Volumetric assessment of tumor size changes in pediatric low-grade gliomas: feasibility and comparison with linear measurements

Purpose We report a retrospective comparison between bi-dimensional RANO criteria and manual volumetric segmentation (MVS) in pediatric low-grade gliomas. Methods MRI FLAIR or T1 post contrast images were used for assessment of tumor response. Seventy patients were included in this single center study, for each patient two scans were assessed (“time 0” and “end of therapy”) and response to therapy was evaluated for both methods. Inter-reader variability and average time for volumetric assessment were also calculated. Results Fourteen (20%) of the 70 patients had discordant results in terms of response assessment between the bi-dimensional measurements and MVS. All volumetric response assessments were in keeping with the subjective analysis of tumor (radiology report). Of the 14 patients, 6 had stable disease (SD) on MVS and progressive disease (PD) on 2D assessment, 5 patients had SD on MVS and partial response (PR) on 2D assessment, 2 patients had PD on MVS and SD on 2D assessment, and 1 patient had PR on MVS and SD on 2D analysis. The number of discordant results rises to 21(30%) if minor response is integrated in the response assessment. MVS was relatively fast and showed high inter-reader concordance. Conclusion Our analysis shows that therapeutic response classification may change in a significant number of children by performing a volumetric tumor assessment. Furthermore, MVS is not particularly time consuming and has very good inter-reader concordance.