Preoperative Radiotherapy Combined with Total Mesorectal Excision for Resectable Rectal Cancer

Abstract & Commentary

Synopsis: Total mesorectal excision (TME) is a surgical technique using sharp dissection of the rectal fascia during resection of rectal cancer. This contrasts with traditional surgical techniques that use blunt dissection and may thus leave tissue fragments behind. The TME technique has been associated with lower local recurrence rates compared with the blunt technique. The Dutch Colorectal Cancer Group, in collaboration with the EORTC, designed a randomized trial to determine whether a short course of preoperative radiotherapy adds to the benefits of TME in early rectal cancer. Initial results at 2-year follow-up show statistically significantly better local control results for the combined modality group, and it is anticipated that an overall survival advantage will emerge with longer follow-up.

Source: Kapiteijn E, et al. N Engl J Med. 2001;345:
638-646.

The dutch colorectal cancer group reasoned that it is possible that rigorous surgical technique may obviate the need for preoperative radiotherapy for rectal cancer. Previously published results from the Swedish Rectal Cancer Trial and a meta-analysis by Camma indicated that there is a local control and survival benefit with preoperative radiotherapy (RT).1,2 While the Swedish trial used a short course of radiotherapy, administered in 5 days, the Camma paper reported on various series which used 1-5 week courses. The Dutch/EORTC trial reported by Kapiteijn and colleagues is the first randomized trial to incorporate the TME technique, and a short radiotherapy schedule was coupled with it for comparison in a combined modality arm.

From January 1996 through December 1999, 1861 patients with nonfixed adenocarcinomas of the rectum were randomized to either TME alone or 25 Gy in 5 days followed by immediate TME. Patients were stratified by treating center and planned surgical procedure, ie, abdominoperineal resection (APR) or low anterior resection (LAR). There were 1805 patients eligible for evaluation, including 908 in the TME arm and 897 in the combined modality arm. Postoperative adjuvant therapy, including chemotherapy, was prohibited except in patients who had positive surgical margins. Median patient age was 65 years, and two thirds were males. Both arms were equally balanced in terms of demographics and tumor features. The majority of patients (n = 1530, 85%) were accrued in the Netherlands, where surgeons were intensively trained to perform TME procedures using workshops, symposia, and videotapes, and world-renowned surgeon proctors were required for the first 5 cases.

Median follow-up was 24.9 months (range, 1.1-56 months). Gross total resection without tumor spillage was achieved in 1748 patients (97%), and 1653 patients (92%) were free of metastatic disease at laparotomy. No additional information on findings at the time of surgery, other than the fact that no tumor was detected at surgery in 28 patients (2%), was provided. Consistent with a previously published early report, there were no differences between the treatment arms in postoperative mortality, and negligible differences in postoperative morbidity slightly favoring the surgery alone group.3

Approximately two-thirds of patients in both treatment arms underwent LAR. Gross total resection rates were similar in both groups (96-97%). Median time from randomization to surgery was 14 days in the TME arm, and 21 days in the combined modality arm. Ninety-one percent of patients were treated per protocol in the combined modality arm, and 94% in the TME arm. The overall postoperative death rate was 3.4%. Two-year overall survival was essentially identical at 81.8% for the combined modality arm, and 82% for the TME arm (P = .84). The rate of distant metastases was also similar, at 14.8% for the combined modality arm, and 16.8% for the TME arm (P = .87). However, there was a statistically significant difference in local control. In the combined modality arm, the 2-year local recurrence rate was 2.4% compared with 8.2% for the TME arm (P < .001).

Kapiteijn et al concluded that short-term preoperative radiotherapy reduces the risk of local recurrence in patients treated with TME for rectal cancer. An effect on overall survival probably has not emerged yet because of the small number of local recurrences and the short follow-up. The benefit from RT applied to tumors at all levels in the rectum, and to all stages studied.

Comment by Edward J. Kaplan, MD

The Dutch trial is the first to incorporate the TME technique into a randomized trial evaluating adjuvant therapy in rectal cancer. Since a short course of 25 Gy followed by immediate surgery was the regimen used in the combined modality arm, fixed lesions could not be included. This is because downstaging of the lesions increases over weeks as reported by Francois in the Lyon R90-01 randomized trial, not over just a few days as per the Dutch trial.4 Even very advanced lesions can be downstaged successfully in preparation for surgery.5 Since downstaging was not a goal of the Kapiteijn study, sphincter-sparing for low-lying lesions could not be accomplished, and thus there were a lot of APRs done in both treatment arms.

It will be interesting to see whether a survival benefit does become apparent as follow-up data accrue. The local control benefit from radiotherapy, in the face of state-of-the-art surgery where the entire visceral fascia is removed, was not totally expected. It is also somewhat unusual that it held for all stages, including stage I and II lesions. Unfortunately, despite the benefit shown from preoperative RT, the results will not be directly applicable to therapy in the United States because we do not abide by the short-course schedule for 2 practical reasons. First, as mentioned above, there is no potential for downstaging and therefore little satisfaction realized by treating locally advanced and/or low-lying lesions. Second, morbidity, like acute lumbosacral plexopathy during and after short-course preoperative RT as reported by Frykholm in a follow-up study from Sweden, is much more apt to occur when large fraction sizes are used.6 In addition, RT here is typically administered concomitantly with chemotherapy either before or after surgery for transmural and/or node-positive lesions, and chemotherapy was expressly forbidden in the Dutch trial.

The 97.6% local control rate for the combined modality arm in the Dutch trial is much better than the 87% local control rate in the previous Swedish trial, which may be attributable to an improvement in surgical technique.1 The Dutch trial results are similar to the recent early results reported from the NSABP R-03 randomized trial, which also showed better disease-free survival in the preoperative-RT arm.7 We will have to wait until data from both trials mature in order to determine whether TME is worthwhile, and whether overall survival is indeed improved with preoperative adjuvant RT.

References

1. The Swedish Rectal Cancer Trial. N Engl J Med. 1997; 336:980-987.

2. Camma C, et al. JAMA. 2000;284:1008-1015.

3. Kapiteijn E, et al. Eur J Surg. 1999;165:410-420.

4. Francois Y, et al. J Clin Oncol. 1999;17:2396-2402.

5. Mohiuddin M, et al. Int J Radiat Oncol Biol Phys. 2000;48:1075-1080.

6. Frykholm G, et al. Radiother Oncol. 1996;38:121-130.

7. Roh MS, et al. NSABP-03 Randomized Trial [abstract]. ASCO. 2001;123a. Abstract 490.