Kang et al. BMC Medical Imaging (2020) 20:50 https://doi.org/10.1186/s12880-020-00450-x RESEARCH ARTICLE Open Access Grading of Glioma: combined diagnostic value of amide proton transfer weighted, arterial spin labeling and diffusion weighted magnetic resonance imaging Xiao-wei Kang1,2†, Yi-bin Xi1,2†, Ting-ting Liu2,3, Ning Wang2,4, Yuan-qiang Zhu2, Xing-rui Wang2,5 and Fan Guo2,6* Abstract Background: To investigate the ability of amide proton transfer (APT) weighted magnetic resonance imaging (MRI), arterial spin labeling (ASL), diffusion weighted imaging (DWI) and the combination for differentiating high-grade gliomas (HGGs) from low-grade gliomas (LGGs). Methods: Twenty-seven patients including nine LGGs and eighteen HGGs underwent conventional, APT, ASL and DWI MRI with a 3.0-T MR scanner. Histogram analyses was performed and quantitative parameters including mean apparent diffusion coefficient (ADC mean), 20th-percentile ADC (ADC 20th), mean APT (APT mean), 90th-percentile APT (APT 90th), relative mean cerebral blood flow (rCBF mean) and relative 90th-percentile CBF (rCBF 90th) were compared between HGGs and LGGs. The diagnostic performance was evaluated with receiver operating characteristic (ROC) analysis of each parameter and their combination. Correlations were analyzed among the MRI parameters and Ki-67. Results: The APT values were significantly higher in the HGGs compared to the LGGs (p < 0.005), whereas ADC values were significantly lower in HGGs than LGGs (P < 0.0001). The ADC 20th and APT mean had higher discrimination abilities compared with other single parameters, with the area under the ROC curve (AUC) of 0.877 and 0.840. Adding ADC parameter, the discrimination ability of APT and rCBF significantly improved. The ADC was negatively correlated with the APT and rCBF value, respectively, while APT value was positively correlated with rCBF value. Significant correlations between ADC values and Ki-67 were also observed. Conclusions: APT and DWI are valuable in differentiating HGGs from LGGs. The combination of APT, DWI and ASL imaging could improve the ability for discriminating HGGs from LGGs. Keywords: Glioma, Magnetic resonance imaging, Arterial spin labeling, Amide proton transfer, Apparent diffusion coefficient * Correspondence: [email protected] †Xiao-wei Kang and Yi-bin Xi contributed equally to this work. 2Department of Radiology, Xijing Hospital, Xi’an, ShaanXi, China 6Key Laboratory of Molecular Imaging of the Chinese Academy of Sciences, Institute of Automation, Chinese Academy of Sciences, Beijing, China Full list of author information is available at the end of the article © The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Kang et al. BMC Medical Imaging (2020) 20:50 Page 2 of 8 Background pathological condition of gliomas still needs further Gliomas are the most common primary tumors of the verification. central nervous system [1]. The ability of discriminating The present study aims to evaluate and compared the low-grade glioma (LGG) from high-grade glioma (HGG) diagnostic performance of ASL, APT and DWI and their is of clinical importance as the prognosis and the standard combination in reflecting the histopathological charac- management is different substantially according to the teristics and differentiating HGGs from LGGs among grade. Surgery is an important treatment for HGGs and gliomas. usually followed by concurrent chemoradiation [2]. Mis- diagnosing HGGs as LGGs could lead to insufficient and Methods less aggressive treatment [3]. The current gold standard Patients for gliomas’ diagnosis and therapeutic decision relies on This retrospective study was approved by the institu- the histopathology, as well as the molecular profile and tional review board of Xijing hospital, and the require- genetic information [4]. However, the histopathological ment for informed consent was waived. The result depends on biopsy or surgical resection, which is preoperative MR imaging data of the 27 consecutive pa- not only invasive, but also affected by the intratumoral tients with gliomas who were identified between Sep- histological heterogeneity and sampling erros, which may tember 2015 and June 2018 were analyzed. The lead to underestimation of the true grades [3]. histopathologic diagnosis and grades were determined In the clinical practice, preoperative magnetic resonance with resection or biopsy specimens according to the imaging (MRI) contributes to the diagnosis and thus treat- WHO criteria by two established neuropathologists, who ment of glioma patients noninvasively. Traditional MRI, were blinded to the imaging findings. A total of 27 pa- such as T2-weighted and contrast enhanced T1-weighted tients (12 males and 15 females, age range from 29 to imaging, is conventionally used for diagnosing for character- 80) including 9 with LGGs (World Health Organization izing gliomas [5]. HGGs usually show moderate to strong [WHO] grade II and 20 with HGGs (WHO grade III or enhancement whereas LGGs showed no or mild enhance- IV) were identified. The characteristics of patients are ment. However, approximately 14–45% HGGs showed no described in Table 1. enhancement after contrast agents administration, whereas The patients’ histological types of gliomas were as fol- about 20% enhanced gliomas were histologically proved to lows: 6 diffuse astrocytomas (3 IDH-mutant, 3 IDH be LGGs [6, 7]. Therefore, the accurate grading for gliomas wild-type), 3 oligodendrogliomas (3 IDH-mutant and are still challenging with traditional MRI. 1p19q-codeleted), 2 anaplastic astrocytomas (1 IDH Advanced magnetic resonance techniques like dynamic wild-type, 1 IDH-mutant), 1 anaplastic oligodendrogli- contrast-enhanced (DCE) MRI, susceptibility-weighted im- oma (IDH-mutant and 1p19q-codeleted), 1 anaplastic aging (SWI), diffusion-weighted imaging (DWI) and intra- voxel incoherent motion (IVIM) have been utilized for Table 1 Demographics of patients – glioma grading [8 11]. Amide proton transfer (APT) LGG (n = 9) HGG (n = 18) weighted imaging is a noninvasive emerging molecular Sex (male: female) 3: 6 9: 9 MRI technique based on chemical exchange saturation Age (years) 48.3 ± 12.2 52.3 ± 13.0 transfer (CEST) between amide protons of proteins and polypeptides and free water protons [12, 13]. Previous study Histologic feature indicated that APT imaging could be used for grading gli- Glioblastoma (WHO IV) 14 omas [3, 4, 14, 15]. It is also valuable for distinguishing Anaplastic Astrocytomas (WHO III) 2 pseudoprogression from true progression in gliomas and Anaplastic Oligodendroglioma (WHO III) 1 reflecting treatment response [16, 17]. Arterial spin labeling Anaplastic Oligoastrocytoma (WHO III) 1 (ASL) offers non-invasive quantitative measurement of Diffuse astrocytomas (WHO II) 6 cerebral tissue blood perfusion using magnetically labeled arterial blood water as an endogenous tracer. As one non- Oligodendrogliomas (WHO II) 3 contrast perfusion technique, ASL might be one reprodu- Enhancement pattern cible, quantitative way in daily practice. Multiple studies Markedly heterogeneous enhancement 13 have reported ASL applied in brain tumor diagnosis Local Enhancement 2 3 and grading [18–20]. Nevertheless, APT and ASL Non-enhancement 7 2 techniques are still limited evaluated in clinical prac- Number of lesion tice and few analysis has been performed with the combination of multiple MRI techniques for grading Single lesion 9 16 gliomas [21]. Whether advanced MRI techniques and Multiple lesion 0 2 their combination could accurately reflect the Values are mean ± standard deviations Kang et al. BMC Medical Imaging (2020) 20:50 Page 3 of 8 oligoastrocytoma (NOS) and 14 glioblastoma multiforme FuncTool software). Data processing was performed by (GBM, 14 IDH wild-type). All patients underwent MR specialists with 10 years of clinical experience in MR contrast-enhanced T1- weighted imaging, APT, DW and pulse sequence. For the APT quantification, after water ASL imaging in their preoperative examinations. The frequency shift correction, magnetization transfer com- interval between the MR imaging and the surgery was < ponent and the APT (Δω = 3.5 ppm) component, asym- 2 weeks in all patients. Immunohistochemistry was used metrical MT ratio (MTRasym) analysis was performed to measure the Ki-67 labeling index and 18 patients were according to previous studies [3, 17]. For the region-of- recorded (6 LGGs and 12 HGGs). interests (ROIs) were manually segmented on FLAIR and matched to ADC, cerebral blood flow (CBF) and MR imaging