Fluoro-2-deoxy- d -glucose PET radiomic analysis predicts outcome in patients with soft tissue sarcoma.
AI interpretation is pending for this paper.
Open original publication →What the AI sees
Not AI summarized yet.
Research significance
Pending deeper interpretation.
Source abstract
BACKGROUND: Treatment and outcome of soft tissue sarcomas (STS) are presently guided by stage and clinicopathological factors. Radiomics, the quantitative analysis of images applied to [ 18 F]fluoro-2-deoxy- d -glucose (FDG) PET could improve prognostication and thereby treatment outcomes, especially due to its ability to quantify intratumoural metabolic heterogeneity. However, the data for its utility in STS is limited. OBJECTIVES: This study aimed to assess the prognostic value of pretreatment FDG PET-derived radiomic features for patients with STS. A retrospective analysis was performed of patients with STS who underwent FDG PET and were treated with curative intent in a sarcoma centre ( n = 44). Tumour segmentation was performed using a 40% maximum standardised uptake value threshold, and 107 radiomic features were extracted. Univariable and multivariate Cox regression evaluated radiomic features against local recurrence-free survival (LRFS), distant recurrence-free survival (DRFS), disease-free survival (DFS), and overall survival. Survival difference was then assessed using Kaplan-Meier survival curves constructed from receiver operating characteristic (ROC) derived thresholds. RESULTS: Our study demonstrated measures of tumour heterogeneity predicted outcome, the strongest predictor being, GLSZMSZNUN, which was a significant predictor on multivariate analysis of DRFS [heart rate (HR) = 0.00031, 95% confidence interval (CI) = 2.1 × 10-7-0.45, P = 0.03] and DFS (HR = 0.00047, 95% CI = 1.3 × 10-6-0.16, P = 0.01). Stratification using ROC-derived thresholds showed survival difference for LRFS ( P = 0.045), DRFS ( χ2 = 4.7, P < 0.001), and DFS ( χ2 = 8.1, P = 0.005). CONCLUSION: Multiple PET radiomic features demonstrated significant prognostic utility independent of standard clinical features and predicted survival outcomes in STS. Larger prospective multicentre cohorts are required to verify our findings and enable integration into clinical decision aids and precision medicine in STS.