MRI Radiomics Predicts Pathological Complete Response to Neoadjuvant Chemoradiotherapy in Rectal Cancer: Meta-Analysis

MRI radiomics predicts pathological complete response (pCR) in rectal cancer after neoadjuvant chemoradiotherapy: pooled sensitivity 0.82, specificity 0.86, AUC 0.85 (38-study meta-analysis). Deep learning models outperformed. Prospective validation still lacking.

• Systematic review and meta-analysis of 38 studies using validation cohorts only; bivariate random-effects model yielded pooled PLR 6.0 (95% CI 4.0–8.9) and NLR 0.21 (95% CI 0.12–0.35), corresponding to a diagnostic odds ratio of 29 (95% CI 14–61).
• Subgroup analyses suggested incremental gains from deep learning approaches and combined clinical-radiomic models over standard radiomics alone (exact subgroup figures not reported in source).
• Key limitations: significant methodological heterogeneity across studies, predominantly retrospective single-center designs, and absence of prospective external validation — all noted by the authors as barriers to clinical translation.

RadPigeon summaries are original and for information only. They are not clinical advice.