Differentiation of recurrent diffuse glioma from treatment-induced change using amide proton transfer imaging: incremental value to diffusion and perfusion parameters

Purpose To evaluate the incremental value of amide proton transfer (APT) imaging to diffusion tensor imaging (DTI), dynamic susceptibility contrast (DSC) imaging, and dynamic contrast-enhanced (DCE) imaging in differentiating recurrent diffuse gliomas (World Health Organization grade II-IV) from treatment-induced change after concurrent chemoradiotherapy or radiotherapy. Methods This study included 36 patients (25 patients with recurrent gliomas and 11 with treatment-induced changes) with post-treatment gliomas. The mean values of apparent diffusion coefficient (ADC), fractional anisotropy (FA), normalized cerebral blood volume (nCBV), normalized cerebral blood flow, volume transfer constant, rate transfer coefficient, extravascular extracellular volume fraction, plasma volume fraction, and APT asymmetry index were assessed. Independent quantitative parameters were investigated to predict recurrent glioma using multivariable logistic regression. The incremental value of APT signal to other parameters was assessed by the increase of the area under the curve, net reclassification index, and integrated discrimination improvement. Results Univariable analysis showed that lower ADC (p= 0.018), higher FA (p= 0.031), higher nCBV (p= 0.021), and higher APT signal (p= 0.009) were associated with recurrent gliomas. In multivariable logistic regression, the diagnostic performance of the model with ADC, FA, and nCBV significantly increased when APT signal was added, with areas under the curve of 0.87 and 0.92, respectively (net reclassification index of 0.77 and integrated discrimination improvement of 0.13). Conclusion APT imaging may be a useful imaging biomarker that adds value to DTI, DCE, and DSC parameters for distinguishing between recurrent gliomas and treatment-induced changes.

Automated summary

APT imaging adds incremental value to DTI, DCE, and DSC parameters in distinguishing between recurrent gliomas and treatment-induced changes, improving the diagnostic performance of the model. This suggests that APT imaging may serve as a useful imaging biomarker in this context.