Treatment AD in the earliest stage: The role of medical nutrition Prof. Dr. Philip Scheltens

Disclosures The Alzheimer Center has received funding from: AEGON, ZONMW, Alzheimer Nederland, Heineken Nederland, ING, Stichting VUmc Fonds, AHAF, ISOA, ISAO, Pfizer, Jansen, Novartis, KLM Royal Dutch Airlines, KPN, KPMG, Twentse Kabel Holding, Stichting Zabawas, RABO Bank Image analysis research and clinical trials are carried out with Nutricia Advanced Medical Nutrition, Jansen Research Foundation, Novartis, Roche, Merck, Lundbeck, Pfizer Dr Scheltens receives no personal compensation from any of the above or others except from the VUmc

Alzheimer: de getallen Nu: ~ 250.000 patiënten in Nederland Belangrijkste risicofactor: Leeftijd Dubbele vergrijzing: méér ouderen worden ouder  snelle toename! Schatting 2040: 500.000 Aantal jong dementerende (<65) neemt toe

De impact van Alzheimer 1 op de 10 > 65 heeft Alzheimer  1 op de 3 > 80 heeft Alzheimer 

De impact van Alzheimer 1 op de 10 > 65 heeft Alzheimer  1 op de 3 > 80 heeft Alzheimer  Iedere 15 minuten krijgt iemand Alzheimer

De impact van Alzheimer 1 op de 10 > 65 heeft Alzheimer  1 op de 3 > 80 heeft Alzheimer  Iedere 15 minuten krijgt iemand Alzheimer Aangenomen dat per patiënt 3 zorgverleners (part time) betrokken zijn; zijn er 1.5 miljoen zorgverleners nodig in 2040. Terwijl de beroepsbevolking in aantal afneemt… ...neemt de belasting voor de maatschappij toe!  Conclusie: er ligt een grote uitdaging voor ons. Oplossingen moeten snel gevonden worden.    7

‘Dementia’ and ‘Neuronutrients’ > 10 Years of Research ‘98 ‘99 ’00 ’01 ’02 ’03 ’04 ’05 ’06 ’07 ‘08 ‘09 ’10 ’11 ‘12 Various national and international preclinical research consortia including ‘Dementia’ and ‘Neuronutrients’ Prof Wurtman, MIT Prof. Scheltens VUMc, NL Prof. Bennett Rush, USA Prof. Scheltens VUMc, NL Collaboration BTS project “Dementie” (1997-2001) Prof Monique Breteler, Erasmus Medical Center, The Netherlands Prof Paul Luiten, Rijksuniversiteit Groningen, The Netherlands BIT project “Neuronutrients” (2000-2004) Prof Richard Wurtman, MIT / Harvard Medical school, Cambridge, USA … Prof. Soininen UEF, Finland Souvenir I received funding from Stichting Technische Wetenschappen, Souvenir II received funding from the NL Food & Nutrition Delta project, FND #10003, LipiDiDiet: Funded by the EU FP7 project LipiDiDiet, Grant Agreement #211696 8

Alzheimer: 3 fundamentele processen Neurofibrillaire kluwens Synapsverlies Seniele plaques

Synaptic Failure in Alzheimer’s disease Control MCI AD 5 10 * -13% -44% # Synapses dentate gyrus (x1010) Synapse loss is an early event in the disease process Synaptic loss is strongest structural correlate with cognitive decline Failure to replace the loss of synaptic contacts leads to the decline in memory Scheff et al., Neurobiol Aging, 2006

Sperling et al., Alz & Dement, 2011 Synapse Loss in AD is an early event Synapse loss is an early event in the AD continuum Synaps loss happens early Current thinking illustrated by AD model, most shown picture on AD conferences, changes in biomarkers preceed functional decline More later re memory loss early symptom in AD, this study illustrates that loss of synapses is associated with memory decline So synapses are important, already early in disease Fig. 3. Hypothetical model of dynamic biomarkers of the AD expanded to explicate the preclinical phase: Ab as identified by cerebrospinal fluid Ab42 assay or PET amyloid imaging. Synaptic dysfunction evidenced by fluorodeoxyglucose (F18) positron emission tomography (FDG-PET) or functional magnetic resonance imaging (fMRI), with a dashed line to indicate that synaptic dysfunction may be detectable in carriers of the 34 allele of the apolipoprotein E gene before detectable Ab deposition. Neuronal injury is evidenced by cerebrospinal fluid tau or phospho-tau, brain structure is evidenced by structural magnetic resonance imaging. Biomarkers change from normal to maximally abnormal (y-axis) as a function of disease stage (x-axis). The temporal trajectory of two key indicators used to stage the disease clinically, cognitive and behavioral measures, and clinical function are also illustrated. Figure adapted with permission from Jack et al [22]. Sperling et al., Alz & Dement, 2011 11

Nutritional precursor control of neuronal membrane synthesis Saturated fatty acid Glycerol Phosphate Choline PUFA Axon neurite dendritic spine

Phospholipids are synthesized by the Kennedy Pathway Saturated fatty acid Glycerol Phosphate Choline PUFA e.g. DHA One way to do that might be to try to increase the formation of neuronal membranes, as membranes are the main constituents of synapses. Already in 1954 prof kennedy discoved this pathway for the synsthesis of neuralnal membrane, showing that it takes 3 precursors that are normally present in the diet to form Pl and neural membranes via several steps. More recently Prof Wurtman discovered that this pathway is controlled by low affinity enzymes, which implies that if you provide more of the substrates you will end up with more end prpoduct. For instance the notion that UMP is rate liminint in this pathway is illustrated by this experiment where it was showen that increased dietary intake of UMP increased the level of CDP chioline in the brain KENNEDY EP, WEISS SB (1956). J. Biol. Chem. 222 (1): 193–214 13

Different nutritional status in patients with mild AD No significant differences were observed for plasma folate, vitamins B6 and B12, choline, vitamins A and E, plasma fatty acids, BMI, and calf circumference. J.W. Sijben, M.G.M Olde Rikkert et al. Poster EFNS 2012 Stockhom

Verzadigde vetzuren Onverzadigde vetzuren

Increase the formation of neuronal membranes Providing the Nutritional Precursors and Co-Factors for Neuronal Membrane Formation UMP DHA, EPA Choline Phospholipids B vitamins Anti-oxidants Hypothesized to: Increase the formation of neuronal membranes Much of early developmental work was conducted By Professor Richard Wurtman at MIT, Boston, USA

Nutritional precursors increase membrane dependent structures: Neurite outgrowth Control Uridine 50 µM Pooler et al (2005) Neuroscience B-vitamins, choline and omega-3 fatty acids also stimulate neurite outgrowth in vitro plaatje: NGF-DIFFERENTIATED PHEOCHROMOCYTOMA CELLS 4 dagen behandeld in vitro Darios et al. (2006) Nature; Wang et al. (2000) Neurosci Lett; Calderon et al. (2004) J Neurochem

Spine Density hippocampus (for 50µM) Nutritional precursors increase membrane dependent structures (dendritic spines in gerbil hippocampus) Spine Density hippocampus (for 50µM) Control (choline) UMP UMP DHA DHA ** * 20 40 60 80 100 We used gerbils (Meriones unguiculatus) for these studies because the metabolism of pyrimidines in this species more closely resembles that in humans than does pyrimidine metabolism in rats: plasma levels of uridine are higher than those of cytidine, both basally and after giving a uridine (Cansev et al., 2005) or cytidine (Wurtman et al., 2000; Cansev and Wurtman, 2005) source, in humans and gerbils but not in rats (Lopez-Coviella et al., 1995). Dendritic spines are small membranous protrusions extending from post synaptic dendrites in neurons, most of which eventually form synapses with presynaptic axon terminals Same picture, , number of DSP per length of dendrite Effect fast Did not affect length of width of indivdual spines http://en.wikipedia.org/wiki/Dendritic_spine Sakamoto et al. (2007) Brain Res 19 19

& Cognition (ADAS-cog) Clinical Trials Prodromal Mild Moderate Memory (WMS-r) & Cognition (ADAS-cog) Memory (NTB) Biomarker EEG & MEG Cognition (ADAS-cog) Cognition (NTB) Biomarker MRI & CSF Trials are registered in the ICMJE compliant www.trialregister.nl

Souvenir I: Proof of Concept Study in Drug-Naive mild AD Multi-country (NL, Bel, Ger, UK, US), randomized, controlled trial Intervention 12 weeks (+ optional 12 wk extension) Co-primary outcomes: WMS-r delayed verbal recall ADAS-cog-13 Baseline Characteristics Control (n = 106) Active Sex (male/female; counts) 52 / 54 54 / 52 Age (y) 73.3 ± 7.8 74.1 ± 7.2 BMI (kg/m2) 26.2 ± 3.5 26.2 ± 4.8 Years of education on top of primary school 6.0 ± 4.0 5.5 ± 3.9 Days since AD diagnosis (median) 31.5 (0 – 1036) 30.0 (0 – 1932) Total MMSE score 24.0 ± 2.5 23.8 ± 2.7 t ≤-3 t=0 6 12 wks n=212 Souvenaid (n=106) Control (n=106) Outcome parameters Values are mean ±SD, unless stated otherwise

Blood Nutritional Parameters Significant changes (p<0.001) in vitamin E and EPA P-values for changes from Baseline to Week 24, control vs active AND changes from Week 24 to Week 48, control-active vs active-active. ITT, data are mean ± SE

No difference in AEs and Compliance Safety No significant differences in the number of (Serious) Adverse Events (S)AEs No clinically relevant differences in blood safety parameters No difference in dropouts (# patients) due to (S)AEs Tolerance No difference in product appreciation (taste and amount) between the Control and Active group Overall product compliance was very high at 95% with no difference between the groups SAEs Control: 14 SAEs in 11 patients, one possibly related (panic attack / hyperventilation) Active: 13 SAEs in 7 patients

Wechsler Memory Score Logisch geheugen  b) Uitgestelde herinnering (na 30 minuten) Verhaal A Anna / Jansen / uit Amsterdam / Zuid / die werkte/ als werkster / op een kantoorgebouw / deed aangifte / op het politiebureau / Singel, / dat zij de vorige avond/ in de Spuistraat / was aangehouden / en van 115 € / was beroofd. / Zij had vier / kleine kinderen / de huur / was nog niet betaald / en ze hadden reeds twee dagen / niet gegeten. / De agenten waren zeer getroffen / door dit verhaal / en hielden een geldinzameling / voor haar. Max. = 25 Totaal verhaal A

Co-Primary Outcome: WMS-r Delayed Verbal Recall Score mild AD after 12 weeks Very Mild AD after 12 weeks (p=0.021) (p=0.019) At baseline, approximately 40% of patients scored 0 [lowest score] on the WMS-r delayed verbal recall scale of 0–25. Given this skewed distribution, it was necessary to substitute the planned mixed-model analysis of 12-week data with nonparametric analyses. Both noncategorical (Mann–Whitney U Wilcoxon W test) and categorical nonparametric analyses (c2 test) gave similar results. ITT, Chi-square *At baseline 40% scored 0 [lowest score], planned MMRM substituted by nonparametric analyses, MWU and Chi-square gave similar results Pre-defined subgroup MMSE 24 - 26, Chi-square Scheltens et al. Alzheimers Dement. 2010 6 (1):1-10.

No significant (p=0.826) effect1 Co-Primary Outcome: ADAS-cog Primary analysis: No significant (p=0.826) effect1 ITT, MMRM, data are mean ± SE 1Scheltens et al. Alzheimers Dement. 2010 6 (1):1-10. 2Kamphuis et al. J Nutr Health Aging. 2011 15(8):720-4.

& Cognition (ADAS-cog) Clinical Trials Prodromal Mild Moderate Memory (WMS-r) & Cognition (ADAS-cog) Memory (NTB) Biomarker EEG & MEG Cognition (ADAS-cog) Cognition (NTB) Biomarker MRI & CSF Trials are registered in the ICMJE compliant www.trialregister.nl

Baseline Characteristics S-Connect study: Mild to Moderate AD using AD medication Principle investigators: David Bennett and Raj Shah, Rush, Chicago Multi-centre (48 sites in the US), randomized, controlled trial Intervention 24 weeks Primary outcome: ADAS-cog-11 Baseline Characteristics Control (n = 262) Active (n = 265) Age (y) 76.9 (8.2) 76.6 (8.2) Sex: males (n[%]) 127 (48.5%) 126 (47.5%) Years of education on top of primary school 6.4 (3.5) 6.7 (3.6) Total MMSE score 19.3 (3.0) 19.5 (3.2) Duration AD since diagnosis (months) 34.9 (29.6) 32.7 (25.0) Acetylcholinesterase inhibitors 243 (92.7%) 251 (94.7%) NMDA antagonist 170 (64.9%) 177 (66.8%) BMI (kg/m2) 26.64 (4.56) 26.19 (4.51) n=527 n=265 Active t ≤-3 0 12 24 wk Outcome parameters n=262 Control Values are mean±SD, unless stated otherwise

No difference in Adverse Events Body System Control (n=262) Active (n=265) p-value Any AE 165 (60.1) 150 (56.8) 0.130 Body as a whole 33 (12.7) 24 (9.1) 0.208 Nervous system 21 (8.1) 27 (10.2) 0.450 Gastro-intestinal 38 (14.6) 41 (15.5) 0.808 Metabolic & nutritional 19 (7.3) 19 (7.2) 1.000 Musculo-skeletal 15 (5.8) 0.183 Psychiatric 43 (16.5) 32 (12.1) 0.170 Respiratory system 42 (16.2) 50 (18.9) 0.423 Skin and appendages 18 (6.9) 8 (3.0) 0.045 Urinary system 25 (9.5) 0.432 Other 27 (10.4) 20 (7.6) 0.287 Occurrence of > 5% in total subjects Overall compliance during 24 weeks was 94% and not different between the groups

No significant effect* (p=0.513) during 24 weeks Primary Outcome: ADAS-cog No significant effect* (p=0.513) during 24 weeks ITT, MMRM, data are mean ±SE *Statistical analysis run by Rush Alzheimer’s Disease Centre, Rush University Medical Centre Shah et al., J Nutr Health Aging, 2011;15; Suppl 1:S30, Manuscript in preparation

& Cognition (ADAS-cog) Clinical Trials Prodromal Mild Moderate Memory (WMS-r) & Cognition (ADAS-cog) Memory (NTB) Biomarker EEG & MEG Cognition (ADAS-cog) Cognition (NTB) Biomarker MRI & CSF Trials are registered in the ICMJE compliant www.trialregister.nl

Baseline Characteristics Souvenir II study: Drug-Naive Mild AD Multi-country (NL, Ger, Bel, Fr, It, Sp), randomized, controlled trial Intervention 24 weeks Primary outcome: Memory Domain NTB (z-score): RAVLT immediate, delayed, recognition and VPA immediate and delayed Baseline Characteristics Control (n = 129) Active (n = 130) Age (y) 73.2 (8.4) 74.4 (6.9) Sex: males (n[%]) 64 (49.6) 68 (52.3) Years of education on top of primary school 6.6 (4.6) 6.5 (4.8) Total MMSE score 25.1 (2.9) 25.1 (2.8) Duration AD since diagnosis (months) (median[range]) 2.0 (0.0 - 88.0) 1.0 (0.0 - 70.0) BMI (kg/m2) 26.7 (4.2) 26.1 (4.1) n=259 n=130 Active t ≤-3 0 12 24 wk n=129 Control Outcome parameters Values are mean±SD, unless stated otherwise

No Difference in Adverse Events Body system Examples Control (n=129) Active (n=130) p-value Body as a whole Fatigue, influenza-like symptoms 20 (15.5%) 11 (8.5%) p=0.125 Central and peripheral nervous system disorders Dizziness, headache 18 (14.0%) p=0.237 Gastro-intestinal system disorders Constipation, diarrhoea, flatulence, nausea 30 (23.3%) 22 (17.1%) p=0.277 Metabolic and nutritional disorders Hyperglycaeemia, weight increase 9 (7.0%) 13 (10.1%) p=0.505 Musculo-skeletal system disorders Arthralgia, ischial neuralgia 10 (7.8%) p=1.000 Psychiatric disorders Anxiety, depression, insomnia 16 (12.4%) 15 (11.6%) Respiratory system disorders Pharyngitis, bronchitis p=0.400 Skin and appendages disorders Rash, skin dry 4 (3.1%) p=0.168 Other Fall, surgical intervention 8 (6.2%) Occurrence of > 5% in total subjects Overall compliance during 24 weeks was 97% and not different between the groups

Primary Efficacy: Memory Domain Score (z-score) of the NTB ITT, MMRM 2df contrast, data are mean ±SE *Statistical analysis re-run by Rush Alzheimer’s Disease Center Scheltens et al. J Alzheimers Dis. 2012 31(1):225-36.

Secondary Efficacy: NTB Total and NTB Executive Domain (z-score) NTB composite score trend (p=0.053) NTB executive domain score no significant effect (p=0.686) ITT, MMRM, 2 df contrast, data are mean ±SE Scheltens et al. J Alzheimers Dis. 2012 31(1):225-36.

Exploratory – Memory domain z-score Double-blind treatment Open-label extension Overall Open Label ITT population. This includes ALL subjects. Key point of this result is that active continues to increase and control catches up after starting on Souvenaid. Week 0, 12, 24 -> SII ITT; Week 36, 48 -> OLE ITT, all subjects Raw means and SE; change from baseline 12 24 36 48 Control (N) - 100 103 85 83 Active (N) 107

Electrical Activity at the Synapse – EEG: Biomarker for Functional Connectivity Basic quantitative EEG analysis - Relative power and Peak Frequency In AD disturbed signal strength1 PLI Phase Leg Index (PLI) as Functional Connectivity measure In AD loss of PLI1 Healthy AD Clustering & Path length are measures of Network Organization In AD disrupted Organization2 1Stam CJ et al. Brain 2009 132, 213-24 2Stam CJ, van Straaten ECW. Clin Neurophysiol 2012 doi:10.1016/j.clinph.2012.01.011

3. Brain Network organization* Network Parameters suggest Preserved Synaptic Formation, Function and Network 1. Peak Frequency 3. Brain Network organization* p=0.019 p=0.009 2. Phase Leg Index p=0.053 p=0.011 Scheltens et al. J Alzheimers Dis. 2012 31(1):225-36. *Manuscript in preparation, Developing topics P4-363 ITT, MMRM, 2 df contrast, data are mean ±SE

& Cognition (ADAS-cog) Clinical Trials Prodromal Mild Moderate Memory (WMS-r) & Cognition (ADAS-cog) Memory (NTB) Biomarker EEG & MEG Cognition (ADAS-cog) Cognition (NTB) Biomarker MRI & CSF Trials are registered in the ICMJE compliant www.trialregister.nl

EU – Funded* LipiDiDiet: Proof of Concept Study in Prodromal AD Principle investigator: H. Soininen (UEF, Kuopio, Finland) 24-Month randomized, controlled, multicenter (11 sites in Fin, Swe, Ger, NL) Drug-naive prodromal AD patients (Dubois et al., 2007) (recruited 240 / 300) Primary Outcome: Neuropsychological Test Battery (NTB) Secondary Outcomes: Progression to AD Functional Abilities CSF and MRI A randomised, controlled, double-blind, parallel group, multicentre study to assess the effect of a Medical Food in prodromal AD subjects Prodromal AD as defined by episodic memory disorder (Pieter Jelle Vissers version B)with -1SD on 2 out of 8 tests including at least 1 memory teest (cut-off?) and evidence for underlying AD pathology either 1. CSF beta-amyloid 1-42/1-40 ratio <1, T-tau >350 pg/ml and phospho tau >60 pg/ml or 2. MRI evidence for hippocampal atrophy greater than normal for age (Lars-Olof, Hilkka) New research criterea for the diagnosis of alzheimer’s, revised criterea to caputre the earlies stages before full-blown dementia as well as memory impairment Must be at least one or more abnormal biomarkers among Structural neuroimaging with mri (medial temperol lobe atrophy) Molecular neuroimaging with PET (reudced glucose or pib) - cerebrospinal fluid analysis of amylod or tau proteins - episodic memory impairment (gradually decline) * Funded by the EU FP7 project LipiDiDiet, Grant Agreement #211696

Mild AD is a New Target Very limited intervention studies in an exclusively mild AD population have been reported1 Van Gool WA et al., Lancet 2001; 358: 455–60. (hydroxychloroquine) Seltzer B et al., Arch Neurol 2004; 61: 1852–6. (Donepezil) Lannfelt L et al., Lancet Neurol 2008; 7: 779–86. (PBT2) Kessler H et al., J Neural Transm 2008; 115: 1181–7. (copper) Green RC et al., J Am Med Assoc 2009; 302: 2557–64. (Tarenflurbil) Hampel H et al., J Clin Psych 2009; 70: 922–31. (lithium) Scheltens P et al., Alzheimers Dement 2010; 6: 1–10. (Souvenaid) Souvenir II is the first study to suggest memory improvement as a predefined primary outcome measure Scheltens P et al., J Alzheimers Dis 2012; 31(1):225-36. (Souvenaid) 1Medline search string: English up to June 2012, Searches were restricted to randomized controlled clinical trials performed in an exclusively mild AD study population. "mild/early Alzheimer's" or "very mild Alzheimer's" or "cognition testing and outcome and Alzheimer’s" and randomized/controlled/single blind/double blind/intervention study/trial" including nutritional intervention studies.

The AD Continuum Abnormal Normal Time Presymptomatic Dementia CSF Aβ42 Amyloid imaging FDG-PET MRI hippocampal volume CSF Tau Cognitive performance Function (ADL) FDG-PET (Synaptic Dysfunction) MRI hippocampal volume CSF Tau Prodromal Modified from Aisen PS Alzheimers Dement. 2010 42

Multi Level Approach: no single magic bullit for AD IMMUNISATION ? LOWERING AMYLOID ? MEDICAL FOOD SYMPTOMATIC APPROACH CARE

VUmc Alzheimer Center