www.thelancet.com/neurology Published online March 1, 2010 DOI:10.1016/S1474-4422(10)70043-0 1 Articles Published Online March 1, 2010 DOI:10.1016/S1474- 4422(10)70043-0 See Online/Refl ection and Reaction DOI:10.1016/S1474- 4422(10)70055-7 Turku PET Centre and Clinical Research Services Turku, University of Turku and Turku University Hospital, Turku, Finland (J O Rinne MD) Division of Neuroscience and Medical Research Council Clinical Sciences Centre, Faculty of Medicine, Imperial College London, London, UK (D J Brooks MD, A A Okello MRCP, S Rodriguez Martinez de LIano MD) Institute of Neurology, University College London, London, UK (M N Rossor FRCP, N C Fox FRCP) Kingshill Research Centre, Swindon, UK (R Bullock MRCPsych) University of Pittsburgh Medical Center, Pittsburgh, PA, USA (W E Klunk MD, C A Mathis PhD) University of G��teborg, Sahlgrenska University Hospital, M��lndal, Sweden (K Blennow MD) California Pacifi c Medical Center, San Francisco, CA, USA (J Barakos MD) Janssen Alzheimer Immunotherapy, South San Francisco, CA, USA (E Liu PhD, K M Gregg PhD, D Schenk PhD, M Grundman MD) Elan Pharmaceuticals, South San Francisco, CA, USA (M Koller MD, D Schenk, M Grundman) and Pfi zer, Collegeville, PA, USA (R Black MD) Correspondence to: Juha O Rinne, Turku PET Centre, University of Turku, PO Box 52, 20521 Turku, Finland juha.rinne@tyks.fi 11C-PiB PET assessment of change in fi brillar amyloid-�� load in patients with Alzheimer���s disease treated with bapineuzumab: a phase 2, double-blind, placebo-controlled, ascending-dose study Juha O Rinne, David J Brooks, Martin N Rossor, Nick C Fox, Roger Bullock, William E Klunk, Chester A Mathis, Kaj Blennow, Jerome Barakos, Aren A Okello, Sofi a Rodriguez Martinez de LIano, Enchi Liu, Martin Koller, Keith M Gregg, Dale Schenk, Ronald Black, Michael Grundman Summary Background Carbon-11-labelled Pittsburgh compound B (����C-PiB) PET is a marker of cortical fi brillar amyloid-�� load in vivo. We used ����C-PiB PET to investigate whether bapineuzumab, a humanised anti-amyloid-�� monoclonal antibody, would reduce cortical fi brillar amyloid-�� load in patients with Alzheimer���s disease. Methods Patients with mild-to-moderate Alzheimer���s disease were randomly assigned to receive intravenous bapineuzumab or placebo in a ratio of seven to three in three ascending dose groups (0��5, 1��0, or 2��0 mg/kg). Each dose group was enrolled after safety review of the previous group. Randomisation was by interactive voice response system masking was achieved with numbered kit allocation. Patients, investigators, study site personnel, sponsor staff , and carers were masked to treatment. Patients received up to six infusions, 13 weeks apart, and had ����C-PiB PET scans at baseline and at weeks 20, 45, and 78. The primary outcome was the diff erence between the pooled bapineuzumab group and the pooled placebo group in mean change from screening to week 78 in ����C-PiB cortical to cerebellar retention ratio averaged across six cortical regions of interest. Analysis was by modifi ed intention to treat. This study is registered with EudraCT, number 2004-004120-12 ISRCTN17517446. Findings 28 patients were assigned to bapineuzumab (n=20) or placebo (n=8). 19 patients in the bapineuzumab group and seven in the placebo group were included in the modifi ed intention-to-treat analysis. Estimated mean ����C-PiB retention ratio change from baseline to week 78 was ���0��09 (95% CI ���0��16 to ���0��02 p=0��014) in the bapineuzumab group and 0��15 (95% CI 0��02 to 0��28 p=0��022) in the placebo group. Estimated mean diff erence in ����C-PiB retention ratio change from baseline to week 78 between the bapineuzumab group and the placebo group was ���0��24��(95% CI ���0��39 to ���0��09 p=0��003). Diff erences between the bapineuzumab group and the placebo group in the individual regions of interest were similar to the overall mean diff erence. Adverse events were typically mild to moderate in severity and transient. Two patients in the 2��0 mg/kg bapineuzumab group had transient cerebral vasogenic oedema. Interpretation Treatment with bapineuzumab for 78 weeks reduced cortical ����C-PiB retention compared with both baseline and placebo. ����C-PiB PET seems to be useful in assessing the eff ects of potential Alzheimer���s disease treatments on cortical fi brillar amyloid-�� load in vivo. Funding Elan Pharmaceuticals and Wyeth Research. Introduction Alzheimer���s disease is characterised neuropathologically by deposits of extracellular amyloid-�� plaques, intraneuronal neurofi brillary tangles, and cerebral neuronal loss.1 Antibodies directed against the N terminus of amyloid �� reduce cerebral amyloid-�� load in transgenic mice2,3 and block the synaptotoxic eff ects of amyloid-�� oligomers.4 Bapineuzumab, an antibody targeted against the N terminus of amyloid ��, is a passive immunotherapy that is being tested in Alzheimer���s disease.5 Bapineuzumab might bind to amyloid �� in the brain and facilitate its clearance.2,3 Amyloid-�� load can be measured in patients with Alzheimer���s disease by use of PET and the radiotracer carbon-11-labelled Pittsburgh compound B (����C-PiB).6,7 PiB is a thiofl avin analogue that binds with low nanomolar a��� nity to aggregated fi brillar deposits of the amyloid-��peptide, enters the brain at concentrations detectable by PET, and clears rapidly from normal brain tissue.6,8 At the low nanomolar concentrations used in PET studies, PiB selectively binds to fi brillar amyloid-�� deposits in post-mortem human brain.7,9 Compared with healthy controls, patients with Alzheimer���s disease have about two times greater retention of ����C-PiB in areas of the association cortex that are targeted by amyloid-�� deposits.6,10 ����C-PiB retention is similar in patients with Alzheimer���s disease and healthy controls in areas unaff ected by amyloid-�� deposition, such as the subcortical white matter, pons, and cerebellum.6

Articles 2 www.thelancet.com/neurology Published online March 1, 2010 DOI:10.1016/S1474-4422(10)70043-0 We aimed to investigate whether bapineuzumab- related changes in cortical amyloid-�� load could be measured in vivo by use of ����C-PiB PET imaging. Methods Patients We did a phase 2, multicentre, randomised, double-blind, placebo-controlled, ascending-dose study at three clinical sites (two in the UK and one in Finland) between August, 2005, and January, 2009. Eligible patients were aged 50���80 years inclusive, met NINCDS-ADRDA criteria for probable Alzheimer���s disease,11 and had amyloid-�� loads in the range expected for patients with Alzheimer���s disease, defi ned as ����C-PiB PET retention ratios relative to the cerebellum of 1��5 or more in at least three brain regions among the anterior cingulate, posterior cingulate, frontal, temporal, and parietal cortices. Additional inclusion criteria were MRI consistent with Alzheimer���s disease, a mini-mental state examination (MMSE) score of 18���26,12 and a Rosen modifi ed Hachinski ischaemic score of at least 4.13 Patients were excluded if they had clinically signifi cant neurological disease other than Alzheimer���s disease had a major psychiatric disorder had a history of stroke or seizures had a Hamilton rating scale for depression score greater than 12 14 or were currently taking anticonvulsants, antiparkinsonian, anticoagulant, or narcotic drugs, recent immunosuppressive or cancer chemotherapy drugs, or cognitive enhancers other than acetylcholinesterase inhibitors or memantine at a stable dose for at least 120 days before screening. The study was approved by the local independent ethics board at each site, and each patient (or a legally authorised representative) gave written informed consent before enrolment. Randomisation and masking Patients were randomly assigned to receive either intravenous bapineuzumab or placebo in one of three dose groups (0��5, 1��0, or 2��0 mg/kg). Patients who completed the screening phase and met all inclusion criteria were eligible for randomisation. An interactive voice response system (IVRS) vendor generated the treatment assignments, and visit-specifi c blinded study drug kits were dispensed by the IVRS. The IVRS vendor personnel had no further contact with study site staff , patients, or carers. On the basis of the information provided by the IVRS vendor, a masked pharmacist or dispenser took the study drug kit with the allocated kit number and prepared the infusion. During the study, patients, investigators (both image analysts and clinical assessors), study site personnel, and sponsor staff were masked to treatment. A four-member independent safety monitoring committee assessed the safety of treatment throughout the trial. The 0��5 mg/kg dose group was fi rst to be enrolled. Each subsequent dose group was enrolled after the safety monitoring committee reviewed safety in the preceding groups. Procedures Patients received study drug as a 1 h intravenous infusion every 13 weeks for up to six infusions. Each patient had ����C-PiB PET, fl uorine-18-labelled-fl uorodeoxyglucose (�����F-FDG) PET, clinical assessments of cognition and function (activities of daily living), volumetric and clinical MRI, and clinical laboratory investigations (eg, haematology). CSF amyloid beta and tau concentrations were measured in a subset of patients. Final assessment was at week 78. Imaging was done at two PET sites (one in the UK and one in Finland). Details of the synthesis of ����C-PiB and PiB PET data collection have been described.10 Briefl y, all ����C-PiB images were obtained by use of an ECAT EXACT HR+ scanner (Siemens, Erlangen, Germany) after an attenuation scan that preceded an intravenous bolus of about 370 MBq ����C-PiB (specifi c activity ���10 GBq/��mol at injection). Images were obtained in 32 frames over 90 min. Cortex to cerebellar ratio images of ����C-PiB retention were generated at a single site (Hammersmith Imanet, GE Healthcare, London, UK) by use of data from 60���90 min after injection.10 ����C-PiB PET images were co-registered to each patient���s MRI, which was normalised into standard stereotactic space (Montreal Neurological Institute, Quebec, Canada). A probabilistic brain atlas was used to create a standard template of regions of Figure 1: Trial profi le *Patient excluded because enrolment in 2��0 mg/kg group was ended by the sponsor. ���15 of 20 patients had PET data at week 78: two withdrew and scans were not completed in three. ���Five of eight patients had PET data at week 78: two withdrew and scans were not completed in one. PiB=Pittsburgh compound B. 20 assigned to receive bapineuzumab (safety population) 7 to 0��5 mg/kg 7 to 1��0 mg/kg 6 to 2��0 mg/kg 8 assigned to receive placebo (safety population) 3 to 0��5 mg/kg 3 to 1��0 mg/kg 2 to 2��0 mg/kg 53 assessed for eligibility 2 withdrew 2 had adverse events 2 withdrew 1 had adverse events 1 owing to loss of caregiver 1 had no post-baseline PET data 1 had no post-baseline PET data 25 failed screening 23 did not meet inclusion criteria 1 withdrew consent 1 other reasons* 28 randomised 19 included in PiB PET analysis (modified intention-to-treat population)��� 7 included in PiB PET analysis (modified intention-to-treat population)���