Copyright: ©Author(s) 2026. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution-NonCommercial (CC BY-NC 4.0) license. No commercial re-use. See permissions. Published by Baishideng Publishing Group Inc.
Artif Intell Gastroenterol. Nov 8, 2026; 7(3): 125015
Published online Nov 8, 2026. doi: 10.35712/aig.125015
Published online Nov 8, 2026. doi: 10.35712/aig.125015
Predicting pathological complete response to neoadjuvant chemoradiotherapy in rectal cancer via artificial intelligence: A systematic review and meta-analysis
Felix Pius Omullo, Department of Clinical Research, St. Teresa Hospital, Kikuyu 1370, Kiambu, Kenya
Thomas Kimanzi Kitheghe, Ismail Abdi Sheikh, Department of Intensive Care Unit, Mama Lucy Kibaki Hospital, Nairobi 1278, Kenya
Abel Mwangi Gakuya, Department of Surgery, Nakuru County Teaching and Referral Hospital, Nakuru 71, Kenya
Joshua Macharia Gitumu, Department of Surgery, Naromoru Hospital, Nyeri 10105, Kenya
Mohamed Ahmed Shiekh, Department of Internal Medicine, Kenyatta National Hospital, Nairobi 20723, Kenya
Wilson Owuor Agai, Department of Paediatrics, Murang’a County Referral Hospital, Murang’a 52, Kenya
Wendy Bitengo Orenge, Department of Pharmacy, Mathari National Teaching and Referral Hospital, Nairobi 40663, Kenya
Harambee Moses Gogo, Department of Surgery, Thika County Referral Hospital, Thika 3, Kiambu, Kenya
Author contributions: Omullo FP contributed to the conception and design of the study, the statistical analyses, and the writing of the manuscript; Kitheghe TK, Sheikh IA, and Gakuya AM contributed to the literature search, data extraction, and the revision of the manuscript; Gitumu JM contributed to the statistical analyses and the revision of the manuscript; Sheikh MA, Agai WO, Orenge WB, and Gogo HM contributed to the interpretation of the data and the revision of the manuscript; all authors contributed to the critical revision of the manuscript and approved the final version.
AI contribution statement: The advanced version of Grammarly and ChatGPT (GPT-4, used as an advanced grammar/Language optimization tool) were employed. They were solely utilized for language polishing and writing assistance. Grammarly was used for checking grammar, punctuation, and conciseness. ChatGPT was only used to rephrase certain sentences to enhance readability, clarity, and academic quality. These operations were carried out after the authors completed the draft of the manuscript. No artificial intelligence tools were used for data analysis (all statistical analyses were completed by the authors in R software). No AI image generation tools were used. The authors assume full responsibility and accountability for the integrity, accuracy, originality, and scientific validity of the manuscript and all submitted materials.
Conflict-of-interest statement: The authors report no competing interests.
PRISMA 2009 Checklist statement: The authors have read the PRISMA 2009 Checklist, and the manuscript was prepared and revised according to the PRISMA 2009 Checklist.
Corresponding author: Felix Pius Omullo, MD, Senior Researcher, Department of Clinical Research, St. Teresa Hospital, Off Kikuyu Thogoto Road, Kikuyu 1370, Kiambu, Kenya.
Received: July 7, 2026
Revised: August 14, 2026
Accepted: October 16, 2026
Published online: November 8, 2026
Processing time: 121 Days and 16.1 Hours
Revised: August 14, 2026
Accepted: October 16, 2026
Published online: November 8, 2026
Processing time: 121 Days and 16.1 Hours
Core Tip
Core Tip: Artificial intelligence (AI) and radiomics models demonstrate pooled accuracy of 0.889 with low between-study heterogeneity (I2 = 11.3%). Performance was consistent between low and high risk-of-bias studies, although externally validated models showed significantly lower pooled accuracy (AUCP: 0.84) than internally/temporally validated models (AUC: 0.91; P = 0.003). However, the positive predictive value (54.4%) remains insufficient for artificial intelligence to confirm pathological complete response independently. Prospective validation with standardized reporting is required before implementation. AI should be integrated with, rather than replace, multidisciplinary assessment.