Radiology reporting of low-grade glioma growth underestimates tumor expansion

Abstract

Background: An important aspect in the management of patients with diffuse low-grade gliomas (LGGs) involves monitoring the lesions via serial magnetic resonance imaging (MRI). However, radiological interpretations of LGG interval scans are often qualitative and thus difficult to use clinically. Methods: To contextualize these assessments, we retrospectively compared radiological interpretations of LGG growth or stability to volume change measured by manual segmentation. Tumor diameter was also measured in one, two, and three dimensions to evaluate reported methods for assessment of glioma progression, including RECIST criteria, Macdonald/RANO criteria, and mean tumor diameter/ellipsoid method. Results: Tumors evaluated as stable by radiologists grew a median volume of 5.1 mL (11.1%) relative to the comparison scan, and those evaluated as having grown had a median volume increase of 13.3 mL (23.7%). Diameter-based measurements corresponded well but tended to overestimate gold standard segmented volumes. In addition, agreement with segmented volume measurements improved from 17.6 ± 8.0 to 4.5 ± 5.8 to 3.9 ± 3.6 mm for diameter and from 104.0 ± 96.6 to 25.3 ± 36.8 to 15.9 ± 21.3 mL for volume with radiological measurements in one, two, and three dimensions, respectively. Measurement overestimation increased with tumor size. Conclusions: Given accumulating evidence that LGG volume and growth are prognostic factors, there is a need for objective lesion measurement. Current radiological reporting workflows fail to appreciate and communicate the true expansion of LGGs. While volumetric analysis remains the gold standard for assessment of growth, careful diametric measurements in three dimensions may be an acceptable alternative.