High-Dose Cisplatin with Dacarbazine and Tamoxifen in the Treatment of Metastatic Melanoma
I read with interest the paper by Buzaid et al.' In 23 patients with metastatic melanoma treated with high-dose cisplatin, dacarbazine, and tamoxifen, they reported a response rate of 13% and concluded that their data did not support a significant interaction between tamoxifen and cisplatin or dacarbazine.
At least with reference to dacarbazine, these conclusions are in contrast with the results of a trial from our Gruppo Oncologico Italiano di Ricerca Clinica (GOIRC)" mentioned in the paper. This was the only randomized trial and the trial with the highest accrual of patients (117 compared with 20- 23 patients in the other series) until now reported on the interesting problem of the relationship of tamoxifen with chemotherapy in metastatic melanoma. Our study demonstrated significantly better results using dacarbazine plus tamoxifen compared with dacarbazine alone not only in response rate, but in survival. The subgroup analysis demonstrated that, on the dacarbazine plus tamoxifen treatment arm (and not at all on the dacarbazine alone treatment arm), the response rate and mostly survival favored women rather than men. In postmenopausal women and in men, survival was significantly more favorable in patients with a higher body mass index. These data suggest that the positive activity of tamoxifen in metastatic melanoma is biologic in nature and probably modulated by the endogenous estrogens.
In my opinion it is not justified to deny a positive interaction with dacarbazine (which has been directly and positively tested in a random position) and to attribute the merit to carmustine, which has not been directly tested until now. Rather, I think that in this small sample of patients, a prior medical treatment in 9 of 23 (39%) and the generalized presence of visceral disease (27 visceral disease sites in 23 patients) could have masked the positive contribution of tamoxifen to the results in terms of response rate. Moreover, any positive interaction of tamoxifen with malignant melanoma, in terms of survival, is totally masked in studies using a nonrandomized format.