Predicting the outcome of liver transplantation in patients with non-alcoholic steatohepatitis cirrhosis: The NASH LT risk-benefit calculator.

BACKGROUND: Non-alcoholic Steatohepatitis (NASH) cirrhosis is the second most common indication for liver transplantation (LT) in the US and often is associated with significant co-morbidities. We validated a model and risk prediction score that reflects the benefit derived from LT for NASH cirrhosis by predicting 5-year survival post-LT. METHODS: We developed a prediction score utilizing 6515 NASH deceased donor LT (DDLT) recipients from 2002 to 2019 from the Scientific Registry of Transplant Recipients (SRTR) database to identify a parsimonious set of independent predictors of survival. Coefficients of relevant recipient factors were converted to weighted points to construct a risk scoring system that was then externally validated. RESULTS: The final risk score includes the following independent recipient predictors and corresponding points: recipient age (5 points for age ≥70 years), functional status (3 points for total assistance), presence of TIPSS (2 points), hepatic encephalopathy (1 point), serum creatinine (5 points if >1.45 mg/dl), need for mechanical ventilation (3 points), and dialysis within 1 week prior to LT (7 points). Diabetes is a stratifying variable for baseline risk. Scores range from 0 to 20 with scores above 13 having an overall survival of <65% at 5 years post-LT. Internal and external validation indicated good predictive ability. CONCLUSION: Our practically useable and validated risk score helps to identify and stratify candidates who will derive the most long-term benefit from LT for NASH cirrhosis.

Impact of living donor liver transplantation on long-term cardiometabolic and graft outcomes in cirrhosis due to nonalcoholic steatohepatitis.

BACKGROUND AND AIM: Non-alcoholic steatohepatitis (NASH) is a leading indication for liver transplantation (LT). This study aimed to determine whether living donor LT (LDLT) recipients experienced less recurrent NASH, cirrhosis, and cardiometabolic complications compared to deceased donor LT (DDLT). METHOD: Patients with LDLT and DDLT for NASH between February 2002 and May 2018 at University Health Network (UHN) were compared. Cox Proportional Hazard model was used to analyze overall survival (OS), Fine and Gray's Competing Risk models were conducted to analyze cumulative incidence of post LT outcomes. RESULTS: One hundred and ninety-nine DDLTs and 66 LDLTs were performed for NASH cirrhosis. Time and rate of recurrence of NAFLD and NASH were comparable in both groups. Graft cirrhosis was more common in DDLT recipients (n = 14) versus LDLT (n = 0) (p < .0001). Significant fibrosis (Fibrosis ≥ F2) developed in 50 recipients (12 LDLT and 38 DDLT) post LT (DDLT vs. LDLT: HR = 1.00, 95% CI = (.52-1.93), p = .91) and there was no difference in time to significant fibrosis (p = .57). There was no difference in development of post-transplant diabetes, dyslipidemia, metabolic syndrome, cardiovascular disease, and cancers. LDLT group had better renal function at 10 years (MDRD eGFR of 57.0 mL/min vs. 48.5 mL/min, p = .047). Both groups had a comparable OS (HR = 1.83 (95% CI = .92-3.62), p = .08). CONCLUSION: Overall, LDLT recipients had significantly better renal function by virtue of having early transplantation in their disease course. LDLT was also associated with significantly less graft cirrhosis, although OS and cardiometabolic outcomes were comparable between LDLT and DDLT.

Predicting Long-term Survival After Liver Transplantation in Patients With NASH Cirrhosis.

Nonalcoholic steatohepatitis (NASH) cirrhosis is the second most common indication for liver transplantation (LT) in the United States.1 Patients are increasingly older at presentation, with higher rates of metabolic syndrome, obesity, hyperlipidemia, diabetes mellitus, and renal failure.2 They are also at higher risk of cardiovascular events and mortality while on the waiting list1 and in the post-transplant period.3,4 We sought to identify predictors of long-term benefit based on 5-year survival post-LT in NASH cirrhosis, thereby delineating those patients that derive a clear benefit from LT versus those in whom LT may be futile.

Development and validation of an ensemble machine learning framework for detection of all-cause advanced hepatic fibrosis: a retrospective cohort study.

BACKGROUND: Cirrhosis is the result of advanced scarring (or fibrosis) of the liver, and is often diagnosed once decompensation with associated complications has occurred. Current non-invasive tests to detect advanced liver fibrosis have limited performance, with many indeterminate classifications. We aimed to identify patients with advanced liver fibrosis of all-causes using machine learning algorithms (MLAs). METHODS: In this retrospective study of routinely collected laboratory, clinical, and demographic data, we trained six MLAs (support vector machine, random forest classifier, gradient boosting classifier, logistic regression, artificial neural network, and an ensemble of all these algorithms) to detect advanced fibrosis using 1703 liver biopsies from patients seen at the Toronto Liver Clinic (TLC) between Jan 1, 2000, and Dec 20, 2014. Performance was validated using five datasets derived from patient data provided by the TLC (n=104 patients with a biopsy sample taken between March 24, 2014, and Dec 31, 2017) and McGill University Health Centre (MUHC; n=404). Patients with decompensated cirrhosis were excluded. Performance was benchmarked against aspartate aminotransferase-to-platelet ratio index (APRI), fibrosis-4 index (FIB-4), non-alcoholic fatty liver disease fibrosis score (NFS), transient elastography, and an independent panel of five hepatology experts (MB, GS, HK, KP, and RSK). MLA performance was evaluated using the area under the receiver operating characteristic curve (AUROC) and the percentage of determinate classifications. FINDINGS: The best MLA was an ensemble algorithm of support vector machine, random forest classifier, gradient boosting classifier, logistic regression, and neural network algorithms, which achieved 100% determinate classifications (95% CI 100·0-100·0), an AUROC score of 0·870 (95% CI 0·797-0·931) on the TLC validation set (fibrosis stages F0 and F1 vs F4), and an AUROC of 0·716 (95% CI 0·664-0·766) on the MUHC validation set (fibrosis stages F0, F1, and F2 vs F3 and F4). The ensemble MLA outperformed all routinely used biomarkers and achieved comparable performance to hepatologists as measured by AUROC and percentage of indeterminate classifications in both the TLC validation dataset (APRI AUROC score 0·719 [95% CI 0·611-0·820], 83·7% determinate [95% CI 76·0-90·4]; FIB-4 AUROC score 0·825 [95% CI 0·730-0·912], 72·1% determinate [95% CI 63·5-80·8]) and the MUHC validation dataset (APRI AUROC score 0·618 [95% CI 0·548-0·691], 75·5% determinate [95% CI 71·5-79·2]; FIB-4 AUROC score 0·717 (95% CI 0·652-0·776), 75·5% determinate [95% CI 0·713-0·797]), and achieving only slightly lower AUROC than transient elastography (0·773 [95% CI 0·699-0·834] vs 0·826 [95% CI 0·758-0·889]). INTERPRETATION: We have shown that an ensemble MLA outperforms non-imaging-based methods in detecting advanced fibrosis across different causes of liver disease. Our MLA was superior to APRI, FIB-4, and NFS with no indeterminate classifications, while achieving performance comparable to an independent panel of experts. MLAs using routinely collected data could identify patients at high-risk of advanced hepatic fibrosis and cirrhosis among patients with chronic liver disease, allowing intervention before onset of decompensation. FUNDING: Toronto General Hospital Foundation.

Sex Disparity in Liver Transplant and Access to Living Donation.

Importance: The Model for End-stage Liver Disease (MELD)-based organ allocation system has significantly decreased mortality on the transplant waiting list for patients with end-stage liver disease. However, women have remained at a disadvantage with respect to access to deceased donor liver transplant (DDLT) even after introduction of the MELD score for organ allocation. Objective: To determine whether availability of living donation in a transplant program can offset inequity in liver transplant (LT) allocation for women. Design, Setting, and Participants: This cohort study retrospectively analyzed adult patients listed for LT at the University Health Network in Toronto, Ontario, Canada. Patients included had a potential living donor (pLD) at the moment of listing. This study was performed from November 13, 2012, to May 31, 2019. A total of 1289 listed patients (830 men; 459 women) were analyzed during the study period. Main Outcomes and Measures: This study performed survival analysis and competing-risk analysis to delineate how access to livers from living donors was associated with events in women vs men on the transplant waiting list (LT, death, or dropout). Results: Of 1289 included patients, 459 (35.6%) were women, and the mean (SD) age was 56.1 (10.0) years at assessment and listing. A total of 783 of 1289 listed patients underwent LT. Among those with no pLD at assessment, there was a higher median (range) Model for End-stage Liver Disease incorporating sodium levels (MELD-Na) score at listing (22 [6-50] vs 19 [6-50]; P < .001) and at LT (27 [6-49] vs 20 [6-52]; P < .001) in women receiving DDLT. Women were at a significant disadvantage without a pLD (hazard ratio [HR], 1.29; 95% CI, 1.04-1.60; P = .01); there was no difference in access to LT with availability of a pLD (HR, 0.93; 95% CI, 0.76,-1.14; P = .44). The instantaneous rate of receiving a transplant in men with a pLD was 1.39 times higher than men who did not have a pLD (HR, 1.39; 95% CI; P < .001) and the instantaneous rate of receiving a transplant in women with a pLD was 1.92 times higher than in women who did not (HR, 1.92; 95% CI, 1.51-2.44; P < .001). The HR was 1.38 times higher in women compared with men across the MELD-Na score strata (HR, 1.38; 95% CI, 1.03-1.84; P = .03) and 2.04 times higher when the MELD-Na score was less than 20 (HR, 2.04; 95% CI, 1.31-3.14; P = .001). Conclusions and Relevance: These study findings suggest that women can overcome the complex problem of allocation inequity with access to livers from living donors. Women with access only to DDLT were much more unwell than men independent of liver disease at the time of listing, dropout, or LT. Therefore, the wider availability of living donation liver transplant would be helpful in addressing the sex disparity in access to LT in the current MELD-Na era.