Populationsbasierte Anteile geheilter Krebspatientinnen und -patienten in Deutschland

Abstract

Background: Population-based assessment of long-term survival prospects of cancer patients in Germany have, so far, been mainly based on estimating relative 5- or 10- year survival rates. Alternatively, parametric cure models allow for determining the proportion of patients who although being affected by cancer do not exhibit higher mortality rates compared to the general population. These patients are thus considered as cured from cancer. Internationally, cure models are highly established in the monitoring of long term survival of cancer patients. Methods: For the present analysis, population-based cancer registry data from patients aged 15 to 79 years (N=453,666) in Germany were used. In order to obtain up-to-date estimates on cure proportions of cancer patients, period analyses with a period window from 2013 to 2014 and a maximal follow-up time of 10 years were conducted. Based on parametric modelling of survival times, sex-specific estimates on cured proportions of cancer patients were calculated for all cancers as well as for the three most common cancer sites among women (breast, colon, lung) and among men (prostate, lung, colon) were calculated. In additional analyses, the extent to which the cured proportion of all cancer patients might have been overestimated due to the exclusion of DCO (Death Certificate Only) cases were quantified.Results: Overall, 64.0% (95%-confidence interval; 63.4–64.6) of women and 56.8% (56.3–57.4) of men were considered as statistically cured of their cancer in the sense that their mortality did not differ significantly from that of the general population. Parametric cured models also converged in all site-specific analyses except breast cancer among women. The cure proportions for colon and lung cancer were higher in women than in men (62.3% vs. 56.2% and 18.5% vs. 14.9%). Moreover, the estimated cure proportion of prostate cancer among men was 94.5%. After correcting the cure fraction of all cancers for the proportion of DCO cases, a maximal drop of 3% was observed among both genders.Conclusions: Based on nation-wide data for the period window 2013–2014, significant sex difference in long-term cancer survival for common cancers can be observed in Germany. Parametric cured models can provide additional insights on remaining or new challenges of cancer care in Germany.

Keywords

• cancer
• cure fraction
• relative survival
• epidemiological cancer registry
• Germany