Beyond PIB: the next generation of amyloid-imaging ligands

Authors


Dr Tetsuya Suhara MD, PhD, Department of Molecular Neuroimaging, Molecular Imaging Center, National Institute of Radiological Science, 4-9-1 Anagawa, Inage-ku, Chiba 263-8555, Japan

Pittsburgh Compound B (PIB) is the commonly used name for N-methyl-[11C]2-(4′-methylaminophenyl)-6-hydroxybenzothiazole (6-OH-BTA-1), an amyloid-imaging ligand that is widely used in the US, Europe, Australia, and Japan. Before Klunk and Mathis1 developed [11C]-PIB successfully at Pittsburgh University, a number of research groups had worked to find appropriate compounds as amyloid-imaging ligands in vivo. Attempts to find suitable compounds for positron emission tomography (PET) or single photon emission computed tomography (SPECT) studies started in the early 1990s. Radiolabelled monoclonal antibodies targeted to β-amyloid (Aβ) peptide were exploited as amyloid-imaging agents in vivo.2 However, the primary limitation of this approach is that large molecular weight biomolecules cannot efficiently cross the blood–brain barrier (BBB) and bind to amyloid plaques contained within the parenchyma of the Alzheimer's disease (AD) brain. Mathis et al.1 at Pittsburgh University initially focused their attention on Congo red, an amyloid-staining agent widely used in histological studies of post-mortem AD brain tissue sections. Congo red is a large molecule, a negatively charged sulfonate under physiological conditions, and is too hydrophilic to penetrate the BBB. Mathis et al.1 then investigated the binding properties of Chrysamine G, a more lipophilic and potent Aβ ligand. However, efforts over more than 10 years to develop radiolabelled analogs of Congo red and Chrysamine G as amyloid-imaging agents for PET and SPECT were hampered by the relatively poor brain penetration of these compounds.

The first successful in vivo attempt to image Aβ in the brain of AD patients used PET and the malononitrile derivative [18F]2-(1-{6-[(2-fluoroethyl)methyl-amino]-2-naphthyl}ethylidene)malononitrile ([18F]-FDDNP), which was developed by Agdeppa et al.3 at the University of California, Los Angeles.4 This ligand is a derivative of 2-dialkylamino-6-acylmalononitrile-substituted napththalenes (DDNP), which is a hydrophobic, viscosity sensitive fluorescent probe. [18F]-FDDNP has a high affinity for both Aβ and neurofibrillar tangles (NFT). Shoghi-Jadid et al.4 reported the initial results from PET studies in 2002. The PET images showed high accumulation of radioactivity in the frontal, temporal, and parietal cortex in AD patients following injection of [18F]-FDDNP. Unfortunately, signal-to-background ratios are low in the brain with this ligand and [18F]-FDDNP PET only showed 9% higher cortical uptake in AD brain than in healthy controls.5

Mathis et al.1 then turned their attention to thioflavin T, a dye for amyloid and with a small molecular weight. They removed positively charged heterocyclic nitrogen from thioflavin T to allow it to readily cross the BBB and 6-hydroxylated it for rapid clearance from the brain without Aβ. The resulting compound, [11C]-6-OH BTA-1, was named ‘PIB’ at the University of Uppsala, where the first clinical study was performed with this ligand and PET. In 2004, Klunk et al. reported the initial results of a [11C]-PIB PET study in 9 healthy volunteers and 16 patients with AD.6[11C]-PIB PET showed a 70% increase in the cerebral cortex in AD patients compared with healthy controls. Many researchers in PET centers in the US, Europe, Australia, and Japan followed their lead to start PET studies with [11C]-PIB. At present, [11C]-PIB PET studies are underway at seven PET centers in Japan and will soon be started at three others.

The [11C]-PIB PET studies showed that almost all AD patients have high PIB binding in the cerebral cortex,6 two-thirds of patients with mild cognitive impairment (MCI) have high PIB binding in the cerebral cortex similar to AD,6 80% of patients with Lewy bodies have high PIB binding in the cerebral cortex,7 and 10–20% of healthy elderly people have high PIB binding in the cerebral cortex.8 Patients with amyloid angiopathy also have high Aβ accumulation in the cerebral cortex, especially in the occipital cortex.9 Longitudinal studies on healthy controls, MCI patients, and AD patients are underway in several institutions, and are expected to reveal the natural course of amyloid deposition in elderly people and AD.

The third successful in vivo attempt to image amyloid in the brain of AD patients, performed in Canada, used the stilbene derivative [11C]-4-N-methylamino-4′-hydroxystilbene ([11C]-SB-13).10 In Japan, Kudo et al.11 investigated benzoxazole derivatives as amyloid-imaging ligands and developed [11C]-2-(2-[2-dimethylaminothiazol-5-yl]ethenyl)-6-(2-[fluoro]ethoxy)benzoxazole ([11C]-BF-227) as a PET tracer. [11C]-BF-227 may be unique because this ligand specifically binds to cored or mature plaques, whereas [11C]-PIB seems to bind not only to neuritic plaques, but also to some extent to diffuse plaques.12 However, these amyloid-imaging compounds have a lower signal-to-background ratio in the brain of AD patients than [11C]-PIB. At present, [11C]-PIB seems to be the best ligand for imaging Aβ in the AD brain by PET.

Does PIB fulfill all the requirements of an amyloid-imaging ligand in routine clinical studies? Carbon-11 is rapidly decayed, with a short half-life of only 20.4 min, which limits the use of PIB to PET centers with an on-site cyclotron and radiochemistry expertise. Fluorine-18 has a half-life of 110 min, meaning that fluorine-18-labelled compounds, such as [18F]-fluoro-d-deoxyglucose ([18F]-FDG) can be delivered to many PET centers from radiopharmaceutical companies. In the US, 95% of PET centers can receive a supply of an [18F]-compound from radiopharmaceutical companies. In Japan as well, many PET centers are covered by the supply of [18F]-FDG from a radiopharmaceutical company. Amyloid-imaging ligands labelled with fluorine-18 would greatly facilitate amyloid imaging with PET and it would be a profitable business.

SPECT is more widely available than PET, based on the fact that SPECT scanners are less expensive. Iodide-123 is a single photon emitter with a half-life of 13 h. 99mTechnetium is also a single photon emitter, has a half-life of 6 h, and could be produced at institutions by milking. Therefore, amyloid-imaging ligands labelled with [123I] or [99mTc] would also greatly facilitate amyloid imaging.

Several companies are interested in the development of amyloid-imaging ligands labelled with [18F], [123I], or [99mTc]. In 2003, GE Healthcare licensed a number of compounds from the University of Pittsburgh, including PIB, and started large multisite trials with fluorine-18-labelled PIB in 2007. Avid Radiopharmaceutical Incorporation is a new venture company specializing in the development of imaging agents for AD and Parkinson's disease. They investigated stilbene derivatives and developed a series of ligands that share common structural features with PIB. Bayer Schering Pharma licensed one of the ligands developed by Avid Inc., known as [18F]-BAY94-9172 or trans-4-(N-methyl-amino)-4′-{2-[2-(2-[18F]-fluoro-ethoxy)-ethoxy]-ethoxy}-stilbene (also as [18F]-AV1/ZK). [18F]-BAY94-9172 has been used at the Center for PET, Austin Health, in Australia.13 PET images similar to [11C]-PIB PET images have been obtained in patients with AD and healthy controls with [18F]-BAY94-9172 and PET. The mean neocortical uptake 90–120 min after injection of [18F]-BAY94-9172 was 57% greater in AD patients than in healthy controls.13

[123I]-IMPY was developed as an amyloid-imaging ligand for SPECT.1 Unfortunately, the signal-to-background ratio in the brain of AD patients is low and there is considerable overlap in the target/cerebellum ratio between healthy controls and patients with AD. The development of better SPECT ligands for Aβ plaque is well underway.

Apart from amyloid imaging with PET or SPECT, Nakada et al.14 recently reported the direct visualization of Aβ plaques in patients with AD in vivo by magnetic resonance microscopy on a 7T clinical system.

In parallel with the development of amyloid-imaging techniques, clinical trials of many anti-amyloid therapeutic drugs are underway.15 A Phase III study with monoclonal antibody therapy (bapineuzumab (AAB-001); Elan Corporation, Athlone, Ireland), is ongoing and this agent may be approved for clinical use within a few years in the US and Europe (see http://clinicaltrials.gov/ct2/results?term=bapineuzumab). Several clinical trials of gamma secretase inhibitors or modulators are also underway.

The development of amyloid-imaging techniques and anti-amyloid therapy may dramatically change the clinical treatment of AD within a few years. Elderly people, perhaps from 60 years on, may receive a brain checkup, including amyloid imaging, every 5 years. If they are found to have amyloid deposition in the cerebral cortex, they will then be treated with anti-amyloid drugs. The number of AD patients may decrease markedly if anti-amyloid therapy is proven effective. Although we still do not know whether the amyloid cascade hypothesis is true, we are waiting further developments with high expectations. A new diagnostic–therapeutic paradigm to successfully address AD and its harbinger, MCI–amnestic type, is emerging.

Ancillary