Face Aging Rate Serves as a Non-invasive Biomarker for Cancer Prognosis

A research team at Mass General Brigham has expanded the capabilities of FaceAge, an artificial intelligence tool initially designed to estimate biological age from a single facial photograph. A new study indicates that by analyzing multiple images of a person taken over time, the tool can provide significant insights into how a patient with cancer might respond to treatment.

The researchers utilized the Face Aging Rate (FAR), a metric that measures changes in biological age based on photos captured at different intervals. By examining 2,279 cancer patients who underwent radiation therapy between 2012 and 2023, the study identified a clear link between higher FAR and reduced survival probability. Specifically, the data showed that facial aging outpaced chronological aging by 40% among the study cohort, with the association between accelerated aging and lower survival being most pronounced when the time interval between photos spanned at least two years.

“Deriving a Face Aging Rate from multiple, routine facial photographs allows for near real-time tracking of an individual’s health,” said Raymond Mak, corresponding author of the study published in Nature Communications. “Our study suggests that measuring FaceAge over time may refine personalized treatment planning, improve patient counseling, and help guide the frequency and intensity of follow-up in oncology.”

The analysis also incorporated FaceAge Deviation (FAD), which assesses biological age relative to chronological age at a single point in time. While both metrics are useful, FAR proved more stable for predicting survival outcomes over longer periods, suggesting that dynamic, longitudinal measurements may be more reliable than singular readings. Integrating FAR with baseline FAD, according to the authors, could offer a more nuanced understanding of a patient's evolving health.

“Tracking FaceAge over time from simple photos offers a non-invasive, cost-effective biomarker with potential to inform individuals of their health,” noted co-author Hugo Aerts.

The team plans to conduct further prospective trials to assess the utility of FaceAge and FAR across diverse populations, different cancer types, and other chronic conditions. A web portal is now available for the public to submit photos and contribute to further technological development.