FDG-PET in immunocompetent patients with primary central nervous system lymphoma: differentiation from GBM and correlation with DWI

Objectives This study retrospectively evaluates the relation between F-18 FDG uptake using PET/CT and the apparent diffusion coefficient (ADC) in patients with Primary central nervous system lymphoma (PCNSL). Besides, the appropriate cut-off value of FDG uptake for distinguishing PCNSLs from glioblastoma multiforme (GBM) was also explored.

Methods We reviewed 91 patients (with 36 PCNSLs and with 55 GBMs) underwent FDG PET scan at onset. Among PCNSLs, 24 patients with 28 lesions also underwent diffusion-weighted imaging. The interval between MR imaging and PET/CT imaging for the same patient ranged from 1 day to 12 days. The maximum standardized uptake value (SUVmax) was calculated to assess the tumor FDG uptake. On ADC map, the ADC ratio was calculated by dividing the minimum ADC of the tumor by the mean ADC of contralateral normal white matter. Pearson's correlation analysis was used to assess the relation between the SUVmax and the ADC value ratio for those PCNSL patients. Receiver operating characteristic (ROC) curve was generated from the SUVmax.

Results Inverse correlation between the SUVmax and the ADC ratio was found for PCNSL cases (P<0.05, r =-0.758). The SUVmax values of PCNSLs were significantly higher than those of the GBMs. ROC analysis indicated area under the curve was 0.938. The sensitivity with a cut-off point of 13.59 was 86.4 %, and the specificity was 93.1 %.

Conclusions An inverse correlation between the SUVmax and the ADC ratio was found. Both FDG PET and DWI are useful modalities to evaluate PCNSL from the perspectives of tumor metabolic activity and cellular density, respectively. The SUVmax might be more useful for differentiating PCNSL from GBM.