MRS And DWI In Assessment Of Early Therapeutic Effect Of Gamma Knife Radiosurgery For Malignant Brain Tumors

Objective:Conventional MRI of T1WI, T2WI and Gd-T1WI have been applied as standard imaging modalities to access brain cancer treatment response according to the morphological changes of tumor volume. However, morphological changes are not sensitive enough to access treatment response, e.g., the apoptosis and necrosis of tumor cells may happen before the morphological changes. Diffusion weighted imaging(DWI) and MR spectroscopy (MRS) provides non invasive metabolic information to characterize the changes of tumor microstructure as well as tumor cells apoptosis or necrosis after treatment. The purpose of this study is to evaluate the gamma knife radiosurgery (GKS) treatment response for malignant gliomas and brain metastases using MRS and to identify the specific metabolic imaging biomarkers.Methods:Twenty-two patients with brain metastases (mean age=62.2years;12females;10males) and six patients with brain gliomas(mean age=28.8years; all males) were recruited in this study. Standard MRI, DWI and MRS were performed using1.5T MR scanner (GE signa) in all patients on one day before and one month later after GKS treatment. For quantitative analysis of DWI and MRS data, the ROIs were placed on the regions with Gd-enhancing on T1WI and contralateral normal brain tissue. The tumor volume was identified on the Gd-T1WI and calculated according to the Ellipsoid cubature formula (V=π/6×length×width×height). The levels of the NAA, Cr, Cho, NAA/Cr, Cho/Cr and Cho/NAA were calculated using standard software (Functool GE workstation ADW4.2). In addition, the ratios of the tumor metabolite level to the corresponding metabolite level of the normal brain (nNAA, nCr, nCho) and the Cho in the tumor to contralateral NAA (cNAA) were calculated. As for DWI, The ADCs of tumors and contralateral white matter were calculated using standard software (Functool GE workstation ADW4.2). And also, the ratios of the tumor ADCs to the corresponding normal brain (rADC) were calculated.All th data were statistically analyzed by student T test. Correlation between changes of tumor volume and DWI as well as MRS was analyzed by Pearson’s correlation test.Results:There is a significant difference of volume changes in brain metastases group before and after GKS (p=0.003). And there is no significant difference of in brain gliomas group. In brain metastases group (n=22), the changes of DWI and MRS parameters after GKS treatment were as follows:There were significantly decreased ratio of nCho, Cho/Cr and Cho/cNAA (both p<0.05) and Cho/NAA (p<0.01) on one month post GKS compared to baseline. However, ratio of nCr, ADC and rADC was significantly increased (both p<0.01). In glioma group (n=6), although the variation trend was similar to the brain metastases group, however, statistical significance was only found in nCho (P<0.05). Pearson’s correlation test showed no significant correlation among changes of tumor volume, DWI and MRS parameters in both groups..Conclusions:DWI and MRS revealed the significant metabolic changes in brain metastases and gliomas on one month post GKS treatment, which is earlier than conventional MRI determined volumetric changes. Quantitative analysis of DWI and MRS parameters may serve to earlier assessment of brain cancer treatment response.