8003 Background: FDG-PET already showed to be an early assessment of anticancer treatment effectiveness in different solid tumours after the first cycle of chemotherapy. Aim of this pilot study in patients (pts) with advanced NSCLC was to test if a second 18F-FDG PET repeated two weeks after the start of chemotherapy is predictive of subsequent response and survival. Patients and Methods: Twenty two pts with clinical stage IIIB (n=1) or IV (n=21) NSCLC receiving front line platinum-based chemotherapy within a clinical trial were consecutively and prospectively enrolled. Male/female ratio was 18/4, adenocarcinoma 11, squamous 6, NSCLC-NOS 5. All pts underwent 18F-FDG PET scan at baseline (PET1), 15 days after the beginning (PET2) and at the end of chemotherapy. Assessment of disease by spiral CT scan of the chest and abdomen was performed at baseline and every other cycle. Tumour response was evaluated according to RECIST criteria and analysis of PET scans (including the % change in the standardized uptake values [SUV] between baseline and subsequent FDG PET in the primary tumour (ΔSUV) as well as an overall assessment of all visible lesions) completed and recorded before response evaluation by CT. Results: Pts were dichotomized in two groups (responders and non-responders) based on the overall assessment of FDG PET changes in all visible lesions between PET1 and PET2. Among PET responders, TC scan performed after 2 cycles showed objective response (OR) in 55% of the cases (6/11) while only 1 OR (10%) in non-responders (Fisher exact test, p=0.04). Mean ΔSUV in the primary tumour in PET responders was -40% and in non responders -9%. Overall PET response overestimate CT response in 6/21 cases while underestimate it in 3/21 cases. Adopting different cut-off values for baseline SUV or ΔSUV of the primary lesion no correlation was found with progression-free or overall survival. A non significant trend for improved PFS was found in overall responders (25 weeks) vs non-responders (19 weeks). Data on OS are still immature. Conclusion: This pilot study indicates that in advanced NSCLC early PET restaging during cycle 1 may predicts subsequent OR to CT scan after 2 cycles. This findings are worth of a prospective validation in a larger study. Author Disclosure Employment or Leadership Consultant or Advisory Role Stock Ownership Honoraria Research Expert Testimony Other Remuneration AstraZeneca, Bayer, Eli Lilly, Roche, sanofi-aventis