Genetic, metabolic score stratifies risk for adverse events in fatty liver

Using genetic markers with metabolic risk factors can predict an individual patient’s risk for developing liver related events due to non-alcoholic fatty liver disease, according to a researcher at The Liver Meeting Digital Experience.

“In NAFLD patients with significant fibrosis, the genetic metabolic staging score combining demographic and metabolic features with common genetic variants accurately predicts the risk for liver-related events,” Grazia Pennisi, PhD student at the University of Palermo, Italy, said in her presentation.

Pennisi presented the Genetic Metabolic Staging (GEMS) score, which combines the presence of metabolic risk factors such as obesity, diabetes, arterial hypertension and low HDL with the presence of high-risk alleles and stage of fibrosis. Each factor carries a score of 0 to 2 for a total possible score of 8. Researchers evaluated events – defined as HCC or hepatic decompensation – through Kaplan-Meier curves and Cox regression analysis to identify baseline variables most associated with

The study looked at 546 consecutive patients with histological diagnosis of NAFLD or NAFLD with cirrhosis who investigators then followed for more than 6 months. In the group in which FIB-4 was less than 1.3, Pennisi showed there was only one liver-related event. But in the group in which FIB-4 was 1.3 or greater, there were 57 liver-related events, so the GEMS score was used in that group with more fibrosis (n = 229).

GEMS score correlated with likelihood of events (AUROC = 0.837; P < .001), Pennisi said. Over a median 71 months, 60 events occurred (55 participants decompensated and five developed HCC) and age greater than 50 years (HR=21.3; 95% CI, 2.83-160.2), platelets less than 110,000 mm³ (HR=5.76; 95% CI, 3.17-10.4) and GEMS score as a continuous variable (HR=1.46; 95% CI, 1.21-1.77) were all independently associated with outcomes.

In looking at the GEMS score by categories of 0 to 3, 4 to 6 and 7 to 8, Pennisi said it was more strongly associated with risk prediction (HR=2.83; 95% CI, 1.74-4.6). In the 0 to 5 group, Pennisi showed a 4% event occurrence while those in the 8 to 10 group had a 91% rate of liver-related events.

“When externally validated, genetic-metabolic scoring can help physicians to stratify the clinical course of NAFLD patients at individual patient level,” Pennisi said.