The aim of this study was to evaluate and compare the rates of grade 2 or worse late effects in patients treated for prostate cancer on Radiation Therapy Oncology Group (RTOG) 9406. The authors previously have reported the results of patients treated on the first 2 dose levels of this study with respect to grade 3 or greater late toxicity. This analysis examines the incidence of grade 2 toxicity in this study. From August 1994 to September 1999, 424 patients were entered on this dose escalation trial of 3-dimensional conformal radiation therapy (3D CRT) for localized adenocarcinoma of the prostate at doses of 68.4 Gy (level I) and 73.8 Gy (level II). All radiation prescriptions were a minimum dose to a planning target volume. Patients were stratified according to clinical stage and risk of seminal vesicle invasion based on Gleason score and presenting prostate-specific antigen. Average time at risk after completion of therapy ranged from 33.1 to 40.1 months for patients treated at dose level I and 15.6 to 34.2 months for patients at dose level II. The frequency of late effects ≥ grade 2 was compared with a similar group of patients treated on RTOG studies 7506 and 7706 with adjustments made for the interval from completion of therapy. The RTOG toxicity scoring scales for late effects were used. The rate of grade 3 or greater late toxicity continues to be low compared with RTOG historical controls. No grade 4 or 5 late complications were reported in any of the 406 evaluable patients during the period of observation. Interestingly, the incidence of grade 2 late toxicity was increased relative to historical controls in all groups and dose levels. In group 1, level I and group 3, level II, the increase in grade 2 complications was statistically significant; 16 complications were observed in group 1, level I when 9.2 were expected (P = .026) and 22 were observed in group 3, level II when 7.6 were expected (P < .0001). When examining all late effects ≥ grade 2, there were no significant differences in the rate of late effects in both groups and both dose levels with the exception of group 1, level II. This, in combination with the statistically significant decrease in late effects ≥ grade 3, suggests that in most circumstances there has been a shift of grade 3 complications to grade 2. In group 1, dose level II there was a statistically significant reduction in ≥ grade 2 late effects, suggesting there was no shift from grade 3 to grade 2 in these patients. In this circumstance there may have been a global reduction in all complications or a shift to late effects less severe than grade 2. In group 2, dose level II there is a trend (P = .085) toward this same result. It is important to continue to examine late effects closely in patients treated on RTOG 9406. The primary objective of dose escalation without an increase rate of ≥ grade 3 complications has been achieved. However, the reduction in grade 3 complications may have resulted in a higher incidence of grade 2 late effects. Because grade 2 late effects may have a significant impact on a patient's quality of life, it is important to reduce these complications as much as possible. Improved conformal treatment delivery with intensity-modulated radiation therapy or the use of radioprotective agents could be considered. Clinical trials should use quality-of-life measures to determine that trade-offs between severity and rates of toxicity are acceptable to patients.