Abstract 1225P
Background
Tumor heterogeneity and systemic inflammation are considered as important prognostic factors in cancers. Texture analysis is a method for extracting information about tumor heterogeneity from diagnostic images. The aims of this study were to evaluate the relationship between texture features of primary tumors, neutrophil to lymphocyte ratio (NLR), platelet to lymphocyte ratio (PLR) and modified Glasgow prognostic score (mGPS) of patients with non-small cell lung cancer (NSCLC) and and to evaluate their prognostic values for predicting clinical outcomes.
Methods
FDG PET/CT images of 90 patients with NSCLC prior to surgery were analyzed retrospectively. We acquired data related to metabolism (SUVmax, SUVmean, MTV, TLG) and texture features of primary tumors. The relationship of metabolic and textural features with NLR, PLR and mGPS categories were evaluated with Mann Whitney U test. Prognostic values of imaging features and systemic inflammation markers for predicting OS were assessed by using the univariate and multivariate Cox proportional hazards regression models.
Results
MTV and TLG had significantly higher values in patients with high NLR (> 3) than in those with low NLR (≤3), in patients with high PLR (> 150) than in those with low PLR (≤150) and in patients with high mGPS (≥ 1) compared to those with mGPS = 0 (p<0.05). Textural heterogeneity features showed significantly higher values in patients with high NLR, high PLR and high mGPS compared to those with low NLR, low PLR and mGPS = 0, respectively (p<0.05). Textural homogeneity (angular second moment) also showed significantly lower values in patients with high NLR compared to those with low NLR (p=0.025). The median follow-up duration after diagnosis was 48.5 months (5-124 months). In the multivariate Cox regression model, grey level non-uniformity (GLNU) and mGPS were found as independent prognostic factors for OS (HR, 1.005, 95% CI, 1.002-1.008, p=0.004; HR, 4.23, 95% CI, 1.9-9.5, p<0.001, respectively).
Conclusions
Metabolic activity and textural heterogeneity of tumors were significantly associated with systemic inflammatory response in patients with NSCLC. PET textural heterogeneity feature (GLNU) and mGPS were found as independent predictors of mortality.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.