Imaging-based biological age estimation predicts competing mortality risk in lung cancer screening
Abstract
Lung cancer screening (LCS) participants have a high competing risk of non-lung cancer related death, which limits screening benefit. We hypothesized that imaging-based biological age can identify individuals at higher risk of non-lung cancer mortality. Low dose computed tomography scans from a large LCS trial were retrospectively analyzed. Bone density loss, muscular fat infiltration, vascular calcification, and visceral fat mass were calculated using segmentations from a nnU-Net based deep learning segmentation model. Each participant’s age gap, defined as the difference between their estimated biological age and chronological age, was estimated with disease course mapping using a Bayesian mixed-effects model. Disease course map construction used all available longitudinal scan data, whilst individual biological age prediction used baseline scan data only. The performance of age gap as a parameter in competing risk models for non-lung cancer related death versus lung cancer diagnosis was evaluated, following the TRIPOD + AI reporting guideline. We included 29,745 scans from 12,478 participants in the final analysis. Median follow-up was 5 years, representing 58,284 person-years at risk. Multivariable survival models of competing risks were constructed, including the estimated age gap. The subdistribution hazard ratio (sHR) for each decade of age gap for non-lung cancer related death was higher (sHR = 2.0, 95% CI 1.8–2.3), compared to lung cancer diagnosis (sHR = 1.3, 95% CI 1.2–1.5). Addition of age gap improved model concordance of non-lung cancer death compared to a model with clinical variables alone (pooled C-index = 0.69 vs 0.73, difference + 0.04, 95% CI 0.03–0.04, p < 0.001). Biological age estimation using an imaging-based disease progression model improves prediction of non-lung cancer related death in lung cancer screening.
Related
- Competing Mortality, Not Breast-Cancer Mortality, Explains the Survival Deficit of Breast Cancer After a Prior Non-Breast Cancer: A SEER Competing-Risks Analysis
- Competing Risk Prediction of COVID-19 Hospital Mortality with Model Diagnostics and Exploratory Evaluation in Mexico
- Changing cause-of-death composition among patients undergoing gastrectomy for stage I–III gastric adenocarcinoma: a SEER landmark competing-risk analysis of mortality from diagnosis
- Association between radiotherapy laterality and cardiovascular disease mortality in male breast cancer patients receiving postoperative radiotherapy: a competing risks analysis and standardized mortality ratio assessment using SEER data.
- Cognitive function predicts all-cause mortality independent of social engagement among community-dwelling elderly Japanese men: a 15-year prospective cohort study
BibTeX
@article{cheng2026imagingbased,
title = {{Imaging-based biological age estimation predicts competing mortality risk in lung cancer screening}},
author = {Daryl O. Cheng and John McCabe and Tanya Patrick and Robert Chapman and Shanshan Wang and Tianqi Yang and Adam Szmul and Mehran Azimbagirad and Shahab Aslani and Monica L. Mullin and Sindhu Bhaarrati Naidu and Chuen Ryan Khaw and Alfred C. P. So and SUMMIT consortium and Aashna Samson and Adefiola Olabode and Alberto Villanueva and Ali Mohammed and Alice Cotton and Andrew Crossingham and Andrew Perugia and Anjeli Ketkar and Anna Sikorska and Anthony Edey and Antonette Andrews and April Neville and Burcu Ozaltin and Camilla Mitchell and Catherine Nestor and Celia Snell and Charlie Sayer and Charlotte Cash and Chimtom Nwaosu and Christine Hosein and Claire Levermore and Columbus Ife and Derya Ovayolu and Dominique Arancon and Domminique Arancon and Eleanor Hellier and Elena Stefan and Elodie Murali and Esther Arthur-Darkwa and Ethaar El-Emir and Fabia Ruggiero and Fahiza Chowdhury and Fahmida Hoque and Fanta Bojang and Farhida Begum and Geoff Bellingan and Georgia Bullock and Graham Robinson and Gulistan Keskinbicak and Gulten Geneci and Harriet Sorrell and Hasti Robbie and Helen Kiconco and Helen Victoria Bowyer and Henna Makwana and Hina Pervez},
journal = {Scientific Reports},
year = {2026},
month = oct,
doi = {10.1038/s41598-026-75832-2},
url = {https://www.nature.com/articles/s41598-026-75832-2},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Imaging-based biological age estimation predicts competing mortality risk in lung cancer screening
AU - Daryl O. Cheng
AU - John McCabe
AU - Tanya Patrick
AU - Robert Chapman
AU - Shanshan Wang
AU - Tianqi Yang
AU - Adam Szmul
AU - Mehran Azimbagirad
AU - Shahab Aslani
AU - Monica L. Mullin
AU - Sindhu Bhaarrati Naidu
AU - Chuen Ryan Khaw
AU - Alfred C. P. So
AU - SUMMIT consortium
AU - Aashna Samson
AU - Adefiola Olabode
AU - Alberto Villanueva
AU - Ali Mohammed
AU - Alice Cotton
AU - Andrew Crossingham
AU - Andrew Perugia
AU - Anjeli Ketkar
AU - Anna Sikorska
AU - Anthony Edey
AU - Antonette Andrews
AU - April Neville
AU - Burcu Ozaltin
AU - Camilla Mitchell
AU - Catherine Nestor
AU - Celia Snell
AU - Charlie Sayer
AU - Charlotte Cash
AU - Chimtom Nwaosu
AU - Christine Hosein
AU - Claire Levermore
AU - Columbus Ife
AU - Derya Ovayolu
AU - Dominique Arancon
AU - Domminique Arancon
AU - Eleanor Hellier
AU - Elena Stefan
AU - Elodie Murali
AU - Esther Arthur-Darkwa
AU - Ethaar El-Emir
AU - Fabia Ruggiero
AU - Fahiza Chowdhury
AU - Fahmida Hoque
AU - Fanta Bojang
AU - Farhida Begum
AU - Geoff Bellingan
AU - Georgia Bullock
AU - Graham Robinson
AU - Gulistan Keskinbicak
AU - Gulten Geneci
AU - Harriet Sorrell
AU - Hasti Robbie
AU - Helen Kiconco
AU - Helen Victoria Bowyer
AU - Henna Makwana
AU - Hina Pervez
PY - 2026
DA - 2026/10/15
JO - Scientific Reports
DO - 10.1038/s41598-026-75832-2
UR - https://www.nature.com/articles/s41598-026-75832-2
AB - Lung cancer screening (LCS) participants have a high competing risk of non-lung cancer related death, which limits screening benefit. We hypothesized that imaging-based biological age can identify individuals at higher risk of non-lung cancer mortality. Low dose computed tomography scans from a large LCS trial were retrospectively analyzed. Bone density loss, muscular fat infiltration, vascular calcification, and visceral fat mass were calculated using segmentations from a nnU-Net based deep learning segmentation model. Each participant’s age gap, defined as the difference between their estimated biological age and chronological age, was estimated with disease course mapping using a Bayesian mixed-effects model. Disease course map construction used all available longitudinal scan data, whilst individual biological age prediction used baseline scan data only. The performance of age gap as a parameter in competing risk models for non-lung cancer related death versus lung cancer diagnosis was evaluated, following the TRIPOD + AI reporting guideline. We included 29,745 scans from 12,478 participants in the final analysis. Median follow-up was 5 years, representing 58,284 person-years at risk. Multivariable survival models of competing risks were constructed, including the estimated age gap. The subdistribution hazard ratio (sHR) for each decade of age gap for non-lung cancer related death was higher (sHR = 2.0, 95% CI 1.8–2.3), compared to lung cancer diagnosis (sHR = 1.3, 95% CI 1.2–1.5). Addition of age gap improved model concordance of non-lung cancer death compared to a model with clinical variables alone (pooled C-index = 0.69 vs 0.73, difference + 0.04, 95% CI 0.03–0.04, p < 0.001). Biological age estimation using an imaging-based disease progression model improves prediction of non-lung cancer related death in lung cancer screening.
SN - 2045-2322
ER -