Comparison between MRI-derived ADC maps and FLT-PET in pre-operative glioblastoma.

Background And Purpose: Among principal MRI sequences used for a better pre-therapeutic characterization of glioblastoma (GBM), DWI-derived ADC is expected to be a good parameter for the evaluation of cellularity, due to restricted water diffusivity. We aimed here to compare ADC maps to FLT-PET, a proliferation tracer, in GBM cases.

Materials And Methods: Patients underwent FLT-PET, followed by multiparametric magnetic resonance imaging (MRI) just prior to surgery. We analysed in this study twenty GBM confirmed patients. The 5th percentile (5p) of the ADC values were thresholded to define the ADC ROI, while the 95th percentile (95p) of the SUV FLT values were used to define the FLT ROI. The statistical and spatial correlations between these two groups of ROIs were analyzed.

Results: We did not observe any significant correlations between ADC and FLT cut-off values (R=0.0285), neither between ADC and FLT ROIs (mean Dice=0.09±0.12). Mean ADC values in the FLT defined ROI were significantly higher than the values in the ADC ROI (P<0.001). Mean FLT values in the FLT ROI were significantly higher than the values in the ADC ROI (P<0.001).

Conclusions: When comparing ADC maps to FLT uptake, we did not observe significant anatomical overlap nor correlation, between the regions of low ADC and high FLT disabling to clearly link ADC values to cellular proliferation. The exact significance of ADC maps in GBM has yet to be elaborated.