| Objective: Epithelial ovarian cancer (EOC) is one of the most commongynecological malignancies, which was estimated to be the seventh mostleading cause of female cancer-related deaths worldwide. The five-yearsurvival rate for patients with advanced disease relapse is about30%. Thestandard treatments for EOC patients consist of surgical debulking andplatinum-based chemotherapy after the surgery. The platinum-based agent isthe first-line chemotherapy regimen, but drug resistance to anticancer drugsgreatly affects the prognosis.The P-glycoprotein(P-gp), which is encoded by MDR1, locates atMembrane. P-gp have the function of pumping the intracellular drugs out ofthe cells, causing the cancer drug resistance. The level of P-g expression maycause the differences in patients’ responses to platinum-based chemotherapyand recovery after treatment. Drug metabolism is another important cause ofplatinum resistance. Once platinum enters the cells, it could be conjugated byglutathione (GSH). The formed complex reduces the toxicity of platinum andcan be more easily transported out of the cells. The conjugation is determinedby the cellular GSH level. Therefore, some GSH synthesis enzymes are relatedwith platinum response. Therefore, understanding the association of thepolymorphisms with platinum response and survival will be beneficial forindividualized chemotherapy.In this study, we investigate the association between vatiants of MDR1and GSTP1genes and clinical outcome of epithelial ovarian cancer (EOC)patients treated with platinum-based chemotherapy. This article aims to findout the tap SNP of EOC, providing important experimental data for the“individualization†of clinical treatment. Methods: The retrospective cohort study included225EOC patients whohad surgery in Fourth Hospital of Hebei Medical University betweenDecember2001and June2010. Five ml of venous blood from each subjectwas drawn in vacutainer tubes. Genomic DNA was extracted by usingproteinase K digestion followed by a salting out procedure. Three SNPs(MDR1C3435Tã€G2677T/A and GSTP1A313G) were assessed using aligase detection reaction(LDR). All patients were followed up for three yearsand had been evaluated according to the FIGO surgical staging system.Statistical analysis was performed via SPSS ver.13.0software package.The association between the genotypes and allelotype and platinum responseof the patients were compared via the χ2test. To adjust for age, grade, stage,tumor residual and histology status, unconditional logistic regression analysiswas performed. Apply logistic regression method to calculate ORs and CIwith95%to estimate the relative risks. Kaplan-Meier curves were constructedto investigate survival difference between the wild-type and variant genotypes.The wild-type alleles as reference, differences were tested for statisticalsignificance via the log-rank test and Breslow analysis. HRs and95%CIswere calculated using multivariate Cox proportional hazard models to adjustfor appropriate variables, such as age, stage, grade, tumor residual, andhistology status. The SPSS software (version.13.0, SPSS Inc., Chicago, IL,USA) was used for all analyses, and all statistical tests were two-sided.P-values <0.05were considered statistically significant.Results:1No significant differences in genotype and allelotype distributions ofthree SNPs (MDR1C3435Tã€G2677T/A and GSTP1A313G) were foundbetween responders and not responders of all patients.2The polymorphisms of MDR1C3435T gene and clinical outcomes ofovarian cancer patients. Separated by recurrence group and non-recurrencegroup, the genotype frequencies of C/C, C/T and T/T were35.8%,47.0%,17.2%and26.4%,57.1%,16.5%. There were no significant different betweenthe two groups. Kaplan-Meier estimates demonstrated that no statistical difference was found between the genotypes and PFS of EOC cases. Cox’smultivariate analysis suggested that there were no associations betweengenotype of MDR1C3435T and PFS of EOC patients (P>0.05). Thegenotypes frequencies distribution of MDR1C3435T SNPs did not have asignificant difference between the death and living groups. Genotypefrequencies of MDR1C3435T were31.6%,52.9%,15.4%and32.6%,48.3%,19.1%in two groups. Survival analysis showed that the genotypes frequenciesdistribution of MDR1C3435T was not associated with the OS of EOCpatients.3The polymorphisms of MDR1G2677T/A gene and clinical outcomesof ovarian cancer patients: separated by recurrence group and non-recurrencegroup, the G/G, G/T, A/G, A/A, A/T, T/T genotype frequencies of MDR1G2677T/A were17.9%,29.1%,14.2%,2.2%,18.7%,17.9%and8.8%,35.2%,13.2%,2.2%,17.6%,23.1%, respectively. Moreover, the patients carrying T/Tand the others polymorphism other than T/T were17.9%,82.1%and23.1%,76.9%, respectively. Cox’s multivariate analysis suggested that patients withT/T genotype had a shorter PFS (P=0.01,HR=3.27,95%CI=1.36-7.88). Thegenotypes frequencies distribution of MDR1G2677T/A SNPs did not have asignificant difference between the death and living group. Genotypefrequencies of MDR1G2677T/A were14.0%,33.1%,14.0%,1.5%,17.6%,19.9%and14.6%,29.2%,13.5%,3.4%,19.1%,20.2%in two groups,respectively. Survival analysis showed that the genotypes frequenciesdistribution of MDR1G2677T/A was not associated with the OS of EOCpatients.4The polymorphisms of GSTP1A313G gene and clinical outcomes ofEOC patients. Separated by recurrence group and non-recurrence group, theA/A,A/G,G/G genotype frequencies of GSTP1A313G were56.7%,38.1%,5.2%å’Œ64.8%,31.9%,3.3%. There were no significant differences betweenthe two groups. Kaplan-Meier estimates demonstrated that no significantdifference was found between the genotypes and PFS of EOC cases. Cox’smultivariate analysis suggested that there were no associations between genotype of GSTP1A313G and PFS of EOC patients (P>0.05). Thegenotypes frequencies distribution of GSTP1A313G SNPs did not have asignificant difference between the death and live group. Genotype frequenciesof GSTP1A313G were60.3%,37.5%,2.2%and59.6%,32.6%,7.9%intwo groups. Survival analysis showed that the genotypes frequenciesdistribution of GSTP1A313G was not associated with the OS of EOCpatients.Conclusion:1No significant differences in genotype and allele distributions of threeSNPs (MDR1C3435T G2677T/A and GSTP1A313G) were found betweenresponders and non-responders of all patients.2There are no associations between the MDR1C3435T polymorphismsand recovery and clinical outcomes of epithelial ovarian cancer patientstreated by platinum-based chemotherapy.3The results indicated that the MDR1G2677T/A polymorphism maycorrelate with relapse of epithelial ovarian cancer, when patients treated byplatinum-based chemotherapy in Northern China. But there was norelationship between the MDR1G2677T/A polymorphism and overallsurvival.4There are no associations between the GSTP1A313G polymorphismsand OS of epithelial ovarian cancer patients treated by platinum-basedchemotherapy. |