Late and very late mortality in 5-year survivors of childhood cancer:changing pattern over four decades--experience from the Nordic countries

• Published in: International Journal of Cancer
• Format: Article in Journal/Newspaper
• Language: English
• Published: 2012
• Subjects: Adolescent; Adult; Age Factors; Cause of Death; Child; Preschool; Cohort Studies; Female; Finland/epidemiology; Follow-Up Studies; Humans; Iceland/epidemiology; Infant; Newborn; Male; Neoplasms/epidemiology; Scandinavian and Nordic Countries/epidemiology; Survivors; Time Factors; Young Adult; Norway; Iceland
• Online Access: https://pure.au.dk/portal/en/publications/939b8eb6-1086-4650-be4d-790356220f22; https://doi.org/10.1002/ijc.27393

Long-term survivors of childhood cancer suffer from a higher mortality than the general population. Here we evaluate late and very late mortality, and patterns of causes of death, in 5-year survivors after childhood and adolescent cancer in cases diagnosed during four decades in the five Nordic countries. The study is population-based and uses data of the nationwide cancer registries and the cause of death registers. There were in all 37,515 incident cases, diagnosed with cancer before the age of 20 years, between 1960 and 1999. The 5-year survivor cohort used in the mortality analyses consisted of 21,984 patients who were followed up for vital status until December 31, 2005 (Norway, Sweden) or 2006 (Denmark, Finland, Iceland). At the latest follow-up, 2,324 patients were dead. The overall standardized mortality ratio was 8.3 and the absolute excess risk was 6.2 per 1,000 person-years. The pattern of causes of death varied markedly between different groups of primary cancer diagnosis, and was highly dependent on time passed since diagnosis. With shorter follow-up the mortality was mainly due to primary cancer, while with longer follow-up, mortality due to second cancer and noncancer causes became more prominent. Mortality between 5 and 10 years after diagnosis continued to decrease in patients treated during the most recent period of time, 1990-1999, compared to previous periods, while mortality after 10 years changed very little with time period. We conclude that improvement of definite survival demands not only reducing early but also late and very late mortality.