Blood test can spot patients with NASH at risk for progressive liver disease

Last Updated: 2020-08-21

By Reuters Staff

NEW YORK (Reuters Health) - A blood test can identify patients with non-alcoholic steatohepatitis (NASH) and significant fibrosis who are at risk for progressive liver disease, new research indicates.

The test, dubbed NIS4, is being developed by Genfit, which funded the study. It uses four blood-based biomarkers of NASH (miR-34a-5p, A2M, YKL-40, and HbA1c) and an algorithm to produce a score to identify patients with or without "at-risk" NASH - defined as non-alcoholic fatty liver disease (NAFLD) activity score 4 or higher and fibrosis stage 2 or higher.

The researchers derived and validated the NIS4 using blood samples, clinical data and liver-biopsy results from three independent cohorts of patients with suspected NAFLD. The discovery cohort included 239 patients and the two validation cohorts included a total of 702 patients.

The mean age across the cohorts ranged from 52 to 55 years; obesity, diabetes, hypertension and dyslipidemia were common. The prevalence of at-risk NASH was 44% in the discovery cohort and 49% in the pooled validation cohorts.

"NIS4 consistently showed high diagnostic performance and low misclassification rates to rule in and rule out at-risk NASH compared with other diagnostic approaches," Dr. Arun Sanyal of Virginia Commonwealth University School of Medicine in Richmond and colleagues report in The Lancet Gastroenterology and Hepatology.

In the discovery set, the area under the receiver operating characteristic curve (AUROC) for NIS4 was 0.80 (95% confidence interval, 0.73 to 0. 85) and did not require adjustment for age, sex, BMI or aminotransferase levels.

For "real-world" use, the researchers say a lower cutoff of less than 0.36 provides a rule-out decision with 80.8% sensitivity, 65.2% specificity and 81.5% negative predictive value; an upper cutoff of 0.63 or higher provides a rule-in decision with 90.4% specificity, 45.2% sensitivity, and 78.3% positive predictive value.

The NIS4 algorithm performed equally well in the validation cohorts. In the pooled validation cohort, a NIS4 value less than 0.36 ruled out at-risk NASH with 81.5% sensitivity and 63.0% specificity, with a negative predictive value of 77.9%; a NIS4 value of more than 0.63 ruled in at-risk NASH with 87.1% specificity and 50.7% sensitivity, with a positive predictive value of 79.2%.

NASH is the most severe form of NAFLD and is estimated to affect nearly 80 million people in the United States. The vast majority of cases remain undetected because liver biopsy is currently the only way to get conclusive information about NASH activity and fibrosis, the authors note.

"There is a major unmet need to establish simple tools that can be used in all patients with risk factors to identify those with non-alcoholic steatohepatitis with enough disease activity and fibrosis to be considered for more intense therapies over and above lifestyle changes," Dr. Sanyal said in a statement from Genfit.

"The current study is a key step in meeting this unmet need and demonstrates that the NIS4 score can be used to enrich the likelihood of identifying this subpopulation amongst those who are overweight or are obese with or without diabetes. This paves the way for development of patient identification and treatment paradigms that are non-invasive and available to all clinics," Dr. Sanyal said.

The study was funded by Genfit. Several authors, including Dr. Sanyal, have financial relationships with the company.

SOURCE: https://bit.ly/3hebr3u Lancet Gastroenterology and Hepatology, online August 4, 2020.