| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Editorial |
1 Department of Surgery, Division of Surgical Oncology, UNC/Lineberger Comprehensive Cancer Center and the University of North Carolina School of Medicine at Chapel Hill, CB#7213, 101 Manning Drive, Chapel Hill, North Carolina 27599–7213
2 Department of Radiation Oncology, UNC/Lineberger Comprehensive Cancer Center and the University of North Carolina School of Medicine at Chapel Hill, CB #7512, 101 Manning Drive, Chapel Hill, North Carolina 27599-7512
Correspondence: Address correspondence and reprint requests to: Joel E. Tepper, MD; E-mail: tepper{at}med.unc.edu.
It has long been held that chemotherapy and radiation are adjuncts to radical en-bloc surgical resection in the treatment of rectal cancer, that surgery is the most important modality of treatment, and, in early-stage disease, that surgery is the only treatment needed. In an effort to lessen the morbidity and potential lifestyle-altering effects of radical surgical resection, other therapeutic approaches have been advocated. Increasing evidence has suggested that chemoradiation followed by less radical (sphincter-sparing) surgery, or no surgery at all, may be adequate treatment, even for T2 and T3 tumors. However, concerns about regional disease in the mesorectum persist, even in T1 tumors, and radical surgery remains part of the therapy in most rectal cancers because we have yet to identify reliably a group of patients in whom we can predict good results without radical resection.
In this issue of Annals of Surgical Oncology, Pucciarelli et al.1 address this issue. They report on a series of 248 rectal cancer patients with 
T2 lesions treated with preoperative chemoradiation followed by surgery from 1994 to 2003. Patients were treated with several combinations of chemotherapy and radiation; the most common was bolus 5-fluorouracil with leucovorin and 45 Gy of radiation in 25 fractions or infusional 5-fluorouracil throughout radiation to a dose of 50.4 Gy over 28 fractions. A total of 235 patients then had either a low anterior resection or an abdominoperineal resection and comprise the study group for this article. Most patients underwent surgery 6 to 8 weeks after completion of chemoradiation. Only 45% of patients were treated with adjuvant postoperative chemotherapy. The focus of this report is the group of 56 patients (24%) who had a pathologic complete response of the primary tumor to neoadjuvant chemoradiation and the low incidence (1.8%) of pathologic lymph node metastases detected in this group. The authors submit that this very low incidence of lymph node metastasis suggests that radical surgery for these patients may be unnecessary and that local excision of the primary tumor to ensure a pathologic complete response may be adequate local/regional treatment.
The search for less invasive approaches in the treatment of rectal cancer is not new. A number of studies have reported excellent outcomes with local excision alone for early rectal cancer. Selected reports have indicated survival rates up to 90% and local recurrence rates of 10% to 20% for T1 and T2 cancers with surgery alone.2–4 However, the results of local excision are difficult to interpret because of the heterogeneity of patients represented and the retrospective nature of most series. More recently, caution has been urged in treating patients with local excision alone because local recurrence rates approaching 40% have been reported for T2 lesions.4 Although surgical salvage can be accomplished in 60% to 90% of these patients, cancer-specific survival seems to decrease as local recurrence increases, despite surgical salvage. The most informative study evaluating local excision is the Cancer and Leukemia Group B trial of T1 patients undergoing local excision alone and T2 patients undergoing local excision followed by postoperative chemoradiation therapy.5 Overall survival in this group of patients was 85%, and disease-free survival was 78% at 6 years. Of those with T1 tumors, 4 (7%) of 59 experienced a recurrence, and only 2 of these recurrences were isolated local recurrences. For T2 tumors, 10 (20%) of 51 recurred, and 7 of these were isolated local recurrences. All patients with an isolated local recurrence had salvage surgery with an abdominoperineal resection, but four of these nine patients died of distant disease.
Despite data suggesting that local excision alone or with adjuvant chemoradiation is appropriate in highly selected patients with T1 and T2 rectal cancer, widespread acceptance of this technique has been limited by concerns over the effect local recurrence has on overall survival and the implications of untreated disease in the mesorectum. Even in T1 lesions, lymph node metastases will be present in 5% to 13% of patients, and this number increases to 10% to 25% in patients with T2 tumors.6–8 The incidence of local recurrence in most local excision series approximates these rates, thus suggesting that local recurrence is often related to untreated disease in the mesorectum. The presence of high-grade tumors and lymphovascular invasion certainly increases the development of lymph node metastases as well, and this precludes local excision as an option in the minds of many. Even for those who have accepted local excision alone or with chemoradiation as appropriate treatment for early rectal cancers, the most commonly accepted criteria for pursuing local excision are that only T1 or T2 tumors that are well or moderately differentiated, that are without lymphovascular invasion, that are 
4 cm, and that encompass less than one third the circumference of the rectum are appropriate candidates. These criteria relegate a large portion of patients with rectal cancer to radical surgery.
In an effort to increase the number of patients who might be candidates for less radical surgery, several investigators have evaluated primary chemoradiotherapy for rectal cancer. Mohiuddin et al.9 reported their experience with preoperative radiation followed by local excision in two groups of patients: those who met criteria for local excision before radiation (small T1 or T2 lesions) and those with larger lesions or T3 lesions who were downstaged with radiation to meet the criteria for local excision. As expected, those starting out with small lesions did quite well, with 5-year survival of 92% and local recurrence of 11%. It is interesting to note that those who began radiation with more advanced tumors also did very well, with survival approaching 90% and no local recurrences at 5 years.
A more recent series has described a similar approach to patients with T2 and T3 rectal cancer, with chemoradiation followed by local excision in patients with a complete clinical response.10 All patients were staged with endoscopic ultrasound and treated to 45 Gy with external beam irradiation in 25 fractions combined with continuous-infusion fluorouracil. Of 95 patients enrolled in the trial, 26 had a complete clinical response and underwent local excision. Of these 26 patients, 20 had T3 tumors, and 7 were node positive by endoscopic ultrasound. Pathologic analysis revealed a complete pathologic response in 65% of these patients. Those with a pathologic partial response were recommended to have a completion abdominoperineal resection, although only two patients out of nine followed this recommendation. With follow-up at 2 years, there has been only one local recurrence (in a patient with a pathologic partial response), and there have been no distant recurrences.
The concept of primary chemoradiation has been taken one step further in a series of patients with a complete clinical response to chemoradiation who then received no surgical treatment.11 In the most recent update, 71 patients have been treated without surgery. All patients had T2 tumors, and most had either T3 (69%) or T4 (11%) lesions. Most patients (85%) were followed up for at least 2 years after treatment. The overall recurrence rate in this group was only 7% (five patients: two isolated local recurrences and three distant recurrences). Overall survival at 5 years was 100%, with a disease-free survival of 92%. Although these are indeed impressive results, they are difficult to reconcile with the knowledge that a complete clinical response historically translates into a pathologic complete response in only a fraction of patients. Although it may be true that the incidence of pathologic complete response is greater if one waits longer before surgery than the usual 6 to 8 weeks after completing radiation, it seems unlikely that this would approach the rate of local control (93%) reported in this trial.
Despite these encouraging data, there is still concern regarding adequate treatment of the mesorectum in patients undergoing primary chemoradiation, either alone or followed by local excision. This takes on particular significance in T3 tumors, in which the incidence of lymph node metastases may approach 70%.7 The unquantifiable factor has been the effect of chemoradiation on disease in the mesorectum. The study reported by Pucciarelli et al.1 addresses these concerns. The cohort of 56 patients with a complete pathologic response to neoadjuvant chemoradiation had a strikingly low incidence of lymph node metastases (only 1.8%) even though most patients had either T3 (79%) or T4 (16%) tumors before treatment. These data suggest that in patients with a primary tumor that responds completely to chemoradiation, disease in the mesorectum may respond similarly. However, other reports examining this question have not shown such impressive results with regard to treatment of mesorectal lymph node metastases, with residual disease detected in 15% of patients with a complete pathologic response of the primary tumor to neoadjuvant chemoradiation in another series.12 The differences between these two reports may be explained by the difficulty in finding lymph nodes in the mesorectum after chemoradiation, with the suggestion that there are in fact more positive lymph nodes present: they just may not be identified by the pathologist.
Although the appeal of sparing patients the morbidity associated with radical surgical resection for rectal cancer is undeniable, we must proceed cautiously while simultaneously embracing the concept. Long-term follow-up of these data will be imperative. As systemic therapy improves, the effect of distant disease on survival may continue to diminish. Local control will then assume an even more important role in treatment, and we will have an even greater opportunity to observe the effect that local recurrence has on survival. In addition, new cytotoxic chemotherapeutic drugs, as well as biologic agents, seem to have significant radiation-sensitizing properties. The addition of these agents to radiation treatment regimens has the potential to improve pathologic complete response rates and to further enhance the ability to control not only the primary disease, but also nodal disease in the mesorectum, thus increasing the percentage of patients who would be candidates for less invasive procedures. The greatest difficulty may lie in identifying patients in whom a less invasive approach is appropriate. Retrospective series provide useful information in beginning to answer these questions, but only with long-term follow-up of prospectively acquired data will we be able to define the optimal manner in which we can minimize the extent of surgery and its associated morbidity in the treatment of rectal cancer. Until then, we cannot throw away the scalpel.
Received for publication November 15, 2004. Accepted for publication December 7, 2004.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
