Methods

Study Design

As shown in Figure 1, for the initial 12 weeks of treatment, eligible patients were randomized in a 3:1 ratio to receive induction therapy, 4 mg of r-hGH dosed daily (group A, DD) or PL (group B, PL). At week 12, subjects in group A were rerandomized in a 1:1 ratio to receive maintenance therapy, 2 mg of r-hGH on AD or PL on AD, from weeks 12 through 36. Subjects from group A, on r-hGH induction and r-hGH maintenance, are referred to as the DD-AD-AD group; those from group A on induction therapy with r-hGH followed by PL maintenance are the DD-PL-PL group. Subjects in group B, who received PL from baseline to week 24 and then received 4 mg/d of r-hGH for weeks 24 through 36, are the PL-PL-DD group.

Figure 1.  (click image to zoom)

Study design.      

The prespecified primary efficacy parameter was the absolute change from baseline to week 12 in area of VAT on a cross-sectional computed tomography (CT) scan at L4-L5. Secondary efficacy endpoints included changes in other body composition parameters, serum lipids, body image, and quality-of-life variables (results for body image and quality of life are reported separately).[13]

Safety data included glucose and insulin parameters, viral load, CD4 cell counts, adverse events (AEs), insulin-like growth factor (IGF)-I and its main binding protein (IGF-BP3), hemoglobin A1c (HbA1c; an indicator of adequacy of glucose control over a 3-month period),[14] and standard hematologic and clinical chemistry parameters.

The primary prespecified criterion of efficacy for maintenance therapy at week 36 was maintenance failure rate, defined as the proportion of subjects originally given 4 mg/d of r-hGH 4 for weeks 1 through 12 who succeeded in losing VAT during baseline to week 12 but then regained more than 50% of their VAT loss by week 36 (see statistical analysis). Failure rates were also compared between those assigned to PL or 2 mg of r-hGH on AD for weeks 12 through 36. Mean changes from baseline to week 36 in VAT, other body composition parameters, lipid profile, and safety parameters were also examined.

Randomization was stratified by gender and implemented by a central system operated by an independent vendor (Clinphone, Nottingham, United Kingdom), assigning patients to treatment using a blinded computer-generated randomization list. The trial was conducted according to the Declaration of Helsinki principles and Good Clinical Practice. Independent institutional review boards approved the protocol at each site. Written informed consent was obtained from each patient before screening.Study Subjects

Eligibility criteria were the same as for the previous trial.[5] Patients were between 18 and 60 years old, had documented HIV infection, had been on stable antiretroviral therapy (ARVT) for ≥30 days, and agreed to continue on ARVT while on study. They also had fasting glucose <110 mg/dL, 2-hour postload glucose on oral glucose tolerance testing (OGTT) results <140 mg/dL, and evidence of excess abdominal adipose tissue as determined by waist/hip ratio (WHR) ≥0.95 and waist circumference (WC) >88.2 cm for men and WHR ≥0.90 and WC >75.3 cm for women.[5] The WHR criteria are similar to those identified as indicative of abdominal adiposity in the literature on obesity and are known to be associated with increased cardiovascular risk.[15,16]

Patients were excluded if they had active systemic infection, unstable or untreated hypertension (≥140/90 mm Hg), acute illness treated in an intensive care unit, a history of pancreatitis, carpal tunnel syndrome (unless resolved by surgical release), diabetes mellitus, malignancy (except for limited cutaneous Kaposi sarcoma or excised basal cell or squamous cell skin carcinoma), angina pectoris, coronary artery disease, or any disorder associated with moderate to severe edema. Patients must not have been receiving insulin or insulin-sensitizing agents, systemic glucocorticoids, or weight reduction agents for 3 months before screening or therapy for HIV-associated wasting (eg, anabolic steroids other than testosterone replacement, appetite stimulants, r-hGH) for 12 months before screening. Lipid-lowering agents were permitted if they were started at least 8 weeks before study entry.Treatment and Assessment

The 4-mg induction dose of r-hGH (Serostim; EMD Serono, Rockland, MA) and its PL were given as 1.0-mL single subcutaneous injections each evening. The 2-mg AD maintenance dose of r-hGH and its PL were given as single 0.5-mL subcutaneous injections every other evening. Active and PL study drug were labeled and packaged identically, and doses were sequentially numbered. Patients were taught to self-inject according to the prescribed sequence. The protocol for dose adjustments for weight and toxicity was the same as used previously.[5]

Study visits were scheduled at screening; baseline; and weeks 2, 4, 12, 16, 24, 26, 28, and 36. CT and dual x-ray absorptiometry (DXA) scans to assess fat distribution were obtained at baseline, week 12, and week 36. OGTT and lipid profiles were obtained after a minimum 12-hour fast. These and serum IGF-I, IGF-BP3, HIV-1 RNA, and testosterone levels and CD4 T-cell count were obtained at baseline, week 12, week 24, and week 36. HbA1c levels were assessed at baseline and weeks 4, 12, and 36 by affinity chromatography. At each visit, standard hematologic and biochemistry panels, physical examinations, and reporting of AEs using the Medical Dictionary for Regulatory Activities (MedDRA), version 8.0 (MSSO, Reston, VA) were conducted. Laboratory testing was performed centrally (Esoterix Laboratories, Calabasas Hills, CA).Statistical Analysis

Data were analyzed in the modified intention-to-treat (ITT) population, which included subjects who received at least 1 dose of study drug and who had follow-up data. There were separate analysis plans for the initial 12-week induction treatment period and the 24-week maintenance period. The week 12 analysis was the primary analysis. The primary efficacy parameter, change from baseline to week 12 in absolute area of VAT, was analyzed using a nonparametric ANCOVA model with effects for treatment and gender, with baseline VAT as a covariate. The major efficacy endpoint for the maintenance phase of the study was the percentage of patients regaining >50% of the VAT they had lost during induction (weeks 1-12). Maintenance therapy was considered efficacious if, during maintenance (weeks 12-36), no more than half of the subjects who had lost VAT regained >50% of the amount they lost. Mean changes in VAT, other body composition parameters, and lipid parameters were examined in the ITT sample from baseline to week 36 and from weeks 12 to 36 as well. The trial was not powered statistically to test differences in changes in study endpoints between maintenance groups, however.

Between-group differences in continuous secondary efficacy parameters were analyzed using raw data with an ANOVA model, including effects for treatment, gender, and treatment-by-gender interaction, when parametric model assumptions were met or, using ranked data, when parametric assumptions were not met. Within-group differences were analyzed using the Wilcoxon signed rank test. Safety results were summarized for the population of patients who received at least 1 dose of study drug and had follow-up data (n = 325 for induction, n = 258 for maintenance). Between-group differences in categoric variables were analyzed using the Fisher exact test.  Printer- Friendly Email This

J Acquir Immune Defic Syndr.  2007;45(3):286-297.  ©2007 Lippincott Williams & Wilkins
This is a part of article Recombinant Human Growth Hormone to Treat HARS Taken from "Actos Pioglitazone" Information Blog

Cardiovascular

correspondence questions PROactive conclusions

Susan Jeffrey
January 11, 2006

London, UK - Correspondence in the January 7, 2006 issue of the Lancet raises questions about the conclusions of the recently published Prospective Pioglitazone Clinical Trial in Macrovascular Events (PROactive) trial.

PROactive compared treatment with pioglitazone (Actos, Takeda Pharmaceutical Company/Eli Lilly) vs placebo in patients with type 2 diabetes at high risk for macrovascular events. The main results, published in the October 8, 2005 issue of the Lancet, showed a nonsignificant 10% reduction in the study's primary end point of all macrovascular events vs placebo but found a significant 16% reduction in the secondary composite end point of death, MI, and stroke with pioglitazone treatment [1].

The authors' conclusion that pioglitazone "reduces the composite of all-cause mortality, nonfatal myocardial infarction, and stroke in patients with type 2 diabetes who have a high risk of macrovascular events" draws fire from Lancet readers in this week's issue.
Concerns about interpretation

Several letters took issue with the fact that this conclusion was based largely on results from the secondary composite end point. For example, Dr Pierre-Jean Guillausseau (University Paris, France) points out, "Although the study is globally negative, a new 'main secondary end point' appeared?ie, a new composite not described in the study protocol published in 2004. The conclusions of the study are based solely on this composite. [2]"

The results should be adjusted for mean differences in hemoglobin A1c (HbA1c) and blood pressure, and there remains a major concern about the increased incidence of heart failure seen with pioglitazone, Guillausseau adds. "Thus several questions remain, and the role of glitazones in type 2 diabetes mellitus is not yet fully defined."

In a separate letter, Drs Peter Gaede, Hans-Henrik Parving, and Oluf Pedersen (Steno Diabetes Center, Copenhagen, Denmark) point out that patients receiving pioglitazone had a greater decrease in HbA1c than those on placebo, although LDL increased with treatment. However, blood pressure was also lower in treated patients, by about 3 mm Hg [3].

"An accurate prediction of the relative risk reduction of a 3-mm-Hg systolic gradient seen in PROactive indicates that this is more than sufficient to explain the whole potential cardiovascular benefit of pioglitazone," they write. "Furthermore, it should be stressed that this prediction is conservative, since diabetic patients are particularly blood-pressure sensitive."
"Hypothesis generating," not "groundbreaking proof"

Drs John S Yudkin (University College London, UK) and Nick Freemantle (University of Birmingham, UK) are blunter [4]. In concentrating on the "main secondary end point," they write, the PROactive authors "ignore the statistically neutral primary outcome. One assumes that, had the 10% reduction in primary-end-point events been significant, Dormandy and colleagues would have felt no need to emphasize the analysis of the main secondary end point," an end point that should be considered "hypothesis-generating" rather than "groundbreaking proof."

An appropriate conclusion from PROactive, Yudkin and Freemantle assert, "is that glitazones reduce cardiovascular event rates with a point estimate of around 10% to 15%, but with a confidence interval including zero?a result in line with the equally uncertain reduction in macrovascular events seen in the United Kingdom Prospective Diabetes Study (UKPDS). Judging from the way in which the results were presented at the European Association for the Study of Diabetes in Athens in September 2005 and from the website (http://www.proactive-results.com), there is a risk that the marketing division of Takeda and Eli Lilly will use these questionable results mercilessly in their promotional material."

Freemantle also voiced his objection to the PROactive conclusions in a letter previously published by the BMJ [5].

Finally, Drs RR Holman, R Retnakaran, A Farmer, and R Stevens (Churchill Hospital, Oxford UK) write that they carried out an analysis using the UKPDS Outcomes Model to assess the expected outcomes given the risk-factor changes reported [6]. They point out that the nonsignificant 10% relative risk reduction seen with pioglitazone treatment on the primary end point was less than the 20% or more on which the power calculation for the study was based.

"Our analysis supports the explanation that any macrovascular benefits seen reflect the modest improvements obtained in established risk factors, with little evidence that changes seen previously in novel risk factors with pioglitazone have any substantive effect," they write. "More worryingly, the estimated macrovascular benefits are offset by an increased risk of heart failure and concerns about increased peripheral revascularization rates."
The authors respond

In a response undersigned by the principal investigator of the PROactive study, Dr John Dormandy (St Georges Hospital, London, UK), the PROactive investigators defend their interpretation of the main secondary end point [7]. "As described in the PROactive report, this end point was prespecified in the statistical analysis plan and submitted to the Food and Drug Administration before unblinding," they write. "It showed a 16% relative risk reduction with pioglitazone (p=0.027).

"We agree that in isolation this finding would not allow any definite conclusion to be drawn," Dormandy et al add. "However, the primary composite end point, which also included silent myocardial infarction, acute coronary syndrome, major leg amputation, and coronary and leg revascularization, also showed a reduction with pioglitazone?10% relative risk reduction (p=0.095)."

To the comment made by Yudkin and Freemantle about how they would have felt no need to present the main secondary-end-point analysis if the primary end point had been significantly positive, they respond, "On the contrary, it is the effect on mortality, myocardial infarction, and strokes that would be of most interest to patients and the regulatory authorities."

As for the increase in heart failure, they point out that the heart-failure events in this study were unadjudicated and the diagnoses made by the investigators, "who might have been sometimes misled by the known increase in edema with pioglitazone," Dormandy et al write.

"We are surprised by the comparison in the Comment [the editorial accompanying the PROactive publication by Dr Hannele Yki-J¿rvinen (University of Helsinki, Finland) (8)] of first events for end points avoided, on the positive side, vs all episodes of heart failure, on the negative side," they add. "Nevertheless, the executive committee is further exploring the issue of heart failure by setting up an independent commission of experts to review, in a blinded way, all source material for patients reported to have had fatal or nonfatal heart failure at any time during the trial."
Disclosure information is provided for correspondents and PROactive authors in the Lancet publication.

Sources

Dormandy JA, Charbonnel B, Eckland DJA, et al on behalf of the PROactive investigators. Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROactive Study (Prospective Pioglitazone Clinical Trial in Macrovascular Events): a randomized controlled trial. Lancet 2005; 366:1279-1289. Guillausseau PJ. PROactive study. Lancet¿2006;¿367:23. Gaede P, Parving HH, and Pedersen O. PROactive study. Lancet¿2006;¿367:23-24. Yudkin JS, Freemantle N. PROactive study. Lancet¿2006;¿367:24-25. Freemantle N. How well does the evidence on pioglitazone back up researchers' claims for a reduction in macrovascular events? BMJ 2005; 331:836-838. Holman RR, Retnakaran R, Farmer A, and Stevens R. PROactive study. Lancet¿2006;¿367:25-26. Dormandy J. PROactive study?Authors' reply. Lancet¿2006;¿367:26-27. Yki-J¿rvinen H. The PROactive study: some answers, many questions. Lancet 2005; 366:1241-1242.
This is a part of article correspondence questions PROactive conclusions Taken from "Actos Pioglitazone" Information Blog