Purpose: We conducted a randomised phase II study to assess the safety and efficacy of standard versus high-dose pemetrexed in platinum-resistant epithelial ovarian cancer (PR-EOC). The expression of ten genes was also examined as potential biomarkers of pemetrexed/platinum activity.
Patients and methods: Patients received pemetrexed 500mg/m(2) (Pem500) or 900mg/m(2) (Pem900) on day 1 of each 21-d cycle. Responses were defined per RECIST for measurable disease or by Gynaecologic Cancer Intergroup (GCIG) CA-125 criteria for non-measurable disease.
Results: Of 102 patients randomised, 98 were evaluable for toxicity (47 Pem500, 51 Pem900) and 91 were evaluable for efficacy (43 Pem500, 48 Pem900) of whom 68 had measurable disease and 23 had CA-125-defined disease. The overall RR was 9.3% (95% CI: 2.6-22.1%) on Pem500 and 10.4% (95% CI: 3.5-22.7%) on Pem900. The median progression-free survival (PFS) was 2.8 months on both arms, and the median survival was 11.9 and 10.3 months, respectively. Lower mRNA expression of excision repair cross-complementation group 1 (ERCC1) and reduced folate carrier 1 (RFC1) were associated with longer PFS and time to treatment failure, respectively. Grade 3/4 toxicities, including fatigue, nausea and vomiting, were numerically greater on Pem900. Pemetrexed-related SAEs occurred in 17% and 28% of Pem500 and Pem900 patients, respectively.
Conclusions: Pemetrexed has activity in PR-EOC equivalent to other agents in platinum-resistant disease; however, Pem500 has the preferable toxicity profile. ERCC1 and RFC1 may merit examination as predictive biomarkers in PR-EOC.