Abstract & Commentary
Synopsis: Standard therapy for invasive distal rectal cancers is radical surgery. Sphincter-sparing may be attempted and can often be facilitated by the use of neoadjuvant therapy. However, when a tumor is found to be too close to, or traverses, the internal sphincter at the time of surgery, a colostomy is unavoidable. This study found that, in patients who were borderline candidates for sphincter-sparing, and in patients deemed inoperable, dose escalation with primary combined radiotherapy achieved acceptable local control with moderate toxicity. They concluded that abdominoperineal resection should rarely be performed for T2 lesions, even when there is invasion of the anal canal.
Source: Gerard J, et al. Int J Radiat Oncol Biol Phys. 2002;54:142-149.
Gerard and colleagues at the centre Hospitlier Lyon Sud in Lyon, France, performed a retrospective analysis of 63 patients who were treated in a pilot study of primary radiotherapy for middle or lower rectal adenocarcinomas from 1986-1998. There were 26 patients who were medically inoperable, 22 who refused APR, and 15 who wished to attempt sphincter preservation for borderline operable lesions. All tumors were ultrasound stage T2-3N0-1 lesions, including 18 patients with perirectal adenopathy. Forty-one patients had T2 tumors, and 22 had T3 tumors. Forty-three percent of patients had tumors involving the anal canal. Patients were excluded from the study for technical reasons if their tumors involved more than two thirds rectal circumference, mainly because of the difficulty treating circumferential lesions with contact x-ray therapy. All patients had a negative metastatic work-up.
Treatment consisted of contact x-ray therapy, external beam radiotherapy, and brachytherapy, which Gerard et al called "combined radiotherapy." Contact x-ray therapy was administered on days 1, 7, and 21 with a 50-kilovoltage photon unit through 1 or 2 fields to a median surface dose of 80 Gy (r, 35-140). The number of fields varied by tumor size and response. Contact between the radiotherapy device and the tumor was achieved via a proctoscope for a median of 2 sessions (r, 1-3). External beam radiotherapy was initiated on Day 14 using 18 MV photons administered via a 3 or 4 field technique encompassing the primary tumor and perirectal lymph nodes up to S1-2 or S2-3. Three centimeter distal and lateral margins were used. The external and common iliac lymph nodes were not included, and neither was the anal canal. Treatment was given at 3 Gy per fraction for 13 fractions over 17 days with 2:1 weighting favoring the posterior over the lateral fields. A 4-fraction concomitant boost to gross tumor at 1 Gy per fraction using an AP/PA approach was given, as well. The dose of external beam RT was felt to be comparable to 50-54 Gy at conventional 2 Gy fractions. Finally, an interstitial iridium-192 boost was undertaken 4-6 weeks following completion of the external beam RT, for a median brachytherapy dose of 20 Gy to the 85% isodose line (r, 10-25) in 22 hours. This was done with either a 5-wire perineal template (n = 38) for the lower lesions, or a 2-wire "rectal fork" (n = 17) applied with a rigid proctoscope with the patient in the knee-chest position without general anesthesia. Seven frail patients who had complete responses of their T2 lesions did not receive a brachytherapy boost. The median dose rate for the template was 1.13 Gy per hour, and for the rectal fork, it was 1.72 Gy per hour.
Median follow-up for the entire group was 54 months, and for surviving patients it was 78 months. Among the 63 patients treated, 58 (92%) exhibited a complete clinical response at 2 months posttherapy. Thirty-three patients had supple rectal walls, 24 had a small area of fibrosis, and 5 had small residual biopsy-proven tumors. Two of the 3 attempts at salvage APR were successful. The 58 complete responders had a 28% local recurrence rate (n = 18), with a median time to failure of 16 months (r, 7-49). Five of the local failures were salvaged with APR or a second course of contact therapy. Only 2 of the persistent failures required a palliative colostomy. Forty patients (63%) achieved local control with primary therapy, and 6 more were salvaged for an overall local control rate of 73%. Overall survival at 5 and 10 years was 64% and 45%. Overall survival and local control at 5 years was 82% and 83% for T2 tumors. Overall survival at 5 years was 35% for T3 tumors. Twelve patients died of cancer. There were no Grade 3-4 acute toxicities. Twenty-four patients developed intermittent rectal bleeding 6 months posttreatment that lasted 2-3 years, but only one required transfusions. No patient required a colostomy because of radiation toxicity. Anorectal function was assessed using the MSKCC Scale, and was good or better in 92% of patients. Statistical analysis of results was not provided, although Gerard et al stated that T-stage was "a strong prognostic factor."
Gerard et al concluded that combined radiotherapy alone is a suitable treatment for inoperable patients with < 5 cm T2-3 lesions. For patients that manifest a poor response by Day 21, the option to pursue surgery may be reassessed. For T3 lesions, RT dose escalation or concurrent chemotherapy should be explored. While surgery remains the mainstay for rectal cancer, APR should rarely be performed for T2 lesions, even if they invade the anal canal.
Comment by Edward J. Kaplan, MD
This paper from the Lyon group is fascinating from many different perspectives, and represents an ongoing effort reflected in serial publications of their experience. Although brachytherapy boosts are not uncommon, very seldom do we see regimens that incorporate 3 radiotherapy delivery systems in a single protocol. Despite the fact that radiotherapy alone has been shown to confer good results with smaller low-lying rectal tumors since the 1980s,1 this approach has never gained popularity. In a situation where there seems to be no alternative to an APR, an attempt to control the disease with primary radiotherapy makes sense.
Though the combined radiotherapy approach used by Gerard et al seems logical, widespread adoption of their technique in the United States seems unlikely. The 2 major problems I see are the fact that Philips Medical no longer manufactures RT-50 contact x-ray units,2 and most radiation oncologists in the United States have never performed brachytherapy for rectal cancer. Looking back on a recent paper by Wallner,3 from Seattle, perhaps at least some radiation oncologists here could adapt to the French group’s style. Wallner et al performed template guided permanent interstitial brachytherapy on awake Veterans Administration prostate cancer patients using local anesthesia, and reported no problems.
Aumock et al from Washington University published their experience with 207 patients treated with primary radiotherapy from 1981-1995.4 In contrast to the Gerard study, tumors involving the anal canal were excluded because it was felt that endocavitary treatment entails the risk of severe pain due to ulceration in the anus. Patients were treated with a mean external beam dose of 45 Gy at 1.6-1.8 Gy per fraction, which preceded contact x-ray therapy with the Philips 50 kV unit. The latter was used to administer two 30 Gy mucosal surface doses 2 weeks apart. Dose penetration at the center of the contact x-ray field was 54% at 0.5 cm and 33% at 1 cm. Local control with RT alone was 71%, and this rose to 81% with surgical salvage. All T1 lesions were controlled, as were 85% of mobile T2 lesions, and 56% of T3 lesions and tethered T2 tumors. No patient received chemotherapy. No patient underwent a brachytherapy boost. Multivariate analysis indicated that N1 disease and absence of gross total endoscopic debulking were negative prognostic factors.
Gerard et al achieved their goal of preserving good anal function in a group of compromised patients by escalating the radiotherapy dose. Their regimen was vastly different from what is considered conventional radiotherapy in the United States, and it was aggressive even by European standards. For example, while large doses per fraction are accepted in Europe, concomitant boosts go a step further. The dose rate for the interstitial boost was also high in comparison to the 0.5-0.6 Gy per hour that we typically use. One way the French group was able to succeed with this aggressive format was by limiting the treatment volume. Had they included the external and common iliac lymph nodes, their patients probably would not have tolerated the therapy as well as they did.
The concept that primary radiotherapy can be successful in patients who would otherwise have little hope of avoiding a colostomy is important, and should not be brushed aside. What we need to focus on is devising a modern protocol that can achieve similar or better results using equipment that is widely available, such as the high dose rate afterloader with an endorectal applicator. As the Gerard group suggested, chemotherapy would probably improve results for the larger T2 lesions, and T3 tumors. Just as the University of Washington group reported, giving the endorectal therapy after the tumor has been downstaged with external beam RT is attractive. This would eliminate the problem that the French had handling circumferential lesions. The beauty of endorectal HDR as a boost therapy is that the entire circumference of the rectum can be treated easily.
Dr. Kaplan is Acting Chairman, Department of Radiation Oncology, Cleveland Clinic Florida, Ft. Lauderdale, FL; Medical Director, Boca Raton Radiation Therapy Regional Center, Deerfield Beach, FL.
1. Lavery IC, et al. Dis Colon Rectum. 1987;30:835-838.
2. Barish RJ, et al. Med Dosim. 1994;19:1-3.
3. Wallner K, et al. Int J Radiat Oncol Biol Phys. 1999; 45:401-406.
4. Aumock A, et al. Int J Radiat Oncol Biol Phys. 2001; 51:363-370.