Objective To explore whether incorporating the Ki-67 expression into the modified grading method improves the accuracy of preoperative MRI features in predicting patient outcomes. Methods This retrospective study analyzed STS patients treated from February 2013 to August 2022. Preoperative MR images, FNCLCC grades, Ki-67 expressions, and follow-up information were recorded. Grade II patients with high Ki-67 expression and all Grade III sarcoma patients were defined as the high-grade group. The clinical usefulness of the four grading methods was also compared. MR images were evaluated to extract 17 oncologic, intratumoral, and peritumoral features. High-grade relevant MRI features were identified by multivariate binary logistic regression. Kaplan–Meier and Cox risk regression methods were performed for prognostic analysis of high-grade relevant MRI feature combinations. Results Overall 207 STS patients were included (mean age ± standard deviation, 59.5 ± 13.2 years). According to the modified grading method, 99 patients were classified in the high-grade group. The modified grading method was superior to the two traditional grading methods in predicting patient outcomes. According to multivariate analysis, a tumor size greater than 10 cm (odds ratio (OR), 5.44), a necrosis volume greater than 50% (OR, 2.71), peritumoral enhancement (OR, 3.06), and a multilobulated configuration (OR, 3.27) were associated with high-grade STS. At least three of the four MRI features were associated with worse overall and metastasis-free survival. Conclusion The modified grading system demonstrated improved reliability in managing sarcoma patients. The combination of preoperative MRI features may predict high-grade sarcoma and, potentially, poor patient survival and metastasis. Key Points <Emphasis Type="BoldItalic">Question</Emphasis> Does an improved grading method that considers Ki-67 expression and FNCLCC grading correlate with MRI features and improve prognostication of soft-tissue sarcoma (STS)? <Emphasis Type="BoldItalic">Findings</Emphasis> MRI features, including tumor size, necrosis volume, peritumoral enhancement, and multilobulated configuration, are associated with high-grade STS, overall survival, and metastasis-free survival. <Emphasis Type="BoldItalic">Clinical relevance</Emphasis> The modified grading method improves the accuracy of predicting patients’ OS and MFS compared to conventional grading methods. Radiologists can suggest STS patients with potentially poor clinical outcomes and guide personalized treatment through high-grade associated MRI features.

Soft-tissue sarcoma: modified grading method improves the accuracy of preoperative MRI in predicting patient outcomes

期刊论文
发表日期：

2025年2月6日

查看全文

Xiangwen Li,
Yiwen Hu,
Yuxue Xie,
Bijuan Huang,
Shuang Chen,
Hongyue Tao

摘要

Objective

To explore whether incorporating the Ki-67 expression into the modified grading method improves the accuracy of preoperative MRI features in predicting patient outcomes.

Methods

This retrospective study analyzed STS patients treated from February 2013 to August 2022. Preoperative MR images, FNCLCC grades, Ki-67 expressions, and follow-up information were recorded. Grade II patients with high Ki-67 expression and all Grade III sarcoma patients were defined as the high-grade group. The clinical usefulness of the four grading methods was also compared. MR images were evaluated to extract 17 oncologic, intratumoral, and peritumoral features. High-grade relevant MRI features were identified by multivariate binary logistic regression. Kaplan–Meier and Cox risk regression methods were performed for prognostic analysis of high-grade relevant MRI feature combinations.

Results

Overall 207 STS patients were included (mean age ± standard deviation, 59.5 ± 13.2 years). According to the modified grading method, 99 patients were classified in the high-grade group. The modified grading method was superior to the two traditional grading methods in predicting patient outcomes. According to multivariate analysis, a tumor size greater than 10 cm (odds ratio (OR), 5.44), a necrosis volume greater than 50% (OR, 2.71), peritumoral enhancement (OR, 3.06), and a multilobulated configuration (OR, 3.27) were associated with high-grade STS. At least three of the four MRI features were associated with worse overall and metastasis-free survival.

Conclusion

The modified grading system demonstrated improved reliability in managing sarcoma patients. The combination of preoperative MRI features may predict high-grade sarcoma and, potentially, poor patient survival and metastasis.

Key Points

Question Does an improved grading method that considers Ki-67 expression and FNCLCC grading correlate with MRI features and improve prognostication of soft-tissue sarcoma (STS)?

Findings MRI features, including tumor size, necrosis volume, peritumoral enhancement, and multilobulated configuration, are associated with high-grade STS, overall survival, and metastasis-free survival.

Clinical relevance The modified grading method improves the accuracy of predicting patients’ OS and MFS compared to conventional grading methods. Radiologists can suggest STS patients with potentially poor clinical outcomes and guide personalized treatment through high-grade associated MRI features.