Therapeutic Drug Monitoring of the HIV/AIDS Drugs Abacavir, Zidovudine, Efavirenz, Nevirapine, Indinavir, Lopinavir, and Nelfinavir

Effects of cytochrome P450 2B6 and constitutive androstane receptor genetic variation on Efavirenz plasma concentrations among HIV patients in Kenya

The effects of genetic variation of cytochrome P450 2B6 (CYP2B6) and constitutive androstane receptor (CAR) on efavirenz (EFV) plasma concentration was evaluated among 312 HIV patients in Nairobi Kenya. The EFV plasma concentration at steady-state were determined using ultra-high-performance liquid chromatography with a tandem quadruple mass spectrometer (LC-MS/MS). Thirteen CYP2B6 (329G>T, 341T>C, 444 G>T/C, 15582C>T, 516G>T, 548T>G, 637T>C, 785A>G, 18492C>T, 835G>C, 1459C>T and 21563C>T) and one CAR (540C>T) single nucleotide polymorphisms (SNPs) were genotyped using real-time polymerase chain reaction. HIV drug resistance mutations were detected using an in-house genotypic assay. The EFV concentration of patients ranged from 4 ng/mL to 332697 ng/mL (median 2739.5 ng/mL, IQR 1878-4891.5 ng/mL). Overall, 22% patients had EFV concentrations beyond therapeutic range of 1000-4000 ng/mL (4.5%% < 1000 ng/mL and 31.7% > 4000 ng/mL). Five SNPs (15582C>T, 516G>T, 785A>G, 983T>C and 21563C>T) were associated with higher EFV plasma concentration while 18492C>T with lower EFV plasma concentration (p<0.05). Strong linkage disequilibrium (LD) was observed for 15582C>T, 516G>T, 785A>G, 18492C>T, 983T>C, 21563C>T, 1459C>T and CAR 540C>T. Sixteen haplotypes were observed and CTGCTTCC, CTGCTTCT, TTGCTTCT and CGACCCCT were associated with high EFV plasma concentration. In multivariate analysis, factors significantly associated with EFV plasma concentration included; the presence of skin rash (β = 1379, 95% confidence interval (CI) = 3216.9-3416.3; p < 0.039), T allele of CYP2B6 516G>T (β = 1868.9, 95% CI 3216.9-3416.3; p < 0.018), the C allele of CYP2B6 983T>C (β = 2638.3, 95% CI = 1348-3929; p < 0.0001), T allele of CYP2B6 21563C>T (β = 1737, 95% CI = 972.2-2681.9; p < 0.0001) and the presence of 5 to 7 numbers of SNPs per patient (β = 570, 95% CI = 362-778; p < 0.0001) and HIV viral load ≤1000 cells/mL (β = -4199.3, 95% CI = -7914.9 --483.6; p = 0.027). About 36.2% of the patients had EFV plasma concentrations beyond therapeutic window, posing high risk of treatment failure or toxicity. The SNPs of CYP2B6 516G>T, CYP2B6 983T>C, 21563C>T, presence of higher numbers of SNPs per patient and haplotypes CTGCTTCC, CTGCTTCT, TTGCTTCT and CGACCCCT could efficiently serves as genetic markers for EFV plasma concentration and could guide personalization of EFV based ART treatment in Kenya.

A fast-screening dispersive liquid-liquid microextraction-gas chromatography-mass spectrometry method applied to the determination of efavirenz in human plasma samples

We demonstrate the suitability of a fast, green, easy-to-perform, and modified sample extraction procedure, i.e., dispersive liquid-liquid microextraction (DLLME) for the determination of efavirenz (EFV) in human plasma. Data acquisition was done by gas chromatography-mass spectrometry (GC-MS) in the selected ion monitoring (SIM) mode. The simplicity of the method lies in, among others, the avoidance of the use of large organic solvent volumes as mobile phases and non-volatile buffers that tend to block the plumbing in high-performance liquid chromatography (HPLC). Chromatographic and mass spectral parameters were optimized using bovine whole blood for matrix matching due to insufficient human plasma. Method validation was accomplished using the United States Food and Drug Administration (USFDA) 2018 guidelines. The calibration curve was linear with a dynamic range of 0.10-2.0 μg/mL and an R² value of 0.9998. The within-run accuracy and precision were both less than 20% at the lower limit of quantification (LLOQ) spike level. The LLOQ was 0.027 μg/mL which compared well with some values but was also orders of magnitude better than others reported in the literature. The percent recovery was 91.5% at the LLOQ spike level. The DLLME technique was applied in human plasma samples from patients who were on treatment with EFV. The human plasma samples gave concentrations of EFV ranging between 0.14-1.00 μg/mL with three samples out of seven showing concentrations that fell within or close to the recommended therapeutic range.

Steady-State Pharmacokinetics of Nevirapine in HIV-1 Infected Adults in India

BACKGROUND AND OBJECTIVES

A variety of demographic factors, sex, and degree of immunosuppression can influence antiretroviral drug concentrations. The authors studied the influence of immune status, sex, and body mass index (BMI) on the steady-state pharmacokinetics of nevirapine delivered as a fixed-dose combination in HIV-1-infected patients in India.

METHODS

Twenty-six HIV-1-infected adult patients undergoing treatment with nevirapine-based highly active antiretroviral therapy regimens participated in the study. Pharmacokinetic variables were compared between patients divided based on CD4 cell counts, sex, and BMI.

RESULTS

Patients with higher BMI had lower peak and trough concentration and exposure of nevirapine than those with lower BMI; none of the differences in the pharmacokinetic variables of nevirapine between the various patient groups was statistically significant.

CONCLUSIONS

Patients' immune status, sex, or BMI had no impact on the pharmacokinetics of nevirapine. Plasma nevirapine concentrations were maintained within the therapeutic range of the drug in the majority of the patients.

Effect of CYP2B6 Gene Polymorphisms on Efavirenz Plasma Concentrations in Chinese Patients with HIV Infection

Objectives

The main aim of this study was to investigate the effect of CYP2B6 gene polymorphisms on efavirenz (EFV) plasma concentrations in Han Chinese patients with human immunodeficiency virus (HIV) infection.

Methods

In total, 322 patients were recruited for study. EFV plasma concentrations at steady-state were determined using high-performance liquid chromatography. Genotyping for seven single nucleotide polymorphisms (SNPs), including 171+967C>A, 171+3212C>T, 171+4335T>C, 516G>T, 785A>G, 1295-913G>A, and *1355A>G of CYP2B6, was performed using ligase detection reaction (LDR). SPSS 18.0 and Haploview 4.2 were applied for statistical analyses.

Results

The average EFV concentration of patients was 2.35±2.09 μg/mL. Overall, 22% patients displayed EFV concentrations out of the therapeutic range of 1–4 μg/mL (13.1% < 1 μg/mL, 9.3% > 4 μg/mL). We observed significant association of 171+967C>A, 171+4335T>C, 516G>T, 785A>G and *1355A>G with high plasma EFV levels (p<.01). The predictive accuracy values of 171+4335CC, 516TT and 785GG for EFV concentrations > 4 μg/mL were 56.7%, 56.7% and 60%, respectively. We observed strong linkage disequilibrium for 171+967C>A, 171+4335T>C, 516G>T and 785A>G, resulting in five haplotypes. The frequencies of the five haplotypes (high to low) were as follows: CCTG (0.328), ACTG (0.280), ACCT (0.189), ATTG (0.186) and ACCG (0.017). The frequency of CCTG (0.524) in patients with EFV plasma concentrations < 1 μg/mL was significantly higher than that in other patient groups, while that of ACCT (0.733) was significantly higher in patients with EFV concentrations > 4 μg/mL, relative to other patient groups. Average EFV concentrations of patients carrying ACTG (1.78 μg/mL), ACCT (7.50 μg/mL), and ATTG (1.92 μg/mL) haplotypes were markedly higher than those of patients carrying the CCTG haplotype. The predictive accuracy of ACCT for EFV > 4 μg/mL was 81%.

Conclusions

Chinese patients administered standard doses of EFV require therapeutic drug monitoring or personalized medication management. Based on the current findings, we propose that 171+4335T>C, 516G>T, 785A>G and haplotype ACCT may be effectively used as genomic markers for EFV, which should aid in improving the efficacy of EFV-containing treatments and reduce the incidence of adverse reactions.

A simple, rapid, economical, and practical method for the determination of efavirenz in plasma of Chinese AIDS patients by reverse phase high-performance liquid chromatography with ultraviolet detector

This study aimed to develop a reverse phase high-performance liquid chromatographic (RP-HPLC) method for the determination of efavirenz in human plasma and to use it for determining the concentrations of efavirenz in Chinese AIDS patient. A simple mobile phase consisting of 0.01 mol/L NaH2PO4 solution and acetonitrile (38:62, V/V) was pumped at a flow rate of 1.0 mL/min through a reverse phase Diamonsil C18 column maintained at 30°C. Diazepam was used as an internal standard and monitored with efavirenz at 247 nm. The protein of 100 μL plasma sample was precipitated before 20 μL of the supernatant was directly injected into the column. The linear response over the concentration ranges 0.10-20.0 μg/mL was obtained and the linear regression equations was Y = 2.2873X ‒ 0.1449 (r = 0.9999). The intra-day and inter-day precisions (1.9-2.6%, 2.2-7.2%, respectively), the relative and absolute recovery (99.3-106.3%, 75.6-80.3%, respectively) met the international standards. Stability of plasma samples were evaluated for short-term (ambient temperature for 16 h) and long-term (-20°C for 30 days) storage conditions and were found to be stable. The mean plasma concentration of efavirenz of the 406 patients was 2.21 ± 1.95 μg/mL, 77.3% of which were within the therapeutic window (1-4 μg/mL), 15.1% were below the window, and 7.6% were over it. In conclusion, the method had advantages of convenience, rapidity, necessary accuracy and precision, high practicality and met the needs for therapeutic drug monitoring and the pharmacokinetic study of efavirenz, especially in underdeveloped countries. For Chinese AIDS patients, it was beneficial to use efavirenz under the guidance of therapeutic drug monitoring.

Targeting of AIDS related encephalopathy using phenylalanine anchored lipidic nanocarrier

Transport of the anti-HIV agents across the blood-brain barrier (BBB) is a prerequisite to treat acquired immunodeficiency syndrome (AIDS) related encephalopathy. In the present study, we explored facilitated transport of efavirenz (EFV, a non-nucleoside reverse transcriptase inhibitor) across BBB using phenylalanine anchored solid lipid nanoparticles (PA-SLN). PA (amino acid micro-nutrient) was used as a ligand which facilitated carrier mediated transport (CMT) via l-amino acid transporter i.e. LAT1 to traverse BBB. PA was coupled to SLN via amide linkage using carbodiimide chemistry and coupling was confirmed by comparative infrared spectroscopic analysis. SLNs (SLN and PA-SLN) were nanometric in size (around 150nm) and possessed good entrapment efficiency (around 70%). In vitro drug release revealed controlled release pattern for more than 24h. In vivo studies showed 2-3-folds and 7-8-folds accumulation of PA-SLN in brain as compared to SLN and EFV, respectively. Further, transcytosis studies confirmed capability of PA-SLN to cross BBB i.e. 10-fold higher transcytosis potential as compared to EFV. Fluorescence microscopic imaging reassured enhanced brain localization of PA-SLN. Thus, PA-SLN improved the EFV bioavailability and maintained therapeutic levels in the brain for an extended period of time that can result in significant eradication of the viral load therein. Such nutrient mediated drug targeting could bring forth advances in biocompatible and biodegradable drug delivery systems.

Pharmacogenetic testing and therapeutic drug monitoring are complementary tools for optimal individualization of drug therapy

Genetic factors contribute to the phenotype of drug response, but the translation of pharmacogenetic outcomes into drug discovery, drug development or clinical practice has proved to be surprisingly disappointing. Despite significant progress in pharmacogenetic research, only a few drugs, such as cetuximab, dasatinib, maraviroc and trastuzumab, require a pharmacogenetic test before being prescribed. There are several gaps that limit the application of pharmacogenetics based upon the complex nature of the drug response itself. First, pharmacogenetic tests could be more clinically applicable if they included a comprehensive survey of variation in the human genome and took into account the multigenic nature of many phenotypes of drug disposition and response. Unfortunately, much of the existing research in this area has been hampered by limitations in study designs and the nonoptimal selection of gene variants. Secondly, although responses to drugs can be influenced by the environment, only fragmentary information is currently available on how the interplay between genetics and environment affects drug response. Third, the use of a pharmacogenetic test as a standard of care for drug therapy has to overcome significant scientific, economic, commercial, political and educational barriers, among others, in order for clinically useful information to be effectively communicated to practitioners and patients. Meanwhile, the lack of efficacy is in this process is quite as costly as drug toxicity, especially for very expensive drugs, and there is a widespread need for clinically and commercially robust pharmacogenetic testing to be applied. In this complex scenario, therapeutic drug monitoring of parent drugs and/or metabolites, alone or combined with available pharmacogenetic tests, may be an alternative or complementary approach when attempts are made to individualize dosing regimen, maximize drug efficacy and enhance drug safety with certain drugs and populations (e.g. antidepressants in older people).

Risk factors for virological failure and subtherapeutic antiretroviral drug concentrations in HIV-positive adults treated in rural northwestern Uganda

BACKGROUND

Little is known about immunovirological treatment outcomes and adherence in HIV/AIDS patients on antiretroviral therapy (ART) treated using a simplified management approach in rural areas of developing countries, or about the main factors influencing those outcomes in clinical practice.

METHODS

Cross-sectional immunovirological, pharmacological, and adherence outcomes were evaluated in all patients alive and on fixed-dose ART combinations for 24 months, and in a random sample of those treated for 12 months. Risk factors for virological failure (>1,000 copies/ml) and subtherapeutic antiretroviral (ARV) concentrations were investigated with multiple logistic regression.

RESULTS

At 12 and 24 months of ART, 72% (n = 701) and 70% (n = 369) of patients, respectively, were alive and in care. About 8% and 38% of patients, respectively, were diagnosed with immunological failure; and 75% and 72% of patients, respectively, had undetectable HIV RNA (<400 copies/ml). Risk factors for virological failure (>1,000 copies/ml) were poor adherence, tuberculosis diagnosed after ART initiation, subtherapeutic NNRTI concentrations, general clinical symptoms, and lower weight than at baseline. About 14% of patients had low ARV plasma concentrations. Digestive symptoms and poor adherence to ART were risk factors for low ARV plasma concentrations.

CONCLUSION

Efforts to improve both access to care and patient management to achieve better immunological and virological outcomes on ART are necessary to maximize the duration of first-line therapy.

Effect of anticoagulants on plasma nevirapine and efavirenz concentrations

Antiretroviral drug measurements as part of therapeutic drug monitoring are performed in several human immunodeficiency virus (HIV) clinics in developed as well as developing countries. Heparin is routinely used as an anticoagulant for plasma drug estimations. EDTA plasma for drug estimations, if found suitable, would have several practical advantages. In this study, plasma nevirapine (NVP) and efavirenz (EFV) levels were compared in paired blood collected in heparin and EDTA Vacutainer tubes (Greiner Bio-one, Austria). The study involved HIV-infected individuals who were receiving antiretroviral treatment at the Government Hospital of Thoracic Medicine, Tambaram, Chennai, India. Their treatment consisted of administration of NVP or EFV, along with lamivudine and stavudine/zidovudine. Blood was collected at the same time in heparin and EDTA Vacutainer tubes. Plasma concentrations of NVP and EFV were estimated, according to validated methods, by high-performance liquid chromatography (HPLC). The mean NVP concentrations in heparin and EDTA plasma were 8.79 and 8.59 microg/mL, respectively. This difference was not statistically significant (P = 0.203). In the case of EFV, the corresponding values were 3.03 and 2.78 microg/mL, respectively; the difference was significant (P < 0.05). However, the percent variation in EFV between both types of plasma was <10%. Blood collected in heparin or EDTA Vacutainer tubes could be used for plasma NVP determination.

Abcg2/Bcrp1 mediates the polarized transport of antiretroviral nucleosides abacavir and zidovudine

The bioavailability and targeted distribution of abacavir (ABC) and zidovudine (AZT) to viral reservoirs may be influenced by efflux transporters. The purpose of this study was to characterize the interaction of these nucleoside reverse transcriptase inhibitors with the Abcg2/Bcrp1 transporter, the murine homolog of human breast cancer resistance protein (BCRP), using a Bcrp1-transfected Madin-Darby canine kidney II cell model. Intracellular accumulation of ABC and AZT was significantly reduced by approximately 90% and approximately 70%, respectively, in Bcrp1-transfected cells compared with the wild-type cells. Both ABC and AZT showed significantly increased basolateral-to-apical (B-to-A) and decreased apical-to-basolateral (A-to-B) transport in Bcrp1 cells compared with wild-type directional flux. The efflux ratio (ratio of B-to-A to A-to-B) in Bcrp1-transfected cells was 22 for ABC and 11 for AZT. N-(4-[2-(1,2,3,4-tetrahydro-6,7-dimethoxy-2-isoquinolinyl)ethyl]-phenyl)-9,10-dihydro-5-methoxy-9-oxo-4-acridine carboxamide (GF120918) inhibited this difference in accumulation between the two cell variants with an EC(50) of 1.32 +/- 0.3 microM for ABC and 0.31 +/- 0.1 microM for AZT. Potent and highly cooperative inhibition by Ko143 (3-(6-isobutyl-9-methoxy-1,4-dioxo-1,2,3,4,6,7,12,12a-octahydropyrazino[1',2':1,6]pyrido[3,4-b]indol-3-yl)-propionic acid tert-butyl ester) was observed with an EC(50) of 121 +/- 5 nM for ABC and 19.2 +/- 1.5 nM for AZT (Hill coefficient approximately 3-6). Probenecid, an organic anion inhibitor known to influence AZT biodistribution, had no effect on cellular accumulation in the Bcrp1 model. These studies characterize the Bcrp1-mediated transport of ABC and AZT and show that prototypical BCRP inhibitors GF120918 and Ko143 can inhibit the Bcrp1-mediated transport of these important antiretroviral compounds. The functional expression of BCRP at critical barriers, such as the intestinal enterocytes, brain capillary endothelium, and target lymphocytes, could influence the bioavailability and targeted delivery of these drugs to sanctuary sites.

Simultaneous determination of 8 HIV protease inhibitors in human plasma by isocratic high-performance liquid chromatography with combined use of UV and fluorescence detection: Amprenavir, indinavir, atazanavir, ritonavir, lopinavir, saquinavir, nelfinavir and M8-nelfinavir metabolite

A simple, accurate and fast method was developed for determination of the commonly used HIV protease inhibitors (PIs) amprenavir, indinavir, atazanavir, ritonavir, lopinavir, nelfinavir, M8-nelfinavir metabolite and saquinavir in human plasma. Liquid-liquid extraction was used with hexane/ethylacetate from buffered plasma samples with a borate buffer pH 9.0. Isocratic chromatographic separation of all components was performed on an Allsphere hexyl HPLC column with combined UV and fluorescence detection. Calibration curves were constructed in the range of 0.025-10 mg/l. Accuracy and precision of the standards were all below 15% and the lowest limit of quantitation was 0.025 mg/l. Stability of quality control samples at different temperature conditions was found to be below 20% of nominal values. The advantages of this method are: (1) inclusion and determination of the newly approved atazanavir, (2) simultaneous isocratic HPLC separation of all compounds and (3) increased specificity and sensitivity for amprenavir by using fluorescence detection. This method can be used for therapeutic drug monitoring of all PIs currently commercialised and is now part of current clinical practice.

Evaluation of an International Pharmacopoeia method for the analysis of indinavir sulfate by liquid chromatography

A gradient LC method for the determination of indinavir sulfate (IDV) and its impurities has been recently published in a consultation document of the International Pharmacopoeia, WHO Drug Information. The method uses a base-deactivated reversed-phase C18 column (25 cm x 4.6 mm i.d.), 5 microm kept at a temperature of 40 degrees C. The mobile phases consist of acetonitrile, phosphate buffer pH 7.5 and water. The flow rate is 1.0 ml/min. UV detection is performed at 220 nm. A system suitability test (SST) is described to govern the quality of the separation. The separation towards IDV components was investigated on 16 C18 columns and correlation was made with the column classification system developed in our laboratory. The method was evaluated using a Hypersil BDS C18 column (25 cm x 4.6 mm i.d.), 5 microm. A central composite design was applied to examine the robustness of the method. The method shows good precision, linearity, sensitivity and robustness. Six commercial samples were examined using this method.

Abacavir hypersensitivity reaction in primary HIV infection

OBJECTIVE

To evaluate risk factors associated with the abacavir hypersensitivity reaction during primary HIV infection (PHI).

DESIGN

Acute HIV Infection and Early Disease Research Program protocol (AIEDRP) AI-02-001 provided antiretroviral therapy including abacavir. This retrospective analysis evaluated variables potentially associated with hypersensitivity in the cohort enrolled in AI-02-001 at the University of Washington Primary Infection Clinic.

METHODS

Cases of suspected hypersensitivity were identified prospectively and reviewed retrospectively using a standardized case definition. Controls were the remaining cohort without hypersensitivity. Univariate analyses were performed by linear logistic regression.

RESULTS

Nine (18%) of 50 individuals treated with abacavir developed suspected hypersensitivity. Two of nine cases and no controls were HLA-B5701 positive. When antiretroviral medications were started, cases had lower mean CD8 T-cell percentage and plasma HIV RNA value. After 2 weeks on abacavir, cases had a lower mean HIV RNA value and a trend towards greater decrease in RNA. Cases began abacavir a median of 103 days after HIV acquisition compared to 48 days for controls. There was no significant in vitro abacavir-specific lymphoproliferation or IFN-gamma production in peripheral blood mononuclear cells from individuals following the suspected hypersensitivity reaction.

CONCLUSIONS

Abacavir use during PHI may be associated with increased risk of hypersensitivity. As in chronic infection, HLA-B5701 is associated with the abacavir hypersensitivity reaction in PHI. Although levels of CD8 T cells and HIV RNA may be risk factors for hypersensitivity, the observed association may be due to correlation with HLA-B5701. The interesting temporal association of hypersensitivity with initiation of abacavir later in PHI merits future investigation.

Evaluation of antiretroviral therapy results in Blantyre, Malawi

We performed a cross sectional study to evaluate treatment results of the paying antiretroviral therapy clinic of Queen Elizabeth Central Hospital, Blantyre. The only antiretroviral therapy was a fixed drug combination of stavudine, lamivudine and nevirapine.

METHODS

Interviews, laboratory tests (CD4 count, viral load, nevirapine plasma levels, transaminases) and data extraction from files. 422 (59 %) of the patients who started antiretroviral therapy since 2000 were lost to follow up. The 176 patients enrolled in the study had good virological and excellent clinical treatment results. The most common side effect was peripheral neuropathy. Nevirapine plasma levels were remarkably high and associated with successful virological treatment results. Two simple adherence questions pertaining to the use of medication in the previous 8 days corresponded well with nevirapine levels. The most important reasons for non-adherence were shortage of drugs in the hospital pharmacy and personal financial constraints.

CONCLUSIONS

Many patients were lost to follow up.High nevirapine levels contributed to good therapy results in those studied.Simple adherence questions predicted sub-therapeutic nevirapine levels.Antiretroviral drug supply needs to be uninterrupted and free of charge, to prevent avoidable non-adherence.

Determination of Nevirapine and Efavirenz in Plasma Using GC/MS in Selected Ion Monitoring Mode

An accurate, selective, and sensitive method for the determination of the nonnucleoside reverse transcriptase inhibitors (NNRTIs) nevirapine (nvp) and efavirenz (efv) in human plasma using gas chromatography-mass spectroscopy in selected ion monitoring mode (GC/MS-SIM) was developed. Solid-phase extraction (SPE) gave extraction yields near 100% for both nvp and efv, and calibration curves were linear over the therapeutic concentration ranges. Variation of intraday and interday precision was below 5%. Intraday and interday inaccuracies varied between 0.6% and 10.4%. The lower limits of detection using a 200-microL plasma sample were 27 ng/mL for nvp and 26 ng/mL for efv, and the lower limits of quantification were 54 ng/mL and 72 ng/mL, respectively. The method was applied to the determination of nvp and efv in plasma specimens of 73 patients in HIV stage III or IV and on antiretroviral treatment in Kigali, Rwanda.

Evaluation of antiretroviral therapy results in a resource-poor setting in Blantyre, Malawi

OBJECTIVE

To evaluate treatment results of the paying antiretroviral therapy (ART) clinic of Queen Elizabeth Central Hospital, a large public and teaching hospital in Blantyre, Malawi. The only ART was a fixed drug combination of stavudine, lamivudine and nevirapine.

METHODS

Cross sectional study with interviews, laboratory tests (CD4 count, viral load, nevirapine plasma levels, transaminases) and data extraction from files.

RESULTS

A total of 422 (59%) of the patients who started ART since 2000 were lost to follow-up. The 176 patients enrolled in the study had good virological and excellent clinical treatment results. The most common side effect was peripheral neuropathy. Nevirapine plasma levels were remarkably high and associated with successful virological treatment results. Two simple adherence questions pertaining to the use of medication in the previous 8 days corresponded well with nevirapine levels. The most important reasons for non-adherence were shortage of drugs in the hospital pharmacy and personal financial constraints.

CONCLUSIONS

(1) Many patients were lost to follow-up. (2) High nevirapine levels contributed to good therapy results in those studied. (3) Simple adherence questions predicted subtherapeutic nevirapine levels. (4) Antiretroviral drug supply needs to be uninterrupted and free of charge, to prevent avoidable non-adherence.