Role of [18F]fluoroethyl-L-tyrosine PET for differentiation of radiation necrosis from recurrent brain metastasis

Objectives To investigate the potential of [F-18]fluoroethyl-L-tyrosine (FET) PET for differentiation of radiation necrosis from recurrent brain metastasis after radiation therapy.

Methods 31 patients with single or multiple contrast-enhancing brain lesions (n=40) on MRI after radiation therapy of brain metastases were investigated with dynamic FET PET. Mean tumor/brain ratios (TBR) of FET uptake was determined. Time activity curves (TAC) were generated and time-to-peak (TTP) was calculated. Furthermore, TACs of each lesion were assigned to one of the following curve patterns: (I) constantly increasing FET uptake; (II) FET uptake peaking early (TTP≤20min) followed by a plateau; and (III) FET uptake peaking early (TTP≤20min) but followed by a constant descent. The diagnostic accuracy of both TBR values and curve patterns to identify recurrent brain metastasis was evaluated by ROC analyses using subsequent histological data (11 lesions) or clinical course and MRI findings (29 lesions) as reference.

Results TBR values were significantly higher in recurrent metastasis (n=19) than in radiation necrosis (n=21) (2.2±0.5 vs. 1.7±0.2; P<0.001). The diagnostic performance of FET PET for the identification of recurrent brain metastasis yielded an accuracy of 85% using TBR (AUC 0.87, SN 81%, SP 89%, cut-off 1.9) and 92% for curve patterns II/III vs. curve pattern I (AUC 0.92, SN 84%, SP 100%). Highest accuracy (93%) to diagnose recurrent metastasis was obtained when both conditions TBR>1.9 and curve pattern II/III were present (AUC 0.96, SN 95%, SP 91%).

Conclusions The data suggest that the combined evaluation of the mean TBR of FET uptake and pattern of TAC can differentiate brain metastasis from radionecrosis with a high accuracy