BioWaitIndex: a novel machine learning-derived biomarker for improved prediction of liver transplant waitlist outcomes.
8 affiliations
- 1Department of Surgery, Houston Methodist Hospital, Houston, TX, United States.
- 2J. C. Walter Jr. Transplant Center, Houston Methodist Hospital, Houston, TX, United States.
- 3Houston Methodist Academic Institute, Houston, TX, United States.
- 4Department of Surgery, Weill Cornell Medicine, Cornell University, New York, NY, United States.
- 5Department of Medicine, Gastroenterology, Houston Methodist Hospital, Houston, TX, United States.
- 6Department of Medicine, Weill Cornell Medicine, Cornell University, New York, NY, United States.
- 7Department of Anesthesiology and Critical Care, Houston Methodist Hospital, Houston, TX, United States.
- 8Department of Surgery, Immunobiology and Transplant Science Center, Houston Methodist Research Institute, Houston Methodist Hospital, Houston, TX, United States.
Abstract
Liver transplantation remains limited by organ shortage, necessitating precise waitlist risk stratification. The Model for End-Stage Liver Disease (MELD) remains the cornerstone of liver allocation and effectively predicts near-term mortality when assessed dynamically during waitlist follow-up. However, MELD measured at initial transplant evaluation may not fully capture baseline multisystem vulnerability, particularly among candidates with lower MELD scores who may remain on the waitlist for prolonged periods. The BioWaitIndex (BWI) was developed using a random forest model in an NHANES development cohort comprising 13,830 participants. The trained model was first temporally validated in a separate, non-overlapping sample from subsequent NHANES cycles comprising 8,151 participants. It was then independently clinically validated in a separate, non-overlapping, IRB-approved cohort of 1,502 adult liver transplant candidates from Houston Methodist Hospital. In the clinical validation cohort, associations between BWI measured at initial transplant evaluation and subsequent waitlist outcomes were assessed using competing-risk regression. In multivariable competing-risks analyses with BWI and MELD both measured at initial evaluation, each unit increase in BWI was associated with a greater likelihood of transplantation (sHR 1.02, 95% CI 1.01-1.03; p < 0.001) and a lower risk of waitlist death or clinical deterioration (sHR 0.97, 95% CI 0.97-0.98; p < 0.001). Adding BWI to MELD at evaluation improved net reclassification for waitlist mortality by 31.2% (p < 0.001) and for transplantation by 11.1% (p = 0.026), with consistent associations across sexes. BWI may complement dynamic MELD assessment by providing earlier information about baseline multisystem vulnerability and identifying candidates whose risk may not be fully reflected by their MELD score at evaluation.
Related
- To TTV or Not to TTV? A Combined Urinary Biomarker Approach in Kidney Transplant Recipients Using Torque Teno Virus, Cell-Free DNA, and BK Polyomavirus Measurements
- Mixed-Effects Machine Learning improved prediction of Obstetric Anal Sphincter Injury: a nationwide multicenter cohort study.
- Continuous Diagnostic Indices from BanffNET Automated Lesion Scores for Kidney Transplant Pathology: Development and Evaluation of a Diagnostic Prediction Model
- Early Transplant Outcomes and Kidney Donor Profile Index Risk Prediction Performance in Kidneys Recovered Using a Uniform Abdominal Normothermic Regional Perfusion Technique.
- From biomarker discovery to prediction models: clinical translation of multi-omics for early prediction of gestational diabetes mellitus
BibTeX
@article{patel2026biowaitindex,
title = {{BioWaitIndex: a novel machine learning-derived biomarker for improved prediction of liver transplant waitlist outcomes.}},
author = {Khush Patel and Ashton A Connor and Mansi Sanghvi and Ahmed Elaileh and Elsie Rizk and Samar Semaan and Youssef E Dib and Jason Todd and Linda W Moore and Luma Succar and Caroline J Simon and Yee Lee Cheah and Constance M Mobley and Ashish Saharia and Kavish R Patidar and Sudha Kodali and David W Victor and Scott Lindberg and Stefano Casarin and Xian C Li and A Osama Gaber and R Mark Ghobrial},
journal = {Frontiers in transplantation},
year = {2026},
month = oct,
doi = {10.3389/frtra.2026.1964045},
url = {https://europepmc.org/article/MED/42952719},
note = {CC BY},
}
RIS
TY - JOUR
TI - BioWaitIndex: a novel machine learning-derived biomarker for improved prediction of liver transplant waitlist outcomes.
AU - Khush Patel
AU - Ashton A Connor
AU - Mansi Sanghvi
AU - Ahmed Elaileh
AU - Elsie Rizk
AU - Samar Semaan
AU - Youssef E Dib
AU - Jason Todd
AU - Linda W Moore
AU - Luma Succar
AU - Caroline J Simon
AU - Yee Lee Cheah
AU - Constance M Mobley
AU - Ashish Saharia
AU - Kavish R Patidar
AU - Sudha Kodali
AU - David W Victor
AU - Scott Lindberg
AU - Stefano Casarin
AU - Xian C Li
AU - A Osama Gaber
AU - R Mark Ghobrial
PY - 2026
DA - 2026/10/14
JO - Frontiers in transplantation
DO - 10.3389/frtra.2026.1964045
UR - https://europepmc.org/article/MED/42952719
AB - Liver transplantation remains limited by organ shortage, necessitating precise waitlist risk stratification. The Model for End-Stage Liver Disease (MELD) remains the cornerstone of liver allocation and effectively predicts near-term mortality when assessed dynamically during waitlist follow-up. However, MELD measured at initial transplant evaluation may not fully capture baseline multisystem vulnerability, particularly among candidates with lower MELD scores who may remain on the waitlist for prolonged periods. The BioWaitIndex (BWI) was developed using a random forest model in an NHANES development cohort comprising 13,830 participants. The trained model was first temporally validated in a separate, non-overlapping sample from subsequent NHANES cycles comprising 8,151 participants. It was then independently clinically validated in a separate, non-overlapping, IRB-approved cohort of 1,502 adult liver transplant candidates from Houston Methodist Hospital. In the clinical validation cohort, associations between BWI measured at initial transplant evaluation and subsequent waitlist outcomes were assessed using competing-risk regression. In multivariable competing-risks analyses with BWI and MELD both measured at initial evaluation, each unit increase in BWI was associated with a greater likelihood of transplantation (sHR 1.02, 95% CI 1.01-1.03; p < 0.001) and a lower risk of waitlist death or clinical deterioration (sHR 0.97, 95% CI 0.97-0.98; p < 0.001). Adding BWI to MELD at evaluation improved net reclassification for waitlist mortality by 31.2% (p < 0.001) and for transplantation by 11.1% (p = 0.026), with consistent associations across sexes. BWI may complement dynamic MELD assessment by providing earlier information about baseline multisystem vulnerability and identifying candidates whose risk may not be fully reflected by their MELD score at evaluation.
SN - 2813-2440
ER -