Objectives To investigate the potential diagnostic value of MR elastography (MRE)-based stiffness to noninvasively predict the status of Ki67 proliferation index (PI) and Cytokeratin-19 (CK19) in hepatocellular carcinoma (HCC). Methods Two hundred and nineteen patients with histologically proven HCCs (high/low Ki67 PI, N = 134/85; +/−CK19, N = 47/172) who underwent preoperative MRI and MRE examinations were retrospectively enrolled. Imaging features and clinical information were analyzed to develop the clinical-radiologic models. Logistic regression and area under the receiver operating characteristic curve (AUC) analyses were used to evaluate the status of Ki67 PI and CK19, and their model performances. Results Tumor stiffness (TS) (p = 0.004) and TS/liver stiffness (LS) (p < 0.001) were significantly independent predictors of high Ki67 PI and positive CK19, respectively. By incorporating TS or TS/LS value, the AUCs of the clinical-radiological models have been improved for identifying high Ki67 PI from 0.796 to 0.819 (p = 0.113), and positive CK19 from 0.817 to 0.852 (p = 0.026). Patients with predicted high Ki67 PI/positive CK19 based on the combined models had worse recurrence-free survival (RFS) and overall survival (OS) than those with low Ki67 PI/negative CK19 (all p < 0.05), respectively. The combined statuses of predicted Ki67 PI and CK19 stratified patients into three distinct subgroups with significantly different RFS and OS (all p < 0.05). Notably, patients with predicted high Ki67 PI and positive CK19 had the worst prognosis. Conclusions Three-dimensional MRE could be a promising tool for evaluating and stratifying the prognosis of HCC patients by predicting the status of Ki67 PI and CK19 preoperatively. Key Points <Emphasis Type="BoldItalic">Question</Emphasis> Can MRE-based stiffness predict Ki67 PI and Cytokeratin 19 (CK19) in HCC? <Emphasis Type="BoldItalic">Findings</Emphasis> MRE-based stiffness is an independent predictor of high Ki67 PI and CK19-positive and improves the diagnostic performance of clinical-radiological models predicting Ki67 PI and CK19 status. <Emphasis Type="BoldItalic">Clinical relevance</Emphasis> MRE-based stiffness is a noninvasive predictor of high Ki67 PI and CK19-positive, and in combination with clinical-radiological models can be used for prognostic analysis, which could refine patients’ selection for adjuvant therapy or surveillance.

Three-dimensional MR elastography-based stiffness for assessing the status of Ki67 proliferation index and Cytokeratin-19 in hepatocellular carcinoma

期刊论文
发表日期：

2025年2月18日

查看全文

Lina Zhang,
Yuanqiang Xiao,
Mengshi Dong,
Mengsi Li,
Haimei Chen,
Jin Wang

摘要

Objectives

To investigate the potential diagnostic value of MR elastography (MRE)-based stiffness to noninvasively predict the status of Ki67 proliferation index (PI) and Cytokeratin-19 (CK19) in hepatocellular carcinoma (HCC).

Methods

Two hundred and nineteen patients with histologically proven HCCs (high/low Ki67 PI, N = 134/85; +/−CK19, N = 47/172) who underwent preoperative MRI and MRE examinations were retrospectively enrolled. Imaging features and clinical information were analyzed to develop the clinical-radiologic models. Logistic regression and area under the receiver operating characteristic curve (AUC) analyses were used to evaluate the status of Ki67 PI and CK19, and their model performances.

Results

Tumor stiffness (TS) (p = 0.004) and TS/liver stiffness (LS) (p < 0.001) were significantly independent predictors of high Ki67 PI and positive CK19, respectively. By incorporating TS or TS/LS value, the AUCs of the clinical-radiological models have been improved for identifying high Ki67 PI from 0.796 to 0.819 (p = 0.113), and positive CK19 from 0.817 to 0.852 (p = 0.026). Patients with predicted high Ki67 PI/positive CK19 based on the combined models had worse recurrence-free survival (RFS) and overall survival (OS) than those with low Ki67 PI/negative CK19 (all p < 0.05), respectively. The combined statuses of predicted Ki67 PI and CK19 stratified patients into three distinct subgroups with significantly different RFS and OS (all p < 0.05). Notably, patients with predicted high Ki67 PI and positive CK19 had the worst prognosis.

Conclusions

Three-dimensional MRE could be a promising tool for evaluating and stratifying the prognosis of HCC patients by predicting the status of Ki67 PI and CK19 preoperatively.

Key Points

Question Can MRE-based stiffness predict Ki67 PI and Cytokeratin 19 (CK19) in HCC?

Findings MRE-based stiffness is an independent predictor of high Ki67 PI and CK19-positive and improves the diagnostic performance of clinical-radiological models predicting Ki67 PI and CK19 status.

Clinical relevance MRE-based stiffness is a noninvasive predictor of high Ki67 PI and CK19-positive, and in combination with clinical-radiological models can be used for prognostic analysis, which could refine patients’ selection for adjuvant therapy or surveillance.