Researchers at Sylvester Comprehensive Cancer Center, part of the University of Miami Miller School of Medicine, have proposed a framework for developing, evaluating, and implementing AI tools to improve cancer survivorship and supportive care. The PRECISION-AI Adjunctive Supportive Care Development Framework is intended to guide the responsible use of AI to address unmet physical, psychosocial, and practical needs among cancer survivors while maintaining clinical oversight. The conceptual perspective was led by Frank Penedo, PhD, Director of Sylvester’s Cancer Survivorship and Supportive Care Institute and Associate Director for Population Sciences.

“For many survivors, the hardest part isn’t always treatment itself. It’s what happens afterward, when patients are managing fatigue, anxiety, uncertainty and other challenges between clinic visits,” Dr. Penedo said in a Sylvester press release. He emphasized that AI could help extend evidence-based support beyond the cancer center without replacing in-person clinical care.

Framework Development and Supporting Research

The researchers examined opportunities and challenges associated with integrating AI into survivorship care, drawing on existing evidence from oncology AI and cancer survivorship research. They considered potential applications for personalized education, symptom monitoring, psychosocial support, and care coordination, distinguishing approaches supported by prospective evidence from those still in early development.

The framework builds on Sylvester’s existing survivorship research and clinical infrastructure, including My Wellness Check, an electronic health record–integrated screening and triage program through which patients report symptoms, quality-of-life concerns, and supportive care needs before clinical appointments. The program has collected longitudinal patient-reported outcomes, supportive care needs, and nutritional data from more than 37,000 ambulatory oncology patients.

In an earlier analysis of 25,592 ambulatory cancer survivors followed over 36 months, researchers developed machine learning models to identify patterns related to symptom burden and unplanned health-care utilization. According to the researchers, advanced analytic and AI-driven techniques bolstered the models' predictive performance for identifying unfavorable outcomes by more than 25%.

Personalizing Survivorship Care With AI

Building on this infrastructure, researchers are developing My Wellness Support, an agentic AI–enabled, risk-stratified supportive care platform designed to complement standard survivorship care.

The platform combines patient-reported outcomes, clinical records, and behavioral information to piece together risk patterns and identify unmet needs. It is designed to deliver personalized, guideline-based educational resources, symptom-management strategies, psychosocial support, and recommendations addressing patients' practical needs.

Patients will interact with an AI companion between clinical appointments, while clinicians and study personnel monitor trends, risk scores, and alerts through a dedicated dashboard.

“A patient may be doing well medically but still be struggling with symptoms, stress, or questions that arise between appointments,” said AkinaNatori, MD, MSPH, a Sylvester oncologist and Assistant Professor in the Miller School’s Division of Medical Oncology at the University of Miami.

The investigators emphasized that integrating AI into survivorship care requires rigorous prospective validation, data protections, transparency, bias mitigation, equitability, and generalizability across diverse patient populations. The proof-of-concept project is still considered in an early testing phase. 

Sara E. Fleszar-Pavlović, PhD, Research Assistant Professor in the Miller School’s Division of Medical Oncology, stressed the importance of scientific validation and continuous evaluation to ensure responsible implementation. “New technologies often move faster than the systems designed to evaluate them,” she observed.

The authors concluded that AI has the potential to expand access to personalized survivorship support and improve health-related quality of life. However, realizing these benefits will require prospective clinical testing and sustained collaboration among clinicians, researchers, cancer survivors, developers, ethicists, and policymakers.

ASCO AI in Oncology is published by Conexiant under a license arrangement with the American Society of Clinical Oncology, Inc. (ASCO^®). The ideas and opinions expressed in ASCO AI in Oncology do not necessarily reflect those of Conexiant or ASCO. For more information, see Policies.