Higher predictive value of tumour and node [18F]-FDG PET metabolic volume and TLG in advanced lung cancer under chemotherapy.

ABSTRACT

INTRODUCTION: The prognostic value of standardized uptake values (SUVs) at initiation of and during chemotherapy remains controversial in lung cancer patients. However, metabolic volume (MV) and total lesion glycolysis (TLG) have shown promise in lung cancer stratification before treatment. Our aim was to define the prognostic value of MV and TLG in a homogenous group of advanced-stage lung cancer patients treated with bevacizumab and paclitaxel. MATERIALS AND METHODS: Fifty (18)F-fluorodeoxyglucose PET-computed tomography examinations were studied. SUV, MV and TLG were measured for each detectable lesion (classified as primary tumour, node or distant metastasis). MV values were added to compute tumour MV (MVt), nodal MV (MVn), metastatic MV (MVm) and whole-body MV (MVwb). TLGt, TLGn, TLGm and TLGwb were computed using the same method. Maximal SUVt, SUVn, SUVm and SUVwb were measured. Patients were stratified according to each parameter. Overall survival was compared between parameter-stratified groups using the log rank test. RESULTS: At initiation of bevacizumab and paclitaxel therapy (n=12), only TLGwb and TLGn had strong prognostic value with hazard ratios (HRs) of 11.8 and 5.6, respectively (P<0.03). During treatment with bevacizumab and paclitaxel (n=38), SUVwb (HR 4.9), MVwb (HR 4.9), TLGwb (HR 8.9), SUVt (HR 15.1), MVt (HR 15.6) and TLGt (HR 15.6) had significant prognostic value (P<0.05). Metastasis measurements showed no significant prognostic value. CONCLUSION: MV and TLG are powerful prognostic factors in advanced-stage lung adenocarcinoma being treated with chemotherapy. Focusing on the intrathoracic primary tumour increases the prognostic value of PET in these patients.