Discordance in Tumor Mutation Burden from Blood and Tissue Affects Association with Response to Immune Checkpoint Inhibition in Real-World Settings.

BACKGROUND: Tumor mutation burden (TMB), a biomarker for immune checkpoint inhibitor (CPI) response, is reported by both blood- and tissue-based next-generation sequencing (NGS) vendors. However, the agreement between TMB from blood (bTMB) and tissue (tTMB) in real-world settings, both in absolute value and association with CPI response, is not known. MATERIALS AND METHODS: This study utilizes Sarah Cannon's precision medicine platform, Genospace, to harmonize clinico-genomic data from 17 206 patients with cancer with NGS results from September 2015 to August 2021. A subset of patients have both bTMB and tTMB results. Statistical analyses are performed in R and include (1) correlation (r) and concordance (ρ) between patient-matched bTMB-tTMB pairs, (2) distribution of total bTMB and tTMB values, and (3) association of bTMB and tTMB with time to CPI therapy failure. RESULTS: In 410 patient-matched bTMB-tTMB pairs, the median bTMB (m = 10.5 mut/Mb) was significantly higher than the median tTMB (m = 6.0 mut/Mb, P < .001) leading to conflicting "high" and "low" statuses in over one-third of cases at a threshold of 10 mut/Mb (n = 410). Significant differences were observed in the distribution of bTMB values from blood-NGS vendors, with guardant health (GH) reporting higher (m = 10.5 mut/Mb, n = 2183) than Foundation Medicine (FMI, m = 3.8 mut/Mb, n = 462, P < .001). bTMB from GH required a higher threshold (≥40 mut/Mb) than bTMB from FMI (≥12 mut/Mb) in order to be associated with CPI response. CONCLUSIONS: This study uncovers variability in bTMB reporting among commercial NGS platforms, thereby evidencing a need for assay-specific thresholds in identifying patients who may respond to CPI therapy.

Limiting adverse birth outcomes in resource-limited settings (LABOR): protocol of a prospective intrapartum cohort study.

Background: Each year, nearly 300,000 women and 5 million fetuses or neonates die during childbirth or shortly thereafter, a burden concentrated disproportionately in low- and middle-income countries. Identifying women and their fetuses at risk for intrapartum-related morbidity and death could facilitate early intervention. Methods: The Limiting Adverse Birth Outcomes in Resource-Limited Settings (LABOR) Study is a multi-country, prospective, observational cohort designed to exhaustively document the course and outcomes of labor, delivery, and the immediate postpartum period in settings where adverse outcomes are frequent. The study is conducted at four hospitals across three countries in Ghana, India, and Zambia. We will enroll approximately 12,000 women at presentation to the hospital for delivery and follow them and their fetuses/newborns throughout their labor and delivery course, postpartum hospitalization, and up to 42 days thereafter. The co-primary outcomes are composites of maternal (death, hemorrhage, hypertensive disorders, infection) and fetal/neonatal adverse events (death, encephalopathy, sepsis) that may be attributed to the intrapartum period. The study collects extensive physiologic data through the use of physiologic sensors and employs medical scribes to document examination findings, diagnoses, medications, and other interventions in real time. Discussion: The goal of this research is to produce a large, sharable dataset that can be used to build statistical algorithms to prospectively stratify parturients according to their risk of adverse outcomes. We anticipate this research will inform the development of new tools to reduce peripartum morbidity and mortality in low-resource settings.

Estimation and Optimization of Composite Outcomes.

There is tremendous interest in precision medicine as a means to improve patient outcomes by tailoring treatment to individual characteristics. An individualized treatment rule formalizes precision medicine as a map from patient information to a recommended treatment. A treatment rule is defined to be optimal if it maximizes the mean of a scalar outcome in a population of interest, e.g., symptom reduction. However, clinical and intervention scientists often seek to balance multiple and possibly competing outcomes, e.g., symptom reduction and the risk of an adverse event. One approach to precision medicine in this setting is to elicit a composite outcome which balances all competing outcomes; unfortunately, eliciting a composite outcome directly from patients is difficult without a high-quality instrument, and an expert-derived composite outcome may not account for heterogeneity in patient preferences. We propose a new paradigm for the study of precision medicine using observational data that relies solely on the assumption that clinicians are approximately (i.e., imperfectly) making decisions to maximize individual patient utility. Estimated composite outcomes are subsequently used to construct an estimator of an individualized treatment rule which maximizes the mean of patient-specific composite outcomes. The estimated composite outcomes and estimated optimal individualized treatment rule provide new insights into patient preference heterogeneity, clinician behavior, and the value of precision medicine in a given domain. We derive inference procedures for the proposed estimators under mild conditions and demonstrate their finite sample performance through a suite of simulation experiments and an illustrative application to data from a study of bipolar depression.

High-Dimensional Precision Medicine From Patient-Derived Xenografts.

The complexity of human cancer often results in significant heterogeneity in response to treatment. Precision medicine offers the potential to improve patient outcomes by leveraging this heterogeneity. Individualized treatment rules (ITRs) formalize precision medicine as maps from the patient covariate space into the space of allowable treatments. The optimal ITR is that which maximizes the mean of a clinical outcome in a population of interest. Patient-derived xenograft (PDX) studies permit the evaluation of multiple treatments within a single tumor, and thus are ideally suited for estimating optimal ITRs. PDX data are characterized by correlated outcomes, a high-dimensional feature space, and a large number of treatments. Here we explore machine learning methods for estimating optimal ITRs from PDX data. We analyze data from a large PDX study to identify biomarkers that are informative for developing personalized treatment recommendations in multiple cancers. We estimate optimal ITRs using regression-based (Q-learning) and direct-search methods (outcome weighted learning). Finally, we implement a superlearner approach to combine multiple estimated ITRs and show that the resulting ITR performs better than any of the input ITRs, mitigating uncertainty regarding user choice. Our results indicate that PDX data are a valuable resource for developing individualized treatment strategies in oncology. Supplementary materials for this article are available online.

Receiver operating characteristic curves and confidence bands for support vector machines.

Many problems that appear in biomedical decision-making, such as diagnosing disease and predicting response to treatment, can be expressed as binary classification problems. The support vector machine (SVM) is a popular classification technique that is robust to model misspecification and effectively handles high-dimensional data. The relative costs of false positives and false negatives can vary across application domains. The receiving operating characteristic (ROC) curve provides a visual representation of the trade-off between these two types of errors. Because the SVM does not produce a predicted probability, an ROC curve cannot be constructed in the traditional way of thresholding a predicted probability. However, a sequence of weighted SVMs can be used to construct an ROC curve. Although ROC curves constructed using weighted SVMs have great potential for allowing ROC curves analyses that cannot be done by thresholding predicted probabilities, their theoretical properties have heretofore been underdeveloped. We propose a method for constructing confidence bands for the SVM ROC curve and provide the theoretical justification for the SVM ROC curve by showing that the risk function of the estimated decision rule is uniformly consistent across the weight parameter. We demonstrate the proposed confidence band method using simulation studies. We present a predictive model for treatment response in breast cancer as an illustrative example.

Estimating Dynamic Treatment Regimes in Mobile Health Using V-learning.

The vision for precision medicine is to use individual patient characteristics to inform a personalized treatment plan that leads to the best possible health-care for each patient. Mobile technologies have an important role to play in this vision as they offer a means to monitor a patient's health status in real-time and subsequently to deliver interventions if, when, and in the dose that they are needed. Dynamic treatment regimes formalize individualized treatment plans as sequences of decision rules, one per stage of clinical intervention, that map current patient information to a recommended treatment. However, most existing methods for estimating optimal dynamic treatment regimes are designed for a small number of fixed decision points occurring on a coarse time-scale. We propose a new reinforcement learning method for estimating an optimal treatment regime that is applicable to data collected using mobile technologies in an out-patient setting. The proposed method accommodates an indefinite time horizon and minute-by-minute decision making that are common in mobile health applications. We show that the proposed estimators are consistent and asymptotically normal under mild conditions. The proposed methods are applied to estimate an optimal dynamic treatment regime for controlling blood glucose levels in patients with type 1 diabetes.

Estimating individualized treatment regimes from crossover designs.

The field of precision medicine aims to tailor treatment based on patient-specific factors in a reproducible way. To this end, estimating an optimal individualized treatment regime (ITR) that recommends treatment decisions based on patient characteristics to maximize the mean of a prespecified outcome is of particular interest. Several methods have been proposed for estimating an optimal ITR from clinical trial data in the parallel group setting where each subject is randomized to a single intervention. However, little work has been done in the area of estimating the optimal ITR from crossover study designs. Such designs naturally lend themselves to precision medicine since they allow for observing the response to multiple treatments for each patient. In this paper, we introduce a method for estimating the optimal ITR using data from a 2 × 2 crossover study with or without carryover effects. The proposed method is similar to policy search methods such as outcome weighted learning; however, we take advantage of the crossover design by using the difference in responses under each treatment as the observed reward. We establish Fisher and global consistency, present numerical experiments, and analyze data from a feeding trial to demonstrate the improved performance of the proposed method compared to standard methods for a parallel study design.

Renal volumes and estimated glomerular filtration rate changes after fenestrated-branched endovascular aortic repair.

OBJECTIVE: To describe changes in renal volumes (RV) and renal function after fenestrated-branched endovascular repair (F-BEVAR) for complex aortic aneurysms. METHODS: The data from patients enrolled in a physician-sponsored investigational device exemption clinical trial for endovascular treatment of complex aortic aneurysms from July 2012 to April 2017 were retrospectively analyzed. Descriptive statistics were calculated using the mean ± standard deviation. The mean estimated glomerular filtration rate (eGFR) and RV were calculated at baseline and 6, 12, and 18 months after F-BEVAR. Variable distributions were evaluated for skewness, and all models required log-transformation. Linear models using generalized estimating equations were used to assess the association between the RV and eGFR over time after adjustment for relevant covariates. We used Kaplan-Meier life-table analysis to calculate survival and branch patency. RESULTS: A total of 139 patients were followed up for 18 months or until death. The mean age was 71 ± 8 years (70% male). The most common risk factor was hypertension (92%). Chronic kidney disease (CKD; eGFR <60 mL/min) was present in 56 patients (40%). Thirty-one patients (22%) had ≥1 accessory renal artery. Of these 31 accessory arteries, 27 (87%) were embolized or covered. On average, the eGFR had decreased over time compared with baseline, with a median change of -4.4 mL/min (interquartile range [IQR], -11.4 to 4.9 mL/min), -2.6 mL/min (IQR, -11.9 to 6.5 mL/min), and -3.4 mL/min (IQR, -11.9 to 5.5 mL/min) at 6, 12, and 18 months postoperatively, respectively. Similarly, the RV had decreased from baseline by 8% ± 17%, 10% ± 17%, and 11% ± 22% at 6, 12, and 18 months, respectively. An increase in the baseline patient age of 5 years was estimated to be associated with a 3% (95% confidence interval [CI], 0.2%-6.0%) decrease in the mean eGFR during the follow-up period, collapsing over time. This change is similar to the natural history of renal deterioration with age. We estimated that an increase in the log-RV of 1 U would be associated with an estimated 26% (95% CI, 3%-52%) increase in the mean eGFR. Preexisting CKD did not affect the average change in RV. Of the 56 patients with previous CKD, 9 (16.1%) showed improvement in the eGFR to >60 mL/min. The median follow-up period was 17.9 months (IQR, 6.3-24.8). The Kaplan-Meier survival rate at 1 and 2 years was 84.7% (95% CI, 78.3%-91.1%) and 78.8% (95% CI, 71.0%- 86.6%), respectively. CONCLUSIONS: The RV and eGFR decreased in patients undergoing repair at the rates expected for patients with complex aortic disease. The eGFR correlated with the RV. Most of the decline in renal function occurred within the first 6 months postoperatively, after which, the renal function had stabilized.

Prepectoral Direct-to-Implant Breast Reconstruction: Early Outcomes and Analysis of Postoperative Pain.

BACKGROUND: Direct-to-implant (DTI) breast reconstruction provides high-quality aesthetic results in appropriate candidates. Most commonly, implants are placed in the subpectoral space which can lead to pain and breast animation. Surgical and technological advances have allowed for successful prepectoral implant placement which may eliminate these trade-offs. OBJECTIVES: Here we present early outcomes from 153 reconstructions in 94 patients who underwent prepectoral DTI. We sought to determine whether these patients have less postoperative pain and narcotic use than subpectoral implant or expander placement. METHODS: A retrospective review was performed for all prepectoral DTI reconstructions at our institution from 2015 to 2016. Data were collected on postoperative pain and narcotic use while in hospital. RESULTS: The average follow-up time was 8.5 months (range, 3-17 months) and the overall complication rate was 27% (n = 41) with the most common complications being skin necrosis (9%, n = 13) and infection (7%, n = 11). No statistically significant difference in complications was found in patients who underwent postmastectomy radiation therapy. Patients who underwent prepectoral DTI reconstruction did not have a statistically significant difference in postoperative pain and narcotic use while in-hospital compared with other techniques. CONCLUSION: Prepectoral DTI reconstruction provides good results with similar complication rates to subpectoral techniques. Prepectoral DTI eliminates the problem of breast animation. Although our series did not reach statistical significance in pain scores or requirement for postoperative narcotics, we believe that it is an important preliminary result and with larger numbers we anticipate a more definitive conclusion.

Multiple-Institution Comparison of Resident and Faculty Perceptions of Burnout and Depression During Surgical Training.

Importance: Prior studies demonstrate a high prevalence of burnout and depression among surgeons. Limited data exist regarding how these conditions are perceived by the surgical community. Objectives: To measure prevalence of burnout and depression among general surgery trainees and to characterize how residents and attendings perceive these conditions. Design, Setting, and Participants: This cross-sectional study used unique, anonymous surveys for residents and attendings that were administered via a web-based platform from November 1, 2016, through March 31, 2017. All residents and attendings in the 6 general surgery training programs in North Carolina were invited to participate. Main Outcomes and Measures: The prevalence of burnout and depression among residents was assessed using validated tools. Burnout was defined by high emotional exhaustion or depersonalization on the Maslach Burnout Inventory. Depression was defined by a score of 10 or greater on the Patient Health Questionnaire-9. Linear and logistic regression models were used to assess predictive factors for burnout and depression. Residents' and attendings' perceptions of these conditions were analyzed for significant similarities and differences. Results: In this study, a total of 92 residents and 55 attendings responded. Fifty-eight of 77 residents with complete responses (75%) met criteria for burnout, and 30 of 76 (39%) met criteria for depression. Of those with burnout, 28 of 58 (48%) were at elevated risk of depression (P = .03). Nine of 77 residents (12%) had suicidal ideation in the past 2 weeks. Most residents (40 of 76 [53%]) correctly estimated that more than 50% of residents had burnout, whereas only 13 of 56 attendings (23%) correctly estimated this prevalence (P < .001). Forty-two of 83 residents (51%) and 42 of 56 attendings (75%) underestimated the true prevalence of depression (P = .002). Sixty-six of 73 residents (90%) and 40 of 51 attendings (78%) identified the same top 3 barriers to seeking care for burnout: inability to take time off to seek treatment, avoidance or denial of the problem, and negative stigma toward those seeking care. Conclusions and Relevance: The prevalence of burnout and depression was high among general surgery residents in this study. Attendings and residents underestimated the prevalence of these conditions but acknowledged common barriers to seeking care. Discrepancies in actual and perceived levels of burnout and depression may hinder wellness interventions. Increasing understanding of these perceptions offers an opportunity to develop practical solutions.

The impact of current smoking on outcomes after infrainguinal bypass for claudication.

OBJECTIVE: Although smoking cessation is a benchmark of medical management of intermittent claudication, many patients require further revascularization. Currently, revascularization among smokers is a controversial topic, and practice patterns differ institutionally, regionally, and nationally. Patients who smoke at the time of revascularization are thought to have a poor prognosis, but data on this topic are limited. The purpose of this study was to evaluate the impact of smoking on outcomes after infrainguinal bypass for claudication. METHODS: Data from the national Vascular Quality Initiative from 2004 to 2014 were used to identify infrainguinal bypasses performed for claudication. Patients were categorized as former smokers (quit >1 year before intervention) and current smokers (smoking within 1 year of intervention). Demographic and comorbid differences of categorical variables were assessed. Significant predictors were included in adjusted Cox proportional hazards models to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) by smoking status for outcomes of major adverse limb event (MALE), amputation-free survival, limb loss, death, and MALE or death. Cumulative incidence curves were created using competing risks modeling. RESULTS: We identified 2913 patients (25% female, 9% black) undergoing incident infrainguinal bypass grafting for claudication. There were 1437 current smokers and 1476 former smokers in our study. Current smoking status was a significant predictor of MALE (HR, 1.27; 95% CI, 1.00-1.60; P = .048) and MALE or death (HR, 1.22; 95% CI, 1.03-1.44; P = .02). Other factors found to be independently associated with poor outcomes in adjusted models included black race, below-knee bypass grafting, use of prosthetic conduit, and dialysis dependence. CONCLUSIONS: Current smokers undergoing an infrainguinal bypass procedure for claudication experienced more MALEs than former smokers did. Future studies with longer term follow-up should address limitations of this study by identifying a data source with long-term follow-up examining the relationship of smoking exposure (pack history and duration) with outcomes.

Laparoscopic Heller Myotomy Versus Peroral Endoscopic Myotomy (POEM) for Achalasia: A Systematic Review and Meta-analysis.

OBJECTIVE: To compare the outcome of per oral endoscopic myotomy (POEM) and laparoscopic Heller myotomy (LHM) for the treatment of esophageal achalasia. BACKGROUND: Over the last 2 decades, LHM has become the primary form of treatment in many centers. However, since the first description of POEM in 2010, this technique has widely disseminated, despite the absence of long-term results and randomized trials. METHODS: A systematic Medline literature search of articles on LHM and POEM for the treatment of achalasia was performed. The main outcomes measured were improvement of dysphagia and posttreatment gastroesophageal reflux disease (GERD). Linear regression was used to model the effect of each procedure on the different outcomes. RESULTS: Fifty-three studies reported data on LHM (5834 patients), and 21 articles examined POEM (1958 patients). Mean follow-up was significantly longer for studies of LHM (41.5 vs. 16.2 mo, P < 0.0001). Predicted probabilities for improvement in dysphagia at 12 months were 93.5% for POEM and 91.0% for LHM (P = 0.01), and at 24 months were 92.7% for POEM and 90.0% for LHM (P = 0.01). Patients undergoing POEM were more likely to develop GERD symptoms (OR 1.69, 95% CI 1.33-2.14, P < 0.0001), GERD evidenced by erosive esophagitis (OR 9.31, 95% CI 4.71-18.85, P < 0.0001), and GERD evidenced by pH monitoring (OR 4.30, 95% CI 2.96-6.27, P < 0.0001). On average, length of hospital stay was 1.03 days longer after POEM (P = 0.04). CONCLUSIONS: Short-term results show that POEM is more effective than LHM in relieving dysphagia, but it is associated with a very high incidence of pathologic reflux.