When Dr. Zak Kinsella first looked at what AI could see in tissue samples from breast cancer patients, he noticed something that standard genomic tests had been missing. In samples processed the same way hospitals use every day, the technology could detect dense clusters of immune cells surrounding tumors — and that density turned out to be a surprisingly powerful predictor of who would actually benefit from chemotherapy.
Research led by RCSI University of Medicine and Health Sciences and University College Dublin, published this month in Nature Communications, found that analyzing these cancer-fighting immune cells — called cytotoxic T-cells — could help doctors more accurately determine which patients with early-stage, hormone receptor-positive breast cancer might safely skip chemotherapy. The condition accounts for roughly 70 percent of all breast cancer diagnoses each year, and for many patients, the decision to add chemotherapy to hormone-blocking therapy has until now remained painfully unclear.
"For patients with an intermediate genomic risk, the decision around chemotherapy is often difficult, and uncertainty frequently leads to treatment that may not have been necessary, impacting quality of life," said Professor Darran O'Connor, research lead at RCSI School of Pharmacy and Biomolecular Sciences. His team worked with an Irish cohort of patients who had received intermediate risk scores — the clinical gray zone where chemotherapy is commonly prescribed as a precaution, despite uncertain benefit.
The findings revealed a striking pattern: patients whose tumors were surrounded by high densities of cytotoxic T-cells actually had poorer outcomes when treated with chemotherapy. That counterintuitive result suggests the immune cells may be signaling a biological vulnerability that current testing methods simply cannot detect.\n "It's really encouraging to see how much additional prognostic information can be extracted from these samples using AI," said Dr. Kinsella, the study's first author. "The density of cytotoxic T-cells in the tumor microenvironment proved to be a remarkably strong predictor of treatment response, and that has real implications for how we approach chemotherapy decisions in this patient group."
The researchers have jointly filed a patent and are now seeking commercial partners to help bring the technology into clinical practice. But before widespread rollout, the team acknowledges that larger validation studies will be needed. Professor William Gallagher, senior author from UCD's School of Biomolecular and Biomedical Science, said the work brings them closer to truly personalized treatment decisions.\n "Because this approach works from tissue samples processed as standard," Professor O'Connor noted, "it has the potential to improve both the precision and the equity of treatment for most women with early-stage breast cancer, regardless of where they are treated."