PURPOSE: To demonstrate the feasibility of preoperative Hyperfractionated Accelerated RadioTherapy (preop-HART) in rectal cancer and to explain the rationales to switch from postoperative HART to preoperative HART. METHODS AND MATERIALS: Fifty-two consecutive patients were introduced in successive Phase I trials since 1989. In trial 89-01, postoperative HART (48 Gy in 3 weeks) was applied in 20 patients. In nine patients with locally advanced rectal cancer, considered unresectable by the surgeon, 32 Gy in 2 weeks was applied prior to surgery (trial 89-02). Since 1991, 41.6 Gy in 2.5 weeks has been applied preoperatively to 23 patients with T3-T4 any N rectal cancer immediately followed by surgery (trial 91-01). All patients were irradiated at the department of radiation-oncology with a four-field box technique (1.6 Gy twice a day and with at least a 6-h interval between fractions). The minimal accelerating potential was 6 MV. Acute toxicity was scored according to the World Health Organization (WHO for skin and small bowel) and the Radiation Therapy Oncology Group criteria (RTOG for bladder). This was done weekly during treatment and every 3 months thereafter. Small bowel volume was estimated by a modified "Gallagher's" method. RESULTS: Acute toxicity was acceptable both in postoperative and preoperative setup. The mean acute toxicity was significantly lower in trial 91-01 compared to 89-01. This difference was due to the smaller amount of small bowel in irradiation field and lower total dose in trial 91-01. Moreover, there was a significantly reduced delay between surgery and radiotherapy favoring trial 91-01 (median delay 4 days compared to 46 days in trial 89-01). Nearly all patients in trial 89-02 and 91-01 underwent surgery (31 out of 32; 97%). Resection margins were negative in 29 out of 32. Hospitalization duration in trial 91-01 was not significantly different from trial 89-01 (19 vs. 21 days, respectively). CONCLUSIONS: Hyperfractionated accelerated radiotherapy immediately followed by surgery is feasible as far as acute toxicity is concerned. Preoperative HART is favored by a significantly lower acute toxicity related, in part, to a smaller amount of irradiated small bowel, and a shorter duration of the delay between radiotherapy and surgery. Moreover, the hospital stay after preoperative HART is not significantly increased.