947

Background. Circulating DNA in plasma can represent a useful marker for clinical management of lung cancer. We assessed the screening performance and the prognostic value of plasma DNA levels in a prospective cohort of heavy smokers monitored with annual spiral-CT for 5 years.
 Methods. Plasma DNA levels were determined through real-time quantitative PCR amplification of human telomerase reverse transcriptase (hTERT) gene in 1035 volunteers of a screening trial at baseline and at time of lung cancer diagnosis. Screening performance of the assay was calculated through the area under the receiver-operating characteristic curve (AUC-ROC). Kaplan-Meier analyses were computed for association with prognosis.
 Results. Median baseline concentration of plasma DNA was similar in CT-detected patients and cancer-free controls (AUC-ROC 0.496, p= 0.9330) whereas was slightly higher at lung cancer diagnosis (AUC-ROC 0.607, p= 0.0369). Plasma DNA was higher in tumors detected during the first year (AUC-ROC 0.80, P<0.0001) and in Stage II-IV tumors detected during the first 2 years of screening (AUC-ROC 0.87, P<0.0001). A longitudinal study of plasma DNA levels showed increased values approaching to lung cancer diagnosis (p=0.0010). Plasma DNA was significantly associated with reduced 5 years survival (p=0.0089).
 Conclusions. This study shows a limited discriminatory power of plasma DNA levels for the identification of lung cancer patients in a CT-screening setting, with better performance in particular subsets. Changes in plasma DNA levels along follow-up appointments might be informative of lung cancer onset over time. Plasma DNA concentration at surgery has prognostic value in CT-detected patients.

99th AACR Annual Meeting-- Apr 12-16, 2008; San Diego, CA