AI Reveals Rectal Cancer Therapy Benefit; Study Finds Online Patient Info Lags
AI identified rectal cancer patients who benefit from adding irinotecan to chemoradiotherapy, cutting recurrence and death risk. Separate research finds most online AI-cancer information is low quality and too complex for patients.
Artificial intelligence developed by researchers at UCL has revealed a promising combination of cancer treatments for patients with locally advanced rectal cancer, while a separate study finds that online information about AI and cancer for patients is largely low quality and difficult to read.
In a peer-reviewed observational study published in eBioMedicine, combining the cancer treatment irinotecan with standard chemoradiotherapy improved survival against advanced rectal cancer in patients with a high concentration of cancerous cells in their tumours. For patients who had a high tumour cell density in biopsy samples prior to starting treatment, adding irinotecan reduced the risk of cancer recurrence by about 43% and reduced the risk of death by about 50% compared with those who received chemotherapy using capecitabine combined with radiation therapy. Those with low concentrations of cancer cells showed no difference.
The effectiveness was revealed using a specially-developed AI trained to distinguish patients by their cancer cell density on tumour samples taken at the time of diagnosis. The researchers developed the free online tool Octopath, where clinicians can upload biopsy slides to be analysed. The AI scanned standard images of biopsies, counted millions of cells, and sorted patients into high and low concentration categories. The research is part of the analysis of the phase III ARISTOTLE trial, which enrolled patients from 75 UK hospitals and featured 414 rectal cancer biopsy samples; the analysis classified 188 as high concentration and 226 as low concentration. Colorectal cancer is the fourth most fatal cancer in the UK, and combining irinotecan with capecitabine is likely to intensify serious side effects such as diarrhoea or low white blood cell counts. The findings show that doctors assisted by AI can pinpoint which patients will likely benefit from the more intensive treatment before it begins.
Separately, researchers from the Abramson Cancer Center of the University of Pennsylvania and Penn's Perelman School of Medicine presented a study at the 2026 American Society of Clinical Oncology Annual Meeting (Abstract 9000) showing that online information about AI and its impact on cancer research and treatment is limited, and that available webpages and videos are largely of low quality, difficult to read, and frequently omit risks of AI use. After screening the first 320 webpages and videos from Google and YouTube searches and removing content not relevant to AI and cancer care or not intended for lay audiences, 52 webpages (31 percent of Google search results) and 29 videos (19 percent of YouTube search results) were included in the final analysis. Only 17 webpages (33 percent) and seven videos (23 percent) were considered high quality. While the American Medical Association and National Institutes of Health recommend a 6-8th grade reading level for consumer health information, median readability of the webpages was college level, and only 15 percent of webpages mentioned the risk of AI hallucinations.
A collection on prospective and interventional clinical evidence in medical AI notes that artificial intelligence is rapidly moving from retrospective benchmarking studies into clinical workflows, but much of the literature still focuses on model accuracy as the primary endpoint. The collection welcomes studies built around AI–action bundles, the combination of an AI result and the predefined action expected from clinicians, patients, or care teams, with an explicit causal pathway from model output to action to outcome.
In head and neck cancer, personalized treatment remains limited despite substantial biological heterogeneity. Using the SuPerTreat project as a case study, researchers outlined a prototype clinical decision support system integrating transcriptomic data and artificial intelligence, and summarized expert consensus on its potential, requirements for accuracy, validation, regulatory alignment, and clinical implementation. The Perspective provides a roadmap to guide future development and responsible integration of clinical decision support systems into precision oncology.