Value of dynamic 18F-FDG PET/CT semi-quantitative parameters in the diagnoses of non-small cell lung cancer subtypes and lymph node metastasis / 中华核医学与分子影像杂志

ABSTRACT

Objective:To explore the values of dynamic 18F-FDG PET/CT semi-quantitative parameters in the differentiation of histological subtypes of non-small cell lung cancer (NSCLC) and the diagnosis of lymph node metastasis. Methods:Twenty-three patients (10 males, 13 females, age (61.5±8.1) years) with NSCLC in Shengjing Hospital of China Medical University were prospectively enrolled between October 2020 and June 2021. All patients underwent 40 min of dynamic PET/CT scan and static scan at 60 min post-injection of 18F-FDG. SUV max, SUV mean, metabolic tumor volume (MTV) and total lesion glycolysis (TLG) of primary lesions at different stages of PET/CT imaging were evaluated. SUV max of lymph nodes were also analyzed. The histopathological results were considered as the gold standard. Parameters of primary lesions and lymph nodes at different PET/CT imaging stages in different groups were compared by independent-sample t test or Mann-Whitney U test. The diagnostic efficiencies of those parameters were tested by ROC curve and compared by DeLong test. Results:A total of 24 lesions in 23 patients were pathologically confirmed as NSCLC, of which 11 were squamous cell carcinoma and 13 were adenocarcinoma. The SUV max (13.5±3.4 vs 9.6±5.1), SUV mean (8.1±2.2 vs 5.8±3.2) at the third stage of dynamic PET/CT (33-40 min) between squamous cell carcinoma and adenocarcinoma were both significantly different ( t values: 2.20, 2.10, P values: 0.039, 0.048). SUV max, TLG of static PET/CT between the 2 groups were also different ( t=2.22, P=0.037; z=-2.17, P=0.030). ROC curves of those parameters showed that AUCs were between 0.727 and 0.762. Fourteen patients underwent surgery and lymph node dissection. According to the pathological results, lymph nodes in 15 areas were metastatic, the rest in 70 areas were benign. There were significant differences in SUV max at the second stage of dynamic PET/CT (19-26 min) between metastatic and benign lymph nodes (4.0(2.8, 6.2) vs 2.3(1.8, 2.8); z=-4.31, P<0.001), as well as SUV max at the third stage of dynamic PET/CT and static PET/CT between the 2 groups ( z values: -4.59, -4.10, both P<0.001). ROC curves of those 3 parameters showed that the AUCs were 0.856, 0.879 and 0.838 respectively, with no significant differences ( z values: 0.78, 0.34, 1.27, P values: 0.434, 0.734, 0.205). The sensitivity, specificity, and accuracy of the SUV max of the third dynamic imaging stage were 13/15, 90.0%(63/70) and 89.4%(76/85), respectively. Conclusion:Dynamic 18F-FDG PET/CT semi-quantitative parameters have certain clinical significance in the identification of histological subtypes of NSCLC and the diagnosis of metastatic lymph nodes, especially the third stage dynamic imaging has a better diagnostic performance.