Abstract
1916
Objectives Determination of the ability of [18F]-fluorothymidine (FLT) to image cellular proliferation using positron emission tomography (PET) in comparison to malignancy parameters obtained from multiple multimodal image-guided brain biopsies.
Methods Thirteen patients with a resectable suspected supratentorial malignant glioma were imaged preoperatively after the injection of [18F]-FLT. PET volumes were delineated from standardized uptake value (SUV) calculation and MRI volumes from a gadolinium enhanced T1-weighted sequence. PET and MRI matched and mismatched volumes were superimposed on neuronavigation image with different colors. Before the opening of dura mater, different specimens were obtained from the different volumes. Histopathological malignancy parameters (MIB-1 proliferative index (PI), cellular atypies, mitose number, necrosis, endothelial hyperplasia) and the final pathological diagnosis of malignant glioma (grade 3 or 4) were rated blindly and compared to imaging data.
Results A total of 149 biopsies (median = 11/patient) were studied : 26 samples from PET - / MRI + regions , 56 from PET + / MRI - regions, 48 from PET + / MRI + regions, 19 from PET - / MRI - regions. Fourty-three % of 56 PET+/MRI- samples and 81 % of 48 PET+/MRI+ samples corresponded to grade 3 or 4. Among 26 samples from PET- / MRI +, 73 % were grade 4. Data from histopathological parameters showed that PI is £ 15% in 38 % of 79 malignant samples.
Conclusions FLT PET is able to detect malignant tissue in MRI-negative volumes with a positive predictive value of 43 %. It may therefore be useful to better identify the margin of malignant gliomas for planning radiotherapy and detect tumoral recurrence in association with MRI