This article is more than 5 years old. Information may no longer be current.

Misclassified fibrosis stages common among patients with chronic HCV

Multiple serum biomarkers, measured by various methods, revealed high misclassification rates when determining fibrosis stage accuracy among patients with chronic hepatitis C virus, according to study results.

Keyur Patel, MD, of the Duke Clinical Research Institute, and colleagues used multiple cohorts for the study. There were 383 treatment-naive patients (median age, 44 years; 69% men) with chronic hepatitis C (CHC) virus in the training set, 100 controls without CHC and 434 treatment-naive patients (median age, 40 years; 68% men) with CHC from the Trent HCV cohort used for a validation analysis.

Keyur Patel

Thirty-seven serum biomarkers were used to determine whether cytokines or extracellular matrix proteins could determine fibrosis stage in the CHC patients. The biomarkers were evaluated through the SearchLight multiplex platform (n=34) and ELISA testing (n=3). Sixty-one percent of the training set had minimal stage fibrosis (F0-F1), and 79% of the validation cohort had minimal to mild fibrosis (FO-F2).

The overall misclassification rate in the training cohort was 38%, with a penalized Obuchowski (OB) measure of 0.85; the validation cohort had a misclassification rate of 29%, with a penalized OV of 0.89. Misclassification rates were greatest among patients with stages F2 and F3 fibrosis in the training (78%) and validation (70%) sets and those patients primarily were misclassified as having stage F0 to F1 (92% and 90%, respectively).

Using FibroSure, the misclassification rate among the training cohort was 50%, with an OB of 0.82. Researchers also determined that the change in fibrosis stage was not associated with a change in the FibroSure index (P=.62), but trends were associated with change in Histologic Activity Index and an increase in ActiTest (P=.0003) and FibroSure (P=.006).

Age and serum markers for hyaluronic acid levels, VCAN1, A2M and RBP4 were associated with fibrosis stage. Area under the curve values for the markers among the training set patients were similar to FibroSure test results: 0.51 vs. 0.53 (F0 vs. F1); 0.6 vs. 0.59 (F1 vs. F2); 0.69 vs. 0.72 (F2 vs. F3); and 0.51 vs. 0.52 (F3 vs. F4).

“Accurate and reliable differentiation of adjacent fibrosis stages with serum markers alone is going to be difficult, and even our study with a large number of patients and a complex array of proteins was not able to significantly improve upon more simple serum marker panels [such as FibroSure] in their diagnostic ability to differentiate adjacent fibrosis stages in chronic hepatitis C infection,” Patel told Healio.com/Hepatology.

Disclosure: See the study for a full list of relevant financial disclosures.