EANM'16 - Annual Congress of the European Association of Nuclear Medicine October 15 – 19, 2016 Barcelona, Spai

AIM: 18F-F PET/CT plays an important role in hilar and mediastinal lymph node staging of patients with non-small cell lung cancer (NSCLC), because it is more accurate than contrast-enhanced computed tomography (ceCT).According to the literature, PET/CT for lymph node staging has 80- 90% sensitivity, 85-95% specificity, and in patients with peripheral NSCLC has greater NPV than ceCT. The purpose of this study was to assess the accuracy of PET/CT for hilar and mediastinal lymph node staging of NSCLC patient eligible for potentially radical surgery according to the T stage, and to identify risk factors correlated to lymph node involvement. MATERIAL S AND METHODS: We reviewed 67 patients (73% male) with NSCLC [adenocarcinoma (ADK), n=44; squamous cell carcinoma (SCC), n=16; miscellaneous, n=7] and node status assessed by surgical pathological findings. From September 2014 to January 2016 all patients underwent preoperative ceCT and PET/CT. The maximum standardized uptake value (SUVmax) and metabolic tumor volume (MTV) of the primary lung cancer were measured. Mediastinal lymph node positivity on PET/TC was defined as focally increased FDG uptake higher than mediastinal blood pool.Multivariate analysis was performed separately in patients with ADK and SCC, for the following factors associated with lymph node metastasis: age, gender, tumor location, grading, SUVmax, MTV. RESULTS: The primary tumor SUVmax was significantly higher in SCC patients (range : 2.7-51) compared with ADK (range:1.5-31). The sensitivity, specificity, accuracy, NPV, PPVof regional lymph node metastasis were respectively 47%, 82%, 74%, 82%, 47% , with PET/CT and 41%, 63%, 58%, 75%, 29%, with ceCT. Multivariate analysis showed that SUVmax (p= 0.0028) and MTV (p=0.0097) were significant predictors of lymph node metastasis only in ADK group. Moreover, in ADK the ROC curve showed an area under curve better for SUVmax, 0.859 (95%CI: 0.717-0.947) than for MTV. No significant differences were found in the SCC group, possibly because of small sample size. For PET/CT, the falsenegative incidence (N1 or N2) was 13.4% (14.9% for ceCT). Interestingly, occult N2 metastaseswere found in 3 of 44ADKpatients (7%) and all these occurred in peripheral ADK. CeCT showed a trend to overstaging (falsepositive rate was twice that of PET/CT). CONCLUSION: In patients with lung ADK the SUVmax andMTV predicts mediastinal nodal involvement. In patients with peripheral ADK with high SUVmax and MTV, and PETnegative mediastinal lymph nodes, our results suggest the need of presurgical mediastinal lymph node biopsy to rule out false-negative N2 status.