gms | German Medical Science

Background: The German Childhood Cancer Registry records all childhood cancer cases in Germany since 1980. A long-term follow has been established with increasing coverage. Reports show that survival has improved considerably since the 1980ies, particularly short term survival up to 15 years after diagnosis. Reports from other countries, especially European countries, show, that as patients reach middle age and beyond, the mortality is consistently higher than in the general population, which becomes particularly noticeable as former patients age. We set out to compare German data to the European experience.

Material: We are including diagnoses by ICCC-3 (International Childhood Cancer Classification) from 1981-2013 in Germany with follow-up until 31.12.2013. These are 52,267 patients with 1,1473 observed deaths, providing 554,979 person years (PY) for analysis. Up to 0.2% of age, sex, birthdate and death date information was missing and imputed. First cancers included were diagnosed at age under 15 years, patients were observed up until the age of 47. Mean age of the survivors at the end of their respective follow-up was 19.

Methods: We calculated the mortality rates, stratified by the covariates of interest. Based on mortality in the general population (1), we obtained absolute excess rates and SMR to assess the long-term effect on mortality relative to the general population. Joint effects of relevant covariates were modelled using Poisson regression.

Results: Mortality rates are highest in the five years following diagnosis and thus in persons under 20 years old. Overall, the mortality rate is lowest, though still more than 6 times higher than expected, when patients are in their twenties, and increases afterwards with increasing age. As general mortality starts to rise after about 40 years of age, the excess mortality after this is still considerable (6.7 per 1000 person years for patients in their forties). The mortality rate differs little between males and females. But as the background mortality of females is considerably lower, the SMR for females is far higher than for males (e.g. 9.1 versus 4.2 at 20-24 years from diagnosis). The mortality rate within 5 years of diagnosis dropped considerably over the study period (from 73/1000 PY to 46/1000 PY) in line with treatment improvements, but the mortality rates and SMR after more than 10 years of follow-up hardly changed and stayed at about 5 times the population mortality.

Discussion: The data clearly show that special care for former childhood cancer patients is needed all their life, their risks do not go back to “normal” even decades after the primary therapy. The lack of improvements in later age mortality since the 1980ies shows that more efforts in late effects care are needed, especially as international data shows that many of these deaths are from diseases which respond well to management, such as heart failure and respiratory diseases.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.