Background:  Ribavirin (RBV) concentration was shown to correlate with hemoglobin decrease and overall early virological response in HIV⁺/hepatitis C virus (HCV)⁺ patients. The relationship between early virological response and RBV plasma exposure according to HCV genotype has not been yet analyzed. Therefore, our aim was to study the effect of RBV trough concentration (C_trough) on both early virological response according to HCV genotype and on early hemoglobin decrease.

Methods:  HIV⁺/HCV⁺ patients placed on association RBV + pegylated interferon (PegIFN)-2a or -2b in 2004 were prospectively evaluated. Qualitative and quantitative HCV RNA, hemoglobin levels, and RBV C_trough were measured at baseline and week 2, 4, 8, and 12. HCV genotype was determined at baseline. Early virological response was considered as a negative qualitative HCV RNA or quantitative HCV RNA decrease >2 log at week 12. Linear and logistic regression analyses were used as needed. RBV effective and toxic concentration were considered as the values associated with 50% or 90% probability of detecting early virological response or hemoglobin decrease by logistic regression. Values were given as median (IQR).

Results:  We included 41 consecutive patients, of whom 21 (51.21%) had genotype 1 or 4. PegIFN-2a (180 mg) and -2b (1.5 mg/kg) were used in 31 and 10 patients, respectively. RBV weight-adjusted dose was 12.3 mg/kg (11.5 to 13.3). HCV RNA and hemoglobin at baseline were 6.3 log (5.94 to 6.64) and 15 g/dL (14 to 16), respectively. Overall, 31 patients (75.6%) showed early virological response, with an HCV RNA decrease of 5.44 log (–6.3, –2.06). Early virological response was observed in 20 of 20 patients with genotype 3 as compared with 11 of 21 patients with genotype 1 or 4 (c² = 12.59, p <0.0001). Overall, no correlation between early virological response and RBV C_trough was found. Nevertheless, RBV C_trough was an independent predictor of early virological response in subgroup of patients with genotype 1 or 4 (p = 0.039). In the latter, RBV effective concentration 50% and effective concentration 90% were 1600 ng/mL and 2500 ng/mL, respectively. Overall, maximum hemoglobin decrease was 2.7 g/dL [–3.65, –4.5], and the lowest hemoglobin value reached was 12.4 g/dL (11.4 to 13.3). Mean RBV C_trough correlated with maximum hemoglobin decrease (R = –0.358, p = 0.02). Moreover, maximum hemoglobin decrease >3 g/dL was predicted by higher RBV C_trough at logistic regression analysis (p = 0.01). Toxic concentration 50% and toxic concentration 90% for maximum hemoglobin decrease >3 g/dL were 1700 ng/mL and 3000 ng/mL, respectively.

Conclusions:  Our study confirmed an overall relationship between hemoglobin decrease and RBV exposure, while, as opposite to previous reports, showed a C_trough-related early virological response of RBV only in patients with genotype 1 or 4. Further studies are warranted in order to define the role of therapeutic drug-level monitoring as a tool to optimize RBV efficacy or tolerability according to the HCV genotype.