A retrospective cohort study published in JAMA Network Open has evaluated whether non-invasive measures of biological age, including artificial intelligence–derived facial age and spirometry-based lung age, are associated with outcomes in older patients with early-stage non–small cell lung cancer (NSCLC) undergoing stereotactic body radiotherapy (SBRT).
Accurately assessing physiological fitness and life expectancy in older patients with early-stage NSCLC remains challenging, particularly as chronological age alone may not reflect underlying health status. In this study, researchers evaluated 670 patients aged 60 years or older treated with definitive SBRT, examining whether biological age metrics could provide additional prognostic value beyond conventional clinical factors.
CLINICAL SUMMARY
What was examined
A retrospective cohort study evaluated whether biological age measures—derived from facial imaging and spirometry—are associated with survival outcomes in older patients with early-stage NSCLC treated with SBRT.
Key findings
- Higher AI-derived face age was independently associated with worse overall survival (HR per decade 1.39) and increased risk of early mortality.
- Chronological age was not independently associated with survival outcomes after adjustment.
- Lung age showed limited correlation with face age, suggesting complementary biological information.n
Clinical implications
- Biological age measures may provide additional prognostic information beyond chronological age in older patients with NSC.LC.
- Non-invasive tools such as facial imaging and spirometry may support risk stratification in conjunction with established clinical factors.
- Prospective validation is required before integration into routine clinical practice.
Face age was estimated using a deep learning algorithm applied to clinical photographs, while lung age was derived from spirometry data in a subset of patients. The median chronological age was 77 years, compared with a median facial age of 79 years. Over a median follow-up of 44 months, median overall survival was 47 months, and approximately one-quarter of patients died within two years.
On multivariable analysis adjusting for key clinical variables, including performance status, stage, smoking history, and histology, higher face age was independently associated with worse overall survival (adjusted hazard ratio [HR] per decade, 1.39; 95% CI, 1.13–1.71; p=0.002). Face age was also associated with an increased risk of early mortality (death within two years) (adjusted HR per decade, 1.35; 95% CI, 1.03–1.78; p=0.03).
Featured Event
In contrast, chronological age was not independently associated with overall survival or early mortality after adjustment for clinical factors. Patients with a face age of 85 years or older had a higher risk of early mortality (adjusted HR, 1.59; 95% CI, 1.06–2.40).
Among patients with available pulmonary function data, lung age showed limited correlation with face age, suggesting that these measures may capture distinct aspects of physiological ageing. While lung age was associated with outcomes in univariate analyses, face age remained independently predictive of survival when both metrics were included in multivariable models.
These findings highlight the potential role of biological age as an additional marker of patient fitness beyond chronological age alone in older patients undergoing SBRT. The use of readily available data, including clinical photographs and spirometry, may offer a potential approach to improving risk stratification.
However, the study is retrospective and conducted within a single academic network, and further validation is required before such tools can be incorporated into routine clinical decision-making. The findings are hypothesis-generating and should be interpreted in the context of established clinical assessment frameworks.
From a clinical perspective, the study contributes to growing interest in personalised treatment decision-making in oncology, particularly in older populations where balancing treatment benefit against competing risks remains complex.
Paper: Lee G, et al. Multimodal Assessment of Biological Age Among Patients With Early-Stage Non–Small Cell Lung Cancer. JAMA Network Open. 2026. Access online here.
