Figure 1. Patient disposition in bapineuzumab Studies 201 and 202. Number of cases in the bapineuzumab and placebo groups in each study are given from the patients receiving at least 1 infusion after randomization to cases that completed the study and had both baseline and week 54 cerebrospinal fluid (CSF) data.
Figure 2. Changes in cerebrospinal fluid (CSF) biomarker levels between baseline and week 54 in the pooled bapineuzumab Studies 201 and 202. Values are given as changes in absolute levels (pg/mL), showing both individual values and mean (standard error of the mean). Significances are based on an analysis of covariance model as described in the methods section: T-tau P = .03 within the bapineuzumab group; P-tau P = .001 within the bapineuzumab group; P = .03 for the differences between the bapineuzumab and placebo groups; and AβX-42 P = .02 within the bapineuzumab group. Aβ indicates β-amyloid; P-tau, phosphorylated tau; and T-tau, total tau.
Figure 3. Pharmacokinetics of bapineuzumab. Cerebrospinal fluid (CSF) to serum concentrations of bapineuzumab at different doses at end of study (week 54). Values given as mean (standard error of the mean) (percentage). Ten cases at a treatment dose of 0.15 mg/kg, 5 cases at 0.5 mg/kg, 6 cases at 1.0 mg/kg, and 4 cases at 2.0 mg/kg. The mean (standard error of the mean) absolute bapineuzumab levels in serum and CSF were 1650 (118) ng/mL and 4.9 (1.2) ng/mL, respectively, in the 0.15 mg/kg cohort; 6290 (1060) ng/mL and 18.1 (6.0) ng/mL, respectively, in the 0.5 mg/kg cohort; 11 460 (1700) ng/mL and 27.2 (5.1) ng/mL, respectively, in the 1.0 mg/kg cohort; and 17 660 (1510) ng/mL and 44.7 (7.7) ng/mL, respectively, in the 2.0 mg/kg cohort.
Blennow K, Zetterberg H, Rinne JO, Salloway S, Wei J, Black R, Grundman M, Liu E, AAB-001 201/202 Investigators FT. Effect of Immunotherapy With Bapineuzumab on Cerebrospinal Fluid Biomarker Levels in Patients With Mild to Moderate Alzheimer Disease. Arch Neurol. 2012;69(8):1002-1010. doi:10.1001/archneurol.2012.90
Author Affiliations: Clinical Neurochemistry Laboratory, Sahlgrenska Academy, University of Gothenburg, Mölndal, Sweden (Drs Blennow and Zetterberg); Turku PET Centre and Clinical Research Services, University of Turku, Turku University Hospital, Turku, Finland (Dr Rinne); Butler Hospital, Warren Alpert Medical School, Brown University, Providence, Rhode Island (Dr Salloway); Janssen Alzheimer Immunotherapy, San Francisco (Drs Wei and Liu) and Global R&D Partners, San Diego (Dr Grundman), California; and Pfizer, Collegeville, Pennsylvania (Dr Black). The Bapineuzumab Study 201 and 202 investigators and their affiliations are listed at the end of this article.
Background Given the slow and variable clinical course of Alzheimer disease, very large and extended clinical trials are needed to identify a beneficial clinical effect of disease-modifying treatments. Therefore, biomarkers are essential to prove that an anti–β-amyloid (Aβ) drug candidate affects both Aβ metabolism and plaque load as well as downstream pathogenic mechanisms.
Objective To evaluate the effect of the anti-Aβ monoclonal antibody bapineuzumab on cerebrospinal fluid (CSF) biomarkers reflecting Aβ homeostasis, neuronal degeneration, and tau-related pathology in patients with Alzheimer disease.
Design Two phase 2, multicenter, randomized, double-blind, placebo-controlled clinical trials of 12-month duration.
Setting Academic centers in the United States (Study 201) and England and Finland (Study 202).
Patients Forty-six patients with mild to moderate Alzheimer disease.
Interventions Patients received either placebo (n = 19) or bapineuzumab (n = 27) in 3 or 4 ascending dose groups.
Main Outcome Measures Changes between end of study and baseline in the exploratory CSF biomarkers Aβ1-42, AβX-42, AβX-40; total tau (T-tau); and phosphorylated tau (P-tau).
Results Within the bapineuzumab group, a decrease at end of study compared with baseline was found both for CSF T-tau (−72.3 pg/mL) and P-tau (−9.9 pg/mL). When comparing the treatment and placebo groups, this difference was statistically significant for P-tau (P = .03), while a similar trend for a decrease was found for T-tau (P = .09). No clear-cut differences were observed for CSF Aβ.
Conclusions To our knowledge, this study is the first to show that passive Aβ immunotherapy with bapineuzumab results in decreases in CSF T-tau and P-tau, which may indicate downstream effects on the degenerative process. Cerebrospinal fluid biomarkers may be useful to monitor the effects of novel disease-modifying anti-Aβ drugs in clinical trials.
Trial Registrations clinicaltrials.gov Identifier: NCT00112073, EudraCT Identifier: 2004-004120-12, and isrctn.org Identifier: ISRCTN17517446.
Alzheimer disease (AD) is a progressive neurodegenerative disease characterized neuropathologically by cerebral neuronal loss, deposits of extracellular β-amyloid (Aβ) plaques, and intraneuronal neurofibrillary tangles with accompanying decreases in cerebrospinal fluid (CSF) Aβ and increases in CSF tau proteins.1 Current marketed therapies for AD aim to improve symptoms by targeting the surviving neurotransmitter neuronal circuitry.2 The pivotal trials for approvals for these cholinesterase inhibitors were of short duration (6 months) owing to expectations of relatively rapid onset of symptomatic improvement. In contrast, therapies aiming to slow the progression of the disease may require clinical trials with longer duration to observe clinical improvement owing to downstream therapeutic effects on the underlying pathophysiological process. In addition to the necessary cognitive and functional end points, measurements of biomarkers reflecting the molecular pathogenesis of the disease would be critical for assessing the efficacy of such disease-modifying therapies.3
β-Amyloid immunotherapies are based on either active immunization with full-length Aβ or Aβ analogues together with an adjuvant or passive immunization with humanized anti-Aβ antibodies or intravenous immunoglobulins.1 The intended effect is thought to be mediated by anti-Aβ antibodies that either bind to Aβ plaques or other forms of Aβ aggregates in the brain, thereby inducing Aβ clearance by microglia or by binding soluble Aβ in the periphery, thereby driving an efflux of Aβ from the brain.1 Preclinical studies in transgenic mice producing excess Aβ have shown that antibodies directed against the N-terminus of Aβ enter the brain and reduce amyloid deposits in both brain tissue and the cerebral vasculature.4,5 These antibodies also block the synaptotoxic effects of Aβ oligomers and improve cognitive performance in amyloid precursor protein transgenic mice.6,7 Previous Aβ immunotherapy trials in humans using active immunization with the full-length Aβ42 peptide suggested clinical benefits.8
Bapineuzumab, an antibody targeted against the N-terminus of Aβ, was previously evaluated in phase 2 multicenter trials on passive Aβ immunotherapy in AD9,10 and is currently being evaluated in phase 3 trials. In this exploratory post hoc pooled analysis, we evaluated whether bapineuzumab impacted the CSF levels of the downstream biomarkers, total tau (T-tau) and phosphorylated tau (P-tau), and the primary biomarker Aβ in these completed trials.9,10
Two phase 2, multicenter, randomized, double-blind, placebo-controlled, multiple-ascending dose studies were conducted at 30 sites in the United States (Study 201)9 and 3 clinical sites in Europe (2 in England and 1 in Finland) (Study 202).10 Study 201 enrolled 234 patients, randomly assigned to either placebo or intravenous bapineuzumab in 4 dose cohorts (0.15, 0.5, 1.0, or 2.0 mg/kg). Study 202 enrolled 28 patients randomized to either placebo or intravenous bapineuzumab in 3 ascending dose cohorts (0.5, 1.0, or 2.0 mg/kg). Patients received 6 infusions, 13 weeks apart, with final assessments at week 78 (Table 1).
Both studies included a substudy in which CSF was collected when clinicians agreed to perform and patients consented to undergo lumbar puncture. Cerebrospinal fluid was taken before treatment initiation and at week 54 (approximately 2 weeks after the week 52 study drug infusion). Study 201 enrolled 35 substudy patients (20 taking bapineuzumab and 15 taking placebo) and Study 202 enrolled 11 CSF substudy patients (7 taking bapineuzumab and 4 taking placebo), with paired baseline and post-baseline samples (Figure 1). Patient demographics are presented in Table 2.
Each patient underwent clinical assessments of cognition and function, volumetric and safety magnetic resonance images (MRIs), and safety evaluations.9,10 Subjects in Study 202 also underwent carbon 11-labeled Pittsburgh Compound B positron emission tomography and [18F]-fluorodeoxyglucose positron emission tomographic imaging. An independent safety monitoring committee assessed the safety of treatment throughout the trial. Eligible patients met the National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer's Disease and Related Disorders association11 criteria for probable AD, had an MRI consistent with AD, and a score of 4 or less on the Rosen-modified Hachinski Ischemic Score.12 Study 201 enrolled subjects who had a score of 16 to 26 on the Mini-Mental State Examination scale, while Study 202 enrolled subjects with a score of 18 to 26.13 Patients were excluded if they had clinically significant neurologic or psychiatric disease other than AD, while treatment with cholinesterase inhibitors or memantine at a stable dose for at least 120 days before screening was allowed. One placebo subject had a CSF T-tau level of 2490 pg/mL at screening and a reduction of 1541 pg/mL at week 54. It was reported that this subject fell within the first week of the study, and he may have fallen previously. A fall due to a minor stroke or with concussion might explain the high CSF T-tau at baseline and its subsequent drop; declines in CSF T-tau of 1500 pg/mL are not normal for patients with AD without some other explanation.14,15 Therefore, all data points from this subject were excluded from the analysis.
Study 201 (clinicaltrials.gov Identifier: NCT00112073) was approved by each site's local institutional review board, while Study 202 (EudraCT Identifier: 2004-004120-12 and isrctn.org Identifier: ISRCTN17517446) was approved by each site's local independent ethics board. For both studies, written informed consent was obtained from each patient (or legally authorized representative).
Cerebrospinal fluid samples were obtained by lumbar puncture. Cerebrospinal fluid was collected in polypropylene tubes to avoid absorbance of proteins to the test-tube walls. All CSF samples were gently mixed to avoid possible gradient effects, centrifuged, aliquoted, frozen, and stored at −80°C, pending biochemical analyses. A corresponding blood sample was obtained at the time of the lumbar puncture.
Cerebrospinal fluid T-tau was analyzed using a sandwich enzyme-linked immunosorbent assay (ELISA) (Innotest hTAU-Ag; Innogenetics), which measures all tau isoforms irrespective of phosphorylation status.16 Phosphorylated tau in CSF was analyzed using a sandwich ELISA method (Innotest Phospho-tau[181P]; Innogenetics).17 Cerebrospinal fluid Aβ1-42 was measured using a sandwich ELISA (Innotest β-amyloid(1-42); Innogenetics), as previously described.18 In this assay, the monoclonal antibody 3D6, which is specific to the N-terminus (epitope Aβ1-5), was used as a detector. Cerebrospinal fluid AβX-42 was analyzed using a modified sandwich ELISA in which the 4G8 monoclonal antibody, which has the epitope Aβ18-22, was used as detector antibody, thus it measures all N-terminally truncated Aβ42 species. The AβX-40 in CSF was analyzed using the MS6000 Human Ultra-Sensitive Aβ40 Kit (Meso-Scale Discovery). All CSF analyses were performed at the end of study at the Clinical Neurochemistry Laboratory, Sahlgrenska University Hospital, Mölndal, Sweden, by certified laboratory technicians. All CSF samples were analyzed in 1 batch, with paired samples from individual patients side by side on the same plate. Serum bapineuzumab concentrations were measured by a validated sandwich ELISA method with a range of quantification from 1.98 to 147 ng/mL, while the bapineuzumab concentration in human CSF was measured by a validated electrochemiluminescence method with a range of quantification from 1.56 to 200 ng/mL.
The analysis compared the bapineuzumab group with the placebo group using an analysis of covariance model. The primary outcome was the change between end of study (week 54) and baseline CSF biomarker levels. The explanatory variables were treatment group (bapineuzumab vs placebo) and screening value of the outcome of interest as a continuous covariate. The analysis was a 2-sided test of the week 54 least squares mean difference and was based on the observed cases (ie, missing values were not imputed). Given the small sample size of each cohort and lack of a clear efficacy dose response, all dose levels of bapineuzumab (or placebo) were pooled in the analysis. In addition, the data from the 2 studies were combined and analyzed using the same analysis of covariance model as for the individual study, except that study was included as an additional exploratory variable to adjust for differences between Study 201 and Study 202. All analyses were performed with SAS version 9.1 (SAS Institute).
Neither Study 201 nor Study 202 were designed or powered as an efficacy study. The sample size for each study was calculated to ensure 80% or greater probability of detecting an adverse event that occurred with a rate of at least 5% within a single bapineuzumab-treated dose cohort. The CSF substudy was optional to participants who consented to the additional procedures, thus the sample size for the substudy was not a priori specified. The analyses were considered exploratory and P values were not adjusted for multiplicity. Statistical differences were noted if the calculated P < .05.
The combined analysis of Study 201 and Study 202 included a total of 27 bapineuzumab-treated and 19 placebo-treated subjects with paired baseline and end of study samples, most of whom had all 6 intended doses of bapineuzumab (Table 1). No significant differences were found in basic demographic variables between the bapineuzumab and placebo groups or between subjects that completed the CSF substudy and the drop-outs (Table 2). Some minor differences in age, height, and duration of disease were found between the subjects in the CSF substudy and those who did not enroll (Table 2), but we found no evidence in the literature that such factors would influence the longitudinal change in CSF biomarker levels during a 1-year period.
In pooled analysis from Study 201 and Study 202 (Table 3, Figure 2), a decrease in CSF T-tau was observed at end of study compared with baseline in the bapineuzumab-treated group (−72.3 pg/mL, P = .03), while there was no difference in the placebo group (−5.6 pg/mL, P = .80). When comparing the change from baseline to end of study, a trend for a decrease in CSF T-tau at end of study was found in bapineuzumab cases compared with placebo cases (−66.7 pg/mL, P = .09).
In pooled analysis of Study 201 and Study 202 (Table 3, Figure 2), a decrease in CSF P-tau was observed at end of study compared with baseline in the bapineuzumab-treated group (−9.9 pg/mL, P = .001), while there was no difference in CSF P-tau levels in the placebo group. When comparing the change from baseline to end of study between the bapineuzumab-treated cases with placebo subjects, a treatment decrease was observed (−7.3 pg/mL, P = .03).
In the pooled analysis of Study 201 and Study 202, there were no notable differences in CSF Aβ1-42 or AβX-40 at end of study (week 54) compared with baseline within the bapineuzumab group, or when comparing the change between end of study and baseline between the bapineuzumab and placebo groups (Table 3, Figure 2). There was a modest increase in CSF AβX-42 (Table 3, Figure 2) at end of study compared with baseline for the bapineuzumab-treated group (32.2 pg/mL, P = .02). However, no difference in CSF AβX-42 levels were seen when comparing the change from baseline to end of study between the bapineuzumab and placebo groups (18.4 pg/mL, P = .89).
At the 4 doses assessed, both serum and CSF bapineuzumab concentrations increased with dose. Thus, the average ratios of CSF to serum bapineuzumab concentrations were consistent at approximately 0.3% (Figure 3). No anti-bapineuzumab antibodies were detected in any subject.
Summaries of safety findings have been previously reported.9,10 In this study, 1 patient in the bapineuzumab group had MRI evidence of amyloid-related imaging abnormalities thought to represent vasogenic edema or sulcal effusions.19 Cerebrospinal fluid samples were taken both at baseline and at the time of the amyloid-related imaging abnormalities–edema/effusion (ARIA-E) for this patient. Cerebrospinal fluid safety biomarkers showed a normal CSF to serum albumin ratio (5.4 at baseline vs 5.0 at time of ARIA-E; reference value <10.2) and IgG index (0.41 at baseline vs 0.51 at time of ARIA-E; reference value <0.63), as well as no oligoclonal IgG bands in CSF, either at baseline or at the time of ARIA-E). The CSF AD biomarkers showed typical values for AD, and they were stable between baseline and at the time of ARIA-E (796 vs 720 pg/mL for T-tau, 78 vs 75 pg/mL for P-tau, and 307 vs 240 pg/mL for Aβ1-42).
Passive immunotherapy with antibodies against Aβ, such as bapineuzumab,9 is one of the major disease-modifying therapeutic approaches being evaluated for AD. Because the clinical course is slowly progressive and highly variable in AD, very large clinical trials with extended treatment duration will be needed to identify clinical effects of disease-modifying drugs. For this reason, theragnostic markers may be valuable to identify and monitor the biochemical effect of a novel drug.20 Biomarker evidence that the drug affects both the primary target and downstream pathogenic mechanisms will also be essential to label the drug as being disease modifying.20 In this context, biomarkers in anti-Aβ clinical trials can be divided into primary (pharmacodynamic) biomarkers used to monitor the specific biochemical mode of action of the anti-Aβ drug and downstream biomarkers used to monitor effects on downstream pathogenic processes, such as the neuronal degeneration or tangle formation, downstream of the drug target.21
The pharmacokinetic characteristics of bapineuzumab include a small distribution volume, slow clearance, and long terminal half-life. The ratio of bapineuzumab in CSF to serum was stable between the dose cohorts, with a mean of approximately 0.3%. The CSF to serum ratio for total IgG is also approximately 0.3%,22 suggesting that bapineuzumab passes the blood-brain barrier at the expected ratio for IgG antibodies.
In our study, we found a decrease in CSF P-tau in the bapineuzumab group, while no change was found in the placebo group. In addition, there was a treatment decrease in CSF P-tau at study end for the bapineuzumab-treated compared with placebo-treated subjects. Phosphorylated tau measured in CSF samples taken during life correlate with neocortical tangle pathology at autopsy,23 and CSF P-tau also correlates with the rate of hippocampal atrophy in the brain.24 The CSF level of P-tau thus seems to reflect the phosphorylation state of tau and the formation of tangles in the brain. Animal studies suggest that Aβ immunotherapy may affect tau pathology.25 A neuropathologic study of patients with AD from the AN1792 trial also suggests that Aβ immunotherapy ameliorates neurite abnormalities and tau pathology through decreased tau phosphorylation.26 The reduction in the downstream biomarker CSF P-tau following treatment with bapineuzumab suggests that bapineuzumab reduces brain levels of P-tau, which may also reduce the formation of tangles in the brain.
In addition, the downstream biomarker CSF T-tau decreased with treatment in the bapineuzumab group, while no change was found in the placebo group. This is consistent with the reduction in CSF T-tau seen in the AN1792 trial with active Aβ immunotherapy.8 Cerebrospinal fluid T-tau levels correlate with the amount of damaged tissue and poor clinical outcome in acute brain disorders,15,27 and high T-tau has also been associated with fast progression from mild cognitive impairment to AD and with rapid cognitive decline and a high mortality rate in patients with AD,28,29 suggesting that the CSF level of T-tau reflects the intensity of the neuronal degeneration. The reduction in the downstream biomarker CSF T-tau with bapineuzumab treatment warrants further investigation into the potential for this drug to attenuate the intensity of the neurodegenerative process.
Bapineuzumab is hypothesized to bind to Aβ in the brain, thereby facilitating its clearance.6 In support of this hypothesis, we previously reported that bapineuzumab treatment results in a modest reduction in cortical binding of the Aβ ligand Pittsburgh Compound B as evaluated by positron emission tomography, both as compared with baseline and with placebo-treated patients.10 Because the population in this study is not identical to that in the positron emission tomography study, the number of cases with both investigations is too small to allow any comparisons. In the AD brain, a relatively large proportion of Aβ is N-terminally truncated or modified.30 For this reason, we analyzed Aβ42 by 2 different ELISA methods, 1 for Aβ1-42 to measure full-length Aβ42 peptides and 1 for AβX-42, which also captures Aβ isoforms that are N-terminally modified. We were not able to identify any change in CSF AβX-42 or in Aβ1-42 or AβX-40 that differed between the bapineuzumab and placebo groups. Similarly, CSF Aβ42 levels were also minimally affected in the AN1792 immunotherapy trial.8 This lack of change in CSF Aβ42 may be attributed to clearance of cortical Aβ through other pathways than CSF or, alternatively, oligomerization of Aβ or binding of Aβ to chaperones or to the therapeutic antibody that masks a change in CSF Aβ with treatment. For these reasons, the Aβ results have to be interpreted with caution.
Cerebrospinal fluid biomarkers may be valuable for safety monitoring in AD clinical trials.20 Magnetic resonance image changes referred to as ARIA-E appears to be a safety concern with bapineuzumab treatment.19 In one study, this MRI abnormality was found in approximately 10% of treated cases.9 In our study, both a baseline CSF sample and a sample at the time of ARIA-E were available in the single patient who developed ARIA-E. This case had normal and stable CSF levels of both albumin and IgG; no oligoclonal IgG bands in CSF; and stable CSF levels of T-tau, P-tau, and Aβ42, which were compatible with the clinical diagnosis of AD. Thus, in this patient, ARIA-E was not accompanied by blood-brain barrier dysfunction, inflammatory reactions, or tau-related signs of additional neuronal damage.
One limitation is that this study is based on pooled analysis of the bapineuzumab trials Study 201 and Study 202. This was done to increase the sample size of patients in each substudy with paired CSF samples. However, for both trials, CSF biomarkers were analyzed in the same laboratory using the same assay formats, CSF samples from individual cases were analyzed side by side on the same ELISA plate, and data were analyzed as the change in biomarker levels between baseline and end of study. This procedure will minimize variation and allow pooling of CSF data.
In summary, in a pooled analysis of CSF data from 2 phase 2 clinical trials on passive immunotherapy with bapineuzumab in patients with mild to moderate AD, a decrease in both P-tau and T-tau at end of study compared with baseline within the bapineuzumab group was observed. For CSF P-tau, a statistically significant treatment difference was observed between the bapineuzumab and placebo groups. These findings may indicate downstream effects of bapineuzumab treatment on the degenerative process. An important question remains whether such changes in CSF biomarkers correlate with clinical benefit.21 This question will be addressed in the ongoing bapineuzumab phase 3 trials.
Correspondence: Kaj Blennow, MD, PhD, Clinical Neurochemistry Laboratory, Institute of Neuroscience and Physiology, Department of Psychiatry and Neurochemistry, Sahlgrenska Academy, University of Gothenburg, Sahlgrenska University Hospital, Mölndal, SE-431 80 Mölndal, Sweden (email@example.com).
Accepted for Publication: January 19, 2012.
Published Online: April 2, 2012. doi:10.1001/archneurol.2012.90
Author Contributions: All authors had access to the study data in the form of statistical tables and figures that were generated by the data analyst (J.W.). Study concept and design: Blennow, Black, Grundman, and Liu. Acquisition of data: Blennow, Zetterberg, Rinne, Salloway, and Liu. Analysis and interpretation of data: Blennow, Zetterberg, Salloway, Wei, Black, Grundman, and Liu. Drafting of the manuscript: Blennow, Grundman, and Liu. Critical revision of the manuscript for important intellectual content: Blennow, Zetterberg, Rinne, Salloway, Wei, Black, Grundman, and Liu. Statistical analysis: Wei and Grundman. Obtained funding: Liu. Administrative, technical, and material support: Blennow, Zetterberg, and Grundman. Study supervision: Blennow and Liu.
Group Members: Investigators in Study 201: R. Doody, S. Rosentree, S. Awosika-Olumo, J. Sims, M. Chowdhury, Baylor College of Medicine, Houston, Tex; M. Brody, D. Watson, C. Bouchard, G. Barbati, L. Brody, Brain Matters Research Inc, Delray Beach, Fla; R. Sperling, G. Marshall, D. Wolk, M. Vander Vliet, S. Salisbury, D. Rentz, Brigham & Women's Hospital, Boston, Mass; S. Salloway, P. Malloy, B. Blackham, S. Correia, D. Cimini, Butler Hospital, Providence, RI; M. Sabbagh, J. Ceimo, J. Lopez, K. Davis, D. Connor, Cleo Roberts Center for Clinical Research/Sun Health Research Institute, Sun City, Ariz; P. Solomon, L. Catapane-Friedman, P. Levin, C. Holland, M. Robinson, Clinical Neuroscience Research Associates Inc, Bennington, VT; K. Bell, L. Honig, E. Dominguez-Rivera, R. Tejeda, A. Canaan, Columbia University, New York City, NY; J. Beyer, R. Doraiswamy, K. Gersing, M. Aiello, H. Husn, Duke University Medical Center, Durham, NC; P. Aisen, C. Ward, K. Behan, B. Reynolds, Georgetown University Medical Center, Washington, DC; A. Hake, R. Evans, E. Van Hook, K. Cavanaugh, D. Rexroth, Indiana University Medical Center, Indianapolis, Ind; N. Graff-Radford, J. Boughey, R. Fletcher, F. Parfitt, L. Makarov, Mayo Clinic, Jacksonville, Fla; D. Knopman, B. Boeve, N. Haukom, K. Johnson, K. Bailey, K. Wytaske, Mayo Clinic, Rochester, Minn; J. Quinn, J. Kaye, T. Gilmore, K. Wild, F. Ferguson, Oregon Health & Science University, Portland, Ore; C. Wilcox, J. Morrissey, A. Hardy, W. Myrant, Pharmacology Research Institute, Los Alamitos, Calif; D. Grosz, A. Schneider, J. Morrissey, W. Myrant, C. Wilcox, Pharmacology Research Institute, Encino, Calif; L. Taber, L. Kirby, T. Williams, C. McCarthy, T. Footer, S. Fadden, J. Christensen, S. Borwege, Pivotal Research Centers, Peoria, Ariz; N. Aggarwal, R. Shah, N. Lopez-Barbera, S. Carroll, M. Vidaurri, Rush Presbyterian St. Luke's Medical Center, Chicago, Ill; J. Ross, Y. Shi, P. Ponugoti, W. Garcia, L. McGlue, K. Doyle, The Memory Enhancement Center, Eatontown, NJ; C. DeCarli, W. Seavery, C. Conover, K. Sweeny, M. Levallois, J. Campos, B. Faulstich, University of California Davis Medical Center, Sacramento, Calif; A. Boxer, B. Miller, M. Koestler, K. Arnold, N. Belfor, R. Gearhart, University of California at San Francisco; N. Barbas, J. Heidebrink, J. Lord, L. Guidotti, A. Freymuth Caveny, University of Michigan Health System, Ann Arbor; J. Corey-Bloom, G. Tong, K. Wetzel, A. Wilson, G. Tong, University of California–San Diego, La Jolla; R. Mulnard, G. Thai, C. McAdams-Ortiz, B. Yanez, University of California at Irvine; O. Lopez, S. Dekosky, W. Klunk, L. Macedonia, B. Sarles, T. Baumgartner, University of Pittsburgh, Penn; A. Porsteinsson, P. Tariot, B. Goldstein, J. LaFountain, L. Terwilliger, University of Rochester/Monroe Community Hospital, Rochester, NY; R. Rosenberg, K. Womack, D. Svetlik, K. Moser, B. Davis, University of Texas Southwestern Medical Center, Dallas; M. Raskind, E. Peskind, S. Eliza, N. Brown, E. Petrie, VA Medical Center, Seattle, Wash; J. Morris, J. Galvin, C. Tomlinson, W. Overkamp, S. Schneider, Washington University School of Medicine, St Louis, MO; and C. van Dyck, R. Tampi, K. Estok, M. MacAvoy, N. Barcelos, A. Benincasa, M. Sazon, Yale University School of Medicine, New Haven, Conn. Investigators in Study 202: T. Lehtinen, M. Kailajarvi, J. Hilli, P. Ruokomäki, K. Virtanen, T. Kaila, M. Koutu, T. Toikka, M. Scheinin, University of Turku, Turku, Finland; and M. Rossor, W. Knight, J. Douglas, J. Warren, B. Ridha, University College London, Institute of Neurology, London; R. Bullock, A. Narayan, K. Ventkatesh, C. Hall, H. Cartwright, Kingshill Research Centre, Swindon, England.
Financial Disclosure: Dr Blennow has served on scientific advisory boards for Innogenetics and Pfizer as well as on a speaker bureau for Janssen Alzheimer Immunotherapy. Dr Rinne has received research support from Elan/Wyeth. Dr Salloway has served as a consultant for AstraZeneca, Athena, Avid-Lilly, Baxter, Bristol-Myers Squibb, Eisai, Elan, Janssen Alzheimer Immunotherapy, Pfizer, and Sanofi. He has also received honorarium from Athena Diagnostics, Bristol-Myers Squibb, Eisai, Elan, and Pfizer, as well as provided data monitoring for Medivation and Merck-Serono. He has also provided research support to Bayer, Bristol-Myers Squibb, Elan, Eisai, GE, Genentech, Janssen Alzheimer Immunotherapy, Medivation, and Pfizer, as well as to the National Institute on Aging's Alzheimer Disease Neuroimaging Initiative and Dominantly Inherited Alzheimer Network, the Alzheimer Association, the Norman and Rosalie Fain Family Foundation, the John and Happy White Foundation, and the Champlin Foundation. Dr Wei is an employee of Janssen Alzheimer Immunotherapy. Dr Black is an employee of Pfizer. Dr Grundman has been a consultant for Adamas, American Life Science Pharmaceuticals, Avid, Biogen-Idec, Elan, Eli Lilly, Helicon, Intellect Neurosciences, Janssen Alzheimer Immunotherapy, Johnson & Johnson, and Teva Pharmaceuticals. He also owns stock in Elan and has been named on patents associated with Alzheimer immunotherapy. Dr Liu is an employee of Janssen Alzheimer Immunotherapy.
Funding/Support: This study was funded by Elan (acquired by Janssen Alzheimer Immunotherapy in 2009) and Wyeth Pharmaceuticals (acquired by Pfizer in 2009).
Role of the Sponsors: Employees of both sponsor companies (Janssen Alzheimer Immunotherapy and Pfizer) were involved in the study design, collection, analysis, and interpretation of data, as well as in the development and submission of this article.
Additional Contributions: We thank the patients and all of the investigators who took part in this study. We acknowledge the following individuals for their assistance: Cherry Lucas, BA, Erika Jones, BA, and Robert Schilling, MS (study management); Kristen Morris, MS (pharmacovigilence); Rezi Zawadski, PhD, and Keith Gregg, PhD (statistical analyses); John Baer, PhD, and Kay Jing, PhD (clinical study summary); Daniel Kidd, PhD, and Bert Blank, MS (bioanalysis); and Monica Christiansson, BS, and Åsa Kjällén, BS (cerebrospinal fluid analyses). We also thank members of the safety monitoring committee: Sid Gilman, MD, FRCP (chair); Chris Clark, MD; Douglas Galasko, MD; and Diane Griffin, MD, PhD.