AI-Enhanced RAIN Protocol: A Systematic Approach to Optimize Drug Combinations for Rectal Neoplasm Treatment

Background Rectal cancers, or rectal neoplasms, are tumors that develop from the lining of the rectum, the concluding part of the large intestine ending at the anus. These tumors often start as benign polyps and may evolve into malignancies over several years. The causes of rectal cancer are diverse, with genetic mutations being a key factor. These mutations lead to uncontrolled cell growth, resulting in tumors that can spread and damage healthy tissue. Age, genetic predisposition, diet, and hereditary conditions are among the risk factors. Treating rectal cancer is critical to prevent severe health issues and death. Untreated, it can cause intestinal blockage, metastasis, and deteriorate the patient’s quality of life. Effective treatment hinges on finding the right drug combinations to improve therapeutic outcomes. Given the intricacies of cancer biology, treatments often combine surgery, chemotherapy, and radiation, with drugs chosen to target different tumor growth mechanisms, aiming to reduce the tumor and limit side effects. The continuous advancement in cancer treatments highlights the need for ongoing research to discover new drug combinations, offering patients improved recovery prospects and a better quality of life. This background encapsulates a detailed yet succinct understanding of rectal neoplasms, their origins, the urgency of treatment, and the quest for effective drug therapies, paving the way for discussions on treatment advancements and patient care impacts. Method This study employed the RAIN protocol, comprising three steps: firstly, utilizing the GraphSAGE model to propose drug combinations for rectal neoplasm treatment Each node in the graph model is a drug or a human gene/protein that acts as potential target for the disease, and the edges are P-values between them; secondly, conducting a systematic review across various databases including Web of Science, Google Scholar, Scopus, Science Direct, PubMed, and Embase, with NLP investigation; and thirdly, employing a meta-analysis network to assess the efficacy of drugs and genes in relation to each other. All implementations was conducted using Python software. Result The study evaluated the efficacy of Oxaliplatin, Leucovorin, and Capecitabine in treating Rectal Neoplasms, confirming their effectiveness through a review of 30 studies. The p-values for individual drugs were 0.019, 0.019, and 0.016 respectively, while the combined use of all three yielded a p-value of 0.016. Conclusion Given the significance of rectal neoplasms, policymakers are urged to prioritize the healthcare needs of affected individuals. Utilizing artificial intelligence within the RAIN protocol can offer valuable insights for tailoring effective drug combinations to better address the treatment and management of rectal neoplasms patients. ![Figure][1] Highlights ### Competing Interest Statement The authors have declared no competing interest. * ### Abbreviations STROBE : Strengthening the Reporting of Observational studies in Epidemiology PRISMA : Preferred Reporting Items for Systematic Reviews and Meta-Analysis RAIN : Systematic Review and Artificial Intelligence Network Meta-Analysis [1]: pending:yes