Clinicians and radiologists interested in T1ρ MRI and cerebral tumors.Glioma is the most common primary cerebral tumor, which is classified by WHO as low-grade (grade Ⅰ and Ⅱ）or high-grade tumors (grade Ⅲ and Ⅳ) based on histopathologic characteristics. Accurate glioma grading is crucial for treatment planning and prognosis. High-grade gliomas are usually treated with surgery followed by radiation therapy and chemotherapy due to dismal prognosis, whereas low-grade gliomas are subject to either strict follow-up or surgery1. In addition, histopathologic assessment has significant limitations: It is an invasive procedure that has inherent sampling error related to stereotactic biopsy, and inability to evaluate residual tumor tissue after surgery2. Conventional MR imaging examination only reveals the morphology of tumors. T1 relaxation time in the rotating frame (T1ρ) is a new quantitative MR examination that can reflect the slow motional interactions between macromolecules and bulk water. Up to now, studies of T1ρ in glioma have only been implemented in animal model3, or utilized to delineate human gliomas at 0.1T4. In this research, we used the T1ρ examination at 3.0T to explore the correlation of the T1ρ value with pathological grades of human glioma. 29 patients (Mean age 38.1 yrs; 17F/12M) who had glioma confirmed by pathology and no history of previous treatment for brain tumor were recruited in this study. All patients underwent MRI scan with a 3.0T clinical scanner (Achieva 3.0T TX, Philips Healthcare, Best, Netherlands) using an 8 channel head coil. T1ρ was performed using Turbo Spin-Echo (TSE) pulse sequence, scanning parameters were as follows: TR/TE= 4800ms/229ms, FOV= 250×250mm2, flip angle= 90°, matrix=240×240, slice thickness=1.8mm, number of slices= 100, spin lock frequency= 500 Hz, spin lock time= 0, 20, 40, 60,80,100 ms respectively. According to 2007 WHO criteria, 18 patients were assigned to the low-grade group and 11 patients were in the high-grade group. All patients were scanned successfully. The representative T1ρ maps of the two groups are shown in Figure 1. Three ROIs were placed on the tumor parenchyma, with the necrotic, liquefacient and cystic region excluded (Figure 2). All the values were normalized to a ratio of tumor/normal parenchyma extracted by placing a ROI in normal white/gray matter of the contralateral hemisphere in a healthy-looking area of brain parenchyma. The mean, the minimal and the maximal T1ρ value of two groups are show in Table 1 respectively. Comparison between two groups was completed with independent sample T-test. The mean, the minimal and the maximal T1ρ values of the low-grade group were significantly higher than those of high-grade group (P＜0.01) (Figure 3, Table1). In this study, we investigated the T1ρ technique for gliomas grading in human. Preliminary results suggested that the T1ρ values are negatively correlated with the grade of gliomas. Macromolecules interaction in gliomas is complex, but we think that high cellularity and narrow extracellular space might impede this interaction and thus lead to a decreased T1ρ values. As a quantitative evaluation method, T1ρ MRI could avoid the mistakes caused by experience and subjective factors, and improving the accuracy and objectivity of pathological diagnosis. Therefore, T1ρ is potential to be a new indicator of malignancy degree in human gliomas. In future, other MRI techniques will be applied to determine diagnositic accuracy such as diffusion weighted imaging, perfusion weighted imaging or spectroscopic imaging, compared with the T1ρ MRI in our study. More patients at different grades of gliomas will also be recruited（gradeⅠ-Ⅳ）and analysed in depth the relation between the glioma grades and the T1ρ values. Our study clearly suggested that the T1ρ values could potentially serve as non-invasive predictors for the preoperative grading of gliomas. In clinical practice, T1ρ MRI may provide valuable reference in planning treatment and judging prognosis of gliomas.