HIV Articles  
Back 
 
 
Risk factors (AZT) for anaemia in human immunodeficiency virus/hepatitis C virus-coinfected patients treated with interferon plus ribavirin
 
 
  Journal of Viral Hepatitis
Volume 14 Issue 9 Page 639-644, September 2007
 
Bani-Sadr11Groupe Hospitalier Universitaire Est, Universite Paris, INSERM U707, Paris,
I. Goderel11Groupe Hospitalier Universitaire Est, Universite Paris, INSERM U707, Paris,
C. Penalba22Hopital Corvisart, Charleville Mezieres,
E. Billaud33Hopital Hotel Dieu, Nantes,
J. Doll44Hopital Andre MIGNOT, Le Chesnay,
Y. Welker55Centre Hospitalier de Saint Germain en Laye,
P. Cacoub66Groupe Hospitalier Universitaire Est, Universite Paris, Paris,
S. Pol77Groupe Hospitalier Universitaire Ouest, Universite Paris, INSERM U370, Paris,
C. Perronne88Centre Hospitalier Universitaire Raymond Poincare, Universite de Versailles, Garches, France,
F. Carrat11Groupe Hospitalier Universitaire Est, Universite Paris, INSERM U707, Paris and
the ANRS HC02 - Ribavic Study team
1Groupe Hospitalier Universitaire Est, Universite Paris, INSERM U707, Paris; 2Hopital Corvisart, Charleville Mezieres; 3Hopital Hotel Dieu, Nantes; 4Hopital Andre MIGNOT, Le Chesnay; 5Centre Hospitalier de Saint Germain en Laye; 6Groupe Hospitalier Universitaire Est, Universite Paris, Paris; 7Groupe Hospitalier Universitaire Ouest, Universite Paris, INSERM U370, Paris; and 8Centre Hospitalier Universitaire Raymond Poincare, Universite de Versailles, Garches, France
 
"...zidovudine discontinuation prior to HCV therapy, particularly when alternative antiretroviral options are available could help to avoid anaemia and thereby permit the use of higher ribavirin doses... In this study of 383 HIV/HCV-coinfected patients, despite the use of a lower ribavirin starting dose (800 mg/day), 15.9% of patients developed anaemia, defined as a haemoglobin concentration below 10 g/dL. A lower baseline haemoglobin level, zidovudine therapy and peg-IFN therapy were associated with a significantly increased risk of anaemia. In contrast, PI-based antiretroviral therapy roughly halved the risk...."
 
Summary. The most frequent and the most troublesome adverse effect of interferon plus ribavirin-based therapy is anaemia. The aim of this analysis was to determine the incidence and risk factors of anaemia (Hb < 10 g/dL) in human immunodeficiency virus/hepatitis C virus (HCV)-coinfected patients receiving anti-HCV therapy. We reviewed all cases of anaemia occurring among 416 patients participating in a randomized, controlled 48-week trial comparing peginterferon (peg-IFN) alpha 2b plus ribavirin with interferon alpha-2b plus ribavirin. Univariate and multivariate analyses were used to identify links with antiretroviral treatments, HCV therapy and clinical and laboratory findings.
 
Sixty-one (15.9%) of the 383 patients who received at least one dose of anti-HCV treatment developed anaemia.
 
In multivariate analysis the risk of anaemia was significantly associated with zidovudine (OR, 3.27 95% CI, 1.64-6.54, P = 0.0008) and peg-IFN (OR, 2.35; 95% CI, 1.16-4.57, P = 0.0179).
 
The risk of anaemia was lower in patients with higher baseline haemoglobin levels (OR, 0.35 95% CI, 0.26-0.49, P < 0.0001) and in patients receiving protease inhibitor-based antiretroviral therapy (OR, 0.51 95% CI, 0.30-0.86, P = 0.0114). Zidovudine discontinuation could help to avoid anaemia associated with anti-HCV therapy.
 
Introduction

Combination therapy with peginterferon (peg-IFN) and ribavirin is currently the standard of care for patients with hepatitis C virus (HCV) infection [1,2]. The most frequent and the most troublesome adverse effect of this combination is anaemia, due to both haemolysis induced by ribavirin and to bone marrow suppression by interferon [3]. As full adherence to HCV therapy is crucial for virological success, we attempted to identify risk factors for anaemia in human immunodeficiency virus (HIV)/HCV-coinfected patients participating in a large randomized-controlled trial comparing pegylated interferon alpha-2b plus ribavirin with conventional interferon alpha-2b plus ribavirin (the ANRS HC02 RIBAVIC study) [1].
 
Results
At least one dose of study medication was received by 383 patients. Three-quarters of the patients were men, and the mean age was 39.7 ± 5.4 years. Intravenous narcotic use was the risk factor for HCV infection in 79% of cases. Three hundred and seventeen patients (83%) were receiving antiretroviral therapy at enrolment. At baseline, the mean haemoglobin level was 14.5 ± 1.4 g/dL. Men had a higher mean haemoglobin concentration than women (14.8 ± 1.3 vs 13.6 ± 1.2 g/dL). Anaemia occurred in 61 patients (15.9%), after <12 weeks of treatment in 42 cases (68.8%). The median time between the outset of anti-HCV therapy and the onset of anaemia was 4 weeks (range, 2-48 weeks). The mean duration of anti-HCV treatment was slightly but not significantly shorter in patients who developed anaemia than in the other patients (247 ± 116 vs 268 ± 104 days; P = 0.067). The rates of sustained virological responses were not significantly different between the groups (19.6% (n = 12/61) vs 26.4% (n = 85/322); P = 0.33). The incidence of ribavirin discontinuation for all causes was similar in the two groups [n = 29/61 (47.5%) vsn = 119/322 (36.9%); P = 0.15], but the incidence of anaemia-related ribavirin discontinuation was higher in the first group (five patients and one patient; P < 0.0001).
 
Univariate analysis showed that anaemia was significantly associated with the following baseline variables: a lower BMI, a higher daily ribavirin dose (per kg body weight), a lower haemoglobin level, lower leucocyte and neutrophil counts, lower alanine aminotransferase levels and lower creatinine clearance, as estimated with the Cockcroft-Gault formula (CrCL) (Table 1). Anaemia was more likely to occur in the peg-IFN alpha-2b arm than in the interferon alpha-2b arm (63.9%vs 48.1%; P = 0.0258).
 
The possible role of nucleoside reverse transcriptase inhibitors (NRTI), non-nucleoside reverse transcriptase inhibitors (NNRTI) and protease inhibitors (PIs) used at the outset of anti-HCV treatment was also examined. Among the NRTIs, only zidovudine was significantly associated with an increased risk of anaemia (P < 0.0001). No association was observed with NNRTI-containing regimens. Patients who were taking a PI-containing regimen were less likely to develop anaemia (27.8%vs 42.5%, P = 0.0334).
 
The following factors were not associated with the risk of anaemia: age, sex, mean duration of HIV and HCV infection, AIDS status, baseline CD4 cell count, HIV viral load below 400 copies/mL, ongoing antiretroviral therapy, the mean duration of previous antiretroviral treatment, the HCV transmission group, the mean METAVIR scores for necroinflammation and fibrosis, cirrhosis, HCV viral load and the HCV genotype, lymphocyte and platelet counts, the AST level, serum GGT and alkaline phosphatase levels, the prothrombin time, and the serum creatinine concentration.
 
In multivariate analysis, zidovudine (OR, 3.27 95% CI, 1.64-6.54, P = 0.0008) and peg-IFN (OR, 2.35; 95% CI, 1.16-4.57, P = 0.0179) remained significantly associated with a higher risk of anaemia. In contrast, a higher baseline haemoglobin level (OR, 0.35 95% CI, 0.26-0.49, P < 0.0001) and PI-based antiretroviral therapy had a protective effect (OR, 0.51 95% CI, 0.30-0.86, P = 0.0114) (Table 2).
 
These factors were also independently associated with a decrease by 25% of the baseline haemoglobin level which was observed in 92 (24%) patients on treatment. Multivariate odds-ratios were 3.72 (95% CI, 1.99-6.96, P < 0.0001) for zidovudine, 1.98 (95% CI, 1.13-3.45, P = 0.016) for peg-IFN and 0.67 (95% CI, 0.045-1.01, P = 0.058) for PI-based antiretroviral therapy.
 
The Kaplan-Meier for anaemia-free survival relative to antiretroviral therapy (with or without zidovudine, with or without PIs) are shown in Fig. 1. Changes in the mean haemoglobin level during HCV therapy in patients treated with zidovudine-based antiretroviral therapy and in patients treated without zidovudine-based antiretroviral therapy are shown in Fig. 2.
 
Discussion
Among HIV-seronegative HCV-infected patients, the haemoglobin level falls below <10 g/dL in 8-9% of cases during peg-IFN-ribavirin combination therapy [4,5]. In this study of 383 HIV/HCV-coinfected patients, despite the use of a lower ribavirin starting dose (800 mg/day), 15.9% of patients developed anaemia, defined as a haemoglobin concentration below 10 g/dL. A lower baseline haemoglobin level, zidovudine therapy and peg-IFN therapy were associated with a significantly increased risk of anaemia. In contrast, PI-based antiretroviral therapy roughly halved the risk.
 
As ribavirin interfere with thymidine analogues such as zidovudine and stavudine, and can aggravate didanosine toxicity, a low ribavirin starting dose (800 mg/day) was used in this trial [6]. However, the mean dose was above 10.6 mg/kg (12.3 ± 2.1 mg/kg), which is the dose associated with sustained virologic responses among HIV-seronegative HCV-infected patients receiving peg-IFN alpha-2b and ribavirin [4].
 
Concomitant treatment with zidovudine was associated with an adjusted 3.3-fold higher risk of anaemia among patients receiving the interferon-ribavirin combination, confirming the results of previous studies [7-9]. Zidovudine affects haematopoietic lineages and reduces globin mRNA synthesis [10]. Zidovudine, contrary to abacavir, lamivudine and stavudine, also increases ribavirin plasma levels [7]. In HCV-monoinfected patients and in HIV/HCV-coinfected patients, the daily ribavirin dose, expressed per kilogram of body weight, is the main determinant of ribavirin plasma levels [7]. In HIV/HCV-coinfected patients treated with peg-IFN plus ribavirin, higher ribavirin plasma levels are independently associated with a fall in the haemoglobin level [7]. We also found a correlation between the daily ribavirin dose per kilogram of body weight and the risk of anaemia, but only in univariate analysis. Nevertheless, our results suggest that zidovudine and ribavirin have synergistic haematologic toxicity. Interestingly, all three patients who stopped taking zidovudine after developing anaemia in our study were able to resume ribavirin therapy at the initial dose. Patients taking stavudine-based antiretroviral therapy were significantly less likely to develop anaemia, confirming the higher risk of haematologic disorders with zidovudine-based regimens compared with stavudine-based regimens [11].
 
In this study, PI-based therapy, but not NNRTI-based therapy, had a preventive effect on anaemia. The introduction of highly active antiretroviral therapy based on PI or NNRTI is associated with a correction of pre-existing anaemia [12,13]. The mechanisms underlying these positive effects are unclear but may involve antiretroviral efficacy. PIs, in addition to their effects on HIV replication, appear to affect various cellular functions. In particular, ritonavir has been associated with decreased apoptosis of haematopoietic progenitors and also directly stimulates progenitor cell growth in vitro [14].
 
Although peg-IFN enhances the suppression of haematopoiesis in all three lineages relative to standard interferon, the mean decrease in the haemoglobin level, and the risk of anaemia (<10 g/dL), is similar with the two interferons in HCV-monoinfected patients [4,5,15]. In contrast, we found that peg-IFN-2b was associated with an adjusted 2.3-fold higher risk of anaemia relative to standard interferon. However, treatment discontinuations and dose adjustments for anaemia were similarly frequent in patients receiving peg-IFN and standard interferon [1].
 
Lower baseline creatinine clearance and older age have been identified as risk factors for anaemia in HCV-monoinfected patients treated with standard interferon plus ribavirin [16]. In our study, lower baseline creatinine clearance, estimated with the Cockcroft-Gault formula, which includes age, sex, body weight and the serum creatinine level, was also associated with anaemia but only in univariate analysis. Anaemia was not associated with age, sex or the serum creatinine level in univariate analysis, suggesting that the most important determinant of anaemia is body weight and, thus, the daily ribavirin dose per kg.
 
In conclusion, clinicians should be aware of the potential interference between treatments for HCV and HIV infection regarding the haemoglobin level. The haemoglobin level should be closely monitored, particularly during the first 3 months of HCV therapy, the period during which 68.8% of the cases of anaemia observed in our study occurred. Anaemia did not negatively impact on the anti-HCV treatment outcome in our study. However, higher doses of ribavirin given in combination with pegylated interferon have been linked to higher rates of viral eradication in several studies [17,18]. If anti-HCV and anti-HIV treatments must be administered concomitantly, then zidovudine discontinuation prior to HCV therapy, particularly when alternative antiretroviral options are available could help to avoid anaemia and thereby permit the use of higher ribavirin doses.
 
 
 
 
  icon paper stack View Older Articles   Back to Top   www.natap.org