The AGENT clinical trials programme
Division of Cardiology/3rd Floor Heart Center, William Beaumont Hospital, Royal Oak, USA
* C.L. Grines, Division of Cardiology/3rd Floor Heart Center, William Beaumont Hospital, 3601 West 13 Mile Road, Royal Oak, MI 48073-6769, USA. Tel.: +1-248-551-7280; fax: +1-248-551-8806
cgrines{at}beaumonthospitals.com
Abstract
AGENT and AGENT 2 are the first randomized, double-blind, placebo-controlled trials studying the benefits of stimulating coronary angiogenesis with gene therapy. To promote collateral circulation, Ad5FGF-4 was infused into coronary arteries of patients with chronic stable angina.
AGENT evaluated doses of Ad5FGF-4 from 3.3x108 to 3.3x1010 vp in 60 patients who tolerated the therapy well. Exercise time on a treadmill (ETT) improved in patients with baseline ETTs 
10 min, but not in pooled data from all doses. Additionally, 50% of patients given 1x1010 vp had increased ETT, versus 16% in the placebo group (
and 16, respectively, 
). Doses of 1010 and 109 vp were chosen for further study.
AGENT 2 determined whether Ad5FGF-4 improved myocardial perfusion compared with placebo. A significant decrease in ischaemic defect size was observed in patients treated with Ad5FGF-4 compared with baseline (
, 21% relative decrease, 
) that was not found in placebo group (
, 
). Again, Ad5FGF-4 was well tolerated.
AGENT 3 and 4 trials were planned to determine the efficacy and safety of Ad5FGF-4 in selected doses in larger populations. An interim review of the data from AGENT 3 indicated no safety concerns. However, differences in ETT were unlikely to reach significance. Therefore, further recruitment in the trials was stopped.
Key Words: Myocardial ischaemia • Stable angina • Angiogenesis • Fibroblast growth factor • Adenovirus • Clinical trials
Introduction
Coronary artery disease is the foremost cause of morbidity and mortality in the developed world. Conventional therapy consists of antianginal and antithrombotic medication, coupled with treatment of risk factors, such as hypertension and hyperlipidaemia. In patients who fail to respond to drug treatment, additional treatment options include revascularization by primary percutaneous coronary angioplasty or coronary artery bypass graft (CABG). However, some patients are not suitable candidates for revascularization, including those with diffuse coronary artery disease, microcirculatory impairment or repetitive CABG failure. There is, therefore, a need to develop alternative treatment strategies for such patients.
Therapeutic angiogenesis – administration of an angiogenic stimulus to promote the growth of collateral blood vessels to provide adequate tissue perfusion – is a promising approach for reducing myocardial ischaemia. Initial studies administering basic fibroblast growth factor (bFGF) as intravascular angiogenic protein growth factor therapy were, however, unsuccessful.1 Although modest benefits were obtained with multiple intracoronary/intravascular administrations of vascular endothelial growth factor (VEGF)-165, treatment effects were limited by the relatively short half-life of the protein.2 Attempts at direct injection of FGF protein into the myocardium of patients during CABG provided some proof of concept, with evidence of collateral blood vessel formation.3 However, the study did not provide evidence for the efficacy or safety necessary for routine use in patients with ischaemic heart disease, and the invasiveness of the technique (injection directly into the heart muscle during surgery) would limit its use.
Gene therapy offers an alternative approach for the delivery of angiogenic proteins, allowing more sustained production of the chosen therapeutic protein from a single administration. An adenoviral construct containing the human FGF-5 gene has been successfully tested in ameroid pig models of myocardial ischaemia, with significantly improved myocardial blood flow and wall thickening observed, and no evidence of hepatic or myocardial toxicity.4 However, the positive results observed in pre-clinical studies in animal models need to be translated into patient benefits. An adenovirus construct, Ad5FGF-4, containing cDNA for the human FGF-4 gene under the control of a strong promoter derived from cytomegalovirus has recently entered clinical testing.
The AGENT (Angiogenic GENe Therapy) trials were planned to test the clinical efficacy and safety of Ad5FGF-4 in patients. Four trials were designed (Table 1). AGENT and AGENT 2 were clinical trials assessing safety and efficacy, and have recently been completed with positive results.5,6 The results of AGENT 3 and AGENT 4 are not yet available.
|
AGENT trial5
The AGENT study was the first randomized, double blind, placebo controlled trial of gene therapy in therapeutic coronary angiogenesis to be conducted, and was designed to assess the safety and anti-ischaemic effects of Ad5FGF-4 in patients with stable exertional angina. An assessment of five doses of Ad5FGF-4, for the purpose of establishing the optimal safe and effective dose for future studies, was also incorporated into the study design.
A total of 79 patients with chronic stable angina (Canadian Cardiovascular Society [CCS] Class II or III), who had been stable for at least 2 months, were enrolled and randomized to receive either Ad5FGF-4 or placebo (the vehicle alone) in a 3:1 ratio. The doses of Ad5FGF-4 used ranged from 3.3x108 viral particles (vp) to 1011 vp increasing in half-log increments. Treatment was administered via 90-s intracoronary infusions through sub-selective catheters into each major coronary artery. Efficacy was assessed by measuring the increase in exercise treadmill testing (ETT) time, with assessments at 4 and 12 weeks post-treatment.
Safety is an important concern in any trial of gene therapy, and it is important to monitor any potential adverse effects from the treatment itself, as well as any systemic effects that might arise from administration of an angiogenic protein. These include atherosclerosis, oedema, retinopathy, and angiogenesis at unwanted sites, for example, in tumours.7 Safety was assessed by means of blood tests and physical examinations, which were performed at baseline and at 1, 2, 4, 8 and 12 weeks, and at 6 and 12 months. At baseline, patients were screened for the presence of common tumours. After angiogenic gene therapy, safety evaluations included: a viral infectivity assay to assess whether any virus entered the systemic circulation, which was performed on pulmonary artery blood (during intracoronary administration), venous blood (1 h post-administration) and urine (0–6 h post-administration); an assessment of FGF-4 levels in serum (24 h post-administration and at all follow-up visits) to establish whether overall FGF production increased, leading to potential systemic effects; adenoviral neutralizing antibody titres (baseline, 4 and 12 weeks); and semen analysis using polymerase chain reaction (PCR) testing for adenoviral DNA (8 weeks).
Tolerability and safety
Overall, doses of Ad5FGF-4 up to 3.3x1010 vp were well tolerated, with no safety concerns. Ad5FGF-4 was detectable in the pulmonary artery, the frequency of detection increasing with dose. Some patients (
40%) in the higher dose groups had detectable virus in the venous blood. No virus could be detected in the urine. There was no evidence of an increase in systemic FGF-4 levels; FGF-4 protein was not detectable at any time in the plasma samples. Semen testing at 8 weeks showed no detectable adenoviral DNA in the 12 patients willing to provide samples. Three patients receiving Ad5FGF-4 at the highest dose exhibited mildly elevated temperatures within 24 h of administration, one requiring an extra day of hospitalization as a result. However, this is not an uncommon occurrence in trials of adenovirus-based gene therapy.8
Long-term safety was also assessed, with a mean follow-up time of 311 days. There was no evidence of myocarditis, no signs or symptoms of new heart failure and no reduction in cardiac function. There were also no cases of retinal neovascularization in any patient during follow-up. There were 14 patients hospitalized for worsening angina, 10 of these from active treatment groups (10/60; 17%) and 4/19 (21%) from the placebo group. Adenovirus is associated with hepatotoxicity, but there was no evidence of this during the study, with only two patients experiencing transient liver abnormalities (one with elevated SGPT and one with elevated bilirubin).
Two patients who had received active treatment were diagnosed with malignancies. One presented with metastatic colon cancer 267 days after treatment and subsequently died of metastatic disease. The other presented with a brain tumour (glioblastoma multiforme) 69 days after receiving treatment. Both cases were unrelated to treatment; the patient with metastatic colon cancer had three first-degree relatives with colon cancer and most likely had hereditary nonpolyposis colon cancer. It was concluded that the tumour in the second patient had most likely been present when treatment was administered. PCR analysis showed that both tumours were negative for Ad5FGF-4 DNA.
Efficacy
Efficacy in AGENT was assessed by changes in ETT times, and primarily by percentage change in ETT time relative to baseline. There was no dose–response effect across the five different dosage groups, most likely as a result of the study being insufficiently powered to detect one. The data from the treatment groups were, therefore, pooled and compared with placebo. These results were highly encouraging. The mean increase in ETT times in the pooled treatment group at Week 4 and Week 12 were greater relative to the placebo group (Fig. 1(a)). However, this increase did not achieve significance. This might be explained by the fact that it is difficult to detect improvements in ETT times when baseline values are already quite high (i.e. approaching 10 min). When a post hoc analysis was performed to exclude all patients with baseline ETT times of over 10 min, the percentage increase in the treated group was significantly greater, relative to placebo, at 4 weeks (27% versus 12%, 
) and at 12 weeks (30% versus 22%, 
; Fig. 2).
|
|
The results from dose group 4 (1x1010 vp) are believed to be indicative of a physiological response, as this dose gave positive results in animal studies. Furthermore, this group included 22 patients as a result of a protocol amendment designed to test whether patients with patent bypass grafts should be included in future trials, and whether qualifying treadmill exercise time should be restricted to 10 mins and whether LVEF could be reduced safely from 40% or less to 30% or less. ETT time was significantly increased in this group compared with the placebo group (Fig. 1(b)). The percentage increase in treatment group 4 was also significantly higher than in the placebo group at 4 weeks (50% versus 16%,
), a trend that continued at 12 weeks (45% versus 21%). Finally, some patients might be less susceptible to gene transfer with adenovirus than others, due to neutralizing antibodies. Patients with lower neutralizing antibody titres to Ad5FGF-4 also did significantly better than those with higher titres in AGENT.
The results from AGENT were very promising, showing both tolerability and some evidence of efficacy, and doses of 109 and 1010 were selected for future study. However, no direct assessment of myocardial perfusion was performed, an issue that would be addressed in subsequent AGENT trials.
AGENT 2 trial6
The second trial in the AGENT programme, AGENT 2, was a randomized, double blind, placebo controlled trial in patients with stable angina (CCS Class II or III) who remained symptomatic despite receiving medication. The primary objective of this study was to determine whether intracoronary administration of Ad5FGF-4 could improve myocardial perfusion compared with placebo. The primary endpoint in this study was the change in stress-related reversible perfusion defect size (RPDS), as assessed by adenosine single positron emission computed tomography (SPECT) imaging at 8 weeks – a direct measure of the angiogenic effects of treatment. Safety was also assessed. A total of 52 patients with documented coronary artery disease, RPDS of more than 9%, left ventricular ejection fraction (LVEF) of at least 30%, and who were not optimal candidates for revascularization were enrolled. Ad5FGF-4 was to be administered at a dose of 1010 vp, and 35 patients were randomized to active treatment and 17 to placebo.
Tolerability 9
As with the AGENT trial, the safety of treatment was a paramount concern, and an extensive series of safety assessments, similar to those in AGENT, were performed. As before, the safety results showed that the treatment was well tolerated, with no safety concerns. One patient (3%) in the treatment group experienced a transient fever for which an extra day of hospitalization was required. Ten patients (29%) in the treatment group and five (29%) in the placebo group experienced a transient rise in liver enzymes (approximately double normal levels). There was no evidence of worsening myocarditis and no evidence of retinal neovascularization or angioma formation. The assessment of FGF-4 levels showed no systemic presence. In the placebo group, four patients (24%) developed worsening angina, three of these requiring angioplasty, compared with only two patients (6%) in the Ad5FGF-4 group. In addition, fewer Ad5FGF-4 patients required revascularization procedures compared with the placebo group (6% versus 18%).
One patient from the Ad5FGF-4 treatment group had a skin lesion removed, diagnosed as a squamous cell carcinoma. This patient had several skin lesions removed previously, all thought to be pre-malignant; however, no histopathology had been performed.
Efficacy
Administration of Ad5FGF-4 resulted in a significant reduction in RPDS relative to baseline at 8 weeks. Patients who received Ad5FGF-4 at a dose of 1010 vp had a 21% relative reduction compared with baseline, compared with a relative decrease of 8% in the placebo group (Fig. 3).10 The change in RPDS between Ad5FGF-4 and placebo was not significant (Fig. 3). However, the placebo group contained one patient who experienced a 50% improvement in RPDS, possibly due to initial poor compliance with anti-anginal medications. When this single outlier was excluded from the analysis, a significant difference between the two treatment groups was observed.
|
The symptoms of angina were also improved in the patients in AGENT 2. Ad5FGF-4 therapy increased the proportion of patients who experienced improved angina at Week 8, relative to placebo, as well as reducing the proportion of patients who required nitroglycerin therapy at Week 8 (Fig. 4). Interestingly, unlike in the AGENT study, patients with neutralizing anti-adenovirus antibodies had no apparent reduction in response to Ad5FGF-4 in AGENT 2.
|
The data from AGENT 2 show that Ad5FGF-4 therapy was well tolerated and provide evidence for an angiogenic effect, with an encouraging trend for improved myocardial perfusion.
AGENT 3 and AGENT 4
Two larger Phase III clinical trials were subsequently designed to evaluate further the safety and efficacy of Ad5FGF-4. The first, AGENT 3, was to be conducted exclusively in the US, planned to recruit patients with stable angina (CCS Class II to IV) with optimal anti-anginal therapy, a baseline ETT time of 3–10 min, and an LVEF of at least 30%. AGENT 4 was to be conducted in Europe, and North and South America and include patients meeting the criteria for inclusion in AGENT 3 but who were poor candidates for revascularization therapy. Both trials were designed as randomized, double blind and placebo controlled studies, with a recruitment goal of 450 patients each. Randomization was to either active treatment (Ad5FGF-4 at a dose of 109 or 1010 vp) or to placebo in a 2:1 ratio, with equal numbers of patients being randomized to the two active treatment groups (Fig. 5).
|
Two primary efficacy endpoints were planned for both studies: the change from baseline in ETT at 12 weeks and the proportion of patients who increase ETT by more than 30% at 12 weeks. Other endpoints were to include additional measures of symptomatic improvement and long-term cardiovascular morbidity and mortality. A more thorough investigation of safety endpoints, with a considerably longer period of follow-up, was also planned for both trials.
At the time of this presentation, the AGENT 3 and 4 studies were ongoing. However, as of Jan 30, 2004, enrollment was stopped because interim data analysis of AGENT 3 indicated that the studies, as currently designed, will provide insufficient evidence of efficacy. No evidence of important safety concerns was found. Patient follow-up will be continued, and data from these trials will be available at a later date.
Conclusions
The results from the first two AGENT trials were promising. Both showed that the Ad5FGF-4 treatment is well tolerated, and the mild adverse events experienced by patients in both studies appear to be consistent. These data are similar to those from other clinical trials of adenoviral gene transfer to the heart and, together, show that no serious adverse events appear to be attributable to infusion of Ad5FGF-4 into the coronary arteries.
It was hoped that the larger Phase III AGENT studies of greater duration would produce more definitive results with regard to the benefits of angiogenic therapy with Ad5FGF-4. When the full dataset is analysed, significant data on the safety of angiogenic gene therapy with Ad5FGF-4 and on efficacy in patient subgroups will be available. Future studies in angiogenic gene therapy will build on knowledge gained from the AGENT trials.
Footnotes
Dr. Grines has received research funding from Berlex Pharmaceuticals for AGENT trials.
References
- Lazarous DF, Shou M, Stiber JA, et al. Pharmacodynamics of basic fibroblast growth factor: route of administration determines myocardial and systemic distribution. Cardiovasc. Res. 1997;36(1):78–85
[Abstract/Free Full Text] - Henry TD, Annex BH, McKendall GR, et al. The VIVA trial: vascular endothelial growth factor in Ischemia for Vascular Angiogenesis. Circulation. 2003;107:1359–1365
[Abstract/Free Full Text] - Ruel M, Laham RJ, Parker JA, et al. Long-term effects of surgical angiogenic therapy with fibroblast growth factor 2 protein. J. Thorac. Cardiovasc. Surg. 2002;124:28–34
[Abstract/Free Full Text] - Giordano FJ, Ping P, McKirnan MD, et al. Intracoronary gene transfer of fibroblast growth factor-5 increases blood flow and contractile function in an ischemic region of the heart. Nat. Med. 1996;2:534–539[CrossRef][ISI][Medline]
- Grines CL, Watkins MW, Helmer G, et al. Angiogenic Gene Therapy (AGENT) trial in patients with stable angina pectoris. Circulation. 2002;105:1291–1297
[Abstract/Free Full Text] - Grines CL, Watkins MW, Mahmarian JJ, et al. A randomized, double-blind, placebo-controlled trial of Ad5FGF-4 gene therapy and its effect on myocardial perfusion in patients with stable angina. J. Am. Coll. Cardiol. 2003;42:1339–1347
[Abstract/Free Full Text] - Isner JM, Vale PR, Symes JF, Losordo DW. Assessment of risks associated with cardiovascular gene therapy in human subjects. Circ. Res. 2001;89:389–400
[Abstract/Free Full Text] - Ylä-Herttuala S, Martin JF. Cardiovascular gene therapy. Lancet. 2000;355:213–222[CrossRef][ISI][Medline]
- Watkins M, Kleiman N, Helmer G, et al. Intracoronary angiogenic gene therapy is well tolerated in patients with coronary artery disease. Eur. Heart J. 2003;24(suppl):P1182 (Poster)
- Grines C, Rubanyi GM, Kleiman NS, et al. Angiogenic gene therapy with adenovirus 5 fibroblast growth factor-4 (Ad5FGF-4): a new option for the treatment of coronary artery disease. Am. J. Cardiol. 2003;92(Suppl.):24N–31N[CrossRef][ISI][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A R Lyon, M Sato, R J Hajjar, R J Samulski, and S E Harding Gene therapy: targeting the myocardium Heart, January 1, 2008; 94(1): 89 - 99. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||