Rates of Memantine Discontinuation Underestimated in Dementia Studies?

Daniel M. Keller, PhD

March 06, 2014

MUNICH — Rates of discontinuation and dose reduction of memantine (Namenda, Forest Laboratories, Inc) due to adverse events in patients with several dementia types may be underestimated in clinical trials, new research suggests.

An analysis of clinical records suggests that patients with Alzheimer's disease (AD) and with mixed type of dementia tend to tolerate memantine better than patients with other forms of dementia. However, side effects leading to memantine discontinuation are common among all groups of treated patients.

Speaking here at the 22nd European Congress of Psychiatry (EPA), Tomasz Sobow, MD, PhD, professor in the Department of Medical Psychology at the Medical University of Lodz, Poland, pointed out that as a group, randomized controlled trials examining the safety of memantine suffered from several limitations, including patient selection bias, reporting bias from companies not publishing negative results, and a lack of data on the relationship of memantine to adverse effects or drug discontinuation. Finally, rigidity of study protocols may not have allowed appropriate dose reductions.

Nevertheless, Dr. Sobow cited literature showing that discontinuation of the drug due to adverse events was never higher than 12% in AD trials, and sometimes adverse events were more prevalent in the placebo group than in those patients receiving active drug.

Agitation More Frequent With Placebo

In trials involving vascular dementia (VaD), adverse event–related discontinuation was 13% with memantine vs 10% with placebo. Combined data for Parkinson's disease dementia/dementia with Lewy bodies (PDD/DLB) showed a high rate of discontinuation for drug and placebo (20% vs 23%, respectively) in 1 trial and 11% vs 10%, respectively, in another.

No single symptom was associated with discontinuation across the board. In trials of moderate to severe AD, agitation was an important factor but occurred more frequently with placebo (3.5%) than with memantine (1.7%).

In VaD, trials showed more discontinuations in the placebo group (6.7%) than with memantine (2.7%). Dr. Sobow said no distinct symptom profile was detected in PDD/DLB trials, but "patients with DLB tended to experience more side effects [15%] than those with PDD [9%]."

From clinical records, Dr. Sobow gathered data on memantine discontinuations and dose adjustments resulting from adverse events, the mean drug dose during stable dosing periods, and the type of adverse events commonly associated with discontinuations or dose reductions.

For each form of dementia (AD, mixed, VaD, and PDD/DLB), mean ages in the study samples ranged from 71.5 to 76.6 years, with an aggregate mean age of 74.7 ± 5.9 years. Mean Mini–Mental State Examination scores ranged from 14.4 to 17.1, with a mean score of 15.9 ± 4.6 for the groups together.

Patients with dementia not otherwise specified or those lost to follow-up were excluded from the analysis.

Table. Memantine Discontinuations and Dose Reductions Because of Adverse Effects

Form of Dementia Discontinuations Dose Reductions
AD 28/161 (17.4%) 33/133 (24.8%)
MxD 18/71 (25.4%) 20/53 (37.7%)
VaD 15/52 (28.8%) 17/37 (45.9%)
PDD/DLB 10/28 (32.1%) 8/18 (44.4%)
Whole sample 75/312 (24%) 85/241 (35.2%)

AD, Alzheimer disease; MxD, mixed dementia; VaD, vascular dementia; PDD/DLB, Parkinson's disease/dementia with Lewy bodies

Comparing AD with the other diagnostic groups, the rates of discontinuations from adverse effects were significantly different only between AD and PDD/DLB (P =.02), with a trend toward fewer discontinuations for AD vs VaD (P = .07).

AD groups had significantly fewer dose reductions vs the VaD groups (P = .01), and there was a trend toward fewer dose reductions when compared with the mixed dementia and PDD/DLB groups (P =.08 for each).

The mean stable doses achieved were highest for AD (14.6 mg) and mixed AD patients (12.8 mg), intermediate for VaD (10.7 mg), and lowest for PDD/DLB patients (8.8 mg). However, for a substantial proportion of VaD and PDD/DLB patients, physicians settled on lower doses and did not try escalating them, compromising any conclusions about tolerability based on the observed data.

"For the whole sample, symptoms most commonly associated with discontinuation were sedation, dizziness, and agitation/sleeplessness, which is quite similar to randomized controlled trials," said Dr. Sobow.

"What we found peculiar was in the PDD/DLB group, a major safety issue were falls and also postural instability with or without dizziness.... Also, sleep disorders were common in patients with PDD/DLB." Headache and abdominal pain were related to discontinuations in VaD and, to a lesser degree, in mixed dementia, he said.

Dr. Sobow noted that randomized clinical trials may have underestimated the rates of memantine discontinuation due to adverse events and that symptoms leading to discontinuation are common for all treated patients.

In addition, he said that patients with AD or mixed type dementia tended to tolerate the drug better than VaD patients and that patients with PDD/DLB had the least tolerance for it. He warned that postural instability and falls among PDD/DLB patients present a reason for special concern.

Balancing Act

Session chair Aye Mu Myint, MD, PhD, Department of Psychiatry at Ludwig Maximilian University in Munich, Germany, who was not involved in the research, told Medscape Medical News that memantine's potential adverse effects have to be balanced against efficacy and that she would have liked to have seen efficacy data.

Dr. Myint added that it would be worth examining the positive effects of the drug and investigating whether there are any potential biomarkers that could predict adverse events, paving the way for potential interventions, including dose adjustment.

Asked whether she had any biomarkers in mind, she noted that memantine is an antagonist against N-methyl-D-aspartate (NMDA) receptors, which are glutamate receptors.

Therefore, glutamate/glutamine ratio imaging could be one important marker. But if imaging is too expensive, she suggested using blood markers such as quinolinic acid, a tryptophan metabolite and an NMDA receptor agonist.

If quinolinic acid levels are high, patients may have increased psychomotor agitation, she said. Using this biomarker may provide a basis for dose adjustment, she suggested.

"So I think biomarker studies should be done there." Any biomarker determinations should be done at baseline, before treatment initiation, because "baseline is always very important for prediction," she said.

Dr. Sobow reports that he has served as a speaker, expert, or reviewer for several pharmaceutical companies, including Pfizer, Novartis, Lundbeck, Merz, Eli Lilly, Servier, sanofi-aventis, Bristol Myers-Squibb, Glaxo, AstraZeneca, Janssen, and several Polish companies offering generic drugs. He stated that no company, particularly any one manufacturing or distributing memantine, took any part in the preparation of his presentation. Dr. Myint reports no relevant financial relationships.

22nd European Congress of Psychiatry (EPA). Session FC08. Presented March 4, 2014.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....