Incidence of New-Onset Seizures in Mild to Moderate Alzheimer Disease FREE

Seizures are more common in patients with Alzheimer disease (AD) than in the general elderly population.^1- 6 For example, a population-based case-control study in Rochester, Minnesota, from 1955 to 1984 reported that a history of dementia was a risk factor for unprovoked seizures (odds ratio, 6.2; 95% CI, 2.2-17.0).³ A population-based cohort study from Bronx, New York, suggested that dementia increased the incidence of epilepsy (hazard ratio [HR], 1.96; 95% CI, 0.70-5.48).⁴ The incidence of unprovoked seizures in 453 consecutive patients with mild AD followed up at Columbia University, Johns Hopkins University, and Massachusetts General Hospital was 418 per 100 000 person-years, representing an 8-fold increased incidence relative to an age-stratified population from Rochester, Minnesota.⁶ Seizures have also been reported as adverse events in AD clinical trials. Establishing the background rate of seizures in AD clinical trial populations can provide a context for emergent safety signals arising in ongoing and future AD clinical trials. We pooled patient-level data from 10 Alzheimer Disease Cooperative Study clinical trials from 1995 to 2010 to obtain an estimate of the background rate and risk factors for new-onset seizures during the course of patient follow-up.

The study population consisted of all subjects randomized in 10 Alzheimer Disease Cooperative Study clinical trials, beginning in October 1995 with 6 to 24 months duration in mild to moderate AD: Simvastatin (clinicaltrials.gov identifier: NCT00053599), homocysteine lowering (folate, vitamin B₆, and vitamin B₁₂) (clinicaltrials.gov identifier: NCT00056225),⁷ rofecoxib/naproxen sodium (clinicaltrials.gov identifier: NCT00004845),⁸ valproate sodium neuroprotection (clinicaltrials.gov identifier: NCT00071721),⁹ docosahexanoic acid (clinicaltrials.gov identifier: NCT00440050),¹⁰ Receptor for Advanced Glycation End (RAGE) products inhibitor (PF-04494700) (clinicaltrials.gov identifier: NCT00566397), intravenous immune globulin (clinicaltrials.gov identifier: NCT00818662), estrogen (clinicaltrials.gov identifier: NCT00000177), prednisone (clinicaltrials.gov identifier: NCT00000178),¹¹ and huperzine A (clinicaltrials.gov identifier: NCT00083590). The subjects were recruited from US Alzheimer Disease Research Centers, academic centers, and clinical trial centers. Inclusion criteria for the clinical trials were similar and included: (1) diagnosis of probable AD according to the definition by the National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer Disease and Related Disorders Association;¹² (2) people aged older than 50 years; (3) available caregiver; (4) informed consent by patient or representative; (5) mild to moderate AD with Mini-Mental State Examination (MMSE) scores ranging from 10 to 28;¹³ and (6) stable use of cholinesterase inhibitors and/or memantine. Exclusion criteria included history of seizures. All clinical trial sites obtained approval from their local institutional review boards. Written informed consent was obtained from study participants, legally authorized representatives, or both, according to local guidelines.

The outcome measure was the rate of incident seizures by person-time of follow up from randomization within the Alzheimer Disease Cooperative Study programs. Seizures were identified by search of the adverse event verbatim terms for the following text strings: seizure, epil, epilepsy, epileptic, status epilepticus, epileptiform, epilepsia, status, tonic, clonic, convuls, convulsion, convulsive, grand mal, and fit. This was followed by clinical review of the full verbatim text. Safety assessments were performed at each study visit every 1 to 6 months. Because this analysis involved reuse of existing clinical trial data sets, additional clinical information was not available for most subjects to formally adjudicate the seizure adverse events or to evaluate nonspecific adverse events such as fall or confusion.

Person-time was calculated from time of randomization until whichever of the following happened first: (1) end of study; (2) withdrawal from study; (3) death; or (4) onset of adverse events of seizure or convulsion. Exact 95% confidence intervals were calculated for the incidence rates. The valproate neuroprotection, intravenous immune globulin, and RAGE inhibitor clinical trials were blinded at the time of analysis. The intravenous immune globulin and RAGE inhibitor clinical trials were ongoing at the time of analysis.

Univariable Cox proportional hazards modeling was performed to determine whether incidence rates varied by age; sex; age at onset of dementia; duration of dementia; MMSE score; use of acetylcholinesterase inhibitors (AchE-I), memantine, antidepressants, anxiolytics, and antipsychotics; treatment arm (placebo, active treatment, and ongoing blinded); and presence of the APOE ε4 allele. Univariable significant terms at the P < .05 level were incorporated in a multivariable Cox proportional hazards model. A fully adjusted multivariable model was also performed as a sensitivity analysis (excluding age at onset because it was a linear combination of age and duration of dementia).

The 10 clinical trials randomized 3078 subjects with a total of 3719 person-years of follow-up (Table 1). Consistent with the inclusion criteria, patients were at a mild to moderate stage of AD at randomization (mean [SD] MMSE score, 20.2 [3.9]; mean [SD] duration of dementia, 4.3 [2.6] years). Seventy-two percent were taking AchE-I overall. Of the 2238 individuals tested, 64% carried the APOE ε4 allele.

Eighteen seizures were identified from the verbatim adverse event listings in the study: Cases identified were reported as seizure (n = 13), possible seizure (n = 2), probable seizure (n = 2), or prodrome preseizure symptoms (n = 1, included as a possible aura [partial seizure]). Eight of the 18 seizure adverse events were reported as serious adverse events. Four subjects discontinued study medication within 1 week of the seizure adverse event.

The incidence rate during 3719 person-years of follow-up was 484 per 100 000 person-years (95% CI, 287-764). Seizure rates within selected patient subgroups are shown in Table 1, although precision of subgroup estimates is limited by the low numbers of events. Subjects with greater cognitive impairment at baseline had a higher incidence rate: Subjects with an MMSE score less than 18 had an incidence of 1110 per 100 000 person-years, while those with an MMSE score of 18 or greater had an incidence rate of 257 per 100 000 person-years. Seventy-two percent of all subjects were taking AchE-I, and 89% of subjects with seizures were taking AchE-I at baseline; the exact confidence intervals for the seizure rate in subjects taking and not taking AchE-I broadly overlapped. The valproate study had the most reported seizures, likely owing to an overrepresentation of subjects with moderate AD as this study had a lower MMSE inclusion range (MMSE score range, 12-20) than the other studies.

By univariable Cox proportional hazards regression, significant univariable predictors of seizure were younger age (HR, 0.80; 95% CI, 0.70-0.91 per every 5 years of age), younger age at dementia onset (HR, 0.79; 95% CI, 0.70-0.90 per every 5 years of age), lower MMSE score (HR, 3.9; 95% CI, 1.5-10.1 for MMSE scores <18 compared with MMSE scores ≥18), memantine use (HR, 3.87; 95% CI, 1.12-13.49), and antipsychotic use (HR, 4.0; 95% CI, 1.5-10.3) (Table 2). Duration of dementia, APOE ε4 allele, treatment arm (omnibus test for placebo, active, or blinded), and cholinesterase inhibitor use were not significant univariable predictors of seizure. In a multivariable model including the factors age, MMSE score (continuous or <18 compared with an MMSE score ≥ 18), antipsychotic drug use, and memantine use, all factors except memantine use remained significant independent predictors of seizure occurrence, with attenuation of the effect estimates for MMSE score, antipsychotic use, and memantine use relative to the univariable analysis (Table 2). Age and antipsychotic use remained significant in the fully adjusted model; MMSE score, though not significant (P = .07), had an HR of 3.15 (95% CI, 0.93-10.75), with wider confidence intervals relative to the other models, suggesting collinearity with variables in the overparameterized model (Table 2).

The overall incidence rate of 484 per 100 000 person-years in this clinical trial population is slightly higher than the rate of validated epileptic events in a recently published observational cohort of subjects with mild AD at baseline (418 per 100 000).⁶ Similar to the latter study, younger age was associated with greater incidence of seizures. Dementia severity at baseline was strongly associated with increased seizure risk, which has also been reported in some studies.⁵

Among the medication treatments, antipsychotic drug use (in univariable and multivariable analyses) and memantine use (in univariable analysis) were associated with an increased incidence of seizures. Memantine is indicated for moderate to severe but not mild dementia. Agitation, hallucinations, and delusions that may require treatment are more common in moderate to severe dementia than mild dementia.¹⁴ Seven of the 18 individuals with seizures took at least 1 antipsychotic drug at baseline, ranging from aripiprazole, chlorpromazine, perphenazine, olanzapine, risperidone, quetiapine fumarate, ziprasidone hydrochloride, and haloperidol. While antipsychotic drugs are reported to reduce the seizure threshold, there is likely residual confounding by indication because use of these drugs or memantine may be markers of more advanced disease.¹⁵

Incidence rates of seizures in mild to moderate AD are higher than incidence rates reported in other elderly populations. Incidence rates of first unprovoked seizure in a Northern Manhattan study ranged from 45 per 100 000 person-years for individuals aged 65 to 74 years to 144.7 per 100 000 person-years for those aged 75 to 84 years.¹⁶ Higher rates were observed in Rochester, Minnesota, from 1975 to 1984, approximately 120 per 100 000 person-years for those aged 65 to 74 years and 280 per 100 000 person-years for those aged 75 and older.¹⁷ Although rates in AD are higher than the general population, our results and those of Scarmeas et al⁶ suggest that seizures are uncommon events in mild to moderate AD.

The strengths of the study include the consistent diagnostic criteria (National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer Disease and Related Disorders Association) used to classify AD in centers experienced with the disease and the standardized clinical assessment of adverse events at least every 6 months. Pooling studies enabled incidence calculations of seizure events that were rare in individual trials. However, the small number of events limited detailed evaluation of risk factors and effect modifiers of the seizure incidence rate. The study assumes that the clinical trial populations were similar. While patients were recruited from similar study sites, the individual clinical trial participants may differ by severity or concomitant medications based on differences in inclusion criteria. The population of subjects with AD recruited into clinical trials may not be a random sample of all subjects with mild to moderate AD in the community. Subjects generally must be free of other neurologic, psychiatric, or systemic diseases that could significantly impair clinical trial participation and may thus be healthier than the general AD population. Nonetheless, the incidence rates of seizures observed in the pooled Alzheimer Disease Cooperative Study clinical trials provide context for seizure safety signals that may arise in future clinical trials in mild to moderate AD.

Seizure diagnoses were based on adverse event reporting by clinical study sites. While this took advantage of the careful collection of adverse event data by study physicians during safety assessments at each visit (every 1-6 months), confirmation and clinical adjudication of seizure diagnoses were not performed. Clinically evident tonic-clonic seizures are more likely to be recognized than nonconvulsive partial seizures. Other alterations of consciousness such as syncope or delirium and motor manifestations such as tremor and myoclonus may be misdiagnosed as seizures. The incidence rates in this meta-analysis reflect diagnosed seizures; other alterations of consciousness that were not diagnosed as seizures by the clinical trial physicians are not included in the estimates.

The seizure rates were pooled across treatment and placebo arms. The seizure rate may be modified by the specific treatment in the clinical trial. For instance, divalproex sodium treatment arms may have had a reduced seizure rate since divalproex is an anticonvulsant. Subjects in the valproate neuroprotection study overall had a higher seizure rate, likely attributable to the greater dementia severity at baseline.

Pooling data across AD clinical trials can provide valuable information regarding rates and risk factors for uncommon medical events. The results suggest that younger age and greater cognitive impairment at baseline are risk factors for seizures in mild to moderate AD. Antipsychotic drug use (in univariable and multivariable analyses) and memantine use (in univariable analysis) were associated with increased risk of seizures, although use of these may be a marker for more advanced disease. Current AD data aggregation initiatives such as the Coalition Against Major Diseases are focusing initially on predictors of diagnosis and disease progression.¹⁸ Supplementing these proposals with adverse events data can further clarify the background rates of adverse events and incident comorbidities in AD, facilitate safety monitoring of clinical trials, and provide insight into the overall medical course of AD.