THE REDUCED ACCURACY OF NON-INVASIVE TESTS FOR SIGNIFICANT FIBROSIS IN CHB PATIENTS WITH MASLD

Background: The new nomenclature of metabolic dysfunction-associated steatotic liver disease (MASLD) has broaden its definition to encompass populations with viral hepatitis. However, the interplay between CHB and MASLD and its implications for non-invasive testings (NITs) accuracy remains under-explored. In this study, we determined the diagnostic accuracy of NITs for significant fibrosis leveraging a large cohort of CHB patients with MASLD.
Methods: Included from eleven medical centers, 1079 adult CHB patients with liver biopsy were categorized into three groups, (A) with hepatic steatosis (HS) only, (B) with MASLD involving 1-3 cardiometabolic risk factors (CMRFs), and (C) with MASLD involving 4-5 CMRFs. Significant fibrosis was defined as Scheuer's classification ≥2. The corresponding optimal cutoff values for three non-invasive tests (NITs), namely the fibrosis-4 index (FIB-4), aspartate aminotransferase-to-platelet ratio index (APRI), and non-alcoholic fatty liver disease fibrosis score (NFS), were determined to diagnose significant fibrosis (≥2) based on CHB combined with HS popultion(group A). The performance of these NITs was evaluated for area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value, and negative predictive value across three groups. The AUCs of these NITs in each group were compared using the DeLong test.
Results: The AUC for diagnosing significant fibrosis (≥2) was 0.73 in group A, using an optimal cutoff value of 1.26. However, the AUCs were significantly lower in group B (0.62) and group C (0.57) (Figure 1A). For APRI, the AUCs in the three groups were 0.75, 0.62, and 0.63, respectively, based on an optimal cutoff value of 0.56. The NFS exhibited more modest performance, with AUCs of 0.67, 0.55, and 0.54 in group A, B, and C, respectively (Figure 1B, 1C). The diagnostic accuracy of the three NITs was all falling below 0.74, with the lowest accuracy observed for FIB-4 at 0.57 in group C. For APRI, the lowest accuracy was 0.63 in both group B and group C. NFS had the lowest accuracy of 0.52 in group B (Table 1). The overall diagnostic accuracy showed a downward trend with the increased number of CMRFs.
Conclusions: We revealed a notable reduction in diagnostic accuracy of three widely-used NITs for significant fibrosis in patients with concomitant MASLD and CHB. Particularly pronounced in patients with an elevated count of CMRFs, this diagnostic gap underscores an urgent need for the development of more refined NIT methodologies. This is especially critical in regions where CHB prevalence is high, signaling a potential shift in clinical strategies for managing these complex patient profiles.