Absence of a Relationship between Tumor 18F-fluorodeoxyglucose Standardized Uptake Value and Survival in Patients Treated with Definitive Radiotherapy for Non–Small-Cell Lung Cancer

INTRODUCTION A recent meta-analysis suggested that patients with non-small-cell lung cancer (NSCLC) whose primary tumors have a higher standardized uptake value (SUV) derived from F-fluorodeoxyglucose positron emission tomography (PET) have a worse prognosis in comparison with those with tumors with lower values. However, previous analyses have had methodological weaknesses. Furthermore, the prognostic significance over the full range of SUV values in patients treated nonsurgically remains unclear. The aim of this retrospective study was to investigate the relationship between survival and maximum SUV (SUV(max)) analyzed as a continuous variable, in patients with NSCLC, staged using PET/computed tomography (CT) and treated with radiotherapy with or without chemotherapy. METHODS Eligible patients had a histological diagnosis of NSCLC, were treated with radical radiotherapy with or without chemotherapy as their primary treatment, and had pretreatment PET/CT scans. SUV(max), defined as the maximum pixel SUV value retrieved from the primary tumor, was analyzed primarily as a continuous variable for overall survival. RESULTS Eighty-eight patients met eligibility criteria: stage I, 19; stage II, 10; and stage III, 59. Median SUV(max) was 15.0 (range, 2.5-56). Higher stage was associated with higher SUV(max) values (p = 0.048). In univariate analysis, there was no evidence of a prognostic effect of SUV(max) (hazard ratio per doubling = 0.83; 95% confidence interval, 0.62-1.11; p = 0.22). Analyzing SUV(max) as a dichotomous variable (median cut point = 15.0), the hazard ratio (high: low) for risk of death was 0.71, with p = 0.18 (95% confidence interval, 0.44-1.15). CONCLUSIONS In this cohort of patients, increasing SUV(max) derived from F-fluorodeoxyglucose-PET/CT was associated with increasing tumor, node, metastasis (TNM) stage. We found no evidence of an association of increasing SUV(max) with a shorter survival. Previous reports of an association between prognosis and SUV(max) may partly be the result of methodological differences between this study and previous reports and an association between stage and SUV(max).