RESUMEN
<p><b>BACKGROUND</b>The outcome of surgical treatment of non-small-cell lung cancer (NSCLC) remains poor. In many patients the biological behavior of NSCLC does not follow a definite pattern, and can not be accurately predicted before treatment. (18)F-fluoro-2-deoxy-glucose ((18)F-FDG) uptake on positron-emission tomography (PET) is associated with the aggressiveness of NSCLC. The present study focused on the role of (18)F-FDG uptake in predicting the outcome of surgically treated patients with NSCLC.</p><p><b>METHODS</b>A retrospective analysis was made of 82 patients who underwent complete resection and preoperative FDG PET. The maximum standardized uptake value (SUV(max)), in addition to five clinicopathological factors and three biomolecular factors, which could possibly influence survival, was compared for possible association with patients' recurrence and survival, by the Log-rank test in univariate analysis and the Cox proportional hazards model in multivariate analysis. The association between SUV(max) and other factors was also analyzed.</p><p><b>RESULTS</b>Patients with SUV(max) more than 11 had a disease-free survival and overall survival shorter than patients with SUV(max) less than 11 in univariate analyses (P < 0.001, P = 0.002). In the multivariate analysis, SUV(max) (dichotomized by 11) was the only significant predictor for tumor recurrence. TNM stage and SUV(max) (dichotomized by 11) were independent predictors for the overall survival. Associations of SUV(max) with p53 overexpression, proliferating cell nuclear antigen (PCNA) labeling index and microvascular density of the tumor were significant in the entire group.</p><p><b>CONCLUSIONS</b>(18)F-FDG uptake on PET may be used to noninvasively assess biological aggressiveness of NSCLC in vivo, identifying the surgically-treated patients with poor prognosis who could benefit from additional therapy.</p>