Defining the optimal therapy for rectal cancer.

For physicians trained in the United States, postoperative combined modality therapy has become the accepted standard for patients with rectal cancer who are at high risk for locoregional and distant recurrence. In 1990, the results of two randomized trials demonstrating improved survival for postoperative chemotherapy with radiation therapy compared with either surgery alone or with postoperative radiation therapy alone led to an National Institutes of Health Consensus Conference recommendation that all patients with stage II or III rectal cancer should receive postoperative combined modality therapy. Two subsequent intergroup trials have retained control arms of postoperative radiation therapy with 5-fluorouracil-based chemotherapy, perpetuating the role of such treatment in clinical practice. However, there is not universal agreement that high-risk patients require all three modalities of therapy—surgery, radiation therapy, and chemotherapy—or even the proper order of such therapy. There is widespread use of preoperative radiation therapy given without chemotherapy in Europe. At least one clinical trial, designed by the European Organization for Research and Treatment of Cancer, is testing the benefit of both preoperative chemotherapy and postoperative chemotherapy when given concurrently or sequentially with radiotherapy, with preoperative radiation alone as the standard treatment arm. Independent of the timing of treatment, it has been a historical fact that most studies have typically asked whether chemotherapy adds to the benefit of adjuvant radiation therapy in high-risk rectal cancer patients. The acceptance of radiation therapy as an additional local therapy after potentially curative surgery is primarily based on clinical observations of the risk of local recurrence and on the significant morbidity of locoregional failure. More than 50 years ago, second-look procedures were instituted in patients who were thought to be at high risk of recurrence after surgery for rectal cancer. Predating currently available radiographic techniques, these procedures allowed identification of local failures and regional lymph node metastases as common initial sites of recurrence (1). Although it was recognized that many patients with initial locoregional recurrence subsequently developed distant metastases as well, the rationale was established for additional local therapy in the form of either preoperative or postoperative radiation therapy. Save for a single randomized trial of preoperative radiation therapy, controlled trials have demonstrated either reduction in local recurrence with postoperative radiation therapy or tumor and lymph node downstaging with preoperative radiation therapy but no improvement in overall survival (2). Although it is accepted that a beneficial outcome of adjuvant treatment of rectal cancer is prevention of clinically relevant local recurrences, most clinicians and patients would agree that overall survival and quality of life are more pertinent end points of treatment. Balancing the potential benefits in reduction of locoregional failure alone after combined modality therapy are the shortand long-term toxic effects of such treatment. Qualityof-life assessments have suggested that the increased complications of postoperative radiation therapy and chemotherapy may be offset by improved recurrence and survival rates when compared with postoperative radiation therapy alone (3). However, retrospective analyses of patients treated with postoperative chemotherapy and radiation therapy suggest that such treatment may also be associated with major long-term alterations in bowel function, an observation that can be supported by any clinician listening carefully to the symptoms of patients who have been treated with the current standard of care (4). It is possible that altering the sequence of treatment may reduce these late morbidities, although two attempts to directly compare preoperative with postoperative chemoradiation in the United States have failed, in large part because of patient or physician preferences and biases. Since it is likely that most of the long-term morbidity of postoperative chemotherapy and radiation therapy is due to the late effects of the radiation therapy, it is reasonable to ask whether all patients should be exposed to this treatment and whether similar clinical outcomes could be achieved with improved surgical techniques or with systemic chemotherapy alone. The establishment of the efficacy of adjuvant chemotherapy for highrisk extrapelvic colon cancer has strengthened the desire to reverse the original combined modality question and to ask whether radiation therapy adds to the benefit from adjuvant chemotherapy. In this issue of the Journal, the National Surgical Adjuvant Breast and Bowel Project (NSABP) reports the mature results of the R-02 trial, which compared postoperative chemotherapy alone with postoperative chemotherapy and radiation therapy (5). This study design was derived from the first NSABP rectal adjuvant trial, which compared surgery alone with postoperative chemotherapy or with postoperative pelvic radiation therapy (6). That trial engendered much discussion, as much for the primary results of the trial as for the exploratory analyses of the effect of treatment based on sex and age. The primary end point for survival showed a marginally significant improvement in overall survival (P 4. 05) for chemotherapy alone compared with surgery. Although radiation therapy was associated with a reduction in locoregional recurrence (25% for surgery alone versus 16% for postoperative radiation therapy; P 4 .06), there was no significant benefit in disease-free or overall survival. Subset analysis of these data suggested that younger males were the primary beneficiaries of the effects of adjuvant chemotherapy with 5-fluorouracil, vincristine, and semustine. These results led to the study design of R-02, in which sex-