- Email: [email protected]
Safety and efficacy of idalopirdine, a 5-HT6 receptor antagonist, in patients with moderate Alzheimer's disease (LADDER): a randomised, double-blind, placebo-controlled phase 2 trial
Articles
Safety and efficacy of idalopirdine, a 5-HT6 receptor antagonist, in patients with moderate Alzheimer’s disease (LADDER): a randomised, double-blind, placebo-controlled phase 2 trial David Wilkinson, Kristian Windfeld, Eskild Colding-Jørgensen
Summary Lancet Neurol 2014; 13: 1092–99 Published Online October 6, 2014 http://dx.doi.org/10.1016/ S1474-4422(14)70198-X See Comment page 1063 Memory Assessment and Research Centre, Moorgreen Hospital, Southampton, UK (D Wilkinson FRCPsych) and H Lundbeck A/S, Valby, Denmark (K Windfeld PhD, E Colding-Jørgensen MD) Correspondence to: Dr David Wilkinson, Memory Assessment and Research Centre, Moorgreen Hospital, Southampton, SO30 3JB, UK [email protected]
Background In human beings, 5-HT6 receptors are almost exclusively expressed in the brain, particularly in areas relevant for cognition, such as the hippocampus and frontal cortex. We assessed the effect on cognitive performance of Lu AE58054 (idalopirdine), a selective 5-HT6 receptor antagonist, in donepezil-treated patients with moderate Alzheimer’s disease. Methods For this randomised, double-blind, placebo-controlled phase 2 trial (LADDER), we recruited patients from 48 outpatient clinical sites in seven countries. Patients were 50 years or older, had moderate Alzheimer’s disease (a mini-mental state examination score of 12–19), and had been stably treated with donepezil 10 mg per day for 3 or more months. Using a computer-generated sequence, we randomly assigned patients (1:1, stratified by site) to receive either idalopirdine 90 mg per day (30 mg thrice daily) or placebo. The primary endpoint was change from baseline in the 11-item Alzheimer’s Disease Assessment Scale–cognitive subscale (ADAS-cog) at week 24. We analysed all efficacy endpoints in the full-analysis set (modified intention-to-treat analysis). This trial is registered with ClinicalTrials.gov, number NCT01019421. Findings Between Dec 8, 2009, and Dec 23, 2011, we randomly allocated 278 patients to treatment: 133 to placebo and 145 to idalopirdine. 132 patients in the placebo group and 140 in the experimental group were included in the final analysis. At week 24, the change from baseline in ADAS-cog total score was +1·38 (SD 0·53) in the placebo group and –0·77 (0·55) in the idalopirdine group (treatment difference of –2·16 points, 95% CI –3·62 to –0·69; p=0·0040). 25 patients (seven taking placebo and 18 taking idalopirdine) discontinued treatment because of adverse events, the difference between groups being mainly due to asymptomatic transient increases in transaminase concentrations in some idalopirdine-treated patients. The most common adverse events (occurring in >3% of patients) were increased γ-glutamyltransferase (14 [10%] patients in the idalopirdine group vs two [2%] in the placebo group), diarrhoea (six [4%] vs nine [7%]), urinary tract infection (three [2%] vs nine [7%]), fall (three [2%] vs eight [6%]), increased alanine aminotransferase (nine [6%] vs none), and benign prostatic hyperplasia (two [5%] vs none). Serious adverse events were reported by 14 (10%) patients in the idalopirdine group and 13 (10%) patients in the placebo group. One death occurred in each treatment group, neither were regarded as being related to treatment. Interpretation Idalopirdine improved cognitive function in donepezil-treated patients with moderate Alzheimer’s disease. Larger studies in a broader population of patients are ongoing to substantiate the effects reported here. Funding H Lundbeck A/S.
Introduction Serotonin (5-HT) receptor antagonism is a promising mechanism for treating cognitive dysfunction. Lu AE58054 (idalopirdine) is a selective 5-HT6 antagonist being investigated as a symptomatic treatment for Alzheimer’s disease. 5-HT6 receptors are almost exclusively expressed in the CNS in frontal and entorhinal cortices, dorsal hippocampus, nucleus accumbens, and striatum, and might be involved in learning and memory,1 potentially through the modulation of cholinergic, monoaminergic, and glutamatergic neurotransmitter systems.2 5-HT6 receptor antagonists have been shown to improve performance in animal models of cognitive function.3 1092
Findings from non-clinical experiments have shown that idalopirdine has a high affinity for the human 5-HT6 receptor, with little affinity for non-targeted pharmacological receptors.4 Idalopirdine reversed cognitive deficits in young and aged rats, mainly in test models of working memory, which suggested that idalopirdine might be effective in treating cognitive deficits in people.4 Idalopirdine has previously been shown to be safe and well tolerated in doses of up to 360 mg (single dose) in clinical pharmacology studies (Ellen Schmidt, Lundbeck, Valby, Denmark, personal communication). For this first phase 2 study of idalopirdine in Alzheimer’s disease, we included patients with a well www.thelancet.com/neurology Vol 13 November 2014
Articles
characterised Alzheimer’s disease diagnosis who were already being stably treated with donepezil at 10 mg. The rationale for the addition of idalopirdine to donepezil was based on non-clinical pharmacological evidence suggesting additive effects when combining idalopirdine with acetyl cholinesterase inhibitors.5 The idalopirdine dose used in the study (30 mg thrice daily) was based on available non-clinical and human pharmacokinetic data (E Schmidt [Lundbeck], personal communication). Our primary objective was to explore the effect on cognitive performance of idalopirdine in donepeziltreated patients with moderate Alzheimer’s disease. Secondary objectives were to explore the effect on global impression, activities of daily living, behavioural symptoms, and caregiver burden of idalopirdine.
Methods Study design and patients For this double-blind, randomised, fixed-dose, placebocontrolled study (LADDER) we enrolled patients from 48 outpatient clinical sites in seven countries (Australia, Canada, the Czech Republic, Germany, Italy, Poland, and Spain). We enrolled patients between Dec 8, 2009, and Dec 23, 2011. Ambulatory patients, both male and female, aged 50 years or older, with probable Alzheimer’s disease consistent with National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association (NINCDS-ADRDA) criteria,6 with a CT or a MRI done within the previous 6 months, a mini-mental state examination (MMSE) score of 12–19 at screening and baseline, with no significant abnormalities on electrocardiogram (ECG) and physical examination and clinical laboratory tests, and who had been treated daily with donepezil for 4 months or more and stable on 10 mg per day for 3 months or more before screening, were eligible for inclusion in the study. Both the patient and a reliable caregiver were required to be fluent in the language used for the psychometric testing. A full list of inclusion and exclusion criteria is provided in the appendix. We excluded patients if they had taken memantine within 2 months before screening, had received treatment with any investigational drug within 30 days of screening, or with a known hypersensitivity to 5-HT6 receptor antagonists. Patients with a current diagnosis of any psychiatric disorder, clinically significant or unstable systemic disease, exposure to toxins or severe medical procedures were also excluded. In addition, patients had to be withdrawn from the study if they withdrew consent, were lost to follow-up, were 75% or less compliant with study drug between any two visits, if the randomisation code was broken, or if serum aspartate aminotransferase or alanine aminotransferase concentrations exceeded the protocol-specified limits. Compliance with study drug was assessed by returned capsules. www.thelancet.com/neurology Vol 13 November 2014
We did the study in accordance with the principles of Good Clinical Practice7 and the Declaration of Helsinki.8 National central ethics committees gave approval in Czech Republic, Germany, Italy, Poland, and Spain; ethics committees or institutional review boards gave approval at each site in Australia and Canada. Eligible patients or their legal representatives provided written informed consent before participating.
Randomisation and masking We randomly assigned patients (1:1) to double-blind treatment with idalopirdine or placebo according to a randomisation list that was computer generated by H Lundbeck A/S in blocks of four, stratified by site. The person who generated the randomisation sequence was not otherwise involved in the trial. The details of the randomisation series were contained in a set of sealed opaque envelopes. At each centre, sequentially enrolled patients were assigned the lowest randomisation number. The study drug or placebo was given as capsules of identical appearance, taste, and smell. All study personnel and participants were masked to treatment assignment. During the study, the randomisation code was broken by the Qualified Person for Pharmacovigilance for 16 patients because of increased liver function test values. Individual patient treatment allocation was not communicated to any other clinical personnel before formal unmasking of the study for analysis. The raters who assessed patients using the Alzheimer’s Disease Cooperative Study Clinical Global Impression of Change (ADCS-CGIC) were masked to other efficacy and safety assessments.
Procedures The study comprised a 2-week screening period followed by a 24-week double-blind treatment period. Eligible patients were given idalopirdine 90 mg per day (30 mg three times a day) or placebo as adjunctive treatment to donepezil 10 mg per day. Patients were seen at baseline and at weeks 2, 4, 6, 8, 12, 16, and 24, with a safety follow-up 4 weeks after last assessment. After the study start, visits at weeks 2 and 6 were added in protocol amendments that also introduced stricter inclusion and withdrawal criteria and follow-up and repeat testing for liver function tests. Patients who withdrew were seen after the decision to withdraw. The cognitive subscale of the 70-point, 11-item Alzheimer’s Disease Assessment Scale (ADAS-cog)9 was rated at baseline and at weeks 4, 12, and 24. All raters underwent formal training in the scoring conventions for the scales and cognitive tests and only personnel trained as a rater were allowed to rate patients.
See Online for appendix
Outcomes The primary endpoint was the change from baseline in the ADAS-cog at week 24. Pre-specified secondary 1093
Articles
efficacy endpoints were the ADCS-CGIC,10 the Alzheimer’s Disease Cooperative Study Activities of Daily Living, 23-items scale (ADCS-ADL23),11 the neuropsychiatric inventory (NPI),12 the Neuropsychological Test Battery (NTB),13 and the Zarit Burden Interview (ZBI), all assessed at week 24.14 The NPI is a 12-item validated structured interview with a caregiver, designed to assess behavioural disturbances in patients with dementia, and includes ten behavioural and two neurovegetative areas. The NTB, modified from that used by Harrison and colleagues,13 was composed of the following six tests (subitems): Category Naming Test (CNT), Controlled Oral Word Association Test (COWAT), Simple Reaction Time (SRT), Choice Reaction Time (CRT), Go/No-Go Paradigm (Go/NoGo), and ADAS-cog Memory Word Recall. The ZBI is a 22-item, caregiver burden, self-rating questionnaire designed to assess the consequences of caring activities in the physical, psychological, and social areas. We recorded all adverse events and vital signs. Qualified personnel coded adverse events using the lowest level term according to MedDRA (version 14.0). We also assessed clinical safety laboratory tests, weight, body-mass index, ECGs, and physical examination findings.
Statistical analysis Safety analyses were based on the all-patients-treated set, comprising all patients who were randomly allocated to and took at least one dose of study drug. Efficacy analyses were pre-specified in the protocol and were based on the intention-to-treat principle, using the full-analysis set, comprising all patients in the allpatients-treated set who had at least one valid postbaseline assessment of the primary efficacy variable (ADAS-cog). All efficacy analyses were pre-specified in the protocol. We regarded a treatment difference of 2 points on the ADAS-cog total score after 24 weeks as a clinically relevant effect. An SD of 6, a withdrawal rate of 15%, a 5% significance level, and a minimum of 135 patients in each treatment group led to 72% power in analysing change from baseline at week 24 for completers. Simulations of various scenarios suggested that applying mixed model repeated measures (MMRM) would increase the power compared with an analysis ignoring data from dropouts. Although lower than what is often used, the study sponsor regarded this level of power as acceptable for this proof-of-concept study. The primary analysis assessed the treatment difference in efficacy based on the change from baseline in the ADAS-cog at week 24. We analysed the changes from baseline at weeks 4, 12, and 24 using MMRM, and estimated the treatment difference at these timepoints. The model included categorical effects of treatment, pooled site, week, and treatment-by-week interaction, plus the continuous, fixed covariate of baseline score 1094
and baseline score-by-week interaction. We used an unstructured covariance matrix to model the variation of the repeated measures within patients. We did a sensitivity analysis of the change from baseline using ANCOVA, with treatment and pooled site as factors and baseline score as covariate, for weeks 4, 12, and 24 separately using completers and last observation carried forward (LOCF). LOCF imputation of missing values has traditionally been used for efficacy analyses in clinical trials supporting registration and, in this particular study, LOCF can be expected to underestimate rather than overestimate treatment effects. Analyses of completers, however, tend to overestimate treatment effects, so in combination the two analyses were felt to be suited to address the sensitivity of the results of the MMRM model to the unobserved missing values. In the factor termed “pooled site”, small sites (ie, sites with one or two participants) not contributing to both treatment groups were pooled into single pooled sites within each country to include all observations in the analysis. We analysed the ADCS-ADL23, ADCS-CGIC, NPI, and ZBI changes in total scores using MMRM, similar to the primary analysis, as well as an ANCOVA with baseline score as covariate and treatment and pooled site as factors. We calculated standardised Z scores for each of the six individual tests comprising the NTB. We then calculated a Z score for each test by dividing the difference from baseline score at each assessment for each patient by the relevant baseline SD. We calculated a combined NTB composite Z score by taking the mean of the Z scores across the six tests (CNT, COWAT, SRT, CRT, Go/No-Go, ADAS-cog Memory Word Recall) at each assessment for each patient. Histograms of residuals and plots of residuals versus fitted values indicated approximate normality of residuals with no relation between fitted values and residuals. We used SAS (version 9.2) for all statistical analysis. This study is registered with ClinicalTrials.gov, number NCT01019421.
Role of the funding source The study funders generated the randomisation sequence and designed the study, collected and analysed the data, and were involved in data interpretation and in the drafting of this report with the support of a medical writer employed by the study funders. Two of the co-authors (KW, EC-J) are employed directly by the funders. All authors, including those employed by the study funders, had full access to all the data in the study and DW had final responsibility for the decision to submit for publication.
Results Between Dec 8, 2009, and Dec 23, 2011, we screened 430 patients, of whom 278 were randomly allocated to treatment: 133 to placebo and 145 to idalopirdine www.thelancet.com/neurology Vol 13 November 2014
Articles
(figure 1). 118 (89%) patients in the placebo group and 114 (79%) patients in the idalopirdine group completed the study. Baseline characteristics were much the same between the two groups (table 1). At baseline, 117 (81%) of 145 patients in the idalopirdine group and 111 (84%) of 133 patients in the placebo group took concomitant drugs (besides donepezil) that they continued to take during the trial, and 52 (36%) patients in the idalopirdine group and 45 (34%) patients in the placebo group continued concomitant drug treatment during the study. The most frequent primary reasons for withdrawal were adverse events and withdrawal of consent (figure 1). About 80% of the patients in each treatment group received study drugs for at least 140 days in the 24-week treatment period. The exposure to study drug was 57·1 patientyears in both treatment groups. In the pre-specified primary efficacy analysis, we saw an improvement in cognitive function with idalopirdine compared with placebo (table 2). At week 24, the change from baseline in ADAS-cog total score was +1·38 (SE 0·53) with placebo and –0·77 (0·55) with idalopirdine, giving a treatment difference of –2·16 points (95% CI –3·62 to –0·69), which was statistically significant (p=0·0040; table 2). Improvement in cognitive function was significantly better with idalopirdine than with placebo by week 12 (p=0·013; figure 2). Differences versus placebo were shown by ANCOVA for completers (difference versus placebo of –2·40, 95% CI –3·89 to –0·91; p=0·0017 at week 24) and after LOCF imputation for withdrawals (–1·85, –3·18 to –0·51; p=0·0070 at week 24). The number of patients randomised at each site ranged from one to 18. The results were not sensitive to the data from sites recruiting fewer than three patients. None of the effects of idalopirdine on the pre-specified secondary efficacy endpoints was statistically significant at week 24 (table 2; figures 3 and 4). In a post-hoc analysis, we found a higher mean MMSE score at week 24 (17·5 [SD 3·8]) than at baseline (16·7 [2·1]) in the idalopirdine group and a lower mean MMSE score at week 24 (16·4 [4·4]) than at baseline (16·9 [2·1]) in the placebo group. Treatment-emergent adverse events reported by at least 3% of the patients in either treatment group are shown in table 3. The higher incidence of treatmentemergent adverse events in the idalopirdine group relative to the placebo group was mainly due to increased liver enzyme values. During the treatment period, 25 patients withdrew owing to treatmentemergent adverse events, seven in the placebo group and 18 in the idalopirdine group (figure 1). 13 patients (all in the idalopirdine group) had increased (greater than two times the upper limit of normal) aspartate aminotransferase or alanine aminotransferase values, most of which were first recorded at week 6 or week 8. Eight of these patients had concentrations greater than three times the upper limit of normal with no concurrent increase in total bilirubin. For 12 patients, alanine aminotransferase or aspartate www.thelancet.com/neurology Vol 13 November 2014
aminotransferase values subsequently decreased to within the normal range within 1–16 weeks. For the thirteenth patient, alanine aminotransferase and aspartate aminotransferase concentrations had decreased to 1·5 times the upper limit of normal by
430 screened 144 excluded 122 met exclusion criteria 14 inclusion failure 4 consent withdrawn 1 lost to follow-up 3 other 286 baseline 8 excluded 6 met exclusion criteria 2 inclusion failure 278 randomised
133 randomly allocated to placebo plus donepezil
145 randomly allocated to idalopirdine 90 mg plus donepezil
15 withdrawn 7 adverse events* 1 lack of efficacy 1 non-compliance 4 consent withdrawn 2 protocol violation
31 withdrawn 18 adverse events 3 non-compliance 9 consent withdrawn† 1 protocol violation
132 analysed in full-analysis set
140 analysed in full-analysis set
Figure 1: Trial profile *Includes one patient withdrawn from full-analysis set due to absence of post-baseline Alzheimer’s Disease Assessment Scale–cognitive subscale (ADAS-cog) assessment. †Includes five patients withdrawn from full-analysis set due to absence of post-baseline ADAS-cog assessment.
Placebo (n=133)
Idalopirdine (n=145)
Age (years)
75 (7·2)
74 (7·5)
Men
44 (33%)
38 (26%)
White
131 (99%)
143 (99%)
Time since diagnosis (years)
2·2 (1·9)
2·1 (1·8)
Duration of donepezil use (years)
1·5 (1·6)
1·4 (1·6)
MMSE score
17 (2·1)
17 (2·1)
ADAS-cog
28 (11)
28 (10)
ADCS-ADL23
51 (16)
53 (14)
ADCS-CGIC*
4·0 (0·8)
3·9 (0·9)
Neuropsychiatric inventory (total score)
9·3 (12·0)
8·4 (8·9)
Neuropsychological test battery
0·00 (0·48)
0·01 (0·53)
Data are mean (SD) or n (%). *This scale is an assessment of change from baseline; the baseline value recorded was a severity score, in response to the question: “Considering your total clinical experience with this particular population, how ill is the patient at this time?”. MMSE=mini-mental state examination. ADAS-cog=Alzheimer’s Disease Assessment Scale cognitive subscale. ADCS-ADL23=Alzheimer’s Disease Cooperative Study Activities of Daily Living 23-Items scale. ADCS-CGIC=Alzheimer’s Disease Cooperative Study Clinical Global Impression of Change scale.
Table 1: Baseline demographic and clinical characteristics
1095
Articles
Overall baseline mean
Week 24 adjusted mean*
Difference to placebo
Placebo
Mean
Idalopirdine
95% CI
p value
Primary analysis ADAS-cog
28·0 (0·65)
29·4 (0·53)
27·2 (0·54)
51·7 (0·88)
47·9 (0·80)
49·6 (0·82)
–2·16 (0·74)
–3·62 to –0·69
0·0040
Secondary analyses ADCS-ADL23 ADCS-CGIC
3·95 (0·05)
Neuropsychiatric Inventory (total score)
8·88 (0·62)
12·2 (0·85)
Neuropsychological test battery (composite Z score)
0·01 (0·03)
–0·04 (0·04)
Zarit burden interview (total score)
22·9 (0·82)
4·47 (0·10)
26·5 (0·94)
1·72 (1·11)
–0·48 to 3·92
0·12
4·25 (0·10)
–0·22 (0·14)
–0·49 to 0·054
0·12
10·8 (0·86)
–1·45 (1·18)
–3·78 to 0·88
0·22
0·07 (0·06)
–0·05 to 0·18
0·24
–0·29 (1·30)
–2·85 to 2·27
0·82
0·03 (0·04) 26·2 (0·94)
Data are mean (SE ) unless otherwise stated. *Adjusted means are least squares means corresponding to the overall baseline mean scores ADAS-cog=Alzheimer’s Disease Assessment Scale cognitive subscale. ADCS-ADL23=Alzheimer’s Disease Cooperative Study Activities of Daily Living 23 Items scale. ADCS-CGIC=Alzheimer’s Disease Cooperative Study Clinical Global Impression of Change scale.
Table 2: Efficacy analyses
Adjusted mean change from baseline in ADAS-cog
–3 –2
p=0·0130 p=0·0040
–1 0 1 2
Donepezil plus idalopirdine 90 mg Donepezil plus placebo
3 0
4
Placebo (n) 132 Idalopirdine (n) 140
8
12 16 Treatment week
131 140
20
125 127
24 118 114
Adjusted mean change from baseline in ADCS-ADL23
Figure 2: Adjusted mean change in ADAS-cog (full-analysis set, MMRM by visit) Error bars are 95% CI. ADAS-cog=11-item Alzheimer’s Disease Assessment Scale, Cognitive Subscale. MMRM=mixed model, repeated measures.
2
Donepezil plus idalopirdine 90 mg Donepezil plus placebo
1 0 –1 –2 –3 –4 –5 –6 0
4
8
12
16
20
24
Treatment week
Figure 3: Adjusted mean change in ADCS-ADL23 (full-analysis set, MMRM by visit) Error bars are 95% CI. ADCS-ADL23=23-item Alzheimer’s Disease Cooperative Study Activities of Daily Living scale. MMRM=mixed model, repeated measures.
1096
week 24, after which no further testing was done. The median time for concentrations to return to normal was 4 weeks (IQR 2–7). Eleven of the 13 patients with increased alanine aminotransferase or aspartate aminotransferase values withdrew. Treatment-emergent serious adverse events were reported by 27 patients (table 3). Serious adverse events reported by more than one patient were syncope (two patients in the idalopirdine group), femoral neck fracture (two patients in the placebo group), and gastroenteritis (two patients in the placebo group). One patient (in the idalopirdine group) was admitted to hospital with epilepsy and one patient (in the placebo group) was admitted to hospital after a convulsion. In the idalopirdine group, we judged two serious adverse events to be probably related to treatment and six serious adverse events to be possibly related to treatment (all patients recovered, one with sequelae). In the placebo group, we judged five serious adverse events to be possibly related to treatment (four patients recovered, one died). Two patients died during this study. A 79-year-old man who had received idalopirdine for 169 days was hit by a motorcycle and incurred a cranio-cerebral injury and died 22 days later from bronchopneumonia. The second patient, an 80-year-old man who had received placebo for 2 days, fell and incurred a femoral neck fracture (serious adverse events). He developed bronchopneumonia, was admitted to hospital, and died. Neither death was considered by the investigator to be related to the study drug. For the clinical safety laboratory tests except for liver function tests, we saw no clinically relevant mean changes over time or differences between the treatment groups (data not shown).
Discussion Our findings suggest that idalopirdine is effective in improving cognitive function in patients with moderate Alzheimer’s disease who are receiving donepezil, with www.thelancet.com/neurology Vol 13 November 2014
Articles
Adjusted mean change from baseline in ADCS-CGIC
5·0
Donepezil plus idalopirdine 90 mg Donepezil plus placebo
4·5
4·0 p=0·0204 3·5
0 0
4
8
12
16
20
24
Treatment week
Figure 4: Adjusted mean ADCS-CGIC (full-analysis set, MMRM by visit) Error bars are 95% CI. ADCS-CGIC=Alzheimer’s Disease Cooperative Study Clinical Global Impression of Change scale. MMRM=mixed model, repeated measures.
a mean difference in ADAS-cog at week 24 of 2·16 points. The improvements in functional and global clinical measures were not statistically significant at week 24, and the study sponsor is doing confirmatory studies in larger populations to elucidate further the clinical effects. Our treatment efficacy effect estimate was based on MMRM, which provides a reliable estimate of efficacy under a missing at random assumption for the missing values.15 The excess withdrawal of patients from the study on idalopirdine versus placebo was driven by the asymptomatic liver function test increases and automatic withdrawal of those patients according to the protocol, and can thus be regarded as missing at random—ie, not related to the unobserved efficacy outcomes. Sensitivity analyses based on carrying the last observation forward (LOCF) for withdrawals and based on completers lent support to the MMRM findings. The improved cognitive performance in the idalopirdine group, as assessed by use of the ADAS-cog total score, was consistent with changes in the MMSE score. Although the MMSE was not a pre-planned efficacy outcome, data were collected at baseline and at week 24 and analysis showed improvement in the idalopirdine group and worsening in the placebo group. However, analyses of the NTB composite Z score, a preplanned efficacy outcome, showed no statistically significant treatment difference. Through their actions on multiple transmitter systems, 5-HT6 receptor antagonists might also improve behavioural symptoms in Alzheimer’s disease.16 In this study, we saw no effect on the NPI score. However, the average baseline NPI total score of the patients was quite low. In some patients, treatment with idalopirdine was associated with asymptomatic liver enzyme increases, particularly aspartate aminotransferase, alanine aminowww.thelancet.com/neurology Vol 13 November 2014
Placebo (n=133)
Idalopirdine (n=145)
Patients with serious adverse events
13 (10%)
14 (10%)
Patients with treatment-emergent adverse events
78 (59%)
96 (66%)
Increased γ-glutamyltransferase
2 (2%)
14 (10%)
Increased alanine aminotransferase
0
Dizziness
3 (2%)
7 (5%)
Aggression
2 (2%)
6 (4%)
Increased aspartate aminotransferase
0
6 (4%)
Diarrhoea
9 (7%)
6 (4%)
Headache
2 (2%)
6 (4%)
Hypertension
5 (4%)
6 (4%)
Vomiting
1 (1%)
6 (4%)
Increased blood alkaline phosphatase
1 (1%)
5 (3%)
Hypercholesterolaemia
2 (2%)
5 (3%)
Nausea
4 (3%)
5 (3%)
Weight loss
2 (2%)
5 (3%)
Fall
8 (6%)
3 (2%)
Urinary tract infection
9 (7%)
3 (2%)
Agitation
5 (4%)
2 (1%)
9 (6%)
Benign prostatic hyperplasia (men; n=38)
0
2 (5%)
Contusion
4 (3%)
0
Data are number of patients (%).
Table 3: Treatment-emergent adverse events occurring in 3% or more of patients in either treatment group (all patients treated set)
transferase, and γ-glutamyltransferase. For patients with increased concentrations, the values decreased towards the reference range later in the study. The incidence of gastrointestinal adverse events (nausea, vomiting and diarrhoea) was low in both treatment groups. Owing to inhibition of CYP2D6, coadministration of idalopirdine and donepezil resulted in an approximate 10% increase in the idalopirdine group compared with the placebo group at week 4 (no between-group difference at week 24) in the mean plasma concentration of donepezil (data not shown). However, we judge the effect of this small increase to be negligible, in view of the slight reported differences in tolerability and efficacy between 5 mg per day and 10 mg per day donepezil, which corresponds to an approximate 100% increase in plasma concentration.17 We tested the effect of idalopirdine in a selected population of donepezil-treated patients with moderate Alzheimer’s disease. The addition of idalopirdine to acetyl cholinesterase inhibitors is being explored in ongoing studies (NCT02006641, NCT01955161, NCT02006654, NCT02079246). Another 5-HT6 receptor antagonist, GSK SB-742457, has previously been explored and showed no significant benefit on ADAS-cog in three monotherapy studies.18–20 However, in one study21 it was assessed as an adjunct to donepezil and was shown to significantly improve ADAS-cog score (panel).21 Our study was a phase 2 study, so inevitably had limitations. The small size of our study could have been 1097
Articles
Panel: Research in context Systematic review We searched PubMed and clinicaltrials.gov for articles in all languages published up to July 9, 2014, that reported any randomised therapeutic trial with a 5-HT6 receptor antagonist. We used the search terms “a 5-HT6 receptor antagonist” and “clinical trial”, and restricted the search to studies in adult human beings. We applied no language restrictions. We identified four completed controlled phase 2 trials18–21 with the 5-HT6 receptor antagonist SB-742457 and articles describing the results of two of these studies.18,19 Neither showed a statistically significant improvement on the ADAS-cog versus placebo from baseline to week 24. Findings from a third study with SB-742457 as monotherapy,20 also showed no significant difference from placebo. A fourth trial (in patients already receiving donepezil) showed statistically significant improvements in patients receiving donepezil and SB-742457 versus patients receiving donepezil and placebo in the ADAS-cog at weeks 24 and 48.21 We also identified a report of the results from a dose-finding trial with SAM-531 (a 5-HT6 receptor antagonist) in patients with mild to moderate Alzheimer’s disease.22 Findings from this trial showed no superiority compared with placebo on the ADAS-cog at week 24.22 Interpretation In four of the five phase 2 trials identified by our search,18–20,22 a 5-HT6 receptor antagonist was used as monotherapy and failed to show a difference from placebo, but our findings and those of one other trial21 together suggest that the approach of using a 5-HT6 receptor antagonist in patients already receiving an acetylcholinesterase inhibitor should be investigated in further studies.
responsible for the absence of effect on secondary measures other than cognition (larger studies powered for these secondary measures are ongoing). Patients in this study had moderate Alzheimer’s disease, so we cannot extrapolate the results to other groups of patients with milder or more severe disease. The results from this study apply only to patients stably treated with donepezil and cannot be generalised to patients treated with other acetyl cholinesterase inhibitors. Additionally, the use of a single idalopirdine dose means we could not establish whether lower doses would have been efficacious. These data add to the evidence that 5-HT6 receptor antagonism is an interesting approach to the symptomatic treatment of Alzheimer’s disease, and phase 3 studies have been initiated with idalopirdine to explore this hypothesis. Contributors DW had the idea for and designed the study. DW and all the study investigators acquired the data. All authors analysed and interpreted the data and directed content of the report after the initial draft was prepared with the support of a medical writer, who provided a draft
1098
version of Methods and Results, and the authors made changes to the draft. All authors revised the report, approved the final version of the report, and vouch for the integrity of the data and analyses. Study investigators Australia: Roger Clarnette, Michael Woodward, Karyn Boundy, Denis Crimmins, and David Ames. Canada: Sharon Cohen, Angeles Garcia, Guy Lacombe, David Patry, William Nisker, Richard Bergeron, and Miranda Du Preez. Czech Republic: Claudia Borzova, Slavomir Pietrucha, Oldrich Vysata, Ladislav Pazdera, Petranek Svojmil, Martin Brunovsky, Zdenek Solle, Michal Bajacek, Jiri Pivejc, and Jaroslav Lestina. Germany: Matthias Riepe, Gerd Reifschneider, Walter Albrecht, Johannes Pantel, David Prulovic, Hermann-Josef Gertz, and Karl-Otto Sigel. Italy: Alessandro Padovani, Giuseppe Bruno, Guido Rodriguez, Amalia Cecilia Bruni, Angelo Bianchetti, Elio Scarpini, and Sandro Sorbi. Poland: Andrzej Szczudlik, Robert Kucharski, Dorota Ussorowska, Arleta Kuczynska-Zardzewialy, Jacek Staszewski, Alicja Klich-Raczka, Roman Chwedorowicz, Maciej Czarnecki, Andrzej Potemkowski, and Grzegorz Opala. Spain: Mercedes Boada, Jordi Alom, Javier Olzaran, Pedro Gil Gregorio, and Juan Jose López Lozano. Declaration of interests DW has received honoraria for serving on scientific advisory boards and financial compensation for speaking at symposia on behalf of H Lundbeck A/S, Nutricia, and MedNet Consulting outside the submitted work. KW (biostatistician) and ECJ are employees of H Lundbeck A/S. Acknowledgments We thank all patients for their participation in the study and D J Simpson (H Lundbeck A/S) for providing support in the preparation, revision, and editing of the first draft of the paper. References 1 Geldenhuys WJ, Van der Schyf CJ. Role of serotonin in Alzheimer’s disease: a new therapeutic target? CNS Drugs 2011; 25: 765–81. 2 Ramírez MJ. 5-HT6 receptors and Alzheimer’s disease. Alzheimers Res Ther 2013; 5: 15. 3 Upton N, Chuang TT, Hunter AJ, Virley DJ. 5-HT6 receptor antagonists as novel cognitive enhancing agents for Alzheimer’s disease. Neurotherapeutics 2008; 5: 458–69. 4 Arnt J, Bang-Andersen B, Grayson B, et al. Lu AE58054, a 5-HT6 antagonist, reverses cognitive impairment induced by subchronic phencyclidine in a novel object recognition test in rats. Int J Neuropsychopharmacol 2010; 13: 1021–33. 5 Herrik KF, de Jong I, Pedersen JT, et al. The 5-HT6 antagonist Lu AE58054 potentiates the effects of the acetylcholinesterase inhibitor donepezil on hippocampal extracellular acetylcholine efflux and oscillatory activity. Alzheimers Dement 2013; 9 (suppl 4): P503. 6 McKhann G, Drachman D, Folstein M, et al. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 1984; 34: 939–44. 7 ICH Harmonised Tripartite Guideline E6: Guideline for Good Clinical Practice. 1996. http://www.fda.gov/downloads/drugs/ guidancecomplianceregulatoryinformation/guidances/ ucm073122.pdf (accessed July 4, 2014). 8 World Medical Association (WMA). Declaration of Helsinki: Ethical principles for medical research involving human subjects. www.wma.net/en/30publications/10policies/b3/ (accessed July 4, 2014). 9 Rosen WG, Mohs RC, Davis KL. A new rating scale for Alzheimer’s disease. Am J Psychiatry 1984; 141: 1356–64. 10 Schneider LS, Olin JT, Doody RS, et al. Validity and reliability of the Alzheimer’s disease cooperative study- clinical global impression of change. Alzheimer Dis Assoc Disord 1997; 11 (suppl 2): S22–32. 11 Galasko D, Bennett D, Sano M, et al. An inventory to assess activities of daily living for clinical trials in Alzheimer disease. The Alzheimer’s disease Cooperative Study. Alzheimer Dis Assoc Disord 1997; 11 (suppl 2): S33–39.
www.thelancet.com/neurology Vol 13 November 2014
Articles
12
13
14 15
16 17
18
Cummings JL, Mega M, Gray K, Rosenberg-Thompson S, Carusi DA, Gornbein J. The Neuropsychiatric Inventory: comprehensive assessment of psychopathology in dementia. Neurology 1994; 44: 2308–14. Harrison J, Minassian SL, Jenkins L, Black RS, Koller M, Grundman M. A neuropsychological test battery for use in Alzheimer disease clinical trials. Arch Neurol 2007; 64: 1323–29. Zarit SH, Orr NK, Zarit JM. The hidden victims of Alzheimer’s disease. New York: New York University Press, 1985. Siddiqui O, Hung HM, O’Neill R. MMRM vs LOCF: a comprehensive comparison based on simulation study and 25 NDA datasets. J Biopharm Stat 2009; 19: 227–46. Dawson LA. The central role of 5-HT6 receptors in modulating brain neurochemistry. Int Rev Neurobiol 2011; 96: 1–26. Eisai Inc. Prescribing information: Aricept (donepezil hydrochloride) tablets. http://www.aricept.com/docs/pdf/aricept_ PI.pdf (accessed Sept 9, 2014). Maher-Edwards G, Zvartau-Hind M, Hunter AJ, et al. Doubleblind, controlled phase II study of a 5-HT6 receptor antagonist, SB-742457, in Alzheimer’s disease. Curr Alzheimer Res 2010; 7: 374–85.
www.thelancet.com/neurology Vol 13 November 2014
19
20
21
22
Maher-Edwards G, Dixon R, Hunter J, et al. SB-742457 and donepezil in Alzheimer disease: a randomized, placebo-controlled study. Int J Geriatr Psychiatry 2011; 26: 536–44. GSK. Study AZ3110865, a study comparing SB742457 or donepezil versus placebo in subjects with mild-to-moderate Alzheimer’s disease. http://www.gsk-clinicalstudyregister.com/result_detail.js p?protocolId=AZ3110865&studyId=242C4974-9729-4D30-8F91327CF0631014&compound=SB742457 (accessed July 4, 2014). Maher-Edwards G, Ascher J, Watson C, et al. 48-week efficacy and tolerability of treatment with SB-742457, a novel 5HT6 receptor antagonist, when added on to donepezil in subjects with mild-tomoderate Alzheimer’s disease (AD). Alzheimer’s Dementia 2011; 7 (suppl 4): e5. Pfizer. Study comparing 3 dosage levels of SAM-531 in outpatients with mild to moderate Alzheimer disease. http://clinicaltrials. gov/ct2/show/results/NCT00895895?term=sam531&rank=1§=X70156#outcome1 (accessed July 4, 2014).
1099
Safety and efficacy of idalopirdine, a 5-HT6 receptor antagonist, in patients with moderate Alzheimer’s disease (LADDER): a randomised, double-blind, placebo-controlled phase 2 trial David Wilkinson, Kristian Windfeld, Eskild Colding-Jørgensen
Summary Lancet Neurol 2014; 13: 1092–99 Published Online October 6, 2014 http://dx.doi.org/10.1016/ S1474-4422(14)70198-X See Comment page 1063 Memory Assessment and Research Centre, Moorgreen Hospital, Southampton, UK (D Wilkinson FRCPsych) and H Lundbeck A/S, Valby, Denmark (K Windfeld PhD, E Colding-Jørgensen MD) Correspondence to: Dr David Wilkinson, Memory Assessment and Research Centre, Moorgreen Hospital, Southampton, SO30 3JB, UK [email protected]
Background In human beings, 5-HT6 receptors are almost exclusively expressed in the brain, particularly in areas relevant for cognition, such as the hippocampus and frontal cortex. We assessed the effect on cognitive performance of Lu AE58054 (idalopirdine), a selective 5-HT6 receptor antagonist, in donepezil-treated patients with moderate Alzheimer’s disease. Methods For this randomised, double-blind, placebo-controlled phase 2 trial (LADDER), we recruited patients from 48 outpatient clinical sites in seven countries. Patients were 50 years or older, had moderate Alzheimer’s disease (a mini-mental state examination score of 12–19), and had been stably treated with donepezil 10 mg per day for 3 or more months. Using a computer-generated sequence, we randomly assigned patients (1:1, stratified by site) to receive either idalopirdine 90 mg per day (30 mg thrice daily) or placebo. The primary endpoint was change from baseline in the 11-item Alzheimer’s Disease Assessment Scale–cognitive subscale (ADAS-cog) at week 24. We analysed all efficacy endpoints in the full-analysis set (modified intention-to-treat analysis). This trial is registered with ClinicalTrials.gov, number NCT01019421. Findings Between Dec 8, 2009, and Dec 23, 2011, we randomly allocated 278 patients to treatment: 133 to placebo and 145 to idalopirdine. 132 patients in the placebo group and 140 in the experimental group were included in the final analysis. At week 24, the change from baseline in ADAS-cog total score was +1·38 (SD 0·53) in the placebo group and –0·77 (0·55) in the idalopirdine group (treatment difference of –2·16 points, 95% CI –3·62 to –0·69; p=0·0040). 25 patients (seven taking placebo and 18 taking idalopirdine) discontinued treatment because of adverse events, the difference between groups being mainly due to asymptomatic transient increases in transaminase concentrations in some idalopirdine-treated patients. The most common adverse events (occurring in >3% of patients) were increased γ-glutamyltransferase (14 [10%] patients in the idalopirdine group vs two [2%] in the placebo group), diarrhoea (six [4%] vs nine [7%]), urinary tract infection (three [2%] vs nine [7%]), fall (three [2%] vs eight [6%]), increased alanine aminotransferase (nine [6%] vs none), and benign prostatic hyperplasia (two [5%] vs none). Serious adverse events were reported by 14 (10%) patients in the idalopirdine group and 13 (10%) patients in the placebo group. One death occurred in each treatment group, neither were regarded as being related to treatment. Interpretation Idalopirdine improved cognitive function in donepezil-treated patients with moderate Alzheimer’s disease. Larger studies in a broader population of patients are ongoing to substantiate the effects reported here. Funding H Lundbeck A/S.
Introduction Serotonin (5-HT) receptor antagonism is a promising mechanism for treating cognitive dysfunction. Lu AE58054 (idalopirdine) is a selective 5-HT6 antagonist being investigated as a symptomatic treatment for Alzheimer’s disease. 5-HT6 receptors are almost exclusively expressed in the CNS in frontal and entorhinal cortices, dorsal hippocampus, nucleus accumbens, and striatum, and might be involved in learning and memory,1 potentially through the modulation of cholinergic, monoaminergic, and glutamatergic neurotransmitter systems.2 5-HT6 receptor antagonists have been shown to improve performance in animal models of cognitive function.3 1092
Findings from non-clinical experiments have shown that idalopirdine has a high affinity for the human 5-HT6 receptor, with little affinity for non-targeted pharmacological receptors.4 Idalopirdine reversed cognitive deficits in young and aged rats, mainly in test models of working memory, which suggested that idalopirdine might be effective in treating cognitive deficits in people.4 Idalopirdine has previously been shown to be safe and well tolerated in doses of up to 360 mg (single dose) in clinical pharmacology studies (Ellen Schmidt, Lundbeck, Valby, Denmark, personal communication). For this first phase 2 study of idalopirdine in Alzheimer’s disease, we included patients with a well www.thelancet.com/neurology Vol 13 November 2014
Articles
characterised Alzheimer’s disease diagnosis who were already being stably treated with donepezil at 10 mg. The rationale for the addition of idalopirdine to donepezil was based on non-clinical pharmacological evidence suggesting additive effects when combining idalopirdine with acetyl cholinesterase inhibitors.5 The idalopirdine dose used in the study (30 mg thrice daily) was based on available non-clinical and human pharmacokinetic data (E Schmidt [Lundbeck], personal communication). Our primary objective was to explore the effect on cognitive performance of idalopirdine in donepeziltreated patients with moderate Alzheimer’s disease. Secondary objectives were to explore the effect on global impression, activities of daily living, behavioural symptoms, and caregiver burden of idalopirdine.
Methods Study design and patients For this double-blind, randomised, fixed-dose, placebocontrolled study (LADDER) we enrolled patients from 48 outpatient clinical sites in seven countries (Australia, Canada, the Czech Republic, Germany, Italy, Poland, and Spain). We enrolled patients between Dec 8, 2009, and Dec 23, 2011. Ambulatory patients, both male and female, aged 50 years or older, with probable Alzheimer’s disease consistent with National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association (NINCDS-ADRDA) criteria,6 with a CT or a MRI done within the previous 6 months, a mini-mental state examination (MMSE) score of 12–19 at screening and baseline, with no significant abnormalities on electrocardiogram (ECG) and physical examination and clinical laboratory tests, and who had been treated daily with donepezil for 4 months or more and stable on 10 mg per day for 3 months or more before screening, were eligible for inclusion in the study. Both the patient and a reliable caregiver were required to be fluent in the language used for the psychometric testing. A full list of inclusion and exclusion criteria is provided in the appendix. We excluded patients if they had taken memantine within 2 months before screening, had received treatment with any investigational drug within 30 days of screening, or with a known hypersensitivity to 5-HT6 receptor antagonists. Patients with a current diagnosis of any psychiatric disorder, clinically significant or unstable systemic disease, exposure to toxins or severe medical procedures were also excluded. In addition, patients had to be withdrawn from the study if they withdrew consent, were lost to follow-up, were 75% or less compliant with study drug between any two visits, if the randomisation code was broken, or if serum aspartate aminotransferase or alanine aminotransferase concentrations exceeded the protocol-specified limits. Compliance with study drug was assessed by returned capsules. www.thelancet.com/neurology Vol 13 November 2014
We did the study in accordance with the principles of Good Clinical Practice7 and the Declaration of Helsinki.8 National central ethics committees gave approval in Czech Republic, Germany, Italy, Poland, and Spain; ethics committees or institutional review boards gave approval at each site in Australia and Canada. Eligible patients or their legal representatives provided written informed consent before participating.
Randomisation and masking We randomly assigned patients (1:1) to double-blind treatment with idalopirdine or placebo according to a randomisation list that was computer generated by H Lundbeck A/S in blocks of four, stratified by site. The person who generated the randomisation sequence was not otherwise involved in the trial. The details of the randomisation series were contained in a set of sealed opaque envelopes. At each centre, sequentially enrolled patients were assigned the lowest randomisation number. The study drug or placebo was given as capsules of identical appearance, taste, and smell. All study personnel and participants were masked to treatment assignment. During the study, the randomisation code was broken by the Qualified Person for Pharmacovigilance for 16 patients because of increased liver function test values. Individual patient treatment allocation was not communicated to any other clinical personnel before formal unmasking of the study for analysis. The raters who assessed patients using the Alzheimer’s Disease Cooperative Study Clinical Global Impression of Change (ADCS-CGIC) were masked to other efficacy and safety assessments.
Procedures The study comprised a 2-week screening period followed by a 24-week double-blind treatment period. Eligible patients were given idalopirdine 90 mg per day (30 mg three times a day) or placebo as adjunctive treatment to donepezil 10 mg per day. Patients were seen at baseline and at weeks 2, 4, 6, 8, 12, 16, and 24, with a safety follow-up 4 weeks after last assessment. After the study start, visits at weeks 2 and 6 were added in protocol amendments that also introduced stricter inclusion and withdrawal criteria and follow-up and repeat testing for liver function tests. Patients who withdrew were seen after the decision to withdraw. The cognitive subscale of the 70-point, 11-item Alzheimer’s Disease Assessment Scale (ADAS-cog)9 was rated at baseline and at weeks 4, 12, and 24. All raters underwent formal training in the scoring conventions for the scales and cognitive tests and only personnel trained as a rater were allowed to rate patients.
See Online for appendix
Outcomes The primary endpoint was the change from baseline in the ADAS-cog at week 24. Pre-specified secondary 1093
Articles
efficacy endpoints were the ADCS-CGIC,10 the Alzheimer’s Disease Cooperative Study Activities of Daily Living, 23-items scale (ADCS-ADL23),11 the neuropsychiatric inventory (NPI),12 the Neuropsychological Test Battery (NTB),13 and the Zarit Burden Interview (ZBI), all assessed at week 24.14 The NPI is a 12-item validated structured interview with a caregiver, designed to assess behavioural disturbances in patients with dementia, and includes ten behavioural and two neurovegetative areas. The NTB, modified from that used by Harrison and colleagues,13 was composed of the following six tests (subitems): Category Naming Test (CNT), Controlled Oral Word Association Test (COWAT), Simple Reaction Time (SRT), Choice Reaction Time (CRT), Go/No-Go Paradigm (Go/NoGo), and ADAS-cog Memory Word Recall. The ZBI is a 22-item, caregiver burden, self-rating questionnaire designed to assess the consequences of caring activities in the physical, psychological, and social areas. We recorded all adverse events and vital signs. Qualified personnel coded adverse events using the lowest level term according to MedDRA (version 14.0). We also assessed clinical safety laboratory tests, weight, body-mass index, ECGs, and physical examination findings.
Statistical analysis Safety analyses were based on the all-patients-treated set, comprising all patients who were randomly allocated to and took at least one dose of study drug. Efficacy analyses were pre-specified in the protocol and were based on the intention-to-treat principle, using the full-analysis set, comprising all patients in the allpatients-treated set who had at least one valid postbaseline assessment of the primary efficacy variable (ADAS-cog). All efficacy analyses were pre-specified in the protocol. We regarded a treatment difference of 2 points on the ADAS-cog total score after 24 weeks as a clinically relevant effect. An SD of 6, a withdrawal rate of 15%, a 5% significance level, and a minimum of 135 patients in each treatment group led to 72% power in analysing change from baseline at week 24 for completers. Simulations of various scenarios suggested that applying mixed model repeated measures (MMRM) would increase the power compared with an analysis ignoring data from dropouts. Although lower than what is often used, the study sponsor regarded this level of power as acceptable for this proof-of-concept study. The primary analysis assessed the treatment difference in efficacy based on the change from baseline in the ADAS-cog at week 24. We analysed the changes from baseline at weeks 4, 12, and 24 using MMRM, and estimated the treatment difference at these timepoints. The model included categorical effects of treatment, pooled site, week, and treatment-by-week interaction, plus the continuous, fixed covariate of baseline score 1094
and baseline score-by-week interaction. We used an unstructured covariance matrix to model the variation of the repeated measures within patients. We did a sensitivity analysis of the change from baseline using ANCOVA, with treatment and pooled site as factors and baseline score as covariate, for weeks 4, 12, and 24 separately using completers and last observation carried forward (LOCF). LOCF imputation of missing values has traditionally been used for efficacy analyses in clinical trials supporting registration and, in this particular study, LOCF can be expected to underestimate rather than overestimate treatment effects. Analyses of completers, however, tend to overestimate treatment effects, so in combination the two analyses were felt to be suited to address the sensitivity of the results of the MMRM model to the unobserved missing values. In the factor termed “pooled site”, small sites (ie, sites with one or two participants) not contributing to both treatment groups were pooled into single pooled sites within each country to include all observations in the analysis. We analysed the ADCS-ADL23, ADCS-CGIC, NPI, and ZBI changes in total scores using MMRM, similar to the primary analysis, as well as an ANCOVA with baseline score as covariate and treatment and pooled site as factors. We calculated standardised Z scores for each of the six individual tests comprising the NTB. We then calculated a Z score for each test by dividing the difference from baseline score at each assessment for each patient by the relevant baseline SD. We calculated a combined NTB composite Z score by taking the mean of the Z scores across the six tests (CNT, COWAT, SRT, CRT, Go/No-Go, ADAS-cog Memory Word Recall) at each assessment for each patient. Histograms of residuals and plots of residuals versus fitted values indicated approximate normality of residuals with no relation between fitted values and residuals. We used SAS (version 9.2) for all statistical analysis. This study is registered with ClinicalTrials.gov, number NCT01019421.
Role of the funding source The study funders generated the randomisation sequence and designed the study, collected and analysed the data, and were involved in data interpretation and in the drafting of this report with the support of a medical writer employed by the study funders. Two of the co-authors (KW, EC-J) are employed directly by the funders. All authors, including those employed by the study funders, had full access to all the data in the study and DW had final responsibility for the decision to submit for publication.
Results Between Dec 8, 2009, and Dec 23, 2011, we screened 430 patients, of whom 278 were randomly allocated to treatment: 133 to placebo and 145 to idalopirdine www.thelancet.com/neurology Vol 13 November 2014
Articles
(figure 1). 118 (89%) patients in the placebo group and 114 (79%) patients in the idalopirdine group completed the study. Baseline characteristics were much the same between the two groups (table 1). At baseline, 117 (81%) of 145 patients in the idalopirdine group and 111 (84%) of 133 patients in the placebo group took concomitant drugs (besides donepezil) that they continued to take during the trial, and 52 (36%) patients in the idalopirdine group and 45 (34%) patients in the placebo group continued concomitant drug treatment during the study. The most frequent primary reasons for withdrawal were adverse events and withdrawal of consent (figure 1). About 80% of the patients in each treatment group received study drugs for at least 140 days in the 24-week treatment period. The exposure to study drug was 57·1 patientyears in both treatment groups. In the pre-specified primary efficacy analysis, we saw an improvement in cognitive function with idalopirdine compared with placebo (table 2). At week 24, the change from baseline in ADAS-cog total score was +1·38 (SE 0·53) with placebo and –0·77 (0·55) with idalopirdine, giving a treatment difference of –2·16 points (95% CI –3·62 to –0·69), which was statistically significant (p=0·0040; table 2). Improvement in cognitive function was significantly better with idalopirdine than with placebo by week 12 (p=0·013; figure 2). Differences versus placebo were shown by ANCOVA for completers (difference versus placebo of –2·40, 95% CI –3·89 to –0·91; p=0·0017 at week 24) and after LOCF imputation for withdrawals (–1·85, –3·18 to –0·51; p=0·0070 at week 24). The number of patients randomised at each site ranged from one to 18. The results were not sensitive to the data from sites recruiting fewer than three patients. None of the effects of idalopirdine on the pre-specified secondary efficacy endpoints was statistically significant at week 24 (table 2; figures 3 and 4). In a post-hoc analysis, we found a higher mean MMSE score at week 24 (17·5 [SD 3·8]) than at baseline (16·7 [2·1]) in the idalopirdine group and a lower mean MMSE score at week 24 (16·4 [4·4]) than at baseline (16·9 [2·1]) in the placebo group. Treatment-emergent adverse events reported by at least 3% of the patients in either treatment group are shown in table 3. The higher incidence of treatmentemergent adverse events in the idalopirdine group relative to the placebo group was mainly due to increased liver enzyme values. During the treatment period, 25 patients withdrew owing to treatmentemergent adverse events, seven in the placebo group and 18 in the idalopirdine group (figure 1). 13 patients (all in the idalopirdine group) had increased (greater than two times the upper limit of normal) aspartate aminotransferase or alanine aminotransferase values, most of which were first recorded at week 6 or week 8. Eight of these patients had concentrations greater than three times the upper limit of normal with no concurrent increase in total bilirubin. For 12 patients, alanine aminotransferase or aspartate www.thelancet.com/neurology Vol 13 November 2014
aminotransferase values subsequently decreased to within the normal range within 1–16 weeks. For the thirteenth patient, alanine aminotransferase and aspartate aminotransferase concentrations had decreased to 1·5 times the upper limit of normal by
430 screened 144 excluded 122 met exclusion criteria 14 inclusion failure 4 consent withdrawn 1 lost to follow-up 3 other 286 baseline 8 excluded 6 met exclusion criteria 2 inclusion failure 278 randomised
133 randomly allocated to placebo plus donepezil
145 randomly allocated to idalopirdine 90 mg plus donepezil
15 withdrawn 7 adverse events* 1 lack of efficacy 1 non-compliance 4 consent withdrawn 2 protocol violation
31 withdrawn 18 adverse events 3 non-compliance 9 consent withdrawn† 1 protocol violation
132 analysed in full-analysis set
140 analysed in full-analysis set
Figure 1: Trial profile *Includes one patient withdrawn from full-analysis set due to absence of post-baseline Alzheimer’s Disease Assessment Scale–cognitive subscale (ADAS-cog) assessment. †Includes five patients withdrawn from full-analysis set due to absence of post-baseline ADAS-cog assessment.
Placebo (n=133)
Idalopirdine (n=145)
Age (years)
75 (7·2)
74 (7·5)
Men
44 (33%)
38 (26%)
White
131 (99%)
143 (99%)
Time since diagnosis (years)
2·2 (1·9)
2·1 (1·8)
Duration of donepezil use (years)
1·5 (1·6)
1·4 (1·6)
MMSE score
17 (2·1)
17 (2·1)
ADAS-cog
28 (11)
28 (10)
ADCS-ADL23
51 (16)
53 (14)
ADCS-CGIC*
4·0 (0·8)
3·9 (0·9)
Neuropsychiatric inventory (total score)
9·3 (12·0)
8·4 (8·9)
Neuropsychological test battery
0·00 (0·48)
0·01 (0·53)
Data are mean (SD) or n (%). *This scale is an assessment of change from baseline; the baseline value recorded was a severity score, in response to the question: “Considering your total clinical experience with this particular population, how ill is the patient at this time?”. MMSE=mini-mental state examination. ADAS-cog=Alzheimer’s Disease Assessment Scale cognitive subscale. ADCS-ADL23=Alzheimer’s Disease Cooperative Study Activities of Daily Living 23-Items scale. ADCS-CGIC=Alzheimer’s Disease Cooperative Study Clinical Global Impression of Change scale.
Table 1: Baseline demographic and clinical characteristics
1095
Articles
Overall baseline mean
Week 24 adjusted mean*
Difference to placebo
Placebo
Mean
Idalopirdine
95% CI
p value
Primary analysis ADAS-cog
28·0 (0·65)
29·4 (0·53)
27·2 (0·54)
51·7 (0·88)
47·9 (0·80)
49·6 (0·82)
–2·16 (0·74)
–3·62 to –0·69
0·0040
Secondary analyses ADCS-ADL23 ADCS-CGIC
3·95 (0·05)
Neuropsychiatric Inventory (total score)
8·88 (0·62)
12·2 (0·85)
Neuropsychological test battery (composite Z score)
0·01 (0·03)
–0·04 (0·04)
Zarit burden interview (total score)
22·9 (0·82)
4·47 (0·10)
26·5 (0·94)
1·72 (1·11)
–0·48 to 3·92
0·12
4·25 (0·10)
–0·22 (0·14)
–0·49 to 0·054
0·12
10·8 (0·86)
–1·45 (1·18)
–3·78 to 0·88
0·22
0·07 (0·06)
–0·05 to 0·18
0·24
–0·29 (1·30)
–2·85 to 2·27
0·82
0·03 (0·04) 26·2 (0·94)
Data are mean (SE ) unless otherwise stated. *Adjusted means are least squares means corresponding to the overall baseline mean scores ADAS-cog=Alzheimer’s Disease Assessment Scale cognitive subscale. ADCS-ADL23=Alzheimer’s Disease Cooperative Study Activities of Daily Living 23 Items scale. ADCS-CGIC=Alzheimer’s Disease Cooperative Study Clinical Global Impression of Change scale.
Table 2: Efficacy analyses
Adjusted mean change from baseline in ADAS-cog
–3 –2
p=0·0130 p=0·0040
–1 0 1 2
Donepezil plus idalopirdine 90 mg Donepezil plus placebo
3 0
4
Placebo (n) 132 Idalopirdine (n) 140
8
12 16 Treatment week
131 140
20
125 127
24 118 114
Adjusted mean change from baseline in ADCS-ADL23
Figure 2: Adjusted mean change in ADAS-cog (full-analysis set, MMRM by visit) Error bars are 95% CI. ADAS-cog=11-item Alzheimer’s Disease Assessment Scale, Cognitive Subscale. MMRM=mixed model, repeated measures.
2
Donepezil plus idalopirdine 90 mg Donepezil plus placebo
1 0 –1 –2 –3 –4 –5 –6 0
4
8
12
16
20
24
Treatment week
Figure 3: Adjusted mean change in ADCS-ADL23 (full-analysis set, MMRM by visit) Error bars are 95% CI. ADCS-ADL23=23-item Alzheimer’s Disease Cooperative Study Activities of Daily Living scale. MMRM=mixed model, repeated measures.
1096
week 24, after which no further testing was done. The median time for concentrations to return to normal was 4 weeks (IQR 2–7). Eleven of the 13 patients with increased alanine aminotransferase or aspartate aminotransferase values withdrew. Treatment-emergent serious adverse events were reported by 27 patients (table 3). Serious adverse events reported by more than one patient were syncope (two patients in the idalopirdine group), femoral neck fracture (two patients in the placebo group), and gastroenteritis (two patients in the placebo group). One patient (in the idalopirdine group) was admitted to hospital with epilepsy and one patient (in the placebo group) was admitted to hospital after a convulsion. In the idalopirdine group, we judged two serious adverse events to be probably related to treatment and six serious adverse events to be possibly related to treatment (all patients recovered, one with sequelae). In the placebo group, we judged five serious adverse events to be possibly related to treatment (four patients recovered, one died). Two patients died during this study. A 79-year-old man who had received idalopirdine for 169 days was hit by a motorcycle and incurred a cranio-cerebral injury and died 22 days later from bronchopneumonia. The second patient, an 80-year-old man who had received placebo for 2 days, fell and incurred a femoral neck fracture (serious adverse events). He developed bronchopneumonia, was admitted to hospital, and died. Neither death was considered by the investigator to be related to the study drug. For the clinical safety laboratory tests except for liver function tests, we saw no clinically relevant mean changes over time or differences between the treatment groups (data not shown).
Discussion Our findings suggest that idalopirdine is effective in improving cognitive function in patients with moderate Alzheimer’s disease who are receiving donepezil, with www.thelancet.com/neurology Vol 13 November 2014
Articles
Adjusted mean change from baseline in ADCS-CGIC
5·0
Donepezil plus idalopirdine 90 mg Donepezil plus placebo
4·5
4·0 p=0·0204 3·5
0 0
4
8
12
16
20
24
Treatment week
Figure 4: Adjusted mean ADCS-CGIC (full-analysis set, MMRM by visit) Error bars are 95% CI. ADCS-CGIC=Alzheimer’s Disease Cooperative Study Clinical Global Impression of Change scale. MMRM=mixed model, repeated measures.
a mean difference in ADAS-cog at week 24 of 2·16 points. The improvements in functional and global clinical measures were not statistically significant at week 24, and the study sponsor is doing confirmatory studies in larger populations to elucidate further the clinical effects. Our treatment efficacy effect estimate was based on MMRM, which provides a reliable estimate of efficacy under a missing at random assumption for the missing values.15 The excess withdrawal of patients from the study on idalopirdine versus placebo was driven by the asymptomatic liver function test increases and automatic withdrawal of those patients according to the protocol, and can thus be regarded as missing at random—ie, not related to the unobserved efficacy outcomes. Sensitivity analyses based on carrying the last observation forward (LOCF) for withdrawals and based on completers lent support to the MMRM findings. The improved cognitive performance in the idalopirdine group, as assessed by use of the ADAS-cog total score, was consistent with changes in the MMSE score. Although the MMSE was not a pre-planned efficacy outcome, data were collected at baseline and at week 24 and analysis showed improvement in the idalopirdine group and worsening in the placebo group. However, analyses of the NTB composite Z score, a preplanned efficacy outcome, showed no statistically significant treatment difference. Through their actions on multiple transmitter systems, 5-HT6 receptor antagonists might also improve behavioural symptoms in Alzheimer’s disease.16 In this study, we saw no effect on the NPI score. However, the average baseline NPI total score of the patients was quite low. In some patients, treatment with idalopirdine was associated with asymptomatic liver enzyme increases, particularly aspartate aminotransferase, alanine aminowww.thelancet.com/neurology Vol 13 November 2014
Placebo (n=133)
Idalopirdine (n=145)
Patients with serious adverse events
13 (10%)
14 (10%)
Patients with treatment-emergent adverse events
78 (59%)
96 (66%)
Increased γ-glutamyltransferase
2 (2%)
14 (10%)
Increased alanine aminotransferase
0
Dizziness
3 (2%)
7 (5%)
Aggression
2 (2%)
6 (4%)
Increased aspartate aminotransferase
0
6 (4%)
Diarrhoea
9 (7%)
6 (4%)
Headache
2 (2%)
6 (4%)
Hypertension
5 (4%)
6 (4%)
Vomiting
1 (1%)
6 (4%)
Increased blood alkaline phosphatase
1 (1%)
5 (3%)
Hypercholesterolaemia
2 (2%)
5 (3%)
Nausea
4 (3%)
5 (3%)
Weight loss
2 (2%)
5 (3%)
Fall
8 (6%)
3 (2%)
Urinary tract infection
9 (7%)
3 (2%)
Agitation
5 (4%)
2 (1%)
9 (6%)
Benign prostatic hyperplasia (men; n=38)
0
2 (5%)
Contusion
4 (3%)
0
Data are number of patients (%).
Table 3: Treatment-emergent adverse events occurring in 3% or more of patients in either treatment group (all patients treated set)
transferase, and γ-glutamyltransferase. For patients with increased concentrations, the values decreased towards the reference range later in the study. The incidence of gastrointestinal adverse events (nausea, vomiting and diarrhoea) was low in both treatment groups. Owing to inhibition of CYP2D6, coadministration of idalopirdine and donepezil resulted in an approximate 10% increase in the idalopirdine group compared with the placebo group at week 4 (no between-group difference at week 24) in the mean plasma concentration of donepezil (data not shown). However, we judge the effect of this small increase to be negligible, in view of the slight reported differences in tolerability and efficacy between 5 mg per day and 10 mg per day donepezil, which corresponds to an approximate 100% increase in plasma concentration.17 We tested the effect of idalopirdine in a selected population of donepezil-treated patients with moderate Alzheimer’s disease. The addition of idalopirdine to acetyl cholinesterase inhibitors is being explored in ongoing studies (NCT02006641, NCT01955161, NCT02006654, NCT02079246). Another 5-HT6 receptor antagonist, GSK SB-742457, has previously been explored and showed no significant benefit on ADAS-cog in three monotherapy studies.18–20 However, in one study21 it was assessed as an adjunct to donepezil and was shown to significantly improve ADAS-cog score (panel).21 Our study was a phase 2 study, so inevitably had limitations. The small size of our study could have been 1097
Articles
Panel: Research in context Systematic review We searched PubMed and clinicaltrials.gov for articles in all languages published up to July 9, 2014, that reported any randomised therapeutic trial with a 5-HT6 receptor antagonist. We used the search terms “a 5-HT6 receptor antagonist” and “clinical trial”, and restricted the search to studies in adult human beings. We applied no language restrictions. We identified four completed controlled phase 2 trials18–21 with the 5-HT6 receptor antagonist SB-742457 and articles describing the results of two of these studies.18,19 Neither showed a statistically significant improvement on the ADAS-cog versus placebo from baseline to week 24. Findings from a third study with SB-742457 as monotherapy,20 also showed no significant difference from placebo. A fourth trial (in patients already receiving donepezil) showed statistically significant improvements in patients receiving donepezil and SB-742457 versus patients receiving donepezil and placebo in the ADAS-cog at weeks 24 and 48.21 We also identified a report of the results from a dose-finding trial with SAM-531 (a 5-HT6 receptor antagonist) in patients with mild to moderate Alzheimer’s disease.22 Findings from this trial showed no superiority compared with placebo on the ADAS-cog at week 24.22 Interpretation In four of the five phase 2 trials identified by our search,18–20,22 a 5-HT6 receptor antagonist was used as monotherapy and failed to show a difference from placebo, but our findings and those of one other trial21 together suggest that the approach of using a 5-HT6 receptor antagonist in patients already receiving an acetylcholinesterase inhibitor should be investigated in further studies.
responsible for the absence of effect on secondary measures other than cognition (larger studies powered for these secondary measures are ongoing). Patients in this study had moderate Alzheimer’s disease, so we cannot extrapolate the results to other groups of patients with milder or more severe disease. The results from this study apply only to patients stably treated with donepezil and cannot be generalised to patients treated with other acetyl cholinesterase inhibitors. Additionally, the use of a single idalopirdine dose means we could not establish whether lower doses would have been efficacious. These data add to the evidence that 5-HT6 receptor antagonism is an interesting approach to the symptomatic treatment of Alzheimer’s disease, and phase 3 studies have been initiated with idalopirdine to explore this hypothesis. Contributors DW had the idea for and designed the study. DW and all the study investigators acquired the data. All authors analysed and interpreted the data and directed content of the report after the initial draft was prepared with the support of a medical writer, who provided a draft
1098
version of Methods and Results, and the authors made changes to the draft. All authors revised the report, approved the final version of the report, and vouch for the integrity of the data and analyses. Study investigators Australia: Roger Clarnette, Michael Woodward, Karyn Boundy, Denis Crimmins, and David Ames. Canada: Sharon Cohen, Angeles Garcia, Guy Lacombe, David Patry, William Nisker, Richard Bergeron, and Miranda Du Preez. Czech Republic: Claudia Borzova, Slavomir Pietrucha, Oldrich Vysata, Ladislav Pazdera, Petranek Svojmil, Martin Brunovsky, Zdenek Solle, Michal Bajacek, Jiri Pivejc, and Jaroslav Lestina. Germany: Matthias Riepe, Gerd Reifschneider, Walter Albrecht, Johannes Pantel, David Prulovic, Hermann-Josef Gertz, and Karl-Otto Sigel. Italy: Alessandro Padovani, Giuseppe Bruno, Guido Rodriguez, Amalia Cecilia Bruni, Angelo Bianchetti, Elio Scarpini, and Sandro Sorbi. Poland: Andrzej Szczudlik, Robert Kucharski, Dorota Ussorowska, Arleta Kuczynska-Zardzewialy, Jacek Staszewski, Alicja Klich-Raczka, Roman Chwedorowicz, Maciej Czarnecki, Andrzej Potemkowski, and Grzegorz Opala. Spain: Mercedes Boada, Jordi Alom, Javier Olzaran, Pedro Gil Gregorio, and Juan Jose López Lozano. Declaration of interests DW has received honoraria for serving on scientific advisory boards and financial compensation for speaking at symposia on behalf of H Lundbeck A/S, Nutricia, and MedNet Consulting outside the submitted work. KW (biostatistician) and ECJ are employees of H Lundbeck A/S. Acknowledgments We thank all patients for their participation in the study and D J Simpson (H Lundbeck A/S) for providing support in the preparation, revision, and editing of the first draft of the paper. References 1 Geldenhuys WJ, Van der Schyf CJ. Role of serotonin in Alzheimer’s disease: a new therapeutic target? CNS Drugs 2011; 25: 765–81. 2 Ramírez MJ. 5-HT6 receptors and Alzheimer’s disease. Alzheimers Res Ther 2013; 5: 15. 3 Upton N, Chuang TT, Hunter AJ, Virley DJ. 5-HT6 receptor antagonists as novel cognitive enhancing agents for Alzheimer’s disease. Neurotherapeutics 2008; 5: 458–69. 4 Arnt J, Bang-Andersen B, Grayson B, et al. Lu AE58054, a 5-HT6 antagonist, reverses cognitive impairment induced by subchronic phencyclidine in a novel object recognition test in rats. Int J Neuropsychopharmacol 2010; 13: 1021–33. 5 Herrik KF, de Jong I, Pedersen JT, et al. The 5-HT6 antagonist Lu AE58054 potentiates the effects of the acetylcholinesterase inhibitor donepezil on hippocampal extracellular acetylcholine efflux and oscillatory activity. Alzheimers Dement 2013; 9 (suppl 4): P503. 6 McKhann G, Drachman D, Folstein M, et al. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 1984; 34: 939–44. 7 ICH Harmonised Tripartite Guideline E6: Guideline for Good Clinical Practice. 1996. http://www.fda.gov/downloads/drugs/ guidancecomplianceregulatoryinformation/guidances/ ucm073122.pdf (accessed July 4, 2014). 8 World Medical Association (WMA). Declaration of Helsinki: Ethical principles for medical research involving human subjects. www.wma.net/en/30publications/10policies/b3/ (accessed July 4, 2014). 9 Rosen WG, Mohs RC, Davis KL. A new rating scale for Alzheimer’s disease. Am J Psychiatry 1984; 141: 1356–64. 10 Schneider LS, Olin JT, Doody RS, et al. Validity and reliability of the Alzheimer’s disease cooperative study- clinical global impression of change. Alzheimer Dis Assoc Disord 1997; 11 (suppl 2): S22–32. 11 Galasko D, Bennett D, Sano M, et al. An inventory to assess activities of daily living for clinical trials in Alzheimer disease. The Alzheimer’s disease Cooperative Study. Alzheimer Dis Assoc Disord 1997; 11 (suppl 2): S33–39.
www.thelancet.com/neurology Vol 13 November 2014
Articles
12
13
14 15
16 17
18
Cummings JL, Mega M, Gray K, Rosenberg-Thompson S, Carusi DA, Gornbein J. The Neuropsychiatric Inventory: comprehensive assessment of psychopathology in dementia. Neurology 1994; 44: 2308–14. Harrison J, Minassian SL, Jenkins L, Black RS, Koller M, Grundman M. A neuropsychological test battery for use in Alzheimer disease clinical trials. Arch Neurol 2007; 64: 1323–29. Zarit SH, Orr NK, Zarit JM. The hidden victims of Alzheimer’s disease. New York: New York University Press, 1985. Siddiqui O, Hung HM, O’Neill R. MMRM vs LOCF: a comprehensive comparison based on simulation study and 25 NDA datasets. J Biopharm Stat 2009; 19: 227–46. Dawson LA. The central role of 5-HT6 receptors in modulating brain neurochemistry. Int Rev Neurobiol 2011; 96: 1–26. Eisai Inc. Prescribing information: Aricept (donepezil hydrochloride) tablets. http://www.aricept.com/docs/pdf/aricept_ PI.pdf (accessed Sept 9, 2014). Maher-Edwards G, Zvartau-Hind M, Hunter AJ, et al. Doubleblind, controlled phase II study of a 5-HT6 receptor antagonist, SB-742457, in Alzheimer’s disease. Curr Alzheimer Res 2010; 7: 374–85.
www.thelancet.com/neurology Vol 13 November 2014
19
20
21
22
Maher-Edwards G, Dixon R, Hunter J, et al. SB-742457 and donepezil in Alzheimer disease: a randomized, placebo-controlled study. Int J Geriatr Psychiatry 2011; 26: 536–44. GSK. Study AZ3110865, a study comparing SB742457 or donepezil versus placebo in subjects with mild-to-moderate Alzheimer’s disease. http://www.gsk-clinicalstudyregister.com/result_detail.js p?protocolId=AZ3110865&studyId=242C4974-9729-4D30-8F91327CF0631014&compound=SB742457 (accessed July 4, 2014). Maher-Edwards G, Ascher J, Watson C, et al. 48-week efficacy and tolerability of treatment with SB-742457, a novel 5HT6 receptor antagonist, when added on to donepezil in subjects with mild-tomoderate Alzheimer’s disease (AD). Alzheimer’s Dementia 2011; 7 (suppl 4): e5. Pfizer. Study comparing 3 dosage levels of SAM-531 in outpatients with mild to moderate Alzheimer disease. http://clinicaltrials. gov/ct2/show/results/NCT00895895?term=sam531&rank=1§=X70156#outcome1 (accessed July 4, 2014).
1099