Published in:
01-06-2015 | Original Article
Prognostic value of volume-based measurements on 11C-methionine PET in glioma patients
Authors:
Kentaro Kobayashi, Kenji Hirata, Shigeru Yamaguchi, Osamu Manabe, Shunsuke Terasaka, Hiroyuki Kobayashi, Tohru Shiga, Naoya Hattori, Shinya Tanaka, Yuji Kuge, Nagara Tamaki
Published in:
European Journal of Nuclear Medicine and Molecular Imaging
|
Issue 7/2015
Login to get access
Abstract
Purpose
11C-methionine (MET) PET is an established diagnostic tool for glioma. Studies have suggested that MET uptake intensity in the tumor is a useful index for predicting patient outcome. Because MET uptake is known to reflect tumor expansion more accurately than MRI, we aimed to elucidate the association between volume-based tumor measurements and patient prognosis.
Methods
The study population comprised 52 patients with newly diagnosed glioma who underwent PET scanning 20 min after injection of 370 MBq MET. The tumor was contoured using a threshold of 1.3 times the activity of the contralateral normal cortex. Metabolic tumor volume (MTV) was defined as the total volume within the boundary. Total lesion methionine uptake (TLMU) was defined as MTV times the mean standardized uptake value (SUVmean) within the boundary. The tumor-to-normal ratio (TNR), calculated as the maximum standardized uptake value (SUVmax) divided by the contralateral reference value, was also recorded. All patients underwent surgery (biopsy or tumor resection) targeting the tissue with high MET uptake. The Kaplan-Meier method was used to estimate the predictive value of each measurement.
Results
Grade II tumor was diagnosed in 12 patients (3 diffuse astrocytoma, 2 oligodendroglioma, and 7 oligoastrocytoma), grade III in 18 patients (8 anaplastic astrocytoma, 6 anaplastic oligodendroglioma, and 4 anaplastic oligoastrocytoma), and grade IV in 22 patients (all glioblastoma). TNR, MTV and TLMU were 3.1 ± 1.2, 51.6 ± 49.9 ml and 147.7 ± 153.3 ml, respectively. None of the three measurements was able to categorize the glioma patients in terms of survival when all patients were analyzed. However, when only patients with astrocytic tumor (N = 33) were analyzed (i.e., when those with oligodendroglial components were excluded), MTV and TLMU successfully predicted patient outcome with higher values associated with a poorer prognosis (P < 0.05 and P < 0.01, respectively), while the predictive ability of TNR did not reach statistical significance (P = NS).
Conclusion
MTV and TLMU may be useful for predicting outcome in patients with astrocytic tumor.