Safety and efficacy of rivastigmine in children with Down syndrome: A double blind placebo controlled trial

RESEARCH ARTICLE

Safety and Efﬁcacy of Rivastigmine in Children

with Down Syndrome: A Double Blind Placebo

Controlled Trial

Gail A. Spiridigliozzi,

Sarah J. Hart,

* James H. Heller,

Heather E. Schneider,

Jane Ann Baker,

Cathleen Weadon,

George T. Capone,

5,6

and Priya S. Kishnani

Department of Psychiatry and Behavioral Sciences, Duke University Medical Center, Durham, North Carolina

Department of Pediatrics, Duke University Medical Center, Durham, North Carolina

Formerly of Department of Surgery, Duke University Medical Center, Durham, North Carolina

Department of Neuropsychology, Kennedy Krieger Institute, Baltimore, Maryland

Down Syndrome Clinic and Research Center, Kennedy Krieger Institute, Baltimore, Maryland

Department of Pediatrics, Johns Hopkins University School of Medicine, Baltimore, Maryland

Manuscript Received: 13 October 2015; Manuscript Accepted: 11 March 2016

Individuals with Down syndrome (DS) have decreased cho-

linergic function and an uneven proﬁle of cognitive abilities,

with more pronounced deﬁcits in learning, memory, and

expressive language. Choline ster ase inhibito rs m ay improv e

cognitive function in adults and adolescents with DS, but

studiesinchildrenwithDShavebeenlimited.Thisstudy

aimed to: (i) investigate the safety and efﬁcacy of rivastigmine

treatment; (ii) build upon our open-label studies in children

with DS in a double-blind, placebo-controlled clinical trial;

and (iii) investigate speciﬁc cognitive domains that

may respond to rivastigmine t reatment. We conducted a

20-week double-blind, placebo-controlled trial to investiga te

the safety and efﬁcacy of rivastigmine in 22 children a nd

adolescents with DS aged 10–17 years. Safety measures inclu-

ded r eports of adverse events, laboratory parameters, and

electrocardiograms. Efﬁcacy measures included parental

assessments of adaptive behavior and executive function,

and direct measures of language and memory. No group

differences were found o n safety m easures and 22 of 24

participants that passed study screening completed the study.

The results did not demonstrate evidence for signiﬁcant

improvement in aspects of cognition, language, or overall

function in the children receiving rivastigmine. Our results

suggest that rivastigmine is safe and well-tolerated for chil-

dren and adolescents with DS, but may not be effective for

improving performance on the selected measures in this study.

However, larger samples and/or alternate measures could

possibly reveal improvements in cognitive function with riva-

stigmine treatment. Further research is needed to deﬁne a

battery of cognitive measures that is sensitive to treatment

effects in DS.

Ó 2016 Wiley Periodicals, Inc.

Key words: Down syndrome; trisomy 21; cholinesterase

inhibitors; cognition; adaptive behavior

INTRODUCTION

Individuals with Down syndrome (DS) show an uneven proﬁle of

cognitive abilities, with more pronounced deﬁcits in learning, mem-

ory, and language [Kishnani et al., 2010]. There is currently an unmet

need for strategies to improve independence and overall performance

in these cognitive domains for individuals with DS. The cognitive

deﬁcits in DS may be associated with an innate cholinergic deﬁcit

or dysfunction, as abnormalities have been demonstrated in both

Conflict of interest: none.

Grant sponsor: Taishoff Family Foundation (Naples, FL); Grant sponsor:

The Anna Michelle Merrills Foundation for Down Syndrome Research

(Durham, NC).



Correspondence to:

Sarah J. Hart, Department of Pediatrics, Duke University Medical

Center, 905 S. LaSalle St., Box 103857, Durham, NC 27710.

E-mail: [email protected]

Article first published online in Wiley Online Library

(wileyonlinelibrary.com): 8 April 2016

DOI 10.1002/ajmg.a.37650

How to Cite this Article:

Spiridigliozzi GA, Hart SJ, Heller JH,

Schneider HE, Baker JA, Weadon C,

Capone GT, Kishnani PS. 2016. Safety and

efﬁcacy of rivasti gmine in children with

Down syndrome: A double blind placebo

controlled trial.

Am J Med Genet Part A 170A:1545–1555.

Ó 2016 Wiley Periodicals, Inc. 1545

peripheral and central cholinergic function [Sacks and Smith, 1989;

Florez et al., 1990; Beccaria et al., 1998]. As the central cholinergic

system is critical in cognition, memory, attention, and mood, enh-

ancing cholinergic function may be an appropriate target for treat-

ment in individuals with DS. DS has been associated with reductions

in cholinergic neurons [Casanova et al., 1985], leading to the hypoth-

esis that decreased neurogenesis or survival of cholinergic neurons

could underlie altered cortical connectivity and delayed maturation of

cortical neurons in children with DS [Becker et al., 1991; Berger-

Sweeney, 2003]. This hypothesis implies that increasing cholinergic

function early in development could potentially improve cognitive

function by altering cortical connectivity [Berger-Sweeney, 2003].

Cholinesterase inhibi tors (ChEIs), including rivastigmine and done-

pezil, have been investigated as potential therapeutic agents to

improve cognitive performance.

Several clinical trials have been completed to test the safety and

efﬁcacy of ChEIs for improving cognitive and language function in

adults with DS who do not have Alzheimer disease. The ﬁrst clinical

trial of ChEIs in individuals with DS demonstrat ed improved

global functioning in a small sample of adults treated with done-

pezil [Kishnani et al., 1999]. A subsequent open-label study of

donepezil indicated that donepezil was generally well-tolerated and

was associated with improvements in expressive language in adults

with DS [Heller et al., 2003]. Additionally, several double-blind,

placebo-controlled studies of donepezil have shown gains in

memory, language, and global function in individuals with DS

[Johnson et al., 2003; Kishnani et al., 2009; Kondoh et al., 2011].

Studies investigating efﬁcacy of ChEIs in children with DS have

yielded mixed resu lts. Using donepezil and rivastigmine tartrate,

open-label trials with children (aged 8–13 years) [Heller et al., 2004 ;

Spiridigliozzi et al., 2007] and adolescents (aged 10–17 years)

[Heller et al., 2006b] with DS have shown similar increases in

expressive language performance, memory, and attention. How-

ever, several recent clinical trials have failed to demonstrate signif-

icant effects of ChEIs on cognition in children with DS. A

long-term follow-up study of rivastigmine in adolescents with

DS indicated no between-group differences in cognitive or lan-

guage performance change over time [Heller et al., 201 0]. However,

at least two participants in this study showed long-term gains in

adaptive function, suggesting that there may be a subset of indi-

viduals that respond to the medication. A double-blind placebo-

controlled study of donepezil in adolescents with DS also found no

group differences in adaptive behavior, with both the experimental

and placebo groups showing signiﬁcant improvement over the

course of the trial [Kishnani et al., 2010].

The previous studies of ChEIs in children with DS have been

limitedby open-labeldesignsor by the speciﬁc cognitive assessments

that were selected. The study by Kishnani et al. [2010] did not

demonstrate improvement in cognition using a limited set of

measures with administration of donepezil. We aimed in this study

to investigate the safety and efﬁcacy of rivastigmine treatment in a

double-blind design and to expand the assessment battery to include

other speciﬁc cognitive domains that may selectively respond to

ChEIs. To our knowledge, there have been no double-blind ran-

domized trials conducted using rivastigmine in children with DS.

MATERIALS AND METHODS

We conducted a 20-week double-blind, placebo-controlled trial for

safety and efﬁc acy of rivastigmine at Duke University Medical

Center and at the Kennedy Krieger Institute at Johns Hopkins

University (Clinical Trials Registry NCT01084135). The study was

prospectively reviewed and approved by both the Duke University

and Johns Hopkins University Institutional Review Boards.

Participants

Eligible participants for enrollment in the study included males and

females ages 10 through 17 years of age with a diagnosis of DS (free

trisomy 21 or Robertsonian translocation for trisomy 21 as deter-

mined by chromosome analysis). Participants were required to be

outpatients with reliable caregivers. Participants were also req uired

to have sufﬁcient verbal skills for stud y participation as deﬁned by

the Expressive One-Word Picture Vocabulary Test (EOWPVT)

[Brownell, 2000] and examiner assessment (described below).

Exclusion criteria included any medical con dition that was con-

sidered both signiﬁcant and unstable, including an active medical

history of gastric ulcer, glaucoma, poorly controlled asthma, and

prolonged use (>2 weeks) of NSAIDs. Exclusion criteria also

included any active or clinically signiﬁcant conditions affecting

absorption, distribution, or metabolism of the study drugs.

Participants were recruited from DS clinics and the community.

A total of 24 participants with DS were enrolled into the study,

including 12 participants enrolled at Johns Hopkins University and

10 participants at Duke University. Two participants at the Johns

Hopkins site declined to continue with the study following the

screening visit. A total of 12 parti cipants received rivastigmine and

10 received placebo. The treatment group included four males and

eight females, and the placebo group included four males and six

females. Participant ages are presented in Table I by treatment

group and gender. No signiﬁcant differences were fou nd between

treatment groups for age (P ¼ 0.24) or gender (P ¼ 1.0). No

participants had a history of epilepsy or seizures. A history of

cardiac surgery for structur al heart defects such as atrioventricular

TABLE I. Demographic Characteristics

Rivastigmine Placebo

Male ¼ 4 Female ¼ 8 Total (n ¼ 12) Male ¼ 4 Female ¼ 6 Total (n ¼ 10)

Mean age (SD) 13.02 (1.79) 13.89 (1.98) 13.6 (1.88) 12.79 (1.51) 12.54 (2.08) 12.64 (1.78)

Age range 10–14 10–17 10–17 10–14 10–16 10–16

1546 AMERICAN JOURNAL OF MEDICAL GENETICS PART A

canal defects and valvular defects was noted for ﬁve individuals in

the treatme nt group and four individuals in the placebo group.

Four individuals in the treatment group had comorbid neuro-

developmental disabilities including ADHD (three participants)

and sensory integration disorder (one participant). In the placebo

group, one individual had ADHD, two had oppositional deﬁant

disorder, and one had pervasive developmental disorder. No

signiﬁcant differe nces were found between treatment groups in

the number of individuals with comorbid neurodevelopmental

disabilities (P ¼ 1.0) or history of cardiac surgery (P ¼ 1.0).

Overview of Study Visits

The study consisted of a screening visit (4 weeks), a baseline visit

(week 0), a safety visit at week 10, and a ﬁnal visit at week 20 (Fig. 1).

Efﬁcacy measures of overall function, language, and cognition

(described below) were administered at the screening, baseline,

and ﬁnal (week 20) visits. Vital signs, physical examinations, and

pregnancy testing for females of childbearing potential were con-

ducted at all visits. An electrocardiogram (EKG) was performed at

the screening and week 20 visits only. Participants receiving placebo

maintained the same schedule as those receiving the medication.

Screening visit. At the screening visit, participants’ caregivers

provided informed consent and the child provided assent prior to

all study activities. Blood was drawn to check thyroid function, liver

function, and comple te blood count. Participants were required to

have normal bloodwork to continue participation in the study. A

12-lead EKG was also performed for all participants.

Participants were administered the examples and items one to nine

of the EOWPVT to determine eligibility to proceed with the study. The

examiner also spoke informally with the child regarding a subject of

interest to the child for approximately 5 min. In order to continue

participation in the study, participants were required to provide correct

verbal responses for seven of nine of the EOWPVT items. The child was

required to be able to put at least two to three words together in

conversational speech. The child’s speech was required to be under-

standable to the examiner for the majority of the time and the child

could not use sign language or an augmentative communication tool as

the primary means of communication. The parent and child were also

interviewed for about 30 min by a clinician about the child’s medical

history and to assess the parent’s impression of the child’s functioning.

Baseline visit. At the baseline visit, participants were random-

ized by the Investigational Pharmacies at Duke University and

Johns Hopkins University to receive liquid rivastigmine or liquid

placebo orally. Blinded data was maintained by each site’s Investi-

gational Pharmacy. Rivastigmine was titrated, based on tolerabil-

ity, on the following schedule: At the baseline visit (week

0), participants began rivastigmine treatment at a dose of

1.5 mg/day (0.75 mg bid). This dose was continued for 2 weeks.

Parents of participants received a phone call from the study

coordinator after 2 weeks to determine the child’s dose tolerance.

The information about the child’s dose tolerance was shared with

the study PI, who determined whether or not to increase the child’s

dose to 3.0 mg/day (1.5 mg bid) for an additional 6 weeks.

Caregivers were provided with study diaries to record drug

administration compliance and changes in medical history. Com-

pliance diaries were reviewed at each visit and adverse events (AEs)

and serious adverse events (SAEs) were recorded. AEs were classi-

ﬁed by the study investigators as being potenti ally related or as

likely unrelated to rivastigmine. A SAE was deﬁned as: (i) results

in death; (ii) is life-threatening; (iii) requires hospitalization;

(iv) results in persistent or signiﬁcant functional disability; or

(v) other medical event that may jeopardize the participant or

require intervention to prevent one of the outcomes listed above.

FIG. 1. Timeline of study visits.

SPIRIDIGLIOZZI ET AL. 1547

All participants were administered the NEPSY Immediate Mem-

ory for Names subtest [Korkman et al., 1998], subtests from the

Clinical Evaluation of Language Fundamentals-Preschool-2

(CELF-P-2) [Semel et al., 2004], and the Test of Verbal Expression

and Reasoning (TOVER) [Heller et al., 2000] . Parents/caregivers

were administered the Vineland Adaptive Behavior Scales , Second

Edition, Survey Interview Form (VABS-II) [Sparrow et al., 2005],

the Behavior Rating Inventory of Executive Function-Preschool

Version (BRIEF-P) [Gioia et al., 2002], and the Rating of Every Day

Activities and Life Skills (REAL) [Spiridigliozzi et al., 2010].

Safety visit. At the week 10 safety visit, the child’s dose toler-

ance was reviewed again. Parents and the child were interviewed for

about 30 min about changes in his/her functioning. Based on the

participant’s response to his/her dose, the dose was increased to

4.5 mg/day (3.0 mg and 1.5 mg). All participants were able to

tolerate the 4.5 mg/day dose.

Final visit. Assessments at study termination (week 20) inc-

luded a full physical exam and medical history intake, vital signs,

routine blood work, and a thyroid proﬁle. Participants were also

given an EKG and liver function tests. Efﬁcacy measures adminis-

tered at the baseline visit with the participants and their parents/

caregivers were repeated at this ﬁnal visit.

Efﬁcacy Measures

Efﬁcacy measures were obtained at the baseline and ﬁnal visits

(Fig. 1) and included both parental report measures and direct

assessments of the child’s performance. Parental report measures

included the VABS-II Survey Interview Form, REAL, and BRIEF-P.

The VABS-II is a measure of adaptive behavior in children,

adolescents, and adults. It yields an overall standard score (Adap-

tive Behavior Composite, ABC) and age standard scores in three

domains (Communication, Daily Living Skills, and Socialization).

These scores have a mean of 100 and a standard deviation of 15

(range ¼ 20–160). Higher scores suggest a higher level of adaptive

functioning. In this study, the change between each participant’s

ABC at baseline and the ﬁnal visit was computed. A rise in standard

scores from baseline to the ﬁnal visit indicates improvement.

The REAL is a 29-item parent report measure of an individual’s

behavior, speech and language ability, mood, social interactions,

and executive functioning/attention during the past month. It was

developed in an attempt to brieﬂy characterize each child’s func-

tioning in those areas that may be sensitive to a treatment effect.

Higher scores suggest a higher level of overall functioning. In this

study, we analyzed subscale scores for behavior and executive

function/attention and the Total Scores only.

The BRIEF-P is a parent report measure of executive function

behaviors in children in their home setting. It yields an overall score

(Global Executive Composite, GEC), that is, based on its ﬁve

clinical scales. Raw scores range from 63 to 189. Higher scores

suggest that an individual’s executive function skills are more

problematic. In this study, the change between each subject’s

raw score at bas eline and the ﬁnal visit was compu ted for the

Global Executive Composite. A decline in raw scores from baseline

to the ﬁnal visit indicates improvement.

Direct assessments of the child’s perf ormance included the

NEPSY Immediate Memory for Names subtest, the CELF-P-2

Recalling Sentences and Word Classes subtests, and the TOVER.

The NEPSY Memory for Names subtestmeasures the child’sability

to learn and recall the names of eight pictured children over three

learning trials. The total number of names recalled immediately after

each learning trial was analyzed. These scores range from 0 to 24.

TABLE II. Treatment-Emergent Adverse Events by Preferred Term,

All Causalities

Treatment group

Any adverse event

Rivastigmine

(n ¼ 12), n

(%)

Placebo

(n ¼ 10), n

(%)

Stomachache 5 (41.7) 3 (30.0)

Diarrhea 3 (25.0) 2 (20.0)

Nausea 3 (25.0) 3 (30.0)

Vomiting 3 (25.0) 1 (10.0)

Cold 3 (25.0) 2 (20.0)

Menstrual cramps 2 (16.7) 0

Stomach flu 2 (16.7) 0

Sleepy 2 (16.7) 0

Dizziness 1 (8.3) 0

Headache 1 (8.3) 1 (10.0)

Trouble sleeping 1 (8.3) 1 (10.0)

Decreased appetite 1 (8.3) 1 (10.0)

Weakness 1 (8.3) 0

Lumps in neck 1 (8.3) 0

Nasal congestion 1 (8.3) 0

Boil 1 (8.3) 0

Cut 1 (8.3) 0

Restless legs 1 (8.3) 0

Pale coloring 1 (8.3) 0

Foot twitching 1 (8.3) 0

Itchy 1 (8.3) 0

Eye twitch 1 (8.3) 0

Rash 1 (8.3) 0

Indigestion 1 (8.3) 0

Constipation 1 (8.3) 0

Worsening acne 1 (8.3) 0

Irritable 1 (8.3) 0

Leg cramps 1 (8.3) 0

Incontinence 1 (8.3) 1 (10.0)

“Assertive, stubborn, more emotional” 1 (8.3) 0

Fainted 1 (8.3) 0

Fatigue 1 (8.3) 0

Shakiness 0 1 (10.0)

UTI, yeast infection 0 1 (10.0)

Increased bilirubin 0 1 (10.0)

Fever 0 2 (20.0)

Worsening alopecia 0 1 (10.0)

Frequent urination 0 1 (10.0)

Weight gain 0 1 (10.0)

Following the initial 10-week period of low dose study drug, two participants

could not have their dose increased to 4.5 mg due to an insufﬁcient supply of the

study drug. The study time was extended for these two participants in order for

them to have a dosage of 4.5 mg for 10 weeks.

1548 AMERICAN JOURNAL OF MEDICAL GENETICS PART A

TABLE III. Electrocardiogram (EKG) Findings for Participants Receiving Rivastigmine or Placebo

Rivastigmine Placebo

Group difference in

change from

screening-final

EKG measure

Screening

average value

(SD)

Final

average

value (SD) P-value



Screening

average value

(SD)

Final average

value

(SD) P-value



P-value

Ventricular rate (BPM) 76.38 (14.5) 73.6 (14.5) 0.25 73.14 (19.1) 72.14 (16.25) 0.75 0.64

P–R interval (ms) 137.0 (21.1) 133.6 (21.1) 0.12 138.46 (24.6) 135.00 (20.41) 0.53 0.62

QRS duration (ms) 88.0 (25.4) 89.2 (28.3) 0.41 85.57 (12.1) 85.07 (13.95) 0.74 0.71

QT interval (ms) 368.38 (73.8) 391.4 (38.4) 0.25 384.42 (39.1) 384.57 (35.77) 0.98 0.45

QTc interval (ms) 431.56 (27.6) 427.6 (19.4) 0.46 415.71 (24.1) 416.00 (31.09) 0.97 0.37

Individual EKG findings present at both screening and final visits

Participant Rivastigmine group

1 AV dual-paced rhythm (Pacemaker)

2 Sinus rhythm; ST elevation anterior leads; possible right ventricular hypertrophy; possible left ventricular hypertrophy

Participant Placebo group

1 Sinus rhythm; first degree AV block; left atrial abnormality; left ventricular hypertrophy

2 Normal sinus rhythm; left axis deviation; borderline prolonged QT interval or TU fusion

3 Sinus bradycardia; left axis deviation; right bundle branch block

No signiﬁcant differences were found on EKG measures between screening and ﬁnal visits for either group and no signiﬁcant differences were found between treatme nt groups.

Left axis deviation was found in two additional participants. Isolated sinus bradycardia was found in one participant.

Isolated mild sinus bradycardia was found in three additional participants.



Two sample

-test assuming equal variance.

TABLE IV. Baseline Performance on Cognitive Measures

Baseline average score (SD)

Outcome masure Rivastigmine Placebo P-value

VABS-II (survey interview form): Standard scores

Communication 67.55 (7.98) 67.20 (4.37) 0.84

Daily living skills 71.80 (17.88) 67.10 (9.39) 0.47

Socialization 74.50 (14.42) 72.00 (9.57) 0.65

Adaptive behavior composite 69.90 (13.60) 67.00 (6.75) 0.55

BRIEF-P: Raw scores

Global executive composite 107.55 (21.03) 100.10 (23.44) 0.45

REAL: Raw scores

Behavior 18.30 (3.56) 17.78 (4.29) 0.78

Executive function/attention 10.27 (3.58) 10.00 (3.97) 0.78

Total 61.10 (12.99) 63.89 (14.78) 0.67

CELF-P-II: Raw scores

Recalling sentences 11.83 (10.54) 6.10 (6.84) 0.16

Word classes expressive 7.75 (6.24) 6.40 (5.56) 0.6

Word classes receptive 13.67 (6.11) 11.0 (5.58) 0.3

Word casses total 21.42 (11.34) 17.4 (10.9) 0.41

NEPSY: Immediate memory for names 10.83 (6.18) 10.20 (5.09) 0.8

TOVER 22.5 (15.92) 12.8 (7.48) 0.09

BRIEF-P, behavior rating inventory of executive function-preschool; CELF-P-II, clinical evaluation of language fundamentals-preschool; REAL, rating of everyday activities and life skills; TOVER, test of verbal

expression and reasoning; VABS-II, Vineland adaptive behavior scales—second edition.

SPIRIDIGLIOZZI ET AL. 1549

The CELF-P-2 Recalling Sentences subtest is a measure of

expressive language where the child is asked to repeat sentences

of increasing length and complexity immediately after hearing

them. The Word Classes subtest requires the child to choose two

words that are related best and then verbalize how they go together.

The Total, Receptive, and Expressive raw scores for this subtest

were analyzed.

The TOVER is an expressive language measure designed to

assess an individual’s abilit y to ﬁnd solutions to questions from

pictured stimuli and to think about answers using visual informa-

tion. It was developed to assess change in response to treatment in

individuals with DS. The TOVER consists of 23 pictures and 64

related questions. The total score (range 0–64) was analyzed.

Analysis

Descriptive statistics were computed for all study variables of

interest (overall and stratiﬁed by treatment group). EKG measure-

ments of ventricular rate, QRS complex duration, QT interval, PR

interval, and QTc interval were also compared between the treat-

ment and placebo groups. Short-term efﬁcacy was determined by

comparing differences in language and cognitive performance

between baseline and week 20 visits across treatment and placebo

groups. Repeated measures ANOVA models were used to examine

pairwise changes between baseline and week 20 visits within and

between groups. Age and gender were included as covariates in the

analyses of changes between baseline and ﬁnal visits. Given

the exploratory nature of the study, we did not correct for multiple

comparisons.

RESULTS

Safety Measures

Adverse events. No SAEs were reported for any in dividuals in

the treat ment or placebo groups and no parti cipants needed to

terminate study participation after starting the study me dication.

Ther e were 11 AEs related or possibly related to the study

medication. A total of six AEs occur red in the ﬁrst 8 weeks of

the treatmen t (one at the 1.5 mg and ﬁve at the 3 mg levels) and ﬁve

occurred in the second 8 weeks of treatment at the 4.5 mg level.

None of the AEs was unexp ected and all were categorized as mild.

Of the 12 participants taking rivastig mine, 1 participant reported

no AEs and 8 partic ipants reported between one and ﬁve mild,

transient AEs considered to be potentially r elated to rivastigmine

effects, including diarrhea, nausea, vomiting, stomachache, inc-

ontinen ce, dec reased appetite, dizziness, weakness, head ache,

irritability, and “assertive, stubborn, and more emoti onal” beha v-

ior (Table II). The re maining three participants ta king rivastig-

mine reported AEs that were considered to be unrelated to the

study medication, including sleepiness, and diarrhea related to

viral i llness.

Electrocardiogram data. EKG ﬁndings ar e presented in

Table III. No group differences were found between the treatment

and placebo groups on EKG measures, including ventricular rate,

QRS complex duration, QT interval, PR interval, and QTc interval.

Within group s, no signiﬁcant change s between baseline and ﬁnalvisits

were foundon EKG measures.Three participants inthe placebo group

and two participants in the treatment group had abnormal EKG

ﬁndingsatbaselineandthesameﬁndingsattheﬁnalvisit.

TABLE V. Be tween-Group Differences on Cognitiv e Measures

Number of participants analyzed Mean difference final-baseline (SD)

Outcome Measure Rivastigmine Placebo Rivastigmine Placebo P-value

VABS-II (survey interview form): Standard scores

Communication 11 10 0.9 (3.3) 2.0 (3.6) 0.45

Daily living skills 10 10 2.3 (4.6) 2.8 (5.2) 0.03



Socialization 10 10 3.2 (5.0) 2.0 (6.1) 0.05

Adaptive behavior composite 10 10 1.7 (3.2) 2.0 (3.6) 0.03



BRIEF-P: Raw scores

Global executive composite 11 10 3.6 (7.7) 6.1 (12.0) 0.58

REAL: Raw scores

Behavior 10 9 0.2 (2.5) 0.9 (1.2) 0.25

Executive function/attention 11 9 0.2 (1.9) 1.1 (3.1) 0.24

Total 10 9 0.8 (7.1) 3.4 (8.3) 0.46

CELF-P-II: Raw scores

Recalling sentences 12 10 0.5 (4.1) 2.0 (1.9) 0.09

Word classes expressive 12 10 3.75 (4.25) 1.3 (3.23) 0.15

Word classes receptive 12 10 1.17 (3.66) 1.0 (3.89) 0.92

Word classes total 12 10 4.92 (7.06) 2.3 (6.31) 0.38

NEPSY: Immediate memory for names 12 10 2.25 (3.67) 1.9 (5.38) 0.86

TOVER 11 10 0.09 (6.07) 1.80 (3.39) 0.44

BRIEF-P, behavior rating inventory of executive function-preschool; CELF-P-II, clinical evaluation of langua ge fundamentals-preschool; REAL, rating of everyday activities and life skills; TOVER, test of verbal

expression and reasoning; VABS-II, Vineland adaptive behavior scales—second edition.



Signiﬁcant group difference at

< 0.05.

1550 AMERICAN JOURNAL OF MEDICAL GENETICS PART A

Efﬁcacy Measures

Performance on baseline efﬁcacy measures did not signiﬁcantly

differ between groups, although there was a trend toward higher

performance on the TOVER at baseline for the rivastigmine group

(Table IV). Group differences on performance change between

baseline and ﬁnal visits are presented in Table V. Three participants

receiving rivasti gmine were missing data for some measures. Two

of these participants did not complete all ﬁnal visit assessments and

the other participant’s data was excluded due to discrepant reports

FIG. 2. Differences in VABS-II scores between ﬁnal and baseline visits for individual study participants (numbered 1–20) for (A) Daily Living

scores and (B) Adaptive Behavior Composite scores. A rise in standard scores from baseline to the ﬁnal visit indicates improvement. [Color

ﬁgure can be seen in the online version of this article, available at http://wileyonlinelibrary.com/journal/ajmga].

SPIRIDIGLIOZZI ET AL. 1551

between parental ratings at the baseline and ﬁnal visits. Addition-

ally, one participant receiving placebo did not complete the REAL

at the baseline visit.

Signiﬁcant differences between the treatment and placebo

groups were found on the overall score for the Vineland

Adaptive Behavior Scale-II and one domain. Signiﬁcant group

differences were found on the standard score changes between

baseline and ﬁnal visits for the Adaptive Behavior Composite

score (P ¼ 0.03) and the Daily Living Skills domain (P ¼ 0.03).

Group differences on the standard score changes for the Soc-

ialization domain were of borderline signiﬁcance (P ¼ 0.05).

However, the patterns of these group differences on all three

VABS-II domain scores indicated relative improvem ents in

performance over time in the placebo group, while the treat-

ment group showed relative decreases in performance over

time. Figure 2 illustrates the differences in VABS-II scores

between ﬁnal and baseline visits across indiv idual participants.

No signiﬁcant group differences were found on the VABS-II

Communicationdomainstandardscoresorontheotherparent

report measures (REAL and BRI EF-P). Additionally, no signif-

icant group differences were found on the direct assessments of

the child’s performance, including t he NEPSY Immediate

Memory for Names subtest score, the CELF-P-2 subtest scores,

or the TOVER total s core.

Within-group analyses of the effects of time (baseline ver sus

ﬁnal visit) are presented in Table VI. Within-group analyses of time

effects indicated that the treatment group showed signiﬁcant

improvement between baseline and ﬁnal visits on the CELF-P-2

Word Classes Expressive Scores (P ¼ 0.01) and Total Scores

(P ¼ 0.03). Figure 3 illustrates the differences from baseline to

ﬁnal visits in CELF-P-2 Word Classes Expressive and Total scores

across individual participants. Differences between baseline and

ﬁnal visits also approached statistical signiﬁcance for the treatment

group on the NEPSY Memory for Names (P ¼ 0.06). In contrast,

only the placebo group showed improvement between the baseline

and ﬁnal visits on the CELF-P-2 Recalling Sentences subtest

(P ¼ 0.01). No signiﬁcant differences were found between baseline

and ﬁnal visits for either the treatment or placebo groups on any of

the VABS-II, BRIEF-P, REAL, or TOVER scores. Finally, no

signiﬁcant effects of age or gender were found for either group

on any outcome measures.

DISCUSSION

This double-blind, placebo-controlled clinical trial of rivastigmine

treatment in children and adolescents with DS did not show

signiﬁcant improvement in aspects of cognition, language, or

overall function. The results suggest that rivastigmine is not

associated with signiﬁcantly increased adverse events or increased

risk for cardiac complications compared to placebo.

The current s tudy’s ﬁndings build upo n p revious investiga-

tions of the efﬁcacy of ChEIs in DS. A previous open-label trial of

TABLE VI. Within-Group Differences Between Baseline and Final Visits on Cognitive Measures

Rivastigmine Placebo

Outcome measure

Baseline average

score (SD)

Final average

score (SD) P-value

Baseline average

score (SD)

Final average

score (SD) P-value

VABS-II (survey interview form):

Standard scores

Communication 67.55 (7.98) 68.45 (8.49) 0.38 67.20 (4.37) 69.20 (4.10) 0.11

Daily living skills 71.80 (17.88) 69.50 (15.17) 0.15 67.10 (9.39) 69.90 (11.0) 0.12

Socialization 74.50 (14.42) 71.3 (12.30) 0.07 72.00 (9.57) 74.00 (6.04) 0.32

Adaptive behavior composite 69.90 (13.60) 68.20 (11.71) 0.13 67.00 (6.75) 69.00 (5.85) 0.11

BRIEF-P: Raw scores

Global executive composite 107.55 (21.03) 103.91 (17.72) 0.15 100.10 (23.44) 94.00 (20.87) 0.14

REAL: Raw scores

Behavior 18.30 (3.56) 18.10 (2.77) 0.80 17.78 (4.29) 18.70 (3.57) 0.05

Executive function/attention 10.27 (3.58) 10.09 (2.39) 0.76 10.00 (3.97) 11.10 (4.31) 0.31

Total 61.10 (12.99) 61.90 (10.50) 0.73 63.89 (14.78) 67.30 (12.85) 0.25

CELF-P-II: Raw scores

Recalling sentences 11.83 (10.54) 11.33 (9.33) 0.68 6.10 (6.84) 8.10 (6.54) 0.01



Word classes expressive 7.75 (6.24) 11.5 (6.52) 0.01



6.40 (5.56) 7.7 (4.32) 0.24

Word classes receptive 13.67 (6.11) 14.83 (7.13) 0.29 11.0 (5.58) 12.0 (5.01) 0.44

Word classes total 21.42 (11.34) 26.3 (13.3) 0.03



17.4 (10.9) 19.7 (9.12) 0.28

NEPSY: Immediate memory for

names

10.83 (6.18) 13.08 (6.96) 0.06 10.20 (5.09) 12.10 (4.48) 0.29

TOVER 22.5 (15.92) 24.45 (15.12) 0.96 12.8 (7.48) 14.6 (7.83) 0.13

BRIEF-P, behavior rating inventory of executive function-preschool; CELF-P-II, clinical evaluation of langua ge fundamentals-preschool; REAL, rating of everyday activities and life skills; TOVER, test of verbal

expression and reasoning; VABS-II, Vineland adaptive behavior scales—second edition.



Signiﬁcant effect of visit at

< 0.05.

1552 AMERICAN JOURNAL OF MEDICAL GENETICS PART A

ChEIs in children with DS [Heller et al., 2004] indicated that

donepezil was a ssociated with signiﬁcant improvements in exp-

ressive and recepti ve language. Similarly, an inve stigation

of riv astigmi ne efﬁcac y in adolescent s with DS indi cated

improvements in overall adaptive function, attention, memory,

and language [Heller et al ., 2006b]. However, these studies were

limited by sample size (with totals of 7 children and 11 adolescent

partici pants, respectively ) and by the lack of a placebo-c ontrolled

comparison group. In the current study, there was a great deal of

variability in the performance of participants who were treated

with rivastigmine. Although the treatment group showed signiﬁ-

cant improvement on the measures emphasizing expressive

language and short-term v isual memory, this change was not

statistically signiﬁca ntly different from the pl acebo g roup. Still,

FIG. 3. Differences in CELF-P-2 scores between ﬁnal and baseline visits for individual study participants (numbered 1–22) for (A) Word

Classes Expressive scores and (B) Word Classes Total scores. A rise in raw scores from baseline to the ﬁnal visit indicates improvement.

[Color ﬁgure can be seen in the online version of this article, available at http://wileyonlinelibrary.com/journal/ajmga].

SPIRIDIGLIOZZI ET AL. 1553

these measures targeting the participant s’ expressive language

skil ls contin ue to hold promise as a way to detect change in la rger

medication trials, as it is possible that only a subset of individuals

with DS show improvements in expressive language w ith riva-

stigmine treatment.

The results of th e current study were consistent with a long-

term follow-up study of rivastigmine efﬁcacy in adolesce nts with

DS [Heller et al., 2010] and a double -blind placebo-controlled

study of donepezil in adolescents with DS [Kishnani et al., 2010].

Neither stud y demonstrated signiﬁcant changes in cognitive or

language performance related to the study medication. Wh ile the

study by Heller et al. [2010] su gges ted the potential for a smal l

subset of individual responders to the med icat ion fo r adaptive

behavior, the data from the current study did not provide

evidence for a larger subset of individual responders in the

treatment group compared to the placebo group. Furthermo re,

perf ormance in the treatment group appear ed to decline on

several VABS-II domains from the baseline to ﬁnal visits, raising

the question of whether a subset of individuals may have had a

detrimental response to rivastigmine. However, thes e pe rfor-

mance differences between visits did not reach statistical signiﬁ-

cance and no individual score changes on the VABS-II were

within a clinically meaningful range. Larger studies may be

needed to address whether a subset of individuals c ould show

clinical ly relevan t changes on cognitive measures related to the

participant’s overall intelligence qu otient (IQ) or other variables,

such as genetic or epigenetic factors, that may modify the

respons e to ChEIs.

While cholinergic deﬁcits have been established to be present

in adults with DS, the timing of the onset of these deﬁcits is less

clear. Data from animal models have suggested that cholinergic

function related to projections from the basal forebrain is

critical for cortical development and establishment of neural

circuits that underlie complex cognitive functions in adulthood

[Berger-Sweeneyand Hohmann, 1997; Berger-Sweeney, 1998].

It is unclear, however, whether the inﬂuence of the cholinergic

basal forebrain system is similar during development and

adulthood [Berger-Sweeney, 2003]. It will be important for

future research to aim to gain an understanding of how the

developing cholinergic system is affected in D S. Additionally, i t

will be critical to charac terize the roles of other modulatory

systems such as glutamate, which has been im plicated in le arn-

ing and memory difﬁculties i n DS [Costa, 2014]. In order t o

develop cognitive interventions that target these systems, fur-

ther research is necessary to characterize the pathogenic roles of

these neurotransmitter systems across different stages of brain

development in DS.

Ther e are multiple challenges inherent in conducting clinical

trials in children with DS. O ne important challenge is determin-

ing appropriate cognit ive testing measures that will be sensitive

to medication effects. DS is as soci ated with changes in a broad

range of cognitive functions, including ad aptive function, exp-

ressive and receptive langua ge, attention, memory, and executive

function. When selecting study measures to investigat e efﬁcacy of

treatment, there is often a trade-off between investigating multi-

ple cognitive domains and obtaining sufﬁcient statist ical power

to demonstrate signiﬁcant effects [Heller et al., 2006a]. The

current s tudy highl ights the current un met need to deﬁne a

set of the most appropriate cognitive measu res that are reliable,

valid, and sensitive to treatment effects. It is also critical to

minimize the variability within each part icipant’s performance

at each time point.

A limitation of the current study is that it may not be adequately

powered to detect small changes in cognitive performance related

to rivastigmine effects. It may be that many hundreds of partic-

ipants are required to detec t more subtle changes in cognitive

functioning in children with DS related to rivastigmine. Larger

studies would be necessary to investigate factors such as dose and

titration tolerance, especially when wide variability in these factors

may be observed [Heller et al., 2006a]. Additionally, the current

study did not investigate the potential role of factors such as

educational or other therapeutic interventions that could inﬂuence

change in cognitive functioning. Future studies may beneﬁt from

including these types of factors as covariates to distinguish poten-

tial medication effects.

This study also highlights the challenges of conducting clinical

trials using liquid rivastigmine. The liquid form of rivastigmine is

helpful for children who may have difﬁculty swallowing pills and

allows for ﬂexibility in dosing and titration [Heller et al., 2006a].

However, we experienced challenges related to a shortage of the

study medication mid-way through the study that required two

participants to follow a modiﬁed protocol and extended study

time. Partnering with the pharmaceutical industry is necessary to

ensure adequate supply of study drug, as well for attaining the

necessary study funding and sample sizes that are required for

conducting clinical trials for enhancement of cognitive function in

DS [Heller et al., 2006a].

CONCLUSIONS

The current study did not demonstrate evidence for efﬁcacy of

rivastigmine in improving cognitive function in children with DS.

However, our results suggested that rivastigmine does appear to be

safe and well-tolerated for children and adolescents with DS.

Future investigations with larger samples of participants may be

needed to deﬁne a proﬁle of speciﬁc cognitive measures that are

reliable and sensitive to treatment effects for use in clinica l trials for

children with DS.

ACKNOWLEDGMENT

This article is dedicated to James H. Heller, whose inspiration and

passion for this research continues to live on despite his untimely

death in 2014.

REFERENCES

Beccaria L, Marziani E, Manzoni P, Arvat E, Valetto MR, Gianotti L, Ghigo

E, Chiumello G. 1998. Further evidence of cholinergic impairment of the

neuroendocrine control of the GH secretion in Down’s syndrome.

Dement Geriatr Cogn Disord 9:78–81.

Becker L, Mito T, Takashima S, Onodera K. 1991. Growth and

development of the brain in Down syndrome. Prog Clin Biol Res

373:133–152.

1554 AMERICAN JOURNAL OF MEDICAL GENETICS PART A

Berger-Sweeney J. 1998. The effects of neonatal basal forebrain lesions on

cognition: Towards understanding the developmental role of the cho-

linergic basal forebrain. Int J Dev Neurosci 16:603–612.

Berger-Sweeney J. 2003. The cholinergic basal forebrain system during

development and its inﬂuence on cognitive processes: Important ques-

tions and potential answers. Neurosci Biobehav Rev 27:401–411.

Berger-Sweeney J, Hohmann CF. 1997. Behavioral consequences of

abnormal cortical development: Insights into developmental disabilities.

Behav Brain Res 86:121–142.

Brownell R. 2000. Expressive one-word picture vocabulary test

(EOWPVT). Ann Arbor, MI: Academic Therapy Publications.

Casanova MF, Walker LC, Whitehouse PJ, Price DL. 1985. Abnormalities

of the nucleus basalis in Down’s syndrome. Ann Neurol 18:310–313.

Costa AC. 2014. The glutamatergic hypothesis for Down syndrome: The

potential use of N-methyl-D-aspartate receptor antagonists to enhance

cognition and decelerate neurodegeneration. CNS Neurol Disord Drug

Targets 13:16–25.

Florez J, del Arco C, Gonzalez A, Pascual J, Pazos A. 1990. Autoradio-

graphic studies of neurotransmitter receptors in the brain of newborn

infants with Down syndrome. Am J Med Genet Suppl 7:301–305.

Gioia GA, Espy KA, Isquith PK. 2002. Behavior rating inventory of

executive function, preschool version (BRIEF-P). Odessa, FL: Psycho-

logical Assessment Resources.

Heller JH, Spiridigliozzi GA, Crissman BG, McKillop JA, Yamamoto H,

Kishnani PS. 2010. Safety and efﬁcacy of rivastigmine in adolescents with

Down syndrome: Long-term follow-up. J Child Adolesc Psychophar-

macol 20:517–520.

Heller JH, Spiridigliozzi GA, Crissman BG, Sullivan-Saarela JA, Li JS,

Kishnani PS. 2006a. Clinical trials in children with Down syndrome:

Issues from a cognitive research perspective. Am J Med Genet Part C

Semin Med Genet 142C:187–195.

Heller JH, Spiridigliozzi GA, Crissman BG, Sullivan JA, Eells RL, Li JS,

Doraiswamy PM, Krishnan KR, Kishnani PS. 2006b. Safety and efﬁcacy

of rivastigmine in adolescents with Down syndrome: A preliminary

20-week, open-label study. J Child Adolesc Psychopharmacol 16:

755–765.

Heller JH, Spiridigliozzi GA, Doraiswamy PM, Sullivan JA, Crissman BG,

Kishnani PS. 2004. Donepezil effects on language in children with Down

syndrome: Results of the ﬁrst 22-week pilot clinical trial. Am J Med Genet

Part A 130A:325–326.

Heller JH, Spiridigliozzi GA, Kishnani PS. 2000. TOVER: Test of verbal

expression and reasoning. Durham, NC: Duke University Medical Center.

Heller JH, Spiridigliozzi GA, Sullivan JA, Doraiswamy PM, Krishnan RR,

Kishnani PS. 2003. Donepezil for the treatment of language deﬁcits in

adults with Down syndrome: A preliminary 24-week open trial. Am J

Med Genet Part A 116A:111–116.

Johnson N, Fahey C, Chicoine B, Chong G, Gitelman D. 2003. Effects of

donepezil on cognitive functioning in Down syndrome. Am J Ment

Retard 108:367–372.

Kishnani PS, Heller JH, Spiridigliozzi GA, Lott I, Escobar L, Richardson S,

Zhang R, McRae T. 2010. Donepezil for treatment of cognitive dysfunc-

tion in children with Down syndrome aged 10–17. Am J Med Genet Part

A 152A:3028–3035.

Kishnani PS, Sommer BR, Handen BL, Seltzer B, Capone GT, Spiridigliozzi

GA, Heller JH, Richardson S, McRae T. 2009. The efﬁcacy, safety, and

tolerability of donepezil for the treatment of young adults with Down

syndrome. Am J Med Genet Part A 149A:1641–1654.

Kishnani PS, Sullivan JA, Walter BK, Spiridigliozzi GA, Doraiswamy PM,

Krishnan KR. 1999. Cholinergic therapy for Down’s syndrome. Lancet

353:1064–1065.

Kondoh T, Kanno A, Itoh H, Nakashima M, Honda R, Kojima M,

Noguchi M, Nakane H, Nozaki H, Sasaki H, Nagai T, Kosaki R, Kakee

N, Okuyama T, Fukuda M, Ikeda M, Shibata Y, Moriuchi H. 2011.

Donepezil signiﬁcantly improves abilities in daily lives of female Down

syndrome patients with severe cognitive impairment: A 24-week

randomized, double-blind, placebo-controlled trial. Int J Psychiatry

Med 41:71–89.

Korkman M, Kirk U, Kemp S. 1998. NEPSY: A developmental neuropsy-

chological assessment, manual. San Antonio, TX: The Psychological

Corporation.

Sacks B, Smith S. 1989. People with Down’s syndrome can be distinguished

on the basis of cholinergic dysfunction. J Neurol Neurosurg Psychiatry

52:1294–1295.

Semel E, Wiig EH, Secord WA. 2004. Clinical evaluation of language

fundamentals-preschool-2. San Antonio, TX: Harcourt Assessment.

Sparrow SS, Cicchetti DV, Balla DA. 2005. Vineland adaptive behavior

scales. Survey forms manual, 2nd edition. Minneapolis, MN: NCS

Pearson, Inc.

Spiridigliozzi GA, Heller JH, Crissman BG, Sullivan-Saarela JA, Eells R,

Dawson D, Li J, Kishnani PS. 2007. Preliminary study of the safety and

efﬁcacy of donepezil hydrochloride in children with Down syndrome: A

clinical report series. Am J Med Genet Part A 143A:1408–1413.

Spiridigliozzi GA, Heller JH, Kishnani PS. 2010. Rating of everyday

activities and life skills. Durham, NC: Duke University Medical Center.

SPIRIDIGLIOZZI ET AL. 1555