TOPLINE:
A machine learning (ML) model using standard laboratory values and clinical data can help predict hepatocellular carcinoma (HCC) risk in patients with metabolic dysfunction-associated steatotic liver disease (MASLD) with high specificity and sensitivity, a new study has found.
METHODOLOGY:
Because the incidence of HCC is increasing and is correlated with MASLD, even in patients without advanced liver fibrosis, researchers developed an ML model that aims to estimate HCC risk for patients with MASLD at any stage of the disease.Data from 1561 patients with MASLD were used to develop and train the pilot predictive model, which was subsequently validated in an independent dataset of 686 patients with MASLD. The patients received care at one of two academic medical centers between 2010 and 2021.Researchers used International Classification of Diseases (ICD) 9/10 Clinical Modification (CM) codes to identify patients with MASLD and HCC in the two institutions’ electronic medical record (EMR) databases.
TAKEAWAY:
HCC developed in 227 patients (14%) in the training cohort and 176 patients (25%) in the validation cohort.Liver fibrosis determined by the noninvasive fibrosis-4 score was the strongest predictive parameter of HCC in the model. The other top predictive parameters were total cholesterol, alkaline phosphate, bilirubin, and hypertension.In the validation cohort, the model combining all four parameters predicted HCC development with 92.06% accuracy, with an area under the curve of 0.97, sensitivity of 74.41%, and specificity of 98.34%.
IN PRACTICE:
“We envision the model will be applicable in a clinical setting as a point-of-care tool as well as for population-level triaging. The tool can be configured to automatically generate a risk prediction score with requisite data from the EMR. The availability of such a score can help providers and patients effectively discuss screening strategies and institute modifiable measures to mitigate risks for the development of HCC,” the authors wrote.
SOURCE:
The study, with first author Souvik Sarkar, MD, University of California Davis, Sacramento, California, was published online on January 22, 2024, in Gastro Hep Advances.
LIMITATIONS:
Although data from two independent cohorts were used, the size of the cohorts was relatively small. The study relied on ICD CM (diagnosis) codes. The model did not incorporate liver biopsy or imaging findings.
DISCLOSURES:
The study had no funding. The authors disclosed no conflicts of interest.
>>> Read full article>>>
Copyright for syndicated content belongs to the linked Source : Medscape – https://www.medscape.com/viewarticle/machine-learning-helps-predict-hcc-risk-masld-2024a10002lj
