Bowen Zhuang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Bowen Zhuang has a total of 94 co-authors. Here are the top co-authors by number of shared publications:
Bin Qiao
Department of Medical Ultrasonics, Institute of Diagnostic and Interventional Ultrasound, The First Affiliated Hospital of Sun Yat-Sen University, Guangzhou, China
8 shared publications- Xiaoyan Xie
Department of Medical Ultrasonics, Institute of Diagnostic and Interventional Ultrasound, First Affiliated Hospital of Sun Yat-Sen University, Guangzhou, China
8 shared publications Ke Lin
Department of Medical Ultrasonics, Institute of Diagnostic and Interventional Ultrasound, The First Affiliated Hospital of Sun Yat-Sen University, Guangzhou, China
6 shared publicationsNan Zhang
Department of Medical Ultrasonics, Institute of Diagnostic and Interventional Ultrasound, The First Affiliated Hospital of Sun Yat-Sen University, Guangzhou, China
6 shared publications- Manxia Lin
Department of Medical Ultrasonics, Sun Yat-sen University First Affiliated Hospital, Guangzhou, Guangdong, China
5 shared publications Xiao-Yan Xie
Department of Medical Ultrasonics, Institute of Diagnostic and Interventional Ultrasound, The First Affiliated Hospital of Sun Yat-Sen University, NO.58 Zhongshan Road 2, Guangzhou 510080, PR China
5 shared publications
Latest Publications
- Open Access
- 6 July 2026
- Published by Elsevier in JHEP Reports
- Vol. 8 (10) , 101954
A risk-based post ablation follow-up strategy for hepatocellular carcinoma
- 6 July 2026
- Published by Elsevier in JHEP Reports
- Vol. 8 (10) , 101954
Open AccessA risk-based post ablation follow-up strategy for hepatocellular carcinoma
- Open Access
- 8 June 2026
- Published by Wolters Kluwer Health in Hepatology
High resolution spatial transcriptomics identifies insufficient radiofrequency ablation induces hepatocellular carcinoma progression via CEBPD / CXCL2 axis
Background & Aims: Radiofrequency ablation (RFA) is becoming a standard treatment for early-stage hepatocellular carcinoma (HCC). However, high rates of postoperative recurrence, particularly following insufficient RFA (iRFA), remains a clinical obstacle. iRFA contributes to the formation of an immunosuppressive tumor microenvironment and facilitates tumor progression, though the molecular mechanisms driving these processes are not fully elucidated. This study seeks to comprehensively characterize the spatial architecture and cellular interactions within iRFA HCC, and to identify key transcriptional mechanisms through which sublethal heat stress promotes immune evasion and recurrence. Approach & Results: We integrated high-definition spatial transcriptomics (Visium HD) with single-cell RNA sequencing data to map the cellular heterogeneity and communication networks in samples from 6 HCC patients who underwent surgical resection after iRFA. Functional validation was performed using in vitro heat stress models, co-cultured assays, and orthotopic mouse models. We provided a high-resolution spatial atlas of iRFA HCC and the ablation zone, identifying a subpopulation of tumor cells exhibiting activation of the transcription factor CEBPD in response to sublethal heat stress. CEBPD directly bound to the promoter of CXCL2 and drove its expression. Up-regulation of CXCL2 diminished immune checkpoint inhibitors response in iRFA HCC through promoting SPP1+ TAM accumulation via the CXCL2/STAT3/SPP1 axis, and enhanced tumor cell invasion through autocrine signaling. Conclusions: Our study provides a comprehensive spatial characterization of HCC following iRFA, identifies the heat stress-CEBPD-CXCL2 axis as a key driver of tumor progression post-RFA, and demonstrating the therapeutic potential of CXCR2 inhibition for preventing tumor recurrence following RFA. Export
- 8 June 2026
- Published by Wolters Kluwer Health in Hepatology
Open AccessHigh resolution spatial transcriptomics identifies insufficient radiofrequency ablation induces hepatocellular carcinoma progression via CEBPD / CXCL2 axis
Background & Aims: Radiofrequency ablation (RFA) is becoming a standard treatment for early-stage hepatocellular carcinoma (HCC). However, high rates of postoperative recurrence, particularly following insufficient RFA (iRFA), remains a clinical obstacle. iRFA contributes to the formation of an immunosuppressive tumor microenvironment and facilitates tumor progression, though the molecular mechanisms driving these processes are not fully elucidated. This study seeks to comprehensively characterize the spatial architecture and cellular interactions within iRFA HCC, and to identify key transcriptional mechanisms through which sublethal heat stress promotes immune evasion and recurrence. Approach & Results: We integrated high-definition spatial transcriptomics (Visium HD) with single-cell RNA sequencing data to map the cellular heterogeneity and communication networks in samples from 6 HCC patients who underwent surgical resection after iRFA. Functional validation was performed using in vitro heat stress models, co-cultured assays, and orthotopic mouse models. We provided a high-resolution spatial atlas of iRFA HCC and the ablation zone, identifying a subpopulation of tumor cells exhibiting activation of the transcription factor CEBPD in response to sublethal heat stress. CEBPD directly bound to the promoter of CXCL2 and drove its expression. Up-regulation of CXCL2 diminished immune checkpoint inhibitors response in iRFA HCC through promoting SPP1+ TAM accumulation via the CXCL2/STAT3/SPP1 axis, and enhanced tumor cell invasion through autocrine signaling. Conclusions: Our study provides a comprehensive spatial characterization of HCC following iRFA, identifies the heat stress-CEBPD-CXCL2 axis as a key driver of tumor progression post-RFA, and demonstrating the therapeutic potential of CXCR2 inhibition for preventing tumor recurrence following RFA. Export
- Open Access
- 23 March 2026
- Published by Elsevier in JHEP Reports
- Vol. 8 (6) , 101824
Elevation of liver elasticity following radiofrequency ablation reflects neutrophil-mediated abscopal effect in liver cancer
- 23 March 2026
- Published by Elsevier in JHEP Reports
- Vol. 8 (6) , 101824
Open AccessElevation of liver elasticity following radiofrequency ablation reflects neutrophil-mediated abscopal effect in liver cancer
- 1 March 2026
- Published by American Society of Clinical Oncology (ASCO) in JCO Precision Oncology
- Vol. 10 (3) , e2500858
Pretreatment KRAS-Mutant Circulating Tumor DNA Predicts Incidence and Patterns of Recurrence in Patients Undergoing Radiofrequency Ablation for Colorectal Liver Metastases: A Prospective Analysis
PURPOSE: This prospective study aimed to evaluate the prognostic significance of pretreatment KRAS -mutant circulating tumor DNA (ctDNA) in predicting recurrence incidence and patterns in patients with colorectal liver metastasis (CRLM) undergoing radiofrequency ablation (RFA). METHODS: Seventy-eight patients with CRLM treated with curative-intent RFA were enrolled. Pretreatment plasma ctDNA was analyzed using droplet digital polymerase chain reaction for KRAS mutations. Clinicopathologic data, recurrence patterns, and progression-free survival (PFS) were correlated with ctDNA and KRAS status. Transcriptomic profiling of tumor tissues was performed to explore biological mechanisms. RESULTS: Pretreatment ctDNA was detected positive in 26.9% (21/78) of patients. Patients who harbored KRAS -mutant ctDNA exhibited significantly higher recurrence rates (71.4% v 31.6%, P = .004) and shorter median PFS (352 days v not reached, P < .001). Multivariate Cox analysis confirmed that >3 hepatic lesions (adjusted hazard ratio [HR], 7.66 [95% CI, 2.70 to 21.8], P < .001) and positive ctDNA status (adjusted HR, 3.20 [95% CI, 1.60 to 6.42], P = .001) were independent predictors of recurrence. Moreover, ctDNA-positive patients predominantly developed lung metastases (60%), while ctDNA-negative patients showed liver-dominant recurrence (61.1%). Transcriptomic analysis linked ctDNA positivity to epithelial-mesenchymal transition and DNA repair pathways, with elevated nicotinamide N-methyltransferase expression correlating with aggressive phenotypes. CONCLUSION: Pretreatment KRAS -mutant ctDNA predicts higher recurrence risk and shows an association with lung-predominant metastasis in patients with CRLM after RFA. Integrating ctDNA and KRAS profiling enhances risk stratification and supports personalized surveillance strategies.
- 1 March 2026
- Published by American Society of Clinical Oncology (ASCO) in JCO Precision Oncology
- Vol. 10 (3) , e2500858
Pretreatment KRAS-Mutant Circulating Tumor DNA Predicts Incidence and Patterns of Recurrence in Patients Undergoing Radiofrequency Ablation for Colorectal Liver Metastases: A Prospective Analysis
PURPOSE: This prospective study aimed to evaluate the prognostic significance of pretreatment KRAS -mutant circulating tumor DNA (ctDNA) in predicting recurrence incidence and patterns in patients with colorectal liver metastasis (CRLM) undergoing radiofrequency ablation (RFA). METHODS: Seventy-eight patients with CRLM treated with curative-intent RFA were enrolled. Pretreatment plasma ctDNA was analyzed using droplet digital polymerase chain reaction for KRAS mutations. Clinicopathologic data, recurrence patterns, and progression-free survival (PFS) were correlated with ctDNA and KRAS status. Transcriptomic profiling of tumor tissues was performed to explore biological mechanisms. RESULTS: Pretreatment ctDNA was detected positive in 26.9% (21/78) of patients. Patients who harbored KRAS -mutant ctDNA exhibited significantly higher recurrence rates (71.4% v 31.6%, P = .004) and shorter median PFS (352 days v not reached, P < .001). Multivariate Cox analysis confirmed that >3 hepatic lesions (adjusted hazard ratio [HR], 7.66 [95% CI, 2.70 to 21.8], P < .001) and positive ctDNA status (adjusted HR, 3.20 [95% CI, 1.60 to 6.42], P = .001) were independent predictors of recurrence. Moreover, ctDNA-positive patients predominantly developed lung metastases (60%), while ctDNA-negative patients showed liver-dominant recurrence (61.1%). Transcriptomic analysis linked ctDNA positivity to epithelial-mesenchymal transition and DNA repair pathways, with elevated nicotinamide N-methyltransferase expression correlating with aggressive phenotypes. CONCLUSION: Pretreatment KRAS -mutant ctDNA predicts higher recurrence risk and shows an association with lung-predominant metastasis in patients with CRLM after RFA. Integrating ctDNA and KRAS profiling enhances risk stratification and supports personalized surveillance strategies.
- 26 February 2026
- Published by Elsevier in Journal of Vascular and Interventional Radiology
- Vol. 37 (5) , 108580
Delayed Rupture of Subcapsular Hepatic Gastrointestinal Stromal Tumor Metastasis Following Percutaneous Biopsy
- 26 February 2026
- Published by Elsevier in Journal of Vascular and Interventional Radiology
- Vol. 37 (5) , 108580
Delayed Rupture of Subcapsular Hepatic Gastrointestinal Stromal Tumor Metastasis Following Percutaneous Biopsy
- Open Access
- 10 January 2026
- Published by American Association for the Advancement of Science (AAAS) in Computational and Structural Biotechnology Journal
- Vol. 32, 34-47
ChatGPT-4V prompt: A tool to enhance junior radiologists' diagnostic capabilities in cystic renal masses to senior-level accuracy
Purpose: This study aimed to evaluate the diagnostic performance of prompt-adapted ChatGPT-4 Vision (ChatGPT-4V) in interpreting contrast-enhanced ultrasound (CEUS) images for the assessment of cystic renal masses (CRMs) using Bosniak classification. Additionally, it tested the ability of the best prompt to assist radiologists with different experience. Materials and methods: This retrospective study included 103 CRMs in patients who underwent CEUS and CT. ChatGPT-4V and six radiologists (three senior and three junior) independently assigned the Bosniak category (BC) based solely on CEUS images. Subsequently, radiologists re-assessed these images while reviewing ChatGPT-4V-generated BC and decided whether to modify their initial assessments. The diagnostic performance of radiologists and prompts was assessed using the area under the receiver operating characteristic curve (AUC). Results: The AUCs for prompts ranged from 0.507 to 0.688, whereas that for radiologists ranged from 0.685 to 0.831. Among all prompts, Reflection of Thoughts (ROT) prompt achieved the highest AUC, demonstrating performance comparable to juniors (0.688 vs. 0.715, P = 0.727). Although the AUC was lower than that of seniors (0.688 vs. 0.832, P = 0.019), ROT improved the AUCs of juniors: from 0.714 to 0.834 for junior 1, from 0.685 to 0.782 for junior 2, and from 0.704 to 0.783 for junior 3, with the post-assistance performance of all three being comparable to that of seniors. Conclusion: Prompt-adapted ChatGPT-4V showed variable performance in interpreting CEUS images. ROT as the best-performing prompt achieved diagnostic performance comparable to juniors, and it could help juniors achieve an AUC comparable to that of seniors.
- 10 January 2026
- Published by American Association for the Advancement of Science (AAAS) in Computational and Structural Biotechnology Journal
- Vol. 32, 34-47
Open AccessChatGPT-4V prompt: A tool to enhance junior radiologists' diagnostic capabilities in cystic renal masses to senior-level accuracy
Purpose: This study aimed to evaluate the diagnostic performance of prompt-adapted ChatGPT-4 Vision (ChatGPT-4V) in interpreting contrast-enhanced ultrasound (CEUS) images for the assessment of cystic renal masses (CRMs) using Bosniak classification. Additionally, it tested the ability of the best prompt to assist radiologists with different experience. Materials and methods: This retrospective study included 103 CRMs in patients who underwent CEUS and CT. ChatGPT-4V and six radiologists (three senior and three junior) independently assigned the Bosniak category (BC) based solely on CEUS images. Subsequently, radiologists re-assessed these images while reviewing ChatGPT-4V-generated BC and decided whether to modify their initial assessments. The diagnostic performance of radiologists and prompts was assessed using the area under the receiver operating characteristic curve (AUC). Results: The AUCs for prompts ranged from 0.507 to 0.688, whereas that for radiologists ranged from 0.685 to 0.831. Among all prompts, Reflection of Thoughts (ROT) prompt achieved the highest AUC, demonstrating performance comparable to juniors (0.688 vs. 0.715, P = 0.727). Although the AUC was lower than that of seniors (0.688 vs. 0.832, P = 0.019), ROT improved the AUCs of juniors: from 0.714 to 0.834 for junior 1, from 0.685 to 0.782 for junior 2, and from 0.704 to 0.783 for junior 3, with the post-assistance performance of all three being comparable to that of seniors. Conclusion: Prompt-adapted ChatGPT-4V showed variable performance in interpreting CEUS images. ROT as the best-performing prompt achieved diagnostic performance comparable to juniors, and it could help juniors achieve an AUC comparable to that of seniors.
- Open Access
- 27 August 2025
- Published by Springer Nature in Cancer Imaging
- Vol. 25 (1) , 1-14
Contrast-enhanced ultrasound for diagnosing subtypes of intrahepatic cholangiocarcinoma: a comparative study with poorly differentiated hepatocellular carcinoma
Pathologically, intrahepatic cholangiocarcinoma (ICC) is classified into small-duct (SD) type and large-duct (LD) type, each with distinct clinicopathological characteristics. The contrast-enhanced ultrasound (CEUS) features of the two ICC types remain insufficiently explored. To evaluate liver CEUS imaging for differentiating the SD and LD types of ICC and further compare them with poorly differentiated hepatocellular carcinoma (pHCC). A single-center retrospective study enrolled 252 patients with SD-type ICC, LD-type ICC, or pHCC between October 2017 and August 2023. Logistic regression analyses identified independent clinical, pathological, ultrasound, and CEUS predictors. Based on these features, a decision tree-based diagnostic model was developed. The model’s performance was evaluated using receiver operating characteristic (ROC) curve analysis in both the training and validation cohorts, as well as in subgroup stratified by tumor size ≤ 5 cm and > 5 cm. Differences in overall survival (OS) and recurrence-free survival (RFS) based on the model were further analyzed. Overall, 252 patients (mean age, 58.4 ± 10.7 years; 174 males) with 140 SD-type ICC, 55 LD-type ICC and 57 pHCC were enrolled. Multivariate analysis revealed that AFP, CEA, CA19-9, HBsAg status, arterial phase enhancement pattern, washout time ≤ 45 s, and marked washout were independent predictors for tumor categories differentiation (all P <.05). The decision tree-based model incorporating the major features demonstrated excellent performance in both the training cohort (AUC 0.89) and validation cohort (AUC 0.88), as well as in tumor size ≤ 5 cm (AUC 0.90) and > 5 cm (AUC 0.84). OS was significantly worse in LD-type ICC patients compared to SD-type and pHCC (P <.05 for both), while RFS showed no significant difference. A user-friendly, decision tree-based diagnostic model was developed to accurately predict ICC subtypes and pHCC, facilitating improved clinical decision-making. The decision tree-based diagnostic model effectively diagnosed small-duct type and large-duct type intrahepatic cholangiocarcinoma, as well as poorly differentiated hepatocellular carcinoma.
- 27 August 2025
- Published by Springer Nature in Cancer Imaging
- Vol. 25 (1) , 1-14
Open AccessContrast-enhanced ultrasound for diagnosing subtypes of intrahepatic cholangiocarcinoma: a comparative study with poorly differentiated hepatocellular carcinoma
Pathologically, intrahepatic cholangiocarcinoma (ICC) is classified into small-duct (SD) type and large-duct (LD) type, each with distinct clinicopathological characteristics. The contrast-enhanced ultrasound (CEUS) features of the two ICC types remain insufficiently explored. To evaluate liver CEUS imaging for differentiating the SD and LD types of ICC and further compare them with poorly differentiated hepatocellular carcinoma (pHCC). A single-center retrospective study enrolled 252 patients with SD-type ICC, LD-type ICC, or pHCC between October 2017 and August 2023. Logistic regression analyses identified independent clinical, pathological, ultrasound, and CEUS predictors. Based on these features, a decision tree-based diagnostic model was developed. The model’s performance was evaluated using receiver operating characteristic (ROC) curve analysis in both the training and validation cohorts, as well as in subgroup stratified by tumor size ≤ 5 cm and > 5 cm. Differences in overall survival (OS) and recurrence-free survival (RFS) based on the model were further analyzed. Overall, 252 patients (mean age, 58.4 ± 10.7 years; 174 males) with 140 SD-type ICC, 55 LD-type ICC and 57 pHCC were enrolled. Multivariate analysis revealed that AFP, CEA, CA19-9, HBsAg status, arterial phase enhancement pattern, washout time ≤ 45 s, and marked washout were independent predictors for tumor categories differentiation (all P <.05). The decision tree-based model incorporating the major features demonstrated excellent performance in both the training cohort (AUC 0.89) and validation cohort (AUC 0.88), as well as in tumor size ≤ 5 cm (AUC 0.90) and > 5 cm (AUC 0.84). OS was significantly worse in LD-type ICC patients compared to SD-type and pHCC (P <.05 for both), while RFS showed no significant difference. A user-friendly, decision tree-based diagnostic model was developed to accurately predict ICC subtypes and pHCC, facilitating improved clinical decision-making. The decision tree-based diagnostic model effectively diagnosed small-duct type and large-duct type intrahepatic cholangiocarcinoma, as well as poorly differentiated hepatocellular carcinoma.
- Open Access
- 16 December 2024
- Published by Frontiers Media SA in Frontiers in Bioengineering and Biotechnology
- Vol. 12, 1514245
Intrinsic anti-inflammatory nanomedicines for enhanced pain management
Introduction: Effective postoperative pain management remains a significant challenge due to the severe side effects of opioids and the limitations of existing analgesic delivery systems. Inflammation plays a critical role in pain exacerbation, highlighting the need for therapies that combine analgesic effects with intrinsic anti-inflammatory properties.Methods: Herein, we develop an intrinsic anti-inflammatory nanomedicine designed to enhance pain management by integrating controlled anesthetic release with inherent anti-inflammatory activity. Our nanoplatform utilizes dendritic mesoporous silica nanoparticles (MSNs) loaded with levobupivacaine and coated with Rg3-based liposomes derived from ginsenoside Rg3, termed LMSN-bupi.Results: The MSNs enable sustained and controlled release of the local anesthetic, while the Rg3-liposome coating provides intrinsic anti-inflammatory effects by inhibiting macrophage activation. In animal models, LMSN-bupi demonstrates significantly prolonged analgesic effects and attenuated inflammatory responses compared to traditional liposome-decorated nanoparticles (TMSN-bupi) (n = 5).Discussion: These findings underscore the potential of intrinsic anti-inflammatory nanomedicines in enhancing pain management, offering a promising strategy to overcome the limitations of current therapies and improve patient outcomes in postoperative care.
- 16 December 2024
- Published by Frontiers Media SA in Frontiers in Bioengineering and Biotechnology
- Vol. 12, 1514245
Open AccessIntrinsic anti-inflammatory nanomedicines for enhanced pain management
Introduction: Effective postoperative pain management remains a significant challenge due to the severe side effects of opioids and the limitations of existing analgesic delivery systems. Inflammation plays a critical role in pain exacerbation, highlighting the need for therapies that combine analgesic effects with intrinsic anti-inflammatory properties.Methods: Herein, we develop an intrinsic anti-inflammatory nanomedicine designed to enhance pain management by integrating controlled anesthetic release with inherent anti-inflammatory activity. Our nanoplatform utilizes dendritic mesoporous silica nanoparticles (MSNs) loaded with levobupivacaine and coated with Rg3-based liposomes derived from ginsenoside Rg3, termed LMSN-bupi.Results: The MSNs enable sustained and controlled release of the local anesthetic, while the Rg3-liposome coating provides intrinsic anti-inflammatory effects by inhibiting macrophage activation. In animal models, LMSN-bupi demonstrates significantly prolonged analgesic effects and attenuated inflammatory responses compared to traditional liposome-decorated nanoparticles (TMSN-bupi) (n = 5).Discussion: These findings underscore the potential of intrinsic anti-inflammatory nanomedicines in enhancing pain management, offering a promising strategy to overcome the limitations of current therapies and improve patient outcomes in postoperative care.
- Open Access
- 23 October 2024
- Published by Wiley in Cancer Medicine
- Vol. 13 (20) , e70344
Machine Learning Model Based on Prognostic Nutritional Index for Predicting Long‐Term Outcomes in Patients With HCC Undergoing Ablation
Aims: To develop multiple machine learning (ML) models based on the prognostic nutritional index (PNI) and determine the optimal model for predicting long‐term survival outcomes in hepatocellular carcinoma (HCC) patients after local ablation.Methods: From January 2009 to December 2019, we analyzed data from 848 primary HCC patients who underwent local ablation. ML models were constructed and evaluated using the concordance index (C‐index), concordance‐discordance area under curve (C/D AUC), and Brier scores. The optimal ML model was interpreted using the partial dependence plot (PDP) and SHapley Additive exPlanations (SHAP) framework. Additionally, the prognostic performance of our model was compared with other models.Results: Alkaline phosphatase, preoperation alpha‐fetoprotein level, PNI, tumor number, and tumor size were identified as independent prognostic factors for ML model construction. Among the 19 ML algorithms tested, the Aorsf model showed superior performance in both the training cohort (C/D AUC: 0.733; C‐index: 0.736; Brier score: 0.133) and validation cohort (C/D AUC: 0.713; C‐index: 0.793; Brier score: 0.117). The time‐dependent AUC of the Aorsf model for predicting overall survival was as follows: 1‐, 3‐, 5‐, 7‐, and 9‐year were 0.828, 0.765, 0.781, 0.817, and 0.812 in the training cohort, 0.846, 0.859, 0.824, 0.845, and 0.874 in the validation cohort, respectively. The PDP and SHAP algorithms were employed for visual interpretation. Furthermore, time‐AUC and decision curve analysis demonstrated that the Aorsf model provided superior clinical benefits compared to other models.Conclusion: The PNI‐based Aorsf model effectively predicts long‐term survival outcomes after ablation therapy, making a significant contribution to HCC research by improving surveillance, prevention, and treatment strategies.
- 23 October 2024
- Published by Wiley in Cancer Medicine
- Vol. 13 (20) , e70344
Open AccessMachine Learning Model Based on Prognostic Nutritional Index for Predicting Long‐Term Outcomes in Patients With HCC Undergoing Ablation
Aims: To develop multiple machine learning (ML) models based on the prognostic nutritional index (PNI) and determine the optimal model for predicting long‐term survival outcomes in hepatocellular carcinoma (HCC) patients after local ablation.Methods: From January 2009 to December 2019, we analyzed data from 848 primary HCC patients who underwent local ablation. ML models were constructed and evaluated using the concordance index (C‐index), concordance‐discordance area under curve (C/D AUC), and Brier scores. The optimal ML model was interpreted using the partial dependence plot (PDP) and SHapley Additive exPlanations (SHAP) framework. Additionally, the prognostic performance of our model was compared with other models.Results: Alkaline phosphatase, preoperation alpha‐fetoprotein level, PNI, tumor number, and tumor size were identified as independent prognostic factors for ML model construction. Among the 19 ML algorithms tested, the Aorsf model showed superior performance in both the training cohort (C/D AUC: 0.733; C‐index: 0.736; Brier score: 0.133) and validation cohort (C/D AUC: 0.713; C‐index: 0.793; Brier score: 0.117). The time‐dependent AUC of the Aorsf model for predicting overall survival was as follows: 1‐, 3‐, 5‐, 7‐, and 9‐year were 0.828, 0.765, 0.781, 0.817, and 0.812 in the training cohort, 0.846, 0.859, 0.824, 0.845, and 0.874 in the validation cohort, respectively. The PDP and SHAP algorithms were employed for visual interpretation. Furthermore, time‐AUC and decision curve analysis demonstrated that the Aorsf model provided superior clinical benefits compared to other models.Conclusion: The PNI‐based Aorsf model effectively predicts long‐term survival outcomes after ablation therapy, making a significant contribution to HCC research by improving surveillance, prevention, and treatment strategies.
- Open Access
- 21 September 2023
- Published by Taylor & Francis in International Journal of Hyperthermia
- Vol. 40 (1) , 2256498
A novel prognostic model based on AFP, tumor burden score and Albumin-Bilirubin grade for patients with hepatocellular carcinoma undergoing radiofrequency ablation
The aim of this study was to develop prognostic scores, including the tumor burden score (TBS) and albumin-bilirubin (ALBI) grade, for evaluating the outcomes of hepatocellular carcinoma (HCC) patients after radiofrequency ablation (RFA). This retrospective study enrolled treatment-naïve HCC patients with BCLC 0-A who underwent RFA between January 2009 and December 2019. Regular follow-up was conducted after RFA to determine progression-free survival (PFS) and overall survival (OS). The patients were randomly allocated to the training or validation datasets in a 1:1 ratio. Preoperative prognostic scores were developed based on the results of multivariate analysis. The discriminatory ability of the scores was assessed using time-dependent AUC and compared with other models. Serum alpha-fetoprotein (AFP) level and TBS were identified as independent prognostic factors for PFS, while serum AFP, TBS, and ALBI were identified as independent prognostic factors for OS in HCC patients after RFA. The time-dependent AUCs of the AFP-TBS score for the 1-, 3-, and 5-year PFS were 0.651, 0.667, and 0.620, respectively, in the training set, and 0.657, 0.687, and 0.704, respectively, in the validation set. For the 1-, 3-, and 5-year OS, the time-dependent AUCs were 0.680, 0.712, and 0.666, respectively, in the training set, and 0.712, 0.706 and 0.726 in the validation set for the AFP-TBS-ALBI score (ATA). The C-indices and AIC demonstrated that the scores provided better clinical benefits compared to other models. The ATA/AT score, derived from clinical and objective laboratory variables, can assist in individually predicting the prognosis of HCC patients undergoing curative RFA.
- 21 September 2023
- Published by Taylor & Francis in International Journal of Hyperthermia
- Vol. 40 (1) , 2256498
Open AccessA novel prognostic model based on AFP, tumor burden score and Albumin-Bilirubin grade for patients with hepatocellular carcinoma undergoing radiofrequency ablation
The aim of this study was to develop prognostic scores, including the tumor burden score (TBS) and albumin-bilirubin (ALBI) grade, for evaluating the outcomes of hepatocellular carcinoma (HCC) patients after radiofrequency ablation (RFA). This retrospective study enrolled treatment-naïve HCC patients with BCLC 0-A who underwent RFA between January 2009 and December 2019. Regular follow-up was conducted after RFA to determine progression-free survival (PFS) and overall survival (OS). The patients were randomly allocated to the training or validation datasets in a 1:1 ratio. Preoperative prognostic scores were developed based on the results of multivariate analysis. The discriminatory ability of the scores was assessed using time-dependent AUC and compared with other models. Serum alpha-fetoprotein (AFP) level and TBS were identified as independent prognostic factors for PFS, while serum AFP, TBS, and ALBI were identified as independent prognostic factors for OS in HCC patients after RFA. The time-dependent AUCs of the AFP-TBS score for the 1-, 3-, and 5-year PFS were 0.651, 0.667, and 0.620, respectively, in the training set, and 0.657, 0.687, and 0.704, respectively, in the validation set. For the 1-, 3-, and 5-year OS, the time-dependent AUCs were 0.680, 0.712, and 0.666, respectively, in the training set, and 0.712, 0.706 and 0.726 in the validation set for the AFP-TBS-ALBI score (ATA). The C-indices and AIC demonstrated that the scores provided better clinical benefits compared to other models. The ATA/AT score, derived from clinical and objective laboratory variables, can assist in individually predicting the prognosis of HCC patients undergoing curative RFA.