Subsequent primary cancer incidence among cancer survivors in the United States, 1975–2019: An age–period–cohort analysis

by Hui G. Cheng, Livingstone Aduse-Poku, Chelsey McGill, Oxana Palesh, Susan Hong

Background

The growing population of cancer survivors faces elevated risks of subsequent primary cancers (SPCs), yet temporal patterns in SPC incidence remain poorly understood. This study aims to characterize age-, period-, and cohort-specific patterns in SPC incidence among US cancer survivors using population-based data.

Methods and findings

We conducted a retrospective cohort study using Surveillance, Epidemiology, and End Results (SEER) 8 registries, identifying 3.36 million individuals diagnosed with a first primary cancer between 1975 and 2019. Survivors were followed through 2022 to estimate the incidence of SPCs. We used age–period–cohort analysis to estimate longitudinal age curves, cohort and period rate ratios, and annual percent changes in SPC incidence. Analyses were stratified by sex and the five most common index cancer sites.During 29.5 million person‑years of follow‑up, 510,340 SPCs were observed. SPC incidence increased with age at index cancer diagnosis, rising among females from 915 per 100,000 person‑years at ages 35–39 years to 1,980 per 100,000 at ages 75–79 years, and among males from 1,228 per 100,000 to 2,945 per 100,000 across the same age groups, demonstrating steeper rises in men. Cohort-specific SPC risk peaked in the 1935–1945 birth cohorts and declined in later cohorts, except among female survivors of lung cancer and male survivors of bladder cancer, where risks continued to rise. Period patterns showed overall declines in SPC incidence, particularly among survivors diagnosed at a younger age, but increasing risks among survivors diagnosed at an older age and survivors of specific index cancer sites. Notably, SPC incidence rose by 60% among female lung cancer survivors between 1975–1979 and 2015–2019 (incidence rate ratio = 1.60, 95% CI [1.22, 2.09]; p < 0.001). Main limitations include the descriptive nature of age–period–cohort analyses and the absence of treatment, genetic, and lifestyle data in SEER.

Conclusions

SPC risk is shaped by complex, site- and sex-specific temporal patterns. These findings underscore the need for tailored survivorship care strategies that incorporate age, cohort, and index cancer site to mitigate future SPC burden.

Source: journals.plos.org