Correspondence

Tamoxifen and Chemotherapy for Refractory Metastatic Malignant Melanoma

N Engl J Med 1993; 328:140-141January 14, 1993

Article

To the Editor:

Cocconi et al. (Aug. 20 issue)1 reported enhancement of survival in patients with metastatic malignant melanoma by some 20 weeks after treatment with dacarbazine and tamoxifen as compared with dacarbazine alone. In a cautious accompanying editorial, Guerry and Schuchter called for confirmatory studies in more uniform prognostic groups2. We have examined the role of tamoxifen in enhancing chemosensitivity in a group with a particularly bad prognosis -- namely, patients with advanced malignant melanoma in whom chemotherapy has failed, a situation in which response rates are less than 10 percent.

Between March 1990 and December 1991, we treated 16 patients (10 men and 6 women; median age, 47 years; range, 20 to 64) with metastatic malignant melanoma, giving them cisplatin and tamoxifen with or without dacarbazine, depending on previous exposure to dacarbazine. All the patients had truly refractory disease, in that they had not responded to at least one chemotherapy regimen (carmustine in 81 percent, dacarbazine in 38 percent, vindesine in 6 percent, mechlorethamine in 6 percent, and melphalan in 6 percent), except one patient who had recurrent disease after a partial response to carmustine. The sites of disease were the lymph nodes (50 percent), subcutaneous tissues (44 percent), liver (38 percent), lung (38 percent), and bone (19 percent). Cisplatin (100 mg per square meter of body-surface area) and dacarbazine (500 to 600 mg per square meter) were given intravenously three to four times weekly in conjunction with continuous oral tamoxifen (20 mg per day) for a median of three treatment cycles (range, one to six). One woman had a complete response in the lung that lasted six months, and three men had partial responses lasting eight months (in one patient with disease in subcutaneous tissues) and four months (in two patients with disease in the lungs, subcutaneous tissues, and lymph nodes). The most frequent toxic effects of grade 2 or above in the World Health Organization classification system were emesis and anemia (72 percent and 38 percent of patients, respectively), and in one patient a deep venous thrombosis developed. The overall response rate was 29 percent (95 percent confidence interval, 3 to 55 percent), and the median survival was 6.4 months (range, 1 to 34); the response of two patients could not be evaluated.

Our results support those of Cocconi and colleagues1 and suggest that the ability of tamoxifen to enhance chemosensitivity is not specific for dacarbazine. The response rate that we report here is unexpectedly high for this group of patients and warrants further study, particularly with regard to determining the optimal schedule of tamoxifen and elucidating the mechanism behind this interaction.

Mark J. McKeage, F.R.A.C.P.
Alison Lorentzos, R.G.N.
Martin E. Gore, Ph.D., M.R.C.P.
Royal Marsden Hospital, London SW3 6JJ, United Kingdom

3 References

1
Cocconi G, Bella M, Calabresi F, et al. Treatment of metastatic malignant melanoma with dacarbazine plus tamoxifen. N Engl J Med 1992;327:516-523
Full Text | Web of Science | Medline

2
Guerry D IV, Schuchter LM. Disseminated melanoma -- is there a new standard therapy? N Engl J Med 1992;327:560-561
Full Text | Web of Science | Medline

3
Lakhani S, Selby P, Bliss JM, Perren TJ, Gore ME, McElwain TJ. Chemotherapy for malignant melanoma: combinations and high doses produce more responses without survival benefits. Br J Cancer 1990;61:330-334
CrossRef | Web of Science | Medline

Author/Editor Response

The authors reply:

To the Editor: With respect to the activity of tamoxifen combined with dacarbazine in metastatic malignant melanoma, McKeage et al. report four objective responses among 16 patients treated with tamoxifen and cisplatin, given with or without dacarbazine, depending on previous treatment with this agent. These results are rather favorable for a population of pretreated patients, but in our opinion they do not prove that tamoxifen contributed to the observed results or that dacarbazine could not have influenced our results.

Cisplatin is active as a single agent in advanced melanoma. In a review of the results of all the phase II studies conducted by the National Cancer Institute in 1985, the overall rate of response to cisplatin as a single agent was 19 percent in 145 patients who could be evaluated, most of whom had been pretreated.1

We reported a higher response rate with tamoxifen plus dacarbazine than with dacarbazine alone (28 percent vs. 12 percent). The most interesting result, however, was the favorable influence of the combination on overall survival and, in the subgroup analysis, on the survival of women as compared with men and, among the men, on the survival of those with a higher rather than a lower body-mass index.

Because the increase in the response rate does not fully explain the prolongation of survival, we hypothesize that a biologic modulation delaying the spontaneous progression of the tumor could be exerted by tamoxifen and endogenous estrogen, with or without dacarbazine. The increased response rate could not be the cause of this biologic phenomenon, but is possibly an indirect effect. Whether the contribution of dacarbazine, if any, to the observed results has to be considered specific for this agent or similar to that possibly achieved with a different chemotherapy could be clarified only by a properly constituted randomized study.

Giorgio Cocconi, M.D.
Mariangela Bella, M.D.
Marisa Bacchi, Ph.D.
Ospedale Maggiore, 43100 Parma, Italy

1 References

1
Wittes RE, Adrianza ME, Parsons R, et al. Compilation of phase II results with single antineoplastic agents. Cancer treatment symposia. Vol. 4. Bethesda, Md.: Department of Health and Human Services, 1985. (NIH publication no. 85-2739.)

Citing Articles (1)