Background:  Orosomucoid (ORM) is a major binding protein for protease inhibitors such as amprenavir (APV). The concentration and the polymorphism of ORM could be responsible for interindividual variation in the plasma binding of APV, which might influence the pharmacokinetics and virological efficacy of APV. The objective was to describe the ORM polymorphism in HIV-infected patients and to determine its possible influence on the APV pharmacokinetics.

Methods:  The study was conducted in a subgroup of 13 HIV-1-infected patients (11 males, 47 [32 to 53] years old) included in the Puzzle 1-ANRS 104 trial and treated by ritonavir (RTV) -boosted APV and NRTI. The phenotypes of ORM variants was determined in sera after desialylation by isoelectric focusing on polyacrylamide gels following by immunoblotting. ORM concentrations were assayed at week 2. APV total (APVt) and unbound plasma concentrations (APVu) were measured at week 2 prior to drug intake (3 patients) and during the dosing interval (10 patients). The unbound fraction was the ratio of APVu and APVt. APV total and unbound apparent clearance (CLt/F, Clu/F) and area under the curve (AUCt, AUCu) were determined using a non-compartmental approach. Relations between ORM phenotypes, ORM concentrations, and APV parameters were tested using an ANOVA test.

Results:  Median ORM concentrations were 1.0 g/L (0.4 to 1.8). Of these patients, 3 had the phenotype ORM1*S-ORM2*A, 6 were ORM1*F1S-ORM2*A, and 4 were ORM1*F1-ORM2*A. The allele ORM1*F2 was not observed. All patients expressed a monomorphic allele on ORM2 locus. Median APVt and APVu were 1648 ng/mL (648 to 5353) and 152 ng/mL (65 to 408), respectively and were not correlated to ORM concentrations or phenotypes. ORM concentrations were not found to be significantly dependant on ORM phenotypes. Mean APV unbound fraction was significantly higher (p = 0.037) in patients presenting the ORM1*S-ORM2*A phenotype than the ORM1*F1S-ORM2*A or the ORM1*F1-ORM2*A (0.12 vs 0.10 and 0.07). Consequently there was a significant increase in mean APV Clu/F (584 L/h vs 314 and 196; p = 0.029) and APV Clt/F (p = 0.001, 50.4 L/h vs 19.8 and 11.6) leading to a decrease in APV AUCt (p = 0.050, 7937ng/mL·h vs 21,758 and 37,391) without significant difference in APV AUCu.

Conclusions:  The different ORM variants affect APV pharmacokinetics parameters. This could be the consequence of different capacities for APV binding, which suggest that the variant ORM1*S has a lower affinity for APV than the variant ORM1*F1. Further studies are warranted to evaluate the impact on virological efficacy.