This study aims to identify patient characteristics that predict effective
eltrombopag dosage for the treatment of Hepatitis C virus (HCV)-related
thrombocytopenia. Demographic, clinical and genetic data collected from thrombocytopenic patients (n = 1463, age ≥ 18 years) with chronic HCV
infection who were able to achieve a target platelet count of > 90 × 10(9) /L following
eltrombopag treatment. Patients were categorized into four groups (25, 50, 75, and 100 mg) based on the
eltrombopag dose needed to achieve the target platelet count.
Eltrombopag dose predictors were identified using a two stage approach. First, bivariate analysis, using anova for continuous variables and Chi-square test for categorical variables, was performed to identify possible predictors of
eltrombopag dose (P < 0.05). Second, ordinal logistic regression with stepwise addition followed by backward deletion was then performed using predictors identified in bivariate analysis step to produce final model containing independent predictors at P < 0.05. Ordinal logistic model identified several predictors of
eltrombopag dose. Predictors of higher
eltrombopag dose include: having a HCV genotype 2 or 3, being Central/South Asian, being White (Caucasian or European heritage), increased weight, and increased spleen length. Predictors of lower
eltrombopag dose include: female gender, increased age, having a higher ALP plasma concentration, increased
creatinine clearance, increased baseline lymphocytes count, and increased baseline platelet count. In conclusion, this study identified patient characteristics that predict effective
eltrombopag dose for the treatment of HCV-related
thrombocytopenia. Early selection of the optimal
eltrombopag dose expedites the initiation of
antiviral therapy. This is expected to improve the
antiviral therapy outcome before the patient progress into liver decompensation.