Intravoxel incoherent motion diffusion weighted imaging in evaluating the radio-sensitivity of nasopharyngeal carcinoma xenografts
Youping Xiao1, Yunbin Chen1, Jianji Pan2, Dechun Zheng1, Xiang Zheng1, and Ying Chen1

1Radiology, Fujian Provincial Cancer Hospital, Fuzhou, China, People's Republic of, 2Radiation Oncology, Fujian Provincial Cancer Hospital, Fuzhou, China, People's Republic of


In this present study, by applying the special mouse coil(4 channel), IVIM-DWI with 14 b-factors(0~1000s/mm2) was successfully conducted on nude mice with different radio-sensitive NPC xenografts(CNE-1 and CNE-2) during the course of fractional radiations. The IVIM-DWI parameters of xenografts were found to change characteristically after fractional radiations and were significantly different between different radio-sensitive NPC xenografts, and their corresponding changes also behaved significant correlations with the pathological features of NPC xenografts. Thus, it is suggested IVIM-DWI parameters be valuable in evaluating the micro-structures and radio-sensitivity of NPC xenografts.


The radio-sensitivity is deem as a significant prognostic indicator for nasopharyngeal carcinomoa(NPC) and evaluation of radio-sensitivity of NPC is valuable in improving patient's treatment outcomes. Intravoxel incoherent motion diffusion weighted imaging (IVIM-DWI) is a novel functional magnetic resonance imaging(MRI) technique which has been demonstrated with excellent diagnostic capability[1,2]. The purpose of this study is to initially investigate the characteristics of parameters in different radio-sensitive NPC xenografts after fractional radiations.


Two types of different radio-sensitive NPC cell lines (CNE-1 and CNE-2) were transplanted to raise xenografts on sixty nude mice (30 of CNE-1 and 30 of CNE-2), which then underwent the fractional radiations with a fraction of 10Gy at the alternative days. Nude mice with CNE-1 and CNE-2 xenografts were sub-categorized into the following five groups: non-radiation group (G0), radiation groups of 10Gy (G1), 20Gy (G2), 30Gy (G3), and three days after 30Gy (G4), respectively. On a 3.0T MR system (Philips Healthcare, Best, The Netherlands), by applying a special four channel mouse coil (Chenguang, Shanghai, China), IVIM-DWI with 14 b-factors (0, 10, 20, 30, 40, 50, 100, 150, 200, 350, 500, 650, 800, and 1000 s/mm2) were performed directly on G0 xenografts and radiation groups (G1, G2, G3 and G4) after their fractional radiations were completed. IVIM-DWI parameters of xenografts were then analyzed and calculated with the IDL 6.3 software (Boulder, Chicago, USA). And the cell density and necrosis proportion of xenografts were also analyzed histopathologically. All statistical analyzes were conducted on the SPSS 18.0 software (Chicago, IL, USA). IIVIM-DWI parameters and pathological features between CNE-1 and CNE-2 xenografts were compared by the Student t test or Mann-Whitney U test; and the correlations between different variables were analyzed with Spearman test.


After fractional radiations, the general changes of D, f and D* values in CNE-2 xenografts were significant (P<0.01), while those of CNE-1 xenografts were not significant (P>0.1). During the course of fractional radiations, D value increased and D* and f values decreased more significantly in CNE-2 xenografts than in CNE-1 xenografts (P<0.01). D values and the necrosis proportion of G3 and G4 of CNE-2 xenografts were significantly higher than those of CNE-1 xenografts (P<0.05), while D* and f values of G4 as well as the cell density of G2, G3 and G4 in CNE-2 xenografts were significantly lower than those in CNE-1 xenografts (P<0.005). Furtheromre, D value behaved a negative correlation with the cell density and a positive correlation with the necrosis proportion of xenografts, while f value correlated positively with the cell density but negatively with the necrosis proportion of xenografts (rs>0.6; P<0.001), respectively.


The IVIM-DWI is a novel DWI technique basing on the multi-b factors scanning model and bi-exponential algorithm model of IVIM theory[3]. The IVIM model allows to evaluate the diffusion and perfusion characteristics simultaneously in a single sequence, and enables derivation of quantitative parameters that separately reflect tissue’s diffusivity and microcapillary perfusion without any contrast agents[3, 4]. In a previous study, Kim S et al.[5] have successfully applied the IVIM-DWI on mouse mammary carcinoma model and suggested that the pseudo-diffusion behaved an inverse correlation with interstitial fluid pressure (IFP). In this present study, IVIM-DWI was also successfully and reliably performed on NPC xenografts of different radio-sensitive cell lines to characterize their diffusion and perfusion features during the course of fractional radiations. After fractional radiations, especially after 30Gy radiation, the higher radio-sensitive CNE-2 xenografts presented more significant changes on both the IVIM-DWI parameters and the pathological features than the lower radio-sensitive CNE-1 xenografts. It is indicated that the changes of IVIM-DWI parameters of NPC xenografts during the course of fractional radiations can indirectly reflect their changes of pathological micro-structure features and potentially predict their radio-sensitivities. Furthermore, D and f values of xenografts also correlated significantly with the micro-structure features of xenografts. Higher radio-sensitive NPC xenografts which presented larger areas of necrosis and less cell density in tumor tissues would appear to exhibit a higher D value and a lower f value after radiations. Thus, D and f values can be more powerful in characterizing the pathological micro-structure features and predicting the radio-sensitivity for NPC xenografts.


IVIM-DWI derived parameters and their corresponding changes after fractional radiations were characteristically and significantly different between different radio-sensitive NPC xenografts of CNE-1 and CNE-2 cell lines, and their corresponding changes behaved significant correlations with the pathological features of xenografts. Thus, IVIM-DWI parameters can be valuable in evaluating the micro-structures of NPC xenografts and potentially predicting their radio-sensitivity.


This study is partly supported by the National Clinical Key Specialty Construction Program and Key Clinical Specialty Discipline Construction Program of Fujian.


[1]Sumi M, Van Cauteren M, Sumi T, et al. Salivary gland tumors: useof intravoxel incoherent motion MR imaging for assessment ofdiffusion and perfusion for the differentiation of benign frommalignant tumors. Radiology. 2012; 263: 770–777.

[2]Zhang SX, Jia QJ, Zhang ZP, et al. Intravoxel incoherent motionMRI: emerging applications for nasopharyngeal carcinoma at theprimary site. Eur Radiol. 2014; 24: 1998–2004.

[3]Le Bihan D. Diffusion, confusion and functional MRI. Neuroimage.2012; 62: 1131–1136.

[4]Koh DM, Collins DJ, Orton MR. Intravoxel incoherent motion in body diffusion-weighted MRI: reality and challenges. AJR Am J Roentgenol. 2011. 196(6): 1351-61.

[5]Kim S; Decarlo L; Cho GY; et al. Interstitial fluid pressure correlates with intra-voxel incoherent motion imaging metrics in a mouse mammary carcinoma model. NMR Biomed. 2012. 25(5):787-94.


IVIM-DWI images on a NPC xenograft of CNE-2 cell line

a. S0 map; b. the fitting curve of IVIM model; c. D map; d. f map; e. D* map; f. goodness of fit R2 map.

Pathological features(cell density) of CNE-2 xenografts before and after fractional radiations.

a. G0 xenograft; b. G1 xenograft; c. G2 xenograft; d. G3 xenograft.

The change trends of IVIM-DWI parameters (D, f, D*) and pathological features (cell density and necrosis proportion) between CNE-1 and CNE-2 xenografts during the course of fractional radiations.

a. D value; b. f value; c. D* value; d. cell density; e. necrosis proportion.

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)