A Bayesian method to estimate variant-induced disease penetrance.

A major challenge emerging in genomic medicine is how to assess best disease risk from rare or novel variants found in disease-related genes. The expanding volume of data generated by very large phenotyping efforts coupled to DNA sequence data presents an opportunity to reinterpret genetic liability of disease risk. Here we propose a framework to estimate the probability of disease given the presence of a genetic variant conditioned on features of that variant. We refer to this as the penetrance, the fraction of all variant heterozygotes that will present with disease. We demonstrate this methodology using a well-established disease-gene pair, the cardiac sodium channel gene SCN5A and the heart arrhythmia Brugada syndrome. From a review of 756 publications, we developed a pattern mixture algorithm, based on a Bayesian Beta-Binomial model, to generate SCN5A penetrance probabilities for the Brugada syndrome conditioned on variant-specific attributes. These probabilities are determined from variant-specific features (e.g. function, structural context, and sequence conservation) and from observations of affected and unaffected heterozygotes. Variant functional perturbation and structural context prove most predictive of Brugada syndrome penetrance.

Missing data and prediction: the pattern submodel.

Missing data are a common problem for both the construction and implementation of a prediction algorithm. Pattern submodels (PS)-a set of submodels for every missing data pattern that are fit using only data from that pattern-are a computationally efficient remedy for handling missing data at both stages. Here, we show that PS (i) retain their predictive accuracy even when the missing data mechanism is not missing at random (MAR) and (ii) yield an algorithm that is the most predictive among all standard missing data strategies. Specifically, we show that the expected loss of a forecasting algorithm is minimized when each pattern-specific loss is minimized. Simulations and a re-analysis of the SUPPORT study confirms that PS generally outperforms zero-imputation, mean-imputation, complete-case analysis, complete-case submodels, and even multiple imputation (MI). The degree of improvement is highly dependent on the missingness mechanism and the effect size of missing predictors. When the data are MAR, MI can yield comparable forecasting performance but generally requires a larger computational cost. We also show that predictions from the PS approach are equivalent to the limiting predictions for a MI procedure that is dependent on missingness indicators (the MIMI model). The focus of this article is on out-of-sample prediction; implications for model inference are only briefly explored.

Effect of balanced crystalloids versus saline on urinary biomarkers of acute kidney injury in critically ill adults.

BACKGROUND: Recent trials have suggested use of balanced crystalloids may decrease the incidence of major adverse kidney events compared to saline in critically ill adults. The effect of crystalloid composition on biomarkers of early acute kidney injury remains unknown. METHODS: From February 15 to July 15, 2016, we conducted an ancillary study to the Isotonic Solutions and Major Adverse Renal Events Trial (SMART) comparing the effect of balanced crystalloids versus saline on urinary levels of neutrophil gelatinase-associated lipocalin (NGAL) and kidney injury molecule-1 (KIM-1) among 261 consecutively-enrolled critically ill adults admitted from the emergency department to the medical ICU. After informed consent, we collected urine 36 ± 12 h after hospital admission and measured NGAL and KIM-1 levels using commercially available ELISAs. Levels of NGAL and KIM-1 at 36 ± 12 h were compared between patients assigned to balanced crystalloids versus saline using a Mann-Whitney U test. RESULTS: The 131 patients (50.2%) assigned to the balanced crystalloid group and the 130 patients (49.8%) assigned to the saline group were similar at baseline. Urinary NGAL levels were significantly lower in the balanced crystalloid group (median, 39.4 ng/mg [IQR 9.9 to 133.2]) compared with the saline group (median, 64.4 ng/mg [IQR 27.6 to 339.9]) (P < 0.001). Urinary KIM-1 levels did not significantly differ between the balanced crystalloid group (median, 2.7 ng/mg [IQR 1.5 to 4.9]) and the saline group (median, 2.4 ng/mg [IQR 1.3 to 5.0]) (P = 0.36). CONCLUSIONS: In this ancillary analysis of a clinical trial comparing balanced crystalloids to saline among critically ill adults, balanced crystalloids were associated with lower urinary concentrations of NGAL and similar urinary concentrations of KIM-1, compared with saline. These results suggest only a modest reduction in early biomarkers of acute kidney injury with use of balanced crystalloids compared with saline. TRIAL REGISTRATION: ClinicalTrials.gov number: NCT02444988 . Date registered: May 15, 2015.

A classical regression framework for mediation analysis: fitting one model to estimate mediation effects.

Mediation analysis explores the degree to which an exposure's effect on an outcome is diverted through a mediating variable. We describe a classical regression framework for conducting mediation analyses in which estimates of causal mediation effects and their variance are obtained from the fit of a single regression model. The vector of changes in exposure pathway coefficients, which we named the essential mediation components (EMCs), is used to estimate standard causal mediation effects. Because these effects are often simple functions of the EMCs, an analytical expression for their model-based variance follows directly. Given this formula, it is instructive to revisit the performance of routinely used variance approximations (e.g., delta method and resampling methods). Requiring the fit of only one model reduces the computation time required for complex mediation analyses and permits the use of a rich suite of regression tools that are not easily implemented on a system of three equations, as would be required in the Baron-Kenny framework. Using data from the BRAIN-ICU study, we provide examples to illustrate the advantages of this framework and compare it with the existing approaches.

A Regression Framework for Causal Mediation Analysis with Applications to Behavioral Science.

We introduce and extend the classical regression framework for conducting mediation analysis from the fit of only one model. Using the essential mediation components (EMCs) allows us to estimate causal mediation effects and their analytical variance. This single-equation approach reduces computation time and permits the use of a rich suite of regression tools that are not easily implemented on a system of three equations. Additionally, we extend this framework to non-nested mediation systems, provide a joint measure of mediation for complex mediation hypotheses, propose new visualizations for mediation effects, and explain why estimates of the total effect may differ depending on the approach used. Using data from social science studies, we also provide extensive illustrations of the usefulness of this framework and its advantages over traditional approaches to mediation analysis. The example data are freely available for download online and we include the R code necessary to reproduce our results.

Acute Kidney Injury and Subsequent Frailty Status in Survivors of Critical Illness: A Secondary Analysis.

OBJECTIVES: Acute kidney injury frequently complicates critical illness and is associated with high morbidity and mortality. Frailty is common in critical illness survivors, but little is known about the impact of acute kidney injury. We examined the association of acute kidney injury and frailty within a year of hospital discharge in survivors of critical illness. DESIGN: Secondary analysis of a prospective cohort study. SETTING: Medical/surgical ICU of a U.S. tertiary care medical center. PATIENTS: Three hundred seventeen participants with respiratory failure and/or shock. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Acute kidney injury was determined using Kidney Disease Improving Global Outcomes stages. Clinical frailty status was determined using the Clinical Frailty Scale at 3 and 12 months following discharge. Covariates included mean ICU Sequential Organ Failure Assessment score and Acute Physiology and Chronic Health Evaluation II score as well as baseline comorbidity (i.e., Charlson Comorbidity Index), kidney function, and Clinical Frailty Scale score. Of 317 patients, 243 (77%) had acute kidney injury and one in four patients with acute kidney injury was frail at baseline. In adjusted models, acute kidney injury stages 1, 2, and 3 were associated with higher frailty scores at 3 months (odds ratio, 1.92; 95% CI, 1.14-3.24; odds ratio, 2.40; 95% CI, 1.31-4.42; and odds ratio, 4.41; 95% CI, 2.20-8.82, respectively). At 12 months, a similar association of acute kidney injury stages 1, 2, and 3 and higher Clinical Frailty Scale score was noted (odds ratio, 1.87; 95% CI, 1.11-3.14; odds ratio, 1.81; 95% CI, 0.94-3.48; and odds ratio, 2.76; 95% CI, 1.34-5.66, respectively). In supplemental and sensitivity analyses, analogous patterns of association were observed. CONCLUSIONS: Acute kidney injury in survivors of critical illness predicted worse frailty status 3 and 12 months postdischarge. These findings have important implications on clinical decision making among acute kidney injury survivors and underscore the need to understand the drivers of frailty to improve patient-centered outcomes.

Acute Kidney Injury as a Risk Factor for Delirium and Coma during Critical Illness.

RATIONALE: Acute kidney injury may contribute to distant organ dysfunction. Few studies have examined kidney injury as a risk factor for delirium and coma. OBJECTIVES: To examine whether acute kidney injury is associated with delirium and coma in critically ill adults. METHODS: In a prospective cohort study of intensive care unit patients with respiratory failure and/or shock, we examined the association between acute kidney injury and daily mental status using multinomial transition models adjusting for demographics, nonrenal organ failure, sepsis, prior mental status, and sedative exposure. Acute kidney injury was characterized daily using the difference between baseline and peak serum creatinine and staged according to Kidney Disease Improving Global Outcomes criteria. Mental status (normal vs. delirium vs. coma) was assessed daily with the Confusion Assessment Method for the ICU and Richmond Agitation-Sedation Scale. MEASUREMENTS AND MAIN RESULTS: Among 466 patients, stage 2 acute kidney injury was a risk factor for delirium (odds ratio [OR], 1.55; 95% confidence interval [CI], 1.07-2.26) and coma (OR, 2.04; 95% CI, 1.25-3.34) as was stage 3 injury (OR for delirium, 2.56; 95% CI, 1.57-4.16) (OR for coma, 3.34; 95% CI, 1.85-6.03). Daily peak serum creatinine (adjusted for baseline) values were also associated with delirium (OR, 1.35; 95% CI, 1.18-1.55) and coma (OR, 1.44; 95% CI, 1.20-1.74). Renal replacement therapy modified the association between stage 3 acute kidney injury and daily peak serum creatinine and both delirium and coma. CONCLUSIONS: Acute kidney injury is a risk factor for delirium and coma during critical illness.

Variables affecting maximum urethral closure pressure (MUCP) and abdominal leak point pressure (ALPP) measurements.

INTRODUCTION AND HYPOTHESIS: The relationship between pelvic floor muscles and measurements of urethral function is not well studied. It is not known whether adjusting for clinical, demographic and urodynamic parameters would improve the association between MUCP and ALPP. Our hypothesis was that pelvic floor muscle strength (PFMS) influences the relationship between MUCP and ALPP. METHODS: This was a retrospective study of women who underwent a complex urodynamic study with evaluation of MUCP and ALPP using ICD-9 codes with documentation of PFMS. RESULTS: Urodynamic stress incontinence was confirmed in 478 patients, of whom 323 had MUCP recorded and 263 had both MUCP and ALPP recorded. Women with higher PFMS had a higher MUCP. In regression analysis ALPP at 150 mL and MUCP were weakly associated (coefficient 0.43, 95% CI 0.08-0.78; p = 0.02), whereas ALPP at capacity and MUCP were moderately associated (coefficient 0.60, 95% CI 0.25-0.95; p < 0.001). CONCLUSIONS: This study showed that MUCP and ALPP at 150 mL were weakly associated and that this improved to a moderate association for ALPP at capacity. MUCP increased with increasing PFMS among women with stress urinary incontinence and decreased with increasing age. There was no evidence that ALPP was associated with PFMS or age. The relationship between MUCP and ALPP was unchanged when accounting for covariates of PFMS (age, parity, BMI, prior procedure, urethral mobility, bladder capacity, stage of cystocele, or stage of uterine or apical prolapse).

Latent variable modeling improves AKI risk factor identification and AKI prediction compared to traditional methods.

BACKGROUND: Acute kidney injury (AKI) is diagnosed based on postoperative serum creatinine change, but AKI models have not consistently performed well, in part due to the omission of clinically important but practically unmeasurable variables that affect creatinine. We hypothesized that a latent variable mixture model of postoperative serum creatinine change would partially account for these unmeasured factors and therefore increase power to identify risk factors of AKI and improve predictive accuracy. METHODS: We constructed a two-component latent variable mixture model and a linear model using data from a prospective, 653-subject randomized clinical trial of AKI following cardiac surgery (NCT00791648) and included established AKI risk factors and covariates known to affect serum creatinine. We compared model fit, discrimination, power to detect AKI risk factors, and ability to predict AKI between the latent variable mixture model and the linear model. RESULTS: The latent variable mixture model demonstrated superior fit (likelihood ratio of 6.68 × 1071) and enhanced discrimination (permutation test of Spearman's correlation coefficients, p < 0.001) compared to the linear model. The latent variable mixture model was 94% (-13 to 1132%) more powerful (median [range]) at identifying risk factors than the linear model, and demonstrated increased ability to predict change in serum creatinine (relative mean square error reduction of 6.8%). CONCLUSIONS: A latent variable mixture model better fit a clinical cohort of cardiac surgery patients than a linear model, thus providing better assessment of the associations between risk factors of AKI and serum creatinine change and more accurate prediction of AKI. Incorporation of latent variable mixture modeling into AKI research will allow clinicians and investigators to account for clinically meaningful patient heterogeneity resulting from unmeasured variables, and therefore provide improved ability to examine risk factors, measure mechanisms and mediators of kidney injury, and more accurately predict AKI in clinical cohorts.

Documentation of an Imperative To Improve Methods for Predicting Membrane Protein Stability.

There is a compelling and growing need to accurately predict the impact of amino acid mutations on protein stability for problems in personalized medicine and other applications. Here the ability of 10 computational tools to accurately predict mutation-induced perturbation of folding stability (ΔΔG) for membrane proteins of known structure was assessed. All methods for predicting ΔΔG values performed significantly worse when applied to membrane proteins than when applied to soluble proteins, yielding estimated concordance, Pearson, and Spearman correlation coefficients of <0.4 for membrane proteins. Rosetta and PROVEAN showed a modest ability to classify mutations as destabilizing (ΔΔG < -0.5 kcal/mol), with a 7 in 10 chance of correctly discriminating a randomly chosen destabilizing variant from a randomly chosen stabilizing variant. However, even this performance is significantly worse than for soluble proteins. This study highlights the need for further development of reliable and reproducible methods for predicting thermodynamic folding stability in membrane proteins.

Simultaneous control of error rates in fMRI data analysis.

The key idea of statistical hypothesis testing is to fix, and thereby control, the Type I error (false positive) rate across samples of any size. Multiple comparisons inflate the global (family-wise) Type I error rate and the traditional solution to maintaining control of the error rate is to increase the local (comparison-wise) Type II error (false negative) rates. However, in the analysis of human brain imaging data, the number of comparisons is so large that this solution breaks down: the local Type II error rate ends up being so large that scientifically meaningful analysis is precluded. Here we propose a novel solution to this problem: allow the Type I error rate to converge to zero along with the Type II error rate. It works because when the Type I error rate per comparison is very small, the accumulation (or global) Type I error rate is also small. This solution is achieved by employing the likelihood paradigm, which uses likelihood ratios to measure the strength of evidence on a voxel-by-voxel basis. In this paper, we provide theoretical and empirical justification for a likelihood approach to the analysis of human brain imaging data. In addition, we present extensive simulations that show the likelihood approach is viable, leading to "cleaner"-looking brain maps and operational superiority (lower average error rate). Finally, we include a case study on cognitive control related activation in the prefrontal cortex of the human brain.

The fellowship effect: how the establishment of a fellowship in female pelvic medicine and reconstructive surgery affected resident vaginal hysterectomy training.

OBJECTIVE: We report on trends in resident-performed vaginal hysterectomies before and after the establishment of a female pelvic medicine and reconstructive surgery fellowship at Vanderbilt University Medical Center. STUDY DESIGN: We examined medical records and resident self-reports concerning all hysterectomies at our institution in an 8-year period: 4 years before fellowship and 4 years after. Route of hysterectomy, resident and fellow involvement, and division of attending surgeon were recorded from the electronic medical record. Resident Accreditation Council for Graduate Medical Education (ACGME) case log data were used to estimate the number of hysterectomies where residents reported themselves as the primary surgeon. RESULTS: During the 8-year period of this study, 3317 hysterectomies were performed at our institution, 41% (1371) before and 59% (1946) after fellowship. Prior to fellowship, 29% (393) were vaginal, 56% (766) were abdominal, and 15% (212) were laparoscopic/robotic. After addition of fellowship, 23% (449) were vaginal, 31% (597) were abdominal, and 46% (900) were laparoscopic/robotic. Of the total vaginal hysterectomies (TVH), there was resident involvement in 98.0% (385) cases before fellowship and 98.2% (441) cases after fellowship. From the ACGME case log data, the resident identified himself/herself as the primary surgeon in 388 cases before and 393 cases after fellowship. During this time period, medical records indicate a fellow was involved in 42% (189) of TVH, with resident involvement in all but 5 of these procedures. CONCLUSION: Frequency of resident involvement in TVH cases, either as primary surgeon or team member, remained constant after the addition of the female pelvic medicine and reconstructive surgery fellowship.

Accuracy of FDG-PET to diagnose lung cancer in areas with infectious lung disease: a meta-analysis.

IMPORTANCE: Positron emission tomography (PET) combined with fludeoxyglucose F 18 (FDG) is recommended for the noninvasive diagnosis of pulmonary nodules suspicious for lung cancer. In populations with endemic infectious lung disease, FDG-PET may not accurately identify malignant lesions. OBJECTIVES: To estimate the diagnostic accuracy of FDG-PET for pulmonary nodules suspicious for lung cancer in regions where infectious lung disease is endemic and compare the test accuracy in regions where infectious lung disease is rare. DATA SOURCES AND STUDY SELECTION: Databases of MEDLINE, EMBASE, and the Web of Science were searched from October 1, 2000, through April 28, 2014. Articles reporting information sufficient to calculate sensitivity and specificity of FDG-PET to diagnose lung cancer were included. Only studies that enrolled more than 10 participants with benign and malignant lesions were included. Database searches yielded 1923 articles, of which 257 were assessed for eligibility. Seventy studies were included in the analysis. Studies reported on a total of 8511 nodules; 5105 (60%) were malignant. DATA EXTRACTION AND SYNTHESIS: Abstracts meeting eligibility criteria were collected by a research librarian and reviewed by 2 independent reviewers. Hierarchical summary receiver operating characteristic curves were constructed. A random-effects logistic regression model was used to summarize and assess the effect of endemic infectious lung disease on test performance. MAIN OUTCOME AND MEASURES: The sensitivity and specificity for FDG-PET test performance. RESULTS: Heterogeneity for sensitivity (I2 = 87%) and specificity (I2 = 82%) was observed across studies. The pooled (unadjusted) sensitivity was 89% (95% CI, 86%-91%) and specificity was 75% (95% CI, 71%-79%). There was a 16% lower average adjusted specificity in regions with endemic infectious lung disease (61% [95% CI, 49%-72%]) compared with nonendemic regions (77% [95% CI, 73%-80%]). Lower specificity was observed when the analysis was limited to rigorously conducted and well-controlled studies. In general, sensitivity did not change appreciably by endemic infection status, even after adjusting for relevant factors. CONCLUSIONS AND RELEVANCE: The accuracy of FDG-PET for diagnosing lung nodules was extremely heterogeneous. Use of FDG-PET combined with computed tomography was less specific in diagnosing malignancy in populations with endemic infectious lung disease compared with nonendemic regions. These data do not support the use of FDG-PET to diagnose lung cancer in endemic regions unless an institution achieves test performance accuracy similar to that found in nonendemic regions.

Accuracy and interpretation time of computer-aided detection among novice and experienced breast MRI readers.

OBJECTIVE: The purpose of this study was to compare the diagnostic accuracy and interpretation times of breast MRI with and without use of a computer-aided detection (CAD) system by novice and experienced readers. SUBJECTS AND METHODS: A reader study was undertaken with 20 radiologists, nine experienced and 11 novice. Each radiologist participated in two reading sessions spaced 6 months apart that consisted of 70 cases (27 benign, 43 malignant), read with and without CAD assistance. Sensitivity, specificity, negative predictive value, positive predictive value, and overall accuracy as measured by the area under the receiver operating characteristic curve (AUC) were reported for each radiologist. Accuracy comparisons across use of CAD and experience level were examined. Time to interpret and report on each case was recorded. RESULTS: CAD improved sensitivity for both experienced (AUC, 0.91 vs 0.84; 95% CI on the difference, 0.04, 0.11) and novice readers (AUC, 0.83 vs 0.77; 95% CI on the difference, 0.01, 0.10). The increase in sensitivity was statistically higher for experienced readers (p = 0.01). Diagnostic accuracy, measured by AUC, for novices without CAD was 0.77, for novices with CAD was 0.79, for experienced readers without CAD was 0.80, and for experienced readers with CAD was 0.83. An upward trend was noticed, but the differences were not statistically significant. There were no significant differences in interpretation times. CONCLUSION: MRI sensitivity improved with CAD for both experienced readers and novices with no overall increase in time to evaluate cases. However, overall accuracy was not significantly improved. As the use of breast MRI with CAD increases, more attention to the potential contributions of CAD to the diagnostic accuracy of MRI is needed.

Racial Disparities in Lung Cancer Stage of Diagnosis Among Adults Living in the Southeastern United States.

BACKGROUND: Black Americans receive a diagnosis at later stage of lung cancer more often than White Americans. We undertook a population-based study to identify factors contributing to racial disparities in lung cancer stage of diagnosis among low-income adults. RESEARCH QUESTION: Which multilevel factors contribute to racial disparities in stage of lung cancer at diagnosis? STUDY DESIGN AND METHODS: Cases of incident lung cancer from the prospective observational Southern Community Cohort Study were identified by linkage with state cancer registries in 12 southeastern states. Logistic regression shrinkage techniques were implemented to identify individual-level and area-level factors associated with distant stage diagnosis. A subset of participants who responded to psychosocial questions (eg, racial discrimination experiences) were evaluated to determine if model predictive power improved. RESULTS: We identified 1,572 patients with incident lung cancer with available lung cancer stage (64% self-identified as Black and 36% self-identified as White). Overall, Black participants with lung cancer showed greater unadjusted odds of distant stage diagnosis compared with White participants (OR,1.29; 95% CI, 1.05-1.59). Greater neighborhood area deprivation was associated with distant stage diagnosis (OR, 1.58; 95% CI, 1.19-2.11). After controlling for individual- and area-level factors, no significant difference were found in distant stage disease for Black vs White participants. However, participants with COPD showed lower odds of distant stage diagnosis in the primary model (OR, 0.72; 95% CI, 0.53-0.98). Interesting and complex interactions were observed. The subset analysis model with additional variables for racial discrimination experiences showed slightly greater predictive power than the primary model. INTERPRETATION: Reducing racial disparities in lung cancer stage at presentation will require interventions on both structural and individual-level factors.

Covariate-specific ROC curve analysis can accommodate differences between covariate subgroups in the evaluation of diagnostic accuracy.

OBJECTIVES: We present an illustrative application of methods that account for covariates in receiver operating characteristic (ROC) curve analysis, using individual patient data on D-dimer testing for excluding pulmonary embolism. STUDY DESIGN AND SETTING: Bayesian nonparametric covariate-specific ROC curves were constructed to examine the performance/positivity thresholds in covariate subgroups. Standard ROC curves were constructed. Three scenarios were outlined based on comparison between subgroups and standard ROC curve conclusion: (1) identical distribution/identical performance, (2) different distribution/identical performance, and (3) different distribution/different performance. Scenarios were illustrated using clinical covariates. Covariate-adjusted ROC curves were also constructed. RESULTS: Age groups had prominent differences in D-dimer concentration, paired with differences in performance (Scenario 3). Different positivity thresholds were required to achieve the same level of sensitivity. D-dimer had identical performance, but different distributions for YEARS algorithm items (Scenario 2), and similar distributions for sex (Scenario 1). For the later covariates, comparable positivity thresholds achieved the same sensitivity. All covariate-adjusted models had AUCs comparable to the standard approach. CONCLUSION: Subgroup differences in performance and distribution of results can indicate that the conventional ROC curve is not a fair representation of test performance. Estimating conditional ROC curves can improve the ability to select thresholds with greater applicability.

Evaluation of resampling-based inference for topological features of neuroimages.

Many recent studies have demonstrated the inflated type 1 error rate of the original Gaussian random field (GRF) methods for inference of neuroimages and identified resampling (permutation and bootstrapping) methods that have better performance. There has been no evaluation of resampling procedures when using robust (sandwich) statistical images with different topological features (TF) used for neuroimaging inference. Here, we consider estimation of distributions TFs of a statistical image and evaluate resampling procedures that can be used when exchangeability is violated. We compare the methods using realistic simulations and study sex differences in life-span age-related changes in gray matter volume in the Nathan Kline Institute Rockland sample. We find that our proposed wild bootstrap and the commonly used permutation procedure perform well in sample sizes above 50 under realistic simulations with heteroskedasticity. The Rademacher wild bootstrap has fewer assumptions than the permutation and performs similarly in samples of 100 or more, so is valid in a broader range of conditions. We also evaluate the GRF-based pTFCE method and show that it has inflated error rates in samples less than 200. Our R package, pbj , is available on Github and allows the user to reproducibly implement various resampling-based group level neuroimage analyses.

The Thoracic Research Evaluation and Treatment 2.0 Model: A Lung Cancer Prediction Model for Indeterminate Nodules Referred for Specialist Evaluation.

BACKGROUND: Appropriate risk stratification of indeterminate pulmonary nodules (IPNs) is necessary to direct diagnostic evaluation. Currently available models were developed in populations with lower cancer prevalence than that seen in thoracic surgery and pulmonology clinics and usually do not allow for missing data. We updated and expanded the Thoracic Research Evaluation and Treatment (TREAT) model into a more generalized, robust approach for lung cancer prediction in patients referred for specialty evaluation. RESEARCH QUESTION: Can clinic-level differences in nodule evaluation be incorporated to improve lung cancer prediction accuracy in patients seeking immediate specialty evaluation compared with currently available models? STUDY DESIGN AND METHODS: Clinical and radiographic data on patients with IPNs from six sites (N = 1,401) were collected retrospectively and divided into groups by clinical setting: pulmonary nodule clinic (n = 374; cancer prevalence, 42%), outpatient thoracic surgery clinic (n = 553; cancer prevalence, 73%), or inpatient surgical resection (n = 474; cancer prevalence, 90%). A new prediction model was developed using a missing data-driven pattern submodel approach. Discrimination and calibration were estimated with cross-validation and were compared with the original TREAT, Mayo Clinic, Herder, and Brock models. Reclassification was assessed with bias-corrected clinical net reclassification index and reclassification plots. RESULTS: Two-thirds of patients had missing data; nodule growth and fluorodeoxyglucose-PET scan avidity were missing most frequently. The TREAT version 2.0 mean area under the receiver operating characteristic curve across missingness patterns was 0.85 compared with that of the original TREAT (0.80), Herder (0.73), Mayo Clinic (0.72), and Brock (0.68) models with improved calibration. The bias-corrected clinical net reclassification index was 0.23. INTERPRETATION: The TREAT 2.0 model is more accurate and better calibrated for predicting lung cancer in high-risk IPNs than the Mayo, Herder, or Brock models. Nodule calculators such as TREAT 2.0 that account for varied lung cancer prevalence and that consider missing data may provide more accurate risk stratification for patients seeking evaluation at specialty nodule evaluation clinics.

Locally advanced breast cancer: MR imaging for prediction of response to neoadjuvant chemotherapy--results from ACRIN 6657/I-SPY TRIAL.

PURPOSE: To compare magnetic resonance (MR) imaging findings and clinical assessment for prediction of pathologic response to neoadjuvant chemotherapy (NACT) in patients with stage II or III breast cancer. MATERIALS AND METHODS: The HIPAA-compliant protocol and the informed consent process were approved by the American College of Radiology Institutional Review Board and local-site institutional review boards. Women with invasive breast cancer of 3 cm or greater undergoing NACT with an anthracycline-based regimen, with or without a taxane, were enrolled between May 2002 and March 2006. MR imaging was performed before NACT (first examination), after one cycle of anthracyline-based treatment (second examination), between the anthracycline-based regimen and taxane (third examination), and after all chemotherapy and prior to surgery (fourth examination). MR imaging assessment included measurements of tumor longest diameter and volume and peak signal enhancement ratio. Clinical size was also recorded at each time point. Change in clinical and MR imaging predictor variables were compared for the ability to predict pathologic complete response (pCR) and residual cancer burden (RCB). Univariate and multivariate random-effects logistic regression models were used to characterize the ability of tumor response measurements to predict pathologic outcome, with area under the receiver operating characteristic curve (AUC) used as a summary statistic. RESULTS: Data in 216 women (age range, 26-68 years) with two or more imaging time points were analyzed. For prediction of both pCR and RCB, MR imaging size measurements were superior to clinical examination at all time points, with tumor volume change showing the greatest relative benefit at the second MR imaging examination. AUC differences between MR imaging volume and clinical size predictors at the early, mid-, and posttreatment time points, respectively, were 0.14, 0.09, and 0.02 for prediction of pCR and 0.09, 0.07, and 0.05 for prediction of RCB. In multivariate analysis, the AUC for predicting pCR at the second imaging examination increased from 0.70 for volume alone to 0.73 when all four predictor variables were used. Additional predictive value was gained with adjustments for age and race. CONCLUSION: MR imaging findings are a stronger predictor of pathologic response to NACT than clinical assessment, with the greatest advantage observed with the use of volumetric measurement of tumor response early in treatment.