Controversial LIGHT Trial Now Published but Fails to Illuminate CV Risks of Contrave
The “final” results of a prematurely terminated study assessing the cardiovascular safety profile of the weight-loss drug Contrave are now published, but given the controversies that dogged this trial, physicians are no closer to knowing if the medication poses a risk for obese and overweight patients—those to whom the drug is targeted.
LIGHT, the cardiovascular safety study of the weight-loss drug Contrave (Orexigen, Takeda Pharmaceuticals), a naltrexone/bupropion combination, broke down over a very public dispute between the study’s leadership, sponsors, and the US Food and Drug Administration.
Steven Nissen, MD (Cleveland Clinic, OH), who led the study, said a second study needs to be done to get the answers LIGHT failed to provide. “Patients and physicians need to be aware that the cardiovascular safety of the drug remains uncertain,” he told TCTMD.
The publication of the second interim—and final—analysis of the LIGHT trial online March 8, 2016, in the Journal of the American Medical Association marks the end of this particularly strange saga, one that saw clinical trial investigators and the FDA castigating the study sponsor for widely distributing confidential interim results and then for later claiming a cardiovascular benefit with Contrave based on a trial that was just 25% completed.
Given those actions, the LIGHT trial was stopped prematurely as investigators believed the study had been compromised and would be unable to provide clinically meaningful results. Now, in JAMA, investigators have published the results of a new interim analysis conducted after 50% of events occurred within the event-driven trial. It represents the last prespecified analysis performed by the Data Safety and Monitoring Board prior to public disclosure.
“This is an unusual situation, and it’s very important that this doesn’t become a pattern,” Joshua Sharfstein, MD (Johns Hopkins University, Baltimore, MD), who wrote an editorial accompanying the LIGHT publication, told TCTMD. “When these studies are set up, they should be allowed to be completed on their own terms. What happened here was a disclosure of data that fundamentally interfered with the conduct of the study.”
Out of the Shadows
The FDA’s consideration of Contrave followed a relatively new path, with the weight-loss medication undergoing expedited consideration on the basis of positive trials using surrogate endpoints—in this case weight loss—for patients considered obese or overweight and at high risk for cardiovascular events. Similar to the pathway developed for diabetes medications, approval was contingent on an additional, large-scale trial being undertaken to prove the drug didn’t increase the risk of adverse cardiovascular events, such as MI or stroke. This was LIGHT, a placebo-controlled, noninferiority cardiovascular outcomes study designed in part to allay concerns raised in phase 3 studies showing that Contrave increased heart rate and blood pressure while reducing weight.
To expedite approval, the FDA allowed an interim analysis of LIGHT after 25% of expected events accrued to rule out a doubling of cardiovascular risk. These results were to remain strictly confidential, with the unblinded data available only to the sponsor’s employees managing regulatory submissions. As part of the FDA mandate for drug approval, investigators would also have to rule out a 40% increased risk of cardiovascular events at the study completion.
The 25% interim analysis ruled out the doubling of cardiovascular risk, and Contrave was approved in 2014. In fact, the interim analysis showed that treatment with naltrexone/bupropion reduced the risk of major adverse cardiovascular events by 41% compared with placebo. This led Orexigen, the original sponsor who partnered with Takeda to distribute the drug, to publicly file a patent and securities application with the Securities and Exchange Commission claiming a positive effect on cardiovascular outcomes unrelated to weight change. In addition, the company publicly disclosed the results of the interim results widely within the company and to additional business partners, according to the FDA.
To TCTMD, Nissen said the steering committee did not want to stop the LIGHT trial, but given the release of data from the 25% interim analysis they simply had no choice. With the data in public domain, patients could leave the trial in order to take the now commercially available weight-loss drug, and physicians would be reluctant to randomize patients in light of media reports suggesting the drug was associated with a 41% reduction in the risk of major adverse cardiovascular events.
“The Data and Safety Monitoring Board and Executive Steering Committee were very upset with the behaviour of the original sponsor,” said Nissen, noting that he and others pleaded with Orexigen not to publicly disclose the results as this would compromise the still-ongoing study.
In 2015, the LIGHT investigators performed an interim analysis after 50% of accrued events; results of this prespecified analysis are the data being published this week. The suggestion of benefit at the 25% interim analysis, which Nissen said could be nothing more than statistical noise from highly unstable data, disappeared. At the 50% time point, the primary prespecified outcome measure—time to first MACE—occurred in 192 patients, including 102 patients in the placebo arm and 90 patients treated with the naltrexone/bupropion combination. The final end-of-study analysis, at 64% of completion, also showed no difference in MACE.
Given the low rates of adherence in both arms—just 37.5% of patients were taking the drug at 1 year versus 26.3% of placebo-treated patients—the researchers also conducted an on-treatment analysis at the 50% time point. This on-treatment analysis also failed to show any cardiovascular benefit with Contrave.
Weight loss with Contrave did occur, with patients taking the combination reporting a mean decrease of 3.9 kg compared with a 1.2-kg loss among placebo-treated patients (P < 0.001).
A Shattered Compromise
When embarking on clinical research such as LIGHT, Nissen told TCTMD, study investigators enter into a compact with patients—patients agreed to participate in a randomized, placebo-controlled trial on the assumption that the results would further medical science, while researchers have a “moral responsibility” to publish the results in a timely manner. This compact is not something investigators should take lightly, he said.
Sharfstein echoed those sentiments, blasting the company for violating their part of the deal in LIGHT.
“The use of interim data is a compromise between two goals, one of which is identifying improved treatments as quickly as possible and the other of which is making sure these treatments do not have untoward cardiovascular side effects,” he said. “The disclosure of data inappropriately really shatters the compromise and makes it much more likely that the FDA would have to choose to make sure it receives all the data first.”
Sharfstein, who wrote the editorial with Bruce Psaty, MD (University of Washington, Seattle), said the sponsor’s public statements about a positive cardiovascular effect were “highly misleading” and data from the interim analysis did not support their conclusions. He said that given the uncertainty of safety with naltrexone/bupropion and the additional delay that will now occur before more data is available, the FDA should impose additional restrictions on use. One such control would be to impose a Risk Evaluation and Mitigation Strategy (REMS) to counter “the misleading information put out by the company,” he noted.
In their editorial, Sharfstein and Psaty concede that trial investigators were in a tough spot when the sponsor released the interim results, but they contend LIGHT could have continued. Instead of stopping the study, researchers should have released data from the interim analysis at the 50% time point to counter the positive spin put forward by the company, they suggest. Although “unconventional,” such a move might have restored integrity to the trial.
Heading Off Future Mistakes
To prevent such clinical-trial breakdowns in the future, Sharfstein said one option would be to eliminate the company from the two-step approval process, leaving the trial investigators to communicate directly with the FDA the results from the interim analysis. Companies, he said, need to realize that when they are given an opportunity to bring a drug to market earlier, “it is not an entitlement,” and that if others continue to behave as Orexigen did, it might be necessary to “pull the plug” on the early-approval process.
“There’s a lot discussion in FDA circles about whether companies see the big picture or can only focus on the shortest-term benefit,” said Sharfstein. “This is an example of an individual company doing something that isn’t really productive to the whole industry. It really shows why it’s important for the FDA to have authority to protect scientific research.”
Sharfstein told TCTMD that some companies do take a broader view, especially if they’re in business for the long-term and will likely meet the FDA again down the road, but these companies are in danger of being undercut by those that don’t. Nissen noted that other drugs have come to market following the guidelines outlined by the FDA and this is the first instance where such a breakdown occurred. He notes that the two-step safety plan has worked successfully with the diabetes drugs saxagliptin (Onglyza, Bristol-Myers Squibb/AstraZeneca) and alogliptin (Nesina, Takeda), both of which came to market after interim analyses ruled out cardiovascular harm.
With the analysis of LIGHT to date—64% of events have accrued in the event-driven trial—Nissen said they can effectively rule out a doubling of cardiovascular risk as mandated by the premarket requirements of the FDA. Not known, however, is whether or not they can also rule out a 40% increased risk because the study was terminated early. This has forced the start of a second trial testing the cardiovascular safety of Contrave, one also being conducted by Nissen and colleagues at the Cleveland Clinic. Known as the CONVENE study, its “full results won’t be known for another 4 or 5 years,” he said.
1. Nissen SE, Wolski KE, Prcela L, et al. Effect of naltrexone-bupropion on major adverse cardiovascular events in overweight and obese patients with cardiovascular risk factors. JAMA. 2016;315:990-1004.
2. Sharfstein JM, Psaty BM. Evaluation of cardiovascular risk of naltrexone-bupropion. JAMA. 2016;315:984-986.
- Former FDA Staffer Says Missing Data Across Antithrombotic Trials Cast Doubt on Drug Effects
- Low US Participation in NIH-Sponsored CVD Trials Cause for Concern
- LIGHT was sponsored by Orexigen Therapeutics and Takeda Pharmaceuticals.
- Nissen reports receiving grants from the Medicines Company, Amgen, Pfizer, AstraZeneca, Esperion Therapeutics, Eli Lilly, and Cerenis. He reports consulting with many pharmaceutical companies but requires any honoraria/fees be paid to charity, for which he receives neither income nor a tax deduction.
- Sharfstein reports serving as principal deputy commissioner of the FDA from 2009 to 2011.
- Psaty reports serving on the data monitoring committee of a clinical trial funded by Zoll, on the steering committee of the Yale Open Data Access Project funded by Johnson & Johnson, and on the FDA Science Board.