We are grateful to Hwang et al. for a well-conducted study, as well as for providing additional insights in response to our editorial comments [1]. The study results clearly demonstrated the prognostic role of microvascular invasion (MVI) in hepatocellular carcinoma (HCC). The authors reported that extensive MVI has greater clinical significance because they demonstrated that mild MVI did not significantly affect prognosis, while severe MVI (≥5 microvessels or ≥50 invaded tumor cells) was associated with decreased overall survival [2]. This finding emphasizes the intricate role that MVI plays in the prognosis of HCC. Furthermore, the categorization of MVI into severe and moderate classifications may offer a valuable basis for future prognostic assessments. Nevertheless, more verification in larger, independent groups is required to confirm the relevance of these findings.

The authors highlight the potential application of stratifying MVI severity in a manner similar to the international federation of gynecology and obstetrics staging system for endometrial carcinoma. This comparison is indeed valuable, as stratification by MVI severity could potentially improve prognostication in HCC. The application of severitybased classification systems in different cancers suggests a promising direction. Further studies are needed to validate whether integrating MVI severity into the American Joint Committee on cancer staging system would offer improved prognostic accuracy for HCC patients, particularly for those with larger tumors or more complex disease presentations [3].

Moreover, the authors’ research results on the correlation between imaging characteristics and MVI align with earlier studies. Regarding the variability in imaging features associated with MVI and the challenge of predicting MVI using magnetic resonance imaging (MRI), previous research has identified various MRI characteristics associated with MVI, but the results have often been inconsistent [4-6]. Although the predictive validity of these traits is restricted, they may nevertheless be useful in identifying individuals who are at a greater risk for MVI. Utilizing artificial intelligence to enhance MVI prediction has great potential for further investigation. Nevertheless, it is crucial to verify the accuracy and effectiveness of artificial intelligence (AI) models using separate datasets to guarantee their suitability for clinical use. The collaboration among pathologists, radiologists, and AI professionals will be essential for advancing these technologies [7]. We anticipate seeing future advancements in this field and the possible medical uses that may emerge from this research.