Prognostic value of quantitative FDG PET/CT uptake metrics in non-small cell lung cancer

Abstract

Objectives The TNM staging system might not be accurate enough as prognostic factor in NSCLC patients. The prognostic significance of SUV remains controversial. This retrospective study evaluates the prognostic value of the SUV, TLG and MTV in NSCLC.

Methods This study includes 78 patients with newly diagnosed NSCLC (stage I: 24.4 %, II: 9 %, III: 41 % and IV: 25.6 %). All patients underwent an 18FDG PET/CT at baseline. SUVmax, SUVmean and MTV were derived after semi-automatic delineation of the most active metabolic intrathoracic lesion using a 50% of SUVmax contour. SUVs were corrected for lean body mass and glycaemia. TLG was defined as SUVmean × MTV.

Results Median follow up was 21 months. Median MTV, TLG50 and SUVmax were 9.28 (3.95 -21.07) ml, 65.80 (17.81 - 180,50) ml and 9.15 (6.00-12.84) respectively. Overall and progression free survival (OS, PFS) were significantly higher in patients with uptake below median MTV, TLG50 and SUVmax.Univariate analysis revealed gender (HR 0.3 95%CI 0.15-0.65 for OS and HR 0.42 95%CI 0.22-0.81) for PFS) and stage (HR 2.34 95%CI 1.64-3.35 for OS and HR 2.17 95%CI 1.56 - 3.03 for PFS) as additional prognostic factors. Based on multivariate analysis most important prognostic factors for OS were TLG50 (HR 2.23 95%CI 1.20-4.15), stage (HR 2.48 95%CI 1.68-3.65) and gender (HR 0.3 95%CI 0.14-0.64) and for PFS SUVmax (HR 2.04 95%CI 1.10-3.77), stage (HR 2.4 95%CI 1.65 - 3.49) and gender (HR 0.30 95%CI 0.15-0.59).

Conclusions TLG50 and SUVmax are independent prognostic factors for OS and PFS, respectively in NSLC patients. Stratification of patients with the same TNM stage using TLG and/or SUVmax may further improve outcome, but need to be substantiated with prospective studies and assessment of optimal cutoff values.