The Antithrombotic Conundrum: Is REG1 a Game Changer?

[4]Cath Lab Digest talks with David J. Mazzo, PhD, President and CEO of Regado Biosciences, Inc., Basking Ridge, New Jersey, about REG1, a rapidly reversible, actively controllable anticoagulant system.

Could you describe the problem you are trying to solve with REG1?

With REG1, Regado Biosciences is addressing what we have coined as “the antithrombotic conundrum”; i.e., how do you prevent harmful clots while allowing healing clots to form and develop naturally? Until now, all agents used in antithrombosis or anticoagulation do very well on one side of the equation while giving less than ideal performance on the other side. A good example is heparin, which is generally accepted as a satisfactory anticoagulant and does very well in that regard, but there is a history of bleeding associated with its use. Bivalirudin, another popular anticoagulant, actually exploits heparin’s weakness and does a very good job of reducing bleeding, but often at the expense of increased risk of ischemic events, as seen in several clinical trials. In today’s world, a physician choosing to use an injectable anticoagulant is making a compromising decision.  
REG1, however, performs optimally by reducing the risk of both ischemia and bleeding. It gives us the ability to tune or titrate the amount of anticoagulation at any time, in real time. REG1 is a two-component system:

1. Pegnivacogin (RB006), the therapeutic aptamer, a Factor IX inhibitor; 
2. Anivamersen (RB007), its specific, complementary, active control agent. 

We administer pegnivacogin like you would administer bivalirudin or heparin: by injection, to anticoagulate the patient. However, pegnivacogin and anivamersen are administered as a bolus injection, while heparin and bivalirudin are administered as IV drips. In our case, with a 1mg/kg dose of pegnivacogin, we can completely inhibit nearly all of the Factor IX in the body, which effectively makes our drug a Factor X knockout and essentially curtails thrombin generation. In practical terms, this means REG1 patients are as anticoagulated as they are physiologically capable of being and are maximally protected during an intervention or invasive procedure, when foreign bodies are put into the vasculature and clots are likely to occur.  Immediately thereafter, if pegnivacogin couldn’t be modulated, that patient could be at a very high risk for bleeding. Since we can then administer anivamersen and completely or partially reverse the anticoagulation effects of pegnivacogin, the anticoagulation is reduced to such a low level that the result is a bleeding profile for REG1 that is likely superior to heparin and perhaps even superior to bivalirudin. We get the best of both worlds because we have real-time control with the administration of anivamersen.

What about timing of the reversal agent?

It is at the physician’s discretion. When physicians would essentially call the procedure done if it was a heparinized patient, at that point in time it’s recommended they administer anivamersen and wait a short period (the RADAR protocol stipulated 10-13 minutes). The arterial sheath can then be immediately pulled and the patient sent on from the lab.

With the appropriate dose of anivamersen, there is no concern regarding bleeding.  The RADAR Phase IIb trial showed a trend for REG1 to reduce major bleeding in comparison to heparin, and in fact, to also reduce ischemia in comparison to heparin. REG1 could then also reduce ischemia much more than bivalirudin and do equally or better on bleeding.

RADAR trial data was presented at the American College of Cardiology Scientific Session in March 2011. Can you share more about the trial results?

The RADAR trial was a Phase IIb study with the specific objective of determining the dose or dose range of anivamersen that provides a bleeding profile superior to heparin after a standard 1mg/kg dose of pegnivacogin. We already know the 1mg/kg dose of pegnivacogin is sufficient to completely anticoagulate a patient. The principal role of RADAR was to define the dose range for the reversal agent. It was an adaptive design study against heparin as the active comparator. We used standard endpoints for bleeding and ischemia. Since this was a dose-ranging trial for the control agent, the primary endpoint was bleeding according to the ACUITY bleeding scale, both total bleeding and major bleeding. A secondary endpoint was a standard ischemic composite, which included death, myocardial infarction, target vessel revascularization and recurrent ischemia. As fairly standard regulatory endpoints, these would also be predictive of REG1’s performance in later clinical development pivotal trials. We also included a number of pharmacoeconomic indicators, such as time to beginning a catheterization, post administration of the anticoagulant, time to removal of the arterial sheath, concomitant medications and use of closure devices.

RADAR had an acute coronary syndrome (ACS) patient population, so these were all non-ST-elevation myocardial infarction (STEMI)/unstable angina patients who were intended for cardiac catheterization within 24 hours. This is indicative of at least one of the populations in which we expect to use REG1 in Phase III, the other population being elective percutaneous coronary intervention (PCI) patients.  RADAR was a multi-national study with 69 enrolling sites in 6 countries, including the U.S., Canada, Poland, France, Germany and The Netherlands.

As an adaptive design trial, it was administered in two parts. The first part was the anticoagulation step, and at that step, patients were randomized either to pegnivacogin or to heparin, and that was done open label, so physicians and patients knew whether they were getting the experimental drug or heparin. At the end of the procedure, a blinded dose of the reversal agent was administered.

In the pegnivacogin cohort, patients could receive one of 4 blinded reversal levels, from 25% reversal, which obviously is not very much reversal, up to 100% reversal, in 25% increments. The heparin group, obviously, didn’t get reversal, but received heparin standard of care. At the end of the procedure for the pegnivacogin group, the arterial sheaths were removed 10-13 minutes after the dose of anivamersen. In the case of the heparin group, the arterial sheaths were removed approximately four to six hours after the end of the procedure, the typical wait period for most cath labs after the use of heparin as the anticoagulant in a PCI. The study was looking to further confirm that the 1mg/kg dose of pegnivacogin provided rapid, reproducible, precise and complete Factor IX inhibition, to determine the dose range of anivamersen that provided safe early sheath removal, and also to give us the chance to assess the efficacy of REG1 in comparison to heparin, based on ischemic events. RADAR was a Phase IIb trial, so it was not powered for an ischemia endpoint. It was powered for the endpoints I described earlier, relative to bleeding and for dose assessment, but RADAR was designed to allow us to make certain statistical assumptions and calculations based upon the other endpoints.  Although RADAR was originally intended to enroll 800 patients, our steering committee (SC) decided there were a sufficient number of events to stop the trial at 640 patients. This evaluation of events was undertaken because of three specific adverse events that occurred late in the trial. After several months of investigation, it was concluded these events could not be specifically attributed to pegnivacogin or anivamersen. We could actually evaluate all the endpoints as-is, and so that was the decision. The target population was achieved. In both groups, the median age was approximately 63 years old; about two-thirds were male. Almost all had the typical cardiovascular risk factors: diabetes, smoking, prior myocardial infarction, prior PCI, many had very large body mass indices, and about a third were over the age of 70. It was a good population in which to test the drug and the hypothesis.  Most were on the standard background medications, which included thienopyridines, aspirin, glycoprotein inhibitors and statins.

At about 100 patients enrolled during the first scheduled data safety and monitoring board (DSMB) evaluation, it was determined that the 25% arm was not being effective, and that arm was eliminated. Allowing this evaluation and elimination was part of the design of the trial. The DSMB did similar evaluations at 200 and 400 patients enrolled, and determined that the remaining arms of 50%, 75%, and 100% were performing adequately, and they were left in. The trial completed with those 3 arms. The results, first using a 30-day endpoint for total bleeding, showed that at least a 50% reversal was necessary in order to achieve hemostasis sufficiently after PCI such that you could immediately remove the sheath without having an increase in bleeding versus heparin. At day 30 for major bleeding according to the ACUITY scale, there was numerically less major bleeding compared to heparin going from 50% to 100%, showing a decrease in major bleeding versus heparin as the reversal increased. If you then looked at the same data through hospital discharge, averaging around three days for this patient population, there was a numerical dose response for total bleeding and for major bleeding. At the 100% reversal dose, there was a statistical significance between major bleeding and heparin, and a trend with a nearly significant p-value for total bleeding and heparin, something that was not expected. We thought we would see the trends, but for an endpoint that was not a primary endpoint (major bleeding) and for which the study was not powered, that we were able to see statistical significance. It was very encouraging on the bleeding side.

“Response through discharge” is expected to be our Phase III endpoint. The FDA, in previous discussion with us, suggested that “through discharge” would be the more relevant endpoint for an acute care interventional procedure, occurring closer to the intervention, as opposed to up to a month away, and so they have been encouraging us to consider a shorter duration of monitoring. The “through discharge” endpoint is one we have evaluated as a means to predict what our Phase III results might look like. RADAR data bode very well for Phase III bleeding.

Similarly, if we look at the secondary (ischemic) endpoint, we see a numerical reduction of almost 50% in ischemic events versus heparin, based on the odds ratio of the endpoint. It is a secondary endpoint and the study is not powered to show statistical significance, but we saw a very consistent reduction through day 30. If we look at that same result through discharge, we see an even greater reduction, one of about 66% in ischemic events through discharge in comparison to heparin.

Based on an assessment of the pharmacoeconomic endpoints in RADAR, REG1 should reduce total costs in the PCI lab. For example, with REG1 we demonstrated the ability to remove the arterial sheath up to 4x faster than with heparin, without incurring adverse events of bleeding; the ability to start a procedure more rapidly, and the likelihood that we would, in the long run, be able to eliminate typical concomitant meds that are often used with heparin, specifically, glycoprotein IIb/IIIa inhibitors, as well as arterial closure devices. In addition, REG1 improves administration convenience. REG1 is an IV bolus as opposed to IV drips and onset of action is essentially instantaneous, as is the reversal. Use of REG1 offers the ability to immediately pull sheaths, and therefore the ability to move patients through the cath lab and hospital recovery areas much more rapidly and safely. In most cases, our patients were ambulated and moving out of the recovery area in two hours. This attribute of REG1 could possibly lead to earlier hospital discharge.

Since we used endpoints that would be classically derived from a regulatory perspective, and we included a second look at the data through discharge, in addition to through day 30, we have a great ability to predict the Phase III performance of our drug. We expect these to be the same endpoints in Phase III, except the primary endpoint will be ischemia and the secondary endpoint will be bleeding, and here, we think that RADAR data portend a high probability of Phase III success. It also allows us to produce a classic mortality/morbidity study for Phase III, with a superiority design versus bivalirudin.

Are there any issues with clearance of the reversal agent?

No. Both pegnivacogin and anivamersen are single-strand oligonucleotides and they are metabolized by blood nucleases. In the case of pegnivacogin, a 40-kilodalton polyethylene glycol (PEG) moiety has been added, which gives that agent a very long half-life. If single-stranded oligonucleotides are not modified to possess a long half-life, they are metabolized within a few minutes. Anivamersen, our reversal agent, has a half-life of less than five minutes, for example. These agents are metabolized by blood nucleases, so there should be no dosing limitations or adjustments based on reduced kidney or liver function. They can be used in a standard fashion, and the phamacokinetics and metabolism will be similar to a non-renally and/or non-hepatically impaired patient.

Did RADAR have 100% femoral access?

Yes, and we chose that because, in the U.S., femoral access is still the predominant approach. In other countries, radial access is growing, but since our primary endpoint in this particular study was bleeding, in order to determine the right dose of the reversal agent, we wanted to use femoral access. In the Phase III program and for commercial application, we will include radial as well as femoral access, and we expect the commercial label will allow it to be deployed no matter what access you use. We still think the benefits are there. Some people may say well, you don’t have as much of a need for a bleeding benefit if you are using radial access, but our story isn’t all about bleeding. Our story is all about balance and the ability to reduce ischemia even further, whether you are using radial or femoral access.

Did RADAR show any differences in the use of closure devices (between heparinized and REG1 patients)?

This point was not studied in the RADAR trial; however, there are some misconceptions we run into when speaking to people about REG1, even in the cath lab. We have had active interventionalists telling us, “You know, we have kind of forgotten about ischemia, because we have been focused on bleeding for so long.  We had low expectations about being able to do anything more with ischemia than is already being done with heparin, so we figured we had reached a plateau there that was never going to be raised.” It’s not a story all about just bleeding or all about just ischemia. It’s about being able to do both, in a manner that is completely under the control of the medical staff in the cath lab.

Dr. David Mazzo can be contacted at (908) 580-2100.

Suggested Reading

1. Povsic TJ, Cohen MG, Chan MY, et al. Dose selection for a direct and selective factor IXa inhibitor and its complementary reversal agent: translating pharmacokinetic and pharmacodynamic properties of the REG1 system to clinical trial design. J Thromb Thrombolysis 2011 Jul;32(1):21-31.
2. Povsic TJ, Cohen MG, Mehran R, et al. A randomized, partially blinded, multicenter, active-controlled, dose-ranging study assessing the safety, efficacy, and pharmacodynamics of the REG1 anticoagulation system in patients with acute coronary syndromes: design and rationale of the RADAR Phase IIb trial. Am Heart J 2011 Feb;161(2):261-268.e1-2.
3. Povsic TJ, Wargin WA, Alexander JH, et al. Pegnivacogin results in near complete FIX inhibition in acute coronary syndrome patients: RADAR pharmacokinetic and pharmacodynamic substudy. Eur Heart J 2011 Jun 30. [Epub ahead of print].