| |
Determinants of nevirapine hypersensitivity and its effect on the association between hepatitis C status and mortality in antiretroviral drug-naive HIV-positive patients
|
| |
| |
AIDS:Volume 21(12)31 July 2007p 1561-1568
Phillips, Elizabetha,b; Gutierrez, Soniac; Jahnke, Nataliea; Yip, Benitaa; Lima, Viviane Da; Hogg, Robert Sa,d; Harrigan, P Richarda,b; Montaner, Julio SGa,b
From the aBritish Columbia Centre for Excellence in HIV/AIDS, Providence Healthcare, St Paul's Hospital, Canada
bDepartment of Medicine, University of British Columbia, Vancouver, Canada
cViral Hepatitis and AIDS Study Group, Department of Internal Medicine, Virgen del Rocio University Hospital, Seville, Spain
dFaculty of Health Sciences, Simon Fraser University, Burnaby, Canada.
Abstract
Objective: To assess risks factors and outcomes associated with nevirapine hypersensitivity reactions, and to determine the effect of hypersensitivity as a modifier of the association between hepatitis C virus (HCV) infection and mortality among antiretroviral drug-naive patients.
Methods: The primary endpoint was hypersensitivity reactions in a population-based cohort of antiretroviral therapy-naive HIV-individuals, 18 years or older in British Columbia, Canada, who started triple antiretroviral therapy with nevirapine between May 1997 and June 2003. Univariate and multivariate analyses were performed to identify predictors of nonaccidental mortality in the subgroup of patients with known HCV serostatus.
Results: A total of 66 (9.6%) of 685 patients met the definition for hypersensitivity reactions. In the univariate logistic regression analysis, no variables were identified as risk factors. In multivariate survival analyses conducted to identify characteristics associated with nonaccidental mortality, patients with both HCV coinfection and hypersensitivity reactions had a higher risk of death (hazard ratio, 7.12; 95% confidence interval, 2.73-18.53; P < 0.001) compared with those who did not have HCV coinfection or hypersensitivity reaction.
Conclusion: Results of this study suggest that the hypersensitivity reaction behaves as an effect modifier of the association between HCV infection and mortality in this cohort of antiretroviral drug-naive HIV-positive patients. These results support the current recommendation against the use of nevirapine in HIV/HCV-coinfected patients.
Results
Baseline characteristics
Between May 1997 and June 2003, a total of 685 antiretroviral drug-naive HIV-infected patients aged 18 years and older initiated antiretroviral therapy containing nevirapine. Table 1 shows the baseline characteristics of these patients. HCV serology, ALT and AST levels were available in 528 (77%), 342 (50%) and 435 (64%) patients, respectively. Patients who tolerated nevirapine made up the nevirapine-tolerant group (619 patients, 90.4%), and those who developed HSR made up the hypersensitivity group (66 patients, 9.6%).
The median follow-up time for the entire cohort was 64 months [interquartile range (IQR), 39-78]. The nevirapine-tolerant group had longer median duration of nevirapine therapy (330 days; IQR, 115-830) than the hypersensitivity group (30 days; IQR, 30-39 days) (P < 0.01). These groups did not differ with respect to other variables such as age, male gender, IDU, aboriginal status, being white, AIDS diagnosis at baseline, HCV coinfection, concomitant use of didanosine or stavudine, elevation of AST and ALT, CD4 cell count, HIV plasma viral load and physicians' experience. There was a nonsignificant difference in the proportion of males in the nevirapine-tolerant and HSR groups (77% and 67%, respectively; P = 0.07).
Overall, there were 277 patients (52.5%) who exhibited antibodies against HCV among 528 patients tested for HCV serology status. In the nevirapine-tolerant group (see Table 1), 251 individuals (52.5%) were HCV positive among the 479 tested patients. In the nevirapine HSR group, 26 individuals (53.1%) were HCV positive among the 49 tested patients (P = 0.94).
Determinants of nevirapine hypersensitivity
Table 2 shows the univariate logistic regression analysis for determinants of nevirapine HSR. No variables were identified as risk factors and, therefore, there was no need to perform a multivariate regression analysis.
Clinical description of hypersensitivity
Detailed information on adverse events reported by treating physicians was available for 23 (35%) of the 66 patients who developed HSR with nevirapine. None was reported as life threatening. Among these 23 reported cases, 16 (70%) patients developed a rash, which was severe in 3; 5 (22%) had abnormal liver function tests, 1 (4%) fever and 4 (17%) gastrointestinal symptoms.
Causes of death
In the group of 528 subjects with known HCV serology, 79 (15%) died. While 66 deaths from the 479 nevirapine-tolerant subjects gave a mortality rate of 14%, 13 deaths from the 49 hypersensitivity group gave a mortality rate of 27%.
Among the 66 deaths in the tolerant group (no HSR), 41 (62%) died from HIV-related causes (infections, malignancies and other specified conditions), 4 (6%) from cardiovascular causes, 4 (6%) from pulmonary causes, 2 (3%) from chronic HCV and the remaining 15 (23%) patients from other causes.
Among the 13 subjects who died in the HSR group, 11 (85%) patients died from HIV related causes (B cell lymphoma, infectious and parasitic diseases, other malignancies, wasting syndrome and other specified conditions), 1 (8%) from alcoholic cirrhosis of the liver, and 1 (8%) from an unspecified cause.
Resumption of therapy
Data are available regarding the approximate dates of resumption of therapy for individuals in the HSR group. Among 49 patients with HSR, 13 (27%) had not resumed antiretroviral therapy by the end of the follow-up period of this study, of these 9 (69%) had died. Of the 22 patients (45%) who restarted antiretroviral therapy within the first 6 months following an HSR, only one died. The remaining 14 (28%) restarted antiretroviral therapy after the first 6 months following an HSR and 3 (21%) of these died.
Mortality analyses
The results of the univariate and multivariate survival analyses for risk factors associated with nonaccidental deaths are shown in Table 3. The analyses were restricted to the 528 subjects for whom HCV serology results were available. Patients were stratified into the four groups as described in the Methods. In univariate analyses, the variables significantly associated with nonaccidental mortality were HCV coinfection with or without HSR (groups HSR-/HCV+ and HSR+/HCV+ compared with the control group HSR-/HCV-), baseline CD4 cell count per 100 cellsμl decrement, viral load ≥ 100 000 copies/ml, age per 10 year increment, physician experience per 100 patient increments and adherence per 10% decrement. Adjusted multivariate analysis showed the following were associated with mortality: older age [hazard ratio (HR) per 10 year increment,1.42; 95% confidence interval (CI), 1.13-1.78], low CD4 cell count (HR per 100 cells/μl decrement, 1.53; 95% CI, 1.28-1.83), and HCV coinfection alone without HSR (group HSR-/HCV+; HR, 3.06; 95% CI, 1.61-5.83). Occurrence of HSR was not found to be a highly significant predictor of death in patients without HCV coinfection (group HSR+/HCV-; HR, 2.09; 95% CI, 0.58-7.45; P = 0.26). Patients with both HCV coinfection and HSR (group HSR+/HCV+) showed a higher risk of death (HR, 7.12; 95% CI, 2.73-18.53; P < 0.01) when compared with those without HSR or HCV coinfection (group HSR-/HCV-). This last finding suggested that HSR is an effect modifier of the association between HCV and mortality in this cohort of HIV-positive patients.
The median and interquartile range of survival estimates for each of the four groups are plotted in Fig. 1. The figure shows that there is a strong association of the combined HSR/HCV variable with mortality (P < 0.01). The survival estimate was the lowest for the group of HSR+/HCV+ in comparison with the other three groups (median, 0.75; interquartile range, 0.64-0.89).
Discussion
Our results show that HIV-positive patients who developed HSR to nevirapine and were coinfected with HCV (HSR+/HCV+) had a higher risk of mortality than those without HSR or HCV coinfection. This finding suggests that nevirapine hypersensitivity is an important effect modifier of the association between HCV coinfection and mortality. We did not identify specific factors significantly associated with risk for developing nevirapine HSR in our cohort.
Several factors have been previously associated with increased risk for developing nevirapine hypersensitivity, including CD4 cell counts > 250 cells/μl, female gender, prolonged exposure to any antiretroviral therapy, hepatitis B virus or HCV coinfection and high ALT levels at baseline [15,20,21,30]. In our study, no risk factors were identified, probably because of the limited power to conduct this analysis. Previous studies have associated higher plasma concentration of nevirapine with improvement in the virological response and increase in toxicity [31-33]. However, our study does not provide data on plasma concentration of nevirapine that could be linked to the development of HSR.
Our rate of HSR is higher than that reported in other studies. A prospective cohort study of 610 patients prescribed a nevirapine-containing regimen reported an estimated incidence of hepatotoxicity (defined as at least a three-fold increase in AST or ALT) of 3.7% at 3 months [20]. However, this was associated with reversible clinical hepatitis in only 1.1% of subjects. Six (3%) patients of the 398 adults randomized to receive an antiretroviral regimen of nevirapine, didanosine or zidovudine versus dual nucleoside therapy developed a clinical picture compatible with HSR [34]. All of these patients were prescribed nevirapine. It is difficult to compare our results with others because of the overlap in their definition of hypersensitivity and liver toxicity, and the diversity of the populations studied.
Regarding mortality, our study showed that causes of death were similar for both nevirapine-tolerant and hypersensitive patients. Deaths were mainly the result of HIV disease progression. The INCAS trial [1] showed that, despite the fact that nevirapine-treated patients had an overall higher rate of clinical and laboratory adverse events, triple drug therapy with two nucleosides and nevirapine allowed for a slower disease progression and fewer deaths compared with both dual nucleoside therapy with zidovudine and didanosine, and therapy with nevirapine and zidovudine.
Predictors of death as reported by our multivariate analysis have also been reported in the literature [35-37]. Interestingly, in our cohort of antiretroviral drug-naive HIV-positive patients starting antiretroviral therapy including nevirapine, we were able to show an unexpected influence of nevirapine hypersensitivity on HCV coinfection in terms of its effect on mortality, even in a setting where only a limited number of patients developed hypersensitivity. These findings support the current recommendation against the use of nevirapine in patients with HCV infection [27]. Regarding adherence to subsequent regimens, our results demonstrate that, after developing hypersensitivity to nevirapine, patients who did not restart antiretroviral therapy with other agents were more likely to die. The adherence variable included in our multivariate model should adjust for the effects of patients' various rates of therapy resumption after developing hypersensitivity.
There are several important aspects of our study that should be highlighted. First, the study was carried out within a province-wide treatment programme where all individuals had access to medical attention, combination antiretroviral therapy and laboratory monitoring free of charge. Therefore, we believe that issues related to access to therapy did not compromise our results. Second, this study was based on treatment-naive individuals; therefore, the results were not confounded by any past antiretroviral drug use. Third, delayed reporting did not play a role in these analyses as the vast majority of deaths were reported within 3 months of occurrence.
Our work has some limitations. Nevirapine hypersensitivity as defined in our study includes those who discontinued nevirapine owing to the development of any combination of rash, fever, hepatitis, or some other reasons, so misclassification bias may have occurred. In addition, potential risk factors for nevirapine HSR such as hepatitis B infection and alcohol intake were not available for inclusion in this study.
In summary, this study could not identify specific factors significantly associated with risk for developing nevirapine HSR in our cohort of HIV-positive antiretroviral drug-naive patients. However, our results show that nevirapine hypersensitivity plays an important role in the association between HCV coinfection and mortality. These results support the current recommendation against using nevirapine in HIV/HCV dually infected individuals.
Introduction
Nevirapine is a nonnucleoside reverse transcriptase inhibitor with proved efficacy in combination therapy for HIV [1-3]. Although nevirapine is usually well tolerated [4], patients initiating nevirapine can experience a hypersensitivity reaction (HSR) that can be a serious adverse event; it is characterized by combinations of rash, fever or hepatitis and typically occurs within the first 6 weeks after initiation [5,6]. Nevirapine HSR can lead to discontinuation of nevirapine and it can be more rapid and severe with nevirapine rechallenge [5,7]. The reaction can lead to morbidity through treatment interruption, inconvenience and loss of productivity. In addition, the long half life of nevirapine increases the risk of nonnucleoside reverse transcriptase inhibitor resistance when patients discontinue all the other antiretroviral drugs in the regimen at the same time [8-10]. This can lead to decreased therapeutic options if nevirapine resistance develops, as this often leads to cross-resistance with other nonnucleoside reverse transcriptase inhibitors.
The incidence of patients receiving nevirapine who developed elevated transaminases in a meta-analysis was 10%, whereas symptomatic hepatic events were seen in 4.9% of patients [6]. According to the product monograph [7], nevirapine-related hepatic events, regardless of severity, occurred in 4% of patients (range, 0-11). Asymptomatic transaminases elevations higher than 5X upper limit of normal (ULN) have been reported to occur in 5.8% of patients taking nevirapine (range, 0-11). Rash is a common adverse effect of nevirapine and approximately 16% of patients experience a mild to moderate rash. Severe rashes occur in about 6.5% of nevirapine-treated patients [11].
There is an overlap between HSR and nevirapine-related hepatotoxicity. Case reports describe examples of HSR that developed 2 to 6 weeks after initiation of nevirapine and resolved after discontinuation of the drug [12,13]. A prospective observational study on the incidence of severe hepatotoxicity in patients receiving nonnucleoside reverse transcriptase inhibitor-containing antiretroviral therapy [14] reported that, although severe hepatotoxicity was observed in 15.6% of patients prescribed nevirapine, hypersensitivity was identified only in one. Therefore, with the currently available data, it is difficult to establish if developing an HSR is associated with a higher risk of mortality.
The aetiology of nevirapine hypersensitivity is not clear, but several factors have been shown to be associated with it, including baseline CD4 cell counts > 250 cells/μl and female gender [15]. Genetic factors have been considered in nevirapine hypersensitivity in recent studies. For example, polymorphisms in the gene MDR1 might predict nevirapine-associated hepatotoxicity [16], and both HLA-DRB1*0101 and higher CD4 cell counts have been associated with an increased risk of nevirapine hypersensitivity [17]. Lower pretreatment CD4 cell counts appear to be protective against the development of hypersensitivity, which accordingly has been reported as more frequent and severe among non-HIV-infected individuals receiving prophylactic nevirapine [18,19]. Other factors associated with nevirapine hypersensitivity are prolonged exposure to any antiretroviral therapy, coinfection with hepatitis B virus or hepatitis C virus (HCV) and abnormal baseline levels of alanine aminotransferase (ALT) [20,21].
The shared transmission route of HIV and HCV is responsible for the high prevalence of coinfection. It has been estimated that there are approximately 10 million coinfected individuals worldwide, of whom 300 000 are in the United States [22]. The rate of coinfection ranges from 10 to 40%, and up to 80 to 95% in highly exposed populations such as hemophiliacs and intravenous drugs users (IDU). In Europe, the average figure for HIV-infected patients living with HCV is 40%, being higher in eastern Europe [23]. As of December 1999, more than 11 000 patients were estimated to be dually infected in Canada, 71% of these were IDU [24]. The Vancouver Injecting Drug Users Study [25] reported HIV and HCV prevalence rates of 23% and 88%, respectively, in a cohort of more than 1000 IDU.
The objective of this study was to assess the risks factors and outcomes associated with nevirapine hypersensitivity in a large population of antiretroviral drug-naive HIV-infected patients who started triple antiretroviral therapy with nevirapine. Survival analyses were performed in the subgroup of patients with known HCV serostatus to identify variables associated with mortality.
Methods
HIV/AIDS drug treatment programme
The distribution and the population-based monitoring of antiretroviral therapy in British Columbia have been extensively described in the literature [26]. Briefly, since October 1992, the distribution of antiretroviral drugs has been the responsibility of the HIV/AIDS Drug Treatment Program of the British Columbia Center for Excellence in HIV/AIDS (the Centre). The Centre distributes antiretroviral drugs based on guidelines generated by the Therapeutic Guidelines Committee [27]. The Centre's HIV/AIDS drug treatment programme has received ethical approval from the University of British Columbia Ethics Review Committee at its St Paul's Hospital site. The programme also conforms with the province's Freedom of Information and Protection of Privacy Act.
Data collection
All antiretroviral treatment recipients are entered into an Oracle-based monitoring and evaluation reporting system that uses standardized indicators to prospectively track antiretroviral drug use and clinical and health status of HIV-positive individuals. Physicians starting an HIV-infected individual on antiretroviral therapy complete a drug request form that compiles information on the applicant's address, past HIV-specific drug history, CD4 cell count, plasma HIV-1 RNA, current drug requests, and enrolling physician data. At the time of the first prescription refill, participants are asked to provide informed consent for accessing medical electronic records and to complete a participant survey that elicits information on sociodemographic characteristics, clinical and health status, and alternative therapy use. Both the consent form and the participant survey are optional and participant's refusal to do either will not limit access to free antiretroviral therapy. At the same time, the treating physicians are asked to complete a clinical staging form using the World Health Organization (WHO) clinical staging system [28].
The Centre recommends that plasma HIV-1 RNA levels and CD4 cell counts be monitored at baseline, at 4 weeks after starting antiretroviral therapy, and every 3 months thereafter [27]. All HIV-1 RNA viral load testing and most CD4 cell count testing in the province are conducted at St. Paul's Hospital and the data files are uploaded daily into the drug treatment programme database. The results of CD4 cell count testing carried out in other laboratories in British Columbia are collected through drug request forms. Plasma HIV-1 RNA levels were determined using the Roche Amplicor Monitor assay (Roche Diagnostics, Laval, Quebec, Canada) using either the standard method or the ultrasensitive adaptation (since 1999). CD4 cell counts were measured by flow cytometry, followed by fluorescent monoclonal antibody analysis (Beckman Coulter, Mississauga, Ontario, Canada).
Study participants
HIV-infected, treatment-naive patients aged 18 years and older starting nevirapine-based antiretroviral therapy between May 1997 and June 2003 were included in the analysis. Participants must also have had a CD4 cell count and plasma HIV-1 RNA measurement within 6 months prior to the first antiretroviral drug start date. Study data from eligible patients were extracted from the drug treatment programme database at the British Columbia Centre for Excellence.
Outcome measures and explanatory variables
The primary endpoint in this analysis was nevirapine HSR. Univariate and multivariate survival analyses were performed for the 528 (77%) subjects with HCV serology results available to identify risk factors associated with nonaccidental death. The restriction of the survival analyses to these individuals can be justified by the high prevalence of HCV infection in our study population, its known association with poorer outcomes [22] and the possibility of HSR being related to liver toxicity and potentially worsening of the patients' liver profile. For this purpose, patients were stratified into four groups: HSR+/HCV+ (patients who developed hypersensitivity and also had a positive HCV serology); HSR-/HCV+ (patients who did not develop hypersensitivity but had a positive HCV serology); HSR+/HCV- (patients who developed hypersensitivity but had a negative HCV serology); and HSR-/HCV- (patients who did not develop hypersensitivity and had a negative HCV serology).
Nevirapine HSR was defined as the patient permanently stopping nevirapine after the first prescription (a) within 30 days with no report of adverse effects, (b) within 90 days with report of adverse drug event, or (c) within 90 days with elevation in alanine aminotransferase (ALT) or aspartate aminotransferase (AST) during treatment, or decrease in either one of these values after stopping nevirapine. ALT and AST levels were examined at up to three time points (baseline, during treatment, and after stopping nevirapine, when available). Each value was graded from 1 to 4 in agreement with the National Institute of Health Division of AIDS statement for grading the severity of adult adverse experiences (grade 1, 1.25-2.50X ULN; grade 2, > 2.5-4.9X ULN; grade 3, 5-10.0X ULN; grade 4, > 10X ULN) [29]. If a patient's value was elevated by at least one grade level between baseline and during treatment, and/or decreased by at least one grade level after discontinuation compared with during nevirapine treatment, the criteria for hypersensitivity was met.
The following explanatory variables were investigated: age, gender, ethnicity, a history of IDU, physician experience, didanosine and stavudine concomitant use, adherence, baseline AIDS diagnosis, baseline CD4 cell count, baseline HIV-1 RNA and baseline ALT and AST levels. Physician experience was defined as the number of HIV-positive patients the physician had previously treated at the time the study subject was enrolled into the HIV/AIDS drug treatment programme. Adherence was defined as the number of days' worth of antiretroviral medications dispensed as a proportion of follow up time during year 1, and expressed as percentage.
Statistical analyses
Logistic regression analysis was used to examine associations between baseline variables and the development of nevirapine hypersensitivity. Categorical variables were compared using Pearson's χ2 test, while continuous variables were compared using the Wilcoxon rank test. For the mortality analysis, parametric survival regression under the exponential distribution was used to examine baseline factors known or suspected to be associated with time to death among individuals for whom HCV serology results were available. The Kruskal-Wallis test was applied to show associations of survival estimates with the combined variable of hypersensitivity and HCV coinfection (HSR/HCV). Death dates were the event dates. Event-free subjects were right censored as of 30 November, 2005. Subjects with last known contact being prior to 30 November, 2005 were censored at the last contact date. Variables found to be significant univariately (P < 0.05) were placed in multivariate models.
In the multivariate analyses, a number of prognostic variables were measured at baseline and treated as either categorical or continuous. Nevirapine hypersensitivity combined with HCV coinfection (i.e., combined variable HSR/HCV), plasma HIV-1 RNA levels (≥ 100 000 versus < 100 000 copies/ml), male gender (yes versus no), IDU (yes versus no), being Caucasian (yes versus no), aboriginal status (yes versus no), concurrent use of didanosine and stavudine (yes versus no), diagnosis of AIDS (yes versus no), ALT baseline (≥ 1.25X ULN versus < 1.25X ULN) and AST baseline (≥ 1.25X ULN versus < 1.25X ULN) were treated as fixed binary variables. Age (per 10 years increment), physician experience (per 100 patients increment), CD4 cell count (per 100 cell/μl decrement) and adherence (10% decrement) were treated as continuous variables.
All analyses were performed using SAS software version 8.02 (SAS, Cary, North Carolina, USA). All tests with a P value < 0.05 indicated that an association was statistically significant.
|
|
| |
| |
|
|
|
