open access

https://hdl.handle.net/2115/59751

The diagnostic role of F-18-FDG PET for primary central nervous system lymphoma

英語

ROC curve

2014-08

Springer

Annals of Nuclear Medicine

28(7)

603-609

F-18-FDG PET has become one of the most important methods for studying malignant lymphoma, but its diagnostic role for primary central nervous system lymphoma (PCNSL) has not been established. The aim of this study was to determine the appropriate cut-off values of FDG uptake and to investigate how corticosteroid administration influences PCNSL. We retrospectively reviewed 82 patients with contrast-enhanced brain tumors who underwent an FDG PET scan at onset, including 19 PCNSLs. FDG uptake of the lesion was assessed by the maximum standardized uptake value (SUVmax) and the ratio of tumor to normal contralateral cortex activity (T/N ratio). Receiver operating characteristic (ROC) curves were generated from the SUVmax and T/N ratios. To investigate the influence of corticosteroid application before a FDG PET scan, we evaluated the association between the FDG uptake of the lesion and the cumulative dose of corticosteroid administration on 13 PCNSL patients who had received steroid treatment before an FDG PET examination. The mean FDG SUVmax and T/N ratio of PCNSLs were 22.6 and 2.79, respectively, and these values were significantly higher than those of the other malignant brain tumors. ROC analysis indicated that the evaluation of FDG uptake using the T/N ratio was more reliable than the SUVmax with respect to the differential diagnosis. When PCNSL patients went without steroid application before FDG PET, the accuracy of the T/N ratio with a cut-off point of 2.0 was 91.1 %, the sensitivity was 94.7 %, and the specificity was 87.3 %. Although there are no significant differences in the FDG T/N ratio for PCNSL patients with or without steroid treatment, a negative correlation was found between the T/N ratio and cumulative dose of corticosteroid before PET study (r = -0.71, p = 0.032). We concluded that the T/N ratio was superior to SUVmax for FDG uptake assessment as for distinguishing PCNSLs from other malignant brain tumors; the appropriate T/N ratio cut-off point was 2.0. In addition, FDG uptake could be influenced by cumulative doses of corticosteroid before a PET scan, and thus this fact should be taken into consideration when evaluating FDG PET for PCNSL diagnosis.

学術雑誌論文

Accepted Manuscript

The final publication is available at link.springer.com

0914-7187