Fisher, Martin FRCP*; Moyle, Graeme J MD, MBBS†; Shahmanesh, Mohsen FRCP‡; Orkin, Chloe MRCP§; Kingston, Margaret FRCP‖; Wilkins, Edmund FRCP¶; Ewan, Jacqueline BSc (Hons)#; Liu, Hui PhD#; Ebrahimi, Ramin MS#; Reilly, Geraldine RN#; for the SWEET (Simplification With Easier Emtricitabine Tenofovir) group UK
Continuous antiretroviral therapy dramatically reduces HIV-associated morbidity and mortality1 but may be complicated by adverse effects including the laboratory and metabolic adverse events (including anemia and dyslipidemia) and clinical adverse events (including limb fat loss and subcutaneous lipoatrophy).2 Adverse events or fear of adverse events remain a key cause of drug interruption or discontinuation. The morphological changes of lipoatrophy are stigmatizing and may lead to reduced adherence or treatment discontinuation.3 The metabolic changes may contribute to an increased risk of cardiovascular events.4 Anemia is associated with morbidities such as fatigue and with “all cause” mortality, even in treated patients.5
Patients express a preference for and may adhere better to once-daily dosing.6-8 Pharmacokinetic “forgiveness,” through prolonged elimination half-lives of some once-daily agents, may also make dose timing accuracy less critical, improving treatment satisfaction.8
Clinical trials of thymidine nucleoside analogue (stavudine or zidovudine)-based regimens indicate that these agents are associated with progressive limb fat decline.9-12 Data from studies using thymidine-sparing regimens have reported few instances of clinical lipoatrophy and normal limb fat mass over prolonged follow-up,10,12 suggesting that, in the future, lipoatrophy may occur less frequently. Zidovudine, the most widely used thymidine analogue, is also associated with hematological toxicity relative to other agents including stavudine13 and tenofovir disoproxil fumarate (DF)11 and greater rises in cholesterol than tenofovir DF in initial therapy.11
The main approach to management of thymidine toxicities is switching therapy from a thymidine analogue to tenofovir DF or abacavir.14-16 Both these agents have shown similar beneficial effects on limb fat restoration in persons with lipoatrophy when used as replacements for a thymidine analogue.16 Data have also indicated that an early switch to abacavir from a thymidine analogue may preserve limb fat and prevent further limb fat loss.17 Use of abacavir is constrained by HLAB*5701-related hypersensitivity reactions, intolerance, or drug resistance. Furthermore, both switch and prospective comparative studies with tenofovir DF-based regimens have suggested more favorable effects on total cholesterol and triglycerides than abacavir.16,18,19 More recent cohort data have reported an association between current or recent abacavir use and coronary heart disease but not stroke risk. Tenofovir DF was not investigated in this study.20 This increased coronary heart disease risk may relate to elevated inflammatory or coagulation markers.21
We sought to investigate the relative benefits of switching from a virologically successful twice-daily zidovudine/lamivudine (dosed separately or as Combivir, CBV) regimen to a once-daily tenofovir DF/emtricitabine (as Truvada, TVD) regimen compared with continued CBV in persons receiving once-daily efavirenz (EFV) as the third antiretroviral agent.
This phase IV, open-label, multicenter, randomized, 48-week trial compared the continuation of CBV with substitution with TVD in patients successfully treated with EFV-based antiretroviral therapy. The study was performed in 23 UK and 1 Irish HIV treatment center.
The primary endpoint was to compare the change in hemoglobin (Hb) from baseline to week 24. Secondary endpoints included changes in fasting cholesterol (total, high-density lipoprotein and calculated low-density lipoprotein) and triglycerides, glucose, creatinine, and estimated glomerular filtration rate [eGFR, by both Cockcroft-Gault and modified diet in renal disease (MDRD) methods] at 24 and 48 weeks. Additional endpoints included changes in total limb fat mass by dual x-ray absorptiometry (DEXA) scan in a randomized subset of subjects with access to a central scanning site (MeDiNova Research, Middlesex, United Kingdom). Additional assessments included the incidence of clinical and laboratory adverse events, virological rebound (2 consecutive HIV-1 RNA measurements ≥ 50 copies/mL), and change in CD4 cell count.
Eligible HIV-1-infected subjects were aged ≥18 years had been stable on azidothymidine and 3TC or CBV plus EFV therapy for >6 months with no known resistance to any of the study medications. Individuals who had received previous stavudine therapy as part of a fully suppressive regimen were included but those who had received nonsuppressive therapy or azidothymidine monotherapy were excluded. Participants had documented HIV-1 RNA of <50 copies per milliliter on 2 occasions before study entry and <400 copies per milliliter for >3 months before screening. Women of childbearing potential were required to use an effective method of contraception. Exclusion criteria included pregnant or lactating females, use of anabolic steroids within the last 90 days, with the exception of testosterone replacement for documented hypogonadism, parvovirus infection, use of erythropoietin or blood transfusion within the last 6 weeks, Karnofsky score <50, history of significant renal disease or osteopenia/osteoporosis, creatinine clearance <60 mL/min, aspartate aminotransferase/alanine aminotransferase >5 × upper limit of normal, previous adefovir dipivoxil or cidofovir therapy. Subjects with active infections, malignancies (except basal cell carcinoma), and coinfection with hepatitis B virus (as determined by presence of detectable HBsAg, HbeAg, or detectable hepatitis B viral RNA) were also excluded. Eligible subjects were randomized to (a) stop CBV and start TVD once daily or (b) continue CBV twice daily and continue EFV once daily.
The randomization list, generated using a random number generator, was held by the sponsor Gilead Sciences (Foster City, CA). After recruitment, each patient's anonymized details were faxed to Gilead who notified the recruiting clinician of the treatment allocation.
Participants were followed at baseline, weeks 4, 12, 24, 36, and 48 for adverse effects, full blood count, biochemistry, liver and renal function, lactate, and fasting lipids. DEXA scans were performed only at baseline and week 48.
Two main analyses were planned: at week 24 when all subjects completed week 24 time point or early discontinued and at week 48 when all subjects completed study or discontinued. End of study analysis of all secondary endpoints included all data collected in the study up to 48 weeks on treatment and a 30-day post-study safety follow-up period.
The treated analysis set included all randomized subjects who took at least 1 dose of study drug. The treated analysis set was the primary population for analyses of safety data including the primary endpoint. The intent-to-treat analysis set included all treated subjects who had no major deviations from the eligibility (inclusion/exclusion) criteria. A major deviation was defined as a viral load ≥1000 copies per milliliter at baseline. The intent-to-treat analysis set was used for efficacy analyses.
Inclusion in the DEXA substudy required an additional consent at the screening visit and required the means to travel, at baseline (± 2 weeks) and week 48 (± 2 weeks), to the central DEXA scanning site. A separate randomization schedule was used for subjects who consented to participate in the DEXA substudy. There were no requirements for preexisting subjective evidence of morphological changes to participate in this substudy.
The analysis of change in hemoglobin utilized the last post-baseline observation carried forward (LOCF) for missing results at weeks 24 (primary endpoint) and 48. Analysis of efficacy endpoint (viral load <50 copies/mL) assumed missing = failure. Secondary sensitivity analyses of the primary and analyses of efficacy endpoints described above excluded missing results (ie, missing = excluded analysis). The primary endpoint was compared, between the treatment groups, using the 2-sample Student t test. The P value from 1-sample paired t test for testing the change from baseline in hemoglobin within each treatment group was also reported. The mean and the 95% CIs for changes from baseline and the difference between treatment groups in mean change from baseline in hemoglobin were reported.
For CD4 and CD8 absolute counts and other laboratory values except HIV RNA, a Wilcoxon signed rank test was used to test for changes from baseline within each treatment group. A Wilcoxon rank sum test compared the differences in changes from baseline between treatment groups at each visit. HIV-1 RNA viral load by visit was summarized categorically (<50, 50 to <400, 400 to <1000, ≥1000 copies/mL, and missing). Exact 95% CIs were calculated for the proportion of subjects with HIV RNA < 50 at week 48 and the proportion of subjects who maintained virological response up to week 48 in each treatment group, and a Fisher exact test was used to test for differences between treatment groups. The 95% confidence interval (CI) for the difference of proportions was calculated based on normal approximation.
In the DEXA substudy, absolute values at baseline and week 48 and change from baseline in total grams of limb fat mass, trunk fat, and whole body fat were analyzed. Limb fat mass was calculated as the sum of the limb fat mass for the right arm, left arm, right leg, and left leg. Two-sided 95% CIs for each treatment regimen were presented for the absolute and change from baseline values along with descriptive statistics. Two-sided 95% CIs for the differences in DEXA markers between treatment groups were calculated based on the normal approximation. The changes from baseline at week 48 in DEXA results were compared between treatment groups using the 2-sample Student t test.
The planned sample size was 220 randomized subjects (110 per treatment group). A sample size of 110 subjects per group, assuming 10% early discontinuation, would have 83% power to detect a difference in means of 0.5 g/dL in hemoglobin assuming that the common SD is 1.2 g/dL using a 2-group t test with a 0.05 2-sided significance level.
The study was approved by appropriate UK and Irish ethics committees. Patients provided written informed consent before study entry. The study was conducted according to Good Clinical Practice guidelines. The study was funded and sponsored by Gilead Sciences. The study was monitored by a data and safety monitoring board.
Demographics and Subject Disposition
Two hundred fifty individuals were recruited to the study, 234 individuals (117 in each group) taking at least 1 dose of study medication. Patients were well matched for baseline characteristics including for weight and hemoglobin levels (Table 1).
The majority of subjects completed the study (Fig. 1). Three subjects (3%) in the TVD group and 6 subjects (5%) in the CBV group experienced adverse events that resulted in study discontinuation, for 2 subjects in each group, these events were thought to be study drug related. No new or unexpected safety findings or deaths were reported during the study.
Primary Endpoint: Changes in Hemoglobin
Switching treatment from CBV to TVD resulted in increases in hemoglobin. Differences between groups in the change from baseline in absolute hemoglobin were statistically significant at weeks 24 and 48 when assessed using LOCF or missing = excluded methods (week 24 mean difference for TVD minus CBV: LOCF 0.37 g/dL, 95% CI: 0.15 to 0.58 g/dL, P < 0.001; missing = excluded 0.36 g/dL, 95% CI: 0.13 to 0.59 g/dL, P = 0.003). Within groups, there was a statistically significant increase in absolute hemoglobin in the TVD group [mean ± SD; week 24: 0.46 ± 0.88 g/dL, P < 0.001 (LOCF)] compared with no statistically significant change from baseline in the CBV group [week 24 0.09 ± 0.78 g/dL, P = 0.19 (LOCF)]. At week 48, differences between groups remained significant [mean difference for TVD minus CBV: LOCF 0.6 g/dL, 95% CI: 0.39 to 0.81 g/dL (P < 0.001); missing = excluded 0.65 g/dL, 95% CI: 0.43 to 0.86 g/dL (P < 0.001)]. Within groups, there was a statistically significant increase in absolute hemoglobin in the TVD group [0.5 ± 0.81 g/dL, P < 0.0001 (LOCF)] compared with no statistically significant change from baseline in the CBV group [−0.1 ± 0.81 g/dL, P = 0.20 (LOCF)]. At week 48, more subjects in the TVD group (22 subjects, 22%) compared with the CBV group (2 subjects, 2%) had increases in absolute hemoglobin of >1 g/dL. Similarly, fewer subjects (2 subjects, 2%) in the TVD group compared with the CBV group (8 subjects, 9%) had decreases in absolute hemoglobin of >1 g/dL.
Statistically significant decreases from baseline in mean corpuscular volume and increases in white blood cell count were observed in the TVD group through week 48 compared with no changes from baseline in the CBV group.
HIV Disease Markers
Viral rebound was infrequent in both groups. Only 1 subject (<1%) in the TVD group and 5 subjects (4%) in the CBV group experienced virological rebound (2 consecutive HIV RNA ≥50 copies/mL). There were no statistically significant differences between groups for the numbers of subjects who maintained virological response at any time point through week 48 (90% of subjects in the TVD group and 84% of subjects in the CBV group, P = 0.25).
The difference between groups in change from baseline to week 48 in CD4 cell counts was statistically significant (median changes from baseline: TVD group −3 cells/mm3, CBV group +33 cells/mm3, P = 0.003). However, the distribution of observed CD4 cell counts was similar in the 2 treatment groups during the study, and there were no statistically significant differences between groups at week 48.
Switching to TVD resulted in decreases in fasting total cholesterol and fasting triglycerides compared with continued CBV. Differences between groups in the changes from baseline in fasting total cholesterol and fasting triglycerides were statistically significant at all visits except week 48. Changes from baseline for fasting total cholesterol in the TVD group were largest at week 4 (median −0.59 mmol/L), but reduced by visit to week 48 (median −0.22 mmol/L). Similar results were seen for fasting total cholesterol and fasting triglycerides when subjects who received concomitant lipid-lowering drug during the study were excluded from the analysis.
Within groups, there were statistically significant decreases from baseline in fasting total cholesterol and fasting triglycerides at all visits for the TVD group compared with no statistically significant changes from baseline in the CBV group. Changes in lipids from baseline to weeks 24 and 48 are shown in Table 2. A post hoc analysis performed at week 24 indicated that declines in total cholesterol were greatest in those with elevated cholesterol at baseline (Fig. 2).
Subjects had similar mean eGFR at baseline (by MDRD TVD 111 mL·min−1·1.73 m−2, CBV 109 mL·min−1·1.73 m−2). In the TVD group, there were larger decreases in MDRD eGFR (median change of −0.90 mL·min−1·1.73 m−2 at week 48) compared with CBV (−0.47 mL·min−1·1.73 m−2) (between groups, P = 0.001). Median values remained within the normal range. Similar differences were observed if the Cockcroft-Gault equation was used (between groups, P < 0.001). Two subjects, both in the CBV group, had stage 3 eGFR during the study. In both cases, the eGFR was stage 2 at baseline.
Similar changes were reported with median creatinine, the baseline to week 48 changes were TVD 3 μmol/L, and CBV −1 μmol/L (between groups, P = 0.002). No significant changes were seen in median serum phosphorus over 48 weeks (TVD −0.01 mmol/L, P = 0.50, CBV −0.04 mmol/L, P = 0.67).
One hundred subjects participated in the DEXA substudy, of whom 50 CBV and 44 TVD patients had a baseline scan and 40 patients in each group had 48-week scans. Patients were well matched at baseline for total, limb, and trunk fat and had similar demographic and treatment characteristics to the overall study population. There was a wide range of limb fat at baseline, with a mean (SD) of 6208 g (4595.4) (range, 1200-22,334 g) in the TVD group and a mean (SD) of 6148 g (3880.4) (range, 2498-22,434 g) in the CBV group. Differences between groups in the change from baseline to week 48 in total limb fat mass were statistically significant (mean difference for TVD minus CBV: 448 g, 95% CI: 57 to 839 g, P = 0.025) (Table 3, Fig. 3A). This difference occurred as a result of subjects in the TVD group having modest increases in limb fat (mean change from baseline 261 g, P = 0.054) compared with subjects in the CBV group having modest decreases in limb fat (mean change from baseline −187 g, P = 0.21). There were no statistically significant differences between or within groups in the changes from baseline to week 48 in trunk fat or whole body fat. However, both groups gained body fat during the study (TVD mean = 393 g and median = 299 g; CBV mean = 170 g and median = 230 g). Gains in the TVD group were predominately in the limbs, whereas trunk fat change from baseline was greater in the CBV group (TVD mean = 130 g and median = 128 g; CBV mean = 358 g and median = 470 g).
Increases in limb fat after changing to TVD were greatest in individuals who had received less than the median baseline duration 3 years exposure to ZDV (Fig. 3B) or had lower baseline limb fat (Fig. 3C). No clinically relevant changes in body weight or body mass index were recorded in either treatment group.
Prevention and management of drug-related adverse events remains a key challenge to the success of long-term antiretroviral therapy. This study documents multiple benefits of switching from CBV to TVD in persons established on EFV. Benefits include increases in hemoglobin, declines in proatherogenic lipids, and preservation and recovery of limb fat. These benefits are gained without loss of virological control, maintenance of CD4 numbers, and facilitate the establishment of a compact once-daily regimen.
Multiple previous switch studies have focused on switching once toxicity has established. Studies of switch for lipoatrophy, for example, have typically included only individuals with clinically evident changes.14-16 Although replacement of stavudine or zidovudine with abacavir or tenofovir DF has led to limb fat gain in multiple controlled randomized trials,14-16 the magnitude of recovery is small and the clinical appearance remains. A previous small study of stavudine recipients suggested that early switching to abacavir could preserve limb fat.17 Our study establishes this preemptive approach using TVD as replacement for CBV. Both modest limb fat recovery and preservation of limb fat are achieved through this approach. In Randomised Abacavir vs. Viread Evaluation (RAVE),16 we observed a smaller magnitude of limb fat recovery in persons receiving zidovudine at baseline compared with stavudine. In this study, we extend those observations. Individuals with longer duration of zidovudine, beyond the median baseline duration of 3 years, recovered less limb fat than those who switched early. Longer duration of zidovudine exposure was also associated with less subsequent fat loss than shorter exposure. These data suggest that fat loss is most rapid with zidovudine in the early years after initiation and long-term zidovudine may lead to a relatively irreversible loss in limb fat. These data underline the need for preemptive switching.
Changes in total cholesterol and triglycerides favored switching to TVD. The changes were modest, although greater declines were observed in those with the highest cholesterol at baseline. The clinical significance of these changes on future cardiovascular risk is not known. Prospective studies of tenofovir DF in initial therapy have also generally observed smaller rises in total, low-density lipoprotein, and high-density lipoprotein cholesterol compared with thymidine analogue-based regimens.10,11
The increases in hemoglobin in individuals switching away zidovudine were small and evaluation of symptoms potentially related to low hemoglobin levels is not sought. These differences observed following switch are similar to differences observed in prospective comparisons of CBV and the components of TVD over 48 weeks.11
Although renal dysfunction has been reported in individuals receiving tenofovir DF,22-24 no differences in changes in creatinine or graded creatinine elevations were observed. These findings are consistent with 3-year data from a randomized controlled study in treatment-naive individuals that, like SWEET, included individuals with normal renal function at baseline.10,11
The study has some limitations. First, the primary endpoint based on hemoglobin changes has relatively limited clinical relevance. Studies in the highly active antiretroviral therapy era indicate that hemoglobin is a predictor of HIV-associated morbidity and mortality5,25-27 and low hemoglobin may impact quality of life.25-27 Second, the key DEXA findings include a relatively small number of individuals in a randomized but geographically selected cohort. The patients in this substudy had similar demographic and treatment characteristics to the overall population and the observations are consistent with previous switch studies.14-17
In summary, switching from CBV to TVD leads to statistically significant improvement in hemoglobin and lipids. Limb fat mass is preserved and partially restored through switching to TVD. Switching is generally well tolerated and virologically effective. Where feasible, consideration should be given to proactive switching.
1. Palella FJ Jr, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med
2. Carr A, Samaras K, Burton S, et al. A syndrome of peripheral lipodystrophy, hyperlipidaemia and insulin resistance in patients receiving HIV protease inhibitors. AIDS
3. Duran S, Savès M, Spire B, et al. Failure to maintain long-term adherence to highly active antiretroviral therapy: the role of lipodystrophy. AIDS
4. Friis-Moller N, Reiss P, Sabin CA, et al. Class of antiretroviral drugs and the risk of myocardial infarction. N Engl J Med
5. Moore R, Forney D. Anaemia in HIV-infected patients receiving highly active antiretroviral therapy. J Acquir Immune Defic Syndr
6. Moyle G. The assessing patients' preferred treatments (APPT-1) study. Int J STD AIDS
. 2003;14(Suppl 1):34-36.
7. Portsmouth S, Osorio J, McCormick K, et al. Better maintained adherence on switching from twice-daily to once-daily therapy for HIV: a 24-week randomized trial of treatment simplification using stavudine prolonged-release capsules. HIV Med
8. Maitland D, Jackson A, Osorio J, et al. 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
9. Dubé MP, Komarow L, Mulligan K, et al. Long-term body fat outcomes in antiretroviral-naive participants randomized to nelfinavir or efavirenz or both plus dual nucleosides. Dual X-ray absorptiometry results from A5005s, a substudy of Adult Clinical Trials Group 384. J Acquir Immune Defic Syndr
10. Gallant JE, Staszewski S, Pozniak AL, et al; 903 Study Group. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: a 3-year randomized trial. JAMA
11. Gallant JE, DeJesus E, Arribas JR, et al. Tenofovir DF, emtricitabine, and efavirenz vs. zidovudine, lamivudine, and efavirenz for HIV. N Engl J Med
12. Shlay JC, Visnegarwala F, Bartsch G, et al. Body composition and metabolic changes in antiretroviral-naive patients randomized to didanosine and stavudine vs. abacavir and lamivudine. J Acquir Immune Defic Syndr
13. Moyle G, Sawyer W, Law M, et al. Changes in haematological parameters and treatment efficacy during thymidine analog-based HAART: a meta-analysis of 6 prospective randomized, comparative studies. Clin Ther
14. Martin A, Smith D, Carr A, et al; for the Mitochondrial Toxicity (MITOX) Study Group. Reversibility of lipoatrophy in HIV-infected patients 2 years after switching from a thymidine analogue to abacavir: the MITOX Extension Study. AIDS
15. Moyle GJ, Baldwin C, Landroudi B, et al. A 48-week, randomized, open-label comparison of three abacavir-based substitution approaches in the management of dyslipidemia and peripheral lipoatrophy. J Acquir Immune Defic Syndr
16. Moyle GJ, Sabin CA, Cartledge J, et al. A randomized comparative trial of tenofovir DF or abacavir as replacement for a thymidine analogue in persons with lipoatrophy. AIDS
17. John M, McKinnon EJ, James IR, et al. Randomized, controlled, 48-week study of switching stavudine and/or protease inhibitors to combivir/abacavir to prevent or reverse lipoatrophy in HIV-infected patients. J Acquir Immune Defic Syndr
18. Smith K, Fine D, Patel P, et al. Efficacy and Safety of Abacavir/Lamivudine Compared to Tenofovir/Emtricitabine in Combination with Once-daily Lopinavir/Ritonavir through 48 Weeks in the HEAT Study. CROI 2008, Boston, MA. 3-6 February, 2008
. Abstract 774.
19. Martinez E, Arranz JA, Podzamczer D, et al. A simplification trial switching from nucleoside reverse transcriptase inhibitors to once-daily fixed-dose abacavir/lamivudine or tenofovir/emtricitabine in HIV-1-infected patients with virological suppression. J Acquir Immune Defic Syndr
20. Sabin CA, Worm SW, Weber R, et al. Use of nucleoside reverse transcriptase inhibitors and risk of myocardial infarction in HIV-infected patients enrolled in the D:A:D study: a multi-cohort collaboration. Lancet
21. Lundgren J, Neuhaus J, Babiker A, et al. Use of Nucleoside Reverse Transcriptase Inhibitors and Risk of Myocardial Infarction in HIV-Infected Patients Enrolled in the SMART Study. 17th International AIDS Conference, Mexico City, 3-8 August, 2008
. Abstract THAB0305.
22. Malik A, Abraham P, Malik N. Acute renal failure and Fanconi syndrome in an AIDS patient on tenofovir treatment-case report and review of literature. J Infect
23. Quimby D, Brito MO. Fanconi syndrome associated with use of tenofovir in HIV-infected patients: a case report and review of the literature. AIDS Read
24. Jones R, Stebbing J, Nelson M, et al. Renal dysfunction with tenofovir disoproxil fumarate-containing highly active antiretroviral therapy regimens is not observed more frequently: a cohort and case-control study. J Acquir Immune Defic Syndr
25. Mocroft A, Kirk O, Barton S, et al. Anaemia is an independent predictive marker for clinical prognosis in HIV-infected patients from across Europe. EuroSIDA study group. AIDS
26. Levine A, Berhane K, Masri-Lavine L, et al. Prevalence and correlates of anaemia in a large cohort of HIV-infected women. Women's interagency HIV study. J Acquired Immune Defic Syndr
27. Semba R, Shah N, Klein R, et al. Prevalence and cumulative incidence of and risk factors for anaemia in a multicenter cohort study of HIV-infected and -uninfected women. Clin Infect Dis
SWEET Study Group, United Kingdom
J. Ainsworth (North Middlesex, London), J. Anderson (Homerton University, London), G. Brook (Central Middlesex, London), A. de Burgh Thomas (Gloucester Royal), S. Das (Coventry & Warwickshire), J. Dhar (Leicester Royal Infirmary), M. Fisher (Brighton & Sussex University), R. Fox (Gartnavel General), V. Harindra (St. Mary's, Portsmouth), P. Hay (St. Georges, London), M. Johnson (Royal Free, London), M. Kapembwa (Northwick Park, London), C. Kilkelly (Wellcome Trust Clinical Research Facility, Manchester), M. Kingston (Manchester Royal Infirmary), N. Larbalestier (St. Thomas's, London), C. Leen (Western General, Edinburgh), R. Maw (Royal Victoria, Belfast), G. Moyle (Chelsea & Westminster, London), F. Mulcahy (St. James's, Dublin), E. Ong (Newcastle General), C. Orkin (Barts & Royal London), M. Shahmanesh (Selly Oak, Birmingham), D. White (Birmingham Heartlands), E. Wilkins (North Manchester General), I. Williams (University College London), and Gilead Sciences Limited, Cambridge. Cited Here...
© 2009 Lippincott Williams & Wilkins, Inc.