Clinicopathological features, risk model and prognosis of 115 cases of epithelioid hemangioendothelioma: A single-center study.

OBJECTIVE

To investigate the clinicopathological features, diagnostic value, risk model and prognostic significance of epithelioid hemangioendothelioma (EHE) in a retrospective cohort of 115 cases.

METHODS

A total of 115 cases of EHE diagnosed in the Cancer Hospital of the Chinese Academy of Medical Sciences (NCC) from 2011 to 2023 were collected. The clinical and pathological features of EHE were reviewed by Fluorescence hybridization (FISH) and Immunohistochemistry (IHC). SPSS 25.0 software for the overall survival (OS) curve. Univariate and multivariate COX proportional risk regression models were used to analyze the prognostic factors.

RESULTS

The male to female ratio of 115 patients was 1.05:1. The age of the patients ranged from 8 to 84 years (median, 47 years; standard deviation (SD), 15.055), and tumor diameter ranged from 8 to 152 mm (median, 20 mm; SD, 29.156).Among them, there were 80 multiple cases, 69 cases of the patients underwent surgery. IHC showed that 92.2% of calmodulin-binding transcription activator 1 (CAMTA1) and 58.0% of transcription factor E3 (TFE3) were positively expressed. The positive rate of the WWTR1::CAMTA1 fusion gene was 86.7% and the positive rate of the TFE3 fracture gene was 13.6% (12/88). The difference test between FISH and IHC showed that the two detection methods have good consistency for CAMTA1 gene detection, while the consistency with TFE3 is poor. Univariate COX regression showed that radical surgical resection, tumor size (>5cm) and age (>50 years), multi-organ involvement, and OS were statistically significant (P<0.05). A proposed 3-tiered risk assessment system using these 5 parameters significantly stratified the patients into low-risk, intermediate-risk and high-risk groups with significantly OS rates.

CONCLUSION

The prognosis for EHE patients with tumor size more than 50 mm or age at diagnosis over 50 years old is unfavorable. In this investigation, we pioneered the development of a prognostic risk model, leveraging five key parameters to anticipate the outcomes for EHE patients.