PET imaging of meningioma with 18F-FLT: a predictor of tumour progression

We have previously reported that PET with 3'-deoxy-3'-18F-fluorothymidine (18F-FLT) provides a non-invasive assessment of cell proliferation in vivo in meningiomas. The purpose of this prospective study was to evaluate the potential of 18F-FLT PET in predicting subsequent tumour progression in asymptomatic meningiomas. Forty-three adult patients harbouring 46 MRI-presumed (n = 40) and residual meningiomas from previous surgery (n = 6) underwent a 60-min dynamic 18F-FLT PET scan prior to radiological surveillance. Maximum and mean tumour-to-blood ratios (TBRmax, TBRmean) of tracer radioactivity were calculated. Tumour progression was defined according to the latest published trial end-point criteria for bidimensional (2D) and corresponding yet exploratory volumetric measurements from the Response Assessment of Neuro-Oncology (RANO) workgroup. Independent-sample t-test, Pearson correlation coefficient, Cox regression, and receiver operating characteristic (ROC) curve analyses were used whenever appropriate. The median follow-up time after 18F-FLT PET imaging was 18 months (range 5-33.5 months). A high concordance rate (91%) was found with regard to disease progression using 2D-RANO (n = 11) versus volumetric criteria (n = 10). Using 2D-RANO criteria, 18F-FLT uptake was significantly increased in patients with progressive disease, compared to patients with stable disease (TBRmax, 5.5 ± 1.3 versus 3.6 ± 1.1, P < 0.0001; TBRmean, 3.5 ± 0.8 versus 2.4 ± 0.7, P < 0.0001). ROC analysis yielded optimal thresholds of 4.4 for TBRmax [sensitivity 82%, specificity 77%, accuracy 78%, and area under curve (AUC) 0.871; P < 0.0001] and 2.8 for TBRmean (sensitivity 82%, specificity 77%, accuracy 78%, AUC 0.848; P = 0.001) for early differentiation of patients with progressive disease from patients with stable disease. Upon excluding patients with residual meningioma or patients with stable disease with less than 12 months follow-up, the thresholds remained unchanged with similar diagnostic accuracies. Moreover, positive correlations were found between absolute and relative tumour growth rates and 18F-FLT uptake (r < 0.513, P < 0.015) that remained similar when excluding patients with residual meningioma or patients with stable disease and shorter follow-up period. Diagnostic accuracies were slightly inferior at 76% when assessing disease progression using volumetric criteria, while the thresholds remained unchanged. Multivariate analysis revealed that TBRmax was the only independent predictor of tumour progression (P < 0.046), while age, gender, baseline tumour size, tumour location, peritumoural oedema, and residual meningioma had no influence. The study reveals that 18F-FLT PET is a promising surrogate imaging biomarker for predicting subsequent tumour progression in treatment-naïve and asymptomatic residual meningiomas.