IPD from four RCTs were included in the meta-analysis. Out of 228 enrolled participants, 220 were eligible and were included in the analysis.
Response evaluation took place at either 2 or 3 months from the start of therapy in all trials. At the time of analysis, 210 of the 220 participants had died.
There was no difference in overall response rate between tamoxifen and ovarian ablation across the four trials (P=0.94), even when only patients assessable for response were included (P=0.88).
The OR for disease progression favoured tamoxifen over ovarian ablation as first-line therapy (OR 0.86, 95% CI: 0.63, 1.10). This was not statistically significant (P=0.32).
The OR for mortality also favoured tamoxifen over ovarian ablation as first-line therapy (OR 0.94, 95% CI: 0.69, 1.20), and again was not statistically significant (P=0.72).
For the association between response to initial treatment with tamoxifen and response to crossover treatment with ovarian ablation (25 patients), the Spearman rank correlation coefficient was 0.398 (0.02<P<0.05). Three (33%) out of 9 patients who initially responded to tamoxifen also responded to ovarian ablation. Three (19%) out of 16 patients who did not respond to tamoxifen responded to ovarian ablation.
For the association between response to initial treatment with ovarian ablation and response to crossover treatment with tamoxifen (47 patients), the Spearman rank correlation coefficient was 0.332 (0.02<P<0.05). One (7%) out of 15 patients who initially responded to ovarian ablation also responded to tamoxifen. Three (9%) out of 32 patients who did not respond to ovarian ablation responded to tamoxifen.