Editorial: FibroTest to predict liver-related mortality in NAFLD. Should this change the diagnostic algorithm in NAFLD?

Globally, nonalcoholic fatty liver disease (NAFLD) is widely prevalent, estimated at around 25% in a recent meta-analysis.1 NAFLD is easy to identify based on steatosis on imaging in the absence of significant alcohol intake, but identifying those at highest risk of progression to cirrhosis is more problematic. Degree of liver fibrosis is the single biggest predictor of disease-specific and overall mortality in NAFLD.2 Liver biopsy can identify patients with steatohepatitis and can stage fibrosis, but cannot be considered for the average NAFLD patient and should be reserved for diagnostic dilemmas or those at high risk for advanced fibrosis.

Non-invasive strategies to identify the high risk include clinical prediction rules, blood-based biomarker panels, and radiographic techniques that measure liver “stiffness” as a surrogate for fibrosis. Transient elastography (TE), the NAFLD fibrosis score, and the FIB-4 index are all good tests to predict fibrosis with ROC curves in the 0.84-0.88 range and “ready for prime time” according to recent expert opinion.3 All three of these are relatively easy to obtain or calculate, although TE depends on access to specialised radiology equipment.

FibroTest is a commercially available panel that uses seven parameters including α2-macroglobulin, haptoglobin, and apolipoprotein-A1 and has a high AUROC (0.88) to predict advanced fibrosis in NAFLD with a few caveats of testing interference from other inflammatory diseases including cholestasis, sepsis, and acute inflammation.4 Munteanu et  al recently provided evidence to suggest that FibroTest provides long-term (10 year) prognostic value for mortality in NAFLD populations with a high AUROC of 0.941 compared to 0.875 for the comparator control group of chronic hepatitis C (CHC) patients.5 Fibrotest was best at predicting survival without liver-related deaths, but also had prognostic value for cardiovascular death and, to a lesser extent, overall mortality. As a secondary endpoint, two additional testing panels were evaluated in a smaller group of patients, the NASHTest-2 and the SteatoTest-2. They were also shown to predict survival, although conclusions from this portion of the analysis were limited by low numbers. The data require external validation, particularly given the unexpected association of high haptoglobin and cardiovascular death in NAFLD patients seen in this group. Ten year survival without liver-related death was quite high in the NAFLD population at 0.965 with only 38 events in 1079 patients. These results suggest that FibroTest could be considered as second line after a cheaper screening tool such as FIB-4, BARD, or NAFLD Fibrosis score was used to eliminate those at low risk from further immediate evaluation. FibroTest may then be useful to identify those who should be considered for biopsy, clinical treatment trials, or maximal lifestyle interventions. It should be noted that TE has proven similarly efficacious in a smaller cross-sectional study in predicting prognosis.6 Other biomarker panels such as ELF or the Pro-C3 tests have not been compared head-to-head with TE or FibroTest but may also prove to be efficacious.7 No single test has shown superiority, including in the data presented here. Accuracy, cost, and ease of administration will all be important in determining the future NAFLD risk stratification tool of choice from the current menu of options.

FUNDING INFORMATION

None