Tumour–node–metastases (TNM) category distribution can be used as a short-term surrogate measure for survival probability to evaluate the impact of interventions for improved prevention and early detection of cancer. We aimed to demonstrate whether or not an association between registration completeness and TNM category distribution exists and whether accurate estimates of the TNM category distribution are possible before complete registration. With respect to the diagnosis year 2002, the number of registered cases of prostatic and testicular cancer at the Bavarian cancer registry was determined every 6 months between 2003 and 2006 together with the respective TNM category distributions and the completeness of registration. For the diagnosis year 2002, 6135 and 480 cases of newly developed prostatic and testicular cancer, respectively, were reported to the registry. After 2003, the proportion of prostatic cancer cases labelled ‘T1’ fluctuated by only 1%. A stable TNM category distribution was seen with a minimum registration completeness of 65–70%. No relevant association between registration completeness and TNM category distribution could be demonstrated for prostatic and testicular cancer. Therefore, in prostatic and testicular cancer, the TNM category distribution can be used for short-term evaluation of quality assurance projects and in health services research.