A model-based estimate of cumulative excess mortality in survivors of childhood cancer

Abstract

Background: Although childhood cancer survival rates have dramatically increased, survivors face elevated risk for life-threatening late effects, including secondary cancer.

Objective: To estimate the cumulative effect of disease- and treatment-related mortality risks on survivor life expectancy.

Design: State-transition model to simulate the lifetime clinical course of childhood cancer survivors.

Setting: Childhood Cancer Survivor Study.

Patients: Five-year survivors of childhood cancer.

Measurements: Probabilities of risk for death from the original cancer diagnosis, excess mortality from subsequent cancer and cardiac, pulmonary, external, and other complications, and background mortality (age-specific mortality rates for the general population) were estimated over the lifetime of survivors of childhood cancer.

Results: For a cohort of 5-year survivors aged 15 years who received a diagnosis of cancer at age 10 years, the average lifetime probability was 0.10 for late-recurrence mortality; 0.15 for treatment-related subsequent cancer and death from cardiac, pulmonary, and external causes; and 0.05 for death from other excess risks. Life expectancy for the cohort of persons aged 15 years was 50.6 years, a loss of 10.4 years (17.1%) compared with the general population. Reduction in life expectancy varied by diagnosis, ranging from 4.0 years (6.0%) for kidney tumor survivors to more than 17.8 years (> or =28.0%) for brain and bone tumor survivors, and was sensitive to late-recurrence mortality risk and duration of excess mortality risk.

Limitation: Estimates are based on data for survivors who received treatment 20 to 40 years ago; patients who received treatment more recently may have more favorable outcomes.

Conclusion: Childhood cancer survivors face considerable mortality during adulthood, with excess risks reducing life expectancy by as much as 28%. Monitoring the health of current survivors and carefully evaluating therapies with known late toxicities in patients with newly diagnosed cancer are needed.

Figure 1. Model structure

At the start of the simulation, a cohort of five-year childhood cancer survivors enters the model. Each month, they face a risk of dying from late recurrence, non-recurrence excess mortality and background mortality. Non-recurrence excess mortality includes risks associated with subsequent cancers, cardiac, pulmonary, external causes and other causes. Individuals are followed throughout their lifetime.

Figure 2. Cause-specific attributable proportion of overall mortality risk

The bar graph depicts the proportion of overall mortality risk attributed to each specific mortality cause at 10, 20, 30, 40, 50 and 60 years after diagnosis. As survivors age, the cumulative proportion of overall mortality attributable to background mortality increases relative to the proportion for all late-effects from cancer or cancer treatment combined.