Various imaging modalities are commonly used for preoperative examinations of brain tumors among which ADC mapping, MRS, and PET provide noninvasive, spatially specific information about tumor cellularity and metabolism. ADC has been shown to be negatively correlated with tumor cell density and is useful in assessing the glioma grade^1-3, whereas MRS is useful in evaluating in vivo biomarkers to predict the tumor grade⁴. 11C-methionine (MET)-, and 18F-fluorodeoxyglucose (FDG)-PET have provided new prognostic information beneficial for identifying metabolically active tumors^5-8.

In this study, we investigated the capacity of DWI, MRS, and PET-CT to predict the WHO glioma grade and identify correlations to the Ki-67 index as a marker of tumor cell proliferation⁹.

We retrospectively reviewed the medical records of 32 glioma patients (13 men and 19 woman; mean age, 55.5 years; age range, 22–85 years) who underwent MRI and MET- and FDG-PET examinations. MRS data were obtained from 29 patients. All patients were pathologically diagnosed via surgical specimens (G-II, 6; G-III, 5; G-IV, 21). The Ki-67 proliferation index was assessed by histochemical staining. All examinations and subsequent surgeries were performed within 2 weeks.

MRI examinations were performed using a 3-T system. DWI was obtained using a single-shot EPI sequence with b values of 0 and 1,000 s/mm² with automatic generation of ADC maps. To obtain the ADC value of the tumor, several round ROIs of 10-20mm² were carefully placed on the ADC map to include the area with the lowest ADC value, as determined with visually, while avoiding cystic, necrotic, or hemorrhagic components of the tumor. The ROI with the lowest ADC was chosen as the minimum ADC (ADCmin).

MRS data was obtained using the PRESS sequence technique (TR = 2000ms, TE = 144ms; volume = 8mL) from the VOI, including the solid components of the tumor and Cho/Cr ratios (Cho/Cr) were recorded.

PET studies were performed in a three-dimensional acquisition mode. MET-PET data were acquired for 20 min after the administration of a MET dose of 5 MBq/kg body weight. For the FDG-PET study, enteral and parental sources of glucose were withheld for at least 6 h before the examination. The PET scan was initiated 90 min after the administration of an FDG dose of 3.5 MBq/kg body weight. Emission data were acquired for 20 min. The highest standard uptake values (SUV) was chosen among the ROIs over the tumor as the maximum SUV (SUVmax) on MET and FDG-PET images. The SUV ratio (MET-SUV and FDG-SUV) was calculated by dividing SUVmax of the tumor by the mean SUV of the contralateral normal cortex.

ADCmin, Cho/Cr, MET-SUV, and FDG-SUV of each glioma grades are presented in Table 1. There were statistically significant deferences in the ADCmin values between each glioma grade and between grade G-II and G-IV by MET- and FDG-SUVs. A box-and-whisker plot of the WHO grade and ADCmin is shown in Figure 1.

The Spearman rank correlation coefficients between the Ki-67 index and these four parameters are presented in Table 2. An inverse correlation was observed between the Ki-67 index and ADCmin (ρ= −0.6898, p < 0.0001, Figure 2), whereas there were positive correlations between the other three parameters.

ADCmin, Cho/Cr, MET-SUV, and FDG-SUV have been well correlated with the histological grade in previous reports^{1-3, 6, 9}. Among these parameters, ADCmin was found to be the best predictor of the histological grade in our patient series. However, any differentiation between MET- and FDG-SUV was only significant between G-II and G-IV. Some authors reported that MET- and FDG-SUV were reliable prognostic factors of glioma^{7, 8}. In this study, we were unable to further investigate the prognoses of our patients because of the limited follow-up duration; therefore, we plan to conduct further investigations with continuous follow-ups.

The antigen Ki-67 index is reportedly the most reliable marker of cellular proliferation because of its expression in almost all phases of the cell cycle, with the exception of the G0 phase⁹. Our results showed that these parameters and the Ki-67 index were significantly correlated with glioma; therefore, ADCmin, Cho/Cr, MET-, and FDG-SUV may be reliable markers of cellular proliferation of in gliomas.

The findings of this limited patient series showed that ADCmin was superior to Cho/Cr, MET0, and FDG-SUV as a preoperative histological predictor of glioma. ADCmin was also significantly negatively correlated with the Ki-67 index and thus, should be considered as a reliable marker of cellular proliferation.