Diffusion kurtosis imaging study on temporal lobe after nasopharyngeal carcinoma radiotherapy

Brain Research 1648 (2016) 387–393

Contents lists available at ScienceDirect

Brain Research journal homepage: www.elsevier.com/locate/brainres

Research report

Diffusion kurtosis imaging study on temporal lobe after nasopharyngeal carcinoma radiotherapy Dan Wang a, Yue-Hua Li a,n, Jie Fu b, He Wang c a Institute of Diagnostic and Interventional Radiology, The Sixth Affiliated People's Hospital, Shanghai Jiao Tong University, No. 600, Yi Shan Road, Shanghai 200233, China b Department of Radiotherapy, The Sixth Affiliated People's Hospital, Shanghai Jiao Tong University, No. 600, Yi Shan Road, Shanghai 200233, China c Philips Research China, Philips Innovation Campus Shanghai, China

art ic l e i nf o

a b s t r a c t

Article history: Received 21 April 2016 Received in revised form 23 July 2016 Accepted 26 July 2016 Available online 8 August 2016

Purpose: Diffusion kurtosis imaging (DKI) is a MRI technique which can measure alterations in the diffusion of water molecules to reflect tissue changes in both white and grey matter. This study evaluated the potential of DKI for the early diagnosis of radiation-induced temporal lobe changes in the grey and white matter of the temporal lobe in patients with nasopharyngeal carcinoma (NPC). Materials and methods: Sixty patients with NPC who had normal MRI brain scans were enrolled and underwent DKI at 1 week (n¼ 20), 6 months (n¼20) or 1 year (n¼20) after radiotherapy; 20 normal control individuals were also evaluated. Nonlinear fitting routines and equations were used to calculate mean diffusion (MD) and mean kurtosis (MK) and fractional anisotropy (FA). Analysis of variance was used to compare the MK/MD/FA values of white and grey matter between groups. Results: Compared to the normal control group, grey and white matter MK values were significantly higher at 1 week after radiotherapy and significantly lower at 6 months and 1 year after radiotherapy in patients with NPC, whereas the grey and white matter MD values were significantly lower at 1 week after radiotherapy and returned to normal by 6 months and 1 year after radiotherapy. Conclusion: DKI can be used to detect radiotherapy-induced changes in both the white and grey matter of temporal lobe in patients with NPC. MK and MD values may represent reliable indicators for the early diagnosis of radiation-induced temporal lobe changes in NPC. & 2016 Elsevier B.V. All rights reserved.

Keywords: Diffusion kurtosis imaging Nasopharyngeal carcinoma Radiotherapy changes Magnetic resonance imaging

1. Introduction The areas with the highest incidence of nasopharyngeal carcinoma (NPC) are in southern China and Southeast Asia (Sarmiento and Mejia, 2014; Zhang et al., 2015). Radiotherapy is the most effective treatment for NPC (Qiu et al., 2016). However, radiationinduced temporal lobe change is a major complication observed after radiotherapy that may further develop into encephalopathy and induce neurological symptoms such as reduced neurological function and cognitive change (Chong et al., 1999; Su et al., 2011). The precise pathogenesis of radiation-induced temporal lobe change is still under exploration, and it is complicated by the fact that clinical diagnosis is limited to pathological examinations (Chong et al., 1999; Zhou et al., 2014). Animal experiments have shown partial demyelination, vascular endothelial karyopyknosis, perivascular edema, swelling of glial cells and local nerve cell degeneration occur after whole brain radiation (Tang et al., 2006). n

Corresponding author. E-mail address: [email protected] (Y.-H. Li).

http://dx.doi.org/10.1016/j.brainres.2016.07.041 0006-8993/& 2016 Elsevier B.V. All rights reserved.

The side effects of radiation-induced change after radiotherapy can be divided into three stages: (i) the acute stage from 24 h to 1 week after radiotherapy, when vascular endothelial cells swell, vessel wall permeability increases and acute intracranial hypertension is a common symptom. In this stage, cell and cell organelle swelling occurs, resulting in increased cell structure complexity; (ii) early delayed effects, typically occurring a few weeks to a few months after radiotherapy, usually observed as inflammation. In this stage, neuronal cell apoptosis, necrosis and nuclear disintegration can lead to decreased cell complexity; and (iii) late delayed effects occurring a few months to more than a decade after radiotherapy, such as neuronal degeneration, necrosis and gliosis, which usually manifest as irreversible neurological dysfunction. In this stage, glial cell proliferation leads to increased cell complexity, however, cell complexity does not recover to the levels observed before radiotherapy (Cheng et al., 2000; Chong et al., 1999; Lee et al., 1988; Su et al., 2011). Currently, the prevention and treatment of radiation-induced temporal lobe change rely on early diagnosis and monitoring, with early diagnosis mainly based on radiological techniques (Lv et al., 2014; Sun et al., 2013). Enhanced MRI and magnetic resonance

388

D. Wang et al. / Brain Research 1648 (2016) 387–393

spectroscopy (MRS) techniques are both used for the diagnosis of radiation-induced temporal lobe change in patients with NPC (Wang et al., 2012). MRS can reveal metabolic changes due to radiation-induced temporal lobe change (Chong et al., 1999; Wang et al., 2012). However, the detection of abnormalities on enhanced MRI usually reflects the occurrence of irreversible radiation-induced temporal lobe change, and identification of grey matter changes due to radiation-induced temporal lobe change is currently limited to morphological measurements based on MRI. The theory underlying DWI and diffusion tensor imaging (DTI) is based on the Gaussian diffusion behavior of water molecules (Wu and Cheung, 2010). Microstructural components in the brain, such as cell membranes, intracellular spaces and extracellular spaces, make the water distribution behave in a non-Gaussian manner. To characterize such non-Gaussian diffusion behavior, DKI, a fourth-order three-dimensional fully symmetric tensor MR imaging technique, is an extension of conventional DWI that requires the use of multiple b-values commonly up to 2000 or 3000 s/mm2 (Minati et al., 2007; Wu and Cheung, 2010). Nervous system diseases can alter tissue microstructures that affect nonGaussian water distribution. DKI can quantify such changes in non-Gaussian water distribution to evaluate pathophysiological changes, and has the potential to be widely applied for the diagnosis of central nervous system diseases (Giannelli et al., 2012; Grinberg et al., 2011). Application of diffusion kurtosis imaging (DKI) to assess microscopic changes in both the grey and white matter of the temporal lobe may provide useful information for the diagnosis of radiation-induced temporal lobe change. Currently DTI is widely used to indicate white matter changes, such as ischemia and hypoxia (Gao et al., 2012), as well as developmental changes in white matter (Hui et al., 2010). However, Cheung et al. (2009) reported that DKI can reflect changes in both grey matter and white matter. Previous studies on radiotherapyinduced changes in brain tissues by Wang et al. (2009) and Chan et al. (2009) using DTI imaging indicate that changes in white matter after radiotherapy may be a major concern. Chiang et al. used diffusion basis spectrum imaging (DBSI) to study axon/myelin integrity and white matter crossing. Axons and myelin are the major components of white matter (Chiang et al., 2014; Wang et al., 2014, 2011); however, grey matter has a high neuronal cell body cell (soma) distribution and is seldom studied. In this research, the potential of DKI as an analysis method for the early diagnosis of radiation-induced temporal lobe white matter and grey matter change was explored in patients with NPC. This study focused on the early-stage changes in grey matter and white matter during radiation-induced temporal lobe change (Table 1).

Table 1 Characteristics of the normal control (NC) group and groups of patients with NPC analyzed at one week, six months or one year after radiotherapy.

Number Male:female ratio Age (years) Median age (years) P-value*

NC

1W

6M

1Y

20 11:9 49.45 7 9.23 52.5

20 10:10 49.47 7.88 48 0.98327

20 11:9 52.9 7 6.14 50 0.81759

20 11:9 50.6 7 3.82 53 0.15066

Number 1 2 6 7 2 0 2 0

Number 2 0 7 7 0 3 1 0

Number 1 2 8 5 1 2 0 1

TNM stage T1N1M0 T1N2M0 T2N1M0 T2N2M0 T3N1M0 T3N2M0 T4N1M0 T4N2M0 *

P-values are for comparison of age vs. normal control group.

Table 2 White matter and grey matter MK values for the normal control (NC) group and the irradiated temporal lobes of the groups of patients with NPC analyzed at one week, six months or one year after radiotherapy.

NC (n¼20) 1W (n¼ 20) 6M (n ¼20) 1Y (n¼ 20)

White matter

Grey matter

Mean

Standard deviation

Mean

Standard deviation

0.90889 1.08771 0.59218 0.52089

0.18475 0.26214 0.15173 0.14729

0.81568 1.05963 0.65543 0.55075

0.13949 0.31532 0.22763 0.1867

Comparison between groups NC vs. 1W NC vs.6M NC vs. 1Y 1W vs. 6M 1W vs. 1Y 6M vs. 1Y *

White matter P-value o 0.001* o 0.001* o 0.001* o 0.001* o 0.001* 024421

Grey matte P-value 0.00107* 0.02832* o 0.001* o 0.001* o 0.001* 0.14833

Significant difference between groups.

Table 3 White matter and grey matter MD values for the normal control (NC) group and the irradiated temporal lobes of the groups of patients with NPC analyzed at one week, six months or one year after radiotherapy.

2. Results Compared to the NC group, the grey and white matter MK values of patients with NPC were significantly higher in the 1W group (P¼ 0.00107 and P o0.001) and significantly lower in the 6M group (P¼ 0.02832 and P o0.001) and 1Y group (all P o0.001). The grey and white matter MK values of the 6M and 1Y groups were significantly lower than those of the 1W group (all P o0.05; Table 2; Figs. 2–4). Compared to the NC group, the grey and white matter MD values of the 1W group of patients with NPC were significantly lower (P ¼0.04804 and P ¼0.04749). However, the grey and white matter MD values of the 6M group were significantly higher than the 1W group (P ¼0.16238 and P ¼0.03779), and the grey and white matter MD values of the 6M and 1Y groups were not significantly different to the NC group (P ¼0.260236 and P¼ 0.62705; Table 3; Figs. 2–4). Unlike the MK and MD values, clear trends were not observed

NC (n¼20) 1W (n¼ 20) 6M (n ¼20) 1Y (n¼ 20)

Comparison between groups NC vs. 1W NC vs. 6M NC vs. 1Y 1W vs. 6M 1W vs. 1Y 6M vs. 1Y *

White matter

Grey matter

Mean

Standard deviation

Mean

Standard deviation

0.90889 1.08771 0.59218 0.52089

0.18475 0.26214 0.15173 0.14729

0.81568 1.05963 0.65543 0.55075

0.13949 0.31532 0.22763 0.1867

White matter P-value 0.04749* 0.92101 0.55871 0.03779* 0.01114* 0.62705

Significant difference between groups.

Grey matte P-value 0.04804* 0.5512 0.08716 0.16238 0.7829 0.26023

D. Wang et al. / Brain Research 1648 (2016) 387–393

Table 4 White matter and grey matter FA values for the normal control (NC) group and the irradiated temporal lobes of the groups of patients with NPC analyzed at one week, six months or one year after radiotherapy. White matter

Grey matter

Mean

Standard deviation

Mean

Standard deviation

NC (n¼ 20) 1W (n ¼20) 6M (n¼ 20) 1Y (n ¼20)

0.40944 0.41232 0.41208 0.50599

0.13852 0.09651 0.14619 0.17364

0.16716 0.23455 0.18764 0.24084

0.05174 0.09822 0.0399 0.07011

NC vs. 1W NC vs. 6M NC vs. 1Y 1W vs. 6M 1W vs. 1Y 6M vs. 1Y

Comparison between groups t-value P-value 0.06433 0.94887 0.05902 0.95309 2.15842 0.03405*  0.00531 0.99578 2.09409 0.03959* 2.0994 0.0391*

t-value 3.10552 0.94389 3.39569  2.16163 0.29018 2.4518

P-value 0.00267* 0.34822 0.00109* 0.0338* 0.77247 0.01651*

*

Significant difference between groups.

for the FA values. However, there was a significant difference between the white matter FA values of the 1Y and 6M groups (P ¼0.0391). There were also significant differences between the grey matter FA values of the 1W and NC groups (P ¼0.00267), 6M and 1W groups (P ¼0.0338), and 1Y and 6M groups (P ¼0.01651) (Table 4; Figs. 2–4).

3. Discussion The changes in the MK values of the irradiated temporal lobe (primarily white and grey matter) at 1 week after radiotherapy observed in this study can be interpreted as the result of shortterm changes in the temporal lobe after irradiation (DeSalvo, 2012). MK, the average apparent kurtosis along all diffusion-encoding directions, has been shown to offer an improved sensitivity for detecting developmental and pathological changes in neural tissues (Giannelli et al., 2012). MK differs from indices of fractional anisotropy measured by DTI, DTI can be used to observe changes in white matter fibers, while DKI can be used as an indicator of organizational complexity (in both grey and white matter) (Hui et al., 2008a,b; Jensen et al., 2005; Lu et al., 2006). MK values depend on the structural complexity of the region of interest. As structures become more complex, non-Gaussian water molecule diffusion becomes more significant, and the MK values will increase. At 1 week after radiotherapy, swelling of mitochondria and

389

organelles can be detected, which – to an extent – will increase the movement of free intracellular water molecules in different directions, and this may lead to an increase in the MK value. MD values related to water molecule diffusivity (Tan et al., 2004; Wang et al., 2012). Irradiation of the temporal lobe induces transient mild edema, which restricts water molecule movement and led to a reduction in the MD values of both grey matter and white matter. These effects explain the significant changes in the MK and MD values observed at 1 week after radiotherapy compared to the normal control group. The FA values underwent small changes at 1 week, which can be interpreted to be caused by white matter edema, resulting in a decreased tissue uniformity; these effects have previously been demonstrated in animal experiments and preclinical studies (Kershaw et al., 2013). By 6 months after radiotherapy, irradiated temporal lobe tissues undergo a period of recovery, during which gliosis occurs to compensate for neuronal cell apoptosis (Atwood et al., 2007; Obenaus et al., 2008). With respect to grey matter, Obenaus et al. and Atwood et al. showed that brain metabolism altered after radiotherapy in animal studies using MRS (Atwood et al., 2007; Obenaus et al., 2008). This study shows that the MK values were significantly lower at 6 months after radiotherapy than 1 week after radiotherapy or the values of the NC group. We believe that the pathological basis of the decrease in the MK values during the recovery period of irradiated temporal lobe tissues (at 6 months and 1 year after radiotherapy) is due to a reduced number of neurons and proliferation of glial cells. Neurons have a much higher complexity than glial cells; therefore, increased numbers of glial cells would decrease the MK value. Grinberg et al. came to similar conclusions when studying Alzheimer disease in human, in which neuronal degeneration, atrophy and apoptosis led to a reduction in the complexity of temporal lobe tissue and the MK values decreased as a result of this process (Grinberg et al., 2011). In line with previous studies, Grinberg et al. found that the MK values of the frontal cortex increased between adolescence to adulthood; this change corresponded to a comprehensive improvement in the organizational structure and complexity of the brain associated with the process of development (Grinberg et al., 2011). Additionally, the MK values of the brain reduced with age due to neuronal degeneration and contraction, which may due to the complexity of brain tissues reduced with age (Grinberg et al., 2011). Between 6 months and 1 year after radiation, the MK and MD values of the patients with NPC underwent small changes but remained relatively similar overall. In this period, cerebral tissue is basically stable, as the neuronal apoptosis and proliferation of glial cells observed during the initial recovery reduce to low levels (Wang et al., 2012).

Fig. 1. A 43 year-old female nasopharyngeal carcinoma patient (1) MK image (The ROIs of the temporal lobe are showed, ROI1-grey matter, ROI2-white matter); (2) MD image; (3) FA image; (4) color FA image.

390

D. Wang et al. / Brain Research 1648 (2016) 387–393

Fig. 2. The comparison of MK/MD/FA values of the temporal lobe between normal control and one week/six months/one year after radiotherapy.

The DKI technique was chosen for this study over DTI technology for a number of reasons. DTI assesses the anisotropic diffusion of water molecules in at least 6 directions, and DKI requires both multiple b-values and multiple directions to measure nonGaussian water molecule movement. At least 3 b-values and 15 directions are required for DKI (Jensen and Helpern, 2010b). In this study, we used 30 directions. The fourth-order diffusional kurtosis tensor and second-order diffusion tensor can both be estimated by DKI. DTI can measure the second-order tensor, thus DTI cannot overcome the multi-nerve fibers crossing-cutting issue. Therefore, DKI is a more useful technique for evaluating grey matter and white matter changes than DTI. In previous studies, Jensen et al. used three b-values collected in 15 directions (Jensen and Helpern, 2010b) and Fieremans et al. used three b-values collected in 30 directions (Laufs et al., 2014).

The DKI sequence of the present study used six b-values (0–2500, 500) collected in 30 directions to enhance the reliability of the data. As the number of b-values increases, especially at larger bvalues, the diffusion of water molecules better reflects the state of non-Gaussian movement. However, noise increased when the bvalue increased, and the noise may becomes too high when the bvalue increases beyond the optimal limit to get the qualified image; hence, the maximum b-value sequence used in this study was 2500. DKI, as a means of detecting the microscopic structural changes in brain tissue after radiotherapy, can be helpful when devising and adjusting the dose of clinical radiotherapy treatment plans. However, the use of DKI to predict neuroradiation change at an early stage needs further improvement and long-term follow-up research.

D. Wang et al. / Brain Research 1648 (2016) 387–393

391

Fig. 3. The comparison of MK/MD/FA values of the temporal lobe between normal control and one week/six months/one year after radiotherapy.

This study has a number of limitations. Firstly, we were unable to obtain pathological specimens to correlate the changes observed using DKI with microscopic changes in the brain tissues; therefore, further animal studies are needed to confirm the results of this study. Secondly, a longer-term follow-up study is required to more fully characterize the trends that occur after radiation change in patients with NPC. In particular, a prospective radiological study is needed.

of the temporal lobe in patients with NPC before detectable manifestation in routine MRI imaging, indicating that DKI may provide a useful imaging evaluation tool for clinical research.

4. Conclusion

The study enrolled 20 normal control and 60 patients with pathologically-confirmed NPC who were aged 34–62 years-old (28 males and 32 females) who all had normal MRI brain scans. All patients had stage T1N0M0 to T4N2M0 NPC according to the 7th

This study demonstrates that DKI may be used to predict the alterations of radiation change in both the grey and white matter

5. Methods and materials 5.1. Patients

392

D. Wang et al. / Brain Research 1648 (2016) 387–393

who were healthy and had normal MRI brain scans (11 males and 9 females; Table 1). The inclusion criteria for the patients with NPC were: (1) patients without metastatic lesions in the brain, liver or other organs; (2) whose conventional plain and enhanced MRI brain scans after radiation therapy were normal; (3) and who completed the entire course of radiation therapy. Exclusion criteria were (1) a history of intracranial vascular diseases, such as cerebral infarction or cerebral hemorrhages; (2) primary intracranial tumors; or (3) systemic metabolic diseases or neurodegenerative diseases, such as diabetes or Alzheimer’s disease. The NC group had no symptoms or history of intracranial lesions and normal head MRI examination signals. Written informed consent for the use of patient data was obtained prior to MRI scanning, or directly from the healthy volunteers. This study was reviewed and approved by the Ethics Review Board of the Shanghai 6th People's Hospital Affiliated to Shanghai Jiao Tong University. 5.2. MR examination and measurement A 3-T MR scanner (MAGETOM; Verio, Siemens Healthcare) was used with a 32-channel head coil. The T1-weighted, T2weighted, enhanced T1-weighted (gadolinium-enhanced T1weighted MRI) imaging and DKI sequences were taken. The enhanced T1-weighted sequence was taken after the DKI sequence. The T2W imaging settings were TR/TE, 6000/95 ms; matrix, 384  384; slice thickness, 6 mm. The T1 3D axial imaging settings: TR/TE, 1500/2.96 ms; voxel size, 0.9  0.9  1 mm3; slice thickness, 1 mm. DKI of the brain was performed in all subjects using 30 gradient directions and six b-values (b¼ 0, 500, 1000, 1500, 2000 and 2500 s/mm2). The other imaging parameters were: FOV, 300 mm; TR/TE, 3900/109 ms; voxel size, 1.8  1.8  2.0 mm3; matrix size 128  128; slice thickness, 2 mm; number of slices, 25; and scan duration, 11 min 40 s. Parametric maps for MK and MD (Fig. 1) were generated using Matlab code (written in-house) from all raw diffusion images (6 b-values, 30 directions), and measured using MRIcron software (http://www. nitrc.org/projects/mricron). Two neuroradiologists (with five and eight years of experience) collaborated to manually outline the temporal lobe white matter and grey matter ROIs on the DKI images (Fig. 1). The ROIs were manually delineated twice and the mean value was calculated.

6. DKI theory Fig. 4. The comparison of MK/MD/FA values of the temporal lobe between normal control and one week/six months/one year after radiotherapy.

edition of AJCC staging system (Lee et al., 2012) and attended hospital for radiotherapy; no patients had any prior history of radiotherapy and all patients were treated with intensity-modulated radiotherapy (IMRT). The total nasopharyngeal dose was 66– 73 Gy in 30–33 fractions over 41–55 days. Routine MRI was performed prior to radiotherapy to confirm the absence of intracranial tumor invasion, vascular disease, significant age-related brain changes or other intracranial tumors. The time after completion of radiotherapy and DKI ranged from 1 week to 12 months. The 60 patients, all of whom had normal MRI brain scans, were classified into three groups based on the length of time after radiotherapy: patients analyzed 1 week (1W; n ¼ 20) after the end of radiotherapy, when early acute temporal lobe edema is likely to occur, and two groups in which late-onset radiotherapy-induced temporal lobe change was analyzed at 6 months (6M; n ¼20) or 1 year (1Y; n ¼20) after the end of radiotherapy. The normal control group (NC group) consisted 20 normal control individuals

Diffusional kurtosis is related to diffusion-weighted signal intensity according to the following equation:

ln⎡⎣ S( b)⎤⎦ = ln⎡⎣ S( 0)⎤⎦ − bD + b2D2K /6 + O b3

( )

(1)

where S(b) is the signal intensity as a function of the b value; the b value is the diffusion s weighting factor. D is the diffusion coefficient and K is the diffusional kurtosis. The parameters D and K can be obtained by fitting Eq. (1) with the help of the LevenbergMarquardt algorithm (Jensen and Helpern, 2010a). Then MD and MK are computed as the average diffusion coefficient and kurtosis along all uniformly distributed diffusion directions.

Conflict of interest The authors confirm they have no conflicts of interest. This study was supported by the National Natural Science Youth Project Foundation of China, No. 81501460, Shanghai Jiaotong University “crossing Biomedical Engineering Research Fund” No.

D. Wang et al. / Brain Research 1648 (2016) 387–393

YG2015MS16, and the National Natural Science Project Foundation of China, No. 1471656.

References Atwood, T., Payne, V.S., Zhao, W., Brown, W.R., Wheeler, K.T., Zhu, J.M., Robbins, M. E., 2007. Quantitative magnetic resonance spectroscopy reveals a potential relationship between radiation-induced changes in rat brain metabolites and cognitive impairment. Radiat. Res. 168, 574–581. Chan, K.C., Khong, P.L., Cheung, M.M., Wang, S., Cai, K.X., Wu, E.X., 2009. MRI of late microstructural and metabolic alterations in radiation-induced brain injuries. J. Magn. Reson. Imaging 29, 1013–1020. Cheng, K.M., Chan, C.M., Fu, Y.T., Ho, L.C., Tsang, Y.W., Lee, M.K., Cheung, Y.L., Law, C. K., 2000. Brain abscess formation in radiation necrosis of the temporal lobe following radiation therapy for nasopharyngeal carcinoma. Acta Neurochir. 142, 435–440 (discussion 440-1). Cheung, M.M., Hui, E.S., Chan, K.C., Helpern, J.A., Qi, L., Wu, E.X., 2009. Does diffusion kurtosis imaging lead to better neural tissue characterization? A rodent brain maturation study. Neuroimage 45, 386–392. Chiang, C.W., Wang, Y., Sun, P., Lin, T.H., Trinkaus, K., Cross, A.H., Song, S.K., 2014. Quantifying white matter tract diffusion parameters in the presence of increased extra-fiber cellularity and vasogenic edema. Neuroimage 101, 310–319. Chong, V.F., Rumpel, H., Aw, Y.S., Ho, G.L., Fan, Y.F., Chua, E.J., 1999. Temporal lobe necrosis following radiation therapy for nasopharyngeal carcinoma: 1H MR spectroscopic findings. Int J. Radiat. Oncol. Biol. Phys. 45, 699–705. DeSalvo, M.N., 2012. Radiation necrosis of the pons after radiotherapy for nasopharyngeal carcinoma: diagnosis and treatment. J. Radiol. Case Rep. 6, 9–16. Gao, J., Li, X., Hou, X., Ding, A., Chan, K.C., Sun, Q., Wu, E.X., Yang, J., 2012. Tractbased spatial statistics (TBSS): application to detecting white matter tract variation in mild hypoxic-ischemic neonates. Conf. Proc. IEEE Eng. Med. Biol. Soc. 2012, 432–435. Giannelli, M., Toschi, N., Passamonti, L., Mascalchi, M., Diciotti, S., Tessa, C., 2012. Diffusion kurtosis and diffusion-tensor MR imaging in Parkinson disease. Radiology 265, 645–646 (author reply 646-7). Grinberg, F., Farrher, E., Kaffanke, J., Oros-Peusquens, A.M., Shah, N.J., 2011. NonGaussian diffusion in human brain tissue at high b-factors as examined by a combined diffusion kurtosis and biexponential diffusion tensor analysis. Neuroimage 57, 1087–1102. Hui, E.S., Cheung, M.M., Qi, L., Wu, E.X., 2008a. Towards better MR characterization of neural tissues using directional diffusion kurtosis analysis. Neuroimage 42, 122–134. Hui, E.S., Cheung, M.M., Qi, L., Wu, E.X., 2008b. Advanced MR diffusion characterization of neural tissue using directional diffusion kurtosis analysis. Conf. Proc. IEEE Eng. Med. Biol. Soc. 2008, 3941–3944. Hui, E.S., Cheung, M.M., Chan, K.C., Wu, E.X., 2010. B-value dependence of DTI quantitation and sensitivity in detecting neural tissue changes. Neuroimage 49, 2366–2374. Jensen, J.H., Helpern, J.A., Ramani, A., Lu, H., Kaczynski, K., 2005. Diffusional kurtosis imaging: the quantification of non-gaussian water diffusion by means of magnetic resonance imaging. Magn. Reson. Med. 53, 1432–1440. Jensen, J.H., Helpern, J.A., 2010a. MRI quantification of non-Gaussian water diffusion by kurtosis analysis. NMR Biomed. 23, 698–710. Jensen, J.H., Helpern, J.A., 2010b. MRI quantification of non-Gaussian water diffusion by kurtosis analysis. NMR Biomed. 23, 698–710. Kershaw, J., Leuze, C., Aoki, I., Obata, T., Kanno, I., Ito, H., Yamaguchi, Y., Handa, H., 2013. Systematic changes to the apparent diffusion tensor of in vivo rat brain measured with an oscillating-gradient spin-echo sequence. Neuroimage 70, 10–20. Laufs, A., Livingstone, R., Nowotny, B., Nowotny, P., Wickrath, F., Giani, G., Bunke, J., Roden, M., Hwang, J.H., 2014. Quantitative liver 31P magnetic resonance spectroscopy at 3T on a clinical scanner. Magn. Reson. Med. 71, 1670–1675. Lee, A.W., Ng, S.H., Ho, J.H., Tse, V.K., Poon, Y.F., Tse, C.C., Au, G.K., O, S.K., Lau, W.H., Foo, W.W., 1988. Clinical diagnosis of late temporal lobe necrosis following

393

radiation therapy for nasopharyngeal carcinoma. Cancer 61, 1535–1542. Lee, A.W., Ng, W.T., Chan, L.K., Chan, O.S., Hung, W.M., Chan, C.C., Cheng, P.T., Sze, H., Lam, T.S., Yau, T.K., 2012. The strength/weakness of the AJCC/UICC staging system (7th edition) for nasopharyngeal cancer and suggestions for future improvement. Oral Oncol. 48, 1007–1013. Lu, H., Jensen, J.H., Ramani, A., Helpern, J.A., 2006. Three-dimensional characterization of non-gaussian water diffusion in humans using diffusion kurtosis imaging. NMR Biomed. 19, 236–247. Lv, X.F., Zheng, X.L., Zhang, W.D., Liu, L.Z., Zhang, Y.M., Chen, M.Y., Li, L., 2014. Radiation-induced changes in normal-appearing gray matter in patients with nasopharyngeal carcinoma: a magnetic resonance imaging voxel-based morphometry study. Neuroradiology 56, 423–430. Minati, L., Aquino, D., Rampoldi, S., Papa, S., Grisoli, M., Bruzzone, M.G., Maccagnano, E., 2007. Biexponential and diffusional kurtosis imaging, and generalised diffusion-tensor imaging (GDTI) with rank-4 tensors: a study in a group of healthy subjects. Magma 20, 241–253. Obenaus, A., Huang, L., Smith, A., Favre, C.J., Nelson, G., Kendall, E., 2008. Magnetic resonance imaging and spectroscopy of the rat hippocampus 1 month after exposure to 56Fe-particle radiation. Radiat. Res. 169, 149–161. Qiu, W.Z., Huang, P.Y., Shi, J.L., Xia, H.Q., Zhao, C., Cao, K.J., 2016. Neoadjuvant chemotherapy plus intensity-modulated radiotherapy versus concurrent chemoradiotherapy plus adjuvant chemotherapy for the treatment of locoregionally advanced nasopharyngeal carcinoma: a retrospective controlled study. Chin. J. Cancer 35, 2. Sarmiento, M.P., Mejia, M.B., 2014. Preliminary assessment of nasopharyngeal carcinoma incidence in the Philippines: a second look at published data from four centers. Chin. J. Cancer 33, 159–164. Su, S.F., Han, F., Zhao, C., Huang, Y., Chen, C.Y., Xiao, W.W., Li, J.X., Lu, T.X., 2011. Treatment outcomes for different subgroups of nasopharyngeal carcinoma patients treated with intensity-modulated radiation therapy. Chin. J. Cancer 30, 565–573. Sun, Y., Zhou, G.Q., Qi, Z.Y., Zhang, L., Huang, S.M., Liu, L.Z., Li, L., Lin, A.H., Ma, J., 2013. Radiation-induced temporal lobe injury after intensity modulated radiotherapy in nasopharyngeal carcinoma patients: a dose-volume-outcome analysis. BMC Cancer 13, 397. Tan, X.P., Zhao, J.Q., Liang, B.L., Xie, B.K., Zhong, J.L., Ye, R.X., 2004. Diagnostic value of MR diffusion tensor imaging on radiation-induced early brain injury of nasopharyngeal carcinoma after radiotherapy. Ai Zheng 23, 1334–1337. Tang, Y.Y., Zhao, S.P., Xu, J., Zhong, Y., Xiao, J.Y., 2006. Combining FCU/5-FU suicide gene/prodrug system and radiation in treating nasopharyngeal carcinoma: an experimental study. Zhonghua Bing Li Xue Za Zhi 35, 483–487. Wang, H.Z., Qiu, S.J., Lv, X.F., Wang, Y.Y., Liang, Y., Xiong, W.F., Ouyang, Z.B., 2012. Diffusion tensor imaging and 1H-MRS study on radiation-induced brain injury after nasopharyngeal carcinoma radiotherapy. Clin. Radiol. 67, 340–345. Wang, S., Wu, E.X., Qiu, D., Leung, L.H., Lau, H.F., Khong, P.L., 2009. Longitudinal diffusion tensor magnetic resonance imaging study of radiation-induced white matter damage in a rat model. Cancer Res. 69, 1190–1198. Wang, X., Cusick, M.F., Wang, Y., Sun, P., Libbey, J.E., Trinkaus, K., Fujinami, R.S., Song, S.K., 2014. Diffusion basis spectrum imaging detects and distinguishes coexisting subclinical inflammation, demyelination and axonal injury in experimental autoimmune encephalomyelitis mice. NMR Biomed. 27, 843–852. Wang, Y., Wang, Q., Haldar, J.P., Yeh, F.C., Xie, M., Sun, P., Tu, T.W., Trinkaus, K., Klein, R.S., Cross, A.H., Song, S.K., 2011. Quantification of increased cellularity during inflammatory demyelination. Brain 134, 3590–3601. Wu, E.X., Cheung, M.M., 2010. MR diffusion kurtosis imaging for neural tissue characterization. NMR Biomed. 23, 836–848. Zhang, L.F., Li, Y.H., Xie, S.H., Ling, W., Chen, S.H., Liu, Q., Huang, Q.H., Cao, S.M., 2015. Incidence trend of nasopharyngeal carcinoma from 1987 to 2011 in Sihui County, Guangdong Province, South China: an age-period-cohort analysis. Chin. J. Cancer 34, 350–357. Zhou, X., Ou, X., Xu, T., Wang, X., Shen, C., Ding, J., Hu, C., 2014. Effect of dosimetric factors on occurrence and volume of temporal lobe necrosis following intensity modulated radiation therapy for nasopharyngeal carcinoma: a case-control study. Int. J. Radiat. Oncol. Biol. Phys. 90, 261–269.