Preoperative prediction of Ki-67 expression in hepatocellular carcinoma by spectral imaging on dual-energy computed tomography (DECT)

Introduction

Hepatocellular carcinoma (HCC) is one of the most common malignant tumors worldwide, and the long-term prognosis of HCC patients remains poor (1). Cyclic nuclear protein Ki-67 participates in the proliferation of tumor cells, and is related to tumor occurrence, progression, metastasis, and patient prognosis (2). Thus, it can be used as an independent prognostic indicator. In the clinic, the expression level of Ki-67 is often obtained from positive cell counts in pathological specimens. However, the Ki-67 test cannot be performed on patients who are not suitable for surgery. Therefore, a non-invasive method needs to be established to predict Ki-67 expression status in vivo to guide the clinical treatment and postoperative monitoring of HCC patients.

Due to its short scan time and use of three-dimensional anatomical imaging, conventional computed tomography (CT) has become a powerful tool in HCC evaluation. The recently developed single-source fast kVp switching dual-energy computed tomography (DECT) performs a spectral analysis in the projection data space and generates functional images, such as monochromatic energy images, spectral curves, material decomposition images, and effective atomic number maps (3). The quantitative data acquired from spectral imaging can be used to analyze the compositions of tissue and lesions, providing radiologic markers for the diagnosis, classification, and prognosis prediction of tumors.

DECT parameters have been used to predict Ki-67 expression in gastric adenocarcinoma, pancreatic ductal adenocarcinoma, and non-small cell lung cancer (4-6), but reports on the use of DECT parameters to predict Ki-67 expression in HCC are rare. This study sought to investigate the value of DECT in predicting the Ki-67 expression in HCC, and to provide an in-vivo method for evaluating cell proliferation in HCC. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-24-461/rc).

Methods

Patient selection

The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the Ethics Committee of Zhongshan Hospital Affiliated to Xiamen University (approval No. 2022-251), and the Xiamen Hospital of Traditional Chinese Medicine also approved the study. The requirement of individual consent for this retrospective analysis was waived.

The data of patients with pathologically confirmed diagnoses of HCC were collected consecutively at the Zhongshan Hospital Affiliated to Xiamen University and the Xiamen Hospital of Traditional Chinese Medicine from December 2017 to December 2023. To be eligible for inclusion in this study, the patients had to meet the following inclusion criteria: (I) have a pathologically confirmed diagnosis of HCC; (II) have undergone double-phase contrast-enhanced DECT scanning within four weeks before surgery; (III) have undergone a Ki-67 immunohistochemical analysis; and (IV) have no contraindications to the use of iodine contrast agents. Patients were excluded from the study if they met any of the following exclusion criteria: (I) had incomplete clinical data (n=25); (II) had poor-quality images (n=10); and/or (III) had received interventional therapy before surgery (n=25). Ultimately, 123 HCC patients were included in this study (Figure 1).

Figure 1 Flowchart of the patient selection process for this study. HCC, hepatocellular carcinoma; CT, computed tomography.

CT acquisition

A contrast-enhanced DECT examination of the upper abdomen was performed with a 256-row CT scanner (Revolution, GE Healthcare, Milwaukee, USA). The scanning range was from the top of the diaphragm to the lower edge of the liver. Scanning parameters included the gem-stone imaging scanning mode, fast switching of tube voltage between 80 and 140 kVp, a tube current of 400 mAs, a detector width of 80 mm, a pitch of 0.992:1, a rotation rate of 0.8 seconds, a matrix of 512×512, and an adaptive statistical iterative reconstruction of 50%. For contrast injection, 65–85 mL of iopromide (370 mg/mL, Bayer, Germany) was injected through the antecubital vein at a flow rate of 3 mL/s. CT value monitoring (Smart Prep technology) in the abdominal aorta at the level of the celiac trunk was used to trigger the scan [threshold: 120 Hounsfield units (HU)], and the arterial phase (AP) scan was started after a delay of 6 seconds. While the portal venous phase (PVP) scan was performed after a delay of 20 seconds after the end of the AP.

Image processing and data analysis

The data were transmitted to an Advanced Workstation 4.7 (GE Healthcare, Milwaukee, USA) for post-processing and measurement. Monochromatic energy images in 130 and 140 keV, water- and fat-based material decomposition images, and effective atomic number maps were generated. Next, two radiologists (with 12 and 5 years of experience in abdominal CT imaging, respectively), who were blinded to the pathological results, independently conducted the image analyses and parameter measurements for all cases. To evaluate the morphologic characteristics, the necrosis in and capsule of each lesion were observed in 130 and 140 keV monochromatic energy images in the PVP. For the measurement of a lesion, the region of interest (ROI) was positioned on the lesion to include the highly enhanced areas as much as possible, while avoiding large blood vessels and artifact areas. The recorded measurements included the CT values in the monochromatic energy images of the lesions (HU_{130 keV} and HU_{140 keV}), and the fat density (D_fat), and the water density (D_water) of the lesions. D_fat was measured on the fat (iodine) images, and D_water was measured on the water (iodine) images. The effective atomic number (EffZ) of the lesions (EffZ_lesion) and the EffZ of the abdominal aorta (EffZ_aorta) in the AP and PVP were also measured (Figures 2,3). To minimize variations among individuals, the EffZ_lesion was normalized by the EffZ_aorta at the same level to derive a normalized effective atomic number (NeffZ). The NeffZ value was calculated using the following formula: NeffZ = EffZ_lesion / EffZ_aorta.

Figure 2 A 66-year-old male with HCC showing a HU_{130 keV}-PVP value for the lesion of 42.60 HU (black circle), a HU_{140 keV}-PVP value for the lesion of 41.40 HU (black circle), a D_water-PVP value for the lesion of 1,035 (mg/mL) (black circle), a NeffZ-AP value for the lesion of 0.65 (black circle and green bars), and a Ki-67% of 10% (IHC staining, ×200). HU_{130 keV} and HU_{140 keV}, CT values in 130 and 140 keV monochromatic energy images, respectively. HCC, hepatocellular carcinoma; HU, Hounsfield units; D_water, water density; PVP, portal venous phase; NeffZ, normalized effective atomic number; AP, arterial phase; IHC, immunohistochemical.

Figure 3 A 62-year-old male with HCC showing a HU_{130 keV}-PVP value for the lesion of 55.70 HU (black circle), a HU_{140 keV}-PVP value for the lesion of 54.50 HU (black circle), a D_water-PVP value for the lesion of 1,041 (mg/mL) (black circle), a NeffZ-AP value for the lesion of 0.68 (black circle and yellow bars), and a Ki-67% of 65% (IHC staining, ×200). HU_{130 keV} and HU_{140 keV}, CT values in 130 and 140 keV monochromatic energy images, respectively. HCC, hepatocellular carcinoma; HU, Hounsfield units; D_water, water density; PVP, portal venous phase; NeffZ, normalized effective atomic number; IHC, immunohistochemical; AP, arterial phase.

Immunohistochemical staining of Ki-67

The surgically resected specimens were fixed with 10% paraformaldehyde solution, embedded in paraffin, then cut into 4 µm-thick slices for the immunohistochemistry analysis of Ki-67 expression. The slices were stained with rabbit monoclonal primary antibody [CONFIRM^TM anti-Ki-67(30-9), Ventana, Arizona, USA], and then covered with a standard avidin-biotin-peroxidase complex, and soaked in 3,3'-diaminobenzidine solution. The positive rate of Ki-67 (Ki-67%) was determined in the tumor regions with the highest cell density using the percentage of immunoreactive cells of 1,000 malignant tumor cells. In this study, in accordance with previous research (7), the patients with a Ki-67% >20% were allocated to the high-expression group (n=74), and those with a Ki-67% ≤20% were allocated to the low-expression group (n=49). The immunochemical results were retrospectively analyzed by a physician who specialized in pathological diagnosis and was blinded to the clinical information.

Statistical analysis

SPSS 26.0 (IBM Corp., Armonk, NY, USA) and GraphPad Prism 9.0 (GraphPad Software, La Jolla, CA, USA) statistical software were used for the statistical analysis. An intraclass correlation coefficient (ICC) analysis and a Bland-Altman analysis were used to analyze the correlation and consistency of the measurements between the two radiologists, and the measurement data of the senior physician were selected for the subsequent statistical analysis. For the ICC analysis, an ICC >0.75 indicated excellent agreement, an ICC of 0.50–0.75 indicated good agreement, and an ICC <0.50 indicated poor agreement. The Kolmogorov-Smirnov test was used to analyze the normality of the measurement data. Continuous variables with a normal distribution are presented as the mean ± standard deviation, and continuous variables with a non-normal distribution are presented as the median/interquartile range. Categorical variables are presented as the number (n). The Spearman correlation coefficient was used to analyze the correlations between the DECT parameters and Ki-67%; an excellent correlation was defined as a rho ranging from 0.75–1.00, a moderate-to-good correlation was defined as a rho ranging from 0.50–0.74, a fair correlation was defined as a rho ranging from 0.25–0.49, and a rho of ≤0.24 indicated little or no correlation (8). The two-sample t-test, Mann-Whitney U test, and χ² test were used to examine the differences between the two groups. A multivariate logistic regression analysis was used to identify independent predictors of high Ki-67 expression. The receiver operating characteristic (ROC) curve was used to evaluate the ability of each parameter to differentiate between high and low Ki-67 expression levels. Binary logistic regression was used to analyze the efficacy of the combined parameters. R software version 4.4.1 (https://www.r-project.org/) was used to construct the nomogram for the combined model. A P value <0.05 indicated a statistically significant difference.

Results

Patient characteristics

Significant differences were found between the Ki-67 high- and low-expression groups in terms of tumor diameter and shape (P<0.05). However, no statistical differences were found between the two groups in terms of age, gender, cirrhosis, alpha-fetoprotein, capsule, and necrosis (P>0.05) (Table 1).

Consistency test

The ICCs between the two radiologists were all >0.75 (0.858–0.937; Table 2). A Bland-Altman analysis was also conducted to examine the measurements of the spectral parameters. The mean biases between the two observers were all <0.30, and most of the differences were within the 95% limit of agreement (the range of the mean bias ±1.96 standard deviation), indicating that the measurements of the two radiologists showed good consistency (Figure 4).

Figure 4 Bland-Altman plots showing good agreement between the two radiologists for measuring the DECT parameters. HU_{130 keV} and HU_{140 keV}, CT values in 130 and 140 keV monochromatic energy images, respectively. HU, Hounsfield units; PVP, portal venous phase; SD, standard deviations; D_water, water density; NeffZ, normalized effective atomic number; AP, arterial phase; D_fat, fat density; DECT, dual-energy computed tomography; CT, computed tomography.

Correlation between the DECT parameters and Ki-67 expression

The NeffZ-AP, HU_{130 keV}-PVP, HU_{140 keV}-PVP, D_fat-PVP, and D_water-PVP were positively correlated with the Ki-67% (all P<0.05, r=0.283–0.412). Among these, D_water-PVP showed the strongest correlation. The correlations between NeffZ-PVP, HU_{130 keV}-AP, HU_{140 keV}-AP, D_fat-AP, D_water-AP, and Ki-67 expression were not statistically significant (Table 3).

Comparison of the DECT parameters between the groups

The NeffZ-AP, HU_{130 keV}-PVP, HU_{140 keV}-PVP, D_fat-PVP, and D_water-PVP values of the Ki-67 high-expression group in HCC were significantly higher than those of the low-expression group (P<0.05). There were no statistical differences between the two groups in terms of the NeffZ-PVP, HU_{130 keV}-AP, HU_{140 keV}-AP, D_fat-AP, and D_water-AP values (P>0.05) (Table 4, Figure 5).

Figure 5 Comparison of the DECT parameters between the high and low Ki-67 expression groups in HCC. HU_{130 keV} and HU_{140 keV}, CT values in 130 and 140 keV monochromatic energy images, respectively. **, P<0.01; ***, P<0.001. HU, Hounsfield units; PVP, portal venous phase; NeffZ, normalized effective atomic number; AP, arterial phase; D_water, water density; D_fat, fat density; DECT, dual-energy computed tomography; HCC, hepatocellular carcinoma; CT, computed tomography.

Multivariate logistic regression model analysis and diagnostic efficacy of the DECT parameters

The multivariate logistic regression analysis showed that D_water-PVP [odds ratio (OR) =1.353, 95% confidence interval (CI): 1.204–1.521, P<0.001] and tumor diameter (OR =1.258, 95% CI: 1.08–1.465, P=0.003) were independent predictors of high Ki-67 expression. The analysis results are shown in Figure 6.

Figure 6 Forest plots showing that tumor diameter and D_water-PVP are independent predictors of high Ki-67 expression. D_water, water density; PVP, portal venous phase; OR, odds ratio; CI, confidence interval.

The area under the curve (AUC), threshold, sensitivity, and specificity of each parameter for differentiating between high and low Ki-67 expression in HCC are shown in Table 5 and Figure 7. Among the parameters, D_water-PVP had the highest AUC (0.810). The combined model of tumor diameter and D_water-PVP had an AUC of 0.846. After combining the DECT parameters and morphological characteristics, the AUC increased to 0.857, the sensitivity was 86.5%, and the specificity was 69.4%. The equation of the combined model was expressed as follows. Logit (Probability) = 0.318 × D_water − PVP + 0.198 × Diameter + 5.530 × Shape − 334.397. A nomogram was also constructed (Figure 8).

Figure 7 Receiver operating characteristic curves of parameters in predicting Ki-67 expression (threshold: 20%); HU_{130 keV} and HU_{140 keV}, CT values in 130 and 140 keV monochromatic energy images, respectively. NeffZ, normalized effective atomic number; AP, arterial phase; D_fat, fat density; PVP, portal venous phase; D_water, water density; Multivariables, D_water-PVP + diameter + shape.

Figure 8 The nomogram for the combined model. In relation to shape, “0” indicates that the tumor shape is round; and “1” indicates that the tumor shape is irregular. D_water, water density; PVP, portal venous phase.

Discussion

In this study, we found that the quantitative parameters NeffZ-AP, HU_{130 keV}-PVP, HU_{140 keV}-PVP, D_fat-PVP, and D_water-PVP based on DECT spectral scanning were correlated with the Ki-67% of HCC. The CT parameters had good efficacy in the preoperative prediction of Ki-67 expression. Moreover, the multivariable analysis combining the spectral parameters and morphological characteristics improved the predictive efficacy of the model in predicting high Ki-67 expression.

Cyclic nuclear protein Ki-67 is present in all active phases of the cell cycle (Gap phase 1, Synthesis phase, Gap phase 2 and Mitosis phase), and is considered a prognostic factor in patients with HCC (9). High Ki-67 expression in HCC is related to a more aggressive tumor phenotype and a worse prognosis. In addition, it frequently leads to more aggressive surgical interventions for resectable tumors and more intensive systemic treatments for unresectable HCC (10). Further, patients with high Ki-67 expression typically require more frequent monitoring and follow-up to detect any early signs of recurrence or disease progression (10). The analysis of Ki-67 expression is very important in HCC evaluation, but the currently used immunochemistry method is not suitable for patients who have lost access to surgery such as those with advanced tumors or those with other serious organic diseases.

Previous studies reported that the parameters obtained from DECT imaging, such as the slope of the spectral curve, normalized iodine concentration, and CT values on monochromatic images, were associated with the Ki-67% in cervical cancer, non-small cell lung cancer, and lung adenocarcinoma, respectively (11-13). Our study found a similar correlation in HCC, which had not been considered in previous studies; however, the statistically significant parameters (i.e., NeffZ-AP, HU_{130 keV}-PVP, HU_{140 keV}-PVP, D_fat-PVP, and D_water-PVP) identified in this study, differed to those identified in other studies, but this may be related to differences among the organs.

EffZ represents the atomic number of the equivalent compound for a mixture with complex components, and can be used to identify different tissues and reflect their structural and compositional characteristics (14). We found that NeffZ-AP was positively correlated with the Ki-67% in HCC. This might be due to the fact that malignant tumors with high Ki-67 expression display active cell proliferation and abnormal angiogenesis (15). Ultimately, tumors with high Ki-67 expression had higher NeffZ-AP values than tumors with low Ki-67 expression.

Our study showed that HU_{130 keV}-PVP and HU_{140 keV}-PVP were positively correlated with the Ki-67%, and the Ki-67 high-expression group had higher monochromatic CT values than the Ki-67 low-expression group. Tumor cells with higher proliferation activity need more nutrition to generate more energy; thus, a rich blood supply and marked contrast enhancement are often observed in progressive tumors (16). Additionally, a high-keV monochromatic CT value had good diagnostic power in identifying high Ki-67 expression. CT attenuation obtained from conventional CT with mixed energies can cause CT value drift. Conversely, DECT provides accurate CT value measurements. In low-keV monochromatic energy images, the lesions contrast sharply to surrounding tissues but have fairly high noise (17). Conversely, in high-keV monochromatic energy images, the lesions have acceptable contrast and little noise. Thus, it is reasonable to make full use of high-keV CT values in differentiating between high and low Ki-67 expression in HCC.

In this study, D_fat was higher in high Ki-67 expressed HCC. The lipid synthesis of tumor cells is closely related to the extracellular microenvironment. The actively proliferated HCC cells facilitate the glucose uptake through the “Warburg effect” to meet the nutrition and energy requirements for rapid tumor growth due to hypoxia and glucose deficiency. The “Warburg effect” also provides sufficient substrates for lipid synthesis. As a result, more triglycerides are synthesized and stored in intracellular lipid droplets (18). D_fat measurement through spectral material decomposition is sensitive to lipid deposition and can reflect lipid distribution in different lesions (19). We inferred that the D_fat measured by DECT indirectly reflects the degree of the “Warburg effect” in HCCs. Tumor cell proliferation, tissue hypoxia, and glucose deficiency were aggravated with up-regulated Ki-67 expression in HCC. Thus, the expression of lipid synthesis-related enzymes was also up-regulated (20), resulting in higher D_fat-PVP in the Ki-67 high-expression group than the Ki-67 low-expression group.

In this study, the HCC patients with high Ki-67 expression showed high D_water-PVP. As previous studies have found, the D_water values derived from the material decomposition images may reflect the physical density of the lesion (21,22). Further, high D_water is associated with active tumor cell proliferation and invasiveness (23). The heightened proliferation and invasive potential of HCC with high Ki-67 expression could result in a higher D_water value than that of HCC with low Ki-67 expression. Further research needs to be conducted to validate the results and examine the mechanism underlying the increased D_water in HCC with high Ki-67 expression.

The multivariate logistic regression analysis showed that the tumor diameter and D_water-PVP were independent predictors of high Ki-67 expression, and the risk of high Ki-67 expression increased as the diameter increased and the D_water-PVP increased. After combining multiple spectral parameters and morphological characteristics, the AUC of the preoperative prediction model reached 0.857, which was higher than the diagnostic efficiency of any single parameter. Thus, the multivariable prediction model could have high clinical application value.

This study had some limitations. First, the small sample size might have led to selection bias. Second, to date, no consensus has been reached as to the critical value of the Ki-67% in HCC (e.g., 10% or 50%) (24,25). Third, we strived to achieve image-pathology matching for the sample analysis; however, perfect correspondence between the pathological samples and DECT ROIs was difficult to achieve, and the Ki-67 counts were also limited by the field of view. Fourth, the present study primarily focused on spectral parameters derived from the AP and PVP. To comprehensively understand the phenomenon, future research should incorporate non-contrast-enhanced CT and the equilibrium phase of enhanced CT data to explore contrast-free parameters, tumor washout rates, and other parameters in HCC Ki-67 expression. Finally, the number of potential covariates in this study were relatively small, but more confounding variables should be considered to improve the predictive efficacy of Ki-67 in further studies.

Conclusions

The DECT parameters (i.e., NeffZ-AP, HU_{130 keV}-PVP, HU_{140 keV}-PVP, D_fat-PVP, and D_water-PVP) were significantly correlated with the Ki-67% in HCC. DECT may provide a new non-invasive method for evaluating the proliferation of HCC cells. The multivariable analysis combining the DECT parameters and morphological characteristics improved the predictive efficacy of the model in predicting high Ki-67 expression and could be used as a tool to assist clinicians to make clinical decisions and evaluate patient prognosis.

Acknowledgments

Funding: This study was funded by the Science and Technology project of Xiamen (grant/award No. 3502Z20189037).

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-24-461/rc

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-24-461/coif). S.C. is employed by GE HealthCare, the manufacturer of the CT system used in this study. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the Ethics Committee of Zhongshan Hospital Affiliated to Xiamen University (approval No. 2022-251), and the Xiamen Hospital of Traditional Chinese Medicine also approved the study. The requirement of individual consent for this retrospective analysis was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Wang Y, Gao B, Xia C, Peng X, Liu H, Wu S. Development of a novel tumor microenvironment-related radiogenomics model for prognosis prediction in hepatocellular carcinoma. Quant Imaging Med Surg 2023;13:5803-14. [Crossref] [PubMed]
2. Dy A, Nguyen NJ, Meyer J, Dawe M, Shi W, Androutsos D, Fyles A, Liu FF, Done S, Khademi A. AI improves accuracy, agreement and efficiency of pathologists for Ki67 assessments in breast cancer. Sci Rep 2024;14:1283. [Crossref] [PubMed]
3. Li M, Li Z, Wei L, Li L, Wang M, He S, Peng Z, Feng ST. Harnessing dual-energy CT for glycogen quantification: a phantom analysis. Quant Imaging Med Surg 2023;13:4933-42. [Crossref] [PubMed]
4. Mao LT, Chen WC, Lu JY, Zhang HL, Ye YS, Zhang Y, Liu B, Deng WW, Liu X. Quantitative parameters in novel spectral computed tomography: Assessment of Ki-67 expression in patients with gastric adenocarcinoma. World J Gastroenterol 2023;29:1602-13. [Crossref] [PubMed]
5. Wen Y, Song Z, Li Q, Zhang D, Li X, Yu J, Li Z, Ren X, Zhang J, Liu Q, Huang J, Zeng D, Tang Z. Development and validation of a model for predicting the expression of Ki-67 in pancreatic ductal adenocarcinoma with radiological features and dual-energy computed tomography quantitative parameters. Insights Imaging 2024;15:41. [Crossref] [PubMed]
6. Dong Y, Jiang Z, Li C, Dong S, Zhang S, Lv Y, Sun F, Liu S. Development and validation of novel radiomics-based nomograms for the prediction of EGFR mutations and Ki-67 proliferation index in non-small cell lung cancer. Quant Imaging Med Surg 2022;12:2658-71. [Crossref] [PubMed]
7. Murakami K, Kasajima A, Kawagishi N, Ohuchi N, Sasano H. Microvessel density in hepatocellular carcinoma: Prognostic significance and review of the previous published work. Hepatol Res 2015;45:1185-94. [Crossref] [PubMed]
8. Wang P, Tang Z, Xiao Z, Wu L, Hong R, Duan F, Wang Y, Zhan Y. Dual-energy CT in predicting Ki-67 expression in laryngeal squamous cell carcinoma. Eur J Radiol 2021;140:109774. [Crossref] [PubMed]
9. Fan Y, Yu Y, Wang X, Hu M, Hu C. Radiomic analysis of Gd-EOB-DTPA-enhanced MRI predicts Ki-67 expression in hepatocellular carcinoma. BMC Med Imaging 2021;21:100. [Crossref] [PubMed]
10. Zhao YM, Xie SS, Wang J, Zhang YM, Li WC, Ye ZX, Shen W. Added value of CE-CT radiomics to predict high Ki-67 expression in hepatocellular carcinoma. BMC Med Imaging 2023;23:138. [Crossref] [PubMed]
11. Pan L, Jia X, Zhao X, Zhang B, Wang S, Fan T, Zhou M, Yuan Y, Wang G, Xue L. Study on the correlation between energy spectrum computed tomography imaging and the pathological characteristics and prognosis of cervical cancer. Transl Cancer Res 2021;10:4096-105. [Crossref] [PubMed]
12. Tian S, Jianguo X, Tian W, Li Y, Hu J, Wang M, Zhang J. Application of dual-energy computed tomography in preoperative evaluation of Ki-67 expression levels in solid non-small cell lung cancer. Medicine (Baltimore) 2022;101:e29444. [Crossref] [PubMed]
13. Cao X, Hu HG, Shen M, Deng K, Wu N. Correlation Between Quantitative Spectral CT Parameters and Ki-67 Expression in Lung Adenocarcinomas Manifesting as Ground-glass Nodules. Curr Med Imaging 2023;19:1052-62. [PubMed]
14. Wang X, Liu D, Zeng X, Jiang S, Li L, Yu T, Zhang J. Dual-energy CT quantitative parameters for evaluating Immunohistochemical biomarkers of invasive breast cancer. Cancer Imaging 2021;21:4. [Crossref] [PubMed]
15. Konietzke P, Steentoft HH, Wagner WL, Albers J, Dullin C, Skornitzke S, Stiller W, Weber TF, Kauczor HU, Wielpütz MO. Consolidated lung on contrast-enhanced chest CT: the use of spectral-detector computed tomography parameters in differentiating atelectasis and pneumonia. Heliyon 2021;7:e07066. [Crossref] [PubMed]
16. Chen J, Tang L, Xie P, Qian T, Huang J, Liu K. Quantitative parameters of dual-layer spectral detector computed tomography for evaluating Ki-67 and human epidermal growth factor receptor 2 expression in colorectal adenocarcinoma. Quant Imaging Med Surg 2024;14:789-99. [Crossref] [PubMed]
17. Park J, Choi YH, Cheon JE, Kim WS, Kim IO, Pak SY, Krauss B. Advanced virtual monochromatic reconstruction of dual-energy unenhanced brain computed tomography in children: comparison of image quality against standard mono-energetic images and conventional polychromatic computed tomography. Pediatr Radiol 2017;47:1648-58. [Crossref] [PubMed]
18. Jaworska M, Szczudło J, Pietrzyk A, Shah J, Trojan SE, Ostrowska B, Kocemba-Pilarczyk KA. The Warburg effect: a score for many instruments in the concert of cancer and cancer niche cells. Pharmacol Rep 2023;75:876-90. [Crossref] [PubMed]
19. Winkelmann MT, Gassenmaier S, Walter SS, Artzner C, Lades F, Faby S, Nikolaou K, Bongers MN. Differentiation of adrenal adenomas from adrenal metastases in single-phased staging dual-energy CT and radiomics. Diagn Interv Radiol 2022;28:208-16. [Crossref] [PubMed]
20. Yamashita T, Honda M, Takatori H, Nishino R, Minato H, Takamura H, Ohta T, Kaneko S. Activation of lipogenic pathway correlates with cell proliferation and poor prognosis in hepatocellular carcinoma. J Hepatol 2009;50:100-10. [Crossref] [PubMed]
21. Thieme SF, Johnson TR, Lee C, McWilliams J, Becker CR, Reiser MF, Nikolaou K. Dual-energy CT for the assessment of contrast material distribution in the pulmonary parenchyma. AJR Am J Roentgenol 2009;193:144-9. [Crossref] [PubMed]
22. Yu Y, Cheng JJ, Li JY, Zhang Y, Lin LY, Zhang F, Xu JR, Zhao XJ, Wu HW. Determining the invasiveness of pure ground-glass nodules using dual-energy spectral computed tomography. Transl Lung Cancer Res 2020;9:484-95. [Crossref] [PubMed]
23. Liu G, Li M, Li G, Li Z, Liu A, Pu R, Cao H, Liu Y. Assessing the Blood Supply Status of the Focal Ground-Glass Opacity in Lungs Using Spectral Computed Tomography. Korean J Radiol 2018;19:130-8. [Crossref] [PubMed]
24. Wu H, Han X, Wang Z, Mo L, Liu W, Guo Y, Wei X, Jiang X. Prediction of the Ki-67 marker index in hepatocellular carcinoma based on CT radiomics features. Phys Med Biol 2020;65:235048. [Crossref] [PubMed]
25. Yang X, Ni H, Lu Z, Zhang J, Zhang Q, Ning S, Qi L, Xiang B. Mesenchymal circulating tumor cells and Ki67: their mutual correlation and prognostic implications in hepatocellular carcinoma. BMC Cancer 2023;23:10. [Crossref] [PubMed]