ATLANTA-In a phase III trial of patients who had undergone surgery for rectal cancer, three regimens of fluorouracil (5-FU)-based chemotherapy and radiation therapy achieved similar rates of relapse-free and overall survival, although the three regimens had somewhat differing toxicity profiles. Lead author Stephen R. Smalley, MD, a radiation oncologist at the Olathe Regional Oncology Center, Olathe, Kansas, reported the results at the 46th Annual Meeting of the American Society for Therapeutic Radiology and Oncology (abstract 14). Recent trials have found that after surgery for rectal cancer, protracted venous infusion (PVI) of 5-FU during radiation therapy improves outcomes relative to bolus infusion (Intergroup 864751), and outcomes are similar with a variety of different biochemically modulated bolus regimens of 5-FU (Intergroup 0114), Dr. Smalley said. The new trial (Intergroup 0144), therefore, sought to answer three questions, he said: "Number one, would a protracted course of 5-FU (prior to and following protracted 5-FU and radiation) lead to further improvement in outcome? Number two, could a biochemically modulated 5-FU program without central venous catheters produce outcomes similar to those of the protracted venous infusion arm? And number three, how would pelvic control be affected, especially in those groups that were most rationally treated with initial surgery?" The investigators enrolled patients who had undergone complete resection of locally advanced but nonmetastatic rectal adenocarcinoma (T3-4, N0, M0 or T1-4, N1-2, M0) in the prior 20 to 70 days; dentate involvement was allowed. Patients were required to have adequate organ function and no prior chemotherapy or radiation therapy. The patients were stratified by type of resection (abdominoperineal vs low anterior), T stage, N stage, and time from surgery. They were then assigned to three treatment arms:

• Arm 1: Bolus 5-FU before and after radiation therapy, with 5-FU by PVI during radiation therapy
• Arm 2: 5-FU by PVI before, during, and after radiation therapy
• Arm 3: Bolus 5-FU, leucovorin, and levamisole before and after radiation therapy, plus bolus 5-FU and leucovorin during radiation therapy

In all, 1,917 patients were enrolled in the trial, and living patients had a median follow-up of 6.1 years, making the data mature, Dr. Smalley noted. Study Results The three treatment arms had statistically indistinguishable rates of overall survival (81% to 83%) and relapse-free survival (67% to 69%) at 3 years follow-up, Dr. Smalley said. For both endpoints, the findings were the same in analyses comparing arm 1 with arm 2, and arm 1 plus arm 2 with arm 3. (See Table 1 for 5-year survival rates.) "It's important to evaluate pelvic control, especially for those patients who are initially appropriately managed by surgery," Dr. Smalley said, noting that some trials have found better pelvic control with preoperative radiation therapy, with or without chemotherapy. "However, preoperative radiotherapy does produce an increased clinically meaningful toxicity, both gastrointestinal and sexual in nature, and it is obviously desirable to avoid these side effects in those who are unlikely to benefit from radiation therapy. This would certainly include patients who are candidates for sphincter- sparing surgery when they present, as well as patients who do not have fixed primary rectal lesions." With the 6.1-year median followup, the rate of pelvic failure was similarly low across treatment arms in the entire study population (5% to 7%) and in the subgroup of patients who were appropriately managed with initial surgery, namely those with non- T4b disease who underwent low anterior surgical resection (3% to 6%). Toxicity Patients in arms 1, 2, and 3 had similar rates of treatment-related death (approximately 1%) and gastrointestinal toxicity of grade 3 or higher (42% to 46%), Dr. Smalley said. In contrast, rates of grade 3 or higher hematologic toxicity differed, with this complication occurring in about half of the patients in arms 1 and 3 (50% to 56%) but in few patients in arm 2 (4%). "However this was primarily a laboratory change because the PVI arm [arm 2] had a 6% grade 3-5 infection rate vs 10% to 11% in the two bolus 5-FU arms," Dr. Smalley noted. As expected, catheter-related toxicity was higher in arms 1 and 2 (2% to 3%) than in arm 3 (less than 0.5%), he added. 'Similar Outcomes' "We conclude that all three dose schedules lead to similar outcomes following resection," Dr. Smalley commented. "Hematologic toxicity was definitely decreased with protracted venous infusion, but it must be balanced against the cost, inconvenience, and risk of catheter-related toxicities with PVI." He told ONI that although other recent trials have found PVI to be superior to bolus therapy, "our trial is by far the biggest and most satisfactorily powered study, and really suggests that any 5-FU regimen plus radiation produces similar survival outcomes." Less Extensive Treatment The findings further suggest that less extensive treatment is safe in some patients, Dr. Smalley pointed out. "Initial surgical management of nonfixed lesions that are amenable to sphincter- preserving surgery at presentation is entirely justified," he said. "Those patients who are at low risk of pelvic recurrence following surgery-certainly this would include T1-2, N0 patients, potentially even selected T3, N0 patients-can avoid the toxicity of radiation."