AI helps identify new cancer applications for common medicines: Oncologist

LAHORE: Artificial intelligence (AI) is enabling researchers to uncover new cancer applications for widely used medicines by analysing tumour biology, genetic mutations and drug–target networks.

This was revealed by Oncologist Dr Munira Shabbir Moosajee of The Aga Khan University while addressing the students and researchers at the Syed Babar Ali School of Science and Engineering (SBASSE), here at the LUMS.

According to her, drug repurposing is becoming increasingly relevant as conventional cancer treatment options run out after initial therapy lines. “In oncology, blocking one pathway often activates another,” she said, adding “That biological complexity creates opportunities to target multiple vulnerabilities using drugs we already understand.”

On the role of AI, she said computational tools are increasingly used to integrate genomic sequencing, transcriptomic data and chemical structures to predict drug sensitivity and toxicity. “AI cannot replace biology,” she said, adding, “But it helps us prioritise where to look.”

Dr Munira Shabbir explained that cancer care is moving toward tumour-agnostic treatment, where therapy is guided by genetic and molecular features rather than the tumour’s location. She cited trastuzumab as a key example, noting that the drug’s use has expanded well beyond breast cancer as HER2 over expression has been identified in other tumours.

She discussed findings from a major international study showing that aspirin significantly reduced recurrence rates in colorectal cancer patients with PI3K mutations, describing it as one of the clearest clinical demonstrations of drug repurposing to date.

Other medicines highlighted included metformin, which has shown biological activity against tumour growth when administered to endometrial cancer patients prior to surgery, and mebendazole, an anti-parasitic drug identified through computational modelling as having VEGF-inhibitory properties relevant to colorectal cancer.

Dr Munira said her team is preparing a clinical trial using the beta blocker propranolol in colorectal cancer patients, modelled on pre-surgical studies conducted internationally. The trial aims to assess how the drug affects tumour biology before surgery.

Despite scientific promise, she warned that translating repurposed drugs into clinical practice remains difficult in Pakistan due to regulatory delays, limited institutional funding and a lack of pharmaceutical support for non-patented therapies.

Daud Haider, a data scientist working on AI-driven clinical trial platforms, said complex eligibility criteria and fragmented databases make it difficult for doctors to identify suitable trials. He said AI-based trial matching systems could reduce search time and improve enrolment.

Other speakers said sustained collaboration between clinicians and data scientists is essential if AI-driven discoveries are to translate into improved patient outcomes.

Copyright Business Recorder, 2026