SPECIAL EDITORIAL SERIES- STATISTICAL ISSUES IN CANCER RESEARCH Quality of life assessment in clinical cancer research

Classical end-points in clinical therapeutic trials in oncology are usually defined by total, recurrence-free or systemic disease-free survival. Primarily, these allow an adequate de-scription of the biological course of the disease. In current oncological research advances in the treatment of cancer patients as measured by these end-points must realistically be assumed to be only of a rather small magnitude. This implies, on the one hand, that prospectively only by large-scale multi-centre trials and retrospectively only by meta-analyses of comparable trials can sufficient numbers of patients be reached to allow a detection of treatment benefits of a realis-tic size. On the other hand, if treatments do not differ much with respect to survival, it is a plausible step forward to extend the classical criteria of assessing treatment efficacy. It is undisputed that the disease and its treatment have a