A randomized study to evaluate injection site reactions using three different enfuvirtide delivery mechanisms (the OPTIONS study)

Transcription

1 Short communication Antiviral Therapy 13: A randomized study to evaluate injection site reactions using three different enfuvirtide delivery mechanisms (the OPTIONS study) Mark A Boyd 1 *, Matt Truman 2, Gillian Hales 2, Jonathan Anderson 3, Dominic E Dwyer 4 and Andrew Carr 5 1 National Centre in HIV Epidemiology and Clinical Research, University of New South Wales, Sydney, NSW 2010, Australia 2 Roche Pty. Ltd. Sydney, NSW 2099, Australia 3 The Carlton Clinic, Melbourne, Victoria 3053, Australia 4 Institute of Clinical Pathology and Medical Research, Westmead Hospital, Sydney, NSW 2145, Australia 5 St. Vincent s Hospital, Sydney, NSW 2010, Australia *Corresponding author: mboyd@nchecr.unsw.edu.au Background: The antiretroviral enfuvirtide (ENF) is injected subcutaneously using a 27-gauge needle. Injection site reactions (ISRs) can affect long-term ENF tolerability. Alternative ENF delivery methods may ameliorate ISRs. Methods: We conducted a multicentre, open-label, randomized controlled trial in which patients receiving ENF were randomized to continue receiving ENF by a 27-gauge needle, a shorter 31-gauge needle or a gaspowered, needle-free injection device (NFID). The primary study endpoint was the proportion of participants with <grade 2 ISR induration at week 12. Results: Sixty patients received treatment and were included in the intention-to-treat population. The cohort was predominantly male (95%) with a mean age of 49.1 (SD ±7.7) years who had injected ENF for a mean of 821 (SD ±561) days. Response rates for ISR induration at week 12 were 38%, 25% and 42% for the 27-gauge, 31-gauge and NFID groups, respectively (all pairwise treatment comparison P-values >0.2). There was no significant between-group difference for any ISR endpoint, except for changes in the composite ISR score (that is, no ongoing pain of grade 1 or ISR for ongoing pain grade 1 with induration ISR <grade 3 and for nodules <grade 2), which favoured the 27-gauge needle and NFID groups over the 31-gauge group (P=0.012 and 0.047, respectively). Plasma HIV RNA load was unaffected. There were seven adverse events related to the delivery system: five attributed to the NFID. At week 12, 85% of participants elected to use the NFID. Conclusion: Needle-free ENF injection offers a reasonable, reliable alternative to needle-based injecting in this population, at least in the short term. Introduction Enfuvirtide (ENF) is the only HIV fusion inhibitor licensed for clinical use. It is potent and useful in patients with extensive prior antiretroviral therapy (ART), particularly those with evidence of at least one resistance mutation within all three oral ART classes. In this situation, and integrated into an optimized salvage regimen, ENF significantly increased the chances of a response to therapy [1 6]. Despite this clinical efficacy, ENF uptake has been limited. This is probably accounted for by the requirement for twice-daily, subcutaneous injection, which has proven a barrier to both participant and prescriber acceptance of ENF. As a result of its subcutaneous delivery, ENF is associated with injection site reactions (ISRs) which, while rarely treatment-limiting, are associated with inconvenience and treatment fatigue [7]. A gas-powered, needle-free injection device (NFID; Biojector B2000, Bioject, Medical Technologies Inc., Tualatin, Oregon, USA) can deliver ENF without altering its pharmacokinetics [8]. In two non-randomized trials, delivery using the NFID or a shorter 31-gauge (31G) needle was associated with a decreased severity of ISRs [9,10]. We evaluated these alternate delivery systems in a randomized, controlled trial International Medical Press

2 MA Boyd et al. Methods Study design The OPTIONS study was a randomized, multicentre, open-label, parallel-group study that investigated whether delivery of ENF by means of an NFID or a shorter 31G needle was associated with greater tolerability than delivery by a standard 27-gauge (27G) needle. The eligible population comprised triple antiretroviral drug class experienced, HIV-infected participants 18 years old receiving stable combination ART that included ENF for at least the preceding 12 weeks with no anticipated need for ART change for the 24 weeks following randomization. Exclusion criteria included widespread inflammatory or other skin disease or topical corticosteroid therapy that might interfere with ISR interpretation, use of immunomodulators, interferon or chemotherapy, use of an ENF delivery system other than a 27G needle in the previous 4 weeks, and any condition that might affect a participant s ability to comply with the study requirements. The trial was approved by each site s Human Research Ethics Committee. After written informed consent was obtained and eligibility verified, participants were randomized 1:1:1 to one of three treatment arms (27G needle, 31G needle or NFID) for the first 12 weeks of the study, after which they had the option to continue with the same injection device to which they had been randomized or to choose one of the other devices. After baseline, participants were reviewed at week 2 (for ISRs and serious adverse events) and at weeks 6 and 12 (symptom-directed physical examination, assessment of ISRs, serious adverse events [SAEs] and specific adverse events attributed to ISRs, CD4 + T-cell count and plasma HIV RNA load [Roche Cobas Amplicor HIV-1 Ultrasensitive Monitor Test version 1.5, Roche, Branchburg, USA], quality of life [MOS-HIV Health Survey], ENF adherence check and documentation of concomitant medications [including treatments to manage ISRs]). Endpoints Induration (that is, a hardening of the skin) was graded in one of five categories: grade 0 (none); grade 1 (slight); grade 2 (present but <25 mm); grade 3 ( 25 mm to 49 mm); or grade 4 ( 50 mm); the extent of induration was determined by measuring the diameter in millimetres of the induration at the site of injection. The primary endpoint was the proportion of participants with <grade 2 induration at week 12. Secondary endpoints included the proportion of participants with ISR <grade 2 at week 12 for each of ongoing pain/discomfort, erythema, pruritis, nodules and ecchymosis and the proportion of participants meeting the definition of either the ISR response (that is, a participant with no ISR grade 2 at week 12) or the composite response (that is, a participant with no ongoing pain of grade 1 or ISR for ongoing pain grade 1 with induration ISR <grade 3 and for nodules <grade 2). All endpoints were determined a priori. Sample size We estimated that a sample size of 20 participants per arm (total 60) would confer at least an 80% power to detect a major reduction in the proportion of participants with ISR grade 2 from 0.70 to 0.20 in either of the test arms (31G or NFD) compared with the 27G arm. Statistical analysis All analyses were performed by intention-to-treat (ITT). Response endpoints were analysed using a logistic regression model incorporating terms for treatment group, baseline weight and a binary variable indicating if baseline CD4 + T-cell count was >250 cells/μl. The odds ratios (responder:non-responder) were calculated for the three pairwise treatment comparisons. Continuous endpoints, including the ISR score, were analysed using analysis of covariance (ANCOVA). For each analysis the overall hypothesis of no treatment effect was tested using α=0.05, as were the three pairwise treatment comparisons with no adjustment for multiplicity being made. The primary treatment comparisons were the 27G needle versus the 31G needle and the 27G needle versus NFID. Comparison between the 31G needle and the NFID was performed as a secondary comparison. Results Sixty-one participants were randomized. Sixty participants received treatment and were included in the ITT population. Participant demographics and other baseline characteristics were evenly distributed between the three arms. Participants were predominantly middleaged, white homosexual men (Table 1). The cohort had injected ENF through a 27G needle for a mean duration of 821 (SD ±561) days. Two randomized participants withdrew before week 12, one from the 27G needle arm because of an insufficient therapeutic response and the other from the 31G needle arm without specifying a reason. Response rates (that is, the prevalence of the defined endpoints of interest) at week 12 including between-arm comparisons are expressed in Table 2. For ISR induration the absolute response rates for the three delivery mechanisms at week 12 were 27G 38%, 31G 25% and NFID 42%. The 27G needle and NFID groups demonstrated similar degrees of improvement from baseline to week 12 for ISR International Medical Press

3 Randomized trial of three different enfuvirtide delivery mechanisms Table 1. Baseline characteristics and prevalence of specified injection site reactions Variable 27G (N=21) 31G (N=20) NFID (N=19) Sex, male n (%) 19 (90) 20 (100) 18 (95) Race, white n (%) 20 (95) 17 (85) 19 (100) Mean age, years (SD) 48 (7.7) 49 (7.9) 50 (7.6) Mean BMI (SD) 23 (3.6) 25 (5.5) 24 (3.9) CDC stage C, n (%) 11 (52) 12 (60) 10 (53) Mode of transmission, MSM n (%) 18 (86) 15 (79) 14 (74) CD4 + T-cell count, cells/μl Mean (SD) 377 (258) 456 (237) 338 (312) Median (IQR) 323 ( ) 400 ( ) 189 ( ) Mean plasma HIV RNA, log 10 (SD) 2.2 (0.8) 2.2 (0.9) 2.0 (0.6) Duration of ENF Mean duration of ENF, days (SD) 771 (581) 904 (582) 789 (535) Median duration of ENF, days (IQR) 660 ( ) ( ) 595 ( ) ISR endpoint ISR induration, n/n (%)* 17/21 (81) 15/20 (75) 14/19 (74) ISR ongoing pain/discomfort, n/n (%)* 12/21 (57) 9/20 (45) 8/19 (42) ISR erythema, n/n (%)* 14/21 (67) 15/20 (75) 12/19 (63) ISR pruritis, n/n (%)* 0/21 0/20 2/19 (11) ISR nodules, n/n (%)* 9/21 (43) 5/20 (25) 7/19 (37) ISR ecchymosis, n/n (%)* 7/21 (33) 6/20 (30) 7/19 (37) Overall ISR response, n/n (%) 0/21 (0) 2/20 (10) 0/19 (0) ISR composite response, n/n (%) 5/21 (24) 6/20 (30) 5/19 (26) *Presence of injection site reaction (ISR) in these instances is defined as the presence of specified ISR at grade 2. Overall ISR response was defined as no ISR response grade 2. Composite response was defined as no ongoing pain of grade 1 or ISR for ongoing pain grade 1 with induration ISR <grade 3 and for nodules <grade 2. BMI, body mass index; ENF, enfuvirtide; IQR, interquartile range; MSM, men who have sex with men; 27G, 27-gauge needle; 31G, 31-gauge needle; NFID, needle-free injection device. induration (19% to 38% and 26% to 42% without induration for the 27G needle and NFID, respectively), whereas those randomized to the 31G needle experienced no change (25% without induration at baseline and week 12). There were no statistically significant between-group differences for ISR induration at week 12 (all P-values >0.2). There was no statistically significant between-group difference for any other ISR endpoint. An analysis of the change from baseline in the ISR score at week 12 favoured the 27G and NFID groups over the 31G group (P=0.012 and 0.047, respectively). Plasma HIV load and CD4 + T-cell count outcomes were comparable across the groups. There were seven delivery system-related adverse events reported in six participants, five of which were attributed to the NFID (three haematomas, all related to the NFID; three severe ISRs, one NFID-related, one related to the 27G needle and one related to the 31G needle; and knee paraesthesia in one, related to NFID); none of these was considered to be serious. There were 12 serious adverse events (SAEs) reported; none was related to either ENF or the ENF-delivery system (data not shown). At week 12, 51/60 (85%) participants elected to use the NFID to deliver ENF, including 16 of the 19 patients (84%) randomized to use the NFID. No significant differences for any measure of quality of life were detected over the 12 weeks of study. Discussion In this randomized study we found that the use of a shorter 31G needle or an NFID had no major effect on ISRs compared with delivery through a standard 27G needle. Two studies have suggested that the NFID is associated with reduced occurrence of ISRs; however, both were observational studies with no randomized comparator arms and were therefore prone to bias [8,9]. We found that there was a reduction in the severity of ISR induration in both the 27G needle and NFID groups after baseline, suggesting that a trialrelated factor such as improved supervision of injection technique may have improved outcomes in general. This was not the case, however, for those randomized to use of the 31G needle. In addition, for the composite ISR score the 31G needle performed significantly worse than the standard 27G needle and the NFID. This suggests that the use of the smaller, 31G needle offers no advantage over conventional delivery by a standard 27G needle or by an NFID. Antiviral Therapy 13:3 451

4 MA Boyd et al. It could be argued that by enrolling a cohort of patients with a median experience of ENF injection of >2 years the cohort contained an inherent selection bias in favour of patients able to tolerate ENF delivery by means of standard 27G needles. However, if this had been the case we would not have expected to see an improvement in the outcomes in those randomized to continue injecting by the standard 27G needle, which in fact we did. It is relevant to consider that the majority of these study participants were in a very late salvage situation with regard to their HIV control and therefore were likely to be putting up with substantial adverse effects of ENF injection because of a beneficial virological and/or immunological and/or clinical response. We observed responses across all three study groups, raising the possibility that some of the findings may reflect a regression to the mean phenomenon. These results contrast with a randomized study comparing the use of an NFID versus standard 27G needle in ENF-naive participants commencing ENF as part of combination ART, where significantly more participants randomized to ENF delivery by standard 27G needle experienced nodules/induration compared with those using the NFID [11]. This difference may be the product of the different populations studied (ENF-experienced versus ENF-naive). The studies are also difficult to directly compare because the study of ENF-naive patients used as the primary endpoint a combination of painful nodules/induration, as opposed to the use of induration alone in our study. Despite the apparent lack of difference between needle-based and NFID delivery of ENF, the great majority of participants (85%) elected to use the NFID after week 12. This decision is likely to have been influenced by the fact that two-thirds of the participants had not had an opportunity to use this novel device in the first 12 weeks of the study. The study has several limitations. It was powered to find a major (50%) reduction in ISRs with the use of alternative delivery mechanisms. We did not find such a difference, but cannot exclude the possibility that a smaller difference exists. It was not possible to blind the study; it is therefore possible that some of the ISR observations were biased. However, the primary endpoint was chosen on the basis that induration is an objective outcome, which should have minimized observer bias. In conclusion, the use of a smaller 31G needle for ENF delivery did not appear to offer any advantage Table 2. Analysis of proportions of participants with specified injection site reaction responses and between-group comparisons Variable Group Response at week 12, % Week 12 comparisons Odds ratio* 95% CI P-value Induration 27G 38 31G versus 27G G 25 NFID versus 27G NFID 42 NFID versus 31G Ongoing pain/ 27G 76 31G versus 27G discomfort 31G 55 NFID versus 27G NFID 79 NFID versus 31G Erythema 27G 38 31G versus 27G G 25 NFID versus 27G NFID 32 NFID versus 31G Pruritis 27G 100 ND ND ND ND 31G 100 ND ND ND ND NFID 100 ND ND ND ND Nodules 27G 67 31G versus 27G G 70 NFID versus 27G NFID 74 NFID versus 31G Ecchymosis 27G 86 31G versus 27G G 80 NFID versus 27G NFID 79 NFID versus 31G Overall ISR response 27G 14 31G versus 27G G 5 NFID versus 27G NFID 16 NFID versus 31G Composite response 27G 43 31G versus 27G G 30 NFID versus 27G NFID 53 NFID versus 31G *Odds ratio >1 indicates a benefit for the first treatment in the comparison. Overall injection site reaction (ISR) response was defined as no ISR response grade 2. Composite response was defined as no ongoing pain of grade 1 or ISR for ongoing pain grade 1 with induration ISR <grade 3 and for nodules <grade 2. CI, confidence interval; ENF, enfuvirtide; 27G, 27-gauge needle; 31G, 31-gauge needle; NFID, needle-free injection device; ND, not done International Medical Press

5 Randomized trial of three different enfuvirtide delivery mechanisms over delivery by conventional 27G needle. In contrast, needle-free injection of ENF offers a reasonable, reliable alternative to needle-based injecting, at least in the short term. Acknowledgements We thank all participants and all OPTIONS study staff (The Alfred Hospital: Jennifer Hoy and Janine Rooney; The Carlton Clinic: Richard Moore and Kaye Lowe; Flinders Medical Centre: Robyn Gilligan; Ground Zero Medical limited: Cassy Workman and Vanessa Rees; Prahran Market Clinic: Norman Roth and Helen Wood; Royal Brisbane and Women s Hospital: Anthony Allworth, Janelle Zillman and Holly Asher; Westmead Hospital: Margaret Piper). Funding for this study was provided by Roche Products Pty. Ltd., Australia. Disclosure statement Mark Boyd has acted as a paid consultant to Roche, from whom he has also received funding for international conference attendance. Andrew Carr has acted as a paid consultant to Roche. Matt Truman and Gillian Hales are employees of Roche. Jonathan Anderson has received funding from Roche for air travel to a domestic conference. References 1. Lalezari JP, Henry K, O Hearn M, et al. Enfuvirtide, an HIV-1 fusion inhibitor, for drug-resistant HIV infection in North and South America. N Engl J Med 2003; 348: Lazzarin A, Clotet B, Cooper DA, et al. Efficacy of enfuvirtide in patients infected with drug-resistant HIV-1 in Europe and Australia. N Engl J Med 2003; 348: Hicks CB, Cahn P, Cooper DA, et al. Durable efficacy of tipranavir-ritonavir in combination with an optimised background regimen of antiretroviral drugs for treatment-experienced HIV-1-infected patients at 48 weeks in the Randomized Evaluation of Strategic Intervention in multi-drug resistant patients with Tipranavir (RESIST) studies: an analysis of combined data from two randomised open-label trials. Lancet 2006; 368: Clotet B, Bellos N, Molina JM, et al. Efficacy and safety of darunavir-ritonavir at week 48 in treatmentexperienced patients with HIV-1 infection in POWER 1 and 2: a pooled subgroup analysis of data from two randomised trials. Lancet 2007; 369: Grinsztejn B, Nguyen BY, Katlama C, et al. Safety and efficacy of the HIV-1 integrase inhibitor raltegravir (MK- 0518) in treatment-experienced patients with multidrug-resistant virus: a phase II randomised controlled trial. Lancet 2007; 369: Dwyer DE, Workman C, Hales G, Amin J, Cooper D, Miller J. Enfuvirtide in HIV-1-infected individuals changing therapy to a nucleoside reverse transcriptase inhibitor sparing regimen: the ALLIANCE Study. Antivir Ther 2006; 11: Trottier B, Walmsley S, Reynes J, et al. Safety of enfuvirtide in combination with an optimized background of antiretrovirals in treatment-experienced HIV-1-infected adults over 48 weeks. J Acquir Immune Defic Syndr 2005; 40: Harris M, Joy R, Larsen G, Valyi M, Walker E, Frick LW, et al. Enfuvirtide plasma levels and injection site reactions using a needle-free gas-powered injection system (Biojector). AIDS 2006; 20: Loutfy M, Harris M, Raboud J, et al. A large prospective study assessing injection site reactions, quality of life and preference in patients using the Biojector vs standard needles for enfuvirtide administration. HIV Med 2007; 8: Shalit P, True A, Thommes JA. Quality of life and tolerability after administration of enfuvirtide with a thin-walled needle: QUALITE Study. HIV Clin Trials 2007; 8: Gottlieb M, True A, Evans R, et al. Needle-free administration of enfuvirtide significantly reduces incidence of painful injection site reactions: results from a single blind, randomized, controlled study. The 46th International Conference on Antimicrobials and Chemotherapy. September 27 30, 2006, San Francisco, CA. Poster H-1905b. Accepted for publication 7 December 2007 Antiviral Therapy 13:3 453

6