Comparison of MR-permeability imaging from 11C-methionine PET in differentiating radiation necrosis from recurrent metastatic tumors of the brain after gamma knife radiosurgery Noriaki Tomura, M.D.1， Toshiyuki Saginoya,M.D. 2，Yasuhiro Kikuchi, M.D.3 Southern Tohoku Research Institute for Neuroscience Southern Tohoku General Hospital　 Department of Neuroradiology1，Radiology2，Neurosurgery3

☑ No conflict of interest

Introduction Stereotactic radiosurgery such as gamma knife and cyber knife is an effective tool for intracranial neoplasms. However, radiation necrosis is a severe local tissue reaction, which generally occurs 3 – 12 months after therapy. Differentiation between radiation necrosis and recurrent tumors is often difficult with conventional imaging technique, such as MRI, CT, and perfusion SPECT1. Recently, several imaging technique using MR-spectroscopy, MR diffusion-weighted imaging (DWI), MR diffusion tensor imaging (DTI), SPECT with 201Tl, PET with 18F-FDG, have been used to differentiate between them. Compared with those modalities, superiority of PET with11C-methionine (MET)2,3 for differentiating between them has been reported. High sensitivity and specificity using MET-PET has been reported. Dynamic contrast-enhanced MRI (DCE-MRI) with contrast agent has characterized integrity in brain tumors4 and stroke. MR-permeability imaging5,6 using DCE-MRI, based on the Tofts model7, has recently been developed and used for cerebrovascular diseases, brain tumors, and tumors in the prostate. In the present study, MR-permeability imaging was compared with MET-PET in differentiating radiation necrosis from recurrent tumors in patients with metastatic brain tumors after gamma knife radiosurgery.

Material and methods The study was performed for 18 lesions from 15 patients with metastatic brain tumors who underwent gamma knife radiosurgery. Ten lesions were identified as recurrent tumors by surgery after both MR-permeability imaging and MET-PET. Eight lesions were diagnosed as radiation necrosis because of a lack of change or a decrease in size by >4 months after radiosurgery. MET-PET was performed immediately before FDG-PET on the same day. The protocol was indicated in Fig. 18. After CT, MET was injected and MET-PET was performed 20 min later. FDG was injected 60 min after MET-PET. MR-permeability imaging and DWI were performed within 1 week before or after PET. DCE-MRI was acquired using gadolinium contrast medium. A 3-dimensional fast spoiled gradient echo sequence (SPGR) was applied for DCE-MRI using a bolus injection of contrast material (total dose: 0.2 mL / kg of body weight, dose rate: 3.0ml / sec). Parameters of DCE-MRI are the followings; TR/TE = minimum (5.7 msec) / minimum (1.3 msec), FA = 20°, FOV = 24 cm, matrix = 256×160, NEX = 1, number of slices = 16 / phase, number of phase = 32, acquisition time 3’59”. DCE data were transferred to a workstation (Advantage Workstation ver. 4.6, General Electric) and analyzed using commercially available software (GenIQ, General Electric) with the general kinetic model based on a two-compartment model and three parameters (vascular space, extracellular extravascular space, and fractional plasma volume).

Material and methods The transfer constant between intra- and extravascular and extracellular spaces (Ktrans) (/min.), the extravascular extracellular space (Ve), the transfer constant from the extracellular extravascular space to plasma (Kep) (/min.), initial area under the signal intensity-time curve (IAUGC), contrast enhancement ratio (CER), bolus arrival time (BAT) (sec), maximum slope of increase (Max. slope) (mMol/sec), and fractional plasma volume (fPV) were calculated after setting a region of interest on the solid portion of the lesion. The apparent diffusion coefficient (ADC) (10-3 mm/s) was also acquired from DWI. On both MET-PET and FDG-PET, the ratio of the maximum standardized uptake value (SUVmax) of the lesion divided by the SUVmax of the symmetrical site in the contralateral cerebral hemisphere was measured (MET-ratio and FDG-ratio, respectively). For measurement of each data, region of interests (ROIs) were manually set on the fused images with SPGR images using Advantage Workstation. Receiver operating characteristic (ROC) analysis was performed to evaluate the utility of those parameters for differentiating radiation necrosis from recurrent tumors.

Protocol for both PET-CTs MET injection FDG injection PET scan CT scan PET scan 10' for head 10' for head 20' 60' This is a protocol for both PET-CTs. FDG-PET scan was performed subsequently after methionine-PET. FDG injection was performed 60 min. after MET injection. 60' Figure 1

Results The minimal, average, and maximum values of each MRI parameter were obtained. After the minimal, average, and maximum values were analyzed by ROC, the average of Ktrans, Ve, Kep, IAUGC, CER, BAT, Max. slope, and fPV was more excellent than the minimum and maximum values of them. In ADC, the minimum value (ADCmin) was more excellent than the average and maximum values. Fig. 2 shows ROC curve of each parameter. Area under the ROC curve (AUC) for differentiating radiation necrosis from recurrent tumors was the most excellent for MET-ratio (0.90) followed by CER (0.81), Max slope (0.80), IAUGC (0.78), fPV (0.76), BAT (0.76), Ktrans (0.74), Ve (0.68), ADCmin (0.60), Kep (0.55), FDG ratio (0.53) (Table 1).　In MET ratio (p<0.01), CER (p<0.01), Max. slope (p<0.05), IAUGC (p<0.05), the AUC value was significantly more excellent (Chi square test) than 0.5 of AUC. The cutoff value for the best combination of sensitivity and specificity was 1.42 with MET ratio, 0.61 with CER, 0.01 with Max. slope, 0.2 with IAUGC, 0.02 with fPV, 44.0 with BAT, 0.05 with Ktrans, 0.27 with Ve, 0.73 with ADCmin, 0.32 with Kep, and 0.97 with FDG ratio (Table 1).

Result Using the cutoff value, the sensitivity and specificity were 0.9 and 0.75 in MET ratio., 0.8 and 0.88 in CER, 0.9 and 0.5 in Max. slope, 0.6 and 1.0 in IAUGC, 0.5 and 0.88 in fPV, 0.3 and 0.7 and 0.75 in BAT, 0.7 and 0.75 in Ktrans, 0.6 and 0.75 in Ve, 0.6 and 0.25 in ADCmin, 0.8 and 0.5 in Kep, 0.4 and 0.5 in FDG ratio (Table 1). Significant difference in MET ratio (p<0.01), CER (p,0.01), Max. slope (p<0.05) and IAUGC (p<0.05) was evident between radiation necrosis and recurrent tumor (Fig. 3).

Figure 2 av, average; min, minimum

(χ2 test compared with AUC 0.5) Table 1   AUC (χ2 test compared with AUC 0.5) 95% CI Cutoff value Sensitivity (95% CI) Specificity MET ratio 0.90 (p<0.01) 0.75-1.05 1.42 0.90 (±0.026) 0.75 (±0.082) CER av 0.81 (p<0.01) 0.58-1.04 0.61 0.80 (±0.035) 0.88 (±0.023) Max. slope av 0.80 (p<0.05) 0.58-1.02 0.01 (±0.014) 0.50 (±0.098) IAUGC av 0.78 (p<0.05) 0.55-1.00 0.2 0.60 (±0.078) 1.0 (0) fPV av 0.76 0.53-0.99 0.02 (±0.017) BAT av 0.48-1.03 44.0 0.70 (±0.046) (±0.041) Ktrans av 0.74 0.49-0.99 0.05 0.67 Ve av 0.68 0.41-0.95 0.27 (±0.052) 0.63 (±0.051) ADC min 0.32-0.88 0.73 (±0.039) 0.25 (±0.061) Kep av 0.55 0.26-0.84 0.32 FDG ratio 0.53 0.23-0.82 0.97 0.40 (±0.059) (±0.049) av, average; min, minimum

Figure 2 MET ratio CER p<0.01 p<0.01 MET- Ratio av CER av Max.Slope av IAUGC av av, average; min, minimum

Representative case 1 78 y.o. male, Brain metastasis from lung ca. Radiation necrosis s/o (arrows) A contrast enhanced lesion in the isthmus of the left cingulate gyrus slightly increases in size 12 months after gamma knife radiosurgery (GKS). MR-CER and IAUGC shows low values in the lesion, and MET-PET shows decreased activity in the lesion. before GKS T1WI C(+) 12 months after GKS T1WI C(+) T2WI CER IAUGC 14 months after GKS MET-PET 14 months after GKS

Representative case 2 43 y.o. female, Brain metastasis from breast ca. Recurrent tumor (arrows) A contrast enhanced lesion in the right frontal lobe slightly increases in size 59 months after GKS. Its surrounding edema also increases. MR-CER and IAUGC shows high values in the lesion, and MET-PET shows increased activity in the lesion. T1WI C(+) T2WI 54 months after GKS 56 months after GKS T1WI C(+) T2WI 59 months after GKS T1WI C(+) T2WI CER IAUGC 59 months after GKS MET-PET

Representative case 3 75 y.o. male, Brain metastasis from lung ca. Radiation necrosis s/o (arrows) A contrast enhanced lesion with surrounding edema is seen in the left frontal lobe 11 months after GKS. MR-CER and IAUGC shows low values in the lesion, and MET-PET shows decreased activity in the lesion. 11 months after GKS FLAIR T1WI C(+) CER IAUGC 11 months after GKS MET-PET 11 months after GKS

Discussion Enlargement of a contrast enhanced lesion on MRI following radiosurgery may be due to radiation necrosis and recurrent tumor. Differential diagnosis between them is extremely important for indication of additional therapy, but it is often difficult by conventional modalities such as CT, MRI, and SPECT1. The present study elucidated that MET-PET was the most promising to differentiate radiation necrosis from recurrent metastatic tumors after gamma knife radiosurgery in comparison of MR-permeability imaging, MRI-DWI, and FDG-PET. In the present study, both MET-PET and FDG-PET were undertaken on a single day. This technique has already been reported. The cross-talk between two tracers was considered to be minimal. MET presumably accumulates to neoplasms with amino acid transporter. In tumors, MET can preferably accumulate due to high density and activity of amino acid transporter in tumors. In recurrent tumors, MET can be accumulated by active transport through cell proliferation2,3. On the other hand, in radiation necrosis, it could be presumably due to passive diffusion through BBB damage4. The different mechanism of MET accumulation in two pathological processes could be the means of distinguishing recurrent tumors from radiation necrosis.

Discussion MR-permeability imaging in the present study used DCE-MRI. Dynamic susceptibility contrast-enhanced MRI (DSC-MRI)9 has also been used for MR perfusion. But, DSC-MRI has a limitation of susceptibility artifacts due to hemorrhage, calcification, and surgical clips. MR-permeability imaging in the present study yields many parameters, but interpretation of results of those parameters remains difficult. Tissue enhancement following contrast agent generally depends on various factors such as vessel density, vascular permeability, blood flow, and interstitial pressure. Although qualitative visual evaluation of the images is possible, quantitative data could improve the result of analysis. MR permeability package used in the present study was an easy tool for imaging and quantification the data. In the present study, CER, a relatively simple data, was the most excellent for differentiating between radiation necrosis and recurrent tumors. IAUGC, which is nearly equal to blood volume, followed CER. Ktrans, the transfer constant between intra- and extravascular and extracellular spaces, has been reported as a feasible parameter in evaluating grading of gliomas and in detecting tumors in the prostate.

Discussion The present study showed that CER and IAUGC were superior to Ktrans for differentiating between radiation necrosis and recurrent tumors. Parameters in MR-permeability imaging such as CER, IAUGC, and Ktrans showed higher values in recurrent tumors than in radiation necrosis. In recurrent tumors, newly proliferative tumor vessels in the tissue with BBB damage could presumably play a role in increasing those parameters. Compared with recurrent tumors, permeability could be mainly due to passive diffusion through BBB damage in radiation necrosis. Feasibility of FDG-PET in evaluating as well as in detecting neoplasms in various organs has well known and it has been the most available tracer. Usefulness of FDG-PET in diagnosis of radiation necrosis in the brain has also been reported in the literature, however, FDG-PET was presently inferior to every parameter of MR-permeability imaging.

Conclusion MET-PET is superior to MR-permeability imaging, ADC, and FDG-PET in differentiating radiation necrosis from recurrent tumors after gamma knife radiosurgery for metastatic brain tumors. In MR-permeability imaging, CER, Max. slope, and IAUGC are superior to other parameters of MR-permeability imaging.

References 1. Shah R, et al. Radiation necrosis in the brain: Imaging features and differentiation from tumor recurrence. Radiographics 2012;32:1343-1359 2. . Takenaka S, et al. Comparison of 11C-methionine, 11C-choline, and 18F-fluorodeoxyglucose- positron emission tomography for distinguishing glioma recurrence from radiation necrosis. Neurol Med Chir (Tokyo) 2014;54:280-289 3. Terakawa Y, et al. diagnostic accuracy of 11C-methionie PET for differentiation of recurrent brain tumors from radiation necrosis after radiotherapy. J Nucl Med 2008;49:694-698 4. Narang J, et al. Differentiating treatment-induced necrosis from recurrent / progressive brain tumor using nonmodel-based semiquantitative indices derived from dynamic contrast-enhanced T1- weighted MR perfusion. Neuro-Oncol 2011;13:1037-1046 5. Vidarsson L, et al. Quantitative permeability magnetic resonance imaging in acute ischemic stroke: how long do we need to scan ? Magnetic resonance imaging 2009;27;1216-1222 6. Verma S, et al. Overview of dynamic contrast-enhanced MRI in prostate cancer diagnosis and management. AJR 2012;198:1277-1288 7. Tofts PS, et al. Estimating kinetic parameters from dynamic contrast-enhanced T1-weighted MRI of a diffusible tracer standardized quantities and symbols. JMRI 1999; 10:223-232 8. Tomura N, et al. PET findings of intramedullary tumors of the spinal cord using [18F]FDG and [11C]methionine. AJNR 2013;34:1278-1283 9. Fatterpekar GM, et al. Treatment-related change versus tumor recurrence in high-grade gliomas: A diagnostic conundrum. Use of dynamic susceptibility contrast-enhanced (DSC) perfusion MRI. AJR 2012;198;19-26