Cognition data for Forest's memantine, or lack thereof, in patients with moderate-to-severe Alzheimer’s dementia will be a central question of the Sept. 24 meeting of FDA’s Peripheral & Central Nervous System Drugs Advisory Committee.
FDA will ask the committee to discuss the fact that one of the two pivotal U.S. trials of memantine did not have a cognition measure as a primary outcome.
The trial (MRZ 9605) did have a secondary cognitive measure, the Severe Impairment Battery (SIB), which showed a "nominally ‘positive’" result, FDA said in briefing documents for the committee, but added that "at least one other cognitive measure, the [Mini-Mental State Examination]…was not even nominally positive." The SIB showed a –4.46 change in function for memantine compared to –9.84 for placebo.
The co-primary outcomes of the 252-patient MRZ 9605 trial were primarily global outcomes. FDA noted that one of the primary outcomes, the Clinician’s Interview-Based Impression of Change with caregiver input (CIBIC-plus), "did not reach the traditional level of statistical significance (p= 064)."
The mean CIBIC-plus score in the trial was 4.48 for memantine compared to 4.73 for placebo (with a score of 1 indicating marked improvement, a score of 4 indicating no change and a score of 7 indicating marked worsening).
"We are very interested to know the committee’s view of the lack of significance on this protocol specified primary outcome measure (we have no reason to believe, for example, that this outcome is any less valid in this population than the [Alzheimer’s Disease Cooperative Study-Activities of Daily Living] ADCS-ADL)," FDA said. The change from baseline on the ADCS-ADL was –3.02 for memantine and –5.08 for placebo (p=.022), indicating a statistically greater decline in activities for placebo patients.
The agency noted that "perhaps of more concern, however, is the finding that patients with MMSE scores of less than 10, when analyzed as a separate group, fail to demonstrate significance on either of the primary global measures."
"We are very interested to know whether or not the Committee believes that this finding raises serious questions about the value of this treatment in severe patients, the one group of patient for whom this treatment is proposed to be uniquely effective," FDA said.
The second U.S. pivotal trial (MEM-MD-02) with 403 patients found a –2 change from baseline in the co-primary endpoint of ADCS-ADL for memantine compared to a –3.4 point change for placebo (p=.028). For the other co-primary endpoint of SIB, patients on memantine had a .9 point increase compared to a –2.5 change (p=.001) for placebo patients. FDA does not appear to have any serious questions about this trial.
FDA did have questions about a third pivotal trial (MRZ 9403), which was conducted in 166 patients in Latvia. "The various scales used were relatively non-standard, [and] crucially, the cognitive scale was created after the data had been analyzed," FDA said.
The committee will also be asked about the more general question of whether it is essential to have a cognitive scale as a primary endpoint in clinical trials for severe Alzheimer’s dementia.
"We have maintained that this same approach is appropriate for drugs to treat patients with severe DAT as well, although some have argued that, given the patients’ condition, a formal assessment of cognitive functioning (or less often, global functioning) is not necessarily appropriate," FDA noted.
"We are interested in the committee’s views on this fundamental issue of trial design." The agency will also ask the committee about the appropriateness of using the SIB and ADCS-ADL scales in moderate-to-severe Alzheimer’s dementia trials.
On Sept. 25, the committee will discuss Cephalon's sNDA for Provigil in improving wakefulness in patients with sleep disorders.
Source: FDA Advisory Committee.