FANCF promoter hypermethylation as predictor of clinical outcomes in advanced epithelial ovarian cancer

Abstract

5563 Background: Platinum remains the most active drug in the treatment of epithelial ovarian cancer. The Fanconi Anaemia (FA) pathway protects cells against cell death induced by cross-linking agents. Epigenetic silencing of FANCF by promoter hypermethylation has been proposed as a mechanism of sensitization to platinum in vitro (Taniguchi et al., Nature Medicine, 2003). This study aims to profile the methylation of FA genes and to investigate their impact on clinical outcomes in epithelial ovarian cancer treated with cisplatin-based chemotherapy. Methods: Promoter hypermethylation and gene expression of 6 FA genes (FANCA, FANCC, FANCD2, FANCE, FANCF and FANCG) were investigated in 9 ovarian cancer cell lines and in 74 primary ovarian tumors (Stage III/IV) using methylation- specific PCR (MSP) and reverse-transcription PCR (RT-PCR). The clinical samples were retrospectively retrieved from a large Phase III RCT (the EORTC 55931/ NCIC OV10), which randomised patients to cisplatin/ cyclophosphamide or cisplatin/ paclitaxel chemotherapy following surgical debulking. Results: Ovarian cancer cell lines: FANCF hypermethylation was associated with downregulation of mRNA expression. Treatment with demethylating agent, 5'-Azacytidine, resulted in gene reactivation. FANCA, FANCC, FANCD2, FANCE and FANCG were not methylated. Primary tumors: FANCF hypermethylation was detected in 7/74 (10%) of the primary tumors. It predicts for adverse PFS in univariate (median PFS 5.6 vs 18.8 mths, p 0.001) and multivariate analyses (HR 3.70, 95% CI 1.43 - 9.54, p 0.007). 5/7 (71%) of FANCF methylated patients had “clinical platinum resistant / refractory‘ disease compared with 13/67 (20%) of unmethylated (RR 3.68, 95% CI 1.65 - 5.29, p 0.008). HR for OS was 1.16 (95% CI 0.43–3.11, p 0.765). The adverse effects of FANCF methylation on PFS and OS was significant only in patients treated with two cross linking drugs, cisplatin/cyclophosphamide, (p=0.004, and p=0.041, respectively), not one (cisplatin/paclitaxel). Conclusions: Epigenetic silencing of FANCF by promoter hypermethylation is an independent predictor of adverse PFS in advanced epithelial ovarian cancer treated with cisplatin-based chemotherapy, and may also predict for “clinical platinum resistance”. No significant financial relationships to disclose.