Predictive model of the first failure pattern in patients receiving definitive chemoradiotherapy for inoperable locally advanced non-small cell lung cancer (LA-NSCLC)

Abstract

Purpose: To analyze patterns of failure in patients with LA-NSCLC who received definitive chemoradiotherapy (CRT) and to build a nomogram for predicting the failure patterns in this population of patients. Materials and methods: Clinicopathological data of patients with LA-NSCLC who received definitive chemoradiotherapy and follow-up between 2013 and 2016 in our hospital were collected. The endpoint was the first failure after definitive chemoradiotherapy. With using elastic net regression and 5-fold nested cross-validation, the optimal model with better generalization ability was selected. Based on the selected model and corresponding features, a nomogram prediction model was built. This model was also validated by ROC curves, calibration curve and decision curve analysis (DCA). Results: With a median follow-up of 28 months, 100 patients experienced failure. There were 46 and 54 patients who experience local failure and distant failure, respectively. Predictive model including 9 factors (smoking, pathology, location, EGFR mutation, age, tumor diameter, clinical N stage, consolidation chemotherapy and radiation dose) was finally built with the best performance. The average area under the ROC curve (AUC) with 5fold nested cross-validation was 0.719, which was better than any factors alone. The calibration curve revealed a satisfactory consistency between the predicted distant failure rates and the actual observations. DCA showed most of the threshold probabilities in this model were with good net benefits. Conclusion: Clinicopathological factors could collaboratively predict failure patterns in patients with LA-NSCLC who are receiving definitive chemoradiotherapy. A nomogram was built and validated based on these factors, showing a potential predictive value in clinical practice.

Multiplexed quantitative proteomics provides mechanistic cues for malaria severity and complexity

“…The elastic net regularized logistic regression method was used to classify: (1) dengue vs. malaria: HC vs. Dengue, HC vs. malaria, dengue vs. malaria; (2) type of malaria: HC vs. FM, HC vs. VM, FM vs. VM; (3) severity of FM: HC vs. NSFM, HC vs. SFM, NSFM vs. SFM; (4) severity of VM: HC vs. NSVM, HC vs. SVM, NSVM vs. SVM; (5) severity of dengue: HC vs. NSD, HC vs. SD, NSD vs. SD. Of note, the elastic net regularized logistic regression model offers more flexibility in two ways -(i) the l 1 norm helps attain parsimony in the sense that it optimally chooses the number of covariates (in the context of present dataset the covariates are proteins) by driving coefficients of unimportant covariates to zero and (ii) l 2 norm helps address the issue of multicollinearity [74][75][76][77][78][79][80][81][82] . k-fold nested cross-validation (k = 10 for malaria vs. dengue, k = 10 for falciparum vs. vivax, and k = 5 for cerebral vs. severe malaria anemia) was used as it provides robust and almost unbiased parameter estimates and model performance evaluation even for small sample sizes 83,84 .…”

Radiomic Phenotypes for Improving Early Prediction of Survival in Stage III Non-Small Cell Lung Cancer Adenocarcinoma after Chemoradiation

“…The development of accurate models to predict disease recurrence in patients with NSCLC treated with definitive chemoradiotherapy, also known as treatment failure, could play a fundamental role in the efficient personalization of treatment. In fact, previous investigators have utilized clinicopathological data to develop predictive models for treatment failures, demonstrating an ability to predict patterns of failure in NSCLC [22] and to identify factors associated with increased or decreased risks of treatment failure [23]. However, tumoral molecular characterization and genomic analysis require tumor biopsies that are invasive and can be associated with life-threatening complications [24].…”

Sampling error in the diagnosis of colorectal cancer is associated with delay to surgery: a retrospective cohort study

“…Predictors of sampling errors were analyzed via univariable and multivariable logistic regression. To avoid overfitting of the multivariable model due to the large number of variables and relatively low number of sampling errors, variables were selected using a regularized (elastic-net) logistic regression model, with elastic mixing and penalization terms estimated via repeated cross-validation performed using R software (R Core Team, 2020 version 4.0.3) with packages “caret” (Version 6.0–86, 2020) and “glmnet” (Version 4.1, 2020) [ 13 , 14 ]. Model fit was determined via area under the receiver operator curve (AUC).…”