Amino acid metabolomics and machine learning-driven assessment of future liver remnant growth after hepatectomy in livers of various backgrounds.

Accurate assessment of future liver remnant growth after partial hepatectomy (PH) in patients with different liver backgrounds is a pressing clinical issue. Amino acid (AA) metabolism plays a crucial role in liver regeneration. In this study, we combined metabolomics and machine learning (ML) to develop a generalized future liver remnant assessment model for multiple liver backgrounds. The liver index was calculated at 0, 6, 24, 48, 72 and 168 h after 70 % PH in healthy mice and mice with nonalcoholic steatohepatitis or liver fibrosis. The serum levels of 39 amino acids (AAs) were measured using UPLC-MS/MS. The dataset was randomly divided into training and testing sets at a 2:1 ratio, and orthogonal partial least squares regression (OPLS) and minimally biased variable selection in R (MUVR) were used to select a metabolite signature of AAs. To assess liver remnant growth, nine ML models were built, and evaluated using the coefficient of determination (R2), mean absolute error (MAE), and root mean square error (RMSE). The post-Pareto technique for order preference by similarity to the ideal solution (TOPSIS) was employed for ranking the ML algorithms, and a stacking technique was utilized to establish consensus among the superior algorithms. Compared with those of OPLS, the signature AAs set identified by MUVR (Thr, Arg, EtN, Phe, Asa, 3MHis, Abu, Asp, Tyr, Leu, Ser, and bAib) are more concise. Post-Pareto TOPSIS ranking demonstrated that the majority of ML algorithm in combinations with MUVR outperformed those with OPLS. The established SVM-KNN consensus model performed best, with an R2 of 0.79, an MAE of 0.0029, and an RMSE of 0.0035 for the testing set. This study identified a metabolite signature of 12 AAs and constructed an SVM-KNN consensus model to assess future liver remnant growth after PH in mice with different liver backgrounds. Our preclinical study is anticipated to establish an alternative and generalized assessment method for liver regeneration.

Amino acid metabolomics and machine learning for assessment of post-hepatectomy liver regeneration.

Objective: Amino acid (AA) metabolism plays a vital role in liver regeneration. However, its measuring utility for post-hepatectomy liver regeneration under different conditions remains unclear. We aimed to combine machine learning (ML) models with AA metabolomics to assess liver regeneration in health and non-alcoholic steatohepatitis (NASH). Methods: The liver index (liver weight/body weight) was calculated following 70% hepatectomy in healthy and NASH mice. The serum levels of 39 amino acids were measured using ultra-high performance liquid chromatography-tandem mass spectrometry analysis. We used orthogonal partial least squares discriminant analysis to determine differential AAs and disturbed metabolic pathways during liver regeneration. The SHapley Additive exPlanations algorithm was performed to identify potential AA signatures, and five ML models including least absolute shrinkage and selection operator, random forest, K-nearest neighbor (KNN), support vector regression, and extreme gradient boosting were utilized to assess the liver index. Results: Eleven and twenty-two differential AAs were identified in the healthy and NASH groups, respectively. Among these metabolites, arginine and proline metabolism were commonly disturbed metabolic pathways related to liver regeneration in both groups. Five AA signatures were identified, including hydroxylysine, L-serine, 3-methylhistidine, L-tyrosine, and homocitrulline in healthy group, and L-arginine, 2-aminobutyric acid, sarcosine, beta-alanine, and L-cysteine in NASH group. The KNN model demonstrated the best evaluation performance with mean absolute error, root mean square error, and coefficient of determination values of 0.0037, 0.0047, 0.79 and 0.0028, 0.0034, 0.71 for the healthy and NASH groups, respectively. Conclusion: The KNN model based on five AA signatures performed best, which suggests that it may be a valuable tool for assessing post-hepatectomy liver regeneration in health and NASH.

Online interpretable dynamic prediction models for clinically significant posthepatectomy liver failure based on machine learning algorithms: A retrospective cohort study.

BACKGROUND: Posthepatectomy liver failure (PHLF) is the leading cause of mortality in patients undergoing hepatectomy. However, practical models for accurately predicting the risk of PHLF are lacking. This study aimed to develop precise prediction models for clinically significant PHLF. METHODS: A total of 226 patients undergoing hepatectomy at a single center were recruited. The study outcome was clinically significant PHLF. Five pre- and postoperative machine learning (ML) models were developed and compared with four clinical scores, namely, the MELD, FIB-4, ALBI, and APRI scores. The robustness of the developed ML models was internally validated using 5-fold cross-validation by calculating the average of the evaluation metrics and was externally validated on an independent temporal dataset, including the area under the curve (AUC) and the area under the precision‒recall curve (AUPRC). SHapley Additive exPlanations analysis was performed to interpret the best performance model. RESULTS: Clinically significant PHLF was observed in 23 of 226 patients (10.2%). The variables in the preoperative model included creatinine, total bilirubin, and Child‒Pugh grade. In addition to the above factors, the extent of resection was also a key variable for the postoperative model. The pre- and postoperative artificial neural network (ANN) models exhibited excellent performance, with mean AUCs of 0.766 and 0.851, respectively, and mean AUPRC values of 0.441 and 0.645, whereas the MELD, FIB-4, ALBI, and APRI scores reached AUCs of 0.714, 0.498, 0.536 and 0.551, respectively, and AUPRC values of 0.204, 0.111, 0.128 and 0.163, respectively. In addition, the AUCs of the pre- and postoperative ANN models were 0.720 and 0.731, respectively, and the AUPRC values were 0.380 and 0.408, respectively, on the temporal dataset. CONCLUSION: Our online interpretable dynamic ML models outperformed common clinical scores and could function as a clinical decision support tool to identify patients at high risk of PHLF pre- and postoperatively.

Laparoscopic hepatectomy for hepatocellular carcinoma in patients with clinically significant portal hypertension: a systematic review and meta-analysis.

OBJECTIVE: To compare the effects of laparoscopic hepatectomy (LH) on the short-term and long-term outcomes in hepatocellular carcinoma (HCC) patients with and without clinically significant portal hypertension (CSPH). METHODS: A systematic literature search of the PubMed, EMBASE, and Cochrane databases was performed for articles published from inception to March 1, 2023. Meta-analysis of surgical and oncological outcomes was performed using a random effects model. Data were summarized as mean difference and risk ratio with 95% confidence intervals. RESULTS: Five cohort studies with a total of 310 HCC patients were included (CSPH 143; Non-CSPH 167). In terms of surgical outcomes, estimated blood loss and the length of hospital stay were significantly lower in the Non-CSPH group than in the CSPH group. There were no significant differences between the two groups regarding other surgical outcomes, including the operative time, ratio of conversion to open surgery, and overall complication rate. In addition, there were also no significant differences between the two groups regarding the oncological outcomes, such as 1-, 3-, and 5-year overall survival. CONCLUSIONS: HCC patients with and without CSPH who underwent LH had comparable surgical and oncological outcomes. LH is a safe and effective treatment for HCC patients with CSPH under the premise of rational screening of patients.

Robotic versus laparoscopic liver resection for liver malignancy: a systematic review and meta-analysis of propensity score-matched studies.

OBJECTIVE: How different surgical procedures, including the robotic-assisted liver resection (RLR) and laparoscopic liver resection (LLR), can affect the prognosis of patients with liver malignancies is unclear. Thus, in this study, we compared the effects of RLR and LLR on the surgical and oncological outcomes in patients with liver malignancies through propensity score-matched cohort studies. METHODS: The PubMed, Embase, and Cochrane databases were searched using Medical Subject Headings terms and keywords from inception until May 31, 2023. The quality of the included studies was assessed using the Newcastle-Ottawa quality assessment scale. The mean difference with 95% confidence interval (95% CI) was used for analysis of continuous variables; the risk ratio with 95% CI was used for dichotomous variables; and the hazard ratio with 95% CI was used for survival-related variables. Meta-analysis was performed using a random-effects model. RESULTS: Five high-quality cohort studies with 986 patients were included (370 and 616 cases for RLR and LLR, respectively). In terms of surgical outcomes, there were no significant differences in the operation time, conversion rate to open surgery, overall complication rate, major complication rate, and length of hospital stay between the RLR and LLR groups. In terms of oncological outcomes, there were no significant differences in the 5-year overall survival and disease-free survival between the two groups. CONCLUSION: Surgical and oncological outcomes are comparable between RLR and LLR on patients with liver malignancies. Therefore, the benefits of applying RLR in patients with liver malignancies need to be further explored.

Expression profiles of lncRNAs, miRNAs, and mRNAs during the proliferative phase of liver regeneration in mice with liver fibrosis.

The role of lncRNAs in the regeneration of fibrotic liver is unclear. To address this issue, we established a 70% hepatectomy model of liver fibrosis in mice, used high-throughput sequencing technology to obtain the expression profiles of lncRNAs, miRNAs, and mRNAs, and constructed a lncRNA-miRNA-mRNA regulatory network. A total of 1329 lncRNAs, 167 miRNAs, and 6458 mRNAs were differentially expressed. On this basis, a lncRNA-miRNA-mRNA ceRNA regulatory network consisting of 38 DE lncRNAs, 24 DE miRNAs, and 299 DE mRNAs was constructed, and a transcription factor (TF) - mRNA regulatory network composed of 20 TFs and 98 DE mRNAs was built. Through the protein network analysis, a core protein interaction network composed of 20 hub genes was derived. Furthermore, Xist/miR-144-3p/Cdc14b and Snhg3/miR-365-3p/Map3k14 axes in the ceRNA regulatory network were verified by Real-Time quantitative PCR. Therefore, we concluded that these new insights may further our understanding of liver regeneration.

Integrated analysis of lncRNA/circRNA-miRNA-mRNA in the proliferative phase of liver regeneration in mice with liver fibrosis.

BACKGROUND: Non-coding RNAs play important roles in liver regeneration; however, their functions and mechanisms of action in the regeneration of fibrotic liver have not been elucidated. We aimed to clarify the expression patterns and regulatory functions of lncRNAs, circRNAs, miRNAs, and mRNAs in the proliferative phase of fibrotic liver regeneration. METHODS: Based on a mouse model of liver fibrosis with 70% hepatectomy, whole-transcriptome profiling was performed using high-throughput sequencing on samples collected at 0, 12, 24, 48, and 72 h after hepatectomy. Hub genes were selected by weighted gene co-expression network analysis and subjected to enrichment analysis. Integrated analysis was performed to reveal the interactions of differentially expressed (DE) lncRNAs, circRNAs, miRNAs, and mRNAs, and to construct lncRNA-mRNA cis- and trans-regulatory networks and lncRNA/circRNA-miRNA-mRNA ceRNA regulatory networks. Real-Time quantitative PCR was used to validate part of the ceRNA network. RESULTS: A total of 1,329 lncRNAs, 48 circRNAs, 167 miRNAs, and 6,458 mRNAs were differentially expressed, including 812 hub genes. Based on these DE RNAs, we examined several mechanisms of ncRNA regulatory networks, including lncRNA cis and trans interactions, circRNA parental genes, and ceRNA pathways. We constructed a cis-regulatory core network consisting of 64 lncRNA-mRNA pairs (53 DE lncRNAs and 58 hub genes), a trans-regulatory core network consisting of 103 lncRNA-mRNA pairs (18 DE lncRNAs and 85 hub genes), a lncRNA-miRNA-mRNA ceRNA core regulatory network (20 DE lncRNAs, 12 DE miRNAs, and 33 mRNAs), and a circRNA-miRNA-mRNA ceRNA core regulatory network (5 DE circRNAs, 5 DE miRNAs, and 39 mRNAs). CONCLUSIONS: These results reveal the expression patterns of lncRNAs, circRNAs, miRNAs, and mRNAs in the proliferative phase of fibrotic liver regeneration, as well as core regulatory networks of mRNAs and non-coding RNAs underlying liver regeneration. The findings provide insights into molecular mechanisms that may be useful in developing new therapeutic approaches to ameliorate diseases that are characterized by liver fibrosis, which would be beneficial for the prevention of liver failure and treatment of liver cancer.

Laparoscopic versus open repeat hepatectomy for recurrent hepatocellular carcinoma: a systematic review and meta-analysis of propensity score-matched cohort studies.

OBJECTIVE: The effectiveness of laparoscopic repeat hepatectomy (LRH) versus open repeat hepatectomy (ORH) on recurrent hepatocellular carcinoma (RHCC) is unclear. We compared the surgical and oncological outcomes of LRH and ORH in patients with RHCC with a meta-analysis of studies based on propensity score-matched cohorts. METHODS: A literature search was conducted on PubMed, Embase, and Cochrane Library with Medical Subject Headings terms and keywords until 30 September 2022. The quality of eligible studies was evaluated with the Newcastle-Ottawa Scale. Mean difference (MD) with a 95% CI was used for the analysis of continuous variables; odds ratio (OR) with 95% CI was used for binary variables; and hazard ratio with 95% CI was used for survival analysis. A random-effects model was used for meta-analysis. RESULTS: Five high-quality retrospective studies with 818 patients were included; 409 patients (50%) were treated with LRH and 409 (50%) with ORH. In most surgical outcomes, LRH was superior to ORH: less estimated blood loss, shorter operation time, lower major complication rate, and shorter length of hospital stay (MD=-225.9, 95% CI=[-360.8 to -91.06], P =0.001; MD=66.2, 95% CI=[5.28-127.1], P =0.03; OR=0.18, 95% CI=[0.05-0.57], P =0.004; MD=-6.22, 95% CI=[-9.78 to -2.67], P =0.0006). There were no significant differences in the remaining surgical outcomes: blood transfusion rate and overall complication rate. In oncological outcomes, LRH and ORH were not significantly different in 1-year, 3-year, and 5-year overall survival and disease-free survival. CONCLUSIONS: For patients with RHCC, most surgical outcomes with LRH were superior to those of ORH, but oncological outcomes with the two operations were similar. LRH may be a preferable option for the treatment of RHCC.

Laparoscopic versus open hepatectomy for intrahepatic cholangiocarcinoma: Systematic review and meta-analysis of propensity score-matched studies.

OBJECTIVE: To compare the effects of laparoscopic hepatectomy (LH) versus open hepatectomy (OH) on the short-term and long-term outcomes of patients with intrahepatic cholangiocarcinoma (ICC) through a meta-analysis of studies using propensity score-matched cohorts. METHODS: The literature search was conducted in PubMed, Embase, and Cochrane Library databases until August 31, 2022. Meta-analysis of surgical (major morbidity, the length of hospital stay, 90-day postoperative mortality), oncological (R0 resection rate, lymph node dissection rate) and survival outcomes (1-, 3-, and 5-year overall survival and disease-free survival) was performed using a random effects model. Data were summarized as relative risks (RR), mean difference (MD) and hazard ratio (HR) with 95% confidence intervals (95% CI). RESULTS: Six case-matched studies with 1054 patients were included (LH 518; OH 536). Major morbidity was significantly lower (RR = 0.57, 95% CI = 0.37-0.88, P = 0.01) and the length of hospital stay was significantly shorter (MD = -2.44, 95% CI = -4.19 to -0.69, P = 0.006) in the LH group than in the OH group, but there was no significant difference in 90-day postoperative mortality between the 2 groups. There were no significant differences in R0 resection rate, lymph node dissection rate, 1-, 3-, and 5-year overall survival or disease-free survival between the LH and OH groups. CONCLUSIONS: LH has better surgical outcomes and comparable oncological outcomes and survival outcomes than does OH on ICC. Therefore, laparoscopy is at least not inferior to open surgery for intrahepatic cholangiocarcinoma.

RNA-seq transcriptome profiling of liver regeneration in mice identifies the miR-34b-5p/phosphoinositide-dependent protein kinase 1 axis as a potential target for hepatocyte proliferation.

Characterized by compensatory hyperplasia dependent on hepatocyte proliferation, the liver will initiate regeneration after partial hepatectomy (PH) and acute or chronic injuries. A variety of genes and noncoding RNAs play pivotal roles in these cell proliferation and growth processes. However, it is still unclear how competition endogenous RNAs (ceRNAs) modulate cellular activities during each phase of liver regeneration, and the specific mechanisms of posttranscriptional gene expression regulation in hepatocyte proliferation remain to be elucidated. To investigate the mechanism of liver regeneration through RNA-seq profiling and to determine the role of miR-34b-5p/PDK1 on hepatocyte proliferation, we established a 2/3 PH mouse model for whole transcriptome profiling based on high-throughput sequencing techniques. We subsequently constructed a lncRNA-miRNA-mRNA ceRNA regulatory network through integrative analyses of RNA interactions. Finally, plasmid transfection in NCTC 1469 cells, dual luciferase reporter gene assay, quantitative real-time PCR, western blotting, Cell Counting Kit-8, and EdU-DNA synthesis cell proliferation assay were used to demonstrate the role of the miR-34b-5p/PDK1 axis in hepatocyte proliferation in vitro. A total of 1443 mRNAs (962 up, 481 down), 48 miRNAs (35 up, 13 down), and 1955 lncRNAs (986 up, 969 down) were identified as significantly differentially expressed. We then successfully constructed a ceRNA regulatory network consisting of 7 lncRNAs, 15 miRNAs, and 347 mRNAs based on the predicted inverse interactions among ceRNAs. Additionally, miR-34b-5p/PDK1 was predicted to be closely related to hepatocyte proliferation. We further demonstrated that miR-34b-5p could bind specifically to the 3'-untranslated region (3'-UTR) of PDK1 using the dual luciferase reporter assay. Ectopic overexpression of miR-34b-5p significantly reduced the mRNA and protein expression of PDK1, while it markedly inhibited the proliferation of mouse NCTC 1469 cells in vitro. In contrast, knocking down miR-34b-5p exhibited the inverse effects on PDK1 expression and hepatocyte proliferation. Through analyzing the ceRNA network during mouse liver regeneration, this study reveals that miR-34b-5p can inhibit hepatocyte proliferation through negatively regulating PDK1 and may be a potential pharmacological intervention target.