Translational cancer research · 2026
Identification of hepatitis B virus-associated membrane proteins as prognostic biomarkers for hepatocellular carcinoma based on retrospective analysis of integrated membrane proteomics.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Liu Zichen, Wan Jiaming, Deng Guoli, Zhang Mingzhu, Zhang Shaohua, Liu Qizhi, Ouyang Liu, Liu Shupeng
Аннотация
BACKGROUND: Hepatitis B virus (HBV) infection is one of the major risk factors for hepatocellular carcinoma (HCC) development. Current prognostic assessment of HBV-associated HCC (HBV-HCC) largely relies on clinicopathological parameters, such as tumor stage, liver function, and serum alpha-fetoprotein (AFP) levels, which have limited accuracy in identifying patients at high risk of recurrence or poor survival. Therefore, reliable molecular biomarkers are urgently needed to improve individualized risk stratification and guide postoperative surveillance and therapeutic decision-making. HBV infection induces chronic inflammation through mechanisms such as genomic integration, ultimately leading to HCC. Cell membrane proteins play critical roles in maintaining cellular homeostasis, signal transduction, and immune recognition, and may be importantly involved in HBV-HCC. However, the prognostic significance of HBV-associated membrane proteins has not been sufficiently investigated. This study aimed to identify HBV-associated membrane proteins through integrated proteomic analysis and to evaluate their prognostic value for recurrence-free survival (RFS) and overall survival (OS) in patients with HBV-HCC.
METHODS: HBV-associated membrane proteins were identified in HepG2 and Hep3B cells using two parallel proteomic approaches: membrane fractionation and Bacillus subtilis (BS2)-based proximity labeling, both coupled with mass spectrometry (MS). These findings were integrated with proteomic analysis of HBV-HCC tissue samples. Prognostic models for RFS and OS were constructed using machine learning algorithms including least absolute shrinkage and selection operator (LASSO)-Cox (proportional hazards model), random survival forest (RSF), and elastic net (EN).
RESULTS: A total of 98 HBV-associated membrane proteins were identified, including canonical membrane proteins as well as several nuclear and cytoplasmic proteins. Functional annotation revealed numerous tumor-associated molecules, transporters, and enzymes. Univariate and multivariate analyses identified proteins such as POLM and TMEM192 as prognostic factors for RFS, and GLYATL1 and TMEM109 for OS. The prognostic models developed using machine learning achieved receiver operating characteristic (ROC) values exceeding 0.7.
CONCLUSIONS: This study identified HBV-HCC-related membrane proteins beyond classical localized molecules using membrane fractionation and BS2 proximity labeling, and established robust prognostic models through multiple machine learning approaches. These findings provide potential therapeutic targets and prognostic biomarkers that may improve the clinical management of HBV-HCC patients. However, the single-cohort design, absence of external validation, and potential model optimism preclude claims of clinical robustness or immediate application. Independent multicenter validation and standardized targeted protein assays are required before these proteins or prediction models can be considered for clinical risk stratification.
METHODS: HBV-associated membrane proteins were identified in HepG2 and Hep3B cells using two parallel proteomic approaches: membrane fractionation and Bacillus subtilis (BS2)-based proximity labeling, both coupled with mass spectrometry (MS). These findings were integrated with proteomic analysis of HBV-HCC tissue samples. Prognostic models for RFS and OS were constructed using machine learning algorithms including least absolute shrinkage and selection operator (LASSO)-Cox (proportional hazards model), random survival forest (RSF), and elastic net (EN).
RESULTS: A total of 98 HBV-associated membrane proteins were identified, including canonical membrane proteins as well as several nuclear and cytoplasmic proteins. Functional annotation revealed numerous tumor-associated molecules, transporters, and enzymes. Univariate and multivariate analyses identified proteins such as POLM and TMEM192 as prognostic factors for RFS, and GLYATL1 and TMEM109 for OS. The prognostic models developed using machine learning achieved receiver operating characteristic (ROC) values exceeding 0.7.
CONCLUSIONS: This study identified HBV-HCC-related membrane proteins beyond classical localized molecules using membrane fractionation and BS2 proximity labeling, and established robust prognostic models through multiple machine learning approaches. These findings provide potential therapeutic targets and prognostic biomarkers that may improve the clinical management of HBV-HCC patients. However, the single-cohort design, absence of external validation, and potential model optimism preclude claims of clinical robustness or immediate application. Independent multicenter validation and standardized targeted protein assays are required before these proteins or prediction models can be considered for clinical risk stratification.