Meta-analysis and systematic review of radiomics imaging for predicting pathological complete response to neoadjuvant therapy in esophageal squamous cell carcinoma.

Abstract

e16083 Background: Esophageal cancer remains a leading cause of cancer related mortality worldwide, with esophageal squamous cell carcinoma (ESCC) being the predominant subtype in high-incidence regions such as Asia. Neoadjuvant therapy(NAT) has emerged as an effective treatment strategy, with pathological complete response (pCR) serving as a critical marker of therapeutic success and improved survival. Accurate prediction of pCR is necessary for better clinical decision making. Radiomics, a novel imaging based approach, enables the extraction of quantitative features from medical images, offering a non-invasive method for predicting pCR. This meta-analysis and systematic review synthesizes current evidence to evaluate the predictive value of radiomics in ESCC patients undergoing neoadjuvant therapy. Methods: A systematic search was conducted using PubMed, Google Scholar, and Scopus. PRISMA guidelines were followed. A boolean expression was constructed to retrieve and select articles from major medical databases. The studies which involved assessment of pathological response following NAT by Radiomic Imaging methods were considered for the final analysis. The R Studio package was used to evaluate the potential of the diagnostic test. The Pooled Sensitivity, Specificity, Accuracy and Area Under the Curve (AUC)were estimated to predict the diagnostic potential of Radiomic Imaging in the prediction of the pathological response of NAT in ESCC. The random effects model via the linear (mixed-effects) model framework was considered for statistical analysis. Results: A total of 15 studies with 77 images assessed through various machine learning algorithms were assessed through this meta-analysis. The Pooled Sensitivity and Specificity were estimated to be 69.70% (63.25; 75.21, 95% CI, p < 0.0001, z = 22.918, SE = 0.03) and 67.65% ([61.72; 73.57], 95% CI, p < 0.0001, z = 22.33, SE = 0.03) utilising the random effects model. The pooled area under the curve was estimated to be 0.7381(0.7097;0.7665, 95% CI, p < 0.0001, z = 50.933, SE = 0.0145). Conclusions: This meta-analysis establishes radiomics based imaging as a promising non-invasive predictor of pathological complete response (pCR) in esophageal squamous cell carcinoma following neoadjuvant therapy. Further standardization and validation are essential for clinical implementation.