Application value of magnetic resonance imaging targetoid enhancement combined with alpha-fetoprotein level in evaluating the expression of cytokeratin 19 in hepatocellular carcinoma
Gao Zhiling,Shu Jianjun,Chen Yong,Sun Changhao,Yang Guanhua
DOI: https://doi.org/10.3760/cma.j.cn115610-20220515-00276
2022-01-01
Abstract:Objective:To investigate the application value of preoperative clinical and magnetic resonance imaging (MRI) targetoid enhancement combined with alpha-fetoprotein in evaluating the expression of cytokeratin 19 (CK19) in patients undergoing radical resection for single hepatocellular carcinoma (HCC) without macrovascular invasion.Methods:The retrospective cohort study was conducted. The clinicopathological data of 220 patients who underwent radical resection for single HCC without macrovascular invasion in the General Hospital of Ningxia Medical University from January 2016 to December 2020 were collected. There were 171 males and 49 females, aged (56±11)years. Of the 220 patients, 52 cases showed positive CK19 expression, while 168 cases showed negative CK19 expression. Observation indicators: (1) MRI and immunohistochemical staining results of patients with different status of CK19 expression; (2) comparison of clinical features of patients with different status of CK19 expression; (3) comparison of MRI features in patients with different status of CK19 expression; (4) analysis of influencing factors for CK19 expression in patients and predictive value. The normality of continuous variables was tested using the Shapiro-Wilk test. Measurement data with normal distribution were expressed as Mean± SD, and comparison between groups was conducted using the t test. Measurement data with skewed distribution were expressed as M( Q1, Q3), and comparison between groups was conducted using the Mann-Whitney U test. Count data were expressed as absolute numbers and (or) percentages, and comparison between groups was conducted using the chi-square test. Comparison of ordinal data was analyzed by the non-parameter rank sum test. The relevant clinical and imaging features with P<0.05 were included in the binary Logistic regression model. The receiver operating characteristic (ROC) curve was drawn and the area under curve (AUC) was used to evaluate the predictive efficiency of the model. Results:(1) MRI and immunohistochemical staining results of patients with different status of CK19 expre-ssion. Results of MRI examination in patients with positive CK19 expression showed the tumors with low-signal intensity on plain T1-weighted imaging, annular high enhancement in the arterial phase, clear boundaries in the portal venous phase, central enhance-ment in the delayed phase and targetoid high signals on diffusion-weighted imaging (DWI). Immuno-histochemical staining revealed a positive CK19 expression. Results of MRI examination in patients with negative CK19 expression showed the tumors with low-signal intensity on plain T1-weighted imaging, non-annular high enhancement in the arterial phase, unclear boundaries in the portal venous phase, low signals compared with peripheral liver tissue in the delayed phase and uniform high signals on DWI. Immunohistochemical staining revealed a negative CK19 expression. (2) Clinical features of patients with different status of CK19 expression. The neutrophil count and cases with alpha-fetoprotein (AFP) ≥400 μg/L were 3.07(2.21,4.41)×10 9/L and 26 in patients with positive CK19 expression, versus 2.72(2.05, 3.51)×10 9/L and 48 in patients with negative CK19 expression, showing significant differences between them ( Z=?2.06, χ2=8.17, P<0.05). (3) Compari-son of MRI features in patients with different status of CK19 expression. Cases with tumor diameter ≥ 5 cm and cases with tumor showing targetoid enhancement were 34 and 22 in patients with positive CK19 expression, versus 82 and 24 in patients with negative CK19 expression, showing significant differences between them ( χ2=4.38, 18.86, P<0.05). (4) Analysis of influencing factors for CK19 expression in patients and predictive value. Results of multivariate analysis showed that AFP ≥ 400 μg/L and targetoid enhance-ment were independent risk factors for positive CK19 expression in HCC patients [ odds ratio=2.09, 3.23, 95% confidence interval ( CI) as 1.06?4.13, 1.49?6.99, P<0.05]. Results of ROC curve analysis showed that the AUC of targetoid enhancement for predicting positive CK19 expression was 0.64 (95% CI as 0.57?0.71), with the sensitivity and specificity as 42.31% and 85.71%. The AUC of AFP ≥400 μg/L for predicting positive CK19 expression was 0.61 (95% CI as 0.53?0.68), with the sensitivity and specificity as 51.00% and 71.43%. The AUC of targetoid enhancement combined with AFP ≥400 μg/L for predicting positive CK19 expression was 0.69 (95% CI as 0.61?0.77), with the sensitivity and specificity as 67.31% and 63.10%, respectively. Conclusions:Targetoid enhancement and AFP ≥400 μg/L are independent risk factors for positive CK19 expression in patients with single HCC without macrovascular invasion. Their combination has clinical value for preoperative evaluation of CK19 expression.