Baylor-Uganda, Paediatric Infectious Diseases Clinic, Mulago Hospital, Kampala, Uganda 4

Transcription

1 Antiviral Therapy 2010; 15: (doi: /IMP1695) Original article Pharmacokinetics and acceptability of once- versus twice-daily lamivudine and abacavir in HIV type-1- infected Ugandan children in the ARROW Trial Victor Musiime 1, Lindsay Kendall 2, Sabrina Bakeera-Kitaka 3,4, Wendy B Snowden 5, Florence Odongo 1, Margaret Thomason 2, Philippa Musoke 3,4, Kimberly Adkison 5, David Burger 6, Peter Mugyenyi 1, Adeodata Kekitiinwa 3, Diana M Gibb 2, A Sarah Walker 2 *, the ARROW Trial team 1 Joint Clinical Research Centre, Kampala, Uganda 2 Medical Research Council, Clinical Trials Unit, London, UK 3 Baylor-Uganda, Paediatric Infectious Diseases Clinic, Mulago Hospital, Kampala, Uganda 4 Department of Paediatrics and Child Health, College of Health Sciences, Makerere University, Kampala, Uganda 5 GlaxoSmithKline, London, UK 6 Radboud University Nijmegen Medical Centre, Nijmegen, the Netherlands *Corresponding author asw@ctu.mrc.ac.uk A list of the ARROW Trial team members can be found via Additional file 1 Background: No data on once-daily dosing of nucleoside analogues in African children currently exist. We compared the pharmacokinetics (PK) of once- versus twicedaily lamivudine and abacavir treatment using the World Health Organization recommended weight band dosing of scored adult tablets. Methods: HIV type-1 (HIV-1)-infected Ugandan children aged 3 12 years receiving antiretroviral therapy that included lamivudine and abacavir twice daily (total mg, mg and 225/ mg daily for 12 <20, 20 <25 and 25 kg, respectively) were enrolled in a crossover study. Plasma PK sampling (at 0, 1, 2, 4, 6, 8 and 12 h after observed morning intake) was performed for the twice-daily regimen at steadystate. Children were then switched to once-daily treatment with PK sampling repeated 4 weeks later (with an additional 24 h sample). Acceptability questionnaires were completed at both time points. Daily area under the curve ( ) and maximum concentrations ( ) were compared by geometric mean ratios (GMRs). Results: A total of 41 HIV-1-infected children (median age of 7 years) and n=23, n=14 and n=4 in 12 <20, 20 <25 and 25 kg weight bands, respectively, were enrolled. Mean was 13.0 and 12.0 mg h/l for once- and twice-daily lamivudine (GMR 1.09, 90% confidence intervals [CI] ) and 15.3 and 15.6 mg h/l for once- and twice-daily abacavir (GMR 0.98, 90% CI ), respectively, with no difference in 3 6 versus 7 12 year olds. was 76% (lamivudine) and 64% (abacavir) higher on once-daily regimens. For both children and caregivers, once-daily dosing of lamivudine plus abacavir was highly acceptable and strongly preferred over twice-daily. Conclusions: In children aged 3 12 years, of lamivudine and abacavir were bioequivalent on onceand twice-daily regimens. Once-daily dosing of abacavir and lamivudine could provide an alternative dosing strategy for HIV-1-infected children, with high acceptability and strong preference suggesting the potential for improved adherence. Introduction Adherence to medication is crucial to the success of antiretroviral therapy (ART) [1 3]. Although adherence can be challenging for adults, it is even more so for children, who generally rely on caregivers to administer drugs, where issues surrounding disclosure and/or stigma are more complex [4], and where formulation and regimen options are more limited [5]. Adherence to ART has also been shown to decrease over time [6], and strategies for improving and then maintaining high adherence and acceptability in adults and children are urgently needed to fully realize the benefits and durability of ART in resource International Medical Press (print) (online) 1115

2 V Musiime et al. limited settings, especially for children who face lifelong treatment. Lamivudine and abacavir are nucleoside reverse transcriptase inhibitors (NRTIs) that are commonly used in first-line therapy. Their active triphosphorylated moieties have a longer intracellular half-life than their non-phosphorylated forms in plasma [7 9]. In adults, lamivudine and abacavir are licensed for once-daily use because they are non-inferior compared with twice daily administration in both treatment-naive [10] and treatment-experienced patients [11,12]. Once-daily dosing has been shown to improve electronically monitored adherence in adults taking lamivudine and abacavir [13], and is the preferred treatment modality. However, neither lamivudine nor abacavir are yet licensed for once-daily administration in children and there have been no large randomized controlled trials with virological (or other) end points. Two European studies have evaluated the pharmacokinetics (PK) of once- and twice-daily lamivudine and abacavir in children. In 2 13 year olds, PENTA 13 [14] demonstrated that once-daily abacavir and lamivudine were non- inferior in terms of PK profiles compared with twice-daily dosing (19 and 14 children with evaluable PK on lamivudine and abacavir, respectively, most using syrups). In younger children aged 3 36 months, PENTA 15 [15] demonstrated bioequivalence of daily area under the concentration time curve ( ) for twice- and once-daily dosing of abacavir, and very similar on twice- and once-daily lamivudine (17 and 18 children with evaluable PK on lamivudine and abacavir respectively, all using syrups). Although numbers were small, no difference in safety and antiviral activity was observed between children on once-daily and twice-daily dosing. Furthermore, the majority of caregivers believed that administration was much easier when the drugs were given once daily [16]. Low regimen complexity (including once-daily dosing) and low pill burden have consistently been identified as important factors for maintaining high levels of adherence [17]. Most HIV-infected children on ART currently live in resource-limited settings and receive ART as syrups or as whole or divided tablets following the World Health Organization (WHO) guidelines [18]: these children frequently move between different caregivers, and flexibility and simplicity of dosing is crucial in maximizing adherence. The PK and feasibility of once-daily ART has not been evaluated in resource-limited settings, in part because to date the most common first-line regimen to be used in children includes nevirapine and/or stavudine/ zidovudine, all of which are dosed twice daily. The only antiretrovirals currently licensed for once-daily administration in children are didanosine, efavirenz and atazanavir; tenofovir disoproxil fumarate is currently not yet licensed for children anywhere in the world. Lamivudine and abacavir are both particularly attractive choices for first-line therapy in children, and are one of the recommended dual NRTI backbones in the current WHO paediatric ART guidelines [18]. Both drugs have solid scored formulations of the formerly unscored adult tablets, enabling specific paediatric dosing and easy administration (tablets can be crushed into food or liquid if necessary) to children weighing >12 kg using weight-band-based dosing derived from the body surface area as recommended by the WHO. In this study, we compared the plasma PK of onceversus twice-daily lamivudine and abacavir in HIV type-1 (HIV-1)-infected children in Uganda, who were dosed according to the WHO recommended weight bands, to produce the first PK study of lamivudine and abacavir scored tablets in children. This was a substudy of the ARROW Trial, an ongoing randomized trial of monitoring practice and first-line induction maintenance ART strategies in 1,207 ART-naive HIV 1 -infected children in Uganda and Zimbabwe. Methods This was a two-period, non-randomized crossover, open-label study conducted in two centres in Uganda (Joint Clinical Research Centre, Kampala, Uganda and the Paediatric Infectious Disease Clinic, Mulago Hospital, Kampala, Uganda). HIV-1-infected children aged 3 12 years who had been on combination ART including lamivudine and abacavir (both twice daily) and efavirenz once daily for 36 weeks, who were not suffering from illnesses that could influence the PK of these drugs (for example, severe diarrhoea and/or vomiting) and who were not expected to change weight band in the next 4 weeks were eligible for enrolment. Children should not have missed any antiretroviral drugs in the 3 days prior to the PK sampling. All caregivers (and children, where appropriate according to age and knowledge of HIV status) gave written informed consent. The protocol was approved by the ethics committee for each participating centre and in the UK. Medication Doses were given using the WHO dosing chart, which is based on weight bands (Table 1). At the first PK day, some children were also taking zidovudine as part of a randomized four-drug induction maintenance strategy in the ARROW Trial. Zidovudine was stopped immediately after the first PK day (36 weeks after ART initiation), with children remaining on two NRTIs (lamivudine and abacavir) and a non-nucleoside reverse transcriptase inhibitor (efavirenz). Children in the 25 kg weight band who had stopped zidovudine were not included in the primary lamivudine PK analyses International Medical Press

3 Once- vs twice-daily lamivudine/abacavir in Ugandan children Table 1. The WHO dosing guidelines for abacavir, lamivudine and Combivir 150 mg lamivudine mg scored 150 mg scored 300 mg zidovudine abacavir tablets lamivudine tablets scored Combivir tablets Weight, kg AM PM AM PM AM PM 12 < < The World Health Organization (WHO) guidelines [18] recommend half a tablet of Combivir at PM for the 25 kg weight band in order to balance exposure to zidovudine and lamivudine in this weight band. because their lamivudine dose had changed according to the WHO weight bands, which have different total lamivudine doses if lamivudine is taken alone (150 mg twice daily) or as the fixed-dose combination Combivir (150 mg AM and 75 mg PM) in order to dose zidovudine appropriately. PK sampling PK profiles were obtained on two occasions from each child, at 36 weeks post-art initiation on twice-daily regimens, after which children immediately switched to once-daily dosing, and at 40 weeks after receiving a oncedaily regimen for 4 weeks. On the once-daily regimen, all study medication was taken in the morning (enrolled children had switched to morning efavirenz 32 weeks after ART initiation). Children and caregivers completed acceptability questionnaires on both PK days. Questionnaires were designed to solicit identical potential issues (timing/taste of, counting out, swallowing, vomiting and remembering tablets as well as taking the whole dose or doses being conspicuous) with both twice- and once-daily dosing in addition to how easy or hard they thought switching from twice- to once-daily would be, and which schedule they thought would be easiest and which they would prefer. Before switching from twiceto once-daily dosing, caregivers were asked what they anticipated once-daily would be like after switching, they were asked what their actual experience was. Children were admitted to the PK unit either the evening before or on the morning of the PK day. Children fasted for at least 3 h in the morning before the first blood sample, which was usually taken between 6 AM and 7 AM, and then the study nurse immediately observed ingestion of the morning dose of all antiretrovirals (NRTIs plus efavirenz; time 0). Serial 1.5 ml blood samples were then drawn at 1, 2, 4, 6, 8 and 12 h after intake. An additional blood sample was taken at 24 h from children on the once-daily dosing. Children were ambulatory during the day, but restricted to the PK unit. Caregivers could stay with their children throughout. Breakfast (non-standardized, but mostly milk or milky tea with samosas, bread or chapati) was provided at 2 h after the morning dose, following the 2 h blood sample. Children on twice-daily dosing received their evening dose after the 12 h blood sample. Children on the once-daily regimen received the following day s dose after the 24 h blood sample. Plasma concentrations of lamivudine and abacavir were determined using validated mass spectrometry and HPLC by GlaxoSmithKline (Research Triangle Park, NC, USA). The lower limit of quantification for lamivudine, abacavir and zidovudine was 2.5 ng/ml. Any child without a full set of PK measurements on both PK days was replaced by another child in the same age strata. PK parameters of lamivudine and abacavir The maximum concentration ( ) and the trough level ( ) were calculated for each child from their twice- and once-daily plasma PK curves. The AUC for 0 12 h (AUC 0 12 ) and 0 24 h ( ) were calculated using the trapezoidal rule. The apparent oral clearance normalized to body weight (CL/F/kg) was calculated by non-compartment methods using WinNonlin (version 5.2; Pharsight, Mountain View, CA, USA). For all children other than those in the 20 <25 kg weight band, the for twice-daily dosing was calculated by multiplying the AUC 0 12 by 2. Children in the 20 <25 kg weight band received unequal twicedaily doses of lamivudine (150 mg plus 75 mg) and abacavir (300 mg plus 150 mg; Table 1); therefore, the twice-daily AUC 0 12 was multiplied by 1.5 to estimate the (assuming that the PK was doseproportional and that there was no diurnal variation). In a secondary analysis of lamivudine PK parameters including children on Combivir in the 25 kg weight band who had changed lamivudine dose between PK days, lamivudine for twice-daily dosing in the 25 kg weight band was similarly calculated by multiplying the twice-daily AUC 0 12 by 1.5. Statistical analyses Participant characteristics at the first PK day were summarized using medians and interquartile ranges Antiviral Therapy

4 V Musiime et al. Table 2. Characteristics at first PK day at 36 weeks after antiretroviral therapy initiation Characteristic Enrolled (n=41) Included in any PK analysis (n=37) Lamivudine (n=35) Abacavir (n=36) Age, years 7.6 ( ) 7.4 ( ) 7.1 ( ) 7.3 ( ) Male gender 17 (41) 15 (41) 14 (40) 15 (42) Body weight, kg 20.0 ( ) 19.5 ( ) 19.0 ( ) 19.3 ( ) Total daily dose, mg/kg NA NA 9.4 ( ) 19.6 ( ) Weight for age z-score (-2.12 to -0.82) (-2.10 to -0.65) (-2.12 to -0.58) (-2.11 to -0.61) Height for age z-score (-2.80 to -1.12) (-2.80 to -1.11) (-2.84 to -1.00) (-2.82 to -1.11) Antiretroviral therapy Lamivudine/abacavir/ 21 (51) 17 (46) 17 (49) 17 (47) efavirenz Lamivudine/abacavir/ 20 (49) 20 (54) 18 (51) 19 (53) efavirenz/zidovudine a Taking cotrimoxazole prophylaxis b 38 (93) 34 (92) 32 (91) 33 (92) Data are median (interquartile range) or n (%). a All stopped zidovudine treatment (taken as Combivir) after first pharmacokinetic (PK) day and switched to once-daily lamivudine/abacavir/efavirenz. b All other children taking dapsone prophylaxis. NA, not applicable. (IQRs). Weight and height for age were calculated using UK reference ranges [19] (the WHO 2006 reference ranges were not used because they only go to 10 years of age and do not cover the whole age range of enrolled children in this study). For PK parameters, geometric means, geometric mean ratios (GMRs) for once- versus twice-daily dosing and corresponding 90% confidence intervals (CIs) for, and CL/F/kg were calculated. The GMR for the and CL/F/kg were considered bioequivalent if the CI fell between 0.80 and Once- and twice-daily, and CL/F/kg were compared between 3 6 and 7 12 year olds using unpaired Student s t-tests on log transforms. PK analyses included only those children with evaluable PK curves; acceptability and safety analyses included all children enrolled in the PK substudy. All analyses were performed in STATA version 10.1 (StataCorp, College Station, TX, USA). Results A total of 41 HIV-1-infected children had two PK profiles: n=23, n=14 and n=4 in the 12 <20, 20 <25 and 25 kg WHO weight bands, respectively (Table 1). There were seven children who were excluded from all or part of the PK analyses. Two were excluded from all PK analyses because they had increased weight band and therefore increased dose of all drugs between the two PK days; two in the 25 kg weight band were excluded from the primary lamivudine analyses because of a dose increase between PK days as a consequence of moving from Combivir to lamivudine when stopping zidovudine as part of the induction maintenance randomization within the ARROW Trial; and a further two and three were excluded from the lamivudine and abacavir analyses, respectively, because of implausible PK time concentrations (for example, possible labelling errors). The overall and per drug group baseline patient characteristics of the 37 children included in any analyses are shown in Table 2. The youngest child enrolled was 3.6 years old and the oldest was 12.5 years old. Of the 37 children, 20 were taking zidovudine at enrolment to the PK substudy (as part of the ARROW Trial) and stopped this immediately following the first PK day. Most children were taking multivitamins and cotrimoxazole, but a few other concomitant medications were taken (four children were taking isoniazid plus ethambutol plus pyridoxine maintenance anti-tuberculosis therapy, two were taking azithryomicin and one was taking albendazole, amoxycillin/clavullanic acid, cetirizine, griseofulvin and salbutamol). No child was known to have been taking traditional herbal medicines. Lamivudine The PK profiles of lamivudine evaluated in 35 children are presented in Table 3, compared with historical data from 19 children aged 2 13 years in the European PENTA 13 Study [14]. In the ARROW Trial PK study median (IQR; range) total daily doses of lamivudine were 9.4 mg/kg ( ; ) on the twice-daily regimen and 9.6 mg/kg ( ; ) on the once-daily regimen, compared with the recommended total daily dose of 8 mg/kg and median 8.2 mg/kg on both twice- and once-daily dosing in PENTA 13 [14]. for once- and twice-daily lamivudine dosing were bioequivalent (GMR 1.09 [90% CI ]) with the 90% CI of the GMR falling within 0.80 and 1.25, as defined by the US Food and Drug Administration (FDA). The lamivudine was higher (76%) on oncedaily compared with twice-daily dosing (Figure 1A) and CL/F/kg values were similar. Lamivudine occurred at 1 and 2 h after observed intake on twice-daily regimens in 14 and 19 children, respectively, compared with 11 and 20, respectively, on once-daily regimens International Medical Press

5 Once- vs twice-daily lamivudine/abacavir in Ugandan children Table 3. PK parameters of lamivudine for both twice- and once-daily dosing ARROW Trial PK study (n=35) Historical control PENTA 13 (n=19) Twice-daily GM Once-daily GM GMR 4 mg/kg twice-daily 8 mg/kg once-daily GMR Parameter (95% CI) (95% CI) (90% CI) GM (95% CI) GM (95% CI) (90% CI), mg h/l ( ) ( ) 1.09 ( ) 8.88 ( ) 9.80 ( ) 1.12 ( ), mg/l 1.80 ( ) 3.17 ( ) 1.76 ( ) 1.11 ( ) 2.09 ( ) 1.90 ( ), mg/l ( ) a ( ) a NA (< ) a (< ) a NA CL/F/kg, l/h/kg 0.79 ( ) 0.72 ( ) 0.90 ( ) 0.90 ( ) 0.80 ( ) 0.89 ( ) a Median (range)., daily area under the curve; CI, confidence interval; CL/F/kg, apparent oral clearance normalized to body weight;, maximum concentration;, minimum concentration; GM, geometric mean; GMR, geometric mean ratio; NA, not applicable; PK, pharmacokinetic. Figure 1. Lamivudine and abacavir time concentration profiles A Geometric mean lamivudine concentration (95% CI) Time, h B Geometric mean abacavir concentration (95% CI) Time, h Twice daily (12 h sample) Once daily (24 h sample) (A) Lamivudine time concentration profile. (B) Abacavir time concentration profile. CI, confidence interval. Estimated GMRs were very similar to those reported in the previous European PENTA 13 Study [14]. However, interestingly, although lamivudine doses received were 17% higher, children in our ARROW Trial PK study had lamivudine and values that were 33% and 52% higher, respectively, on once-daily dosing compared with those observed in PENTA 13, whereas CL/F/kg values were 10% lower in the ARROW Trial PK study (a similar observation was made for twice-daily dosing; Table 3). PK parameters in the ARROW Trial PK study were also generally more similar to those previously observed in adults (for example, ARROW Trial PK study values on twice-daily dosing were only 8% higher than a recent study in Ugandan adults [20], 12% lower than a study in adolescents and young adults [21]; however, findings from two other adult twice- to once-daily studies have been more variable, with results 30% higher [22] and 30% lower [8] than our study). Of note, 14 of 19 children in PENTA 13 were taking lamivudine syrup, dosed at 8 mg/kg daily, whereas all ARROW Trial children were taking scored adult tablets, with a 15 17% higher average dose (32/35 and 29/35 on twice- and once-daily, respectively, achieving 8 mg/kg). In the ARROW Trial PK study, there were no statistically significant or important differences in PK parameters between 3 6 and 7 12 year olds on either twice- or once-daily regimens (Table 4). Inclusion in the analysis of the two children who had received an increased lamivudine dose between PK days (according to the WHO guidelines because they had changed from Combivir to lamivudine tablets) did not change the results, with lamivudine for twice- and once-daily dosing remaining bioequivalent (GMR 1.12 [90% CI ]). Abacavir Abacavir PK profiles on 36 children are presented in Table 5 and are compared with historical data from 14 Antiviral Therapy

6 V Musiime et al. children in the European PENTA 13 Study [14]. In the ARROW Trial PK study, median (IQR; range) total daily doses of abacavir were 19.6 mg/kg ( ; ) on the twice-daily regimen and 19.1 mg/ kg ( ; ) on the once-daily regimen, compared with the recommended total daily dose of 16 mg/kg, median 16.2 mg/kg on the twice-daily regimen and 16.4 mg/kg on the once-daily regimen in PENTA 13 [14]. and CL/F/kg values for once- and twicedaily abacavir dosing were also bioequivalent (GMR 0.98 [90% CI ]), with the 90% CI of the GMR falling within 0.80 and The abacavir was higher (64%) on once-daily compared with twicedaily dosing (Figure 1B), whereas CL/F/kg values were similar. There was more variability in the plasma levels observed at 1 h after observed intake on once-daily compared with twice-daily regimens, and hence wider Table 4. Lamivudine PK parameters in children aged 3 6 versus 7 12 years Lamivudine twice daily Lamivudine once daily Aged 3 6 years Aged 7 12 years Aged 3 6 years Aged 7 12 years Parameter (n=17) (n=18) P-value (n=17) (n=18) P-value Body weight, kg 16.5 ( ) 21.8 ( ) NA 17.0 ( ) 22.5 ( ) NA Dose, mg/kg 9.37 ( ) 9.58 ( ) ( ) 9.57 ( ) 0.57, mg h/l ( ) ( ) ( ) ( ) 0.43, mg/l 1.69 ( ) 1.91 ( ) ( ) 3.34 ( ) 0.45, mg/l ( ) ( ) NA ( ) ( ) NA CL/F/kg, l/h/kg 0.84 ( ) 0.75 ( ) ( ) 0.69 ( ) 0.47 Data are presented as geometric means (GM; 95% confidence interval) except for body weight, which is presented as median (interquartile range) and minimum concentration ( ) values, which are presented as median (range)., daily area under the curve; CL/F/kg, apparent oral clearance normalized to body weight;, maximum concentration; NA, not applicable; PK, pharmacokinetic. Table 5. PK parameters of abacavir for both twice- and once-daily dosing ARROW Trial PK study (n=36) Historical control PENTA 13 (n=14) Twice-daily GM Once-daily GM GMR 8 mg/kg twice-daily GM 16 mg/kg once-daily GM GMR Parameter (95% CI) (95% CI) (90% CI) (95% CI) (95% CI) (90% CI), mg h/l ( ) ( ) ( ) ( ) ( ) ( ), mg/l ( ) ( ) ( ) ( ) ( ) ( ), mg/l NA <0.015 NA (< ) a (< ) a (< ) a (< ) a CL/F/kg, l/h/kg ( ) ( ) ( ) ( ) ( ) ( ) a Median (range)., daily area under the curve; CI, confidence interval; CL/F/kg, apparent oral clearance normalized to body weight;, maximum concentration;, minimum concentration; GM, geometric mean; GMR, geometric mean ratio; NA, not applicable; PK, pharmacokinetic. Table 6. Abacavir PK parameters in children aged 3 6 versus 7 12 years Abacavir twice daily Abacavir once daily Parameter Aged 3 6 years (n=17) Aged 7 12 years (n=19) P-value Aged 3 6 years (n=17) Aged 7 12 years (n=19) P-value Body weight, kg 16.5 ( ) 23.0 ( ) NA 17.0 ( ) 23.5 ( ) NA Dose, mg/kg 18.8 ( ) 19.6 ( ) ( ) 19.1 ( ) 0.18, mg h/l ( ) ( ) ( ) ( ) 0.79, mg/l 4.12 ( ) 4.23 ( ) ( ) 6.85 ( ) 0.98, mg/l ( ) (< ) NA (< ) (< ) NA CL/F/kg, l/h/kg 1.20 ( ) 1.26 ( ) ( ) 1.24 ( ) 0.92 Data are presented as geometric means (GM; 95% confidence interval) except for body weight, which is presented as median (interquartile range) and minimum concentration ( ) values, which are presented as median (range)., daily area under the curve; CL/F/kg, apparent oral clearance normalized to body weight;, maximum concentration; NA, not applicable; PK, pharmacokinetic International Medical Press

7 Once- vs twice-daily lamivudine/abacavir in Ugandan children 95% CIs. A total of 28 children had their abacavir at 1 h after observed intake on twice- and oncedaily regimens. Like lamivudine, estimated GMRs for abacavir were very similar to the previous PENTA 13 Study [14]. Once-daily abacavir doses received were 16% higher, and children in our ARROW Trial PK study had abacavir and that were 14% and 42% higher, respectively, on once-daily dosing compared with those observed in PENTA 13; clearance was 7% higher in our ARROW Trial PK study (with slightly larger differences for twice-daily dosing; Table 5). PK parameters in both the ARROW Trial PK study and PENTA 13 were higher than those previously observed in adults (for example, the ARROW Trial PK study and PENTA values on oncedaily dosing were 79% and 57% higher, respectively, than those in an adult once- or twice-daily PK study [23]). A total of 13 out of 14 children in PENTA 13 were taking abacavir syrup, dosed at 16 mg/kg daily, whereas all ARROW Trial PK study children were taking scored adult tablets, with a 16 21% higher average dose (34/36 and 31/36 on twice- and oncedaily, respectively, achieving 16 mg/kg). In the ARROW Trial PK study, there were no statistically significant or important differences in abacavir PK parameters between 3 6 and 7 12 year olds at either twice- or once-daily dosing (Table 6). Acceptability All 41 caregivers and children completed the acceptability questionnaires administered just before and at 4 weeks after switching from a twice- to once-daily dosing schedule. Before switching to once-daily dosing, in response to a question about how dosing frequency might affect their lives, 37/41 (90%) caregivers thought switching to once-daily lamivudine and abacavir would be a lot easier for them and 28 (68%) thought it would be a lot easier for their child; these numbers were 39/41 (95%) and 36 (88%), with increases of 5% and 20%, respectively, after switching to once-daily. After switching, 30 (73%) children and 40 (98%) caregivers stated that once-daily was their preferred dosing schedule; 10 (24%) children indicated no preference (once child with missing preference) and one caregiver preferred twice-daily dosing. All the children with no preference (or preference missing) were under 8 years old and 6 were under 5 years of age: comments from 4/6 caregivers indicated the children were too young to decide on a preference. Before switching, 18 (44%) caregivers reported at least one problem with twice-daily dosing compared with only 5 (12%) reporting problems with once-daily dosing after 4 weeks. Problems with the timing of tablet administration were substantially reduced after switching to once-daily regimens (reported by 11 caregivers twice-daily versus 0 once-daily), as were problems with conspicuousness of dose-taking (6 twice-daily versus 3 once-daily), the taste of tablets (5 twice-daily versus 1 once-daily), vomiting doses (5 twice-daily versus 1 once-daily), swallowing tablets (5 twice-daily versus 0 once-daily) and remembering to give the dose (3 twice-daily versus 0 once-daily). In contrast to previous reports from Europe [16], where one-half preferred morning and one-half preferred evening dosing, 34 (83%) Ugandan caregivers preferred once-daily dosing in the morning because this fitted in better with their daily routines, provided greater flexibility in their day and reduced anxiety about being back in time to give the evening dose. By contrast, 7 (17%) caregivers preferred evening dosing so that side effects occurred during sleep rather than at school. Therefore, for children and their caregivers, once-daily dosing of lamivudine and abacavir was both highly acceptable and preferred over twicedaily dosing. Up to 30 November 2009, all 41 enrolled children remained in active follow-up: only 1 child returned to twice-daily dosing after 8 weeks on once-daily dosing at the request of her father who felt more comfortable with twice-daily dosing. However, the family would now prefer to return to once-daily dosing for convenience. Of the 41 enrolled children, 4 changed to evening once-daily dosing at the end of the PK study (3 of the 7 who said that they would prefer evening dosing plus one other child). Safety and efficacy During follow-up after switch to once-daily dosing (49.3 child years up to 30 November 2009; median follow- up time per child 1.15 years [IQR ]), 3 children experienced 1 new Grade 3/4 adverse events. One child had a single episode of asymptomatic neutropaenia (Grade 4), one child had a single episode of asymptomatic hyponatraemia (Grade 4) and one child had a single episode of asymptomatic raised liver enzymes (Grade 4) with Grade 3 bilirubin. None of the adverse events were judged as definitely/probably or uncertain whether related to any antiretroviral drug. The children did not interrupt or change ART doses or regimens and the abnormal laboratory values resolved spontaneously. No child had a new or recurrent WHO Stage 3 or 4 event after switching to once-daily dosing. CD4 + T-cell counts and CD4% continued to increase strongly after switching to once-daily dosing, as expected because switch to once-daily dosing occurred only 36 weeks after ART initiation. Mean (±se) CD4 + T-cell count increases were 261 cells/mm 3 (±65; n=36) and 225 cells/mm 3 (±74; n=38), and CD4% increases were 4.9% (±0.9) and 7.7% (±1.7) after 24 and 48 weeks on once-daily dosing, respectively (P<0.01). Antiviral Therapy

8 V Musiime et al. Discussion In this first PK study of scored adult tablets of lamivudine, abacavir and Combivir for paediatric ART dosing, we have demonstrated that the values of both lamivudine and abacavir are bioequivalent on once-daily and twice-daily dosing in Ugandan children aged 3 12 years. This is consistent with findings among children aged 3 36 months in the European PENTA 15 Study [15] and is similar to the results from children aged 2 13 years in the PENTA 13 Study [14], where values were, if anything, slightly higher on once-daily dosing. Although these studies were conducted in Europe, most children were from various countries in Africa. As expected, the of both lamivudine and abacavir were higher with oncedaily dosing, but no attributable Grade 3 or 4 adverse events were observed after switching and no child discontinued ART because of adverse events. The only other published paediatric study of a once-daily regimen containing abacavir/lamivudine/efavirenz together with didanosine also found once-daily lamivudine and abacavir to be a safe and effective option in children [24]. Viral loads were not performed in real-time in the ARROW Trial; however, samples are stored and investigation of viral suppression over the long-term in these children is planned. Although this study did not measure intracellular levels of lamivudine and abacavir because of the large volumes of blood required, our results in plasma are likely to generalize as studies in adults have shown that once- and twice-daily regimens lead to similar levels of the active intracellular anabolites [23]. Overall drug levels for both lamivudine and abacavir were slightly higher in the ARROW Trial PK study compared with PENTA 13 [14], which is consistent with the 16 21% higher average doses provided by the scored tablets used in the ARROW Trial as a consequence of the WHO recommended doses targeting the minimum rather than the average exposure in a weight band to licensed mg/kg or mg/m 2 doses. This is a key feature of weight-band-based tablet dosing compared with syrups, as failure to increase doses of syrups as children grow is a major source of antiretroviral underdosing errors in children [25]. For abacavir, dose differences appeared to almost completely account for differences in exposure; for lamivudine, dose differences appeared to explain much of the difference, but there was still an approximate 10% difference in exposure even after dose differences were taken into account. This could be the result of a lower bioavailability of the syrup in children. In the PENTA 13 Study, lamivudine and values were lower in 10 children aged 2 6 years compared with 9 children aged >6 13 years [14] for both once- and twice-daily dosing despite similar mg/kg dosing. These findings were supported by a further meta-analysis including data from an additional 32 children [26]. In this ARROW Trial PK study, plasma concentrations of lamivudine for children aged 3 6 years and 7 12 years were equivalent, and and values were more similar to those seen in >6 13 year olds in PENTA 13. One explanation for the lower plasma drug levels in younger but not older children in PENTA 13 might be because many younger children in PENTA 13 and the metaanalysis were taking syrups, whereas all children in this ARROW PK substudy were taking tablets. Previous observations indicate that there is a tendency for syrups to be less bioavailable than solid formulations, particularly in children. For example, the mean ±sd oral bioavailability of a lamivudine solution was 67.9 ±28.1% in 53 HIV-infected children aged 6 months to 17 years receiving doses ranging from 0.5 to 10 mg/ kg [27], compared with 80 85% in adult patients. In a further crossover PK substudy within the ARROW Trial, plasma levels of both lamivudine and abacavir will be measured when children weighing kg switch from syrups to scored tablets in order to investigate this further. Once-daily dosing was highly acceptable to the children and their caregivers, and was strongly preferred over twice-daily dosing. The majority of Ugandan caregivers preferred once-daily dosing in the morning because this allowed them to give the day s dose before they went for their daily activities, permitting flexibility on the timing of returning home without worrying about an evening dose. Of note, all the ARROW Trial PK study children were taking efavirenz, which is taken once-daily, such that the entire regimen taken by these children was once daily. In the PENTA 13 Study, where some children still took other drugs twice daily, the caregivers preferred once-daily dosing only if all the drugs in the regimen were taken once daily [16]. Interestingly, in PENTA 13 caregivers preferred to take once-daily regimens in the evening. Our findings demonstrate that once-daily dosing of lamivudine and abacavir as whole or divided scored adult tablets is feasible in HIV-1-infected children in resource-limited settings. Simplified scheduling might, in turn, lead to improved adherence or to maintenance of high adherence over long periods on first-line ART that will be necessary to maximize the benefits and durability of efficacy of ART, which is particularly important for children who will need lifelong therapy. The PENTA 13 and PENTA 15 studies showed that virological suppression was maintained over time after switching to once-daily dosing [14,15]. In resourcelimited settings where viral load monitoring is generally not available, the risks and benefits of once- versus International Medical Press

9 Once- vs twice-daily lamivudine/abacavir in Ugandan children twice-daily lamivudine and abacavir will need further evaluation. We are therefore currently undertaking a larger longer-term randomized substudy within the ARROW Trial, including retrospective viral load measurements, to evaluate the efficacy of once- versus twice-daily lamivudine and abacavir. This trial will also be able to investigate the effect of all versus only part of the regimen being once daily on acceptability and adherence as children taking both efavirenz and nevirapine will be enrolled, with those taking nevirapine continuing to take this twice daily. Acknowledgements We thank the children and their caregivers for participating in this study. We also particularly thank Mohammed Lamorde, Desire Kabamba and Carol Chijoka for their help in training for paediatric PK studies. Disclosure statement The ARROW Trial is funded by the UK Medical Research Council (G ) and the UK Department for International Development. GlaxoSmithKline donated first-line drugs for the ARROW Trial and provided additional funding for conducting this PK study. WBS and KA are employees of GlaxoSmithKline. DB has received honoraria for serving on advisory boards, speaker s fees and educational grants for clinical research from GlaxoSmithKline, the manufacturer of lamivudine and abacavir. All other authors declare no competing interests. Additional file Additional file 1: A list of the ARROW Trial team members can be found at com/uploads/documents/avt-10-oa-1608_musiime_ Add_file.pdf References 1. Bangsberg DR, Perry S, Charlebois ED, et al. Nonadherence to highly active antiretroviral therapy predicts progression to AIDS. AIDS 2001; 15: Gross R, Yip B, Lo Re V, III, et al. A simple, dynamic measure of antiretroviral therapy adherence predicts failure to maintain HIV-1 suppression. J Infect Dis 2006; 194: Nachega JB, Hislop M, Dowdy DW, et al. Adherence to highly active antiretroviral therapy assessed by pharmacy claims predicts survival in HIV-infected South African adults. J Acquir Immune Defic Syndr 2006; 43: Bikaako-Kajura W, Luyirika E, Purcell DW, et al. Disclosure of HIV status and adherence to daily drug regimens among HIV-infected children in Uganda. AIDS Behav 2006; 10 Suppl 1:S85 S Committee on Pediat ric AIDS, Section on International Child Health. Increasing antiretroviral drug access for children with HIV infection. Pediatrics 2007; 119: Byakika-Tusiime J, Crane J, Oyugi JH, et al. Longitudinal antiretroviral adherence in HIV+ Ugandan parents and their children initiating HAART in the MTCT-plus family treatment model: role of depression in declining adherence over time. AIDS Behav 2009; 13 Suppl 1: Kewn S, Hoggard PG, Sales SD, et al. Development of enzymatic assays for quantification of intracellular lamivudine and carbovir triphosphate levels in peripheral blood mononuclear cells from human immunodeficiency virus-infected patients. Antimicrob Agents Chemother 2002; 46: Yuen GJ, Lou Y, Bumgarner NF, et al. Equivalent steady-state pharmacokinetics of lamivudine in plasma and lamivudine triphosphate within cells following administration of lamivudine at 300 milligrams once daily and 150 milligrams twice daily. Antimicrob Agents Chemother 2004; 48: Harris M, Back D, Kewn S, Jutha S, Marina R, Montaner JS. Intracellular carbovir triphosphate levels in patients taking abacavir once a day. AIDS 2002; 16: Moyle GJ, DeJesus E, Cahn P, et al. Abacavir once or twice daily combined with once-daily lamivudine and efavirenz for the treatment of antiretroviral-naive HIV-infected adults: results of the Ziagen Once Daily in Antiretroviral Combination Study. J Acquir Immune Defic Syndr 2005; 38: Lamarca A, Clumeck N, Plettenberg A, et al. Efficacy and safety of a once-daily fixed-dose combination of abacavir/ lamivudine compared with abacavir twice daily and lamivudine once daily as separate entities in antiretroviralexperienced HIV-1-infected patients (CAL30001 Study). J Acquir Immune Defic Syndr 2006; 41: Sosa N, Hill-Zabala C, Dejesus E, et al. Abacavir and lamivudine fixed-dose combination tablet once daily compared with abacavir and lamivudine twice daily in HIVinfected patients over 48 weeks (ESS30008, SEAL). J Acquir Immune Defic Syndr 2005; 40: Maitland D, Jackson A, Osorio J, Mandalia S, Gazzard BG, Moyle GJ. Switching from twice-daily abacavir and lamivudine to the once-daily fixed-dose combination tablet of abacavir and lamivudine improves patient adherence and satisfaction with therapy. HIV Med 2008; 9: Bergshoeff A, Burger D, Verweij C, et al. Plasma pharmacokinetics of once- versus twice-daily lamivudine and abacavir: simplification of combination treatment in HIV-1-infected children (PENTA-13). Antivir Ther 2005; 10: Paediatric European Network for Treatment of AIDS (PENTA). Pharmacokinetic study of once- versus twicedaily abacavir and lamivudine in HIV-1-infected children aged 3 <36 months. Antivir Ther 2010; 15: LePrevost M, Green H, Flynn J, et al. Adherence and acceptability of once daily lamivudine and abacavir in human immunodeficiency virus type-1-infected children. Pediatr Infect Dis J 2006; 25: Mills EJ, Nachega JB, Bangsberg DR, et al. Adherence to HAART: a systematic review of developed and developing nation patient-reported barriers and facilitators. PLoS Med 2006; 3:e World Health Organization. Antiretroviral therapy of HIV infection in infants and children: towards universal access: recommendations for a public health approach. (Accessed 15 November 2010.) Available from pub/guidelines/paediatric pdf 19. Cole TJ, Freeman JV, Preece MA. British 1990 growth reference centiles for weight, height, body mass index and head circumference fitted by maximum penalized likelihood. Stat Med 1998; 17: Byakika-Kibwika P, Lamorde M, Kalemeera F, et al. Steadystate pharmacokinetic comparison of generic and branded formulations of stavudine, lamivudine and nevirapine in HIV-infected Ugandan adults. J Antimicrob Chemother 2008; 62: Sleasman JW, Robbins BL, Cross SJ, et al. Abacavir pharmacokinetics during chronic therapy in HIV-1-infected adolescents and young adults. Clin Pharmacol Ther 2009; 85: Antiviral Therapy

10 V Musiime et al. 22. Bruno R, Regazzi MB, Ciappina V, et al. Comparison of the plasma pharmacokinetics of lamivudine during twice and once daily administration in patients with HIV. Clin Pharmacokinet 2001; 40: Moyle G, Boffito M, Fletcher C, et al. Steady-state pharmacokinetics of abacavir in plasma and intracellular carbovir triphosphate following administration of abacavir at 600 milligrams once daily and 300 milligrams twice daily in human immunodeficiency virus-infected subjects. Antimicrob Agents Chemother 2009; 53: Scherpbier HJ, Bekker V, Pajkrt D, Jurriaans S, Lange JM, Kuijpers TW. Once-daily highly active antiretroviral therapy for HIV-infected children: safety and efficacy of an efavirenzcontaining regimen. Pediatrics 2007; 119:e705 e Menson EN, Walker AS, Sharland M, et al. Underdosing of antiretrovirals in UK and Irish children with HIV as an example of problems in prescribing medicines to children, : cohort study. BMJ 2006; 332: Burger DM, Verweel G, Rakhmanina N, et al. Agedependent pharmacokinetics of lamivudine in HIV-infected children. Clin Pharmacol Ther 2007; 81: Lewis LL, Venzon D, Church J, et al. Lamivudine in children with human immunodeficiency virus infection: a Phase I/II study. The National Cancer Institute Pediatric Branch-Human Immunodeficiency Virus Working Group. J Infect Dis 1996; 174: Accepted 16 May 2010; published online 25 November International Medical Press