Resurrecting rosiglitazone: FDA panel recommendation could affect its use

After years of debate and controversy over the troublesome past and uncertain future of rosiglitazone, the FDA’s Endocrinologic and Metabolic Drugs Advisory Committee and Drug Safety and Risk Management Advisory Committee voted to ease existing restrictions on the drug. In the weeks following the vote — a result of 2 days of hearings in June — Endocrine Today spoke with diabetes and cardiology experts who have considered the data that prompted the global controversy.

Current Risk Evaluation Mitigation Strategy with Elements to Assure Safe Use (REMS/ETASU) restrictions require health care professionals to submit forms on each patient prescribed rosiglitazone (Avandia, GlaxoSmithKline), and patients must acknowledge that they understand potential cardiovascular risks before they can receive the drug. Rosiglitazone is available to new patients only if other medications have failed.

The committee’s vote included 13 members in favor of modified restrictions, seven for removal of all restrictions, five for unchanged restrictions and one for withdrawal of the drug from the US market. The agency will consider the committee’s vote before making a final decision, but is not required to follow the recommendations of the advisory committee.

Use of rosiglitazone, a thiazolidinedione indicated as an adjunct to diet and exercise for patients with type 2 diabetes, has been controversial since 2006 when data emerged suggesting a potential risk for CV events, including myocardial infarction and stroke.

“This meeting supported the sanctity of performing randomized trials to assess the safety of these drugs. It emphasized we cannot rely upon underpowered meta-analyses from small observational studies to reach sweeping conclusions,” C. Michael Gibson, MD, FACC, the director of PERFUSE Study Group, associate professor at Harvard Medical School, and chief of clinical research in the division of cardiology, told Endocrine Today.

Recommendations were based on the discussion of a highly anticipated, independent analysis of the FDA-requested Rosiglitazone Evaluated for Cardiovascular Outcomes and Regulation of Glycemia in Diabetes (RECORD) trial.

“I think that justice may have been served,” Robert H. Eckel, MD, the Charles A. Boettcher II endowed chair in atherosclerosis, professor of medicine with appointments in the division of endocrinology, metabolism and diabetes and the division of cardiology, and professor of physiology and biophysics at the University of Colorado Denver (UCD), told Endocrine Today with regard to the decision.

Eckel, who is also the director of the Lipid Clinic at the University of Colorado Hospital and director of the Discovery Translation Program of the NIH-funded Colorado Clinical Translational Sciences Institute, said he co-wrote a paper in Circulation about 3 years ago based on the RECORD data, reporting that he and colleagues were not entirely sure the evidence was convincingly incriminatory.

“We were being as objective as possible at that time. I have no bias for or against the drug; we simply felt that the evidence was not sufficient to have the drug pulled from the market. The warnings may have been reasonable at the time; however, now that the RECORD trial has been objectively reviewed and found to be safe for patients, I believe it is appropriate to take a step back from the black box warnings,” he said. “I don’t think any of us really know what the new REMS 
will include. It remains doubtful, however, that prescribing patterns will change now ”

The 2-day meeting was called due to “the public interest in Avandia, the extensive history of the product and the continued uncertainty of the risk surrounding this drug,” Janet Woodcock, MD, director of the FDA Center for Drug Evaluation and Research (CDER), wrote in an FDA blog.

Reanalysis of data

RECORD was a more than 5-year study of more than 4,400 patients designed to evaluate long-term effects of rosiglitazone on CV outcomes and blood glucose control compared with metformin and sulfonylureas. Initial results reported in 2009 demonstrated no increased risk for CV morbidity or mortality with rosiglitazone.

The independent readjudication of RECORD was conducted by the Duke Clinical Research Institute (DCRI) at the request of the FDA. The analysis included follow-up for mortality for 25,833 person-years, including an additional 328 person-years identified during a re-evaluation. Overall, researchers reported 184 CV or unknown-cause deaths (88 rosiglitazone, 96 metformin/sulfonylurea), 128 patients with MI (68 rosiglitazone, 60 metformin/sulfonylurea) and 113 patients with stroke (50 rosiglitazone, 63 metformin/sulfonylurea). The HR for rosiglitazone vs. metformin/sulfonylurea for the endpoint of CV death/MI/stroke was 0.95 (95% CI, 0.78-1.17) vs. 0.93 (95% CI, 0.74-1.15) for the original RECORD results. Treatment comparisons for MI (HR=1.13; 95% CI, 0.8-1.59) and mortality (HR=0.86; 95% CI, 0.68-1.08) also were the same compared with the original results.

“Observed HRs and CIs from these analyses using the original RECORD or new FDA endpoint definitions showed similar treatment effects of rosiglitazone compared with the original RECORD results,” the DCRI researchers concluded in the analysis, which was published online in American Heart Journal.

During the advisory panel meeting, most members agreed that the HRs and CIs from the analyses were consistent and that the drug did not increase risk for all-cause or CV mortality. Even so, the RECORD trial design, reliability of CV endpoints, the drug’s contraindication with statins and the absence of data on major adverse CV events came under fire.

“RECORD was too inadequately designed and conducted to provide any reassurance about the CV safety of rosiglitazone,” Thomas A. Marciniak, MD, of the division of cardiovascular and renal products at the Office of Drug Evaluation I, Office of New Drugs and CDER at the FDA, said during the meeting.

Panel member T. Mark Woods, PharmD, clinical coordinator and residency program director in the pharmacy department of Saint Luke’s Hospital in Kansas City, Mo., voted in favor of REMS/ETASU modification. “While [readjudication of] the RECORD trial helped reassure us in some aspects, it also introduced new questions. For example, the statin issue is very much still up in the air, and given the number of diabetic patients that will end up on statins, I have lingering concerns about toxicity,” he said.

Experts weigh in

“Regarding safety, the story of rosiglitazone is a cautionary tale. With 14 years since the FDA approval, [the drug] is still lacking conclusive evidence,” Ralph G. Brindis, MD, MPH, FACC, president-elect of the American College of Cardiology, said during the open public hearing portion of the advisory meeting.

“Rehashing clinical trial data only goes so far,” Brindis said. “Improved real-time surveillance is critical.”

In a comment after the meeting, American Heart Association national spokesman Jorge Plutzky, MD, said, “Those voting to modify current restrictions may believe it would enable researchers to more definitively assess the potential risk for heart attack and other major heart events with this drug. Many panelists expressed the need for more data, which would be difficult to obtain under the current level of restriction.”

In an interview after the vote, George L. Bakris, MD, FASH, FASN,Endocrine Today Editorial Board member, professor of medicine and director of the Comprehensive Hypertension Center at the University of Chicago School of Medicine said, “After reviewing the quality of data put together by DCRI and the vote cast by the metabolic unit of the FDA, it is clear that a more definitive opinion about CV safety could not be confirmed. I applaud the effort of DCRI and the FDA committee to try and get at the truth with [proliferator-activated receptor] gamma agents, which when used in lower doses are not associated with untoward effects, and the benefits outweigh the risk.”

Robert E. Ratner, MD, FACP, FACE, chief scientific and medical officer of the American Diabetes Association, said he does not anticipate that the vote will have a “major change” on the treatment of patients with diabetes.

“I don’t see a major future for rosiglitazone from a clinical standpoint. The more important aspect of this particular meeting is that the agency is reinforcing the appropriate process by which data need to be reviewed and how logical decisions, as opposed to emotional decisions, are made,” Ratner said.

Previous research

As an investigator in the Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial, George Grunberger, MD, FACP, FACE, chairman of the Grunberger Diabetes Institute and clinical professor of internal medicine and of molecular medicine and genetics at Wayne State University School of Medicine, told Endocrine Today that, to his knowledge, there was no risk for CVD associated with the use of rosiglitazone in ACCORD.

The trial compared intensive therapy with standard approaches to target normal HbA1c for reduced CV risk in diabetes and demonstrated that intensive therapy for 3.5 years was associated with increased mortality and did not significantly reduce major CV events. However, researchers reported that they were unsure of what caused the increased risk for mortality, although, unlike the glycemia trial, the ACCORD BP and lipid trials did not suggest serious harm associated with intensive therapy.

“Think about all of the patients enrolled in the ACCORD trial who were placed on rosiglitazone for years; they had advanced diabetes and already had established CVD or several risk factors,” Grunberger said. “There was no signal from ACCORD.”

Similarly, Grunberger said the results from the Bypass Angioplasty Revascularization Investigation in Type 2 Diabetes (BARI 2D) trial add to the lack of evidence that rosiglitazone increases CV events. The data demonstrated that there were no differences in outcome between patients assigned to insulin-sensitizers, such as rosiglitazone, and those assigned to insulin-providers (77.7% vs. 75.4%; P=.13)

In 2010, the Thiazolidinedione Intervention with Vitamin D Evaluation (TIDE) trial was halted due to CV safety concerns associated with rosiglitazone. It was proposed by GlaxoSmithKline to meet the CV outcomes trial requirement from the FDA and would have included data from roughly 16,000 patients with type 2 diabetes. However, no such data are available due to the termination of the trial.

“The problem is that the reputation of the drug has been tainted and the TIDE trial was supposed to answer the question definitively, but because of the publicity involved, it would be difficult to recruit patients in sufficient numbers to answer the question of safety,” Robert A. Vigersky, MD, past-president of The Endocrine Society and director of the Diabetes Institute at the Walter Reed National Military Medical Center in Bethesda, Md., told Endocrine Today.

Days after TIDE was stopped in September 2010, the European Medicines Agency (EMA) chose to ban rosiglitazone in Europe.

Clinical effects of controversy

In the midst of the controversies stemming from a meta-analysis published in The New England Journal of Medicine by Steven E. Nissen, MD, and Kathy Wolski, MPH, of the Cleveland Clinic in 2007, clinicians were faced with the decision to continue prescribing rosiglitazone for their patients or consider alternatives. At the same time, some said the negative publicity placed on rosiglitazone may have taken away from novel new therapies and further clinical studies.

“Clearly, the ramifications of the RECORD trial and the fiasco which followed are still making waves. You can argue that it has set back the progress and treatment of diabetes for many years. I believe alogliptin (Nesina, Takeda Pharmaceuticals) would’ve been on the market for years now if it weren’t for the repercussions of the meta-analysis. It can also be argued that we didn’t improve the health of people with diabetes thanks to this new requirement for intense CV outcomes studies related to diabetes drugs,” Grunberger said.

And according to Vigersky, some institutions terminated their planned protocol to use rosiglitazone and other TZDs as a potential arm in comparator effectiveness trials based on Nissen’s publication in 2007.

“The point is that much of the data can be garnered about safety in postmarketing studies that will ultimately determine whether one or another drug should or should not remain on the market because of the risks outweighing the benefits,” Vigersky said. “Unfortunately, I don’t think we’re going to get that. It’s nearly impossible that we will ever obtain that information from rosiglitazone studies.”

The future of rosiglitazone

Pending the FDA’s decision, rosiglitazone will be available through the REMS/ETASU program to physicians and appropriate patients, according to a statement from GlaxoSmithKline.

Only about 4,600 patients in the United States are currently using rosiglitazone, Robert Bigelow, PhD, DCRI senior statistician for the re-evaluation, said at the FDA panel meeting. For this reason, many researchers now think of it as an orphan drug. However, it is too early to tell how the proposed loosening of the REMS criteria will affect practice, according to Eckel.

“However, the reduction in all-cause mortality, even in an unblinded trial, is reassuring. There were no statistically significant increases in the risk for MI,” Eckel said. “Again, all cardiologists place much more credence in data from prospective randomized comparisons. And I hope that in the future we can avoid being swayed by faulty meta-analyses.”

Grunberger has a message for clinicians: Forget the headlines and consider the facts when assessing the risks associated with diabetes. Pioglitazone (Actos, Takeda Pharmaceuticals) is another TZD that remains available despite the news of its purported risk for bladder cancer. “As a clinician, one does not have the luxury either to wait for decades for the ultimate safety data or place a patient with diabetes on placebo,” he said.

“However, we certainly still have patients on Actos. I often adopt the practice of the physicians that I work with because it helps with the acceptance of my recommendations. We rarely ever start anybody on a TZD anymore,” Nathan A. Painter, PharmD, CDE, practicing clinical pharmacist and associate clinical professor of the Skaggs School of Pharmacy and Pharmaceutical Sciences at the University of California, San Diego, told Endocrine Today.

Painter said by the time he is confronted by a patient, he is already taking three to four medications, or he is titrated to insulin.

“I become involved once we’re past the discussion of whether or not a TZD should be added. When I am asked questions about it, I always mention TZDs as an option. The uptake by physicians is poor,” he said.

By the time patients who are assigned TZDs see Painter, they have been on their medication a while, are doing fairly well, have a low risk or no risk for heart failure and are not taking other medications that could increase their risk for edema, he said.

“I don’t like to call them ‘healthy’ diabetes patients, but I tend to believe that TZDs are for the healthier, younger or slimmer patients with type 2 diabetes,” Painter said.

The history of rosiglitazone has undoubtedly shaped the way clinicians and the FDA now view CV event risk in diabetes medications. Although the future of rosiglitazone is uncertain, clinicians and experts in the diabetes and cardiology fields have echoed the sentiment that it will not change treatment strategies.

According to Vigersky, there are three lessons to be learned by what has happened in the case of rosiglitazone.

“One of the things we learned is that the pharmaceutical industry must be transparent with their data. Secondly, trials should be conducted in collaboration with the FDA and carefully planned to look at all of the relevant potential safety issues. Third, there should always be active comparators because we’re dealing with real-world patients.”

He said the modification of the REMS is essentially not going to affect the use of rosiglitazone.

“The likelihood that providers will begin to initiate treatment with rosiglitazone is slim to none,” Vigersky said. – by Samantha Costa, with additional reporting by Katie Kalvaitis

Are TZDs still useful?

Yes

I don’t think Avandia will ever be resurrected. That being said, with the new diabetes drugs in the pipeline, there is still a role for the TZD pioglitazone in the treatment of patients with type 2 diabetes.

First of all, TZDs are very powerful at lowering HbA1c; more powerful than some of the new agents (ie, DPP-IV inhibitors and glucagon-like peptide 1 receptor agonists). Pioglitazone can often lower the HbA1c two or three points, whereas the newer agents typically drop HbA1c one or two. Of course, TZDs have adverse events; they include fluid retention, exacerbating congestive heart failure, and there is some concern about its effect on osteoporosis. Yet, in general, most patients currently on TZDs do fairly well on them. Specifically, pioglitazone has a favorable ratio of raising HDL and lowering LDL, so it has beneficial CV effects that the other drugs don’t have.

Pioglitazone has been around a fairly long time, and many diabetologists and endocrinologists are familiar with using it. Further, pioglitazone is a pill as opposed to the GLP-1 drugs, which are injection; it is obviously more favorable in that respect.

Another important factor to consider is that many patients do not want to take insulin, and many doctors do not want to start patients on insulin. I expect that there are going to be more combinations of the oral drugs, such as metformin, sulfonylureas, TZDs, GLP-1 agonists, DPP-IV inhibitors, to spare patients from going on insulin therapy. I think having TZDs drugs in the armamentarium of a diabetologist or endocrinologist is still useful.

Not as first-line therapy

After everything we’ve heard about Avandia, no patient is going to be interested in taking it. There are real downside risks with the drug in terms of fluid retention and bone fracture, to say nothing of the weight gain. At this point in time, there may be better treatment choices.

TZDs are unmatched in their ability to improve insulin sensitivity and reduce the insulin resistance found in many patients with type 2 diabetes, as this resistance lies at the core of the progression of type 2 diabetes and loss of insulin secretory capacity. They are also unmatched in their ability to preserve and protect residual beta-cell function. Unfortunately, owing to the controversies created by the now apparently unsubstantiated allegations regarding CV risk with rosiglitazone, it seems unlikely that any significant number of patients would be willing to be started on this agent without some demonstrable evidence of benefit to the patient. Owing to the appreciable weight gain seen with rosiglitazone, as with all TZDs, together with the increase in bone fracture rates, particularly in women, and especially those who are postmenopausal in nature, there is a clear negative connotation for the use of TZDs early in the course of type 2 diabetes. For some patients with severe insulin resistance in whom a TZD insulin sensitizer greatly modifies their response to injectable insulin, these agents might be useful but are unlikely to be first-line agents in the bulk of patients with type 2 diabetes.