Researchers have developed a new, simple scoring system that determined the severity of fibrosis in patients with hepatitis B virus genotype B or C infection and also positive for the hepatitis B e antigen, according to data published in Clinical Gastroenterology and Hepatology.
In a retrospective study, researchers sought to identify any associations between baseline variables with the severity of fibrosis in 710 patients from a phase 3 clinical trial, using multivariate logistic regression analyses. The researchers used significant variables to build two predictive scoring systems (PS1 and PS2) using optimal cut-off values. PS1 analyzed data on HBV genotype and compared genotypes B and C, patient age, level of hepatitis B surface antigen and alanine aminotransferase levels. In contrast, the PS2 scoring system analyzed data on age and level of HBsAg.
Researchers then worked to validate the final model using a similar patient cohort from a phase 4 clinical trial known as NEPTUNE (n = 465).
“Both studies included treatment-naive HBeAg-positive [chronic hepatitis B] patients with a HBV DNA level greater than 500,000 copies/mL or greater than 100,000 IU/mL, an ALT level of 1 to 10 times the upper limit of normal and biopsy-confirmed liver disease consistent with [chronic hepatitis B],” the researchers wrote.
Overall, the PS1 scoring system identified patients with low severity of fibrosis (F0 to F1) vs. moderate to severe patients with fibrosis (F2 to F4) with a specificity greater than 87% and a positive predictive value (PPV) greater than 75, according to the research. Additionally, PS1 identified patients with F0 to F2 to F2 to F4 fibrosis with a specificity of 95% and a PPV of 97%.
The PS1 scoring system was more accurate in identifying patients with F0 to F1 fibrosis with less than PS2. For PS2, the specificity and negative predictive value for F0 to F1 vs. F2 to F4 fibrosis was lower compared with PS1 (85.9% vs. 54.2% for the phase 3 and 88.5% vs. 49.2% for NEPTUNE, respectively). The sensitivity and PPV values for PS2 were also lower compared with PS1 (35.8% vs. 74.2% for the phase 3 and 25.4% vs. 73% for NEPTUNE cohorts, respectively), according to the research. However, the PS2 system did identify patients with F0 to F2 fibrosis with a very similar sensitivity and PPV.
“The use of this prediction score may help guide clinical decision making and potentially reduce the need for invasive liver biopsies in the management of a relevant proportion of patients with HBV genotype B- or C- associated [chronic hepatitis B],” the researchers wrote. “The high specificity of both prediction scores if applied to identify non-severe fibrosis limits the risk of overlooking patients with severe fibrosis. The noninvasive prediction score approach may be of particular benefit in low-resource settings.”
Article: healio.com/hepatology/chron...