Insulin resistance impairs rapid virologic response in HIV/hepatitis C virus coinfected patients

Insulin resistance impairs rapid virologic response in HIV/hepatitis C virus coinfected patients on peginterferon-alfa-2a



"In this study, insulin resistance was an independent predictor of RVR in HIV/HCV coinfected patients.....insulin resistance (HOMA-IR) reduced odds of RVR by 90%.....Most of the subjects studied were treated with a protease inhibitor. Protease inhibitor use at the beginning of PegIFN/RBV treatment was also found to be negatively related to SVR [24]. The reasons for this are still controversial. The correlation between protease inhibitor exposure and higher HCV viral load is suggested as a hypothesis [24]. Insulin resistance could be another possible link between protease inhibitor exposure and lower SVR rate."


AIDS:

23 April 2008



"In the multivariate analysis the independent predictors of rapid virologic response were: genotype 1 or 4 [adjusted odds ratio 0.18 (0.06-0.55)], hepatitis C virus-RNA < 400.000 UI/ml [adjusted odds ratio 0.229 (0.09-0.92)] and homeostasis model of assessment-insulin resistance more than 3.00 [adjusted odds ratio 0.1 (0.05-0.6)].The GLUT- 4 (glucose transporter 4) activity is impaired by HCV and by some protease inhibitors......In HIV-infected patients the incidence of diabetes mellitus is four times higher than in HIV negative persons [2] and insulin resistance is recognized in 50% of protease inhibitor treated patients [20].......Protease inhibitor-based HAART at the beginning of anti-HCV treatment was negatively related to RVR [AOR 0.3 (95% confidence interval (CI) 0.01-1.04) P 0.04].....In persons with HIV infection, glucose abnormalities can be induced by highly active antiretroviral therapy (HAART). Induction of insulin resistance was reported in 61% of subjects on protease inhibitor based HAART [5]. A prolonged use of zidovudine and stavudine (thymidine nucleoside analogues) could lead insulin resistance through mitochondrial dysfunction [6]. Thus, in HIV/HCV coinfected patients, insulin resistance could be caused by both HCV infection and HAART [7].



Nasta, Paola; Gatti, Francesca; Puoti, Massimo; Cologni, Giuliana; Bergamaschi, Viviana; Borghi, Federica; Matti, Alessandro; Ricci, Antonella; Carosi, Giampiero

From the Institute of Infectious and Tropical Diseases, University of Brescia, Brescia, Italy.

Correspondence to Paola Nasta, MD, Institute of Infectious and Tropical Diseases, University of Brescia, P. le Spedali Civili, 1, 25123 Brescia, Italy. Tel: +39 030 3995665; fax: +39 030 303061; e-mail: paola.nasta@tin.it




Abstract



Objectives: To investigate the association between insulin resistance and rapid virologic response.



Design: All consecutive HIV/hepatitis C virus coinfected patients who started peg-interferon alpha-2a (180 μg/week) and ribavirin 1000-1200 mg/day were analysed.



Methods: Insulin resistance was defined according to the homeostasis model of assessment-insulin resistance calculated as fasting insulin (mIU/l) × fasting glucose (mmol/l)/22.5. Rapid virologic response was defined as testing negative for hepatitis C virus-RNA after 4 weeks of therapy. Fasting levels of insulin and glucose in plasma were measured in all patients on the first day of treatment. Hepatitis C virus-RNA was determined by quantitative PCR assay (version 3.0). Hepatitis C virus-RNA was measured by qualitative PCR assay (COBAS 2.0) after 4 weeks of treatment.



Results: Seventy-four HIV/hepatitis C virus coinfected patients were enrolled [mean age 41.7 years (SD 5.3), 61 men, 54.1% with advanced fibrosis (F3-4 according to METAVIR classification), 52.4% with infection by hepatitis C virus genotype 1 or 4]. Rapid virologic response was reached by 30 subjects. In the multivariate analysis the independent predictors of rapid virologic response were: genotype 1 or 4 [adjusted odds ratio 0.18 (0.06-0.55)], hepatitis C virus-RNA < 400.000 UI/ml [adjusted odds ratio 0.229 (0.09-0.92)] and homeostasis model of assessment-insulin resistance more than 3.00 [adjusted odds ratio 0.1 (0.05-0.6)].



Conclusion: The homeostasis model of assessment-insulin resistance score should be evaluated and possibly corrected before starting anti-hepatitis C virus therapy.



Introduction



Antiretroviral therapies, and HIV and hepatitis C virus (HCV) infections have been associated with abnormalities in carbohydrate metabolism including insulin resistance and diabetes mellitus [1,2].



Prevalence of diabetes mellitus in patients with chronic hepatitis C is higher than in the general population [3] and HCV-positive patients older than 40 years have three times the risk of developing type 2 diabetes mellitus than HCV-negative subjects [4].



In persons with HIV infection, glucose abnormalities can be induced by highly active antiretroviral therapy (HAART). Induction of insulin resistance was reported in 61% of subjects on protease inhibitor based HAART [5]. A prolonged use of zidovudine and stavudine (thymidine nucleoside analogues) could lead insulin resistance through mitochondrial dysfunction [6]. Thus, in HIV/HCV coinfected patients, insulin resistance could be caused by both HCV infection and HAART [7].



Liver-related deaths are increasing among HIV patients with hepatitis coinfection [8] and anti-HCV treatment with peginterferon (PegIFN) and ribavirin is mandatory for eligible subjects [9]. Among positive predictors of anti-HCV treatment success, the rapid HCV RNA decline, defined as rapid virologic response (RVR), is recognized as one of the most important [10]. RVR is associated with a rapid first and second phase decline of HCV RNA expressing the sensitivity of the virus-host system to antiviral treatment [10]. It is the key for both achieving end-of-treatment response and to preventing relapse after clearance induced by PegIFN and ribavirin. On the other hand, genotype, HCV viral load, level of fibrosis and, more recently, a higher homeostasis model of assessment-insulin resistance (HOMA-IR) score have been shown to be related to poor virological response [11,12].



On the basis of these findings, the association between HOMA-IR [13] and RVR was assessed to investigate the role of insulin resistance in the early outcome of PegIFN and ribavirin treatment in HIV/HCV coinfected patients.



Discussion



Treatment with PegIFN and ribavirin should be implemented in HIV/HCV coinfected patients because of the high rate of liver-related mortality [8].



Nevertheless, the sustained viral response rate is lower in these patients than in HCV monoinfected subjects: [17 versus 46% in genotype 1 and 67 versus 76% in genotype 2-3] and the treatment withdrawal incidence seems to be similar in the two groups (22 versus 25%) [14,15]. Thus optimization of anti-HCV treatment and identification of cofactors associated with the reduced response observed in HIV-infected patients are urgent and unmet needs. As in monoinfected patient, the predictors of low sustained virologic response (SVR) in HIV/HCV coinfected patients are three characteristics that cannot be modified: HCV genotype 1-4 [16,17], high HCV viral [14,15] and advanced fibrosis [18,19]. Insulin resistance could be influenced by lifestyle and HAART regimens in HIV-infected persons. We demonstrated that insulin resistance is associated with a reduced sensitivity to anti-HCV treatment in HIV/HCV coinfected persons. In HIV-infected patients the incidence of diabetes mellitus is four times higher than in HIV negative persons [2] and insulin resistance is recognized in 50% of protease inhibitor treated patients [20]. Insulin resistance is multifactorial in HIV/HCV coinfected persons. The GLUT- 4 (glucose transporter 4) activity is impaired by HCV and by some protease inhibitors. The lack of glucose intake is mediated directly by HCV through inhibition of IRS1-2 [21] and by the increase of suppressor of cytokines signalling (SOCS)-3 [17] and indirectly by both HIV and HCV through increased cytokine levels (tumour necrosis factor α and interleukin-6) [16]. In addition, the lack of β oxidation due to HCV, PPAR α and γ (peroxisome proliferator activation receptor α and γ), inhibition by HCV [22], steroil and carbohydrate regulatory binding protein (SREBP and ChreBP) activation by antiretroviral agents and NUCs mitochondrial toxicity favour insulin resistance [23].



In this study, insulin resistance was an independent predictor of RVR in HIV/HCV coinfected patients. Most of the subjects studied were treated with a protease inhibitor. Protease inhibitor use at the beginning of PegIFN/RBV treatment was also found to be negatively related to SVR [24]. The reasons for this are still controversial. The correlation between protease inhibitor exposure and higher HCV viral load is suggested as a hypothesis [24]. Insulin resistance could be another possible link between protease inhibitor exposure and lower SVR rate.



In conclusion, our results suggest that insulin resistance should be evaluated in HIV/HCV coinfected patients and corrected before starting anti-HCV treatment by changing life styles and, when necessary, HAART regimens in order to enhance 'virus-host system' sensitivity to antiviral treatment.



Results



Seventy-four HIV/HCV co-infected patients started PegIFN α-2a and ribavirin between January 2005 and March 2007.



The baseline characteristics of the study population are summarized in Table 1. The mean time of HIV exposure was 13.5 years (SD 5.9); 91.9% of the subjects were on HAART and the mean T lymphocyte CD4+ cell count was 492 (SD 213) cells/mm3. HIV-RNA was under the limit of detection (<50 copies/ml) in all treated patients.



Thirty-five out of 74 (47.2%) subjects had HCV genotype 2-3; 40 (54%) subjects had an advanced liver fibrosis, defined as grades 3-4 at METAVIR score. A low HCV viral load <400 000 UI/ml was found in 43/74 (58.1%) subjects.



RVR was obtained in 30 patients [22/30 (73.3%) of genotype 2-3 and 8/30 (26.6%) with genotype 1-4]. In multivariate analysis variables significantly related to RVR were genotype 1-4 [adjusted odds ratio (AOR) 0.04 (0.009-0.2)], HCV-RNA > 400 000 UI/μl [AOR 0.229 (0.09-0.92)] and HOMA-IR score greater than 3 [AOR 0.1(0.05-0.6)] (Table 1). Protease inhibitor-based HAART at the beginning of anti-HCV treatment was negatively related to RVR [AOR 0.3 (95% confidence interval (CI) 0.01-1.04) P 0.04]. A sensitivity analysis introducing HCV-RNA levels and HOMA-IR scores as continuous variables did not change the results. Figure 1 shows the rate of RVR based on HOMA-IR score, HCV viral load and HCV genotype.



Fig 1.


Table 1.






Aims and methods



All HIV/HCV coinfected subjects who were naïve for HCV therapy and started anti-HCV combination therapy with PegIFN α-2a (180 μg/week) and ribavirin (1000-1200 mg/day) (RBV) from January 2005 were enrolled into an observational prospective study.



Body mass index (BMI) was calculated as weight (kg) divided by height (m2). An overnight (12 h) fasting blood sample was drawn for the measurement of routine laboratory values such as plasma lipids, creatinine, glucose, transaminases, insulin, CD4+ T cells and HIV-RNA. Serum insulin levels were measured by electro-chemo-illuminescence immuno-assay using an autoanalayser. The HOMA-IR was calculated on the basis of fasting values of plasma glucose and insulin according to the HOMA equation: fasting insulin (IU/l) × fasting glucose (mmol/l)/22.5. Insulin resistance was defined as a HOMA value greater than 3.00, on the basis of normal values, (range 0.7-2.25), and one standard deviation [13]. HCV-RNA by quantitative PCR assay (version 3.0) at baseline and HCV-RNA by qualitative PCR assay (COBAS 2.0) were measured after 4 weeks of treatment. RVR was defined as undetectable HCV-RNA by qualitative PCR assay after 4 weeks of treatment.



Statistical analysis



Continuous variables are expressed as mean and standard deviation. Categorical variables are expressed as number of cases (percentage). The associations between RVR and HOMA were analysed. Likewise, we also appraised the relationship between RVR and the following covariates: sex, age, HIV risk factors, smoking, alcohol consumption, body weight, BMI, HIV and HCV exposure, Centers for Disease Control and Prevention (CDC) status, time of antiretroviral (ARV) exposure, type of antiretroviral therapy at baseline (protease inhibitor or non-nucleoside reverse transcriptase inhibitor based), level of liver fibrosis and HCV genotype (T lymphocytes CD4+ nadir and zenith).



In the bivariate analysis, the comparisons between continuous variables were carried out using the Student's t-test if they followed a normal distribution, or the Mann-Whitney U test if they did not. Frequencies were compared using the χ2 test or by applying Fisher's test when there were cells with expected frequencies lower than five. The two-tailed P value was selected in all cases and those less than 0.05 were considered significant. Variables associated with RVR with a P value less than or equal to 0.08 in the bivariate analysis were included in a multivariate stepwise logistic regression analysis where the dependent variable was RVR. A sensitivity analysis was performed including HOMA-IR score and baseline HCV-RNA levels as continuous variables. Patients' data were collected in a computerized database and analysed using Epi Info version 3.3.2 (9 February 2005