associated CRC. We used a likelihood-based approach to integrate reported measures of CRC risk and deconvolved aggregated information to estimate gene- and sex-specific risk. RESULTS: Our comprehensive search identified 10 studies (8 on MLH1, 9 on MSH2, and 3 on MSH6). We estimated the cumulative risk of CRC by age and sex in heterozygous mutation carriers. At age 70 years, for male and female carriers, respectively, risks for MLH1 were 43.9% (95% confidence interval [CI] = 39.6% to 46.6%) and 37.3% (95% CI = 32.2% to 40.2%), for MSH2 were 53.9% (95% CI = 49.0% to 56.3%) and 38.6% (95% CI = 34.1% to 42.0%), and for MSH6 were 12.0% (95% CI = 2.4% to 24.6%) and 12.3% (95% CI = 3.5% to 23.2%). CONCLUSIONS: Our results provide up-to-date and comprehensive age-specific CRC risk estimates for counseling and risk prediction tools. These will have a direct clinical impact by improving prevention and management strategies for both individuals who are MMR mutation carriers and those considering testing.

A flexible nonlinear mixed effects model for HIV viral load rebound after treatment interruption.

Characterization of HIV viral rebound after the discontinuation of antiretroviral therapy is central to HIV cure research. We propose a parametric nonlinear mixed effects model for the viral rebound trajectory, which often has a rapid rise to a peak value followed by a decrease to a viral load set point. We choose a flexible functional form that captures the shapes of viral rebound trajectories and can also provide biological insights regarding the rebound process. Each parameter can incorporate a random effect to allow for variation in parameters across individuals. Key features of viral rebound trajectories such as viral set points are represented by the parameters in the model, which facilitates assessment of intervention effects and identification of important pretreatment interruption predictors for these features. We employ a stochastic expectation-maximization (StEM) algorithm to incorporate HIV-1 RNA values that are below the lower limit of assay quantification. We evaluate the performance of our model in simulation studies and apply the proposed model to longitudinal HIV-1 viral load data from five AIDS Clinical Trials Group treatment interruption studies.

Quantitative Analysis of Oncology Professional Learning Preferences.

PURPOSE: ASCO is the premier and largest global professional society for oncology care professionals. In 2015, ASCO launched a longitudinal Learning Cohort Pilot Project to catalog and better understand the learning behaviors and preferences of oncology health care providers. A secondary goal was to assess learner preferences and utilization related to ASCO's portfolio of educational resources. METHODS: The Learning Cohort Pilot Project was conducted between November 2015 and August 2016 with 49 ASCO members. Participants were selected via convenience sampling and stratified random sampling to generate a cohort that mirrored the demographic distribution of overall ASCO membership. Participants completed a different ASCO resource-specific feedback activity each month, which measured professional educational needs, sources sought, and preferences for educational resources. Responses were organized by demographic variables in our participant pool to identify trends in provider learning preferences. Fisher's exact test was used to assess the association between participant demographics and practice setting and responses. Holm's procedure was used to adjust for multiple testing. RESULTS: The Learning Cohort Pilot Project revealed statistically significant relationships between participant demographic variables and learning preferences. Age and practice setting were the demographic variables most consistently associated with the different preferences explored throughout the targeted activities. CONCLUSION: The results of this pilot cohort reinforced the hypothesis that oncology care providers have different professional educational needs and preferences that can be potentially anticipated and met with tailored resources. Delivering solutions to meet these needs represents an opportunity for further research and resource development.

JAK1/2 pathway inhibition suppresses M2 polarization and overcomes resistance of myeloma to lenalidomide by reducing TRIB1, MUC1, CD44, CXCL12, and CXCR4 expression.

Monocytes polarize into pro-inflammatory macrophage-1 (M1) or alternative macrophage-2 (M2) states with distinct phenotypes and physiological functions. M2 cells promote tumour growth and metastasis whereas M1 macrophages show anti-tumour effects. We found that M2 cells were increased whereas M1 cells were decreased in bone marrow (BM) from multiple myeloma (MM) patients with progressive disease (PD) compared to those in complete remission (CR). Gene expression of Tribbles homolog 1 (TRIB1) protein kinase, an inducer of M2 polarization, was increased in BM from MM patients with PD compared to those in CR. Ruxolitinib (RUX) is an inhibitor of the Janus kinase family of protein tyrosine kinases (JAKs) and is effective for treating patients with myeloproliferative disorders. RUX markedly reduces both M2 polarization and TRIB1 gene expression in MM both in vitro and in vivo in human MM xenografts in severe combined immunodeficient mice. RUX also downregulates the expression of CXCL12, CXCR4, MUC1, and CD44 in MM cells and monocytes co-cultured with MM tumour cells; overexpression of these genes is associated with resistance of MM cells to the immunomodulatory agent lenalidomide. These results provide the rationale for evaluation of JAK inhibitors, including MM BM in combination with lenalidomide, for the treatment of MM patients.

Using Machine Learning and Natural Language Processing to Review and Classify the Medical Literature on Cancer Susceptibility Genes.

PURPOSE: The medical literature relevant to germline genetics is growing exponentially. Clinicians need tools that help to monitor and prioritize the literature to understand the clinical implications of pathogenic genetic variants. We developed and evaluated two machine learning models to classify abstracts as relevant to the penetrance-risk of cancer for germline mutation carriers-or prevalence of germline genetic mutations. MATERIALS AND METHODS: We conducted literature searches in PubMed and retrieved paper titles and abstracts to create an annotated data set for training and evaluating the two machine learning classification models. Our first model is a support vector machine (SVM) which learns a linear decision rule on the basis of the bag-of-ngrams representation of each title and abstract. Our second model is a convolutional neural network (CNN) which learns a complex nonlinear decision rule on the basis of the raw title and abstract. We evaluated the performance of the two models on the classification of papers as relevant to penetrance or prevalence. RESULTS: For penetrance classification, we annotated 3,740 paper titles and abstracts and evaluated the two models using 10-fold cross-validation. The SVM model achieved 88.93% accuracy-percentage of papers that were correctly classified-whereas the CNN model achieved 88.53% accuracy. For prevalence classification, we annotated 3,753 paper titles and abstracts. The SVM model achieved 88.92% accuracy and the CNN model achieved 88.52% accuracy. CONCLUSION: Our models achieve high accuracy in classifying abstracts as relevant to penetrance or prevalence. By facilitating literature review, this tool could help clinicians and researchers keep abreast of the burgeoning knowledge of gene-cancer associations and keep the knowledge bases for clinical decision support tools up to date.