Estimation of the incidence of late bladder and rectum complications after high-dose (70-78 Gy) conformal radiotherapy for prostate cancer, using dose-volume histograms

Purpose: To investigate whether Dose-Volume Histogram (DVH) parameters can be used to identify risk groups for developing late gastrointestinal (GI) and genitourinary (GU) complications after conformal radiotherapy for prostate cancer. Methods and Materials: DVH parameters were analyzed for 130 patients with localized prostate cancer, treated with conformal radiotherapy in a dose-escalating protocol (70-78 Gy, 2 Gy per fraction). The incidence of late (>6 months) GI and GU complications was classified using the RTOG/EORTC and the SOMA/LENT scoring system. In addition, GI complications were divided in nonsevere and severe (requiring one or more laser treatments or blood transfusions) rectal bleeding. The median follow-up time was 24 months. We investigated whether rectal and bladder wall volumes, irradiated to various dose levels, correlated with the observed actuarial incidences of GI and GU complications, using volume as a continuous variable. Subsequently, for each dose level in the DVH, the rectal wall volumes were dichotomized using different volumes as cutoff levels. The impact of the total radiation dose, and the maximum radiation dose in the rectal and bladder wall was analyzed as well. Results: The actuarial incidence at 2 years for GI complications greater than or equal to Grade II was 14% (RTOG/EORTC) or 20% (SOMA/LENT); for GU complications greater than or equal to Grade III 8% (RTOG/EORTC) or 20% (SOMA/LENT). Neither for GI complications greater than or equal to Grade II (RTOG/EORTC or SOMA/LENT), nor for GU complications greater than or equal to GradeIII (RTOG/EORTC or SOMA/LENT), was a significant correlation found between any of the DVH parameters and the actuarial incidence of complications. For severe rectal bleeding (actuarial incidence at 2 years 3%), four consecutive volume cutoff levels were found, which significantly discriminated between high and low risk. A trend was observed that a total radiation dose greater than or equal to 74 Gy (or a maximum radiation dose in the rectal wall >75 Gy) resulted in a higher incidence of severe rectal bleeding (p = 0.07). Conclusions: These data show that dose escalation up to 78 Gy, using a conformal technique, is feasible. However, these data have also demonstrated that the incidence of severe late rectal bleeding is increased above certain dose-volume thresholds. (C) 1998 Elsevier Science Inc.