Does Trimodality Therapy Offer Benefit in Locally Recurrent Rectal Cancer?
Does Trimodality Therapy Offer Benefit in Locally Recurrent Rectal Cancer?
abstract & commentary
Source: Valentini V, et al. Cancer 1999;86:2612-2624.
Forty-seven patients with a biopsy-proven local recurrence of rectal cancer were selected for a study of multimodality therapy. The disease had to be confined to the pelvis, excluding the perineum, para-aortic lymph nodes, and the small bowel. The prior surgery was either an abdominoperineal resection in 30% or an anterior resection in 70%. In addition to surgery, previous therapy included radiation alone (12 patients) or chemotherapy alone (5 patients), with only one patient having previously received both modalities.
Treatment of the recurrent rectal cancer began with the combination of radiation therapy and chemotherapy. The radiation field included both the posterior pelvis and a 15 mm margin around the gross tumor volume. The 34 radiation-naive patients received two cycles of chemotherapy concurrently with 45-47 Gy either during a split course (1989-1992) or a single course (1992-1997). Those patients previously radiated received 23.4 Gy and one cycle of concurrent chemotherapy. The concurrent chemotherapy included 5-FU 1000 mg/m2/d as a 96-hour infusion and mitomycin-C 10 mg/m2 on the first infusion day.
After chemoradiation, only 21 (45%) of the patients underwent a radical resection with curative intent. Including an additional four patients who underwent palliative surgery, operative complications included two deaths (8%), two cutaneous fistulas, and two pelvic abscesses. Eleven patients received (in a non-randomized fashion) 10-15 Gy via intraoperative radiation therapy (IORT).
Post-operative chemotherapy was given to 66% of the patients; the remainder either refused or were medically unable to proceed. This consisted of 6-9 cycles of 5-FU and leucovorin boluses given for five consecutive days every four weeks (350 mg/m2 and 100 mg/m2, respectively).
One of the conclusions from the study was that clinical outcome in locally recurrent rectal cancer could be predicted by a classification system. This "modified Suzuki classification" stratified patients was based on CT scan estimates (rather than an intraoperative assessment) of pelvic side wall infiltration. For example, the five-year disease-free survival rates from the best to worst of the five subcategories were 100, 54.7, 8.6, 11.8, and 0 percent. However, only two patients were in the category with the best prognosis, and only three patients were in the worst category. Valentini and colleagues also conclude that "combined modality therapy was well tolerated and improved resectability, local control, and survival."
COMMENT By Kenneth W. Kotz, MD
In this study, chemotherapy was given both pre- and post-operatively. Whereas chemotherapy has an established role in the adjuvant treatment of non-metastatic rectal cancer, its role in resected stage IV disease or the treatment of a local relapse is unproven. The preoperative chemoradiation resulted in only a 53% response rate with no complete responses. In these minimally pretreated patients who received both chemotherapy and radiation therapy, a higher response rate may have been expected. After preoperative therapy, 45% of the patients were considered resectable. Unfortunately, the contribution of preoperative therapy to resectability is unknown because the percent considered resectable at the initial presentation is not reported.
The role of the postoperative chemotherapy was also unclear. There was no advantage in survival for the 66% of patients who received "adjuvant" chemotherapy, even though the comparison group included not only patients too ill to receive chemotherapy but also two patients who died perioperatively. The distant relapse rate of 43% seen in this study suggests that future attempts to improve chemotherapy in this setting, such as the use of longer infusional schedules of 5-FU or oxaliplatin- or irinotecan-containing regimens, are warranted. Interestingly, a recent article showed that postoperative hepatic arterial infusion of floxuridine combined with intravenous 5-FU improved the outcome of patients with resected hepatic metastases from colorectal cancer.1
If a previously unirradiated patient with a local recurrence of rectal cancer had an operable tumor, radiation therapy to improve local control would routinely be considered. However, the role of radiation in the management of a local recurrence in a previously radiated patient is undefined. Although we do not know the original radiation doses or fields in this study, the 13 previously irradiated patients received an additional 23.4 Gy with only one patient (7%) experiencing Grade 3-4 toxicity. There was no increase in perioperative complications noted in four of these 13 patients who went on to have surgery. A second course of radiation may be feasible, perhaps more so with the use of conformal techniques, but only a randomized trial would determine if it is beneficial.
In a non randomized fashion, 11 patients received IORT. Despite the small numbers, these patients obtained an advantage in five-year local control that was statistically significant. However, the trend towards a survival advantage was not statistically significant.
Local recurrences of rectal cancer may be seen more frequently in the future, particularly with the emerging role of local excision. This study highlights the need to determine the optimal treatment of this situation including not only the need for but also the timing, sequence, and doses of radiation and chemotherapy.
Reference
1. Kemeny N, et al. N Engl J Med 1999;341:2039-2048.
Regarding the rectal cancer trimodality report, which of the following is not true?
a. In this study, intraoperative radiation therapy was associated with improvement in local control.
b. In this study, an additional 23.4 Gy in patients previously radiated for rectal cancer was feasible.
c. In this study, patients who received the post-operative chemotherapy survived longer than those who did not.
d. A modification of the Suzuki classification was used in this study to attempt to predict prognosis based on a pretreatment estimate of pelvic sidewall infiltration as measured by CT scans.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.