Prediction of progression-free and overall survival following temozolomide chemoradiation using FET PET-based parameters including radiomics in patients with glioblastoma

Background Early after surgery and completion of first-line radiotherapy with concomitant temozolomide, the prediction of progression-free and overall survival (PFS, OS) is of considerable interest for managing patients with glioblastoma. Methods Sixty-three newly diagnosed patients with glioblastoma (age range, 19-82 years) who received PET imaging using the radiolabeled amino acid O(2[18F]fluoroethyl)Ltyrosine (FET) after surgery or biopsy and completion of radiotherapy with concomitant temozolomide were evaluated. Static FET PET parameters, i.e., maximum and mean tumor-to-brain ratios (TBRmax, TBRmean), metabolic tumor volumes (MTV), and the dynamic FET PET parameters time-to-peak (TTP) and slope were obtained. Additionally, n=1,303 FET PET radiomics features were extracted per patient, of which 15 robust features were selected for further evaluation based on test-retest analysis. The prognostic values of FET PET parameters and radiomics features were evaluated using receiveroperatingcharacteristic (ROC) analyses regarding a favorable PFS and OS. Subsequently, univariate and multivariate survival estimates were performed to assess the prognostic value of these parameters in predicting a significantly longer PFS and OS. Results ROC analyses revealed that static parameters (i.e., TBRmax, MTV) and one radiomics feature were the most powerful parameters to predict a significantly longer PFS (all P=0.002) and OS (all P≤0.02). In addition, the dynamic parameter TTP predicted a significantly longer OS (P≤0.03), but not PFS (P>0.05). TBRmax, MTV, and one radiomics feature remained significant in multivariate survival analysis (all P≤0.03). Conclusion Our results suggest that FET PET parameters, including radiomics, are highly prognostic in patients with glioblastoma at an early stage of first-line therapy.