ASSOCIATION OF GLUTATHIONE-S-TRANSFERASE GSTM1 AND GSTT1 NULL GENOTYPES WITH CLINICAL OUTCOME IN EPITHELIAL OVARIAN-CANCER

Citation
Rej. Howells et al., ASSOCIATION OF GLUTATHIONE-S-TRANSFERASE GSTM1 AND GSTT1 NULL GENOTYPES WITH CLINICAL OUTCOME IN EPITHELIAL OVARIAN-CANCER, Clinical cancer research, 4(10), 1998, pp. 2439-2445
Citations number
36
Categorie Soggetti
Oncology
Journal title
ISSN journal
10780432
Volume
4
Issue
10
Year of publication
1998
Pages
2439 - 2445
Database
ISI
SICI code
1078-0432(1998)4:10<2439:AOGGAG>2.0.ZU;2-O
Abstract
Epithelial ovarian cancer is generally associated with a poor outcome, although the mechanisms that determine survival and progression-free interval (PFI) are unclear. Data from ovarian tumors showing associati ons between (a) null genotypes at the glutathione S-transferase GSTM1 and GSTT1 loci and expression of p53 protein and (b) outcome and expre ssion of p53 suggest that polymorphism at these loci is a factor deter mining outcome. Accordingly, we have studied the association between t he GSTM1 null and GSTT1 null genotypes and survival and PFI in 148 wom en with epithelial ovarian cancer. Although we did not find an associa tion between individual genotypes and outcome, women with both GSTM1 n ull and GSTT1 null genotypes demonstrated poorer survival (P = 0.001) and reduced PFI (P = 0.003). Thus, no cases with both these genotypes survived past 42 months postdiagnosis. In contrast, 43% of the women w ithout this combination survived beyond this time. Because response to chemotherapy is a major factor determining outcome in ovarian cancer, we also examined the data for associations between the glutathione S- transferase genotypes and response to such treatment. Thus, in 78 pati ents treated with chemotherapy, the combination of GSTM1 null and GSTT 1 null was associated with unresponsiveness to primary chemotherapy (P = 0.004); none of the eight patients with both these genotypes respon ded, compared with 38 of 70 (54%) of patients with other genotype comb inations. The effect of the combination of genotypes on survival and P FI was lost in a multivariate model that included response to chemothe rapy as a confounding factor. This suggests that the combination of GS TM1 null/GSTT1 null is associated with outcome because of its influenc e on response to chemotherapy. These preliminary findings may provide a basis for the selection of patients for treatment with chemotherapeu tic agents.