A framework for leveraging machine learning tools to estimate personalized survival curves.

The conditional survival function of a time-to-event outcome subject to censoring and truncation is a common target of estimation in survival analysis. This parameter may be of scientific interest and also often appears as a nuisance in nonparametric and semiparametric problems. In addition to classical parametric and semiparametric methods (e.g., based on the Cox proportional hazards model), flexible machine learning approaches have been developed to estimate the conditional survival function. However, many of these methods are either implicitly or explicitly targeted toward risk stratification rather than overall survival function estimation. Others apply only to discrete-time settings or require inverse probability of censoring weights, which can be as difficult to estimate as the outcome survival function itself. Here, we employ a decomposition of the conditional survival function in terms of observable regression models in which censoring and truncation play no role. This allows application of an array of flexible regression and classification methods rather than only approaches that explicitly handle the complexities inherent to survival data. We outline estimation procedures based on this decomposition, empirically assess their performance, and demonstrate their use on data from an HIV vaccine trial. Supplementary materials for this article are available online.

Stem cell transcriptional profiles from mouse subspecies reveal cis-regulatory evolution at translation genes.

A key goal of evolutionary genomics is to harness molecular data to draw inferences about selective forces that have acted on genomes. The field progresses in large part through the development of advanced molecular-evolution analysis methods. Here we explored the intersection between classical sequence-based tests for selection and an empirical expression-based approach, using stem cells from Mus musculus subspecies as a model. Using a test of directional, cis-regulatory evolution across genes in pathways, we discovered a unique program of induction of translation genes in stem cells of the Southeast Asian mouse M. m. castaneus relative to its sister taxa. We then mined population-genomic sequences to pursue underlying regulatory mechanisms for this expression divergence, finding robust evidence for alleles unique to M. m. castaneus at the upstream regions of the translation genes. We interpret our data under a model of changes in lineage-specific pressures across Mus musculus in stem cells with high translational capacity. Our findings underscore the rigor of integrating expression and sequence-based methods to generate hypotheses about evolutionary events from long ago.

Statistical approaches for the integration of external controls in a cystic fibrosis clinical trial: a simulation and an application.

Development of new therapeutics for a rare disease such as cystic fibrosis (CF) is hindered by challenges in accruing enough patients for clinical trials. Using external controls from well-matched historical trials can reduce prospective trial sizes, and this approach has supported regulatory approval of new interventions for other rare diseases. We consider three statistical methods that incorporate external controls into a hypothetical clinical trial of a new treatment to reduce pulmonary exacerbations in CF patients: 1) inverse probability weighting, 2) Bayesian modeling with propensity score-based power priors, and 3) hierarchical Bayesian modeling with commensurate priors. We compare the methods via simulation study and in a real clinical trial data setting. Simulations showed that bias in the treatment effect was <4% using any of the methods, with type 1 error (or in the Bayesian cases, posterior probability of the null hypothesis) usually <5%. Inverse probability weighting was sensitive to similarity in prevalence of the covariates between historical and prospective trial populations. The commensurate prior method performed best with real clinical trial data. Using external controls to reduce trial size in future clinical trials holds promise and can advance the therapeutic pipeline for rare diseases.

Predicting Out-of-Hospital Cardiac Arrest in the General Population Using Electronic Health Records.

BACKGROUND: The majority of out-of-hospital cardiac arrests (OHCAs) occur among individuals in the general population, for whom there is no established strategy to identify risk. In this study, we assess the use of electronic health record (EHR) data to identify OHCA in the general population and define salient factors contributing to OHCA risk. METHODS: The analytical cohort included 2366 individuals with OHCA and 23 660 age- and sex-matched controls receiving health care at the University of Washington. Comorbidities, electrocardiographic measures, vital signs, and medication prescription were abstracted from the EHR. The primary outcome was OHCA. Secondary outcomes included shockable and nonshockable OHCA. Model performance including area under the receiver operating characteristic curve and positive predictive value were assessed and adjusted for observed rate of OHCA across the health system. RESULTS: There were significant differences in demographic characteristics, vital signs, electrocardiographic measures, comorbidities, and medication distribution between individuals with OHCA and controls. In external validation, discrimination in machine learning models (area under the receiver operating characteristic curve 0.80-0.85) was superior to a baseline model with conventional cardiovascular risk factors (area under the receiver operating characteristic curve 0.66). At a specificity threshold of 99%, correcting for baseline OHCA incidence across the health system, positive predictive value was 2.5% to 3.1% in machine learning models compared with 0.8% for the baseline model. Longer corrected QT interval, substance abuse disorder, fluid and electrolyte disorder, alcohol abuse, and higher heart rate were identified as salient predictors of OHCA risk across all machine learning models. Established cardiovascular risk factors retained predictive importance for shockable OHCA, but demographic characteristics (minority race, single marital status) and noncardiovascular comorbidities (substance abuse disorder) also contributed to risk prediction. For nonshockable OHCA, a range of salient predictors, including comorbidities, habits, vital signs, demographic characteristics, and electrocardiographic measures, were identified. CONCLUSIONS: In a population-based case-control study, machine learning models incorporating readily available EHR data showed reasonable discrimination and risk enrichment for OHCA in the general population. Salient factors associated with OCHA risk were myriad across the cardiovascular and noncardiovascular spectrum. Public health and tailored strategies for OHCA prediction and prevention will require incorporation of this complexity.

Machine learning models for predicting blood pressure phenotypes by combining multiple polygenic risk scores.

We construct non-linear machine learning (ML) prediction models for systolic and diastolic blood pressure (SBP, DBP) using demographic and clinical variables and polygenic risk scores (PRSs). We developed a two-model ensemble, consisting of a baseline model, where prediction is based on demographic and clinical variables only, and a genetic model, where we also include PRSs. We evaluate the use of a linear versus a non-linear model at both the baseline and the genetic model levels and assess the improvement in performance when incorporating multiple PRSs. We report the ensemble model's performance as percentage variance explained (PVE) on a held-out test dataset. A non-linear baseline model improved the PVEs from 28.1 to 30.1% (SBP) and 14.3% to 17.4% (DBP) compared with a linear baseline model. Including seven PRSs in the genetic model computed based on the largest available GWAS of SBP/DBP improved the genetic model PVE from 4.8 to 5.1% (SBP) and 4.7 to 5% (DBP) compared to using a single PRS. Adding additional 14 PRSs computed based on two independent GWASs further increased the genetic model PVE to 6.3% (SBP) and 5.7% (DBP). PVE differed across self-reported race/ethnicity groups, with primarily all non-White groups benefitting from the inclusion of additional PRSs. In summary, non-linear ML models improves BP prediction in models incorporating diverse populations.

Clinical performance and utility: A microsimulation model to inform the design of screening trials for a multi-cancer early detection test.

OBJECTIVES: Designing cancer screening trials for multi-cancer early detection (MCED) tests presents a significant methodology challenge, as natural histories of cell-free DNA-shedding cancers are not yet known. A microsimulation model was developed to project the performance and utility of an MCED test in cancer screening trials. METHODS: Individual natural history of preclinical progression through cancer stages for 23 cancer classes was simulated by a stage-transition model under a broad range of cancer latency parameters. Cancer incidences and stage distributions at clinical presentation in simulated trials were set to match the data from Surveillance, Epidemiology, and End Results program. One or multiple rounds of annual screening using a targeted methylation-based MCED test (GalleriⓇ) was conducted to detect preclinical cancers. Mortality benefit of early detection was simulated by a stage-shift model. RESULTS: In simulated trials, accounting for healthy volunteer effect and varying test sensitivity, positive predictive value in the prevalence screening round reached 48% to 61% in 6 natural history scenarios. After 3 rounds of annual screening, the cumulative proportions of stage I/II cancers increased by approximately 9% to 14%, the incidence of stage IV cancers was reduced by 37% to 46%, the reduction of stages III and IV cancer incidences was 9% to 24%, and the reduction of mortality reached 13% to 16%. Greater reductions of late-stage cancers and cancer mortality were achieved by five rounds of MCED screening. CONCLUSIONS: Simulation results guide trial design and suggest that adding this MCED test to routine screening in the United States may shift cancer detection to earlier stages, and potentially save lives.

In vitro modulator responsiveness of 655 CFTR variants found in people with cystic fibrosis.

BACKGROUND: In 2017, the US Food and Drug Administration initiated expansion of drug labels for the treatment of cystic fibrosis (CF) to include CF transmembrane conductance regulator (CFTR) gene variants based on in vitro functional studies. This study aims to identify CFTR variants that result in increased chloride (Cl-) transport function by the CFTR protein after treatment with the CFTR modulator combination elexacaftor/tezacaftor/ivacaftor (ELX/TEZ/IVA). These data may benefit people with CF (pwCF) who are not currently eligible for modulator therapies. METHODS: Plasmid DNA encoding 655 CFTR variants and wild-type (WT) CFTR were transfected into Fisher Rat Thyroid cells that do not natively express CFTR. After 24 h of incubation with control or TEZ and ELX, and acute addition of IVA, CFTR function was assessed using the transepithelial current clamp conductance assay. Each variant's forskolin/cAMP-induced baseline Cl- transport activity, responsiveness to IVA alone, and responsiveness to the TEZ/ELX/IVA combination were measured in three different laboratories. Western blots were conducted to evaluate CFTR protein maturation and complement the functional data. RESULTS AND CONCLUSIONS: 253 variants not currently approved for CFTR modulator therapy showed low baseline activity (<10 % of normal CFTR Cl- transport activity). For 152 of these variants, treatment with ELX/TEZ/IVA improved the Cl- transport activity by ≥10 % of normal CFTR function, which is suggestive of clinical benefit. ELX/TEZ/IVA increased CFTR function by ≥10 percentage points for an additional 140 unapproved variants with ≥10 % but <50 % of normal CFTR function at baseline. These findings significantly expand the number of rare CFTR variants for which ELX/TEZ/IVA treatment should result in clinical benefit.

Machine learning models for blood pressure phenotypes combining multiple polygenic risk scores.

We construct non-linear machine learning (ML) prediction models for systolic and diastolic blood pressure (SBP, DBP) using demographic and clinical variables and polygenic risk scores (PRSs). We developed a two-model ensemble, consisting of a baseline model, where prediction is based on demographic and clinical variables only, and a genetic model, where we also include PRSs. We evaluate the use of a linear versus a non-linear model at both the baseline and the genetic model levels and assess the improvement in performance when incorporating multiple PRSs. We report the ensemble model's performance as percentage variance explained (PVE) on a held-out test dataset. A non-linear baseline model improved the PVEs from 28.1% to 30.1% (SBP) and 14.3% to 17.4% (DBP) compared with a linear baseline model. Including seven PRSs in the genetic model computed based on the largest available GWAS of SBP/DBP improved the genetic model PVE from 4.8% to 5.1% (SBP) and 4.7% to 5% (DBP) compared to using a single PRS. Adding additional 14 PRSs computed based on two independent GWASs further increased the genetic model PVE to 6.3% (SBP) and 5.7% (DBP). PVE differed across self-reported race/ethnicity groups, with primarily all non-White groups benefitting from the inclusion of additional PRSs.

A SIEVE STOCHASTIC GRADIENT DESCENT ESTIMATOR FOR ONLINE NONPARAMETRIC REGRESSION IN SOBOLEV ELLIPSOIDS.

The goal of regression is to recover an unknown underlying function that best links a set of predictors to an outcome from noisy observations. in nonparametric regression, one assumes that the regression function belongs to a pre-specified infinite-dimensional function space (the hypothesis space). in the online setting, when the observations come in a stream, it is computationally-preferable to iteratively update an estimate rather than refitting an entire model repeatedly. inspired by nonparametric sieve estimation and stochastic approximation methods, we propose a sieve stochastic gradient descent estimator (Sieve-SGD) when the hypothesis space is a Sobolev ellipsoid. We show that Sieve-SGD has rate-optimal mean squared error (MSE) under a set of simple and direct conditions. The proposed estimator can be constructed with a low computational (time and space) expense: We also formally show that Sieve-SGD requires almost minimal memory usage among all statistically rate-optimal estimators.