Background and purpose: Magnetic resonance imaging (MRI) diffusion kurtosis imaging (DKI) is a new non-invasive functional imaging technology

which is the extension and development of diffusion tensor imaging (DTI) technology. This study aimed to analyse MRI-DKI technology parameters to predict the early response to radiotherapy in nasopharyngeal carcinoma (NPC). Methods: Sixty cases of nasopharyngeal carcinoma treated in Hainan General Hospital from November 2014 to August 2017 were collected. All patients underwent conventional MRI and DKI before and after radiotherapy. Based on MRI and biopsy results 3 months after radiotherapy

these patients were divided into response group (RG) and no-response group (NRG). Results: The maximum diameters of tumors in RG and NRG patients were similar prior to radiotherapy (P=0.057). The pretreatment axis diffusion coefficient (D

ax

)

mean diffusion coefficient (D

mean

) and radius diffusion coefficient (D

rad

) were higher in RG than in NRG patients (P=0.02

0.03 and 0.03). Conversely

the pretreatment fractional anisotropy (FA)

axis kurtosis coefficient (K

ax

)

kurtosis fractional anisotropy (K

FA

)

mean kurtosis coefficient (K

mean

)

radius kurtosis coefficient (K

rad

) and mean kurtosis tensor (Mkt) were lower in RG than in NRG patients (P=0.03

0.02

0.01

0.00

0.00 and 0.00). The K

rad

coefficient (0.76) was the best parameter to predict the radiotherapy response. Based on receiver operating characteristic curve analysis

K

rad

showed 77.8% sensitivity and 93.7% specificity ［area under curve (AUC)=0.913

95%CI: 0.811-1.000］. Multivariate analysis indicated DKI parameters were independent prognostic factors for the short-term effect in NPC. Conclusion: MRI-DKI technology could be used to predict the early response to radiotherapy in NP

C patients.