Peripheral blood cellular dynamics of rheumatoid arthritis treatment informs about efficacy of response to disease modifying drugs.

Rheumatoid arthritis (RA) is an autoimmune disease characterized by systemic inflammation and is mediated by multiple immune cell types. In this work, we aimed to determine the relevance of changes in cell proportions in peripheral blood mononuclear cells (PBMCs) during the development of disease and following treatment. Samples from healthy blood donors, newly diagnosed RA patients, and established RA patients that had an inadequate response to MTX and were about to start tumor necrosis factor inhibitors (TNFi) treatment were collected before and after 3 months of treatment. We used in parallel a computational deconvolution approach based on RNA expression and flow cytometry to determine the relative cell-type frequencies. Cell-type frequencies from deconvolution of gene expression indicate that monocytes (both classical and non-classical) and CD4+ cells (Th1 and Th2) were increased in RA patients compared to controls, while NK cells and B cells (naïve and mature) were significantly decreased in RA patients. Treatment with MTX caused a decrease in B cells (memory and plasma cell), and a decrease in CD4 Th cells (Th1 and Th17), while treatment with TNFi resulted in a significant increase in the population of B cells. Characterization of the RNA expression patterns found that most of the differentially expressed genes in RA subjects after treatment can be explained by changes in cell frequencies (98% and 74% respectively for MTX and TNFi).

Molecular signature of methotrexate response among rheumatoid arthritis patients.

Background: Methotrexate (MTX) is the first line treatment for rheumatoid arthritis (RA), but failure of satisfying treatment response occurs in a significant proportion of patients. Here we present a longitudinal multi-omics study aimed at detecting molecular and cellular processes in peripheral blood associated with a successful methotrexate treatment of rheumatoid arthritis. Methods: Eighty newly diagnosed patients with RA underwent clinical assessment and donated blood before initiation of MTX, and 3 months into treatment. Flow cytometry was used to describe cell types and presence of activation markers in peripheral blood, the expression of 51 proteins was measured in serum or plasma, and RNA sequencing was performed in peripheral blood mononuclear cells (PBMC). Response to treatment after 3 months was determined using the EULAR response criteria. We assessed the changes in biological phenotypes during treatment, and whether these changes differed between responders and non-responders with regression analysis. By using measurements from baseline, we also tried to find biomarkers of future MTX response or, alternatively, to predict MTX response. Results: Among the MTX responders, (Good or Moderate according to EULAR treatment response classification, n = 60, 75%), we observed changes in 29 partly overlapping cell types proportions, levels of 13 proteins and expression of 38 genes during treatment. These changes were in most cases suppressions that were stronger among responders compared to non-responders. Within responders to treatment, we observed a suppression of FOXP3 gene expression, reduction of immunoglobulin gene expression and suppression of genes involved in cell proliferation. The proportion of many HLA-DR expressing T-cell populations were suppressed in all patients irrespective of clinical response, and the proportion of many IL21R+ T-cells were reduced exclusively in non-responders. Using only the baseline measurements we could not detect any biomarkers or prediction models that could predict response to MTX. Conclusion: We conclude that a deep molecular and cellular phenotyping of peripheral blood cells in RA patients treated with methotrexate can reveal previously not recognized differences between responders and non-responders during 3 months of treatment with MTX. This may contribute to the understanding of MTX mode of action and explain non-responsiveness to MTX therapy.

ImmUniverse Consortium: Multi-omics integrative approach in personalized medicine for immune-mediated inflammatory diseases.

Immune mediated inflammatory diseases (IMIDs) are a heterogeneous group of debilitating, multifactorial and unrelated conditions featured by a dysregulated immune response leading to destructive chronic inflammation. The immune dysregulation can affect various organ systems: gut (e.g., inflammatory bowel disease), joints (e.g., rheumatoid arthritis), skin (e.g., psoriasis, atopic dermatitis), resulting in significant morbidity, reduced quality of life, increased risk for comorbidities, and premature death. As there are no reliable disease progression and therapy response biomarkers currently available, it is very hard to predict how the disease will develop and which treatments will be effective in a given patient. In addition, a considerable proportion of patients do not respond sufficiently to the treatment. ImmUniverse is a large collaborative consortium of 27 partners funded by the Innovative Medicine Initiative (IMI), which is sponsored by the European Union (Horizon 2020) and in-kind contributions of participating pharmaceutical companies within the European Federation of Pharmaceutical Industries and Associations (EFPIA). ImmUniverse aims to advance our understanding of the molecular mechanisms underlying two immune-mediated diseases, ulcerative colitis (UC) and atopic dermatitis (AD), by pursuing an integrative multi-omics approach. As a consequence of the heterogeneity among IMIDs patients, a comprehensive, evidence-based identification of novel biomarkers is necessary to enable appropriate patient stratification that would account for the inter-individual differences in disease severity, drug efficacy, side effects or prognosis. This would guide clinicians in the management of patients and represent a major step towards personalized medicine. ImmUniverse will combine the existing and novel advanced technologies, including multi-omics, to characterize both the tissue microenvironment and blood. This comprehensive, systems biology-oriented approach will allow for identification and validation of tissue and circulating biomarker signatures as well as mechanistic principles, which will provide information about disease severity and future disease progression. This truly makes the ImmUniverse Consortium an unparalleled approach.

Antitumor Necrosis Factor-like Ligand 1A Therapy Targets Tissue Inflammation and Fibrosis Pathways and Reduces Gut Pathobionts in Ulcerative Colitis.

BACKGROUND: The first-in-class treatment PF-06480605 targets the tumor necrosis factor-like ligand 1A (TL1A) molecule in humans. Results from the phase 2a TUSCANY trial highlighted the safety and efficacy of PF-06480605 in ulcerative colitis. Preclinical and in vitro models have identified a role for TL1A in both innate and adaptive immune responses, but the mechanisms underlying the efficacy of anti-TL1A treatment in inflammatory bowel disease (IBD) are not known. METHODS: Here, we provide analysis of tissue transcriptomic, peripheral blood proteomic, and fecal metagenomic data from the recently completed phase 2a TUSCANY trial and demonstrate endoscopic improvement post-treatment with PF-06480605 in participants with ulcerative colitis. RESULTS: Our results revealed robust TL1A target engagement in colonic tissue and a distinct colonic transcriptional response reflecting a reduction in inflammatory T helper 17 cell, macrophage, and fibrosis pathways in patients with endoscopic improvement. Proteomic analysis of peripheral blood revealed a corresponding decrease in inflammatory T-cell cytokines. Finally, microbiome analysis showed significant changes in IBD-associated pathobionts, Streptococcus salivarius, S. parasanguinis, and Haemophilus parainfluenzae post-therapy. CONCLUSIONS: The ability of PF-06480605 to engage and inhibit colonic TL1A, targeting inflammatory T cell and fibrosis pathways, provides the first-in-human mechanistic data to guide anti-TL1A therapy for the treatment of IBD.

Early prediction of clinical response to anti-TNF treatment using multi-omics and machine learning in rheumatoid arthritis.

OBJECTIVES: Advances in immunotherapy by blocking TNF have remarkably improved treatment outcomes for Rheumatoid arthritis (RA) patients. Although treatment specifically targets TNF, the downstream mechanisms of immune suppression are not completely understood. The aim of this study was to detect biomarkers and expression signatures of treatment response to TNF inhibition. METHODS: Peripheral blood mononuclear cells (PBMCs) from 39 female patients were collected before anti-TNF treatment initiation (day 0) and after 3 months. The study cohort included patients previously treated with MTX who failed to respond adequately. Response to treatment was defined based on the EULAR criteria and classified 23 patients as responders and 16 as non-responders. We investigated differences in gene expression in PBMCs, the proportion of cell types and cell phenotypes in peripheral blood using flow cytometry and the level of proteins in plasma. Finally, we used machine learning models to predict non-response to anti-TNF treatment. RESULTS: The gene expression analysis in baseline samples revealed notably higher expression of the gene EPPK1 in future responders. We detected the suppression of genes and proteins following treatment, including suppressed expression of the T cell inhibitor gene CHI3L1 and its protein YKL-40. The gene expression results were replicated in an independent cohort. Finally, machine learning models mainly based on transcriptomic data showed high predictive utility in classifying non-response to anti-TNF treatment in RA. CONCLUSIONS: Our integrative multi-omics analyses identified new biomarkers for the prediction of response, found pathways influenced by treatment and suggested new predictive models of anti-TNF treatment in RA patients.

Genome-wide analysis of blood lipid metabolites in over 5000 South Asians reveals biological insights at cardiometabolic disease loci.

BACKGROUND: Genetic, lifestyle, and environmental factors can lead to perturbations in circulating lipid levels and increase the risk of cardiovascular and metabolic diseases. However, how changes in individual lipid species contribute to disease risk is often unclear. Moreover, little is known about the role of lipids on cardiovascular disease in Pakistan, a population historically underrepresented in cardiovascular studies. METHODS: We characterised the genetic architecture of the human blood lipidome in 5662 hospital controls from the Pakistan Risk of Myocardial Infarction Study (PROMIS) and 13,814 healthy British blood donors from the INTERVAL study. We applied a candidate causal gene prioritisation tool to link the genetic variants associated with each lipid to the most likely causal genes, and Gaussian Graphical Modelling network analysis to identify and illustrate relationships between lipids and genetic loci. RESULTS: We identified 253 genetic associations with 181 lipids measured using direct infusion high-resolution mass spectrometry in PROMIS, and 502 genetic associations with 244 lipids in INTERVAL. Our analyses revealed new biological insights at genetic loci associated with cardiometabolic diseases, including novel lipid associations at the LPL, MBOAT7, LIPC, APOE-C1-C2-C4, SGPP1, and SPTLC3 loci. CONCLUSIONS: Our findings, generated using a distinctive lipidomics platform in an understudied South Asian population, strengthen and expand the knowledge base of the genetic determinants of lipids and their association with cardiometabolic disease-related loci.

What Is the Persistence to Methotrexate in Rheumatoid Arthritis, and Does Machine Learning Outperform Hypothesis-Based Approaches to Its Prediction?

OBJECTIVE: The objectives of this study were to assess the 1-year persistence to methotrexate (MTX) initiated as the first ever conventional synthetic disease-modifying antirheumatic drug in new-onset rheumatoid arthritis (RA) and to investigate the marginal gains and robustness of the results by increasing the number and nature of covariates and by using data-driven, instead of hypothesis-based, methods to predict this persistence. METHODS: Through the Swedish Rheumatology Quality Register, linked to other data sources, we identified a cohort of 5475 patients with new-onset RA in 2006-2016 who were starting MTX monotherapy as their first disease-modifying antirheumatic drug. Data on phenotype at diagnosis and demographics were combined with increasingly detailed data on medical disease history and medication use in four increasingly complex data sets (48-4162 covariates). We performed manual model building using logistic regression. We also performed five different machine learning (ML) methods and combined the ML results into an ensemble model. We calculated the area under the receiver operating characteristic curve (AUROC) and made calibration plots. We trained on 90% of the data, and tested the models on a holdout data set. RESULTS: Of the 5475 patients, 3834 (70%) remained on MTX monotherapy 1 year after treatment start. Clinical RA disease activity and baseline characteristics were most strongly associated with the outcome. The best manual model had an AUROC of 0.66 (95% confidence interval [CI] 0.60-0.71). For the ML methods, Lasso regression performed best (AUROC = 0.67; 95% CI 0.62-0.71). CONCLUSION: Approximately two thirds of patients with early RA who start MTX remain on this therapy 1 year later. Predicting this persistence remains a challenge, whether using hypothesis-based or ML models, and may yet require additional types of data.

Prediction of response of methotrexate in patients with rheumatoid arthritis using serum lipidomics.

Methotrexate (MTX) is a common first-line treatment for new-onset rheumatoid arthritis (RA). However, MTX is ineffective for 30-40% of patients and there is no way to know which patients might benefit. Here, we built statistical models based on serum lipid levels measured at two time-points (pre-treatment and following 4 weeks on-drug) to investigate if MTX response (by 6 months) could be predicted. Patients about to commence MTX treatment for the first time were selected from the Rheumatoid Arthritis Medication Study (RAMS). Patients were categorised as good or non-responders following 6 months on-drug using EULAR response criteria. Serum lipids were measured using ultra-performance liquid chromatography-mass spectrometry and supervised machine learning methods (including regularized regression, support vector machine and random forest) were used to predict EULAR response. Models including lipid levels were compared to models including clinical covariates alone. The best performing classifier including lipid levels (assessed at 4 weeks) was constructed using regularized regression (ROC AUC 0.61 ± 0.02). However, the clinical covariate based model outperformed the classifier including lipid levels when either pre- or on-treatment time-points were investigated (ROC AUC 0.68 ± 0.02). Pre- or early-treatment serum lipid profiles are unlikely to inform classification of MTX response by 6 months with performance adequate for use in RA clinical management.

Standard machine learning approaches outperform deep representation learning on phenotype prediction from transcriptomics data.

BACKGROUND: The ability to confidently predict health outcomes from gene expression would catalyze a revolution in molecular diagnostics. Yet, the goal of developing actionable, robust, and reproducible predictive signatures of phenotypes such as clinical outcome has not been attained in almost any disease area. Here, we report a comprehensive analysis spanning prediction tasks from ulcerative colitis, atopic dermatitis, diabetes, to many cancer subtypes for a total of 24 binary and multiclass prediction problems and 26 survival analysis tasks. We systematically investigate the influence of gene subsets, normalization methods and prediction algorithms. Crucially, we also explore the novel use of deep representation learning methods on large transcriptomics compendia, such as GTEx and TCGA, to boost the performance of state-of-the-art methods. The resources and findings in this work should serve as both an up-to-date reference on attainable performance, and as a benchmarking resource for further research. RESULTS: Approaches that combine large numbers of genes outperformed single gene methods consistently and with a significant margin, but neither unsupervised nor semi-supervised representation learning techniques yielded consistent improvements in out-of-sample performance across datasets. Our findings suggest that using l2-regularized regression methods applied to centered log-ratio transformed transcript abundances provide the best predictive analyses overall. CONCLUSIONS: Transcriptomics-based phenotype prediction benefits from proper normalization techniques and state-of-the-art regularized regression approaches. In our view, breakthrough performance is likely contingent on factors which are independent of normalization and general modeling techniques; these factors might include reduction of systematic errors in sequencing data, incorporation of other data types such as single-cell sequencing and proteomics, and improved use of prior knowledge.

Identification of novel pheno-groups in heart failure with preserved ejection fraction using machine learning.

OBJECTIVE: Heart failure with preserved ejection fraction (HFpEF) is a heterogeneous syndrome. We aimed to derive HFpEF phenotype-based groups ('phenogroups') based on clinical and echocardiogram data using machine learning, and to compare clinical characteristics, proteomics and outcomes across the phenogroups. METHODS: We applied model-based clustering to 32 echocardiogram and 11 clinical and laboratory variables collected in stable condition from 320 HFpEF outpatients in the Karolinska-Rennes cohort study (56% female, median 78 years (IQR: 71-83)). Baseline proteomics and the composite end point of all-cause mortality or heart failure (HF) hospitalisation were used in secondary analyses. RESULTS: We identified six phenogroups, for which significant differences in the prevalence of concomitant atrial fibrillation (AF), anaemia and kidney disease were observed (p<0.05). Fifteen out of 86 plasma proteins differed between phenogroups (false discovery rate, FDR<0.05), including biomarkers of HF, AF and kidney function. The composite end point was significantly different between phenogroups (log-rank p<0.001), at short-term (100 days), mid-term (18 months) and longer-term follow-up (1000 days). Phenogroup 2 was older, with poorer diastolic and right ventricular function and higher burden of risk factors as AF (85%), hypertension (83%) and chronic obstructive pulmonary disease (30%). In this group a third experienced the primary outcome to 100 days, and two-thirds to 18 months (HR (95% CI) versus phenogroups 1, 3, 4, 5, 6: 1.5 (0.8-2.9); 5.7 (2.6-12.8); 2.9 (1.5-5.6); 2.7 (1.6-4.6); 2.1 (1.2-3.9)). CONCLUSIONS: Using machine learning we identified distinct HFpEF phenogroups with differential characteristics and outcomes, as well as differential levels of inflammatory and cardiovascular proteins.

An Unbiased Lipid Phenotyping Approach To Study the Genetic Determinants of Lipids and Their Association with Coronary Heart Disease Risk Factors.

Direct infusion high-resolution mass spectrometry (DIHRMS) is a novel, high-throughput approach to rapidly and accurately profile hundreds of lipids in human serum without prior chromatography, facilitating in-depth lipid phenotyping for large epidemiological studies to reveal the detailed associations of individual lipids with coronary heart disease (CHD) risk factors. Intact lipid profiling by DIHRMS was performed on 5662 serum samples from healthy participants in the Pakistan Risk of Myocardial Infarction Study (PROMIS). We developed a novel semi-targeted peak-picking algorithm to detect mass-to-charge ratios in positive and negative ionization modes. We analyzed lipid partial correlations, assessed the association of lipid principal components with established CHD risk factors and genetic variants, and examined differences between lipids for a common genetic polymorphism. The DIHRMS method provided information on 360 lipids (including fatty acyls, glycerolipids, glycerophospholipids, sphingolipids, and sterol lipids), with a median coefficient of variation of 11.6% (range: 5.4-51.9). The lipids were highly correlated and exhibited a range of associations with clinical chemistry biomarkers and lifestyle factors. This platform can provide many novel insights into the effects of physiology and lifestyle on lipid metabolism, genetic determinants of lipids, and the relationship between individual lipids and CHD risk factors.

Molecular Profiling of Ulcerative Colitis Subjects from the TURANDOT Trial Reveals Novel Pharmacodynamic/Efficacy Biomarkers.

BACKGROUND AND AIMS: To define pharmacodynamic and efficacy biomarkers in ulcerative colitis [UC] patients treated with PF-00547659, an anti-human mucosal addressin cell adhesion molecule-1 [MAdCAM-1] monoclonal antibody, in the TURANDOT study. METHODS: Transcriptome, proteome and immunohistochemistry data were generated in peripheral blood and intestinal biopsies from 357 subjects in the TURANDOT study. RESULTS: In peripheral blood, C-C motif chemokine receptor 9 [CCR9] gene expression demonstrated a dose-dependent increase relative to placebo, but in inflamed intestinal biopsies CCR9 gene expression decreased with increasing PF-00547659 dose. Statistical models incorporating the full RNA transcriptome in inflamed intestinal biopsies showed significant ability to assess response and remission status. Oncostatin M [OSM] gene expression in inflamed intestinal biopsies demonstrated significant associations with, and good accuracy for, efficacy, and this observation was confirmed in independent published studies in which UC patients were treated with infliximab or vedolizumab. Compared with the placebo group, intestinal T-regulatory cells demonstrated a significant increase in the intermediate 22.5-mg dose cohort, but not in the 225-mg cohort. CONCLUSIONS: CCR9 and OSM are implicated as novel pharmacodynamic and efficacy biomarkers. These findings occur amid coordinated transcriptional changes that enable the definition of surrogate efficacy biomarkers based on inflamed biopsy or blood transcriptomics data.ClinicalTrials.gov identifierNCT01620255.

SUBSTRA: Supervised Bayesian Patient Stratification.

MOTIVATION: Patient stratification methods are key to the vision of precision medicine. Here, we consider transcriptional data to segment the patient population into subsets relevant to a given phenotype. Whereas most existing patient stratification methods focus either on predictive performance or interpretable features, we developed a method striking a balance between these two important goals. RESULTS: We introduce a Bayesian method called SUBSTRA that uses regularized biclustering to identify patient subtypes and interpretable subtype-specific transcript clusters. The method iteratively re-weights feature importance to optimize phenotype prediction performance by producing more phenotype-relevant patient subtypes. We investigate the performance of SUBSTRA in finding relevant features using simulated data and successfully benchmark it against state-of-the-art unsupervised stratification methods and supervised alternatives. Moreover, SUBSTRA achieves predictive performance competitive with the supervised benchmark methods and provides interpretable transcriptional features in diverse biological settings, such as drug response prediction, cancer diagnosis, or kidney transplant rejection. AVAILABILITY AND IMPLEMENTATION: The R code of SUBSTRA is available at https://github.com/sahandk/SUBSTRA. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Profiling of Gene Expression Biomarkers as a Classifier of Methotrexate Nonresponse in Patients With Rheumatoid Arthritis.

OBJECTIVE: Approximately 30-40% of rheumatoid arthritis (RA) patients who are initially started on low-dose methotrexate (MTX) will not benefit from the treatment. To date, no reliable biomarkers of MTX inefficacy in RA have been identified. The aim of this study was to analyze whole blood samples from RA patients at 2 time points (pretreatment and 4 weeks following initiation of MTX), to identify gene expression biomarkers of the MTX response. METHODS: RA patients who were about to commence treatment with MTX were selected from the Rheumatoid Arthritis Medication Study. Using European League Against Rheumatism (EULAR) response criteria, 42 patients were categorized as good responders and 43 as nonresponders at 6 months following the initation of MTX treatment. Data on whole blood transcript expression were generated, and supervised machine learning methods were used to predict a EULAR nonresponse. Models in which transcript levels were included were compared to models in which clinical covariates alone (e.g., baseline disease activity, sex) were included. Gene network and ontology analysis was also performed. RESULTS: Based on the ratio of transcript values (i.e., the difference in log2 -transformed expression values between 4 weeks of treatment and pretreatment), a highly predictive classifier of MTX nonresponse was developed using L2-regularized logistic regression (mean ± SEM area under the receiver operating characteristic [ROC] curve [AUC] 0.78 ± 0.11). This classifier was superior to models that included clinical covariates (ROC AUC 0.63 ± 0.06). Pathway analysis of gene networks revealed significant overrepresentation of type I interferon signaling pathway genes in nonresponders at pretreatment (P = 2.8 × 10-25 ) and at 4 weeks after treatment initiation (P = 4.9 × 10-28 ). CONCLUSION: Testing for changes in gene expression between pretreatment and 4 weeks post-treatment initiation may provide an early classifier of the MTX treatment response in RA patients who are unlikely to benefit from MTX over 6 months. Such patients should, therefore, have their treatment escalated more rapidly, which would thus potentially impact treatment pathways. These findings emphasize the importance of a role for early treatment biomarker monitoring in RA patients started on MTX.

ProGeM: a framework for the prioritization of candidate causal genes at molecular quantitative trait loci.

Quantitative trait locus (QTL) mapping of molecular phenotypes such as metabolites, lipids and proteins through genome-wide association studies represents a powerful means of highlighting molecular mechanisms relevant to human diseases. However, a major challenge of this approach is to identify the causal gene(s) at the observed QTLs. Here, we present a framework for the 'Prioritization of candidate causal Genes at Molecular QTLs' (ProGeM), which incorporates biological domain-specific annotation data alongside genome annotation data from multiple repositories. We assessed the performance of ProGeM using a reference set of 227 previously reported and extensively curated metabolite QTLs. For 98% of these loci, the expert-curated gene was one of the candidate causal genes prioritized by ProGeM. Benchmarking analyses revealed that 69% of the causal candidates were nearest to the sentinel variant at the investigated molecular QTLs, indicating that genomic proximity is the most reliable indicator of 'true positive' causal genes. In contrast, cis-gene expression QTL data led to three false positive candidate causal gene assignments for every one true positive assignment. We provide evidence that these conclusions also apply to other molecular phenotypes, suggesting that ProGeM is a powerful and versatile tool for annotating molecular QTLs. ProGeM is freely available via GitHub.

Using transfer learning from prior reference knowledge to improve the clustering of single-cell RNA-Seq data.

In many research areas scientists are interested in clustering objects within small datasets while making use of prior knowledge from large reference datasets. We propose a method to apply the machine learning concept of transfer learning to unsupervised clustering problems and show its effectiveness in the field of single-cell RNA sequencing (scRNA-Seq). The goal of scRNA-Seq experiments is often the definition and cataloguing of cell types from the transcriptional output of individual cells. To improve the clustering of small disease- or tissue-specific datasets, for which the identification of rare cell types is often problematic, we propose a transfer learning method to utilize large and well-annotated reference datasets, such as those produced by the Human Cell Atlas. Our approach modifies the dataset of interest while incorporating key information from the larger reference dataset via Non-negative Matrix Factorization (NMF). The modified dataset is subsequently provided to a clustering algorithm. We empirically evaluate the benefits of our approach on simulated scRNA-Seq data as well as on publicly available datasets. Finally, we present results for the analysis of a recently published small dataset and find improved clustering when transferring knowledge from a large reference dataset. Implementations of the method are available at https://github.com/nicococo/scRNA.

From hype to reality: data science enabling personalized medicine.

BACKGROUND: Personalized, precision, P4, or stratified medicine is understood as a medical approach in which patients are stratified based on their disease subtype, risk, prognosis, or treatment response using specialized diagnostic tests. The key idea is to base medical decisions on individual patient characteristics, including molecular and behavioral biomarkers, rather than on population averages. Personalized medicine is deeply connected to and dependent on data science, specifically machine learning (often named Artificial Intelligence in the mainstream media). While during recent years there has been a lot of enthusiasm about the potential of 'big data' and machine learning-based solutions, there exist only few examples that impact current clinical practice. The lack of impact on clinical practice can largely be attributed to insufficient performance of predictive models, difficulties to interpret complex model predictions, and lack of validation via prospective clinical trials that demonstrate a clear benefit compared to the standard of care. In this paper, we review the potential of state-of-the-art data science approaches for personalized medicine, discuss open challenges, and highlight directions that may help to overcome them in the future. CONCLUSIONS: There is a need for an interdisciplinary effort, including data scientists, physicians, patient advocates, regulatory agencies, and health insurance organizations. Partially unrealistic expectations and concerns about data science-based solutions need to be better managed. In parallel, computational methods must advance more to provide direct benefit to clinical practice.

A Genome-Wide Association Study of Diabetic Kidney Disease in Subjects With Type 2 Diabetes.

Identification of sequence variants robustly associated with predisposition to diabetic kidney disease (DKD) has the potential to provide insights into the pathophysiological mechanisms responsible. We conducted a genome-wide association study (GWAS) of DKD in type 2 diabetes (T2D) using eight complementary dichotomous and quantitative DKD phenotypes: the principal dichotomous analysis involved 5,717 T2D subjects, 3,345 with DKD. Promising association signals were evaluated in up to 26,827 subjects with T2D (12,710 with DKD). A combined T1D+T2D GWAS was performed using complementary data available for subjects with T1D, which, with replication samples, involved up to 40,340 subjects with diabetes (18,582 with DKD). Analysis of specific DKD phenotypes identified a novel signal near GABRR1 (rs9942471, P = 4.5 × 10-8) associated with microalbuminuria in European T2D case subjects. However, no replication of this signal was observed in Asian subjects with T2D or in the equivalent T1D analysis. There was only limited support, in this substantially enlarged analysis, for association at previously reported DKD signals, except for those at UMOD and PRKAG2, both associated with estimated glomerular filtration rate. We conclude that, despite challenges in addressing phenotypic heterogeneity, access to increased sample sizes will continue to provide more robust inference regarding risk variant discovery for DKD.

Robust phenotype prediction from gene expression data using differential shrinkage of co-regulated genes.

Discovery of robust diagnostic or prognostic biomarkers is a key to optimizing therapeutic benefit for select patient cohorts - an idea commonly referred to as precision medicine. Most discovery studies to derive such markers from high-dimensional transcriptomics datasets are weakly powered with sample sizes in the tens of patients. Therefore, highly regularized statistical approaches are essential to making generalizable predictions. At the same time, prior knowledge-driven approaches have been successfully applied to the manual interpretation of high-dimensional transcriptomics datasets. In this work, we assess the impact of combining two orthogonal approaches for the discovery of biomarker signatures, namely (1) well-known lasso-based regression approaches and its more recent derivative, the group lasso, and (2) the discovery of significant upstream regulators in literature-derived biological networks. Our method integrates both approaches in a weighted group-lasso model and differentially weights gene sets based on inferred active regulatory mechanism. Using nested cross-validation as well as independent clinical datasets, we demonstrate that our approach leads to increased accuracy and generalizable results. We implement our approach in a computationally efficient, user-friendly R package called creNET. The package can be downloaded at https://github.com/kouroshz/creNethttps://github.com/kouroshz/creNet and is accompanied by a parsed version of the STRING DB data base.

From the Cover: Fenretinide, Troglitazone, and Elmiron Add to Weight of Evidence Support for Hemangiosarcoma Mode-of-Action From Studies in Mice.

Pharmaceuticals and chemicals produce hemangiosarcomas (HS) in mice, often by nongenotoxic, proliferative mechanisms. A mode-of-action (MOA) for hemangiosarcoma was proposed based on information presented at an international workshop (Cohen et al., Hemangiosarcoma in rodents: Mode-of-action evaluation and human relevance. Toxicol. Sci. 111, 4-18.). Five key elements of the MOA were articulated and included hypoxia, macrophage activation, increased angiogenic growth factors, dysregulated angiogenesis/erythropoiesis, and endothial cell proliferation. The goal of the current study was to add to the weight-of-evidence for the proposed MOA by assessing these key elements with 3 different compounds of varying potency for HS induction: fenretinide (high), troglitazone (intermediate), and elmiron (low). Multiple endpoints, including hypoxia (hyproxyprobe, transcriptomics), endothelial cell (EC) proliferation, and clinical and anatomic pathology, were assessed after 2, 4, and 13-weeks of treatment in B6C3F1 mice. All 3 compounds demonstrated strong evidence for dysregulated erythropoiesis (decrease in RBC and a failure to increase reticulocytes) and macrophage activation (4- to 11-fold increases); this pattern of hematological changes in mice might serve as an early biomarker to evaluate EC proliferation in suspected target organs for potential HS formation. Fenretinide demonstrated all 5 key elements, while troglitazone demonstrated 4 and elmiron demonstrated 3. Transcriptomics provided support for the 5 elements of the MOA, but was not any more sensitive than hypoxyprobe immunohistochemistry for detecting hypoxia. The overall transcriptional evidence for the key elements of the proposed MOA was also consistent with the potency of HS induction. These data, coupled with the previous work with 2-butoxyethanol and pregablin, increase the weight-of-evidence for the proposed MOA for HS formation.