This is a retrospective study of a prospective database. Patients operated on non-small cell lung cancer (NSCLC) with N1 disease by PET-CT scan were included. None of them had previous induction treatment, but they underwent standard surgical resection plus systematic lymphadenectomy.
There were 51 patients with FDG-PET-CT scan N1 disease. 44 (86.3%) patients were male with a mean age of 64.1 ± 10.8 years. Type of resection: pneumonectomy = 4 (7.9%), lobectomy/bilobectomy = 44 (86.2%), segmentectomy = 3 (5.9%). Histology: adenocarcinoma = 26 (51.0%), squamous = 23 (45.1%), adenosquamous = 2 (3.9%). Lymph nodes after surgical resection: N0 = 21 (41.2%), N1 = 12 (23.5%), N2 = 18 (35.3%). Mean ratio of the SUVmax of N1 lymph node to the SUVmax of the primary lung tumor (SUVmax N1/T ratio) was 0.60 (range 0.08–2.80). ROC curve analysis to obtain the optimal cut-off value of SUVmax N1/T ratio to predict N2 disease was performed. At multivariate analysis, we found that a ratio of 0.46 or greater was an independent predictor factor of N2 mediastinal lymph node metastases with a sensitivity and specificity of 77.8% and 69.7%, respectively.
SUVmax N1/T ratio in NSCLC patients correlates with mediastinal lymph node metastasis (N2 disease) after surgical resection. When SUVmax N1/T ratio on integrated PET-CT scan is equal or superior to 0.46, special attention should be paid on higher probability of N2 disease.