Prediction model for severe maternal morbidity in pregnant women with hypertensive disorders of pregnancy.

Objectives: To investigate risk factors for severe maternal morbidity (SMM) in pregnant women with hypertensive disorders of pregnancy (HDP) and to develop a risk prediction model. Methods: A prospective observational cohort study was conducted among pregnant women who were hospitalized for hypertensive disorders of pregnancy (HDP) between January 2016 and December 2020 in Fujian College of Clinical Medicine for Obstetrics & Gynecology and Pediatrics, Fujian Province, China (a training set), and a risk predictive model was constructed. Pregnant women with HDP who were hospitalized between January 2021 and December 2021 were selected as a validation set. Concordance index (C-index) and calibration curves were used to test predictive model discrimination and calibration. Results: We included 970 pregnant women (790 in the training set and 180 in the validation set). Least absolute shrinkage and selection operator regression was used to screen for nine related variables such as intra-uterine growth retardation (IUGR), diastolic blood pressure (DBP) and systolic blood pressure (SBP) at suspected diagnosis, total bilirubin, albumin (ALB), uric acid, total cholesterol, serum magnesium, and suspected gestational age. SBP at suspected diagnosis (OR =1.22, 95%CI:1.08-1.42) and total cholesterol (OR = 1.78, 95%CI:1.17-2.80) were independent risk factors of severe maternal morbidity in pregnant women with HDP. A nomogram was constructed, and internal validation of the nomogram model was done using the bootstrap self-sampling method. C-index in the training and the validation set was 0.798 and 0.909, respectively. Conclusion: Our prediction model can be used to determine gestational hypertension severity in pregnant women.

Fast Lasso method for large-scale and ultrahigh-dimensional Cox model with applications to UK Biobank.

We develop a scalable and highly efficient algorithm to fit a Cox proportional hazard model by maximizing the $L^1$-regularized (Lasso) partial likelihood function, based on the Batch Screening Iterative Lasso (BASIL) method developed in Qian and others (2019). Our algorithm is particularly suitable for large-scale and high-dimensional data that do not fit in the memory. The output of our algorithm is the full Lasso path, the parameter estimates at all predefined regularization parameters, as well as their validation accuracy measured using the concordance index (C-index) or the validation deviance. To demonstrate the effectiveness of our algorithm, we analyze a large genotype-survival time dataset across 306 disease outcomes from the UK Biobank (Sudlow and others, 2015). We provide a publicly available implementation of the proposed approach for genetics data on top of the PLINK2 package and name it snpnet-Cox.

Pitfalls of the concordance index for survival outcomes.

Prognostic models are useful tools for assessing a patient's risk of experiencing adverse health events. In practice, these models must be validated before implementation to ensure that they are clinically useful. The concordance index (C-Index) is a popular statistic that is used for model validation, and it is often applied to models with binary or survival outcome variables. In this paper, we summarize existing criticism of the C-Index and show that many limitations are accentuated when applied to survival outcomes, and to continuous outcomes more generally. We present several examples that show the challenges in achieving high concordance with survival outcomes, and we argue that the C-Index is often not clinically meaningful in this setting. We derive a relationship between the concordance probability and the coefficient of determination under an ordinary least squares model with normally distributed predictors, which highlights the limitations of the C-Index for continuous outcomes. Finally, we recommend existing alternatives that more closely align with common uses of survival models.

New C-indices for assessing importance of longitudinal biomarkers in fitting competing risks survival data in the presence of partially masked causes.

Competing risks survival data in the presence of partially masked causes are frequently encountered in medical research or clinical trials. When longitudinal biomarkers are also available, it is of great clinical importance to examine associations between the longitudinal biomarkers and the cause-specific survival outcomes. In this article, we propose a cause-specific C-index for joint models of longitudinal and competing risks survival data accounting for masked causes. We also develop a posterior predictive algorithm for computing the out-of-sample cause-specific C-index using Markov chain Monte Carlo samples from the joint posterior of the in-sample longitudinal and competing risks survival data. We further construct the Δ $$ \Delta $$ C-index to quantify the strength of association between the longitudinal and cause-specific survival data, or between the out-of-sample longitudinal and survival data. Empirical performance of the proposed assessment criteria is examined through an extensive simulation study. An in-depth analysis of the real data from large cancer prevention trials is carried out to demonstrate the usefulness of the proposed methodology.

Bayesian criterion-based assessments of recurrent event models with applications to commercial truck driver behavior studies.

Multitype recurrent events are commonly observed in transportation studies, since commercial truck drivers may encounter different types of safety critical events (SCEs) and take different lengths of on-duty breaks in a driving shift. Bayesian nonhomogeneous Poisson process models are a flexible approach to jointly model the intensity functions of the multitype recurrent events. For evaluating and comparing these models, the deviance information criterion (DIC) and the logarithm of the pseudo-marginal likelihood (LPML) are studied and Monte Carlo methods are developed for computing these model assessment measures. We also propose a set of new concordance indices (C-indices) to evaluate various discrimination abilities of a Bayesian multitype recurrent event model. Specifically, the within-event C-index quantifies adequacy of a given model in fitting the recurrent event data for each type, the between-event C-index provides an assessment of the model fit between two types of recurrent events, and the overall C-index measures the model's discrimination ability among multiple types of recurrent events simultaneously. Moreover, we jointly model the incidence of SCEs and on-duty breaks with driving behaviors using a Bayesian Poisson process model with time-varying coefficients and time-dependent covariates. An in-depth analysis of a real dataset from the commercial truck driver naturalistic driving study is carried out to demonstrate the usefulness and applicability of the proposed methodology.

Radiomics model for preoperative prediction of 3-year survival-based CT image biomarkers in esophageal cancer.

OBJECTIVE: This work aimed to develop a radiomics nomogram to predict 3-year overall survival of esophageal cancer patients after chemoradiotherapy. METHODS: A total of 109 esophageal cancer patients, diagnosed from November 2012 to February 2015, were enrolled in this retrospective study. They were randomly divided into training set (77 cases) and verification set (32 cases). Image standardization was performed prior to feature extraction. And then, about 1670 radiomics features were extracted from the pretreatment diagnostic computed tomography image. A radiomics signature was constructed with the lasso algorithm; then, a radiomics score was calculated to reflect survival probability using the radiomics signature for each patient. A radiomics nomogram was developed by incorporating the radiomics score and clinical factors. A clinical model was constructed using clinical factors only. The performance of the nomogram was assessed with respect to its calibration and discrimination. Kaplan-Meier survival analysis was performed. RESULTS: Sixteen radiomics features were selected to build the radiomics signature. The radiomics nomogram showed better calibration and classification capacity than the clinical model with AUC 0.96 vs. 0.72 for the training cohort, and 0.87 vs. 0.67 for the validation cohort. The model showed good discrimination with a Harrell's Concordance Index of 0.76 in the training cohort and 0.81 in the validation cohort. Decision curve analysis demonstrated the clinical usefulness of the radiomics nomogram. A significant difference (p value < 0.05; log-rank test) was observed between the survival curves of the nomogram-predicted survival and non-survival groups. CONCLUSIONS: The present study proposed a radiomics-based nomogram involving the radiomics signature and clinical factors. It can be potentially applied in the individual preoperative prediction of 3-year survival in esophageal cancer patients.

External validation of the albumin, C-reactive protein and lactate dehydrogenase model in patients with metastatic renal cell carcinoma receiving second-line axitinib therapy in a Japanese multi-center cohort.

PURPOSE: To externally validate the utility of the albumin, C-reactive protein and lactate dehydrogenase model to predict the overall survival of previously treated metastatic renal cell carcinoma patients. PATIENTS AND METHODS: The ability of the albumin, C-reactive protein and lactate dehydrogenase model to predict overall survival was validated and compared with those of other prognostication models using data from 421 metastatic renal cell carcinoma patients receiving second-line axitinib therapy at 36 hospitals belonging to the Japan Urologic Oncology Group. RESULTS: The following factors in this cohort were independently associated with poor overall survival in a multivariate analysis: a low Karnofsky performance status, <1 year from diagnosis to targeted therapy, a high neutrophil count, and low albumin, elevated C-reactive protein, and elevated lactate dehydrogenase, and the Japan Urologic Oncology Group model was newly developed based on the presence/absence of these independent factors. In this cohort, 151 (35.9%), 125 (27.7%) and 145 (34.4%) patients were classified into the favorable, intermediate and poor risk groups, respectively, according to the albumin, C-reactive protein and lactate dehydrogenase model; however, the proportions of patients in the intermediate risk group stratified by the Japan Urologic Oncology Group, Memorial Sloan Kettering Cancer Center and International Metastatic Renal Cell Carcinoma Database Consortium models were >50%. The superiority of the albumin, C-reactive protein and lactate dehydrogenase model to the Memorial Sloan Kettering Cancer Center and International Metastatic Renal Cell Carcinoma Database Consortium models, but not the Japan Urologic Oncology Group model, was demonstrated by multiple statistical analyses. CONCLUSIONS: The utility of the albumin, C-reactive protein and lactate dehydrogenase model as a simple and objective prognostication tool was successfully validated using data from 421 metastatic renal cell carcinoma patients receiving second-line axitinib.

The c-index is not proper for the evaluation of $t$-year predicted risks.

We show that the widely used concordance index for time to event outcome is not proper when interest is in predicting a $t$-year risk of an event, for example 10-year mortality. In the situation with a fixed prediction horizon, the concordance index can be higher for a misspecified model than for a correctly specified model. Impropriety happens because the concordance index assesses the order of the event times and not the order of the event status at the prediction horizon. The time-dependent area under the receiver operating characteristic curve does not have this problem and is proper in this context.

ROC-guided survival trees and ensembles.

Tree-based methods are popular nonparametric tools in studying time-to-event outcomes. In this article, we introduce a novel framework for survival trees and ensembles, where the trees partition the dynamic survivor population and can handle time-dependent covariates. Using the idea of randomized tests, we develop generalized time-dependent receiver operating characteristic (ROC) curves for evaluating the performance of survival trees. The tree-building algorithm is guided by decision-theoretic criteria based on ROC, targeting specifically for prediction accuracy. To address the instability issue of a single tree, we propose a novel ensemble procedure based on averaging martingale estimating equations, which is different from existing methods that average the predicted survival or cumulative hazard functions from individual trees. Extensive simulation studies are conducted to examine the performance of the proposed methods. We apply the methods to a study on AIDS for illustration.

A numerical strategy to evaluate performance of predictive scores via a copula-based approach.

Assessing and comparing the performance of correlated predictive scores are of current interest in precision medicine. Given the limitations of available theoretical approaches for assessing and comparing the predictive accuracy, numerical methods are highly desired which, however, have not been systematically developed due to technical challenges. The main challenges include the lack of a general strategy on effectively simulating many kinds of correlated predictive scores each with some given level of predictive accuracy in either concordance index or the area under a receiver operating characteristic curve area under the curves (AUC). To fill in this important knowledge gap, this paper is to provide a general copula-based numeric framework for assessing and comparing predictive performance of correlated predictive or risk scores. The new algorithms are designed to effectively simulate correlated predictive scores with given levels of predictive accuracy as measured in terms of concordance indices or time-dependent AUC for predicting survival outcomes. The copula-based numerical strategy is convenient for numerically evaluating and comparing multiple measures of predictive accuracy of correlated risk scores and for investigating finite-sample properties of test statistics and confidence intervals as well as assessing for optimism of given performance measures using cross-validation or bootstrap.

A practical perspective on the concordance index for the evaluation and selection of prognostic time-to-event models.

Developing a prognostic model for biomedical applications typically requires mapping an individual's set of covariates to a measure of the risk that he or she may experience the event to be predicted. Many scenarios, however, especially those involving adverse pathological outcomes, are better described by explicitly accounting for the timing of these events, as well as their probability. As a result, in these cases, traditional classification or ranking metrics may be inadequate to inform model evaluation or selection. To address this limitation, it is common practice to reframe the problem in the context of survival analysis, and resort, instead, to the concordance index (C-index), which summarises how well a predicted risk score describes an observed sequence of events. A practically meaningful interpretation of the C-index, however, may present several difficulties and pitfalls. Specifically, we identify two main issues: i) the C-index remains implicitly, and subtly, dependent on time, and ii) its relationship with the number of subjects whose risk was incorrectly predicted is not straightforward. Failure to consider these two aspects may introduce undesirable and unwanted biases in the evaluation process, and even result in the selection of a suboptimal model. Hence, here, we discuss ways to obtain a meaningful interpretation in spite of these difficulties. Aiming to assist experimenters regardless of their familiarity with the C-index, we start from an introductory-level presentation of its most popular estimator, highlighting the latter's temporal dependency, and suggesting how it might be correctly used to inform model selection. We also address the nonlinearity of the C-index with respect to the number of correct risk predictions, elaborating a simplified framework that may enable an easier interpretation and quantification of C-index improvements or deteriorations.

Boosting the discriminatory power of sparse survival models via optimization of the concordance index and stability selection.

BACKGROUND: When constructing new biomarker or gene signature scores for time-to-event outcomes, the underlying aims are to develop a discrimination model that helps to predict whether patients have a poor or good prognosis and to identify the most influential variables for this task. In practice, this is often done fitting Cox models. Those are, however, not necessarily optimal with respect to the resulting discriminatory power and are based on restrictive assumptions. We present a combined approach to automatically select and fit sparse discrimination models for potentially high-dimensional survival data based on boosting a smooth version of the concordance index (C-index). Due to this objective function, the resulting prediction models are optimal with respect to their ability to discriminate between patients with longer and shorter survival times. The gradient boosting algorithm is combined with the stability selection approach to enhance and control its variable selection properties. RESULTS: The resulting algorithm fits prediction models based on the rankings of the survival times and automatically selects only the most stable predictors. The performance of the approach, which works best for small numbers of informative predictors, is demonstrated in a large scale simulation study: C-index boosting in combination with stability selection is able to identify a small subset of informative predictors from a much larger set of non-informative ones while controlling the per-family error rate. In an application to discover biomarkers for breast cancer patients based on gene expression data, stability selection yielded sparser models and the resulting discriminatory power was higher than with lasso penalized Cox regression models. CONCLUSION: The combination of stability selection and C-index boosting can be used to select small numbers of informative biomarkers and to derive new prediction rules that are optimal with respect to their discriminatory power. Stability selection controls the per-family error rate which makes the new approach also appealing from an inferential point of view, as it provides an alternative to classical hypothesis tests for single predictor effects. Due to the shrinkage and variable selection properties of statistical boosting algorithms, the latter tests are typically unfeasible for prediction models fitted by boosting.

Comparison of two ordinal prediction models: A cancer staging system example.

BACKGROUND: Choosing to replace or maintain an existing cancer staging system is a difficult task. The system plays a critical role in patient counselling and treatment decision making because the staging system conveys prognosis. METHODS: Many issues may be considered when deciding the preferred system (i.e. old or new), such as the level of evidence for one or more factors included in the system or the general opinions of expert clinicians. However, given the major objective of estimating prognosis on an ordinal scale, we argue that the rival staging system candidates should be compared on their ability to predict outcome. We sought to outline an algorithm that would compare two rival ordinal systems on their predictive ability. RESULTS: We devised an algorithm based largely on the concordance index, which is appropriate for comparing two models in their ability to rank observations. We demonstrate our algorithm with a prostate cancer staging system example. CONCLUSION: We have provided an algorithm for selecting the preferred staging system based on prognostic accuracy. It appears to be useful for the purpose of selecting between two ordinal prediction models.

A gene expression signature of epithelial tubulogenesis and a role for ASPM in pancreatic tumor progression.

BACKGROUND & AIMS: Many patients with pancreatic ductal adenocarcinoma (PDAC) develop recurrent or metastatic diseases after surgery, so it is important to identify those most likely to benefit from aggressive therapy. Disruption of tissue microarchitecture is an early step in pancreatic tumorigenesis and a parameter used in pathology grading of glandular tumors. We investigated whether changes in gene expression during pancreatic epithelial morphogenesis were associated with outcomes of patients with PDAC after surgery. METHODS: We generated architectures of human pancreatic duct epithelial cells in a 3-dimensional basement membrane matrix. We identified gene expression profiles of the cells during different stages of tubular morphogenesis (tubulogenesis) and of PANC-1 cells during spheroid formation. Differential expression of genes was confirmed by immunoblot analysis. We compared the gene expression profile associated with pancreatic epithelial tubulogenesis with that of PDAC samples from 27 patients, as well as with their outcomes after surgery. RESULTS: We identified a gene expression profile associated with tubulogenesis that resembled the profile of human pancreatic tissue with differentiated morphology and exocrine function. Patients with PDACs with this profile fared well after surgery. Based on this profile, we established a 6-28 gene tubulogenesis-specific signature that accurately determined the prognosis of independent cohorts of patients with PDAC (total n = 128; accuracy = 81.2%-95.0%). One gene, ASPM, was down-regulated during tubulogenesis but up-regulated in human PDAC cell lines and tumor samples; up-regulation correlated with patient outcomes (Cox regression P = .0028). Bioinformatic, genetic, biochemical, functional, and clinical correlative studies showed that ASPM promotes aggressiveness of PDAC by maintaining Wnt-ß-catenin signaling and stem cell features of PDAC cells. CONCLUSIONS: We identified a gene expression profile associated with pancreatic epithelial tubulogenesis and a tissue architecture-specific signature of PDAC cells that is associated with patient outcomes after surgery.

Predicting asthma in preschool children with asthma-like symptoms: validating and updating the PIAMA risk score.

BACKGROUND: The Prevention and Incidence of Asthma and Mite Allergy (PIAMA) risk score predicts the probability of having asthma at school age among preschool children with suggestive symptoms. OBJECTIVE: We sought to externally validate the PIAMA risk score at different ages and in ethnic and socioeconomic subgroups of children in addition to updating it. METHODS: We studied 2877 children with preschool asthma-like symptoms participating in the multiethnic, prospective, population-based cohort study Generation R. The PIAMA risk score was assessed at preschool age, and asthma was predicted at age 6 years. Discrimination (concordance index [C-index]) and calibration were calculated. The PIAMA risk score was updated, and its performance was similarly analyzed. RESULTS: At age 6 years, 6% (168/2877) of the children had asthma. The discriminative ability of the original PIAMA risk score to predict asthma in Generation R was similar compared with that in the PIAMA cohort (C-index = 0.74 vs 0.71). The predicted risks by using the original PIAMA risk score for having asthma at the age of 6 years tended to be slightly higher than the observed risks (8% vs 6%). No differences in discriminative ability were found at different ages or in ethnic and socioeconomic subgroups (P > .05). The updated PIAMA risk score had a C-index of 0.75. CONCLUSIONS: The PIAMA risk score showed good external validity. The discriminative ability was similar at different ages and in ethnic and socioeconomic subgroups of preschool children, which suggests good generalizability. Further studies are needed to reproduce the predictive performance of the updated PIAMA risk score in other populations and settings and to assess its clinical relevance.

Comparison between single-muscle evaluation and cross-sectional area muscle evaluation for predicting the prognosis in patients with oral squamous cell carcinoma: a retrospective cohort study.

Introduction: The most effective method of assessing sarcopenia has yet to be determined, whether by single muscle or by whole muscle segmentation. The purpose of this study was to compare the prognostic value of these two methods using computed tomography (CT) images in patients with oral squamous cell carcinoma (OSCC). Materials and methods: Sex- and age-adjusted Cox proportional hazards models were employed for each parameter of sarcopenia related to overall survival, disease-free survival, and disease-specific survival. Harrell's concordance index was calculated for each model to assess discriminatory power. Results: In this study including 165 patients, a significant correlation was found between the CT-based assessment of individual muscles and their cross-sectional area. Single muscle assessments showed slightly higher discriminatory power in survival outcomes compared to whole muscle assessments, but the difference was not statistically significant, as indicated by overlapping confidence intervals for the C-index between assessments. To further validate our measurements, we classified patients into two groups based on intramuscular adipose tissue content (P-IMAC) of the spinous process muscle. Analysis showed that the higher the P-IMAC value, the poorer the survival outcome. Conclusion: Our findings indicate a slight advantage of single-muscle over whole-muscle assessment in prognostic evaluation, but the difference between the two methods is not conclusive. Both assessment methods provide valuable prognostic information for patients with OSCC, and further studies involving larger, independent cohorts are needed to clarify the potential advantage of one method over the other in the prognostic assessment of sarcopenia in OSCC.

The Concordance Index decomposition: A measure for a deeper understanding of survival prediction models.

The Concordance Index (C-index) is a commonly used metric in Survival Analysis for evaluating the performance of a prediction model. In this paper, we propose a decomposition of the C-index into a weighted harmonic mean of two quantities: one for ranking observed events versus other observed events, and the other for ranking observed events versus censored cases. This decomposition enables a finer-grained analysis of the relative strengths and weaknesses between different survival prediction methods. The usefulness of this decomposition is demonstrated through benchmark comparisons against classical models and state-of-the-art methods, together with the new variational generative neural-network-based method (SurVED) proposed in this paper. The performance of the models is assessed using four publicly available datasets with varying levels of censoring. Using the C-index decomposition and synthetic censoring, the analysis shows that deep learning models utilize the observed events more effectively than other models. This allows them to keep a stable C-index in different censoring levels. In contrast to such deep learning methods, classical machine learning models deteriorate when the censoring level decreases due to their inability to improve on ranking the events versus other events.

Analyzing risk factors for second malignancies in early gastric carcinoma from the SEER database.

This retrospective study analyzed a large population of gastric cancer (GC) patients treated between 2010 and 2015 to investigate the clinical features and predictive risk factors for developing secondary primary malignancies (SPMs). The cumulative incidence of SPM was assessed using Kaplan-Meier analysis. Competing risk analyses adjusted for mortality were conducted using stratified Cox proportional hazard regression models and multivariate analyses to identify independent predictors of SPM. A total of 3289 out of 167,747 GC patients were included in the analytic cohort, with 155 patients diagnosed with SPM. Patients whose histologic type other than adenocarcinomas (AC) and signet ring cell carcinoma (SRCC) emerged as an independent risk factor for developing SPM (hazard ratio [HR] 2.262, 95% confidence interval [CI] 1.146-4.465, P = 0.019) in multivariate Cox regression analysis. The surgical method, including biopsy/local excision (HR 2.3, [CI] 1.291-4.095, P = 0.005) and subtotal/total resection ([HR] 1.947, [CI] 1.028-3.687, P = 0.041), chemotherapy ([HR] 1.527, [CI] 1.006-2.316, P = 0.047), and histologic type ([HR] 2.318, [CI] 1.193-4.504, P = 0.013)), were identified as independent risk factors in the competitive risk model. Subgroup analyses, stratified by chemotherapy, revealed an increased risk of SPM among older patients. Furthermore, a nomogram was developed and internally validated to predict the cumulative incidence of SPM in GC patients (C-index = 0.73 for 72 months). These findings suggested that in specific histologic types of GC, the lymph node infiltration region missed after local surgical resection, and concomitant chemotherapy would have an increased risk of SPM for cancer survivors.

Factors that predict short-term intensive care unit mortality in patients with cirrhosis.

BACKGROUND & AIMS: Despite advances in critical care medicine, the mortality rate is high among critically ill patients with cirrhosis. We aimed to identify factors that predict early (7 d) mortality among patients with cirrhosis admitted to the intensive care unit (ICU) and to develop a risk-stratification model. METHODS: We collected data from patients with cirrhosis admitted to the ICU at Indiana University (IU-ICU) from December 1, 2006, through December 31, 2009 (n = 185), or at the University of Pennsylvania (Penn-ICU) from May 1, 2005, through December 31, 2010 (n = 206). Factors associated with mortality within 7 days of admission (7-d mortality) were determined by logistic regression analyses. A model was constructed based on the predictive parameters available on the first day of ICU admission in the IU-ICU cohort and then validated in the Penn-ICU cohort. RESULTS: Median Model for End-stage Liver Disease (MELD) scores at ICU admission were 25 in the IU-ICU cohort (interquartile range, 23-34) and 32 in the Penn-ICU cohort (interquartile range, 26-41); corresponding 7-day mortalities were 28.3% and 53.6%, respectively. MELD score (odds ratio, 1.13; 95% confidence interval [CI], 1.07-1.2) and mechanical ventilation (odds ratio, 5.7; 95% CI, 2.3-14.1) were associated independently with 7-day mortality in the IU-ICU. A model based on these 2 variables separated IU-ICU patients into low-, medium-, and high-risk groups; these groups had 7-day mortalities of 9%, 27%, and 74%, respectively (concordance index, 0.80; 95% CI, 0.72-0.87; P < 10(-8)). The model was applied to the Penn-ICU cohort; the low-, medium-, and high-risk groups had 7-day mortalities of 33%, 56%, and 71%, respectively (concordance index, 0.67; 95% CI, 0.59-0.74; P < 10(-4)). CONCLUSIONS: A model based on MELD score and mechanical ventilation on day 1 can stratify risk of early mortality in patients with cirrhosis admitted to the ICU. More studies are needed to validate this model and to enhance its clinical utility.

Impact of clinical and histopathological parameters on disease specific survival in patients with collecting duct renal cell carcinoma: development of a disease specific risk model.

PURPOSE: Collecting duct renal cell carcinoma is a rare, aggressive histological subtype of renal cell carcinoma. Since few groups have evaluated the oncological prognosis in these patients based on clinical and pathological parameters, we assessed parameters prognostic for disease specific mortality. MATERIALS AND METHODS: From a cohort of 14,047 patients with renal cell carcinoma we retrieved the records of 95 with collecting duct renal cell carcinoma at a total of 16 European and American centers of the CORONA (Collaborative Research on Renal Neoplasms Association) and SATURN (Surveillance and Treatment Update Renal Neoplasms) projects, and another 2 centers. Multivariable Cox regression analysis was applied to determine the influence of parameters on disease specific mortality. Median followup was 48.1 months (IQR 24-103). RESULTS: The disease specific survival rate at 1, 2, 5 and 10 years was 60.4%, 47.3%, 40.3% and 32.8%, respectively. American Society of Anesthesiologists (ASA) score 3-4, tumor size greater than 7 cm, stage M1, Fuhrman grade 3-4 and lymphovascular invasion independently predicted disease specific mortality. Based on these parameters, patients were divided into 26 (27%) at low, 13 (14%) at intermediate and 56 (59%) at high risk with a 5-year disease specific survival rate of 96%, 62% and 8%, respectively (bootstrap corrected c-index 0.894, 95% CI 0.820-0.967, p <0.001). CONCLUSIONS: While patients with collecting duct renal cell carcinoma are commonly diagnosed at advanced stage and have poor prognosis after surgery, a subset has excellent survival. Histopathological features can help risk stratify patients based on the described, highly accurate risk model to predict disease specific mortality, facilitating patient counseling and risk based clinical decision making for adjuvant therapy and clinical trial inclusion.