Correlation identification between radiotracer uptake and pathological features in non-small cell lung cancer (NSCLC)

Background: SUV values from PET/CT can reflect disease progression; however, there are few detailed and comprehensive studies.In this study, we hope to provide data to complement the analysis of diagnostic indicators for NSCLC, LUAD, and LUSC by analyzing the correlation between SUVmax and various pathological parameters.Patients and Methods: Patients (n=298) with lung lesions were retrospectively studied and clinicopathological index of these patients were collected.Radiomics texture features were extracted by PET/CT scanning and histological features of each patient was collected.The values of SUVmax were obtained and the inner correlation was analyzed.Evaluation and scoring were performed by calculating area under the curve (AUC), sensitivity, specificity and accuracy.Results: PET data were counted and analyzed for positive/negative relationship between SUV value and T stage and histology (P<0.05).Mean SUVmax was 13.326.41mm; the SUVmax was 14.015.72 for male and 11.197.95for female.The results showed that LUSC tumors were smaller and more homogeneous, but with more uptake and greater PET variability.In contrast, LUAD have lower and weaker uptake, variability and homogeneity.Conclusion: By meticulously grouping nearly 300 NSCLC samples, AUC values were calculated to indicate the diagnostic value in NSCLC, LUAD and LUSC.It provides ideas and basis for pathological staging analysis of NSCLC.