Professor Ajay Aggarwal watched as an artificial intelligence system analyzed a CT scan and, in just over an hour, produced a complete radiotherapy plan that would have taken his team weeks to create by hand. The moment marked a turning point in global cancer care: at hospitals across India, South Africa, Jordan, and Malaysia, the ARCHERY trial had proven that AI could plan life-saving radiation treatment with the precision of the world's most skilled oncologists and physicists.
The stakes of this breakthrough could not be higher. Cervical cancer kills 350,000 women annually, and 94 percent of those deaths occur in low- and middle-income countries where trained radiotherapy specialists are scarce. While radiotherapy is the main curative treatment for cervical cancer, a devastating gap exists: only 10 percent of people who need it in low-income countries actually receive it, compared to 40 percent in middle-income nations. The shortage of skilled professionals—the oncologists and physicists required to design each treatment—remains one of the greatest barriers to care.
The ARCHERY trial, led by researchers at University College London and the London School of Hygiene & Tropical Medicine, tested whether AI could close this gap. Traditional radiotherapy planning requires an oncologist to outline tumors on imaging scans and identify surrounding tissue at risk, followed by a physicist determining the precise position, size, and shape of radiation beams. These complex tasks typically demand many hours spread across days or weeks. The AI-based software automates this entire process, identifying target structures and calculating optimal beam configurations in a fraction of the time.
The results, presented at the European Society for Radiotherapy and Oncology congress in Stockholm, exceeded expectations. For cervical cancer, the AI technology planned radiotherapy to a high standard in more than 95 percent of cases—meeting the international best-practice benchmarks set by leading oncology centers. For prostate cancer, the technology achieved a 85 percent success rate, still considered suitable for routine clinical use. The trial, which involved more than 1,000 cancer patients across three cancer types, was deliberately designed to test the technology in real-world settings where the need is greatest, not merely in elite research hospitals.
"In a usual workflow, planning radiotherapy can take many hours over several weeks," Professor Aggarwal explained. "This AI technology can reduce that time to just over an hour." That acceleration matters profoundly in countries where patients may wait months for treatment. By producing high-quality plans quickly, the technology enables hospitals to treat far more patients with existing staff, effectively bridging the workforce gap that has left millions without access to curative care.
The implications extend beyond cervical cancer. Radiotherapy helps cure approximately 40 percent of all cancer cases globally, yet millions lack access to it. Professor Mahesh Parmar of UCL's Institute of Clinical Trials noted that if all patients who needed radiotherapy could receive it, more than a million lives could be saved annually. This trial demonstrates that AI can support the World Health Organization's cervical cancer elimination initiative—a target that has seemed impossibly distant given current resource constraints.
What distinguishes this research is its rigor and global scope. Unlike many AI trials, which are often small-scale, single-center studies conducted in wealthy countries, ARCHERY was deliberately multi-national and large. It fills a critical gap in evidence, providing the kind of robust proof needed to support implementation of AI tools where need is greatest. As hospitals worldwide consider how to expand access to radiotherapy, this technology offers a concrete path forward.