New liver stiffness (LS) thresholds offer accurate prediction of disease progression and clinical outcomes in patients with nonalcoholic steatohepatitis (NASH) and advanced fibrosis, according to investigators.
These new LS thresholds are more reliable because they are based on high-quality prospective data drawn from four randomized controlled trials, reported lead author Rohit Loomba, MD, of the University of California, San Diego, and colleagues.
“Retrospective studies report that increasing baseline LS by VCTE [vibration-controlled transient elastography] is associated with the risk of disease progression in patients with NAFLD [non-alcoholic fatty liver disease], but prospective data in well-characterized NASH cohorts with biopsy-confirmed advanced fibrosis are limited,” the investigators wrote in Gut. “The optimal LS thresholds for prognostication of fibrosis progression and decompensation are unknown.”
Seeking clarity, Dr. Loomba and colleagues leveraged data from two phase 3 placebo-controlled trials for selonsertib and two phase 2b placebo-controlled trials for simtuzumab.
“While the studies were discontinued prematurely due to lack of efficacy, the prospectively collected data in these well-characterized participants with serial liver biopsies provides a unique opportunity to study the association between baseline LS by VCTE and disease progression,” the investigators wrote.
Across all four studies, bridging fibrosis (F3) was present in 664 participants, while 734 individuals had cirrhosis (F4). In the selonsertib studies, fibrosis was staged at baseline and week 48. The simtuzumab studies measured liver fibrosis at baseline and week 96. Out of the 664 participants with bridging fibrosis, 103 (16%) progressed to cirrhosis. Among the 734 patients with cirrhosis, 27 (4%) experienced liver-related events. Comparing these outcomes with LS data at baseline and throughout the study revealed optimal LS thresholds.
The best threshold for predicting progression from bridging fibrosis to cirrhosis was 16.6 kPa. According to the authors, the sensitivity, specificity, positive predictive value, and negative predictive value of this threshold for progression to cirrhosis were 58%, 76%, 31%, and 91%, respectively. Among patients at or above 16.6 kPa, 31% progressed to cirrhosis, compared with 9.1% of those under that threshold. Furthermore, individuals with a baseline LS at or above 16.6 kPa had nearly four times greater risk of developing cirrhosis (adjusted hazard ratio, 3.99; 95% CI, 2.66-5.98; P < .0001).
For patients with cirrhosis at baseline, the optimal threshold for predicting liver-related events, such as ascites, hepatic encephalopathy, and portal hypertension–related GI bleeding, liver transplantation, or mortality, was 30.7 kPa. The sensitivity, specificity, PPV, and NPV of this threshold for liver-related events were 62%, 87%, 10%, and 99%, respectively, according to the authors. Patients with an LS above this mark were 10 times as likely to experience liver-related events (aHR, 10.13; 95% CI, 4.38-23.41; P < .0001).
Scott L. Friedman, MD, chief of the division of liver diseases and dean for Therapeutic Discovery at the Icahn School of Medicine at Mount Sinai, New York, called the study “an important effort” that offers valuable insights for both researchers and practitioners.
“For clinical trials, [these thresholds] really allow for greater refinement or enrichment of patients who are suitable for enrollment in the trial because they’re at a higher risk of clinical problems that might be mitigated if the drug is effective,” Dr. Friedman said in an interview. “For clinical practice, it might indicate that the patient should either be fast tracked for a clinical trial or, more importantly, maybe needs to be referred for evaluation for a liver transplant. It may also indicate – although they didn’t look at it in this study – that there’s a need to begin or accelerate screening for liver cancer, which becomes an encroaching risk as the fibrosis advances to later stages.”
The study was funded by Gilead Sciences. The investigators disclosed additional relationships with Amgen, Eli Lilly, CohBar, and others. Dr. Friedman reported no relevant conflicts of interest.