Prospective Phase II Results of Chemoradiation for Merkel Cell Carcinoma are Encouraging

Abstract & Commentary

Synopsis: Merkel cell carcinoma is a tumor of the skin that is well known for its metastatic potential. Because of its scarcity, no trials have been reported. Poulsen and colleagues from Australia conducted a unique multi-institutional Phase II prospective trial of concomitant chemoradiation for patients with high-risk disease and concluded that their regimen resulted in disease control and survival rates that are better than those reported in the literature.

Source: Poulsen M, et al. J Clin Oncol. 2003;21:4371-4376.

Poulsen and associates in Australia conducted a prospective Phase II trial using chemoradiation for primary, adjuvant, or salvage therapy of merkel cell carcinoma (MCC). They based their approach on the similarities in natural history and biologic behavior between MCC and small-cell lung cancer. The study, which was carried out between 1996 and 2001, accrued 53 patients with high-risk disease from 6 institutions. High risk was defined as primary tumor > 1 cm, presence of lymphvascular invasion, involved lymph nodes (Stage II), subtotal resection, or local recurrence following surgery. Patients with low-risk disease, poor performance status, or distant metastases were excluded. Surgery was frequently performed prior to referral to the oncology department. Clear margins were not mandatory. Median age was 67 years (range, 43-86 years). There were 41 patients with primary disease (77%) and 12 with recurrences (23%). Forty-two percent (n = 22) of the MCC lesions were on the head and neck, 30% (n = 16) were at the extremities, 24% were occult primary tumors with disease in the lymph nodes (n = 13), and 4% were located on the trunk (n = 2). Seventy-two percent (n = 38) were treated adjuvantly, and 28% (n = 15) were treated with primary chemoradiation. Among the patients who underwent surgery, 28/53 (53%) achieved clear margins. Sixty-two percent of patients (n = 33) had involved N1 and/or N2 lymph nodes.

For those patients treated adjuvantly, the median interval from surgery to chemoradiation was 47 days (range, 14-87). The radiation therapy portals included the primary tumor or tumor bed with a 3-5 cm margin and lymph nodes located < 20 cm away. Dose was 50 Gy in 25 fractions (range, 44-60 Gy) over 5 weeks to the operative bed or gross disease and 45 Gy in 25 fractions to areas of suspected microscopic disease. Four cycles of chemotherapy were administered, with the option to continue chemotherapy beyond the fourth cycle. Chemotherapy was given on weeks 1, 4, 7, and 10, with the latter 2 cycles given after the end of the RT. The carboplatin dose was calculated according to the Calvert formula, and etoposide was given at 80 mg per square meter.

Outcomes were measured from the start of chemoradiation. Median follow-up was 48 months (range, 11-70 months). No patients were lost to follow-up. Eighty-three percent of patients received RT per protocol, with no unacceptable major deviations, and 87% completed all 4 cycles of chemotherapy. Sixty-four percent (n = 34) of patients suffered Grade 3 or 4 skin reactions (4 with Grade 4), and 35% experienced febrile neutropenia (n = 19). Grade 3 or higher late skin toxicity occurred in 15% at 3 years. Three-year overall survival was 76%, 3-year disease-free survival was 65%, and 3-year actuarial local control was 75%. Local control was 77% for patients treated adjuvantly and 71% for patients receiving primary chemoradiation (P value not given). The crude 3-year distant metastatic rate was 17% (n = 9).

On multivariate analysis, age, lymph node status, disease site, and presence of gross disease were evaluated. Lymph node status was significant in terms of overall survival (P = .001) but not local control. Lesions on the lower extremities did statistically significantly worse (P = .002), and the presence of gross disease at treatment was not a significant factor for overall survival.

Poulsen et al concluded that chemoradiation was tolerable for the majority of MCC patients and that the survival and disease control rates demonstrated in their study appeared to be superior to other reported series. They suggested that any delay in chemoradiation in order to perform further surgery (eg, to obtain clear margins) would be detrimental to treatment efficacy. They also suggested that the low rate of distant metastases seen in this trial is consistent with a positive influence exerted by the chemotherapy. However, given the heterogeneity of the high-risk patients treated, Poulsen et al advised great caution in interpreting their results. They are now considering a Phase III trial comparing chemoradiation with carboplatin and RT followed by carboplatin/etoposide as the experimental arm, vs radiotherapy alone as the standard arm, recognizing that febrile neutropenia was their worst toxicity.

Comment by Edward J. Kaplan, MD

The issue of adjuvant therapy for MCC remains controversial. Many studies have advocated postoperative radiotherapy following resection of MCC.1-3 Fenig et al from Israel published one of the only papers addressing combined modality therapy for MCC, using a sequential approach with surgery or chemotherapy followed by RT.4 In the largest retrospective study on MCC reported to date, investigators from MSKCC published their findings on 102 MCC patients treated between 1969 and 1996.5 Very few patients received adjuvant therapy, either with RT or with chemotherapy. The most common site of recurrence was in regional lymph nodes, which were affected in 40/55 recurrences (73%). Median follow-up was 35 months, 5-year overall survival was 76%, and 5-year disease-free survival was 74%. Based on these results, the authors recommended an aggressive approach toward the treatment of clinically negative lymph nodes. Poulsen et al were presumably motivated by this type of recommendation, with impressive results. First, they organized a trial for a rare disease, and second, they managed an annual accrual rate nearly 3 times that in the MSKCC paper. Comparing the 65% 3-year DFS from Poulsen’s group to the 74% 5-year DFS from MSKCC seems to reflect an advantage in favor of the MSKCC data, until one factors in that every one of the Australian patients had high risk disease. Looking at it from that perspective, the chemoradiation data begin to look rather good. The very low rate of DM seen in the Australian trial is particularly striking given that almost two-thirds of the patients in that study had positive lymph nodes, compared with the 9-33% reported elsewhere in retrospective studies, including the 22% cited in the MSKCC study.

Further efforts are necessary to amplify upon these early results, while at the same time improving upon the toxicity profile. The Australian team is contemplating a randomized trial, and they indicated that they would likely solicit international cooperation. Hopefully, not only will their perseverance lead to higher cure rates for MCC, but their work can serve as an inspiration to others in the development of trials for other rare tumor types.

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. Nathu RM, et al. Radiat Oncol Investig. 1998;6:233-239.

2. Bischof M, et al. Strahlenther Onkol. 1999;175: 611-615.

3. Gillenwater AM, et al. Arch Otolaryngol Head Neck Surg. 2001;127:149-154.

4. Fenig E, et al. Cancer. 1997;80:881-885.

5. Allen PJ, et al. Annals of Surg. 1999;229: 97-105.