Tolerability and Safety of Souvenaid in Patients with Mild Alzheimer’s Disease: Results of Multi-Center, 24-Week, Open-Label Extension Study

Journal of Alzheimer’s Disease 44 (2015) 471–480
DOI 10.3233/JAD-141305
IOS Press
471
Tolerability and Safety of Souvenaid in
Patients with Mild Alzheimer’s Disease:
Results of Multi-Center, 24-Week,
Open-Label Extension Study
Marcel G.M. Olde Rikkerta,∗
, Frans R. Verheyb
, Rafael Blesac
, Christine A.F. von Arnimd
,
Anke Bongerse
, John Harrisonf
, John Sijbene
, Elio Scarpinig
, Maurits F.J. Vandewoudeh
, Bruno Vellasi
,
Renger Witkampj
, Patrick J.G.H. Kamphuise
and Philip Scheltensk
aRadboud Alzheimer Centre, Department of Geriatric Medicine, Radboud University Hospital, Nijmegen,
The Netherlands
bAlzheimer Centre Limburg, Maastricht University Medical Centre, Maastricht, The Netherlands
cDepartment of Neurology, Hospital de la Sta Creu i St. Pau, Barcelona, Spain
dDepartment of Neurology, Ulm University, Ulm, Germany
eNutricia Advanced Medical Nutrition, Nutricia Research, Utrecht, The Netherlands
fMetis Cognition Ltd, Kilmington, UK & Imperial College, London, UK
gUniversity of Milan, Fondazione Cà Granda, IRCCS Ospedale Policlinico, Milan, Italy
hUniversity Department of Geriatrics, Antwerp Hospital Network (ZNA St Elisabeth), University of Antwerp,
Belgium
iGerontopole, INSERM U 1027, Toulouse, France
jDivision of Human Nutrition, Wageningen University, Wageningen, The Netherlands
kAlzheimer Center, VU University Medical Center, Amsterdam, The Netherlands
Handling Associate Editor: Thomas Shea
Accepted 2 September 2014
Abstract.
Background: The medical food Souvenaid, containing the specific nutrient combination Fortasyn Connect, is designed to
improve synapse formation and function in patients with Alzheimer’s disease (AD). Two double-blind randomized controlled
trials (RCT) with Souvenaid of 12 and 24 week duration (Souvenir I and Souvenir II) showed that memory performance was
improved in drug-naïve mild AD patients, whereas no effects on cognition were observed in a 24-week RCT (S-Connect) in
mild to moderate AD patients using AD medication. Souvenaid was well-tolerated in all RCTs.
Objective: In this 24-week open-label extension (OLE) study to the 24-week Souvenir II RCT, long-term safety and intake
adherence of the medical food Souvenaid was evaluated.
Methods: Patients with mild AD (n = 201) received Souvenaid once-daily during the OLE. Main outcome parameters were
safety and product intake adherence. The memory domain z-score from a revised neuropsychological test battery was continued
as exploratory parameter.
Results: Compared to the RCT, a similar (low) incidence and type of adverse events was observed, being mainly (68.3%) of mild
intensity. Pooled data (RCT and OLE) showed that 48-week use of Souvenaid was well tolerated with high intake adherence
(96.1%). Furthermore, a significant increase in the exploratory memory outcome was observed in both the active-active and
control-active groups during Souvenaid intervention.
∗Correspondence to: Marcel Olde Rikkert, MD, PhD, Radboud
Alzheimer Center, Department of Geriatric Medicine, Rad-
boud University Hospital, P.O. Box 9101, 6500 HB Nijmegen,
The Netherlands. Tel.: +31 24 361 67 72; Fax: +31 24 361 74 08;
E-mail: M.Olde-Rikkert@ger.umcn.nl.
ISSN 1387-2877/15/$27.50 © 2015 – IOS Press and the authors. All rights reserved
This article is published online with Open Access and distributed under the terms of the Creative Commons Attribution Non-Commercial License.

472 M.G.M. Olde Rikkert et al. / Tolerability and Safety of Souvenaid in Mild AD
Conclusion: Souvenaid use for up to 48-weeks was well tolerated with a favorable safety profile and high intake adherence. The
findingsinthisOLEstudywarrantfurtherinvestigationtowardthelong-termsafetyandefficacyofSouvenaidinawell-controlled,
double-blind RCT.
Keywords: Alzheimer’s disease, clinical trial, dietary management, intervention studies, long-term, medical nutrition therapy,
memory, patient adherence, safety
INTRODUCTION
Progressive synaptic dysfunction and synapse loss
are key hallmarks of Alzheimer’s disease (AD) [1–3].
They are present from the very early stages of the dis-
ease, even before the emergence of clinical symptoms
[4], and strongly correlate with cognitive deterioration
[5]. As such, synaptic dysfunction is a useful target for
interventions in AD.
In animal models, dietary enrichment of nutrients
that are precursors of synaptic membrane phospho-
lipids can enhance phosphatide synthesis [6], the
number of dendritic spines, and the levels of pre- or
post-synaptic proteins [7], which are all prerequisites
for new synapse formation. Further, preclinical studies
showed that administration of specific combinations
of the phospholipid synthesis-promoting nutrients can
enhance neurotransmission and cognitive function
[8–11], which are indicative of improved synaptic
function. These and other observations, together with
the notion that the circulating levels of these nutrients
are lower in AD compared with healthy controls [12,
13], led to the creation of the specific nutrient combina-
tion Fortasyn® Connect. This comprises the precursors
and cofactors for the synthesis of neuronal membranes,
and is designed to support synapse formation and func-
tion in patients with AD [14]. In order to test the
hypothesis that administering these nutrients enhances
synaptic function and ameliorates cognitive symptoms
in AD, the medical food Souvenaid® (Nutricia N.V.,
Zoetermeer, The Netherlands), which contains Forta-
syn Connect, was developed.
Memory loss is an important clinical manifestation
in early AD [15, 16], and may be the result of synap-
tic loss in specific areas of the brain [17]. Memory
performance was therefore selected as a (co-)primary
outcome in two double-blind randomized controlled
trials (RCTs) with Souvenaid in drug-naïve patients
with mild AD (‘Souvenir I’ and ‘Souvenir II’). These
studies both showed that Souvenaid statistically signif-
icantly improved memory performance, which may be
related to disease specific nutrient requirements [18,
19]. In addition, cognitive performance was selected
as primary outcome in a double-blind RCT in patients
withmild-to-moderateADreceivingstandard-careAD
medication (‘S-Connect), showing that 24-week use
of Souvenaid did not slow cognitive decline in these
patients [20].
It was postulated that the nutrient levels provided
by once-daily use of Souvenaid are within a safe range
of intake, and that Souvenaid therefore would be well
tolerated. This was supported by the results of Souvenir
I and Souvenir II [18, 19] as well as the add-on S-
Connect study [20]. In addition, the 24-week product
intake adherence was high in all these studies (≥94%)
[18–20].
While these data provide evidence that Souvenaid
is well tolerated, the intervention period of the studies
presented so far is limited to 24 weeks. The study pre-
sented here is a 24-week open-label extension (OLE)
to the Souvenir II RCT, designed to evaluate the long-
term safety and compliance to Souvenaid in patients
with mild AD. In addition, the study explored the long-
term impact of Souvenaid on memory performance in
an uncontrolled design.
MATERIAL AND METHODS
Study setting
Twenty-six out of the 27 AD centers that recruited
participants for the RCT also recruited participants
for the OLE study. In the center that did not include
study participants, none of the three potentially eli-
gible patients were willing or able to continue in the
OLE study. The ethics committee of each participat-
ing center reviewed and approved the protocol. The
study was conducted in accordance with the Declara-
tion of Helsinki and the International Conference on
Harmonisation of Good Clinical Practice guidelines
as appropriate to nutritional products and legislation
of the country in which the research was conducted.
Participants
The patient population of the multi-center, multi-
country Souvenir II RCT has been reported previously
[19]. Briefly, major eligibility criteria included: 1)
being drug-free for AD medication; 2) a probable
diagnosis of AD (National Institute of Neurological

M.G.M. Olde Rikkert et al. / Tolerability and Safety of Souvenaid in Mild AD 473
DisordersandStroke-Alzheimer’sDiseaseandRelated
Disorders Association [21]); and 3) mild AD as defined
by a Mini-Mental State Examination (MMSE) [22]
score of ≥20. Following completion of the RCT,
patients willing to continue into the OLE study were
required to have a responsible caregiver available, and
both had to reconfirm their written informed consent.
Exclusion criteria included the use of other investiga-
tional products, alcohol or drug abuse in the opinion of
the investigator, or the investigator’s uncertainty about
the willingness, ability, or medical status of the patient
to comply with the protocol. Eligibility criteria for the
OLE allowed patients to use nutritional supplements
and/or AD medication, thereby reflecting a more real-
life clinical setting.
Study design
All patients originally randomly assigned to active
(Souvenaid) or control product who completed the
24-week, double-blind RCT (Dutch Trial Registra-
tion number NTR1975) were eligible to participate
in the 24-week OLE study (Dutch Trial Registra-
tion number NTR2571). In the OLE, all patients
received the active product once-daily in the form
of a 125 ml drink (vanilla or strawberry flavor). The
active product contains the specific nutrient combi-
nation Fortasyn Connect, comprising 625 mg uridine
monophosphate, 400 mg choline, 106 mg phospho-
lipids, 1200 mg docosahexaenoic acid (DHA), 300 mg
eicosapentaenoic acid (EPA), 40 mg vitamin E (␣-
tocopherol equivalents), 80 mg vitamin C, 60 ␮g
selenium, 1 mg vitamin B6, 3 ␮g vitamin B12, and
400 ␮g folic acid) per 125 ml, in addition to other vita-
mins, minerals, trace elements, and macronutrients.
Study staff, patients, and caregivers remained blinded
to the patient’s initial randomized study group alloca-
tion throughout the 48 weeks. The screening/baseline
visit for the OLE study was combined with the final
visit (Week 24) of the RCT. Study parameters were
assessed at the screening/baseline visit, at Week 12
(presented as Week 36), and 24 (presented as Week
48). Adverse events (AEs) and changes in concomitant
medication and nutritional supplements were recorded
every six weeks. Recording of product intake was per-
formed daily.
Outcome measures
The main outcome measures of the OLE study
were safety and product intake adherence. Safety
assessments included the recording of (serious) AEs
([S]AEs), and the monitoring of vital signs and lab-
oratory parameters. Product intake adherence was
measured by studying caregiver-supervised patient-
reported daily diaries, in which was recorded whether
the patient did or did not take the study product on each
particular day. A product evaluation questionnaire (7-
point Likert scale; 1 = “dislike it very much”, 7 = “like
it very much”) was also completed by the patient to
assess the appreciation of the taste and convenience of
the product.
At the start of the OLE study, a new MMSE assess-
ment was performed for the characterization of the
study population. All other patient characteristics were
derived from the Souvenir II RCT. The primary out-
come of the RCT, the memory domain composite
z-score, was assessed as exploratory outcome in the
OLE. This domain is based on the Neuropsychological
Test Battery (NTB) [23] and consists of Rey Auditory
Verbal Learning Test immediate recall, delayed recall
and recognition performance, and Wechsler Mem-
ory Scale-revised verbal paired associates immediate
and delayed recall. In addition, the Disability Assess-
ment for Dementia scale [24] was an exploratory
outcome for functional ability. Study staff perform-
ing the assessments received appropriate training on
outcome assessments via a study specific test man-
ual. In addition, local rater trainings were provided
for the assessment of the neuropsychological tests.
Other parameters included specific nutritional blood
markers (plasma vitamin E, erythrocyte DHA and
EPA, and plasma homocysteine). Venous blood sam-
ples were taken, processed, and stored in a −80◦C
freezer until batch shipped on dry ice. After extract-
ing lipids from erythrocyte membranes, the fatty acid
profile in erythrocyte membranes was assayed by gas
chromatography. Plasma vitamin E levels were mea-
sured using high performance liquid chromatography
to determine content of alpha-tocopherol. Homocys-
teine levels (free and protein-bound) were measured
using high performance liquid chromatography with
fluorescence detection after preparing a derivate [25,
26].
Statistical analyses
Analyses were performed on all available data from
patients participating in the OLE study. OLE data are
presented according to the intervention received dur-
ing the double-blind study period (control-active and
active-active). The following (differences in) effects
on the exploratory NTB memory domain scores were
statistically tested: a) between-group comparisons

including control-active group versus active-active
group in the total 48 week and 24–48 week intervals
and the comparison of the baseline-24 week interval in
the active-active group versus the 24–48 week interval
in the control-active group, and b) within-group com-
parisons including within the control-active group the
comparison of the active intervention (week 24–48)
versus control intervention (baseline-week 24), within
the active-active group the comparison of the baseline-
week 24 interval versus the week 24–48 interval and
both within the control-active and active-active groups
the change from 24 to 48 weeks. In addition, two
exploratory key groups of interest were deﬁned for
the analysis of effects on the NTB memory domain
score: 1) patients who did not start using AD medica-
tion during the OLE study, and 2) patients with a large
study gap. See supplementary material for detailed
information.
Comparisons were statistically tested using t-tests
and/or by using a mixed model for repeated measures
(MMRM) in which time was treated as a categorical
variable and represented by dummies. Correction for
multiple testing was not applied due to the explorative
nature of this study, the small sample size, and the high
correlation between outcome parameters.
Statistical analyses were performed using SAS®
statistical software (SAS Enterprise Guide 4.3 for Win-
dows, SAS Institute Inc., Cary, NC, USA).
RESULTS
Patient demographics
A total of 259 patients were originally random-
ized to receive either active or control product for 24
weeks in the double-blind RCT. Of the 238 patients
(92%) who completed the RCT, 201 patients (84.5%)
entered the OLE study between July 2010 and October
2011 (Fig. 1). The overall discontinuation rate during
the OLE study was 7.5% (15 patients). Seven (3.5%)
patients discontinued due to (S)AEs: 6 patients in
the control-active group (enlarged abdomen, bronchi-
tis, avascular bone necrosis, impaired concentration,
malignant lymphoma, and diarrhea), and one patient in
the active-active group (diverticula). Of these (S)AEs,
only diarrhea was considered to be related to the use
of the study product. Additional reasons for discon-
tinuation included withdrawal of informed consent (2
patients), loss to follow-up (3 patients), and ‘other’ (3
patients).
Baseline characteristics for the study populations
enrolled in the double-blind RCT and in the OLE
study are summarized in Table 1. The mean age of
the patients in the OLE study was 74.2 (±7.6) years,
51% of patients were male and the mean MMSE
score was 25.1 (±3.3), representing a population with
very mild AD. Baseline characteristics of patients who
Fig. 1. Subject ﬂow.

Table 1
Baseline characteristics of the Souvenir II RCT population and for
patients (not) continuing in the OLE
Control(-Active) Active(-Active)
Souvenir II RCT
intent-to-treat population
n 129 130
Male, n (%) 64 (49.6) 68 (52.3)
Age, years 73.2 (8.4) 74.4 (6.9)
BMI, kg/m2 26.7 (4.2) 26.1 (4.1)
Years of education beyond
primary school
6.6 (4.6) 6.5 (4.8)
Duration of AD since
diagnosis, months,
median [range]
2.0 [0.0–88.0] 1.0 [0.0–70.0]
ApoE ␧4 carrier, n (%)
No 58 (49.2) 62 (51.2)
Yes 60 (50.8) 59 (48.8)
Unknown 11 9
Total MMSE score 25.0 (2.8) 24.9 (2.9)
Souvenir II RCT completers
enrolled in OLE
n 104 97
Male, n (%) 52 (50.0) 51 (52.6)
Age, years † 73.9 (8.3) 74.5 (6.8)
BMI, kg/m2 † 27.3 (4.2) 26.9 (4.2)
primary school
6.7 (4.7) 6.2 (4.8)
diagnosis, months,
median [range]
2.0 [0.0–88.0] 1.0 [0.0–70.0]
No 46 (48.9) 41 (44.1)
Yes 48 (51.1) 52 (55.9)
Unknown 10 4
Total MMSE score – Start
RCT
25.1 (2.9) 25.0 (2.8)
Total MMSE score – Start
OLE
25.1 (3.4) 25.1 (3.3)
Souvenir II RCT completers
not enrolled in OLE
n 16 21
Male, n (%) 6 (37.5) 10 (47.6)
Age, years † 72.1 (8.7) 76.7 (5.8)
BMI, kg/m2 † 27.8 (4.5) 25.5 (4.1)
primary school
5.9 (4.6) 7.6 (4.7)
diagnosis, months,
median [range]
2.5 [0.0–13.0] 0.0 [0.0–25.0]
No 8 (50.0) 15 (83.3)
Yes 8 (50.0) 3 (16.7)
Unknown 0 3
Total MMSE score 25.4 (2.5) 24.3 (2.9)
OLE, open-label extension; RCT, randomized controlled trial; AD,
Alzheimer’s disease; ApoE, Apolipoprotein E; BMI, body mass
index;MMSE,Mini-MentalStateExamination.Dataaremean(stan-
dard deviation) unless stated otherwise. †At week 24 of the Souvenir
II RCT / start of the OLE study.
completed the RCT but did not enter the OLE are also
provided in Table 1. For each of these populations,
baseline characteristics were not significantly differ-
ent between the control(-active) and active(-active)
groups. In addition, there were no significant differ-
ences in baseline characteristics between the RCT and
OLE study populations.
Forty-five patients (22.4%) began taking AD med-
ication after starting intervention in the OLE study:
9 patients (8.7%) received donepezil in the control-
active group compared with 18 patients (18.6%) in
the active-active group (p = 0.061, Fisher’s Exact test);
2 (1.9%) received galantamine in the control-active
group compared with 4 (4.1%) in the active-active
group (p = 0.432, Fisher’s Exact test); and 6 (5.8%)
received rivastigmine in the control-active group com-
pared with 6 (6.2%) in active-active group (p = 1.000,
Fisher’s Exact test). Ten patients (9.6%) in the control-
active group and 6 patients (6.2%) in the active-active
group started using AD medication after completion
of the OLE study (p = 0.440, Fisher’s Exact test).
There were no significant or relevant between-group
differences in the initiation of other medications or
nutritional supplements during the OLE study.
Outcome parameters
Safety
Table 2 summarizes the number and proportion of
patients experiencing one or more AEs in the RCT
and OLE. In total, 105 patients (52.2%) experienced
at least one AE during the OLE study; including 57
patients (54.8%) with 115 AEs in the control-active
group and 48 patients (49.5%) with 74 AEs in the
active-active group. The AE profile in the OLE study
was comparable to the RCT, except for the occurrence
of any AEs related to the gastrointestinal system and
respiratory system, which was significantly lower in
the OLE study compared to the RCT (p = 0.005 and
p = 0.013, respectively, Fisher’s Exact test). There was
no significant or relevant difference in the proportion of
patients experiencing at least one AE in the subgroup
of patients who started using AD medication during
the OLE study (22 patients [48.9%]) compared with
patients who remained AD drug-naïve (83 patients
[53.2%]).
The most frequent AEs in the OLE study included
dizziness/vertigo (4.0%), surgical intervention (2.5%),
diarrhea (2.5%), hypercholesterolemia (2.0%), hyper-
tension (2.0%), back pain (2.0%), headache (2.0%),
andanginapectoris(2.0%).MostAEsintheOLEstudy
were of mild (68.3%) or moderate (22.2%) intensity,

Table 2
Number (%) of patients experiencing one or more AEs, by affected body system†
Body system (example AEs) Control Control-Active Active Active-Active
(n = 129) (n = 104) (n = 129) (n = 97)
RCT OLE RCT OLE
Body as a whole 20 (15.5%) 13 (12.5%) 11 (8.5%) 9 (9.3%)
(back pain, syncope, fatigue)
Central and peripheral nervous system disorders 18 (14.0%) 7 (6.7%) 11 (8.5%) 9 (9.3%)
(dizziness/vertigo, headache)
Gastrointestinal system disorders 30 (23.3%) 10 (9.6%) 22 (17.1%) 10 (10.3%)
(diarrhea, dyspepsia, constipation, flatulence)
Metabolic and nutritional disorders 9 (7.0%) 3 (2.9%) 13 (10.1%) 6 (6.2%)
(hypercholesterolemia, weight decrease, weight increase)
Musculo-skeletal system disorders 9 (7.0%) 7 (6.7%) 10 (7.8%) 5 (5.2%)
(fracture, myalgia, skeletal pain)
Psychiatric disorders 16 (12.4%) 14 (13.5%) 15 (11.6%) 7 (7.2%)
(aggressive reaction, anorexia, anxiety, depression)
Respiratory system disorders 15 (11.6%) 6 (5.8%) 10 (7.8%) 1 (1.0%)
(bronchitis, pharyngitis)
Other 8 (6.2%) 7 (6.7%) 8 (6.2%) 3 (3.1%)
(fall, surgical intervention)
OLE, open-label extension; RCT, randomized controlled trial; AE, adverse event. †Only affected body systems reported by at least 5% of patients
in either group of the OLE population are shown. Adverse events occurring in less than 5% of patients in either group were: application site
disorders, cardiovascular disorders, hearing and vestibular disorders, heart rate and rhythm disorders, liver and biliary disorders, myo-, endo-,
pericardial and valve disorders, neoplasm, platelet, bleeding and clotting disorders, red blood cell disorders, reproductive disorders (male),
resistance mechanism disorders, skin and appendages disorders, urinary system disorders, vascular (extracardiac) disorders, vision disorders.
with no clinically relevant differences between the
control-active group and the active-active group.
Twelve (10.4%) of the AEs were rated as severe in the
control-active group compared with 5 (6.8%) in the
active-active group, of which none were considered to
be related to the use of the study product by the inves-
tigator. The majority of AEs were rated as being ‘not
related’ (74.1%) or ‘unlikely to be related’ (19.6%) to
the study product. Eleven AEs (5.8%) were considered
to be ‘possibly related’ (4 [3.5%] in the control-active
group, 7 [9.5%] in the active-active group). The most
frequent of these were: weight decrease (3 events), flat-
ulence (2 events), and anorexia (2 events). One AE,
flatulence, in the control-active group was considered
to be ‘probably related’ to the use of the product. There
was no indication that AEs considered to be related to
the use of the study product during the RCT intensified
during the OLE study.
Twenty-one SAEs occurred in 19 patients during the
study; 10 SAEs in 9 patients in the control-active group
and 11 SAEs in 10 patients in the active-active group.
No SAEs were considered to be related to the use of
the study product by the investigator. Two patients
discontinued the study due to SAEs (avascular bone
necrosis in the control-active group and diverticula in
the active-active group), and one patient in the control-
active group died due to a malignant lymphoma. No
clinically relevant changes in laboratory safety mea-
sures and vital signs were observed: mean systolic
blood pressure, diastolic blood pressure, and heart rate
remained within normal limits.
Intake adherence and product evaluation
Overall product intake adherence during the OLE
study was 95.7% (95% Confidence Interval [CI]
94.7–96.7%), without significant differences between
the control-active and active-active groups (between-
group difference; t-test: t183 = 1.36, p = 0.174). When
combined with the Souvenir II RCT, the overall
48-week intake adherence for the active-active
group was 96.1% (95% CI, 95.0–97.1%). The high
compliance was supported by the biochemical out-
come parameters; erythrocyte DHA (Fig. 2A), EPA
and plasma vitamin E levels (Fig. 2B) remained
at consistently elevated levels in the active-active
group and were significantly increased in the
control-active group in the OLE phase (48-week
versus 24-week; 3.13 ± 1.94% (mean ± standard
deviation), paired t-test: t91 = −15.41, p < 0.001
[DHA]; 0.87 ± 0.48%, t91 = −17.42, p < 0.001 [EPA];
7.97 ± 6.34 ␮mol/l, t91 = −12.05, p < 0.001 [vitamin
E]). Moreover, plasma homocysteine levels sig-
nificantly decreased in the active-active (48-week
versus 24-week; −1.16 ± 3.17 ␮mol/l, paired t-test:
t84 = 3.38, p = 0.001) and in the control-active group
(48-week versus 24-week; −5.13 ± 5.27 ␮mol/l,
paired t-test: t91 = 9.34, p < 0.001).

Fig. 2. A) Percentage docosahexaenoic acid (DHA) of total fatty acids in erythrocyte membrane. 24–48 week change (paired t-test): p = 0.790
active-active, p < 0.001 control–active. B) Plasma vitamin E concentration (␮mol/l). 24–48 week change (paired t-test): p = 0.187 active–active,
p < 0.001 control–active. Data are mean and standard error (SE).
Responses from the product evaluation question-
naires, completed by patients with their caregivers,
indicated that the taste of the study product was well
appreciated, with no significant difference in mean
(±standard deviation) scores between the vanilla and
strawberry flavor (5.1 ± 1.5 compared with 4.9 ± 1.6).
These scores were comparable with the responses at
the end of the RCT, as well as with the appreciation
of the taste of the control products used in the RCT
(vanilla: 5.5 ± 1.4, strawberry: 4.9 ± 1.7).
Exploratory outcome measures
NTB memory domain and individual item scores
across the RCT and OLE are presented in Supplemen-
tary Table 1. The fact that a proportion of the subjects
in the RCT continued in the OLE resulted in an unbal-
anced design (different number of observations in the
different cells of the design). In unbalanced designs,
estimated marginal means are less likely to be biased
than unadjusted means. Figure 3 shows the marginal
means as estimated by the MMRM to illustrate the
course of the NTB memory domain scores in the total
48 week period. In the 24-week RCT, the primary out-
come of the trajectory over time of the NTB memory
domain score was statistically significantly increased
in the active group versus control group, which was pri-
marily driven by the changes from Week 12 to Week
24. In the OLE, the exploratory NTB memory domain
score in the active-active group appeared to show con-
tinued improvement (48-week versus 24-week scores;
paired t-test: t81 = −2.28, p = 0.025; MMRM contrast:
t996 = 2.43, p = 0.015). In addition, the change in NTB
memory domain scores between the RCT baseline
assessment and Week 48 was statistically significant
in the active-active group (paired t-test: t82 = −6.63,
p < 0.001; MMRM contrast: t996 = 5.87, p < 0.001).
The control-active group showed a statistically signif-
icant increase of the NTB memory domain score after
switching to the active product upon entry into the
OLE (48-week versus 24-week scores; paired t-test:
t86 = −2.72, p = 0.008; MMRM contrast: t996 = 2.71,
p = 0.007). All other between- and within group statis-
tical comparisons mentioned in the section ‘Statistical
analyses’ of the exploratory NTB memory domain
score were not statistically significant. Results of the
analyses of change in NTB memory domain score in
the two exploratory key groups of interest are shown
in Supplementary Figure 1 and described in the Sup-
plementary Material.
No effect on the Disability Assessment for Demen-
tia scale was observed in both the control-active and
active-active group.
DISCUSSION
The results of this multi-center OLE following the
Souvenir II RCT show that once-daily intake of Sou-
venaid is well tolerated with a favorable safety profile
in patients with mild AD for up to 48 weeks. Product

Fig. 3. NTB memory domain composite z-scores for both the double-blind RCT and OLE study period in (A) observed values and (B) change
from baseline (RCT week 0). 24–48 week change for active-active group; paired t-test: p = 0.025; MMRM: p = 0.015 and for control-active
group; paired t-test: p = 0.008; MMRM: p = 0.007. Data are estimated marginal mean (EMM) and standard error (SE).
intake adherence to Souvenaid was very high (≥95%),
reflecting its good tolerability and ease of use.
Souvenaid was well tolerated, with a similar per-
centage of patients experiencing AEs during the OLE
as in the original RCT. Most AEs were of mild intensity
and were not related to product intake. Overall, the type
and frequency of AEs in the OLE were comparable to
those reported in the RCT, while the overall percentage
of patients experiencing gastrointestinal-related events
during the OLE was even lower than during the RCT,
whichmaysuggestthatthesetypeofAEsdecreaseover
time. Overall gastrointestinal AE rates in the current
study were relatively low. In comparison, a Cochrane
review on the use of acetylcholinesterase inhibitors in
AD suggested that more gastrointestinal-related AEs
occur in acetylcholinesterase inhibitor groups than in
placebo [27]. However, these data are predominantly
based on studies in mild-to-moderate AD, so any com-
parisons should be made with caution. Besides the
favorable AE profile, laboratory safety parameters did
notgiveanyindicationforAEs,andthediscontinuation
rate in the OLE was similarly low as in the RCT. There
was no difference in the proportion of patients who
experienced AEs between the 22% of patients receiv-
ing both AD medication and Souvenaid and those who
remained AD-drug naïve. This extends the previous
observations that Souvenaid was well tolerated when
taken in conjunction with AD medication in mild-to-
moderate AD [20]. Overall, the current data support
results of previous studies and indicate that 48-week
use of Souvenaid is safe and well-tolerated.
Good compliance to an intervention is of particular
importance in clinical practice, where adherence to AD
therapies is often poor and represents a major problem
[28]. This OLE study reported a high product intake
adherence (≥95%), which is line with previous studies
[18–20] and supported by biochemical data. Although
intake adherence is likely to be higher in a clinical trial
population compared with a real-world clinical setting,
high compliance was further supported by the low rate
of discontinuation and these observations highlight the
potential of Souvenaid as a useful intervention for AD
in clinical practice.
The main efficacy assessment from the RCT, the
NTB memory domain z-score, was continued as an
exploratory outcome measure in the OLE. While there
are limits on the extent to which effectiveness can be
assessedinanopen-labelstudy,theincreaseintheNTB
memory score over a 48-week period is an encouraging
result.Episodicverbalmemorydeficitsareaclearchar-
acteristic of the cognitive deficits seen in early AD. The
NTB memory measure helpfully employed to detect
further improvements in memory during the OLE
phase, following the statistically significant improve-

ment observed during the RCT [19]. This finding is
underlined by the improvement seen during the OLE
in the control-active group, especially as patients were
not aware of whether they initially received control
or active intervention during the RCT. The observa-
tions on memory performance presented here extend
the findings of the two previous RCTs [18, 19]. How-
ever, it should be noted that the current exploratory
results should be interpreted carefully, since the direc-
tion of the effect during the OLE phase was similar to
the effect in this group when they used control prod-
uct in the RCT phase [19], and correction for multiple
testing was not applied.
There are limitations inherent to any open-label
study that should be considered when interpreting the
results. Firstly, although baseline characteristics were
comparable between study groups, there is a potential
for a selection bias on entry into the OLE. Secondly,
OLE studies are primarily designed to provide longer-
term safety and tolerability data applicable to routine
clinical practice. As such, there is no control group
and therefore efficacy cannot be adequately assessed.
A limitation of this specific OLE study compared to
the RCT is that there was no restriction to start with
any nutritional supplements and AD medication. Even
though an imbalance was observed in the number of
patients in each group starting AD medication during
the OLE, this appeared to be reversed after completion
of the study, making it difficult to fully compare the
RCT and the OLE studies.
In conclusion, the present open-label extension
study to assess the safety, compliance, and exploratory
effectiveness of Souvenaid in mild AD, showed that
the use of Souvenaid for a total period of 48 weeks
was well tolerated with a favorable safety profile and
associated with a preliminary suggestion of continued
improvement in memory performance. These find-
ings warrant further investigation toward the long-term
safety and efficacy of Souvenaid in a well-controlled,
randomized, blinded setting, like the ongoing 24-
month EU funded LipiDiDiet study (Dutch Trial
Register #NTR1705) in prodromal AD.
ACKNOWLEDGMENTS
We sincerely thank the patients and their caregivers
for their participation in the study and are indebted to
the other investigators of the Souvenir II OLE study
group: Dr. F.H. Bouwman; Dr. P.L.J. Dautzenberg; Dr.
W.M.W. Sipers; Dr. D. van Asselt; Dr. M. Stevens; Dr.
P. Tack; Dr. P. Bourgeois; Dr. E. Mulleners; Dr. med.
R. Horn; Prof. Dr. med. H-J. Gertz; Prof. Dr. med. J.
Schröder; PD Dr. med. K-C. Steinwachs; Dr. P.G. Gil;
Dr. M.R. Boada; Prof. P. Mecocci; Prof. G. Rodriguez;
Dr. P-J. Ousset; Dr. M.M. Vercelletto; Prof. M.P.M.
Ceccaldi.
Study design and conduct (collection, management,
analysis and interpretation of the data, and preparation,
review, and approval of the manuscript) were carried
out in conjunction with the sponsor, Nutricia Research,
on behalf of Nutricia Advanced Medical Nutrition. The
sponsor also provided the study products and funding
for the research and data collection.
Authors’ disclosures available online (http://www.j-
alz.com/disclosures/view.php?id=2537).
SUPPLEMENTARY MATERIAL
The supplementary material is available in the elect-
ronic version of this article: http://dx.doi.org/10.3233/
JAD-141305.
REFERENCES
[1] Selkoe DJ (2002) Alzheimer’s disease is a synaptic failure.
Science 298, 789-791.
[2] Terry RD, Masliah E, Salmon DP, Butters N, DeTeresa R, Hill
R, Hansen LA, Katzman R (1991) Physical basis of cognitive
alterations in Alzheimer’s disease: Synapse loss is the major
correlate of cognitive impairment. Ann Neurol 30, 572-580.
[3] Scheff SW, Neltner JH, Nelson PT (2014) Is synaptic loss a
unique hallmark of Alzheimer’s disease? Biochem Pharmacol
88, 517-528.
[4] Sperling RA, Aisen PS, Beckett LA, Bennett DA, Craft S,
Fagan AM, Iwatsubo T, Jack CR, Jr., Kaye J, Montine TJ,
Park DC, Reiman EM, Rowe CC, Siemers E, Stern Y, Yaffe
K, Carrillo MC, Thies B, Morrison-Bogorad M, Wagster MV,
Phelps CH.(2011) Toward defining the preclinical stages of
Alzheimer’s disease: Recommendations from the National
Institute on Aging-Alzheimer’s Association workgroups on
diagnostic guidelines for Alzheimer’s disease. Alzheimers
Dement 7, 280-292.
[5] DeKosky ST, Scheff SW (1990) Synapse loss in frontal cortex
biopsies in Alzheimer’s disease: Correlation with cognitive
severity. Ann Neurol 27, 457-464.
[6] Wurtman RJ, Ulus IH, Cansev M, Watkins CJ, Wang L,
Marzloff G (2006) Synaptic proteins and phospholipids are
increasedingerbilbrainbyadministeringuridineplusdocosa-
hexaenoic acid orally. Brain Res 1088, 83-92.
[7] Sakamoto T, Cansev M, Wurtman RJ (2007) Oral
supplementation with docosahexaenoic acid and uridine-5’-
monophosphate increases dendritic spine density in adult
gerbil hippocampus. Brain Res 1182, 50-59.
[8] Holguin S, Martinez J, Chow C, Wurtman R.(2008) Dietary
uridine enhances the improvement in learning and memory
produced by administering DHA to gerbils. FASEB J 22,
3938-3946.
[9] Savelkoul PJ, Janickova H, Kuipers AA, Hageman RJ,
Kamphuis PJ, Dolezal V, Broersen LM (2012) A spe-
cific multi-nutrient formulation enhances M1 muscarinic

acetylcholine receptor responses in vitro. J Neurochem 120,
631-640.
[10] Wang L, Albrecht MA, Wurtman RJ (2007) Dietary
supplementation with uridine-5’-monophosphate (UMP), a
membrane phosphatide precursor, increases acetylcholine
level and release in striatum of aged rat. Brain Res 1133,
42-48.
[11] de Bruin NM, Kiliaan AJ, de Wilde MC, Broersen LM (2003)
Combined uridine and choline administration improves cog-
nitive deficits in spontaneously hypertensive rats. Neurobiol
Learn Mem 80, 63-79.
[12] Lopes da Silva S, Vellas B, Elemans S, Luchsinger J, Kam-
phuis P, Yaffe K, Sijben J, Groenendijk M, Stijnen T.(2013)
Plasma nutrient status of patients with Alzheimer’s disease:
Systematic review and meta-analysis. Alzheimers Dement 10,
485-502.
[13] Olde Rikkert MGM, Verhey FR, Sijben JWC, Bouwman FH,
Dautzenberg PLJ, Lansink M, Sipers WMW, van Asselt DZB,
van Hees AMJ, Stevens M, Vellas B, Scheltens P (2014) Dif-
ferences in nutritional status between very mild Alzheimer’s
disease patients and healthy controls. J Alzheimers Dis 41,
261-271.
[14] van Wijk N, Broersen LM, de Wilde MC, Hageman RJ, Groe-
nendijkM,SijbenJW,KamphuisPJ(2014)Targetingsynaptic
dysfunction in Alzheimer’s disease by administering a spe-
cific nutrient combination. J Alzheimers Dis 38, 459-479.
[15] Jack CR, Knopman DS, Jagust WJ, Shaw LM, Aisen PS,
Weiner MW, Petersen RC, Trojanowski JQ (2010) Hypo-
thetical model of dynamic biomarkers of the Alzheimer’s
pathological cascade. Lancet Neurol 9, 119-128.
[16] WeintraubS,WicklundAH,SalmonDP(2012)Theneuropsy-
chological profile of Alzheimer disease. Cold Spring Harb
Perspect Med 2, a006171.
[17] Scheff SW, Price DA, Schmitt FA, Mufson EJ (2006) Hip-
pocampal synaptic loss in early Alzheimer’s disease and mild
cognitive impairment. Neurobiol Aging 27, 1372-1384.
[18] Scheltens P, Kamphuis PJ, Verhey FR, Olde Rikkert MG,
WurtmanRJ,WilkinsonD,TwiskJW,KurzA.(2010)Efficacy
of a medical food in mild Alzheimer’s disease: A randomized,
controlled trial. Alzheimers Dement 6, 1-10 e11.
[19] Scheltens P, Twisk J, Blesa R, Scarpini E, Vor Arnim C,
Bongers A, Harrison J, Swinkels S, Stam C, De Waal H, Wurt-
man R, Wieggers R, Vellas B, Kamphuis P. (2012) Efficacy
of Souvenaid in mild Alzheimer’s disease: Results from a
randomized, controlled trial. J Alzheimers Dis 31, 225-236.
[20] Shah RC, Kamphuis PJ, Leurgans S, Swinkels SH, Sad-
owsky CH, Bongers A, Rappaport SA, Quinn JF, Wieggers
RL, Scheltens P, Bennett DA (2013) The S-Connect study:
Results from a randomized, controlled trial of Souvenaid in
mild-to-moderate Alzheimer’s disease. Alzheimers Res Ther
5, 59.
[21] McKhann G, Drachman D, Folstein M, Katzman R, Price
D, Stadlan EM (1984) Clinical diagnosis of Alzheimer’s dis-
ease: Report of the NINCDS-ADRDA Work Group under the
auspices of Department of Health and Human Services Task
Force on Alzheimer’s Disease. Neurology 34, 939-944.
[22] Folstein MF, Folstein SE, McHugh PR (1975) "Mini-mental
state". A practical method for grading the cognitive state of
patients for the clinician. J Psychiatr Res 12, 189-198.
[23] Harrison J, Minassian SL, Jenkins L, Black RS, Koller M,
Grundman M. (2007) A neuropsychological test battery for
use in Alzheimer disease clinical trials. Arch Neurol 64, 1323-
1329.
[24] Gelinas I, Gauthier L, McIntyre M, Gauthier S (1999) Devel-
opment of a functional measure for persons with Alzheimer’s
disease: The disability assessment for dementia. Am J Occup
Ther 53, 471-481.
[25] Kuo K, Still R, Cale S, McDowell I (1997) Standardization
(external and internal) of HPLC assay for plasma homocys-
teine. Clin Chem 43, 1653-1655.
[26] Krijt J, Vackova M, Kozich V. (2001) Measurement of homo-
cysteineandotheraminothiolsinplasma:Advantagesofusing
tris(2-carboxyethyl)phosphine as reductant compared with
tri-n-butylphosphine. Clin Chem 47, 1821-1828.
[27] Birks J. (2006) Cholinesterase inhibitors for Alzheimer’s dis-
ease. Cochrane Database Syst Rev, CD005593.
[28] Small G, Dubois B (2007) A review of compliance to
treatment in Alzheimer’s disease: Potential benefits of a trans-
dermal patch. Curr Med Res Opin 23, 2705-2713.

Tolerability and Safety of Souvenaid in Patients with Mild Alzheimer’s Disease: Results of Multi-Center, 24-Week, Open-Label Extension Study

More Related Content

What's hot (20)

Similar to Tolerability and Safety of Souvenaid in Patients with Mild Alzheimer’s Disease: Results of Multi-Center, 24-Week, Open-Label Extension Study (13)

Recently uploaded (20)

Tolerability and Safety of Souvenaid in Patients with Mild Alzheimer’s Disease: Results of Multi-Center, 24-Week, Open-Label Extension Study