“Younger people are getting cancer more and more,” said Sai Jasti, senior vice president and head of data science and AI at Bayer. “And what that means is people may have to stay on therapies for longer than what we traditionally have been used to.”
That demographic shift is shaping how artificial intelligence is applied in oncology as 2025 draws to a close. While generative AI has attracted public attention, clinicians and executives in cancer research say the more consequential changes are coming from machine learning, deep learning, and large-scale data integration aimed at making treatment more precise and less harmful.
Jasti, who oversees AI across Bayer’s pharma R&D, argues that rising incidence among younger patients raises the bar for drug design. If patients are expected to remain on treatment for years rather than months, toxicity and long-term side effects become central considerations. “That translates to the fact that we also have to make kinder medicines,” he said. “So that is one of the hypotheses behind our oncology strategy, to go towards precision drug development.”
Precision in oncology is evolving slowly and incrementally. Rather than bespoke drugs for every patient, the trend reflects a finer-grained understanding of disease biology enabled by multimodal datasets—molecular, cellular, clinical and real-world evidence—and AI systems that detect complex patterns across those sources.
The pace of clinical adoption is constrained by the realities of medicine. Ofer Sharon, MD, CEO of precision oncology company OncoHost, said healthcare lags consumer tech because clinical AI systems must be validated through lengthy, costly trials and navigate regulatory pathways not built for models that evolve with new data. “With medicine, we are still behind,” Sharon said.
Advances so far have come mainly from machine learning systems that integrate genomics, proteomics, imaging and electronic health records to reveal patterns clinicians cannot see unaided. Those capabilities are nudging oncology toward more personalized care, but most patients are still treated according to broad clinical guidelines derived from large cohort trials.
Sharon said the near-term trend is increasing the resolution of those guidelines by breaking broad categories into smaller subgroups defined by molecular, clinical or protein-level markers. Biomarkers are emerging as the practical bridge between AI insights and routine decision-making. “If we want to see AI implemented into the clinical day-to-day, it needs to be part of the mainstream,” he said.
Regulators are beginning to align with that direction. Agencies such as the U.S. Food and Drug Administration are showing momentum toward requiring biomarkers alongside new drugs and streamlining approval pathways for diagnostic tools that support treatment choices.
At large drugmakers, the impact is uneven across the development pipeline. Jasti said the biggest gains have been in early-stage research—AI-first molecule design and target validation—while clinical development remains more challenging. Patient identification using real-world data is an area of growing momentum.
Bayer is developing a Microsoft-backed agentic AI framework for the Israeli market designed to coordinate diagnosis, economic assessment, policy considerations and care pathways to improve access to medicines, with plans to bring the system to market next year. “All in all, it is contributing towards achieving that ambition,” Jasti said, adding he expects three to five years before the full potential of these approaches is realized.
Both executives warned against overly futuristic expectations such as fully personalized drugs produced on demand. “Biology is super, super complicated,” Jasti said. “It’s not an internet world wherein you can just scrape through the entire wide web, train a big model, and come up with a large language model.”
For now, the trajectory of AI in oncology is defined by infrastructure: better data, improved models and tighter integration between diagnostics, drug development and clinical decision-making. As cancer increasingly affects younger populations, the pressure to make treatments both effective and sustainable over the long term is likely to intensify.
Leave a Reply