DNA Repair Gene Polymorphisms Predict Favorable Clinical Outcome in Advanced Non–Small-Cell Lung Cancer

Aristea Kalikaki; Maria Kanaki; Helen Vassalou; John Souglakos; Alexandra Voutsina; Vassilis Georgoulias; Dimitris Mavroudis


Clin Lung Cancer. 2009;10(2):118-123. 

In This Article

Patients and Methods


Patients with histologically documented NSCLC treated with first-or second-line platinum-based chemotherapy were enrolled in this retrospective analysis. Pretreatment evaluation included medical history, physical examination, complete blood cell count with dif-ferential, routine chemistry, and computed tomography scan of the chest, abdomen, and central nervous system. Assessment of tumor response was carried out by computed tomography scan after every three chemotherapy cycles according to the Response Evaluation Criteria in Solid Tumors (RECIST). Hematologic and nonhematologic toxicities were evaluated after each chemotherapy cycle. For the pur-pose of this analysis, the worst toxicity grade was recorded according to the National Cancer Institute Common Toxicity Criteria (NCI-CTC) criteria. The analysis of all blood samples was carried out in a blinded fashion relative to the patients' clinical outcome.

Table 1 summarizes the clinicopathologic characteristics of the 119 patients enrolled in the study. Among the 119 NSCLC patients, 69 (58%) and 50 (42%) received platinum-based chemo-therapy as first- and second-line treatment, respectively. Charac-teristics are shown for all patients and for those in each treatment group (Table 1). The median age of patients was 61 years (range, 39–85), and 101 (84.8%) patients were male. The majority of patients (66.4%) had stage IV disease. Histologic types included adenocarcinoma (60.5%), squamous-cell carcinoma (23.5%), and others (15.9%). There was no statistically significant difference in patient characteristics between those who received platinum-based chemotherapy as first- or second-line treatment (Table 1). Overall survival was calculated from the day of starting first-line treatment to the day of the last follow up (death or clinic visit). Overall, the median follow-up time was 28.5 months (range, 0.4-60), and the median survival was 11.3 months.

DNA Isolation and Genotyping

Blood samples in EDTA were obtained from all patients at the time of enrollment to the study protocol. DNA was extracted from peripheral blood mononuclear cells (PBMC) using the MasterPure Complete DNA/RNA Purification kit (EPICENTRE; Biotechnol-ogies, Madison, WI) according to the manufacturer's instructions. All genotypes were determined using a PCR-RFLP assay.[15,16] The polymerase chain reaction (PCR) primers and the necessary restric-tion enzymes for restriction fragment length polymorphism (RFLP) are given in Table 2. For quality purposes, 20% random samples of each genotype were also determined by direct sequencing analysis. In particular, ERCC1 C8092A and XPD D312N polymorphisms were confirmed for all subjects by sequencing analysis.

Statistical Analysis

The Χ2 test was used to compare the frequencies of different genotypes between patients and treatment response. Patients with complete or partial response were characterized as ‘‘responders,'' whereas patients with stable or progressive disease were charac-terized as ‘‘nonresponders.'' We performed multivariate (logistic regression) analyses of the various contributing factors for response, such as gender, tumor histology, disease stage, and polymorphisms. Odds ratios (OR) and their 95% confidence intervals (CI) were calculated using the binary logistic regression model. A Χ2 test was also used to test whether the alleles at each locus were conformed to Hardy-Weinberg equilibrium. The degree of linkage disequilib-rium was quantified by the Lewontin standardized disequilibrium coefficient D' (LDA software) to evaluate allelic association among SNPs.[17] The genotype effect on patient's survival was analyzed using the Kaplan-Meier method. Comparison of survival curves was made using the log-rank and Breslow tests. Hazard ratios (HRs) were determined using the Cox proportional hazard model. A probability level of less than 0.05 was used as the criterion of significance. All analyses were performed using the SPSS software (Version 16, SPSS Inc., Chicago, IL).


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.