City of Hope’s custom AI model saves clinicians thousands of hours

City of Hope has launched HopeLLM, its own proprietary generative AI model designed specifically for cancer care, after finding no commercial AI solutions that met its complex oncology needs. The tool has already saved clinicians “thousands” of hours in its first week of deployment and has attracted interest from pharmaceutical companies seeking to leverage its clinical trial-matching capabilities.

What you should know: HopeLLM addresses the unique challenges of cancer care by processing vast, complex medical records that can span decades of treatment history.

• Cancer patients typically have electronic health records containing 10-20 years of test results and visit notes, with one in five patients having EHRs the size of Herman Melville’s Moby Dick (over 206,000 words).
• The tool assists with patient onboarding, summarizes medical records in seconds, matches patients with clinical trials, and pulls relevant data for research.
• City of Hope, one of the nation’s largest cancer research and treatment organizations, onboarded more than 150,000 new patients in 2024, each with complex medical histories requiring extensive record review.

How it works: The AI model integrates multiple commercial and open-source models directly into City of Hope’s Epic electronic health record system.

• Individual providers can receive personalized summaries of patient records at the point of care.
• The health system can analyze longitudinal care journeys and combine insights with social determinants of health and genomics data.
• The tool has been deployed across City of Hope’s enterprise locations in Los Angeles, Chicago, Phoenix, and Atlanta.

What they’re saying: Healthcare professionals have responded enthusiastically to the efficiency gains from HopeLLM.

• “I got hugged the other day,” said Simon Nazarian, City of Hope’s chief digital and technology officer, describing providers’ reactions to the tool.
• “We get a lot of accolades [from providers] about the fact that they’re not spending as much ‘pajama time,’ and they have this available at the point of care.”
• “Whether in care delivery or research, we want to use AI in order to accelerate discovery to continuously improve our quality of care, and deliver more precise therapies,” Nazarian explained.

Why this matters: Few health systems have the resources and infrastructure to develop their own AI platforms, making City of Hope’s approach particularly significant for specialized care.

• The system has been creating AI-driven predictive models since 2017, focusing on risk identification for sepsis and surgical complications.
• HopeLLM represents an evolution toward comprehensive record analysis and cross-departmental integration rather than just risk assessment.
• The tool’s precision therapy capabilities could advance personalized cancer treatment by accounting for patients’ complete medical backgrounds.

Commercial interest: Pharmaceutical companies are particularly interested in HopeLLM’s clinical trial-matching capabilities.

• City of Hope is exploring partnerships with external stakeholders to make the tool available to others in the oncology market.
• The health system continues to monitor the platform through ongoing AI governance protocols involving physicians, legal, and compliance experts.
• “There is no way for us to predict where a revolutionary technology like this is going to go,” Nazarian said, “but we are staying close to it and will be at the forefront.”