Correlations between Diffusion-Weighted and Intravoxel Incoherent Motion Parameters with Ki-67 Index in Gastric Adenocarcinoma

doi:10.21203/rs.3.rs-1413456/v1

PDF

Research Article

Correlations between Diffusion-Weighted and Intravoxel Incoherent Motion Parameters with Ki-67 Index in Gastric Adenocarcinoma

https://doi.org/10.21203/rs.3.rs-1413456/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background：Intravoxel incoherent motion imaging (IVIM) and DWI have been applied for gastric adenocarcinoma.

Purpose: To investigate correlation between parameters of IVIM and DWI with Ki-67 in gastric adenocarcinoma.

Materials and methods: A total of 64 gastric adenocarcinoma patients were enrolled into the research and received DWI and IVIM at 3.0T MRI before the surgery. Based on the setting of ROIs, the DWI and IVIM parameters (including ADC, ture diffusion coefficient (D), perfusion fraction (f) and pseudo-diffusion coefficient (D^*)) of each patient's gastric adenocarcinoma were measured and calculated. According to expression of Ki-67, participated subjects were divided into two groups (low Ki-67 group and high Ki-67 group). The intraclass correlation coeffificient (ICC) was used to assess the consistency of each parameter measured by two radiologists. The independent-sample t-test was used to compare differences of each parameter between groups. Spearman correlation coefficient was calculated to reveal the correlation between Ki-67 with parameters. By using the area under the curve (AUC) of the receiver operating characteristic (ROC) curve, each parameter was compared.

Results: The ADC, D and f values of high Ki-67 group were all lower than those of low Ki-67 group [(10.89±0.89) × 10⁻² mm²/s vs. (14.65±0.46)× 10⁻² mm²/s, (9.95±0.92) × 10⁻² mm²/s vs. (12.48±0.83) × 10⁻² mm²/s, (2.30±0.51) × 10⁻² mm²/s vs. (3.75±0.85) × 10⁻² mm²/s, all P < 0.05]. Except for D^*, ADC, D and f were negatively correlated with the expression of Ki-67 (r = −0.725, −0.733, -0.666, all P < 0.05). The areas under the curve (AUC) of ADC, D and f were 0.901, 0.923 and 0.657 respectively (Table 5). The difference of AUC value between ADC and D was not statistically significant (AUC_D= 0.923 vs. AUC_ADC= 0.901, Z=0.290, P=0.772). The difference of AUC value between ADC and f was statistically significant (AUC_ADC= 0.901 vs. AUC_f= 0.657, Z=2.329, P=0.025). The difference of AUC value between D and f was statistically significant (AUC_D= 0.923 vs. AUC_f= 0.657, Z=2.124, P=0.034) (Table 6).

Conclusions: DWI and IVIM can effectively reflect the proliferative characteristics of gastric adenocarcinoma. D and ADC have higher diagnostic efficiency than f.

gastric adenocarcinoma

DWI

intravoexl incoherent motion imaging

Ki-67 index.

Gastric cancer has become the fourth most common cancer in the world and the second largest cause of death in Asia. At present, the new annual cases of gastric cancer in China account for more than 40% of the world's new cases. Gastric adenocarcinoma is one of the common pathological types of gastric cancer. MRI has been playing an essential role in preoperative evaluation, development of treatment plans and assessment of effects and prediction in gastric adenocarcinoma (1–3). Intravoxel incoherent motion (IVIM) has been using dual index spread model to analyze the spread of the water element under the condition of non-Gaussian distribution, which means that under the circumstance of low b value, the perfusion of microvessels has a greater impact on DWI and based on the double exponential model, IVIM can effectively eliminate the effect of micro perfusion through a series of low b values, it can more accurately and quantitatively evaluate the micro movement inside and outside cells (4–6). As an important tumor symbol, the expression of Ki67 is closely related to the growth and transfer of gastric tumor (7–8). And, high Ki-67 in early gastric cancer might suggest a poor prognosis (9–10). Previous studies of MRI mainly focus on malignant tumor such as breast cancer and lung cancer. Scholars have found that there was a negative relationship between ADC and Ki-67 index (11–12). Therefore, the ADC has the potential to reflect the growth of the tumor cell to a certain extent while it is limited as an alternative marker to reflect the proliferation potential (13). At present, there are few reports on the correlation between IVIM and DWI with Ki-67 in gastric adenocarcinoma and this study aims to analyze correlation between IVIM and DWI parameters with Ki-67 in gastric adenocarcinoma and explore possibility of its response to proliferative activity.

Patient data

From December 2020 to August 2021, a total of 76 patients with gastric adenocarcinoma were selected for breast MR imaging and of these, 12 patients were excluded for the following reasons: 1) radiotherapy, chemotherapy, or surgery before scanning (5 patients); 2) failure to obtain clear immunohistochemistry and pathological results after scanning (3 patients); and 3) showed poor lesion visibility on DWI and IVIM parameter maps for analysis (4 patients). At last, a total of 64 hospitalized patients with gastric adenocarcinoma in Shandong provincial hospital were selected (including 30 males and 34 females).

Data acquisition

A 3.0T MRI scanner (Siemens, Germany, 3T, Prisma) with a body phased-array coil were used in the study. The patients were supine with the head in front and were fasted for 6–8 hours and received regular breathing and breath holding training before the examination.And patients drank 600–1000 ml water to ensure the filling of the stomach. The scanning range included the whole stomach. MRI scanning sequence includes T1WI, T2WI, DWI and IVIM sequence. (T1WI, T2WI, DWI and IVIM scanning parameters are shown in Table 1).

Image analysis

All images were uploaded to a Simens post-processing workstation (Syngo. Via Workstation, Simens Healthineers) and the IVIM and DWI images were processed and analyzed using relevant software (Body diffusion imaging 1.3.1). Identifification of cancers and manual drawing of the region of interest (ROI) on the software were independently performed by two gastrointestinal radiologists (P.Z and S.D with 10 and 8 years of experience, respectively) blinded to the clinical-pathologic features. First, the ROI was delineated by outlining the solid part of the tumor tissue on the axial DWI image using the T1WI and T2WI sequence as references and then merging the images of the IVIM and DWI parameters with the DWI images of the same layer(Fig. 1). The ROI included the solid region of the tumor, while regions with cystic changes, large blood vessels, and bleeding were avoided. Using the transferred ROIs, the mean values for IVIM and DWI parameter maps were calculated. The DWI and IVIM parameters were calculated using the Eq. 1 and Eq. 2 respectively :

Eq1: ADC = (lnS0/S1) /(b1-b0)

S0 and S1 stand for the signal intensity obtained using the b0 and b1 values. ADC is the diffusion coefficient obtained under the influence of tissue signal attenuation, the diffusion attenuation of water molecules, the uniformity of diffusion gradient field and blood perfusion.

Eq2: S / S₀ = f_IVIM exp [-b (D^*+D_blood)] + (1- f_IVIM) exp (-b×D)

S / S₀ is the sum of tissue and blood components, fivim is the flowing blood fraction, D is the water diffusion coefficient in tissue, D_blood is the water diffusion coefficient in blood and D^* represents the false diffusion coefficient related to microcirculation perfusion factors such as capillary structure in tissue. Among them, f_IVIM and D^* can effectively reflect micro perfusion in tissue. The final value of each parameter for each lesion was calculated using the following formula: the final value = the sum of the parameter values of lesions at different slices / the number of slices.

Pathological analysis

Immunohistochemical sections were diagnosed by two doctors who had over 10 years of pathological diagnosis experience. Ki-67 index refers to the percentage of Ki-67 positive cells in visual field. According to previous studies, we know that high Ki-67 (Ki-67 ≥ 55%) suggests a poor prognosis and higher mortality of gastric adenocarcinoma (14–16) therefore patients were divided into two groups: the low expression group of Ki-67 with Ki-67 index < 55%, and Ki-67 high expression group with Ki-67 index ≥ 55%. And 30 cases were set in low Ki-67 group and 34 cases in high Ki-67 group.

Statistical analysis

Statistical software SPSS 23.0 and Medcalc 19.0.4 were used for data analysis. Firstly, background information, DWI and DKI parameters of patients with Gastric adenocarcinoma were tested for normality and the data were expressed as mean ± standard deviation. Meanwhile, The intraclass correlation coeffificient (ICC) was used to assess the consistency of each parameter measured by two radiologists, and interpreted as follows: 0.00–0.20, poor agreement; 0.21–0.40, fair agreement; 0.41–0.60, moderate agreement; 0.61–0.80, good agreement; and 0.81–1.00, excellent agreement. The independent-sample t-test was used to compare differences of each parameter between groups. Spearman correlation analysis was used to analyze the correlation between each parameter with Ki-67 index. And the ROC curve was analyzed and characteristic values including AUC value, threshold, sensitivity and specificity were obtained. The diagnostic efficiency of different parameters was compared. The above-mentioned analysis was statistically significant on the premise of P < 0.05.

Interclass correlation coefficient

The data measured by the two observers had good intraobserver and interobserver reproducibility (Table 2) (P < 0.05).

Differentiation of low Ki-67 group and high Ki-67 group

As is shown in Table 3, there were no significant difference in height, weight, age and male female ratio between two groups (P > 0.05). There were no significant difference between the two groups in the degree of pathological differentiation, pathological type and Lauren classification (P > 0.05). Gastric adenocarcinoma with higher Ki-67 tends to have a larger mass than the other group (P < 0.05). This is similar to the results of previous research (8, 16).

As is shown in Table 4, except for D^*, there were significant differences in ADC, D and f between the two groups (P < 0.05). The ADC, D and f values of high Ki-67 group were all lower than those of low Ki-67 group [(10.89 ± 0.89) × 10^− 2 mm²/s vs. (14.65 ± 0.46)× 10^− 2 mm²/s, (9.95 ± 0.92) × 10^− 2 mm²/s vs. (12.48 ± 0.83) × 10^− 2 mm²/s, (2.30 ± 0.51) × 10^− 2 mm²/s vs. (3.75 ± 0.85) × 10^− 2 mm²/s, all P < 0.05].

Correlation Analysis

Except for D^*, ADC, D and f were negatively correlated with the expression of Ki-67 (r = − 0.725, − 0.733, -0.666, all P < 0.05).

Comparison of Diagnostic Effificacy

As a result of the difference of D^* between groups was not statistically significant(P > 0.05) (Table 4), ROC curve analysis was not performed for D^*. The areas under the curve (AUC) of ADC, D and f were 0.901, 0.923 and 0.657 respectively (Fig. 2, Table 5). The difference of AUC value between ADC and D was not statistically significant (AUC_D = 0.923 vs. AUC_ADC= 0.901, Z = 0.290, P = 0.772). The difference of AUC value between ADC and f was statistically significant (AUC_ADC= 0.901 vs. AUC_f= 0.657, Z = 2.329, P = 0.025). The difference of AUC value between D and f was statistically significant (AUC_D= 0.923 vs. AUC_f= 0.657, Z = 2.124, P = 0.034)(Table 6).

As an important means to effectively reflect diffusion of water molecules inside and outside cells, DWI has been widely used in the differential diagnosis of benign from malignant gastrointestinal tumors, preoperative evaluation and evaluation of the effect of radiotherapy and chemotherapy (17–18). As a tumor marker with wide clinical application value, higher Ki-67 index often indicates the characteristics of active proliferation, strong infiltration and poor prognosis of tumor cells (7–8). Studies have suggested that with increase of Ki-67 level, the ADC value of tumor continues to decline (19). However, because ADC is limited by the diffusion movement of water molecules that affected by the microstructure inside and outside the cell and the influence of capillary micro perfusion, ADC cannot completely truly reflect the diffusion of water molecules inside and outside the cell. The multi b-value IVIM model proposed by Le Bihan et al can obtain the D value that reflects the real water molecule diffusion and can effectively eliminate the influence of micro perfusion.And, D^* and f values can also be obtained (20) to reflect the patients’ blood perfusion. At present, the correlation between IVIM and Ki-67 index has been applied to malignant tumors such as extrahepatic cholangiocarcinoma, lung cancer, nasopharyngeal carcinoma and rectal cancer. The results show that D has better correlation with Ki-67 than that of ADC. This study found that the results are similar to above tumors and the difference of AUC value between D and ADC value was not statistically significant (Table 6). And in terms of sensitivity and specificity, the results of D is higher than that of ADC. The possible reason is that IVIM effectively eliminates the influence of capillary micro perfusion on diffusion coefficient compared with single index DWI. Based on the obvious negative correlation between ADC and D value with Ki-67, the increase of Ki-67 expression may suggest a stronger tumor cell proliferation activity, tighter cell arrangement, smaller extracellular space, higher nucleocytoplasmic ratio and more organelles, which significantly reduces the intracellular space, shows more significant restriction on the free diffusion of water molecules and therefore shows lower ADC and D (21–24).

D^* and f values mainly reflect micro perfusion of new capillaries in tumor cell space, which is related to cell density, cell necrosis and bleeding. The study found that there was a negative correlation between f and Ki-67 in gastric adenocarcinoma and no significant correlation was between D^* and Ki-67. By contradicting the related studies of lung cancer and nasopharyngeal carcinoma, the results of the study is consistent with the similar research in extrahepatic cholangiocarcinoma and rectal cancer. In the study on the correlation between lung cancer and nasopharyngeal carcinoma and Ki-67, there was no correlation between D^* and f with the expression of Ki-67, which may be out of the histopathological differences. Histopathological differences of nasopharyngeal carcinoma is that prone to bleeding and necrosis due to their strong proliferative ability and there is obvious heterogeneity in the proportion of capillaries in the extracellular space due to the uncertainty of necrosis and bleeding. This might be one of the reasons for the above results in gastric adenocarcinoma. And, because D^* and f values mainly reflect extracellular micro perfusion, there is no direct correlation with cell density, which may explain why the correlation between f with Ki-67 expression is not as good as the relationship between D and Ki-67 (25). At present, although adenocarcinoma is the main type of lung cancer, current studies are mainly on the differences between non-small cell lung cancer and small cell lung cancer. Due to many different pathological types of non-small cell lung cancer, there are great differences in vascular proliferation in intercellular space and the differences in degeneration and necrosis of lung cancer with different pathological types and degrees of differentiation. Therefore, no correlation between D^* and f with the expression of Ki-67 in lung cancer (21) can be well explained. As a result of gastric adenocarcinoma, cholangiocarcinoma and rectal cancer are mainly adenocarcinoma, although the tissue structure arrangement is more loose than nasopharyngeal carcinoma, it will further squeeze the space of extracellular capillaries because it secretes more mucin, causing the proportion of microvascular perfusion in the intercellular space is not high; the cell density increases as the increase of Ki-67, resulting in the narrowing of microvascular space that accompanied by a certain degree of necrosis and bleeding, which may be the reason for the negative correlation between f and Ki-67 (26–28). In gastric adenocarcinoma, there is no significant correlation between D ^* and Ki-67. The possible reason is that D^* is more significantly related to the diameter of blood vessels, but not directly related to the number of microvessels and cell proliferation. And the poor image quality and low signal-to-noise ratio of D^* should also be considered (Fig. 1D) (29).

However, this study has some limitations. First, there is no final conclusion about the optimal number of b values and the optimal size of b values in IVIM sequence. More accurate results can be obtained by increasing the number of b values. Secondly, the effect of different parameter combinations on diagnostic efficacy was not included in the study. Therefore, further studies should be conducted.

In conclusion, DWI and IVIM can effectively reflect the proliferative characteristics of gastric adenocarcinoma. D and ADC have higher diagnostic efficiency than f. Therefore, IVIM and DWI are of great significance in deciding the treatment plan and judging prognosis of gastric adenocarcinoma.

Intravoxel incoherent motion imaging (IVIM), apparent diffusion coefficient (ADC), ture diffusion coefficient (D), perfusion fraction (f) and pseudo-diffusion coefficient (D^*).

Ethics: The experimental protocol was established, according to the ethical guidelines of Shandong Provincial Hospital and was approved by the Human Ethics Committee of Shandong Provincial Hospital. All subjects signed the informed consent form.

Consent for publication: Not applicable.

Availability of data and materials: All data generated or analysed during this study are included in this published article.

Competing interests: The authors declare that they have no competing interests.

Funding: Not applicable.

Authors' contributions:

Xiangtao Lin and Peng Zhao are responsible for the design of research scheme and the screening of target patients.

Hui Ma, Shuai Duan and Ruiyuan Diao are Responsible for patient scanning and sequence debugging.

Shanshan Sun and Shuo Cai are responsible for the statistics and analysis of data

Letian Yuan is responsible for writing article.

Acknowledgement: Not applicable.

Rahman R, Asombang AW, Ibdah JA, et al. Characteristics of gastric cancer in Asia. World J Gastroenterol 2014, 20: 4483–4490.
Zhang YJ, Yu JC. The role of MRI in the diagnosis and treatment of gastric cancer. Diagn Interv Radiol 2020; 26: 176–182.
Ma L, Xu XW, Zhang M, et al. Dynamic contrast-enhanced MRI of gastric cancer: Correlations of the pharmacokinetic parameters with histological type, Lauren classification, and angiogenesis. Magn Reson Imaging 2017; 37: 27–32.
Zeng Q, Hong YL, Cheng J, et al. Quantitative study of preoperative staging of gastric cancer using intravoxel incoherent motion diffusion-weighted imaging as a potential clinical index. Eur J Radiol 2021; 141: 109627.
Song XL, Kong HK, Jeong GW, et al. Intravoxel incoherent motion diffusion-weighted imaging for monitoring chemotherapeutic efficacy in gastric cancer. World J Gastroenterol 2016; 22: 5520–5531.
Szubert-Franczak AE, Naduk-Ostrowska M, Cieszanowski A, et al. Intravoxel incoherent motion magnetic resonance imaging: basic principles and clinical applications. Pol J Radiol 2020; 85: 624–635.
Li LT, Jiang G, Chen Q, et al. Ki67 is a promising molecular target in the diagnosis of cancer (review). Mol Med Rep 2015; 11: 1566–1572.
Luo, GY, Hu YZ, Zhang ZQ et al. Clinicopathologic significance and prognostic value of Ki-67 expression in patients with gastric cancer: a meta-analysis. Oncotarget, 2017. 8(30): 50273–50283.
Ko GH, Go SI, Lee WS, et al. Prognostic impact of Ki-67 in patients with gastric cancer-the importance of depth of invasion and histologic differentiation. Medicine (Baltimore). 2017, 96(25): 7181.
Xiong DD, Zeng CM, Jiang L, et al. Ki-67/MKI67 as a Predictive Biomarker for Clinical Outcome in Gastric Cancer Patients: an Updated Meta-analysis and Systematic Review involving 53 Studies and 7078 Patients. J Cancer. 2019,10(22):5339–5354.
Huang YN, ZHAO ZH, MAO HJ, et al. Correlation between DCE-MRI quantitative perfusion histogram parameters, apparent diffusion coefficient and Ki-67 in different pathological types of lung cancer. Chin J Medicine 2019; 99:1645–1650.
Aydin H, Guner B, Bostanci IE, et al. Is there any relationship between adc values of diffusion-weighted imaging and the histopathological prognostic factors of invasive ductal carcinoma? Br J Radiol 2018; 91: 20170705.
Surov A, Meyer HJ, Wienke A, et al. Correlation between apparent diffusion coefficient (ADC) and cellularity is different in several tumors: a meta-analysis. Oncotarget 2017; 8: 59492–59499.
Huang G, Chen S, Wang D, et al. High Ki67 Expression has Prognostic Value in Surgically-Resected T3 Gastric Adenocarcinoma. Clin Lab. 2016;62(1–2):141–153.
Kikuyama S, Kubota T, Shimizu K, et al. Ki-67 antigen expression in relation to clinicopathological variables and prognosis in gastric cancer. Oncol Rep. 1998;5(4):867–870.
Giaginis C, Giagini A, Tsourouflis G, et al. MCM-2 and MCM-5 expression in gastric adenocarcinoma: clinical significance and comparison with Ki-67 proliferative marker. Dig Dis Sci. 2011;56(3):777–785.
Liu S, Guan WX, Wang H, et al. Apparent diffusion coefficient value of gastric cancer by diffusion-weighted imaging: correlations with the histological differentiation and Lauren classification. European journal of radiology 2014; 83:2122–2128.
Rao A, Sharma C, Parampalli R, et al. Role of diffusion-weighted MRI in differentiating benign from malignant bone tumors. BJR Open 2019; 1: 20180048.
Wang YT, Bai GJ, Zhang X, et al. Correlation analysis of apparent diffusion coefficient value and P53 and Ki-67 expression in esophageal squamous cell carcinoma. Magn Reson Imaging 2020; 68:183–189.
Le BD, Breton E, Lallemand D, et al. Separation of diffusion and perfusion in intravoxel incoherent motion MR imaging. Radiology 1988; 168: 497–505.
Zheng Y, Huang WJ, Zhang XL, et al. A Noninvasive Assessment of Tumor Proliferation in Lung cancer Patients using Intravoxel Incoherent Motion Magnetic Resonance Imaging. J Cancer 2021; 12: 190–197.
Cui XY, Tang QF, Fang XM, et al. Correlation of apparent diffusion coefficient and intravoxel incoherent motion imaging parameters with Ki-67 expression in extrahepatic cholangiocarcinoma. Magn Reson Imaging 2019; 63: 80–84.
Xiao ZB, Wang R, Qian W, et al. Standard diffusion-weighted, diffusion kurtosis and intravoxel incoherent motion MR imaging of sinonasal malignancies: correlations with Ki-67 proliferation status. Eur Radiol 2018; 28: 2923–2933.
Meyer HJ, Höhn AK, Andric M, et al. Associations between IVIM histogram parameters and histopathology in rectal cancer. Magn Reson Imaging 2021; 77:21–27.
Chen YP, Chan AT, Le QT, et al. Nasopharyngeal carcinoma. Lancet 2019; 394: 64–80.
Smyth EC, Nilsson M, Grabsch HI, et al. Gastric cancer. Lancet 2020; 396: 635–648.
Werner M, Becker KF, Keller G, et al. Gastric adenocarcinoma: pathomorphology and molecular pathology. J Cancer Res Clin Oncol 2001; 127: 207–216.
Krasinskas AM et al. Cholangiocarcinoma. Surg Pathol Clin 2018; 11: 403–429.
Surov A, Meyer HJ, Garnov N, et al. Correlations between intravoxel incoherent motion (IVIM) parameters and histological findings in rectal cancer: preliminary results. Oncotarget 2017; 8: 21974–21983.

Tables 1 to 6 are available in the Supplemental Files section.

No competing interests reported.

Table.docx

PDF

Reviews received at journal
06 May, 2022
Reviewers agreed at journal
27 Apr, 2022
Reviewers invited by journal
26 Mar, 2022
Editor assigned by journal
26 Mar, 2022
Editor invited by journal
09 Mar, 2022
Submission checks completed at journal
09 Mar, 2022
First submitted to journal
02 Mar, 2022

You are reading this latest preprint version

Correlations between Diffusion-Weighted and Intravoxel Incoherent Motion Parameters with Ki-67 Index in Gastric Adenocarcinoma

Status:

Version 1

Abstract

Figures

Background

Materials And Methods

Patient data

Data acquisition

Image analysis

Pathological analysis

Statistical analysis

Results

Interclass correlation coefficient

Differentiation of low Ki-67 group and high Ki-67 group

Correlation Analysis

Comparison of Diagnostic Effificacy

Discussion

Conclusion

Abbreviations

Declarations

References

Tables

Competing Interests

Supplementary Files

Status:

Version 1