| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIALS |
From the Division of Colon and Rectal Surgery, Rochester, Minnesota.
Correspondence: Address correspondence to: Heidi Nelson, MD, Division of Colon and Rectal Surgery, Mayo Clinic, 200 First St. SW, Rochester, MN 55905; Fax: 507-284-1794; E-mail: nelson.heidi{at}mayo.edu
Routine use of adjuvant or neoadjuvant chemoradiation for all stage II and III rectal adenocarcinomas likely represents overtreatment. Radiation therapy for high-risk rectal cancer to reduce rates of local recurrence and cancer-related deaths has become standard. That radiation therapy is associated with both short-term and long-term morbidity and posttreatment dysfunctions supports the avoidance of such morbidities when the benefits are marginal. Increasing evidence, however, indicates that it is possible to be more selective in the application of radiation therapy, especially for T1–2N1 and T3N0 tumors. We propose that it is critical to treat, but equally important not to overtreat, patients with rectal cancer.
Although surgery remains the primary curative modality in the treatment of rectal cancer, its limitation as a single curative modality has long been recognized. Historically, local recurrence rates after proctectomy for rectal cancer have been high, which prompted trials investigating the impact of adjuvant chemotherapy and radiation therapy (e.g., the Mayo/North Central Cancer Treatment Group [NCCTG] 7947511 and Gastrointestinal Tumor Study Group [GITSG] 7175.2) With time, it became clear that postoperative radiation and chemotherapy significantly reduced local recurrence and improved survival rates when compared with surgery alone, leading to the National Institutes of Health (NIH) consensus conference statement in 1990.3 This consensus conference concluded that all patients with tumor, node, metastasis (TNM) stage II and III rectal cancer should be treated with postoperative radiotherapy and chemotherapy. Perhaps, as a result of these breakthrough improvements in cancer outcomes, considerable clinical research activity has ensued in this field during the last 13 years. Accordingly, it seems prudent to review and reconsider these consensus recommendations in light of current knowledge. For example, surgery in the previous adjuvant trials was not standardized and local recurrence rates with surgery alone were high. Furthermore, patients selected for these trials represented a somewhat biased population, with only select patients with stage II and stage III cancer (those at high risk for local and distant recurrence) considered for enrollment. We now know that the surgeon is an important prognostic variable and that patients with stage II and stage III cancer have variable risk according to both "T" and "N" classification. It is time to determine if new practice guidelines are warranted, taking into the equation improvements in surgery and risk classification.
The critical contribution of proper oncologic surgery to outcomes in rectal cancer has recently come to the forefront with the focus principally on mesorectal excision, radial margins, and lymphadenectomy. Although no prospective, randomized trial has compared total mesorectal excision (TME) with conventional proctectomy, both single institution series and at least one multicenter trial have demonstrated excellent local control rates with surgery alone. The Dutch TME study4 reported that, with appropriate training, the TME technique could achieve local recurrence rates as low as 8.2% with surgery alone, which is consistent with rates of 4% to 8%5,6 from single institution reports. Just as adequate clearance of the distal bowel margin helps prevent anastomotic recurrences,7 it is now known that clearance of the radial margin is critical for the prevention recurrences of pelvic cancers. Unlike the bowel margin, the radial margin can be involved despite proper wide oncologic resection (i.e., in the setting of a large bulky advanced pelvic tumor). A positive radial margin is associated with high recurrence rates of 66% to 85%8,9; the margin can be involved from inadequate surgical clearance or from an extensive, deeply penetrating tumor in the setting of widest possible surgical clearance. Finally, increasing emphasis is being placed on the adequacy of lymph node harvest. When a tumor is considered node negative and fewer than 14 nodes are examined, both staging accuracy and survival outcomes are compromised.10 The lack of surgical standardization and limitations of surgical information in previous rectal cancer adjuvant studies is well documented.11 The question remains: Could "good" risk tumors be treated without radiation in the setting of "good" surgery? We now know more about what constitutes good surgery. What do we know about good risk tumors?
Emerging evidence, in fact, indicates that not all stage II and stage III tumors behave in the same manner and good risk tumors can be identified. A better understanding of risk stratification comes from the work of Gunderson et al.12 who performed a pooled analysis of the NCCTG 794751, NCCTG 864751, and Intergroup (INT) 114 rectal adjuvant trials. Using this analysis, they determined that intermediate risk tumors (T1–2N1, T3N0) compared with moderately high risk (T1–2N2, T3N1, T4N0) and high risk (T3N2, T4N1–2) tumors had far lower local recurrence rates (6% to 8% vs. 8% to 15% and 15% to 22%) and improved overall survival rates (74% to 81% vs. 61% to 69% and 33% to 48%), respectively. When Tepper et al.13 reported the final analysis of the Intergroup 0114 study, they found similar results and concluded that routine adjuvant radiation therapy may not be required for T1–2N+ or T3N0 tumors located high in the rectum when TME with negative radial margins and appropriate lymphadenectomy had been performed. Drawing firm conclusions from these data is limited because all patients received radiation therapy. To examine this same issue from a different perspective, Gunderson et al.14 performed another pooled analysis, this time including five phase III North American trials (NCCTG 794751, NCCTG 864751, INT 114, National Surgical Adjuvant Breast and Bowel Project [NSABP] R01, NSABP R02). This allowed a comparison of outcomes according to stage subsets and according to six different treatment regimens: surgery alone, surgery and radiation, surgery, radiation, and chemotherapy (three different chemotherapy regimens), or surgery and chemotherapy. This revealed no additional benefit in overall or disease-free survival with the addition of radiation to chemotherapy after surgery for T1–2N1 and T3N0 tumors. With surgery and chemotherapy, without radiation therapy, the 5-year overall survival for patients with T1–2N1 tumors was 85% and for T3N0 it was 84%.14 Based on these pooled data from prospective randomized trials, it seems reasonable to propose that patients with T1–2N1 and T3N0 disease might best be treated with surgery and chemotherapy without radiation therapy.
The time seems right for a formal investigation of the possibility that patients with T1–2N1 and T3N0 tumors might be treated with surgery and chemotherapy without radiation therapy. At least two large pooled analyses would support this approach and we propose a prospective trial. The most logical schema for such a trial would require that patients with clinically staged T1–2N1 and T3N0 tumors have surgery first, followed by chemotherapy alone or radiation and chemotherapy. This would provide two equal groups of patients with pathologically staged tumors. Several challenges to such a study exist, fortunately none seems prohibitive.
The first difficulty regards the availability and accuracy of preoperative staging. Depending on the quality of magnetic resonance imaging (MRI) and endorectal ultrasound (ERUS) preoperative staging, some patients with cancer clinically staged as T1–2N1 and T3N0 will be found at surgical pathology, in fact, to have stage I disease, requiring no further treatment; others will have more advanced stage disease, requiring standard postoperative adjuvant therapy. Indeed, more patients will have to be considered for the trial before surgery than can be enrolled for randomization after surgery. This should be a minor problem overall.
The second difficulty likely will be the problem of recruiting investigators to participate if they prefer neoadjuvant therapy. A study design incorporating neoadjuvant therapy would have to compare chemotherapy alone with radiation plus chemotherapy before surgery. It is doubtful that investigators or patients would accept a delay in surgery for 2 to 4 months for the administration of chemotherapy alone. Considering the importance of the question and the fact that the superiority of preoperative versus postoperative chemoradiation has not been established, we would hope that investigators could accept this minor and temporary change in practice to advance this field.
In summary, the time seems right to move ahead with a prospective, randomized comparison of surgery and chemotherapy versus surgery, radiation therapy, and chemotherapy for patients with T1–2N1 and T3N0 rectal cancer. Analyses of pooled data sets from past adjuvant studies suggest that we have reached a state of equipoise regarding the management of intermediate risk rectal cancers. Because we now have the tools to monitor surgical quality and to provide accurate preoperative staging, it is time to address this question.
Received for publication November 4, 2003. Accepted for publication November 24, 2003.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
