AI pathology for cancer recurrence prediction

In early July 2026 an artificial‑intelligence system was reported that could assess the risk of meningioma recurrence using routine haematoxylin‑and‑eosin‑stained pathology slides. The approach demonstrated that AI could extract prognostic information from standard histology without the need for genomic data.

Later that month a more detailed study described a comparable AI‑driven computational‑pathology platform applied to colorectal cancer. The tool generated a recurrence‑risk score and identified a subgroup of patients with locally advanced rectal cancer who benefited substantially from adding irinotecan to chemoradiotherapy, cutting recurrence by roughly 43 % and mortality by about 50 % in that group. Both applications aim to personalize follow‑up and adjuvant treatment, sparing low‑risk patients from unnecessary toxicity while intensifying therapy for those most likely to profit. The technologies are undergoing broader validation across multiple international centres and will require regulatory approval before routine clinical use.