Blood test differentiates between NAFLD, NASH
Click Here to Manage Email Alerts
A novel combinations of glycans, lipids and hormonal variables can simultaneously diagnose the presence of nonalcoholic fatty liver or nonalcoholic steatohepatitis with up to 90% accuracy, according to findings published in Metabolism Clinical and Experimental.
“Today, physicians assess liver health with liver biopsy — a costly method that carries morbidity and mortality risks for the patient and is not always representative of the actual status of the entire liver,” Christos S. Mantzoros, MD, DSc, director of the human nutrition unit at Beth Israel Deaconess Medical Center, told Healio. “Moreover, it would be impossible to do biopsies in the 80 million at-risk Americans, and even if we did, it would result in tens of thousands of individuals experiencing complications and about 16,000 deaths each year from complications. Finding easy-to-obtain, relatively inexpensive and reliable biomarkers, which can be measured with less invasive techniques, is an urgent, unmet clinical need.”
Mantzoros and colleagues performed a lipidomic, glycomic and free fatty acid analysis using fasting serum samples from 49 healthy adults and 31 individuals with biopsy-proven nonalcoholic fatty liver disease (NAFLD; 15 with NAFL; 16 with nonalcoholic steatohepatitis [NASH]). Researchers analyzed participant data combined with measurements of four hormonal parameters using two different platforms and five machine learning tools.
“We measured as many circulating molecules as reasonably possible and then let machine learning and artificial intelligence pick the best sets of molecules that would most accurately predict outcomes,” Mantzoros said in a press release. “Although the number of subjects appears small given conventional study designs, employing powerful and novel artificial intelligence models allowed us to derive accurate results, as high as 98% in some cases. These models may serve as low-risk, cost-effective alternative method to liver biopsy for diagnosis and monitoring NALFD.”
Using mass-spectrometry and liquid chromatography, researchers identified 365 lipids, 61 glycans and 23 fatty acids and assessed differences in blood concentrations among individuals with and without liver disease.
The researchers noted “robust differences” in the concentrations of specific lipid species observed between healthy adults and those with NAFL and NASH. Individuals with NAFL and NASH had increased concentrations of diglycerides, phosphatidylglycerols and phosphatidic acids, whereas the concentrations of acylcarnitines, cholesterol esters, coenzyme Q10, lysophosphatidylcholines and sphingomyelines were reduced when compared with healthy adults. Results persisted after excluding healthy, normal-weight controls.
“One-vs-Rest” support vector machine (SVM) models with recursive feature elimination including 29 lipids or combining lipids with glycans and/or hormones (20 or 10 variables total) could differentiate with between participants with NAFL, NASH or no disease with up to 90% accuracy, according to the researchers.
“Our predictive models are able to diagnose between three (and not two) conditions, ie, healthy status, NAFL and NASH,” the researchers wrote. “Thus, they allow the evaluation of a subject with a single blood draw without the need of an imaging modality for diagnosing NAFL. This concept is especially attractive for routine screening of the population at high risk for the development of the disease, such as obese patients or patients with type 2 diabetes, in a primary care setting and without the need of specially trained personnel.” – by Regina Schaffer
For more information:
Christos S. Mantzoros, MD, DSc, can be reached at Beth Israel Deaconess Medical Center, 330 Brookline Ave., East Campus, Boston, MA 02215; email: cmantzor@bidmc.harvard.edu.
Disclosure: Mantzoros reports he serves as a consultant for Ansh Labs, Intarcia and Novo Nordisk.
Read more about
Click Here to Manage Email Alerts