Development and validation of a risk prediction tool for second primary lung cancer

Journal of the National Cancer InstituteChoi E, Sanyal N, Ding VY, et al. | July 14, 2021

Advertisement

According to mounting evidence, there is a high risk of second primary lung cancer (SPLC) in lung cancer (LC) survivors, thus, researchers herein focused on developing and validating a prediction tool for clinical use to identify high-risk LC survivors for SPLC. They used cause-specific Cox regression and data from 6,325 ever-smokers in the Multiethnic Cohort (MEC) who were diagnosed with initial primary lung cancer (IPLC) in 1993–2017, and finally succeeded in developing a prediction model for 10-year SPLC risk post-IPLC diagnosis. This model showed a high predictive accuracy as well as discrimination (AUC = 81.9%) on the basis of bootstrap validation in MEC. By stratification according to the estimated risk quartiles, a statistically significantly higher SPLC incidence was noted in the 4th vs 1st quartile (9.5% vs 0.2%). An AUC of 78.8% and 72.7% were yielded in external validation using PLCO and NLST, respectively. Overall, this study offers a validated SPLC prediction model based on large population-based cohorts. This model can assist in detecting high-risk LC patients for SPLC and can be included into clinical decision-making for SPLC surveillance as well as screening.

Read the full article on Journal of the National Cancer Institute

Advertisement