Predicting Gene Promoter Methylation In Lung Tumors Through Examination Of Sputum And Serum.

7208 Background: Personalized medicine may be a key approach in improving survival for advanced lung cancer. Success for this approach is seen with the response of cancer patients with an activating mutation within the epidermal growth factor receptor to the growth fact inhibitor, gefitinib. Genes involved in all types of normal cell function are targeted for inactivation by promoter hypermethylation during lung cancer development. This makes gene-specific promoter hypermethylation an attractive target for novel treatment strategies. One obstacle to targeted therapy is accessibility of tissue from peripheral tumors for methylation analysis. The purpose of this study was to determine the predictive power of sputum and serum to detect NSCLC through analysis for methylation of genes in these fluids. Methods: Tissue, serum, and sputum were obtained from 72 stage III NSCLC patients participating in a Phase III trial of Carboplatin, Paclitaxel and Radiotherapy, With or Without Thalidomide (ECOG 3598). Methylation specific PCR assessed methylation of the p16, MGMT, DAPK, RASSF1A, PAX5-α, PAX5-β, H-cadherin, and GATA5 genes. Results: At least one gene was methylated in 89% of tumors. Prevalence for methylation ranged from 15–47% and did not differ between gender. Methylation of ≥ 3 genes was seen in ∼50% of tumors. The agreement between sputum or serum in classifying methylation status of any gene in the tumor ranged from 54– 69%. The reduced sensitivity was presumably due in part to absence of tumor DNA in the biological fluids of methylation positive tumors. Logistic regression models revealed 3.1 (CI, 1.2, 8.2) and 4.2 (1.1, 16.1) increased odds for methylation of the p16 and MGMT gene, respectively in tumors if the serum or sputum was positive for methylation. The effect of tumor histology for predicting methylation using sputum and serum is being examined. Conclusion: These studies demonstrate for genes such as p16 and MGMT the possibility of interrogating biological fluids to predict for methylation in the primary tumor. (Supported by CA-89551). No significant financial relationships to disclose.