Abstract
Based on the World Health Organization (WHO) 2021 brain tumor pathology classification, we evaluated the usefulness of dual-energy CT (DECT) for differentiating malignancy grades in malignant gliomas and examined its correlation with amide proton transfer (APT). A prospective observational study was conducted on 49 patients: 16 with glioblastoma (GBM, grade 4), 13 with astrocytoma (A3, grade 3), 10 with oligodendroglioma (O3, grade 3), and 10 with oligodendroglioma (O2, grade 2). Effective atomic number (Zeff) and electron density (ED) from DECT were analyzed for tumor grade differentiation and correlation with APT. High-grade gliomas, which are influenced by vascular endothelial growth factor (VEGF) and microvascular proliferation, showed significant differences and correlations in post-contrast Zeff. A correlation between Zeff and MIB-1 suggests its potential as an indicator of cell proliferation. Additionally, correlations between Zeff and APT, as well as between ED and APT, indicate that DECT may be useful for grading malignant gliomas.
Keywords: Dual-energy CT, amide proton transfer imaging, effective atomic number, electron density, glioblastoma, astrocytoma, oligodendroglioma, World Health Organization 2021
Introduction
The difficulty in establishing the differential diagnosis of malignant gliomas has been a longstanding challenge. 1 If the degree of malignancy could be determined before surgery, postoperative treatment strategies would be more straightforward. However, it remains difficult to assess malignancy accurately using functional imaging tests alone. Morphological imaging, particularly Magnetic Resonance Imaging (MRI), plays a crucial role in diagnosing malignant gliomas. 2
With the revision of the WHO 2021 brain tumor pathology classification, 3 the need for advanced functional imaging techniques has become apparent. Dual-energy CT (DECT), 4 the focus of this issue, is one such technique. DECT captures images of an object at two different energies (tube voltages), enabling substance discrimination or the creation of reference material images based on differences in absorption coefficients. It allows for various types of analyses, including effective atomic number analysis, electron density analysis, and material discrimination.
A key requirement for this analysis is that the data from the two different energy levels must be identical in both time and space. DECT (Aquilion ONE, Canon Otawara, Japan) can switch between 135 kV and 80 kV voltages in less than 0.2 s using a 2-turn kV-switching method. Two types of analysis are possible: image-based and raw database analysis. Using DECT, it is possible to determine the effective atomic number (Zeff) and electron density (ED) of a subject. The effective atomic number corresponds to a voxel when it is assumed that the voxel is composed of a single atom. By contrast, ED represents the number of electrons considered to be present in a unit volume.
For example, in water, Zeff measures 7.42 and ED measures 3.34. In this study, we hypothesized that by measuring Zeff and ED within a region of the malignant glioma tumor, these values may change depending on the malignancy. Therefore, we believe that DECT may hold promise as a new biomarker for malignant gliomas.5,6 We thus sought to verify whether DECT could aid in determining the differential diagnosis of malignancy, based on the diagnostic results of the WHO 2021 brain tumor pathological classification.
Recently, amide proton transfer (APT) imaging has gained attention as a promising functional imaging technique. APT is a form of MRI that exploits the Chemical Exchange Saturation Transfer (CEST) phenomenon. 7 During MRI imaging, the protons of a particular substance are saturated with a unique presaturation pulse. These protons exchange with those of free water, indirectly reducing the signal intensity of water—the target of MRI. This exchange phenomenon is known as CEST, and the imaging techniques that use are referred to as CEST imaging. APT imaging specifically focuses on amides, reflecting changes in the concentration, pH, and temperature of substances, and it is expected to have clinical applications such as distinguishing between recurrence and pseudo-progression and diagnosing gliomas based on malignancy grade.8–10 The increased concentration of amides in brain tumors is thought to result from the abundant cytoplasm within the tumor.11,12 In cases of cerebral ischemia, a decrease in the value of pH, due to anaerobic metabolism, and an increase in temperature, resulting from metabolic changes, impaired heat exchange, and early inflammatory responses, are also factors that may influence APT imaging. To create an APT image, a z-spectrum is generated, plotting changes in signal values. The Magnetization Transfer Ratio asymmetry (MTRasym), an index of CEST strength, is subsequently calculated and mapped to create the APT image.13,14 For example, APT signal measurement involves calculating the left-right difference by comparing the signal drop at +3.5 ppm, which is the resonance frequency of the amide group, with that at −3.5 ppm on the opposite side of the center frequency. Since APT images have shown potential as a potential biomarker for malignant gliomas, we also investigated whether a correlation exists between DECT and APT.
Case study
This study was approved by the Ethics Review Committee of the hospital (Clinical Research Approval Number: 20220099). The study design was a prospective observational study, and consent was obtained within 1 week prior to surgery. The study subjects were patients who attended or were admitted to our neurosurgery department between January 2023 and July 2024. Among the malignant glioma patients who underwent DECT and APT, 16 patients with glioblastoma grade 4 (GBM), 13 patients with astrocytoma grade 3 (A3), 10 patients with oligodendroglioma grade 3 (O3), and 10 patients with oligodendroglioma grade 2 (O2) were included, totaling 49 cases Table 1.
Table 1.
Cases with preoperative DECT and APT (January 2023–July 2024).
| Cases with preoperative DECT and APT imaging (WHO 2021 brain tumor pathology classification) | ||
|---|---|---|
| Grade 1 | Oligodendroglioma IDH-mutant 1p/19q-codeleted | 10 |
| Grade 1 | Astrocytoma IDH-mutant | 13 |
| Grade 1 | Oligodendroglioma IDH-mutant 1p/19q-codeleted | 10 |
| Grade 1 | Glioblastoma IDH-wildtype | 16 |
| Total | 49 | |
The study subjects were patients who attended or were admitted to our neurosurgery department between January 2023 and July 2024. Among the malignant glioma patients who underwent DECT and APT, 49 patients were included: 16 patients with Glioblastoma (GBM, grade 4), 13 patients with astrocytoma (A3, grade 3) 10 patients with oligodendroglioma (O3, grade 3), and 10 patients with oligodendroglioma (O2, grade 2).
The average age of patients was 48.6 ± 13.2 years, with a median age of 50 (26–74). The cohort consisted of 31 males and 18 female patients Table 2.
Table 2.
Patient demographics.
| Cases with preoperative DECT and APT imaging | |
|---|---|
| Age | |
| Average value (AV) ± standard deviation (SD) | 48.6 ± 13.2 |
| Median (range) | 50.0 (26–74) |
| Sex | |
| Male | 31 |
| Female | 18 |
The mean age was 48.6 ± 13.2 years, with a median age of 50 (range: 26–74). The cohort included 31 male and 18 female patients. The study was designed as a prospective observational study, and consent was obtained within 1 week prior to surgery.
Equipment used and inspection methods
The CT imaging equipment used was a Canon Aquilion ONE. The imaging conditions for DECT were as follows: tube voltage of 135 kV, tube current of 100 mA, imaging time of 0.5 s, image slice thickness of 0.5 mm, and the reconstruction interval was 0.5 mm.
At a tube voltage of 80 kV, the tube current was 570 mA, the imaging time was 0.5 s, the image slice thickness was 0.5 mm, and the reconstruction interval was 0.5 mm.
To account for potential changes in Zeff and ED values based on contrast media presence, DECT was performed before and after the preoperative cerebrovascular CT (arterial phase + venous phase: CT artery + CT vein) examination.
In our hospital, cerebrovascular CT is mandatory for patients with malignant gliomas Figure 1.
Figure 1.
DECT and cerebrovascular CT examination methods DECT was performed before and after the preoperative cerebrovascular CT examination (arterial + venous phases). Since cerebrovascular CT is mandatory for malignant gliomas patients at our hospital, Zeff and ED values were assessed with and without contrast media.
The fractional dose method (mgI/sec/kg), which specifies the rate of contrast injection per body weight, was used for contrast agent administration. Iopamiron 370 and iopamidol 370 (Bayer HealthCare, Osaka, Japan) were used as contrast media. The contrast medium auto-injector was a Dual-Shot Injector (Nemoto Kyorindo, Tokyo, Japan). The fractional dose was 25.9 mgI/sec/kg, the contrast agent was injected over 15 s, and the delay time between the arterial and venous phases (CT artery + CT vein) was 7 s. The MRI equipment used was an Ingenia 3.0 T (Philips Japan, Tokyo, Japan), with a 15-channel ds head coil phased array. The imaging conditions for APT were as follows: 3D, SE, TR6120, TE7.8, matrix 128 × 100, TSE es/shot (ms):7.8/1369, FOV 230 × 180, SENSE 1.6. No contrast agent was required, and the imaging time was 3 min 30 s.
Evaluation method
The measurement methods for Zeff, 15 ED, 16 and APT images were as follows: T2-weighted image (T2), fluid-attenuated inversion recovery (FLAIR), and T1-weighted (T1) gadolinium (Gd) images were referenced to set a region of interest in the tumor area, and Zeff, ED, and APT images were then measured. DECT and APT analyses were performed using raw data. Zeff and ED images were generated directly on the CT console, while APT images were created on the MRI console Figure 2.
Figure 2.
Evaluation method (glioblastoma IDH-wildtype grade 4). A: ED, B: Zeff, C: APT, D: H&E. Zeff, ED, and APT measurements were taken using T2-weighted (T2), fluid-attenuated inversion recovery (FLAIR), and T1-weighted (T1) gadolinium-enhanced (Gd) images to define the tumor region of interest. DECT and APT analyses were on the CT and MRI consoles, respectively.
Verification items
The following items were verified.
(1) Grade differentiation in Zeff: The differentiation of GBM, A3, O3, and O2 was assessed using Zeff. Additionally, the relative ratio of Zeff (brain tumor area/normal brain tissue: rZeff) was analyzed.
(2) Grade differentiation in ED: The differentiation of GBM, A3, O3, and O2 was evaluated using ED.
The relative ratio of ED (brain tumor area/normal brain tissue: rED) was also examined.
(3) Grade differentiation in APT: The differentiation of GBM, A3, O3, and O2 was assessed using APT.
(4) Correlations with MIB-1: Correlations (including relative ratios) between Zeff and MIB-1, ED and MIB-1, and APT and MIB-1 were analyzed.
(5) Correlations (including relative ratios) between Zeff and APT as well as ED and APT were examined.
Results
For the differential diagnosis of malignancy, a two-sample t-test was used to compare data by malignancy grade. Pearson’s correlation coefficient was used to assess correlations. Data visualization was performed using one-dimensional scatter plots. The Computed Tomography Dose Index (CTDI) 17 for DECT in this case was 18.8 mGy, while the CTDI for routine preoperative head CT was 36.5 ± 6.6 mGy.
(1) The mean and SD of non-contrast Zeff were GBM (7.66 ± 0.11), A3 (7.73 ± 0.19), O3 (7.68 ± 0.20), and O2 (7.78 ± 0.11). No significant differences were found Figure 3.
Figure 3.
GBM, A3, O3, and O2 grading in non-contrast Zeff. The mean and SD of non-contrast Zeff were GBM (7.66 ± 0.11), A3 (7.73 ± 0.19), O3 (7.68 ± 0.20), and O2 (7.78 ± 0.11). No significant differences were found in any of the cases. (a) Glioblastoma grade 4 (GBM) and astrocytoma grade 3 (A3). (b) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 3 (O3). (c) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 2 (O2). (d) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (e) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (f) Oligodendroglioma grade 3 (O3) and oligodendroglioma grade 2 (O2).
The mean and SD of contrast Zeff were GBM (8.05 ± 0.20), A3 (7.84 ± 0.18), O3 (7.84 ± 0.15), and O2 (7.79 ± 0.14). Here, significant differences were observed between GBM and A3 (p < 0.01), GBM and O3 (p < 0.05), and GBM and O2 (p < 0.01) Figure 4.
Figure 4.
GBM, A3, O3, and O2 grading in contrast-enhanced to Zeff. The mean and SD of contrast-enhanced Zeff were GBM (8.05 ± 0.20), A3 (7.84 ± 0.18), O3 (7.84 ± 0.15), and O2 (7.79 ± 0.14). Significant differences were observed between GBM and A3 (p < 0.01), GBM and O3 (p < 0.05), and GBM and O2 (p < 0.01). (a) Glioblastoma grade 4 (GBM) and astrocytoma grade 3 (A3). (b) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 3 (O3). (c) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 2 (O2). (d) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (e) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (f) Oligodendroglioma grade 3 (O3) and oligodendroglioma grade 2 (O2).
Furthermore, the mean and standard deviation of the non-contrast rZeff (x10−1) were GBM (10.11 ± 0.08), A3 (10.06 ± 0.25), O3 (10.04 ± 0.24), and O2 (10.03 ± 0.10). Significant differences were observed between GBM and O2 (p < 0.05) Figure 5.
Figure 5.
GBM, A3, O3, and O2 grading in non-contrast rZeff (rZeff: brain tumor area/normal brain tissue). The mean and standard deviation of the non-contrast rZeff (x10−1) were GBM (10.11 ± 0.08), A3 (10.06 ± 0.25), O3 (10.04 ± 0.24), and O2 (10.03 ± 0.10). Significant differences were observed between GBM and O2 (p < 0.05). (a) Glioblastoma grade 4 (GBM) and astrocytoma grade 3 (A3). (b) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 3 (O3). (c) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 2 (O2). (d) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (e) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (f) Oligodendroglioma grade 3 (O3) and oligodendroglioma grade 2 (O2).
Finally, the mean and standard deviation of contrast-enhanced rZeff (x10−1) were GBM (10.55 ± 0.21), A3 (10.15 ± 0.14), O3 (10.19 ± 0.12), and O2 (10.07 ± 0.16). In this case, significant differences were observed between GBM and A3 (p < 0.0001), GBM and O3 (p < 0.0001), GBM and O2 (p < 0.0001), and O3 and O2 (p < 0.001) Figure 6.
2. The mean and SD of non-contrast ED were GBM (3.43 ± 0.02), A3 (3.40 ± 0.03), O3 (3.41 ± 0.02), and O2 (3.38 ± 0.03). Significant differences were observed between GBM and A3 (p < 0.01), GBM and O2 (p < 0.0001), and O3 and O2 (p < 0.01) Figure 7.
Figure 6.
GBM, A3, O3, and O2 grading in contrast-enhanced rZeff. The mean and standard deviation of contrast-enhanced rZeff (x10-1) were GBM (10.55 ± 0.21), A3 (10.15 ± 0.14), O3 (10.19 ± 0.12), and O2 (10.07 ± 0.16). Significant differences were observed between GBM and A3 (p < 0.0001), GBM and O3 (p < 0.0001), GBM and O2 (p < 0.0001), and O3 and O2 (p < 0.001). (a) Glioblastoma grade 4 (GBM) and astrocytoma grade 3 (A3). (b) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 3 (O3). (c) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 2 (O2). (d) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (e) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (f) Oligodendroglioma grade 3 (O3) and oligodendroglioma grade 2 (O2).
Figure 7.
GBM, A3, O3, and O2 grading in non-contrast ED. The mean and SD of non-contrast ED were GBM (3.43 ± 0.02), A3 (3.40 ± 0.03), O3 (3.41 ± 0.02), and O2 (3.38 ± 0.03). Significant differences were observed between GBM and A3 (p < 0.01), GBM and O2 (p < 0.0001), and O3 and O2 (p < 0.01). (a) Glioblastoma grade 4 (GBM) and astrocytoma grade 3 (A3). (b) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 3 (O3). (c) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 2 (O2). (d) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (e) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (f) Oligodendroglioma grade 3 (O3) and oligodendroglioma grade 2 (O2).
The mean and SD of contrast ED were GBM (3.43 ± 0.02), A3 (3.40 ± 0.02), O3 (3.40 ± 0.02), and O2 (3.37 ± 0.02). Significant differences were observed between GBM and A3 (p < 0.01), GBM and O3 (p < 0.01), GBM and O2 (p < 0.0001), A3 and O2 (p < 0.05), and O3 and O2 (p < 0.05) Figure 8.
Figure 8.
GBM, A3, O3, and O2 grading in contrast-enhanced ED. The mean and SD of the contrast-enhanced ED were GBM (3.43 ± 0.02), A3 (3.40 ± 0.02), O3 (3.40 ± 0.02), and O2 (3.37 ± 0.02). Significant differences were observed between GBM and A3 (p < 0.01), GBM and O3 (p < 0.01), GBM and O2 (p < 0.0001), A3 and O2 (p < 0.05), and O3 and O2 (p < 0.05). (a) Glioblastoma grade 4 (GBM) and astrocytoma grade 3 (A3). (b) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 3 (O3). (c) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 2 (O2). (d) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (e) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (f) Oligodendroglioma grade 3 (O3) and oligodendroglioma grade 2 (O2).
The mean and standard deviation of non-contrast rED (x10−1) were GBM (10.00 ± 0.06), A3 (9.92 ± 0.08), O3 (9.95 ± 0.10), and O2 (9.92 ± 0.05). Significant differences were observed between GBM and A3 (p < 0.01) and GBM and O2 (p < 0.01) Figure 9.
Figure 9.
GBM, A3, O3, and O2 grading in non-contrast rED. The mean and standard deviation of the non-contrast rED (x10-1) were GBM (10.00 ± 0.06), A3 (9.92 ± 0.08), O3 (9.95 ± 0.10), and O2 (9.92 ± 0.05). Significant differences were observed between GBM and A3 (p < 0.01), and GBM and O2 (p < 0.01). (a) Glioblastoma grade 4 (GBM) and astrocytoma grade 3 (A3) (b) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 3 (O3). (c) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 2 (O2). (d) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (e) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (f) Oligodendroglioma grade 3 (O3) and oligodendroglioma grade 2 (O2).
The mean and SD of contrast rED (x10−1) were GBM (10.01 ± 0.06), A3 (9.93 ± 0.06), O3 (9.93 ± 0.06), and O2 (9.89 ± 0.05). Significant differences were observed between GBM and A3 (p < 0.01), GBM and O3 (p < 0.01), and GBM and O2 (p < 0.0001) Figure 10.
(3) The mean and SD of APT were GBM (4.10 ± 0.63), A3 (3.06 ± 0.43), O3 (2.69 ± 0.44), and O2 (1.59 ± 0.41). Significant differences were observed between GBM and A3 (p < 0.0001), GBM and O3 (p < 0.0001), GBM and O2 (p < 0.0001), A3 and O2 (p < 0.0001), and O3 and O2 (p < 0.0001) Figure 11.
Figure 10.
GBM, A3, O3, and O2 grading in contrast-enhanced rED. The mean and SD of contrast-enhanced rED (x10-1) were GBM (10.01 ± 0.06), A3 (9.93 ± 0.06), O3 (9.93 ± 0.06), and O2 (9.89 ± 0.05). Significant differences were observed between GBM and A3 (p < 0.01), GBM and O3 (p < 0.01), and GBM, and O2 (p < 0.0001). (a) Glioblastoma grade 4 (GBM) and astrocytoma grade 3 (A3). (b) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 3 (O3). (c) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 2 (O2). (d) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (e) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (f) Oligodendroglioma grade 3 (O3) and oligodendroglioma grade 2 (O2).
Figure 11.
GBM, A3, O3, and O2 grading in APT. The mean and SD of APT were GBM (4.10 ± 0.63), A3 (3.06 ± 0.43), O3 (2.69 ± 0.44), and O2 (1.59 ± 0.41). Significant differences were observed between GBM and A3 (p < 0.0001), GBM and O3 (p < 0.0001), GBM and O2 (p < 0.0001), A3 and O2 (p < 0.0001), and O3 and O2 (p < 0.0001). (a) Glioblastoma grade 4 (GBM) and astrocytoma grade 3 (A3). (b) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 3 (O3). (c) Glioblastoma grade 4 (GBM) and oligodendroglioma grade 2 (O2). (d) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (e) Astrocytoma grade 3 (A3) and oligodendroglioma grade 3 (O3). (f) Oligodendroglioma grade 3 (O3) and oligodendroglioma grade 2 (O2).
4. Significant correlations were observed between contrast Zeff and MIB-1 (r = 0.61, p < 0.0001), contrast-enhanced rZeff and MIB-1 (r = 0.68, p < 0.0001), non-contrast ED and MIB-1 (r = 0.39, p < 0.01), contrast ED and MIB-1 (r = 0.47, p < 0.001), non-contrast rED and MIB-1 (r = 0.28, p < 0.05), contrast rED and MIB-1 (r = 0.46, p < 0.001), and APT and MIB-1 (r = 0.7, p < 0.0001) Figures 12–14.
Figure 13.
Correlations between ED and MIB-1, and rED and MIB-1. Correlations were found in non-contrast ED and MIB-1 (r = 0.39, p < 0.01), contrast-enhanced ED and MIB-1 (r = 0.47, p < 0.001), non-contrast rED and MIB-1 (r = 0.28, p < 0.05) and contrast-enhanced rED and MIB-1 (r = 0.46, p < 0.001). (a) ED non-contrast and MIB-1. (b) ED contrast-enhanced and MIB-1. (c) rED (x10-1) non-contrast and MIB-1. (d) rED (x10-1) contrast-enhanced and MIB-1.
Figure 12.
Correlations between Zeff and MIB-1, and rZeff and MIB-1. Correlations were observed in Zeff of contrast-enhanced and MIB-1 (r = 0.61, p < 0.0001) and rZeff of contrast-enhanced and MIB-1 (r = 0.68, p < 0.0001). (a) Zeff non-contrast and MIB-1. (b) Zeff contrast-enhanced and MIB-1. (c) rZeff (x10-1) non-contrast and MIB-1. (d) rZeff (x10-1) contrast-enhanced and MIB-1.
Figure 14.
Correlations between APT and MIB-1. A correlation was found between APT and MIB-1 (r = 0.7, p < 0.0001).
5. Significant correlations were found between contrast Zeff and APT (r = 0.53, p < 0.0001), contrast-enhanced rZeff and APT (r = 0.69, p < 0.0001), non-contrast ED and APT (r = 0.53, p < 0.0001), contrast ED and APT (r = 0.6, p < 0.0001), and contrast rED and APT (r = 0.45, p < 0.001) Figures 15–16.
Figure 15.
Correlations between Zeff and APT. Correlations were observed in Zeff of contrast-enhanced and APT (r = 0.53, p < 0.0001) and rZeff of contrast-enhanced and APT (r = 0.69, p < 0.0001). (a) Zeff non-contrast and APT. (b) Zeff contrast-enhanced and APT. (c) rZeff (x10−1) non-contrast and APT. (d) rZeff (x10−1) contrast-enhanced and APT.
Figure 16.
Correlations between Zeff and APT. Correlations were observed in Zeff of contrast-enhanced and APT (r = 0.53, p < 0.0001) and rZeff of contrast-enhanced and APT (r = 0.69, p < 0.0001). (a) Zeff non-contrast and APT. (b) Zeff contrast-enhanced and APT. (c) rZeff (x10−1) non-contrast and APT. (d) rZeff (x10−1) contrast-enhanced and APT.
Discussion
In this study, we examined Zeff values. Non-contrast rZeff (x10−1) was significantly different between GBM and O2. Contrast Zeff also showed significant differences between GBM and A3, GBM and O3, and GBM and O2. Similarly, contrast-enhanced rZeff (x10−1) was significantly different between GBM and A3, GBM and O3, GBM and O2, and O3 and O2.
High-grade malignant gliomas, particularly GBM, exhibit a hypoxic tumor microenvironment and high vascular endothelial growth factor (VEGF) expression. 18 VEGF is a protein that promotes angiogenesis by inducing vascular endothelial cell division, migration, and differentiation, resulting in the formation of new blood vessels that branch off from existing blood vessels (angiogenesis). Tumors secrete VEGF to stimulate angiogenesis in the surrounding area, thereby increasing blood vessel formation are facilitating nutrient and oxygen supply. Histopathologically, GBM is characterized by necrosis surrounded by microvascular proliferation. Tumor vascular endothelial cells exhibit a glomeruloid structure resembling kidney glomeruli, with multilayered, disorganized arrangements, nuclear atypia, and mitotic figures. VEGF, produced by tumor cells and macrophages around necrotic tissue, plays a dual role in vascular endothelial cell proliferation and increased vascular permeability, contributing significantly to tumor angiogenesis in GBM. Considering the dependence of high-grade malignant gliomas on contrast uptake due to VEGF activity and microvascular proliferation, Zeff and rZeff values were higher post-contrast for higher-grade tumors, resulting in numerous significant differences.
ED analysis
We also analyzed ED values. Non-contrast ED showed significant differences between GBM and A3, GBM and O2, and O3 and O2. Non-contrast rED (x10−1) was significantly different between GBM and A3 and GBM and O2. Contrast-enhanced ED demonstrated significant differences between GBM and A3, GBM and O3, GBM and O2, A3 and O2, and O3 and O2. Furthermore, contrast-enhanced rED (x10−1) was significantly different between GBM and A3, GBM and O3, and GBM and O2.
Due to increased tumor cell density and nucleocytoplasmic ratio in high-grade malignant gliomas, particularly GBM, due to VEGF and microvascular proliferation,5,19 ED values increased with malignancy. Given that ED represents the number of electrons per unit volume, it remained largely unaffected by contrast media, showing little change in pre- and post-contrast measurements.
Therefore, we recommend non-contrast DECT for ED analysis.5,19
APT analysis
APT values were significantly higher in higher-grade tumors, with significant differences observed between GBM and A3, GBM and O3, GBM and O2, A3 and O2, and O3 and O2. High-grade gliomas exhibited elevated APT signals within the tumor region, likely due to increased amide concentrations from free proteins in highly proliferative tumor sites. This high proliferative potential of these tumors results in dense cytoplasm, contributing to the observed increase in APT signals.11,12
Correlations of MIB-1 with Zeff, ED, and APT
In this study, we found positive correlations between contrast Zeff and MIB-1, contrast-enhanced rZeff and MIB-1, non-contrast ED and MIB-1, non-contrast rED and MIB-1, contrast-enhanced ED and MIB-1, contrast-enhanced rED and MIB-1, and APT and MIB-1. 20
The MIB-1 Index quantifies proliferative activity using Ki-67 immunostaining, whereas higher values indicate higher tumor proliferative rates. Ki-67 is a cell cycle-associated nuclear protein that is expressed during active cell cycles (G1, S, G2, and M phases), but not in quiescent G0-phase cells, making it a crucial marker for tumor cell proliferation. Our findings suggest that Zeff, ED, and APT serve as indicators of tumor cell proliferative potential.
Correlations between Zeff and APT, and ED and APT
The positive correlations between Zeff and APT, as well as ED and APT, suggest that Zeff and ED hold potential for distinguishing malignancy in gliomas. These correlations arise from the presence of high-density cytoplasm in highly proliferative tumors, linking Zeff and ED to tumor grade differentiation (Table 3).
Table 3.
Imaging equipment and examination method.
| DECT | APT | |
|---|---|---|
| Equipment used | Canon Aquilion ONE | Philips Ingenia 3.0 T |
| Contrast medium | Iopamiron 370 or Iopamidol 370 | Non-contrast MRI |
| Imaging time, etc. | 135 KV+100 mA、80 KV+570 mA Slice thickness 0.5 mm Scan speed 0.5 sec |
Imaging time: 3 min and 30 s |
| Measurement method | Set the region of interest at the tumor site and measure the actual value of Zeff or ED signal | Set the region of interest at the tumor site and measure the actual value of APT signal |
The CT imaging equipment was a Canon Aquilion ONE. The DECT imaging conditions were as follows: tube voltage 135 kV; tube current: 100 mA; imaging time: 0.5 s; image slice thickness: 0.5 mm; and reconstruction interval: 0.5 mm. Additional DECT parameters, at a tube voltage of 80 kV, included: tube current: 570 mA; imaging time: 0.5 s; image slice thickness:0.5 mm; and reconstruction interval: 0.5 mm. The contrast media used were Iopamiron 370 and iopamidol370 (Bayer HealthCare, Osaka, Japan). The MRI imaging equipment used was Ingenia 3.0 T (Philips Japan, Tokyo, Japan), and the coil used was a 15-channel ds head coil phased array. No contrast agent was required, and the imaging time was 3 min 30 s.
Radiation dose considerations
In our study, the CTDI for DECT was 18.8 mGy, whereas the CTDI for standard preoperative head CT was 36.5 ± 6.6 mGy (Table 4).
Table 4.
Radiation dose comparison. DECT CTDI was 18.8 mGy and the standard head CT CTDI before surgery was 36.5 ± 6.6 mGy. DECT CTDI was approximately half the radiation compared to standard head CT.
| 1/2023–7/2024 | |
|---|---|
| Dose of simple CT of the head in the present study | |
| Exposure dose (CTDI※) | 36.5 ± 6.6 mGy |
| Dose in DECT | |
| Tube voltage | 135 KV + 80 KV |
| Tube current | 100 mA for 135 KV and 570 mA for 80 KV |
| Slice thickness | 0.5 mm |
| Scanning speed | 0.5 sec |
| Exposure dose (CTDI※) | 18.8 mGy |
CTDI: Computed Tomography Dose Index.
The CTDI for DECT was approximately half that of routine head CT. Additionally, the International Atomic Energy Agency (IAEA) guidance level for head CT is 50 mGy, indicating that our DECT dose was well below the recommended diagnostic reference level.21,22 Given that for functional imaging primarily measures function rather than morphology, maintaining a minimal radiation is crucial.
Study limitations
Small sample size
The conduct of this study was approved by the Ethics Review Committee of our hospital.
The study design was a prospective observational study, and consent was obtained within 1 week prior to surgery.
The study population consisted of 49 patients with malignant glioma who attended or were admitted to our neurosurgery department in approximately 18 months from January 2023 to July 2024 and for whom DECT and APT were performed. Due to the rarity of these tumors, the initial study cohort was around 80 patients, but after excluding those with only DECT and APT tests, the final sample size was 49. Future studies with larger sample sizes are necessary to validate our findings.
Comparison with other functional imaging modalities
Previous studies have explored functional imaging such as Magnetic Resonance Spectroscopy (MRS), 23 MRI-perfusion, 24 CT-perfusion, 25 and Arterial Spin Labeling image (ASL) 26 for glioma diagnosis. However, these tests are not routinely performed at our neurosurgery department, limiting direct comparisons in this study.
Study contributions
Previous studies by Kaichi et al. 5 and Chakrabarti R et al. 19 explored ED measurement in DECT for glioma differentiation. However, to our knowledge, no prior research has assessed both non-contrast and contrast-enhanced ED and Zeff measurements for differentiating glioma malignancy. Moreover, correlations between DECT ED and APT, and Zeff and APT, have not been previously reported. While previous studies5,19 focused on glioma classification under the WHO 2016 classification, our study evaluated gliomas from Grade 2 to Grade 4 using the WHO 2021 classification, uncovering significant differences in malignancy grading.
Conclusion
Differentiating glioma malignancy remains challenging, despite extensive research on functional imaging modalities such as MRS, MRI-perfusion, CT-perfusion, ASL, and 11C-Methionine-Positron Emission Tomography. 27 Our findings suggest that Zeff, ED, and APT obtained from DECT could serve as valuable, non-invasive biomarkers for differentiating glioma malignancy, aiding in preoperative diagnosis and treatment planning. Future research should focus on integrating morphological and functional imaging for precise tumor characterization. Improved diagnostic accuracy will aid in tailoring postoperative treatment strategies, ultimately enhancing patient outcomes.
Footnotes
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The authors received no financial support for the research, authorship, and/or publication of this article.
Ethical considerations
This retrospective study was approved by the Ethical Review Committee of Tokyo Women’s Medical University (Clinical Research Approval Number: 20220099).
ORCID iD
Masami Shirota https://orcid.org/0009-0004-4439-6446
References
- 1.Watanabe A, Muragaki Y, Maruyama T, et al. Usefulness of 11 C-methionine positron emission tomography for treatment-decision making in cases of non-enhancing glioma-like brain lesions. J Neuro Oncol 2016; 126: 577–583. [DOI] [PubMed] [Google Scholar]
- 2.Lee MD, Jain R, Galbraith K, et al. T2-FLAIR mismatch sign predicts DNA methylation subclass and CDKN2A/B status in IDH-mutant astrocytomas. Clin Cancer Res 2024; 30: 3512–3519. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Shibahara I, Kumabe T. Glioblastoma, IDH-wildtype. No shinkei geka. Neurolog Surg 2023; 51: 821–828. [DOI] [PubMed] [Google Scholar]
- 4.Tatsugami F, Higaki T, Nakamura Y, et al. Dual-energy CT: minimal essentials for radiologists. Jpn J Radiol 2022; 40: 547–559. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kaichi Y, Tatsugami F, Nakamura Y, et al. Improved differentiation between high-and low-grade gliomas by combining dual-energy CT analysis and perfusion CT. Medicine 2018; 97: e11670. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lv Y, Zhou J, Lv X, et al. Dual-energy spectral CT quantitative parameters for the differentiation of Glioma recurrence from treatment-related changes: a preliminary study. BMC Med Imag 2020; 20: 5–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Jones CK, Schlosser MJ, Van Zijl PC, et al. Amide proton transfer imaging of human brain tumors at 3T. Magn Reson Med 2006; 56: 585–592. [DOI] [PubMed] [Google Scholar]
- 8.Ward KM, Aletras AH, Balaban RS. A new class of contrast agents for MRI based on proton chemical exchange dependent saturation transfer (CEST). J Magn Reson 2000; 143: 79–87. [DOI] [PubMed] [Google Scholar]
- 9.Togao O, Hiwatashi A, Yamashita K, et al. Grading diffuse gliomas without intense contrast enhancement by amide proton transfer MR imaging: comparisons with diffusion-and perfusion-weighted imaging. Eur Radiol 2017; 27: 578–588. [DOI] [PubMed] [Google Scholar]
- 10.Togao O, Yoshiura T, Keupp J, et al. Amide proton transfer imaging of adult diffuse gliomas: correlation with histopathological grades. Neuro Oncol 2014; 16: 441–448. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Van Zijl PC, Yadav NN. Chemical exchange saturation transfer (CEST): what is in a name and what isn't? Magn Reson Med 2011; 65: 927–948. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Zhou J, Blakeley JO, Hua J, et al. Practical data acquisition method for human brain tumor amide proton transfer (APT) imaging. Magn Reson Med 2008; 60: 842–849. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kamimura K, Nakajo M, Yoneyama T, et al. Amide proton transfer imaging of tumors: theory, clinical applications, pitfalls, and future directions. Jpn J Radiol 2019; 37: 109–116. [DOI] [PubMed] [Google Scholar]
- 14.Wu B, Warnock G, Zaiss M, et al. An overview of CEST MRI for non-MR physicists. EJNMMI Phys 2016; 3: 19–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kawahara D, Ozawa S, Yokomachi K, et al. Synthesized effective atomic numbers for commercially available dual-energy CT. Rep Practical Oncol Radiother 2020; 25: 692–697. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Schaeffer CJ, Leon SM, Olguin CA, et al. Accuracy and reproducibility of effective atomic number and electron density measurements from sequential dual energy CT. Med Phys 2021; 48: 3525–3539. [DOI] [PubMed] [Google Scholar]
- 17.Botwe BO, Schandorf C, Inkoom S, et al. National indication-based diagnostic reference level values in computed tomography: preliminary results from Ghana. Phys Med 2021; 84: 274–284. [DOI] [PubMed] [Google Scholar]
- 18.Zheng F, Chen B, Zhang L, et al. Radiogenomic analysis of vascular endothelial growth factor in patients with glioblastoma. J Comput Assist Tomogr 2023; 47: 967–972. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Chakrabarti R, Gupta V, Vyas S, et al. Correlation of dual energy computed tomography electron density measurements with cerebral glioma grade. NeuroRadiol J 2022; 35: 352–362. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Xu J, Sheng Y, Li H, et al. A data-driven intravoxel mean diffusivities distribution approach for molecular classifications and MIB-1 prediction of gliomas. Med Phys 2024; 51: 7332–7344. [DOI] [PubMed] [Google Scholar]
- 21.Stadnyk L, Nosyk O, Tonkopi E, et al. Survey of computed tomography practice in Ukraine and establishment of national diagnostic reference levels for most common CT examinations of adults. Radiat Protect Dosim 2023; 199: 1142–1150. [DOI] [PubMed] [Google Scholar]
- 22.Alhujaili SF, Alshabibi AS, Alafer F, et al. Establishment of local diagnostic reference levels for head CT imaging in the madina region, Saudi Arabia. Diagnostics 2024; 14: 2882. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Zhang HW, Zhang HB, Liu XL, et al. Clinical assessment of magnetic resonance spectroscopy and diffusion-weighted imaging in diffuse glioma: insights into histological grading and IDH classification. Can Assoc Radiol J 2024; 75: 868–877. [DOI] [PubMed] [Google Scholar]
- 24.Zhao K, Huang H, Gao E, et al. Distributed parameter model of dynamic contrast-enhanced MRI in the identification of IDH mutation, 1p19q codeletion, and tumor cell proliferation in glioma patients. Front Oncol 2024; 14: 1333798. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Wang K, Li Y, Cheng H, et al. Perfusion CT detects alterations in local cerebral flow of glioma related to IDH, MGMT and TERT status. BMC Neurol 2021; 21: 460. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Flies CM, Snijders TJ, Van Seeters T, et al. Perfusion imaging with arterial spin labeling (ASL)–MRI predicts malignant progression in low-grade (WHO grade II) gliomas. Neuroradiology 2021; 63: 2023–2033. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Saito T, Maruyama T, Muragaki Y, et al. 11C-methionine uptake correlates with combined 1p and 19q loss of heterozygosity in oligodendroglial tumors. Am J Neuroradiol 2013; 34: 85–91. [DOI] [PMC free article] [PubMed] [Google Scholar]