Takeda Rozerem Clinical Significance Debated In FDA Review

The "curious" failure of subjective secondary endpoints to agree with the efficacy of Takeda's Rozerem on objective sleep measures contributed to a debate at FDA over the clinical significance of the insomnia therapy.

FDA approved Rozerem (ramelteon) July 22, one day before the user fee goal. Despite the first-cycle approval, however, review documents show disagreement within the Division of Anesthesia, Analgesia & Rheumatology Products, with Deputy Division Director Rigoberto Roca, MD, dissenting from the approval recommendations of the primary clinical reviewer and division director.

"Ramelteon has a statistically significant treatment effect that is of marginal clinical significance," Roca stated in a June 30, 2005 memo that recommended "approvable" status instead of approval.

Roca pointed to the eight minute difference in latent time to persistent sleep as measured by polysomnography, the primary endpoint, in a transient insomnia trial and approximately 16 minute differences in three chronic insomnia studies.

Furthermore, Rozerem did not show consistent statistically significant effects on secondary subjective measures including subjective total sleep time and sleep quality, he observed.

"The patients who are currently being targeted by the proposed indication do not seem to recognize any benefit from treatment with ramelteon," Roca said.

Division Director Bob Rappaport, MD, disagreed with Roca on the evidence of clinical significance. "While the mean differences in latency to sleep onset were small, this is not unusual for analyses that compare the means of different treatment groups," Rappaport wrote in a July 18 memo.

"Indeed, review of the raw data demonstrates a wide range of outcomes, many of indisputable clinical relevance," Rappaport added.

While acknowledging that Rozerem "curiously…did not show much subjective improvement even when objective measures by polysomnography did," Office of Drug Evaluation II Director Robert Meyer, MD, reiterated the approval recommendations of Rappaport and medical reviewer D. Elizabeth McNeil, MD.

"Ultimately, since this is a medication taken for subjective reasons, patients will need to decide if they perceive a benefit," Meyer stated.

"This product has an immediate hypnotic effect," McNeil concluded in her June 29, 2005 review.

Too Subtle To Appreciate?

McNeil addressed the discrepancy between the objective and subjective endpoints, observing that "insomnia is an interesting disorder as it is one of the few conditions with objective and subjective means of measuring the same endpoint."

"The case could be made that from a clinical standpoint, the subjective measures are perhaps more important," she continued.

However, she concluded, Rozerem "appears to have a subtle mechanism of action that makes it difficult for the end-user to appreciate its beneficial effects."

McNeil quoted extensively from Takeda's explanation that Rozerem's novel melatonin agonist mechanism of action was behind conflicting objective and subjective measurements.

"Subjects with insomnia tend to overestimate sleep latency and underestimate sleep duration relative to PSG measurement," she reports Takeda asserting. "Subjects who are experienced with the use of benzodiazepines, in particular, may anticipate cues such as sedation and equate these sensations with falling asleep."

"Given that the subjective assessment techniques in these studies were originally developed for compounds with GABAergic mechanisms of action, the absence of subjective anxiolytic, sedative, and muscle-relaxant effects prior to sleep onset may make the sleep promoting effects of ramelteon more difficult to detect subjectively," Takeda continued.

Roca found Takeda's argument less persuasive. "Even if the applicant is correct, the end result is the same in that the patients who are currently being targeted by the proposed indication do not seem to recognize any benefit from treatment with ramelteon," he wrote.

"Ordinarily, a marginally clinically significant treatment effect would not preclude an approval," Roca stated, but the FDA review "would then focus even more on the safety profile."

Postmarketing Assessment vs. New Trials

Given "the potential for an association with hyperprolactinemia, it was my opinion that the applicant had not adequately demonstrated a favorable risk:benefit ratio," Roca stated in a July 22 addendum to his memo - the day of Rozerem approval.

Approval should require additional studies that either defined a population in which Rozerem was clinically significant or clarified endocrine safety, he said.

McNeil also filed an addendum to her review July 22 containing additional adverse event tables, but she declared that her recommendation for approval was unchanged. She noted that the addendum had been discussed with Drs. Roca and Rappaport.

Rappaport and Meyer stressed handling both efficacy and endocrine concerns postmarketing through labeling and monitoring.

"Patients who are dissatisfied with the efficacy of the product will simply discontinue taking the medication," Rappaport said.

"Patients presenting with symptoms or signs suggestive of [hyperprolactinemia] can be tested, and the drug discontinued," he continued.

Rozerem labeling is consistent with a June 17, 2005 consult from Metabolic & Endocrine Division medical officer Mary Parks, MD, who recommended monitoring for prolactin elevation "based only on clinical complaints/presentation."

Rappaport also disagreed with McNeil's recommendation for required postmarketing studies of hyperprolactinemia and neoplasia incidence, although he notes that "it will be important to closely watch for any signals of more significant morbidity in the post-marketing period."

While Rozerem was approved with no formal Phase IV commitments regarding prolactin levels, concerns about endocrine effects led FDA to delay pediatric data requirements.

"Due to the effect of this melatonin-like drug on prolactin levels, we will need to be vigilant in our post-marketing assessments for any signs of important endocrine disruption which might signal more of an issue with this drug than we now perceive," Meyer explained.

"It seems prudent to delay the requirement of pediatric data," he concluded. The approval letter notes that pediatric study submission is delayed until 2012.

Approved Indication Narrower Than Desired

Rozerem's approval for "insomnia characterized by difficulty with sleep onset" is not the broad insomnia indication the trials were originally planned to support, the review documents show.

While the documents are redacted, it is clear that the company was seeking another claim based on a secondary endpoint measuring wake time after sleep onset (WASO), according to McNeil's summary of the July 16, 2002, end-of-Phase II meeting. The primary endpoint, measuring sleep latency by PSG, would support a sleep onset claim, she noted.

The WASO endpoint supports sleep maintenance claims of the type received by recently-approved competitors like Sanofi's Ambien CR and Sepracor's Lunesta .

Ambien CR labeling states that it is indicated for insomnia "characterized by difficulty with sleep onset and/or sleep maintenance (as measured by wake time after sleep onset)." Lunesta's indication notes that the drug "decreased sleep latency and improved sleep maintenance."

Takeda, however, has not "provided any evidence that their product is effective" for the redacted claim, Rappaport stated. "They did not study outcome measures that would…allow for adequate assessment."

Sanofi and Sepracor have both characterized Rozerem as not a direct competitor to their insomnia drugs.

Rozerem's product profile is to "recruit patients from the OTC market into the prescription market," Sepracor President Jim O'Shea told an Oct. 25, 2005 earnings call.

Takeda, which launched Rozerem in late September, is highlighting the drug's novel mechanism of action and non-scheduled status.