Mental health issues, such as anxiety, are highly prevalent among medical graduate students. However, the potential relationship between social capital and anxiety symptoms in medical graduate students has not yet been investigated. Therefore, data were collected in 2024 from a multicenter sample of 2,587 medical graduate students obtained through multistage stratified sampling across northern, central, and southern Anhui Province. Anxiety symptoms were assessed using the 20-item Self-Rating Anxiety Scale (SAS-20). Social capital was evaluated across six dimensions: social participation, social support, social connection, trust, sense of belonging, and reciprocity. Latent profile analysis (LPA) was used to identify latent anxiety severity profiles. Multinomial logistic regression was used to examine associations between social capital dimensions and anxiety severity-profile membership. The prevalence of anxiety symptoms among participants was 30.5%. Three latent anxiety severity profiles were identified: low (50.6%), moderate (18.6%), and high (30.8%). Lower levels of social support, social connection, and trust were associated with greater odds of membership in the moderate-and high-anxiety severity profiles. A higher sense of belonging was associated with lower odds of membership in the high-anxiety profile, whereas higher social participation was associated with greater odds of membership in the high-anxiety profile. Compared with the moderate-anxiety profile, membership in the high-anxiety profile was also associated with higher social participation and lower trust. Cognitive social capital, particularly trust and supportive connections, was consistently associated with lower anxiety severity, whereas structural participation showed a context-dependent association. These findings suggest that the quality and context of social relationships may warrant as much attention as the quantity of participation.
Modern systemic regimens have increased objective response in unresectable hepatocellular carcinoma (HCC), producing a small but clinically important population in whom curative-intent local treatment can be considered. And yet response is not synonymous with resectability, and resectability is not synonymous with an indication for hepatectomy. This review begins at the point of making a post-response decision rather than ranking conversion regimens or proposing a general framework for longitudinal treatment sequencing. We separate technical, liver-functional, and oncological unresectability and propose a 5D framework-Disease extent and biology, Depth and durability of response, Definitive disease clearance, Durable liver reserve, and Decision timing-to distinguish patients who merely respond from those for whom resection is both feasible and rational. Retrospective series support feasibility and occasionally durable disease control in selected patients. The TALENTOP phase 3 trial now provides the first randomized comparative signal favoring resection in a narrowly defined post-induction population already judged suitable for R0 resection; however, overall survival remains immature, and the trial has not demonstrated a general causal survival benefit for all responders. The next phase of research should therefore validate reproducible decision rules, prespecify resectability assessments, and anchor analyses to the time surgical eligibility is achieved. Conversion hepatectomy should be understood not as a strict requirement after tumor shrinkage but as an irreversible switch in treatment justified by concordant tumor biology, depth and durability of response, complete disease clearance, hepatic safety, and superiority over available alternatives.
