DOI: 10.1007/s00259-017-3777-2Pages: 2084-2093

Grading and outcome prediction of pediatric diffuse astrocytic tumors with diffusion and arterial spin labeling perfusion MRI in comparison with 18F–DOPA PET

1. Istituto Giannina Gaslini, Neuroradiology Unit

2. Ente Ospedaliero Ospedali Galliera, Nuclear Medicine Unit

3. Ente Ospedaliero Ospedali Galliera, Clinical Trial Unit, Scientific Directorate

4. Istituto Giannina Gaslini, Pathology Unit

5. Istituto Giannina Gaslini, Neurosurgery Unit

6. Istituto Giannina Gaslini, Neuro-oncology Unit

Correspondence to:
Giovanni Morana
Tel: (39) 010 56362516




The aim of this study was to investigate MRI-derived diffusion weighted imaging (DWI) and arterial spin labeling (ASL) perfusion imaging in comparison with 18F–dihydroxyphenylalanine (DOPA) PET with respect to diagnostic performance in tumor grading and outcome prediction in pediatric patients with diffuse astrocytic tumors (DAT).


We retrospectively analyzed 26 children with histologically proven treatment naïve low and high grade DAT who underwent ASL and DWI performed within 2 weeks of 18F–DOPA PET. Relative ASL-derived cerebral blood flow max (rCBF max) and DWI-derived minimum apparent diffusion coefficient (rADC min) were compared with 18F–DOPA uptake tumor/normal tissue (T/N) and tumor/striatum (T/S) ratios, and correlated with World Health Organization (WHO) tumor grade and progression-free survival (PFS). Statistics included Pearson’s chi-square and Mann-Whitney U tests, Spearman’s rank correlation, receiver operating characteristic (ROC) analysis, discriminant function analysis (DFA), Kaplan-Meier survival curve, and Cox analysis.


A significant correlation was demonstrated between rCBF max, rADC min, and 18F–DOPA PET data (p < 0.001). Significant differences in terms of rCBF max, rADC min, and 18F–DOPA uptake were found between low- and high-grade DAT (p ≤ 0.001). ROC analysis and DFA demonstrated that T/S and T/N values were the best parameters for predicting tumor progression (AUC 0.93, p < 0.001). On univariate analysis, all diagnostic tools correlated with PFS (p ≤ 0.001); however, on multivariate analysis, only 18F–DOPA uptake remained significantly associated with outcome (p ≤ 0.03), while a trend emerged for rCBF max (p = 0.09) and rADC min (p = 0.08). The combination of MRI and PET data increased the predictive power for prognosticating tumor progression (AUC 0.97, p < 0.001).


DWI, ASL and 18F–DOPA PET provide useful complementary information for pediatric DAT grading. 18F–DOPA uptake better correlates with PFS prediction. Combining MRI and PET data provides the highest predictive power for prognosticating tumor progression suggesting a synergistic role of these diagnostic tools.

To access the full text, please Sign in

If you have institutional access, please click here

  • Accepted: Jul 10, 2017
  • Online: Jul 27, 2017

Article Tools