Questioning the acceptance of serum sodium changes as a surrogate endpoint for hyponatremia drugs, the FDA's Cardiovascular and Renal Drugs Advisory Committee (CRDAC) wasn't sold on the clinical efficacy of Cornerstone Therapeutics Inc.'s lixivaptan.

The committee voted 3-5 Thursday against recommending approval of the drug for hyponatremia in association with syndrome of inappropriate antidiuretic hormone secretion (SIADH), even though lixivaptan met the surrogate endpoint and raised no safety concerns in the SIADH population.

The eight-member panel voted unanimously against approval of the vasopressin V2 receptor antagonist for hyponatremia associated with congestive heart failure (CHF), citing an "unimpressive" benefit in the BALANCE trial and potential safety concerns.

In both instances, the committee backed away from serum sodium correction as a surrogate endpoint in place of a demonstrated clinical benefit. In the past, CRDAC has accepted the surrogate, and the FDA has used it to approve two other vaptans, Vaprisol (conivaptan, Astellas Pharma US Inc.) and Samsca (tolvaptan, Otsuka Pharmaceutical Co. Ltd.). (See BioWorld Today, Sept. 12, 2012.)

Lixivaptan missed secondary endpoints that would have detected clinical benefit in three Phase III trials. The BALANCE trial, for instance, didn't demonstrate a change in days alive or out of hospital. And none of the trials detected a favorable treatment effect on cognitive function.

In voting against lixivaptan in CHF, the panel generally cited the small benefit over placebo, but a few members also expressed concerns over an imbalance of deaths – nine patients in the treatment arm vs. one on placebo – within the first five days of therapy.

When the sponsor saw the imbalance, it convened an independent panel of cardiologists to review the blinded data. As a result, it found that the treatment arm also had an imbalance of seriously ill patients, which could help explain the difference in deaths.

That didn't matter to Daniel Gillen, an associate statistics professor at the University of California-Irvine. Since the BALANCE trial didn't demonstrate clinical benefit, there's no need to establish harm, he said.

Linda Fried, an associate professor of epidemiology at the Veterans Administration Pittsburgh Healthcare System, said her no vote also was based on the lack of benefit, as she didn't perceive a safety concern. Other panelists suggested further study in the CHF population.

Fried was one of the three committee members supporting approval in SIADH. The two trials in that population raised no safety signals and showed that lixivaptan treated the pathology of the disease, she said.

The other two yes votes were based on a sense of regulatory fair play. The sponsor used the surrogate endpoint because that's what the committee and the FDA have applied in the past, said CRDAC Chairman A. Michael Lincoff, director of the Cleveland Clinic Coordinating Center for Clinical Research. In doing so, the sponsor was following what appeared to be the ground rules for hyponatremia trials, he added.

Scott Emerson, a biostatistics professor at the University of Washington, agreed. While he doesn't think serum sodium is an appropriate surrogate, he said it is the recognized endpoint. Since there was no marked differences between lixivaptan and the other two drugs that have been approved on the basis of that surrogate, Emerson voted for approval in SIADH.

In its briefing document for the meeting, the FDA didn't question the surrogate. In fact, its reviewer recommended approval of lixivaptan for the SIADH indication – based on it meeting the surrogate – provided the therapy is initiated in an inpatient setting where volume status, fluid balance and serum sodium concentration could be closely monitored.

Since the two approved vaptans must be administered in an inpatient setting, they're not approved for chronic hyponatremia. Citing the unmet need outside the hospital setting, Cornerstone was seeking outpatient initiation of lixivaptan. Such approval would make lixivaptan more cost-effective, said Cornerstone consultant Joseph Verbalis, director of Georgetown-Howard Universities Center for Clinical and Translational Sciences.

Fried agreed, admitting that she doesn't use some of the other hyponatremia drugs because of their cost, especially in asymptomatic patients who don't necessarily need to be hospitalized. However, she questioned the feasibility of the required monitoring in the outpatient setting.

Following the meeting, Cornerstone CEO Craig Collard said, "We view today as another step in the process of bringing lixivaptan to market . . . using the established surrogate endpoint of serum sodium correction."

With an Oct. 29 PDUFA date, the Cary, N.C.-based company, which obtained worldwide rights to lixivaptan in last year's acquisition of Cardiokine Inc., said it will continue to work with the FDA throughout the approval process.

Mixed Support for Pre-1938 Drug

In the afternoon, an expanded CRDAC turned its attention to West-Ward Pharmaceuticals Inc.'s publication-based 505(b)(2) application for phenylephrine hydrochloride injection to increase blood pressure in acute hypotensive states, such as shock and peri-operative hypotension.

While the committee was comfortable with the drug's use in the peri-operative setting, it voted 2-8 against approval for a broader indication that would include shock. Of the 50 studies the Eatontown, N.J., drugmaker cited in the application, only eight involved patients with sepsis or septic shock, so the safety data for that population are limited.

The session was as much a discussion on what evidence is needed to approve pre-1938 grandfathered drugs as it was on the specific merits of phenylephrine, which has been used for nearly 80 years in emergency settings. During the public hearing, Seth Mailhot, a life sciences attorney, urged the FDA to be consistent in requiring prospective trials in addition to published studies for all 505(b)(2) applications.

While panel members recommended a clinical trial for the shock indication, they recognized the impossibility of doing a placebo-controlled study, given the widespread use of phenylephrine. Even finding a comparator for a noninferiority trial would be difficult, said Joseph Tobin, professor of anesthesiology and pediatrics at Medical Center Boulevard.

Emerson noted that there's not a lot of financial motivation for a sponsor to do a large trial for a drug that's been used for years. "Are we fooling ourselves if we say, 'Yes, we're going to see those studies'?" he asked.

When it comes to basing approval on published studies, the FDA has concerns about publication bias, use of unapproved comparators and variable mentions of adverse events. But where phenylephrine is concerned, the agency recognized that it can't just pull the drug from the market. And if it were approved for a narrower indication, it would still be broadly used off label.

Not having phenylephrine available would be a detriment to public health, Norman Stockbridge, director of the FDA's Division of Cardiovascular and Renal Drug Products, told the committee. Each year, the drug is used as a standby for about 34 million adults, he said. As a result, the agency has decades of safety information.