Background: Oophorectomy is one of the treatments of choice for premenopausal women with advanced breast cancer. However, in recent years LH-RH analogs have replaced surgical castration (or ovarian irradiation) on the basis of the comparable therapeutic activity shown by these drugs in phase II studies. Moreover, the combination of tamoxifen and LH-RH analogs has gained popularity among clinicians attempting to obtain a 'total estrogen blockade' according to the same rationale previously proposed for advanced prostatic cancer. However, it has thus far not been proven that medical castration is as effective as oophorectomy or ovarian irradiation, nor is there enough evidence that tamoxifen could improve the efficacy of ovarian ablation.
Patients and methods: Eighty-five perimenopausal patients with estrogen receptor or unknown positive metastatic breast cancer were randomly allocated to receive one of the following treatments: surgical castration (or ovarian irradiation); goserelin; surgical castration (or ovarian irradiation) plus tamoxifen; goserelin plus tamoxifen. The study was performed according to a 2 x 2 factorial randomised design.
Results: While overall there was no significant difference in the response rates observed after two by two grouping, a trend did favour oophorectomy (or ovarian ablation) with respect to treatment activity. In fact, the best response rate was observed in patients allocated to this treatment (46.6% OR -95% CL: 21.2-72.9) while the lowest rate was observed in patients treated with oophorectomy plus tamoxifen (11.1% OR: CL: -3.4-25.6). Response to goserelin and goserelin plus tamoxifen was 27.2% (+/- 18.6) and 45% (+/- 21.8), respectively. Logistical regression analysis suggested that there might be a different interaction between tamoxifen and goserelin or oophorectomy (ovarian irradiation), respectively. Nevertheless, patient survivals were comparable, irrespective of allocated treatment. This indicates that two by two grouping has some value with respect to treatment efficacy and shows that oophorectomy (or ovarian irradiation) and goserelin have comparable efficacies. Tamoxifen did not improve the efficacy of gonadal ablation, although it did enhance the activity of goserelin treatment.
Conclusions: The results of the present study confirm prospectively that the efficacy of chemical castration is comparable to that of oophorectomy (or ovarian irradiation). The concurrent use of tamoxifen can probably enhance the activity of goserelin, but it also induces more side effects. However, it doesn't appear that 'total estrogen blockade' is more effective than gonadal ablation alone. Indeed, the question of whether chemical and surgical castration have the same effect in breast cancer is still open as is the one concerning the interaction between tamoxifen and gonadal ablation. Both questions should be addressed by prospective studies.