Incidence of and factors related to late complications in conformal and conventional radiation treatment of cancer of the prostate.

PURPOSE: The fundament hypothesis of conformal radiation therapy is that tumor control can be increased by using conformal treatment techniques that allow a higher tumor dose while maintaining an acceptable level of complications. To test this hypothesis, it is necessary first to estimate the incidence of morbidity for both standard and conformal fields. In this study, we examine factors that influence the incidence of late Grade 3 and 4 morbidity in patients treated with conformal and standard radiation treatment for prostate cancer. METHODS AND MATERIALS: Six hundred sixteen consecutive patients treated with conformal or standard techniques between 1986 and 1994 to doses greater than 65 Gy and with more than 3 months follow-up were analyzed. No patients treated with prostatectomies were included in the analysis. The conformal technique includes special immobilization by a cast, careful identification of the target volume in three dimensions, localization of the inferior border of the prostate using a retrograde urethrogram, and individually shaped portals that conform to the Planning Target Volume (PTV). Multivariate analysis using a proportional hazards model compares differences in the incidence of Radiation Therapy Oncology Group/European Organization for Research and Center Treatment (RTOG/EORTC) Grade 3 and 4 late gastrointestinal (GI) and genitourinary (GU) morbidity by technique, T-stage, grade, age, hormonal treatment, irradiated volume, dose, and comorbid conditions. Grade 3 rectal bleeding was defined as requiring three or more cautery procedures. RESULTS: The overall actuarial incidence of genitourinary (GU) toxicities at 5 years was 3.4%, with the crude incidence being six cases in 616 patients satisfying the selection criteria; for gastrointestinal (GI) toxicities, the overall actuarial incidence was 2.7%, with the crude incidence being 13 cases out of 616 patients. The average time to complication for our patients was 12.8 months for GI toxicity and 32.9 months for GU toxicity (p < 0.001). No factors were found that were predictive for GU morbidity. The only factors significantly related to incidence of late GI morbidity on multivariate analysis of our data were dose and age. The central axis dose was a more significant variable than the dose prescribed to the Treated Volume. Age was negatively correlated with late GI morbidity, with older patients having a reduced incidence of toxicity. The median tolerance dose for GI complications was estimated to be 92.8 Gy, and the dose for 10% incidence was estimated to be 80.2 Gy. Treating the pelvis to 45 Gy did not increase the incidence of late morbidity. Late GI and GU toxicities were not correlated. CONCLUSION: The conformal technique has been associated with fewer acute Grade 2 toxicities (6). The use of conformal fields did not decrease the incidence of late GI morbidity; however, patients with this technique invariably had higher doses. Because of the dose response for this complication and the correlation between the dose and the use of conformal fields, one would not expect to demonstrate an advantage to conformal fields in this data set. On the other hand, no dose effect was observed for late GU morbidity. In this case, there appears to be an advantage for conformal treatment that has not reached statistical significance because the follow-up time is shorter than for the patients treated with conventional fields and the latency for GU morbidity is long.

Duke Scholars

Author W. Robert Lee Radiation Oncology