Overall and late mortality among 24 459 survivors of adolescent and young adult cancer in Alberta, Canada: a population-based cohort study

Background: Adolescent and young adult (AYA) cancer survivors are at an increased risk of premature mortality due to their cancer and its treatment. Herein, we aimed to quantify the excess risks of mortality among AYA cancer survivors and identify target populations for intervention.

Methods: The Alberta AYA Cancer Survivor Study is a retrospective, population-based cohort of individuals diagnosed with a first primary neoplasm at age 15-39 years in Alberta, Canada, between 1983 and 2017. We assessed cancer survivors (ie, all individuals included in the cohort) overall and for 2-year and 5-year survivorship subpopulations. We calculated standardised mortality ratios and absolute excess risks (AERs; per 10 000 person-years) compared with the general population, and cumulative mortality probability. Causes of death were categorised as deaths due to recurrence or progression (of the first primary neoplasm), deaths due to a subsequent primary neoplasm (SPN), and deaths due to non-neoplastic causes.

Findings: Among the 24 459 individuals included in the cohort, 5916 deaths were observed, which was 11·4 times (95% CI 11·1-11·7) that expected for the general population, equating to 191·6 (186·2-196·9) excess deaths; correspondingly, 5-year survivors had 4·2 times (4·0-4·4) more deaths than expected, equating to 74·3 (69·8-78·8) excess deaths. Increased age at diagnosis, poorer neighbourhood income quintile at diagnosis, first primary neoplasm type, and initial treatment plan were identified as important risk factors for mortality. While recurrence or progression was the main cause of excess mortality (AER 172·2 [167·4-177·1]), the majority of deaths beyond 10 years from diagnosis were due to SPNs and non-neoplastic causes among survivors of endometrial cancer, testicular cancer, and Hodgkin lymphoma. The cumulative mortality probability significantly decreased among more recently diagnosed survivors for all-cause mortality (p<0·0001) as well as recurrence or progression deaths (p<0·0001) and SPN deaths (p=0·0070), suggesting that long-term survival is improving.

Interpretation: AYA cancer survivors have substantial excess mortality. Given the high burden of late SPN and non-neoplastic deaths, survivors of endometrial cancer, testicular cancer, and Hodgkin lymphoma are notable populations that might benefit from primary, secondary, and tertiary prevention strategies.

Funding: None.