Jinyuan Zhou1, Erik Tryggestad2,
Zhibo Wen1, Bachchu Lal3, Tingting Zhou1,
Rachely Grossman4, Kun Yan1, Silun Wang1,
De-Xue Fu5, Eric Ford2, John Laterra3, Peter
C.M. van Zijl1
1Department of Radiology, Johns Hopkins
University, Baltimore, MD, United States; 2Department of Radiation
Oncology, Johns Hopkins University, Baltimore, MD, United States; 3Department
of Neurology, Kennedy Krieger Institute, Baltimore, MD, United States; 4Department
of Neurosurgery, Johns Hopkins University, Baltimore, MD, United States; 5Department
of Oncology, Johns Hopkins University, Baltimore, MD, United States
We
show that it is possible to differentiate between glioma and radiation
necrosis using the amide proton signals of endogenous cellular proteins and
peptides as imaging biomarker. Using a radiation necrosis model (dose, 40 Gy;
area, 10x10 mm2) and a SF188/V+ human glioma model in rats, tumors and
radiation necrosis had similar conventional MRI features. However, gliomas
were consistently hyperintense on amide proton transfer (APT) images, while
radiation necrosis (observed about six months post-radiation) was hypointense
to isointense. APT MRI as an imaging biomarker for tumor presence provides
unique visual information for assessing active tumor versus treatment-related
injury, such as radiation necrosis.