RESUMEN
Background and Aims: There is a high unmet need to develop noninvasive tools to identify nonalcoholic fatty liver disease/nonalcoholic steatohepatitis (NAFLD/NASH) patients at risk of fast progression to end-stage liver disease (ESLD). This study describes the development of a machine learning (ML) model using data around the first clinical evidence of NAFLD/NASH to identify patients at risk of future fast progression. Methods: Adult patients with ESLD (cirrhosis or hepatocellular carcinoma) due to NAFLD/NASH were identified in Optum electronic health records (2007-2018 period). Patients were stratified into fast (0.5 and 3 years) and standard progressor (6-10 years) cohorts based on retrospectively established progression time between ESLD and the earliest observable disease, and characteristics were reported using descriptive statistics. Two ML models predicting fast progression were created, performance was compared, and top predictive features from the final model were compared between cohorts. Results: Among a total of 4013 NAFLD patients with cirrhosis or hepatocellular carcinoma (mean age 58.6 ± 12.5; 65% female), 24% were fast (n = 951) and 25% standard (n = 992) progressors that were used for modeling. The cohorts were comparable for gender, body mass index, type 2 diabetes, and arterial hypertension, but differed significantly for obesity, hyperlipidemia, and age at index. The final model (NASH FASTmap) is a 44 feature light gradient boosting model which performed better (area under the curve [0.77], F1-score [0.74], accuracy [0.71], and precision [0.71]) than eXtreme gradient boosting model to predict fast progression. Conclusion: Future fast progression to ESLD in NAFLD/NASH patients can be predicted from clinical data using ML. Electronic health record implementation of NASH FASTmap could support clinical assessment for risk stratification and potentially improve disease management.
RESUMEN
The NASHmap model is a non-invasive tool using 14 variables (features) collected in standard clinical practice to classify patients as probable nonalcoholic steatohepatitis (NASH) or non-NASH, and here we have explored its performance and prediction accuracy. The National Institute of Diabetes and Digestive Kidney Diseases (NIDDK) NAFLD Adult Database and the Optum Electronic Health Record (EHR) were used for patient data. Model performance metrics were calculated from correct and incorrect classifications for 281 NIDDK (biopsy-confirmed NASH and non-NASH, with and without stratification by type 2 diabetes status) and 1,016 Optum (biopsy-confirmed NASH) patients. NASHmap sensitivity in NIDDK is 81%, with a slightly higher sensitivity in T2DM patients (86%) than non-T2DM patients (77%). NIDDK patients misclassified by NASHmap had mean feature values distinct from correctly predicted patients, particularly for aspartate transaminase (AST; 75.88 U/L true positive vs 34.94 U/L false negative), and alanine transaminase (ALT; 104.09 U/L vs 47.99 U/L). Sensitivity was slightly lower in Optum at 72%. In an undiagnosed Optum cohort at risk for NASH (n = 2.9 M), NASHmap predicted 31% of patients as NASH. This predicted NASH group had AST and ALT mean levels above normal range of 0-35 U/L, and 87% had HbA1C levels > 5.7%. Overall, NASHmap demonstrates good sensitivity in predicting NASH status in both datasets, and NASH patients misclassified as non-NASH by NASHmap have clinical profiles closer to non-NASH patients.
Asunto(s)
Diabetes Mellitus Tipo 2 , Enfermedad del Hígado Graso no Alcohólico , Adulto , Humanos , Enfermedad del Hígado Graso no Alcohólico/diagnóstico , Biopsia , Alanina Transaminasa , HígadoRESUMEN
OBJECTIVE: To develop a computer model to predict patients with nonalcoholic steatohepatitis (NASH) using machine learning (ML). MATERIALS AND METHODS: This retrospective study utilized two databases: a) the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) nonalcoholic fatty liver disease (NAFLD) adult database (2004-2009), and b) the Optum® de-identified Electronic Health Record dataset (2007-2018), a real-world dataset representative of common electronic health records in the United States. We developed an ML model to predict NASH, using confirmed NASH and non-NASH based on liver histology results in the NIDDK dataset to train the model. RESULTS: Models were trained and tested on NIDDK NAFLD data (704 patients) and the best-performing models evaluated on Optum data (~3,000,000 patients). An eXtreme Gradient Boosting model (XGBoost) consisting of 14 features exhibited high performance as measured by area under the curve (0.82), sensitivity (81%), and precision (81%) in predicting NASH. Slightly reduced performance was observed with an abbreviated feature set of 5 variables (0.79, 80%, 80%, respectively). The full model demonstrated good performance (AUC 0.76) to predict NASH in Optum data. DISCUSSION: The proposed model, named NASHmap, is the first ML model developed with confirmed NASH and non-NASH cases as determined through liver biopsy and validated on a large, real-world patient dataset. Both the 14 and 5-feature versions exhibit high performance. CONCLUSION: The NASHmap model is a convenient and high performing tool that could be used to identify patients likely to have NASH in clinical settings, allowing better patient management and optimal allocation of clinical resources.