Highly active antiretroviral treatment (HAART) regimens containing 3 antiretroviral drugs have been shown in clinical trials to maximally suppress plasma HIV-1 RNA levels within 6 to 12 months of initiation of therapy in 70% to 90% of antiretroviral-naive patients infected with human immunodeficiency virus type 1 (HIV-1). 1,2 However, findings from clinical trials evaluating HAART may not accurately reflect the real-world efficacy and tolerability of these medications in patients seen in clinical practice. Indeed, the exclusion criteria for clinical trials may select out a great many patients for such reasons as restrictions on magnitude of baseline HIV-1 RNA level, particular concurrent medications and diseases, drug or alcohol abuse, intravenous drug use, previous history of repeated nonadherence, history of failure to respond to certain antiretroviral agents, or limits on patient age range. Ethnic minorities, women, and injection drug users are often underrepresented in efficacy and safety studies although these populations are the ones in which the greatest increase in cases of HIV have been diagnosed during the past decade. 3–5 Finally, clinical practice studies occasionally have revealed antiretroviral toxicity profiles, 6 efficacy, 7 and adherence 8 to be at variance with what has been reported in clinical trials.
The nucleoside analogue abacavir has been shown to be useful when combined with other nucleoside reverse transcription inhibitors (NRTIs) or with nonnucleoside reverse transcription inhibitors (NNRTIs) or protease inhibitors (PIs) as part of new or intensification therapy. 9–12 Abacavir is generally well tolerated, with mild to moderate nausea being the primary adverse event observed. Hypersensitivity reactions have been reported rarely (an incidence of approximately 5%) following initiation of abacavir-containing regimens. These reactions usually (over 93% of cases) occur within the first 6 weeks of the start of abacavir treatment and typically resolve without sequelae once abacavir is discontinued. 13
To date, most of the safety and efficacy information about abacavir-containing regimens has come from controlled clinical trials. The purpose of the Ziagen Optimal Regimen and Resistance Observational (ZORRO) trial was to assess the clinical safety/tolerability and efficacy of abacavir as part of HAART regimens in a very large heterogeneous population of HIV-infected patients treated at US-based clinical practice sites. The primary objective of ZORRO was to evaluate the types and frequencies of serious adverse events in patients receiving abacavir in the clinical setting. To do this, study sites were chosen geographically, with no more than 5 enrolled patients allowed per site, to acquire a representative sample of patients. Most study sites had little or no clinical research experience, as is consistent with the average clinical practice setting. An assessment of risk factors possibly contributing to the occurrence of abacavir-associated hypersensitivity reactions was also performed.
Male and nonpregnant female outpatients were eligible for study enrollment if they were at least 18 years, had HIV-1 infection documented by HIV-1 antibody enzyme-linked immunosorbent assay and confirmed by Western blot test of HIV-1 antibody or positive HIV-1 blood culture, and CD4 cell counts ≥50/mm3. Women of childbearing potential had to be willing to use an adequate method of contraception during the study. Patients could be either antiretroviral-naive or antiretroviral-experienced. Among the antiretroviral-experienced group, patients were subclassified as early therapy-experienced if they had a viral load > 400 copies/mL and were failing their first antiretroviral regimen, or more therapy-experienced if they had a viral load > 1000 copies/mL and were failing their second or subsequent treatment regimen. At the discretion of the investigator, more experienced patients had the opportunity to have a baseline genotype interpreted using the VIRCOGEN software program (Virco Group, Research Triangle Park, NC). This program analyzed genetic sequences and generated a virtual phenotype to match the genotype. In genotyped patients, the prevalence of abacavir resistance-associated mutations was ascertained (the findings of this aspect of the ZORRO trial are presented in a separate publication). Patients were excluded from this clinical trial if they had previous abacavir experience, previously showed phenotypic resistance to abacavir (50% inhibitory concentration indicating >8-fold change from wild type), showed lack of susceptibility to at least 2 other agents, or were pregnant or breast-feeding females.
Study Design and Treatment
This phase IV, prospective, open-label, multicenter cohort study was conducted between May 2000 and February 2001 at 265 outpatient treatment sites in the United States. Study candidates underwent a medical history and physical examination at the preentry visit (2 weeks prestudy [week-2]) for determination of study eligibility. All enrolled patients were assigned for 16 weeks to one abacavir 300-mg tablet twice daily in combination with recommended doses of other antiretroviral agents, the choice of the latter being totally at the discretion of the investigator. Abacavir was supplied as Ziagen (Glaxo Wellcome, Research Triangle Park, NC). All patients provided written informed consent to participate, and the protocol for the study was approved by either a central institutional review board in Kansas City, Missouri or the institutional review board operating at the treatment site.
Patients were monitored for adverse events, laboratory abnormalities, and any HIV-related illnesses at baseline and at weeks 8 and 16. A telephone contact or a clinic visit was also required at weeks 3 to 5 based on standard clinic practice. Adverse events were classified as nonserious if they were unfavorable or unintended signs or symptoms that were not life-threatening and did not incur disability/incapacity, inpatient hospitalization, or prolongation of existing hospitalization. Conversely, an adverse event was classified as serious if it incurred any of the above or resulted in death. Adverse events were elicited from patients by asking them the following non-leading open-ended questions: “Have you had any medical problems since your last visit/assessment?” and “Have you had to take any new medicines other than those given to you in this study since your last visit or assessment?” Patients who experienced AIDS Clinical Trials Group grade 4 toxicities 14 were withdrawn from the study, as were patients who failed to achieve resolution of AIDS Clinical Trials Group grade 3 toxicities 14 following interruption of therapy. A hypersensitivity reaction to abacavir was considered when the following multiorgan signs and symptoms appeared in a patient: fever, rash, gastrointestinal symptoms (nausea, vomiting, diarrhea, or abdominal pain), lethargy, or malaise with or without concomitant respiratory symptoms (dyspnea, sore throat, cough), musculoskeletal symptoms (myalgia, myolysis, arthralgia), headache, paresthesia, and edema. All suspected hypersensitivity reactions were considered serious adverse events regardless of changes in therapy. Suspected hypersensitivity reactions required discontinuation of abacavir without subsequent rechallenge.
The primary efficacy measure was the proportion of patients who attained plasma HIV-1 RNA levels below the 400-copy/mL lower limit of quantitation of the HIV-1 MONITOR Version 1.0 polymerase chain reaction assay (Roche, Nutley, NJ) at 16 weeks. Change in HIV-1 RNA from baseline was measured and tabulated at 8 and 16 weeks and compared between treatment-naive and treatment-experienced patients. HIV-1 RNA levels were also assessed by calculation of area under the HIV-1 RNA curve minus baseline. The secondary efficacy measure evaluated in this study was change in CD4 lymphocyte cell count from baseline at weeks 8 and 16. A subanalysis of efficacy in early therapy-experienced versus more therapy-experienced HIV-infected patients was not done because this information was not collected by the case report form.
Data from the intent-to-treat (ITT) population were analyzed. An ITT: observed analysis was used for assessing proportion of patients with plasma HIV-1 RNA level <400 copies/mL. In this type of analysis, only available assessments were used (no imputation for missing values), regardless of whether the patient was still receiving their original therapy. An ITT: missing equals failure analysis was not deemed appropriate because of the nature of this type of trial; the lack of clinical investigation experience by personnel at the study sites was expected to result in a much higher rate of missing data compared with what is seen in controlled clinical trials, thus making an ITT: missing equals failure analysis of uncertain clinical relevance. No formal power calculations were performed, although a large study population of approximately 750 patients was sought for enrollment based on the 3% to 5% incidence of abacavir-related hypersensitivity reactions reported in clinical trials and the width of a 95% confidence interval (CI) on this incidence rate being about ±1%. About one third of the patients enrolled at study sites were to be antiretroviral-naive, and about two thirds were to be antiretroviral-experienced. Descriptive statistics were applied to summarize demographics and baseline characteristics. Means (±SD) and medians were calculated for plasma HIV-1 RNA levels and CD4 cell counts and were tabulated along with minimum and maximum values. Differences between antiretroviral-naive and antiretroviral-experienced patients regarding relative change from baseline in HIV-1 RNA and CD4 cell counts were examined, but were not statistically analyzed. Risk factors for the hypersensitivity reaction were evaluated prospectively using logistic regression models to calculate odds ratios, with 95% CI, in patients who developed such reactions compared with patients who did not. If the 95% CI for the odds ratio did not include 1, statistical significance for the covariate was affirmed, corresponding to a P value of <0.05. Each covariate was studied using a univariate logistic regression model. Covariates evaluated in this risk factor analysis included sex, race, baseline CDC classification, prior antiretroviral experience (naive compared with experienced), concurrent NNRTI, concurrent PI, age, baseline HIV-1 RNA, baseline CD4 cell count, history of drug allergy, history of food allergy, and family members(s) with drug allergy, current asthma, hay fever, or eczema. In addition, a multivariable logistic regression model that select for significant covariates by the stepwise selection method (with an entry alpha level of 0.10 and an exit alpha level of 0.05) was explored. All of the aforementioned covariates were explored in the multivariable model, except for family member with current asthma, hay fever, and eczema. These 3 covariates were not included because of the vast amount of missing data (>73% of the ITT population).
Patient Characteristics and Disposition
Of the 833 patients enrolled in the study, 791 had documented evidence of having taken at least one dose of abacavir (ITT safety population). Most of the patients were males (84%) and antiretroviral-experienced (62%), about one half were Caucasian (48%), and one third (33%) were African-American (Table 1). Median plasma HIV-1 RNA level was 4.34 log10 copies/mL and median CD4 cell count was 321 cells/mm3. Slightly over one half of the patients were HIV Category A (57%), and less than one third (28%) were Category C. In the antiretroviral-experienced subset, 85% of patients had previously been treated with NRTIs, 66% had received PIs, and 33% had received NNRTIs (Table 2). Prior to the study, the most commonly used NRTIs were zidovudine (56%) and lamivudine (53%); the most commonly used PIs were indinavir (34%) and nelfinavir (31%).
A total of 158 different abacavir-containing regimens were prescribed. Abacavir + COM was the regimen most prescribed in the total patient population (31.0%), followed by abacavir + other NRTIs + any NNRTI (22.4%), abacavir + other NRTIs + any PI (26.0%), and abacavir + NRTIs other than COM or lamivudine + zidovudine (8.2%). A greater proportion of antiretroviral-naive patients than antiretroviral-experienced patients received abacavir + COM during the study (53.4% vs. 17.3%).
Of the 791 patients in the ITT safety population, 583 (74%) completed the study and were evaluable; 200 patients (25%) discontinued treatment for the reasons delineated in Table 1.
Table 3 lists treatment-related adverse events reported in ≥1% of patients. The most frequently observed abacavir-related nonserious adverse events were nausea (12%), fatigue (4%), rash (4%), headache (3%), and diarrhea (3%). Most of these adverse events were mild to moderate in severity.
Suspected abacavir-related hypersensitivity reactions were reported at a rate of 8% by the investigators. However, when cases were excluded in which abacavir was continued or reintroduced uneventfully, or in which a definitive alternative diagnosis was made or only single symptoms were present, the rate of suspected hypersensitivity reactions was adjusted downward to 7% (59/791). Seven of the 59 patients (12%) with suspected abacavir hypersensitivity reactions and 2 of 8 with possible hypersensitivity reactions received nevirapine concurrently with abacavir during the study. Hypersensitivity reactions included fever as a manifestation in 39 cases (66%), malaise in 38 (64%), gastrointestinal symptoms in 36 (61%), rash in 32 (54%), myalgia or weakness in 21 (36%), respiratory symptoms (cough or dyspnea) in 16 (27%), and cardiac symptoms (tachycardia or hypotension) in 13 (22%). The most frequently reported triad of symptoms seen in the hypersensitivity cases was fever/chills/malaise (17 patients [29%]). Time to onset of suspected abacavir-related hypersensitivity reactions varied from 1 to 100 days (mean, 21.5 days) following initiation of abacavir [onset ≥ 75 days postinitiation in 5 patients]), and time until resolution of hypersensitivity signs/symptoms varied from 2 to 41 days (mean, 10.4 days) following discontinuation of abacavir. In all patients who developed suspected hypersensitivity reactions, signs and symptoms of hypersensitivity resolved without sequelae after abacavir was stopped.
There were no deaths attributable to abacavir during this study.
Analysis of Risk Factors for Suspected Abacavir-Related Hypersensitivity Reaction
Table 4 shows the estimates of the odds ratios from the univariate analyses, exploring the association of abacavir-related hypersensitivity reaction with each covariate. The odds of developing a suspected hypersensitivity reaction to abacavir were 1.77 (95% CI, 1.05–2.97) for patients with a history of a drug allergy, and 2.59 (95% CI, 1.15–5.82) for patients with a history of food allergy, when compared with patients without such allergies (P < 0.05). Age, sex, race, baseline HIV-1 RNA, baseline CD4, patient history of asthma or eczema, and family history of drug allergy, food allergy, asthma, or eczema did not appear to be associated with hypersensitivity to abacavir.
In the multivariable logistic regression analysis that looked at all of the risk factors at once (controlling for confounding among the risk factors), food allergy was the only covariate that was selected to remain in the final model based on its statistically significant contribution to predict the increased risk of hypersensitivity. From this model, the odds of developing a suspected hypersensitivity reaction to abacavir were 3.41 (95% CI, 1.48–7.87) for patients with food allergy, when compared with patients without this allergy (P = 0.004).
At baseline, median HIV-1 RNA was 4.34 log10 copies/mL and was higher in antiretroviral-naive patients than in antiretroviral-experienced patients (4.72 vs. 3.98 log10 copies/mL) (Fig. 1A). By week 16, median HIV-1 RNA was reduced below baseline by 1.64 log10 copies/mL in the total treatment population (Figure 1B). The decrease in HIV-1 RNA was approximately 2-fold greater in antiretroviral-naive patients than antiretroviral-experienced patients (−2.40 vs. −1.05 log10 copies/mL) (Figure 1B), as was the median area under the HIV-1 RNA curve minus baseline (−1.64 vs. −0.76 log10 copies/week/mL). ITT: observed analysis showed the percentage of patients achieving HIV-1 RNA <400 copies/mL was very similar in the total (68%), antiretroviral-naive (69%), and antiretroviral-experienced (67%) populations (Figure 1C). At least a 1-log10 decrease in HIV-1 RNA from baseline occurred by week 16 in 64%, 92%, and 50% of patients in these groups, respectively.
Immunologic Measurements of Efficacy
At baseline, the median CD4 count was 321 cells/mm3 in the total population, and was similar in antiretroviral-naive and antiretroviral-experienced patients (287 and 348 cells/mm3, respectively) (Figure 2A). By week 16, CD4 counts had increased above baseline by 78 cells/mm3 in the total population. CD4 count increases above baseline were approximately 2-fold higher in antiretroviral-naive than antiretroviral-experienced patients (120 vs. 54 cells/mm3) (Figure 2B), as was the median area under the HIV-1 RNA curve minus baseline (75.00 vs. 37.25 cells/week/mm3).
The results of this study show that the safety, tolerability, and efficacy of abacavir-containing HAART in patients treated at clinical practice sites mirrors that reported previously in controlled clinical trials. Unlike most studies of abacavir to date, which have evaluated abacavir’s safety and efficacy when combined with the nucleosides zidovudine and lamivudine, this study pooled safety and efficacy data from 158 different abacavir-containing regimens that included other NRTIs, NNRTIs, and PIs. As only 31% of the total study population received the twice-daily abacavir/lamivudine/zidovudine regimen, one might have expected a different safety profile to emerge because of the additive toxicities of the other concomitantly administered drugs. However, a different profile was not seen, and gastrointestinal adverse events were the most frequently reported events just as they have proven to be in clinical trials.
The incidence of nonserious abacavir related adverse events was comparable to that reported in the abacavir expanded access study in which abacavir was combined with many other types of antiretroviral agents (n = 2527 [nonserious adverse event database]) (nausea, 12.2%; rashes, 10.9%; diarrhea, 9.5%; malaise/fatigue, 8%; and fever, 6.5%). 15 As for serious adverse events, a suspected abacavir-related hypersensitivity reaction was observed in 7% of patients. This fits within the 0% to 14% range observed in a safety analysis by Symonds et al 16 of all clinical trials (through the second quarter of 2000) involving patients with ≥24 weeks of abacavir exposure (n = 5332 from 25 trials). However, this incidence figure is higher than the rate of suspected hypersensitivity reported in the abacavir-expanded access trial (4.3%; n = 31,897 [serious adverse event database]) 15 and an analysis of this reaction by Hernandez et al 17 as reported in clinical trials in which abacavir was combined only with Combivir (3.9%; n = 540 from 4 trials). Several possible reasons may account for this discrepancy. First, the likeliest explanation is that most of the investigators who assessed suspected abacavir-related hypersensitivity reactions in this study were less experienced than investigators who have participated in many controlled clinical trials. This inexperience may have prompted overreporting of hypersensitivity, with hypersensitivity applied to adverse events in which certain key signs and symptoms of the reaction could have been absent. Second, unlike the expanded access study, our study included many antiretroviral-naive patients, who, according to one risk analysis, 16 may be more likely to have an abacavir-related hypersensitivity reaction than therapy-experienced patients. However, this explanation is unlikely because our risk factor analysis failed to identify an increased risk for naive patients when compared with experienced patients. Third, our study included a substantial number of patients who received the NNRTI nevirapine concurrently with abacavir (12%). As nevirapine has been reported in a safety review to cause a rash in approximately 16% of patients, 18 it is possible that nevirapine-related rashes may have been factored into the abacavir-associated hypersensitivity incidence figure, thereby pushing it upward.
Our risk factor analysis of abacavir-related hypersensitivity is the first evaluation of this particular safety issue that has been conducted prospectively. As was observed by Symonds et al 16 in their multivariate risk analysis (n = 5248), we found that age, gender, CDC class, concurrent NNRTI, concurrent PI, baseline HIV-1 RNA, and baseline CD4 cell count did not affect the risk of the abacavir-related hypersensitivity reaction. However, unlike Symonds et al, we did not see a significantly lower risk of abacavir-related hypersensitivity in antiretroviral-experienced versus antiretroviral-naive patients or in African-Americans versus other races. Although the reason for this discrepancy is unclear, it is possible that differences in risk regarding certain covariates may only be apparent when very large populations are analyzed (the population analyzed by Symonds was over 6-fold larger than ours). The risk analysis by Hernandez et al 17 (n = 540) reported a higher incidence of the hypersensitivity reaction (2.77 odds ratio) in Hispanics than other races. However, our analysis did not include enough Hispanic patients to permit this analysis.
Our risk analysis indicated a higher incidence of abacavir-related hypersensitivity reactions in patients with a history of food allergy. This covariate was not included in earlier risk analyses. Allergy and immunologic factors have been postulated to contribute to the mechanism of abacavir-related hypersensitivity reactions, although genetic and metabolic factors may also be involved. 13 These findings from different studies suggest that no individual risk factor is useful to predict which patients are likely to develop a hypersensitivity reaction to abacavir. Rather, all patients starting abacavir should be counseled about how to recognize the syndrome to allow its prompt diagnosis and discontinuation of the drug.
This study did not evaluate genetic susceptibility to abacavir-related hypersensitivity reactions. However, MHC region typing performed in 200 abacavir-exposed HIV-infected patients in the Western Australian HIV Cohort Study showed that HLA-B*5701 was present in 14 (78%) of the 18 patients with abacavir-related hypersensitivity, and in 4 (2%) of the 167 abacavir-tolerant patients (odds ratio 117 [95% CI, 29–481], P < 0.0001). 19 The HLA-DR7 and HLA-DQ3 combination was found in 13 (72%) of hypersensitive and 5 (3%) of the abacavir-tolerant patients (73 [20–268], P < 0.0001). Withholding abacavir in patients with all 3 MHC-specific alleles was estimated to reduce the prevalence of hypersensitivity from 9% to 2.5% without inappropriately denying abacavir to any patient. Hetherington et al 20 did a retrospective case-control study to identify multiple markers in the vicinity of HLA-B associated with abacavir-related hypersensitivity reactions. HLA-B57 was present in 39 (46%) of 84 patients versus 4 (4%) of 113 controls (P < 0.0001). However, because of low numbers of women and other ethnic groups enrolled, the findings from both studies related largely to white men. The lower sensitivity of HLA-B57 for predicting abacavir-related hypersensitivity identified in this study compared with that of the Western Australian HIV Cohort Study highlights that predictive values for markers can vary across populations.
Our study was unique in that efficacy and safety were evaluated and compared in both antiretroviral-naive and antiretroviral-experienced patients. Controlled clinical trials have usually evaluated either one or the other type of patient, but not both types at the same time. In general, virologic and immunologic response is more pronounced in antiretroviral-naive patients than antiretroviral-experienced patients because the latter have developed response-limiting mutations and cross-resistance to drugs included in HAART regimens. 21 Consistent with this pattern, in our study reduction in HIV-1 RNA and elevation in CD4 cell counts were greater among antiretroviral-naive than antiretroviral-experienced patients. However, the percentage of patients achieving plasma HIV-1 RNA levels <400 copies/mL was similar in the 2 populations (69% [naive] versus 67% [experienced]). This may have been due to 2 reasons. First, the percentage of our naive patients achieving HIV-1 RNA levels <400 copies/mL at week 16 was lower than that reported in clinical trials evaluating twice-daily abacavir 300 mg in combination with twice-daily zidovudine 300 mg and lamivudine 150 mg (generally >80% in ITT: observed and as-treated analyses 22–27. The poorer virologic suppression in the naive population may have occurred because abacavir combinations other than abacavir/zidovudine/lamivudine (used by approximately 70% of our patients) may be less effective. Second, many of the patients who were classified as antiretroviral-experienced in this study were actually early-experienced, that is, they had limited prior experience with antiretroviral agents. Therefore, the magnitude of virologic suppression among the early-experienced patients may have been greater (making the percentage <400 copies/mL higher in the total experienced group) than has been observed in studies evaluating patients with more antiretroviral experience. Katlama et al 11 similarly found no significant differences in the relative percentages of antiretroviral-naive and antiretroviral-experienced patients achieving HIV-1 RNA levels <400 copies/mL in a 16-week clinical trial in which abacavir 300 mg twice daily was added to stable background antiretroviral therapy of various NRTIs, NNRTIs, and PIs rather than just zidovudine and lamivudine.
The design of this study permitted an evaluation of abacavir-containing regimens in a real world setting, reflecting the safety and efficacy of these regimens outside the constraints of a randomized clinical trial. The population evaluated was larger than that included in a controlled clinical trial and this permitted an analysis of risk factors for the abacavir-related hypersensitivity reaction. No more than 5 patients could be enrolled per study site, and study sites were scattered geographically to be representative of HIV infection in the US population. Thus, biases that could occur from uneven enrollment across geographical study sites were unlikely to occur. Besides their value in assessing relatively rare safety issues, clinical experience trials have allowed more thorough evaluations of medication-taking habits and preference in HIV-infected patients. 28 With respect to efficacy, clinical experience trials have occasionally shown higher 29 or lower efficacy 7 with some antiretroviral agents or combinations than has been reported in controlled clinical trials.
Because this study was conducted over only a 16-week period, this may have resulted in some limitations. The safety and efficacy profiles possibly could have changed if the study had been conducted for a longer period. However, as 93% of abacavir-related hypersensitivity reactions occur within 6 weeks after the start of abacavir (median, 11 days), 13 the study was long enough to evaluate this adverse event. In view of the established direct relationship between antiretroviral adherence and virologic suppression, 30 an assessment of adherence in this study would have been valuable, especially to see if differences in adherence were observable between the antiretroviral-naive and antiretroviral-experienced groups or between particular abacavir-containing combinations. Inclusion of more women in this study (16% were included) would have made the study population more representative of today’s HIV epidemic. Lastly, it would have been interesting to see how virologic suppression would have compared between the antiretroviral-naive and antiretroviral-experienced groups if an assay with a 50-copy/mL lower limit of quantitation had been used to evaluate HIV-1 RNA levels in place of, or in addition to the 400-copy/mL lower limit of quantitation assay we used.
In conclusion, the efficacy and safety profile of abacavir-containing HAART in clinical practice patients appears similar to that previously reported in patients evaluated in controlled clinical trials.
The authors thank the study participants and study investigators and and the following study team personnel: Margaret Becker, Irvie Bullock, Julie Fleming, Carol Humphries, and Anne Stokley. We also thank Gary E. Pakes, PharmD, for his assistance in the writing of this manuscript.
1. Deeks SG, Hecht FM, Swanson M, et al. HIV RNA and CD4 cell count response to protease inhibitor therapy in an urban AIDS clinic: response to both initial and salvage therapy. AIDS. 1999; 13:F35–F43.
2. Lucas GM, Chaisson RE, Moore RD. Highly active antiretroviral therapy in a large urban clinic: risk factors for virologic failure and adverse drug reactions. Ann Intern Med. 1999; 131:81–87.
3. Centers for Disease Control and Prevention (CDC). HIV/AIDS Surveill Rep. 2000; 12( 2):1–44.
4. UNAIDS/WHO. AIDS Epidemic Update, December 2000.
5. Gifford AL, Cunningham WE, Heslin KC, et al. Participation in research and access to experimental treatments by HIV-infected patients. N Engl J Med. 2002; 346:1373–1382.
6. Moreno A, Perez-Elias MJ, Casado JL, et al. Toxicity profile of antiretroviral drugs in naive patients starting highly active antiretroviral therapy in routine clinical practice. Antiviral Ther. 2001; 6(suppl 4):63.
7. Knobel H, Guelar A, Carmona A, et al. Virologic outcome and predictors of virologic failure of highly active antiretroviral therapy containing protease inhibitors. AIDS Patient Care STDs. 2001; 15:193–199.
8. Brook MG, Dale A, Tomlinson D, et al. Adherence to highly active antiretroviral therapy in the real world: experience of twelve English HIV units. AIDS Patient Care STDs. 2001; 15:491–494.
9. Hervey PS, Perry CM. Abacavir—a review of its clinical potential in patients with HIV infection. Drugs. 2000; 60:447–479.
10. McMahon D, Lederman M, Haas DW, et al. Antiretroviral activity and safety of abacavir in combination with selected HIV-1 protease inhibitors in therapy-naive HIV-1-infected adults. Antiviral Ther. 2001; 6:105–114.
11. Katlama C, Clotet B, Plettenberg A, et al. The role of abacavir (ABC, 1592) in antiretroviral therapy-experienced patients: results from a randomized, double-blind trial. AIDS. 2000; 14:781–789.
12. Khanna N, Klimkait T, Schiffer V, et al. Salvage therapy with abacavir plus a nonnucleoside reverse transcriptase inhibitor and a protease inhibitor in heavily pre-treated HIV-1-infected patients. AIDS. 2000; 14:791–799.
13. Hetherington S, McGuirk S, Powell G, et al. Hypersensitivity reactions during therapy with the nucleoside reverse transcriptase inhibitor abacavir. Clin Ther. 2001; 23:1603–1614.
14. Table for Grading of Adult Adverse Experiences: Washington, DC: National Institute of Allergy and Infectious Diseases, Division of AIDS, August 1992.
15. Kessler HA, Johnson J, Follansbee S, et al. The abacavir expanded access program for adult patients with HIV-1 infection. Clin Infect Dis. 2002; 34:535–542.
16. Symonds W, Cutrell A, Edwards M, et al. Risk factor analysis of hypersensitivity reactions to abacavir. Clin Ther. 2002; 24:565–573.
17. Hernandez J, Williams V, Sawyerr G, et al. The rate of hypersensitivity reactions to abacavir is similar in under-represented populations and incarcerated subjects (GW Protocols NZTA4002, NZTA4005, NZTA4006 & NZTA4007). Presented at the 5th International Congress on Drug Therapy in HIV Infection, Glasgow, UK, October 22–26, 2000. Poster 184.
18. Pollard RB, Robinson P, Dransfield K. Safety profile of nevirapine, a nonnucleoside reverse transcriptase inhibitor for the treatment of human immunodeficiency virus infection. Clin Ther. 1998; 20:1071–1092.
19. Mallal S, Nolan D, Witt C, et al. Association between presence of HLA-B*5701, HLA-DR7, and HLA-DQ3 and hypersensitivity to HIV-1 reverse-transcriptase inhibitor abacavir. Lancet. 2002; 359:727–732.
20. Hetherington S, Hughes AR, Mosteller M, et al. Genetic variations in HLA-B region and hypersensitivity reactions to abacavir. Lancet. 2002; 359:1121–1122.
21. Panel on Clinical Practices for Treatment of HIV Infection. Guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents. July 14, 2003. Website: http://www.hivatis.org/trtgdlns.html
. Last accessed November 5, 2003.
22. Staszewski S, Keiser P, Montaner J, et al. Abacavir-lamivudine-zidovudine vs. indinavir-lamivudine-zidovudine in antiretroviral-naive HIV-infected adults: a randomized equivalence trial. JAMA. 2001; 285:1155–1163.
23. Matheron S, Descampa D, Boud F, et al. Triple nucleoside combination zidovudine/lamivudino/abacavir versus zidovudine/lamivudina/nelfinavir as first-line therapy in HIV-1-infected adults: a randomized trial. Antlvir Ther. 2003; 8:163–171.
24. Vibhagool A, on behalf of the CNA3014 International Study Team. Abacavir/Combivir( (ABC/COM) is comparable to indinavir/Combivir (IDV/COM) in HIV-1-infected antiretroviral therapy naive adults: results of a 48-week open-label study (CNA3014) [abstract/poster 063]. Program and abstracts of the 1st International AIDS Society (IAS) Conference on HIV Pathogenesis and Treatment. Buenos Aires, Argentina, July 8–11, 2001.
25. Bowonwatanuwong C, Mootsikapun P, Supparatpinyo K, et al. A randomised, open-label study to investigate abacavir (ABC) and lamivudine (3TC) as once daily (qd) components of a triple combination regimen (EPV40001). Presented at the 1st International AIDS Society (IAS) Conference on HIV Pathogenesis and Treatment, Buenos Aires, Argentina, July 8–11, 2001. Poster 004.
26. Fischl M, Greenberg S, Clumeck N, et al. Ziagen (abacavir, ABC, 1592) combined with 3TC & ZDV is highly effective and durable through 48 weeks in HIV-1-infected antiretroviral-therapy-naive subjects (CNAA3003). Presented at the 6th Conference on Retroviruses and Opportunistic Infections, Chicago, Illinois, January 31–February 4, 1999. Abstract 19.
27. Kirkland LR, Fischl MA, Tashima KT, et al. Efficacy, tolerability, and adherence in antiretroviral-naive incarcerated patients with HIV infection receiving directly-observed treatment with a lamivudine/zidovudine combination tablet plus abacavir twice daily. Clin Infect Dis. 2002; 34:511–5118.
28. DeJesus E, Pistole M, Fetchick R, et al. Norvir-indinavir combination evaluation. The NICE Study. Presented at the XIII International AIDS Conference, Durban, South Africa, July 9–14, 2000. Presentation/Poster WeOrlB482.
29. Carosi G, Castelli F, Suter F, et al. Antiviral potency of HAART regimens and clinical success are not strictly coupled in real life conditions: evidence from the MASTER-1 study. HIV Clin Trials. 2001; 2:399–407.
© 2004 Lippincott Williams & Wilkins, Inc.
30. DeMasi RA, Graham NM, Tolson JM, et al. Correlation between self-reported adherence to highly active antiretroviral therapy (HAART) and virologic outcome. Adv Ther. 2001; 18:163–173.