Non-Gaussian diffusion imaging with a fractional order calculus model to predict response of gastrointestinal stromal tumor to second-line sunitinib therapy

作者：Tang, Lei; Sui, Yi; Zhong, Zheng; Damen, Frederick C.; Li, Jian; Shen, Lin; Sun, Yingshi; Zhou, Xiaohong Joe^*

来源：Magnetic Resonance in Medicine, 2018, 79(3): 1399-1406.

DOI：10.1002/mrm.26798

• GIST
• targeted therapy response
• diffusion imaging
• FROC model
• sunitinib

PurposeTo demonstrate the clinical value of a non-Gaussian diffusion model using fractional order calculus (FROC) for early prediction of the response of gastrointestinal stromal tumor to second-line sunitinib targeted therapy. @@@ MethodsFifteen patients underwent sunitinib treatment after imatinib resistance. Diffusion-weighted imaging with multiple b-values was performed before treatment (baseline) and 2 weeks (for early prediction of response) after initiating sunitinib treatment. Conventional MRI images at 12 weeks were used to determine the good and poor responders according to the modified Choi criteria for MRI. Diffusion coefficient D, fractional order parameter (which correlates to intravoxel tissue heterogeneity), and a microstructural quantity mu were calculated using the FROC model. The FROC parameters and the longest diameter of the lesion, as well as their changes after 2 weeks of treatment, were compared between the good and poor responders. Additionally, the pretreatment FROC parameters were individually combined with the change in D (D) using a logistic regression model to evaluate response to sunitinib treatment with a receiver operating characteristic analysis. @@@ ResultsForty-two good-responding and 32 poor-responding lesions were identified. Significant differences were detected in pretreatment (0.67 versus 0.74, P=0.011) and D (45.7% versus 12.4%, P=0.001) between the two groups. The receiver operating characteristic analysis showed that D had a significantly higher predictive power than the tumor size change (area under the curve: 0.725 versus 0.580; 0.95 confidence interval). When D was combined with pretreatment , the area under the curve improved to 0.843 with a predictive accuracy of 75.7% (56 of 74). @@@ ConclusionsThe non-Gaussian FROC diffusion model showed clinical value in early prediction of gastrointestinal stromal tumor response to second-line sunitinib targeted therapy. The pretreatment FROC parameter can increase the predictive accuracy when combined with the change in diffusion coefficient during treatment. Magn Reson Med 79:1399-1406, 2018.

• 出版日期2018-3
• 单位北京大学; 北京市肿瘤防治研究所