This study was based on 528 consecutive patients who were scheduled for any kind of a curative radiotherapy treatment (external beam irradiation, temporary or permanent brachytherapy) due to cT1-3N0M0 prostatic carcinoma in the years 2003-2007. All patients answered a validated questionnaire, the Expanded Prostate Cancer Index Composite (EPIC), before the treatment [11
]. The questionnaire was handed over to the patients personally by one of the physicians. It comprises 50 items concerning the urinary, bowel, sexual and hormonal domains for function and bother. The multi-item scale scores were transformed lineary to a 0-100 scale, with higher scores representing better health-related quality of life. For a detailed analysis, questions in the urinary domain were classified into incontinence and irritation/obstruction subscales [11
Data has been acquired in compliance with the Helsinki Declaration. The evaluation was approved by the local ethics committee. Informed consent was obtained from patients for the treatment. Patients were informed about the voluntary participation in the evaluation of quality of life data. An explicit statement about the voluntary character of the participation was heading every page of the questionnaire.
The EPIC was administered in German. It has been translated in the Department of Radiation Oncology and validated in the department of Medical Psychology and Medical Sociology of our institution [13
]. Several studies have been already published [14
]. These studies very well support the sensitivity of the questionnaire for quality of life changes after treatment in all domains. The relation between individual questions and quality of life scores in different domains as well the correlation of function and bother domains or changes with time have been elaborated in detail. In accordance with data published by Osoba et al. [21
], mean HRQOL changes of below 5 points were defined as clinically not significant. For patients who indicated "a little" change, the mean change in scores was about 5 to 10, for "moderate" change, about 10 to 20; and for "very much" change, greater than 20. The EORTC (European Organization for Research and Treatment of Cancer) QLQ (Quality of Life Questionnaire)-C30, evaluated by Osoba et al., uses comparably to the EPIC a 0 to 100 scale. Though the assessment of a clinically significant threshold remains controversial, a 5-point threshold proved likewise to be clinically significant in our own experience [14
The group has been divided into the age groups up to 65 (n = 116), 66-70 (n = 144), 71-75 (n = 158) und >75 years (n = 110) with the aim to discriminate purely age-related health-related quality of life (HRQOL) changes. Five-year increments have been estimated to be relevant intervals with an adequate patient number in each group to detect significant age-related differences. The Charlson Comorbidity Index (CCI) was calculated as a scale that considers both comorbidities and the patient age (starting at 50 years of age, each decade is counted as an extra point) [22
Neoadjuvant hormonal therapy (NHT) was administered in 192 cases. In a preliminary analysis, mean HRQOL scores were found to be significantly lower in the sexual (function scores: -18 points) and hormonal (function scores: -11 points) domains. Therefore, all evaluations of sexual and hormonal symptoms and scores were limited to the group of patients without NHT (n = 336; i.e. prostate cancer not treated with any method before).
Statistical analysis was performed using the SPSS 14.0 (SPSS, Chicago, Ill), software. To explore statistical HRQOL score differences between different subgroups, the Mann-Whitney-U-test was used. Contingency table analysis with the chi-square test was performed to compare treatment groups with respect to categorical variables. In a logistic forward stepwise multivariate analysis, different risk factors were tested for their independence. For the evaluation of age and comorbidities, multivariate analysis was performed twice: 1. considering age and the presence of any comorbidity; 2. considering age and various specific comorbidities. As the CCI scale is considering both age and comorbidities, it cannot be considered in this multivariate analysis. Hazard ratios in dependence on the CCI scale were therefore reported separately. The global significance level for all statistical test procedures conducted was chosen as α = 5%. All statistical analyses were conducted in an explorative manner. Thus, with consideration of the explorative character of the analysis, p-values < 0.05 can be interpreted as statistically significant test results.