Background: Functional variants of inosine triphosphatase (ITPA) were recently found to protect against ribavirin (RBV)-induced hemolytic anemia. However, no definitive data are yet available on the role of plasma RBV concentrations on hemoglobin (Hb) decrement. Moreover, no data have been published on the possible interplay between these 2 factors.
Methods: A retrospective analysis included 167 patients. The ITPA variants rs7270101 and rs1127354 were genotyped and tested using the χ2 test for association with Hb reduction at week 4. We also investigated, using multivariate logistic regression, the impact of RBV plasma exposure on Hb concentrations.
Results: Both single nucleotide polymorphisms were associated with Hb decrease. The carrier of at least 1 variant allele in the functional ITPA single nucleotide polymorphisms was associated with a lower decrement of Hb (−1.1 g/dL), as compared with patients without a variant allele (−2.75 g/dL; P = 4.09 × 10−8). RBV concentrations were not influenced by ITPA genotypes. A cut-off of 2.3 μg/mL of RBV was found to be associated with anemia (area-under-receiver operating characteristic = 0.630, sensitivity = 50.0%, and specificity = 69.5%, P = 0.008). In multivariate logistic regression analyses, the carrier of a variant allele (P = 0.005) and plasma RBV concentrations <2.3 μg/mL (P = 0.016) were independently associated with protection against clinically significant anemia at week 4.
Conclusions: Although no direct relationship was found between ITPA polymorphisms and plasma RBV concentrations, both factors were shown to be significantly associated with anemia. A multivariate regression model based on ITPA genetic polymorphisms and RBV trough concentration was developed for predicting the risk of anemia. By relying upon these 2 variables, an individualized management of anemia seems to be feasible in recipients of pegylated interferon–RBV therapy.
*Department of Infectious Diseases, University of Turin, Amedeo di Savoia Hospital
†Department of Gastroenterology, S. Giovanni Battista (Molinette) Hospital, Turin, Italy.
This study was not supported.
The authors declare no conflict of interest.
Correspondence: Antonio D'Avolio, BSc, MSc, SM, Laboratory of Clinical Pharmacology and Pharmacogenetics, Padiglione Q, Ospedale, Amedeo di Savoia, C.so Svizzera 164, 10149 Torino, Italy (e-mail: email@example.com).
Received September 27, 2011
Accepted January 19, 2012