research paper need to be edit
sarrona91
Florbetapir F-18: A Histopathologically Validated Beta-Amyloid Positron Emission Tomography Imaging Agent John Lister-James, PhD,* Michael J. Pontecorvo, PhD,* Chris Clark, MD,* Abhinay D. Joshi, PhD,* Mark A. Mintun, MD,* Wei Zhang, PhD,* Nathaniel Lim, PhD,* Zhiping Zhuang, PhD,* Geoff Golding, PhD,* Seok Rye Choi, PhD,* Tyler E. Benedum, PhD,* Paul Kennedy, PhD,* Franz Hefti, PhD,* Alan P. Carpenter, PhD,* Hank F. Kung, PhD,† and Daniel M. Skovronsky, MD, PhD*
Florbetapir F-18 is a molecular imaging agent combining high affinity for !-amyloid, pharmacokinetic properties that allow positron emission tomography (PET) imaging within a convenient time after dose administration, and the wide availability of the radionuclide fluorine-18. Florbetapir F-18 is prepared by nucleophilic radiofluorination in approximately 60 minutes with a decay-corrected yield of 20%-40% and with a specific activity typically exceeding 100 Ci/mmol. The florbetapir F-18 dissociation constant (Kd) for binding to !-amyloid in brain tissue from Alzheimer’s disease (AD) patients was 3.7 " 0.3 nmol/L, and the maximum binding capacity (Bmax) was 8800 " 1600 fmol/mg protein. Autoradiography studies have shown that florbetapir F-18 selectively binds to !-amyloid aggregates in AD patient brain tissue, and the binding intensity is correlated with the density of !-amyloid quantified by standard neuropathologic techniques. Studies in animals revealed no safety concerns and rapid and transient normal brain uptake (6.8% injected dose/g at 2 minutes and 1.9% injected dose/g at 60 minutes in the mouse). Florbetapir F-18 has been well- tolerated in studies of more than 2000 human subjects. Biodistribution studies in humans revealed predominantly hepatobiliary excretion. The whole body effective dose was 7 mSv from a dose of 370 MBq. The pharmacokinetic of florbetapir F-18 make it possible to obtain a PET image with a brief (10 minutes) acquisition time within a convenient time window of 30-90 minutes after dose administration. Clinical studies have demonstrated a clear correlation between in vivo PET imaging with florbetapir F-18 and postmortem histopatho- logic quantitation of !-amyloid in the brain. Semin Nucl Med 41:300-304 © 2011 Elsevier Inc. All rights reserved.
Alzheimer’s disease (AD) is the most common cause ofdementia in elderly people.1 Despite the improvements in clinical diagnosis effected by the application of standard- ized criteria,2,3 definitive diagnosis remains dependent on pathologic confirmation at autopsy.4 Diagnostic tools to as- sist in in-life diagnosis of this dreaded affliction are much sought after. Currently accepted pathologic definitions of AD1-5 require the presence of abnormal levels of !-amyloid neuritic plaques in the brain, and the absence of this patho-
logic marker is inconsistent with the diagnosis of AD. Con- sequently, an in vivo test that could rule out the presence of pathologically significant levels of !-amyloid plaque in the brain could thus rule out a diagnosis of AD, even in patients with clinical signs and symptoms consistent with this most common form of dementia. This could lead to more appro- priate treatment decisions for those with non-AD causes of dementia. It has been suggested also that a test that could indicate abnormal levels of !-amyloid in the brain could also add confidence to a clinical diagnosis of AD.6
This article summarizes the chemistry, pharmacology, and clinical experience with florbetapir F-18 (18F-AV-45), a !-amyloid–avid imaging agent that was selected from a series of styryl pyridines (azastilbenes) developed by Kung, Skov- ronsky and colleagues7,8. The results of a clinical study dem-
*Avid Radiopharmaceuticals, Philadelphia, PA. †Department of Radiology, University of Pennsylvania, Philadelphia, PA. Address reprint requests to Dr John Lister-James, Avid Radiopharmaceuti-
cals, 3711 Market St, Philadelphia, PA 19104. E-mail: listerjames@ avidrp.com
300 0001-2998/11/$-see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1053/j.semnuclmed.2011.03.001
onstrating a clear correlation between in vivo positron emis- sion tomography (PET) imaging with florbetapir F-18 and postmortem histopathologic quantitation of !-amyloid in the brain has been published recently.9 A New Drug Application for florbetapir F-18 injection (AmyvidTM; Avid Radiopharma- ceuticals, Philadelphia, PA) was under review by the U.S. Food and Drug Administration at the time of preparation of this manuscript.
Chemistry of Florbetapir F-18 Florbetapir F-18 (Fig. 1) is a pyridyl stilbene derivative that contains fluorine-18 substituted for the terminal hydroxyl group of a triethylene glycol side-chain. Nonradioactive flor- betapir F-19 has been fully characterized, including by sin- gle-crystal radiograph structural analysis (Fig. 2).
The radiosynthesis of florbetapir F-18 has been described in detail.8 In principle, the chemistry (Fig. 3) involves the nucleophilic displacement of a tosylate group of the precur- sor by radioactive fluoride, followed by removal of a tert- butoxycarbonyl protecting group by using acid. In common with other radiofluorination, the radioactive fluoride is first isolated by using ion exchange chromatography, taken up in a cryptand 2.2.2/potassium carbonate/aqueous acetonitrile solution and then evaporated to dryness in a reaction vessel. Radiofluorination is carried out by heating the precursor and the dried radiofluoride in dimethyl sulfoxide. Deprotection is performed by heating in aqueous HCl. Purification is per- formed by reversed-phase chromatography, first with a dis- posable cartridge and then with high-performance liquid chromatography (HPLC). HPLC solvent is removed by ad- sorbing the purified florbetapir F-18 on another reversed- phase cartridge, rinsing with water and elution with ethanol. Ascorbate is included in all of the purification steps to mini- mize radiolytic degradation. The process has been adapted to most commercially available radiosynthesizers of both the clean-in-place fluid-path type (eg, Siemens Explora GN/LC [Siemens AG, Munich, Germany] and GE TRACERlab FXF-N
[GE Healthcare, St Giles, UK]) and the disposable cassette fluid-path type (eg, ORA Neptis [ORA sprl, Brussels, Bel- gium] and [modified] GE TRACERlab MX). Radiosynthesis can be accomplished in 50-70 minutes, depending on the type of equipment used, and decay-corrected yields are typ- ically 20%-40% at the 0.5-1.5 Ci scale. The radiopharmaceu- tical is produced no-carrier-added, and specific activity typ- ically exceeds 100 Ci/mmol.
Florbetapir F-18 is formulated in physiological saline with 10% ethanol to maintain solubility and ascorbate to enhance radiolytic stability. The final drug product is sterilized by 0.22 "m sterile-filtration. The quality of florbetapir F-18 in- jection is controlled through extensive end-product testing, including tests for radiopharmaceutical identity (by compar- ison with fully characterized florbetapir F-19 reference stan- dard), chemical and radiochemical purity, radionuclide identity and purity, identity of formulation components, po- tential synthesis impurities and bacterial endotoxins, and for sterility. Stability studies have confirmed that florbetapir F-18 injection meets all release specifications through its 10- hour shelf-life when stored at room temperature.
Pharmacology of Florbetapir F-18 The pharmacology of florbetapir has been described in de- tail.8 In vitro studies with both homogenates and sections of brain tissue from patients with neuropathologic diagnoses of AD according to National Institute on Aging–Reagan Insti- tute Consensus Group criteria2 have shown that florbetapir binds selectively and saturably to !-amyloid in the brain tissues of these patients. By using the previously character- ized !-amyloid ligand, 125I-2-(4=-dimethylaminophenyl)-6- iodo-imidazo[1,2-a]pyridine (IMPY), as the probe, the disso- ciation constant for florbetapir displacement of 125I-IMPY from !-amyloid in AD brain tissue homogenates was 5.5 ! 0.7 nmol/L (unpublished data). The inhibition constant (Ki) of PIB was 2.8 ! 0.5 nmol/L by using the same procedures.10
Again by using AD patient brain tissue homogenates, the Ki for inhibition of binding of florbetapir F-18 by florbetapir F-19 was 2.9 ! 0.2 nmol/L. BAY94-9172, GE-067, and Pitts- burgh compound B (PIB) Ki values were 2.2 ! 0.5, 0.7 !
Figure 1 Florbetapir F-18.
Figure 2 ORTEP drawing of florbetapir with 30% probability ther- mal ellipsoids.
Figure 3 Florbetapir F-18 synthesis.
Florbetapir F-18: a !-amyloid PET imaging agent 301
0.4, and 0.9 ! 0.2 nmol/L, respectively, in this assay, sug- gesting overlapping binding sites. In further equilibrium binding studies with brain tissue homogenates from 4 AD patients, the florbetapir F-18 dissociation constant (Kd) was 3.7 ! 0.3 nmol/L, the maximum binding capacity (Bmax) was 8800 ! 1600 fmol/mg protein, and the data fit a single bind- ing site model. As part of the studies performed to support the safety of florbetapir, no binding with a Kd of "1 "mol/L was found with any of a battery of 46 G protein– coupled and ion channel receptors, further supporting the specificity of florbetapir for !-amyloid.
Autoradiography studies in brain tissue sections ob- tained post mortem from neuropathologically diagnosed AD patients (Fig. 4) incubated with florbetapir F-18 ex vivo revealed a clear correlation between localization of radioactivity and !-amyloid plaques visualized by thiofla- vin-S staining.
The correlation between florbetapir binding density and !-amyloid density was studied by using postmortem brain tissue sections for a broad range of AD patients. Tissue sec- tions were stained by using generally accepted methods for staining for !-amyloid, namely Bielschowsky’s silver stain, thioflavin-S, and immunohistochemistry with monoclonal antibodies specific for 3 different !-amyloid epitopes. All studies showed that florbetapir F-18 binding correlated in a statistically significant manner with conventional methods of !-amyloid plaque localization (correlation coefficient, 0.72- 0.94). A confocal microscopic study with thioflavin-S fluo- rescence and florbetapir F-18 autoradiography provided ad- ditional convincing evidence of the localization of florbetapir on !-amyloid plaques. Furthermore, the autoradiography studies showed that florbetapir F-18 localized specifically to !-amyloid plaques and not to fibrillary tangles that are com- posed predominantly of tau protein. Thus, these studies have shown that florbetapir F-18 selectively binds to !-amyloid aggregates in human brain tissue, and the binding intensity is quantitatively correlated with the density of !-amyloid quan- tified by standard neuropathologic techniques.
Nonclinical Safety Studies of Florbetapir As might be expected considering the extremely low mass of a dose of florbetapir F-18, no toxic effects were observed in single-dose (at 100 times the maximum human dose [MHD], allometrically scaled) and 28-day repeated-dose (at 25 times the MHD, allometrically scaled) toxicity studies performed in rats and dogs, and there were no observations during cardio- vascular (dogs), respiratory (dogs), or central nervous system (rats) safety pharmacology studies at up to 100# the MHD. Drug-drug interaction studies were performed to assess the potential for drugs commonly used in the AD patient popu- lation to interfere with florbetapir binding to !-amyloid. Drugs tested included the most commonly used nonsteroidal anti-inflammatory drugs and acetylcholinesterase inhibitors and the most commonly used cholesterol-lowering drug, an- tidiabetic, antipsychotic, anxiolytic, and antidepressant drugs. None of these drugs affected florbetapir binding at their anticipated maximum plasma concentrations.
Pharmacokinetics, Distribution, and Metabolism of Florbetapir F-18 Nonclinical Studies In biodistribution studies in the normal mouse, after intrave- nous injection of florbetapir F-18, the concentration of radio- activity in the brain was highest at 2 minutes after injection and then washed out rapidly (6.8% injected dose [ID]/g at 2 minutes, reducing to 1.9% at 60 minutes, 1.7% at 120 min- utes and 1.5% at 180 minutes), whereas concentration of radioactivity in blood remained relatively unchanged (1.9%- 2.5% ID/g) during the first 120 minutes after injection.8 Ra- dioactivity in blood, brain, and other major regions of uptake during the 3-hour period after injection are summarized in Fig. 5. Thus, radioactivity distributed predominantly to the liver with eventual accumulation in the gastrointestinal tract.
Figure 4 Anti–!-amyloid immunohistochemistry with 4G8 and florbetapir F-18 autoradiography from adjacent sec- tions of human brain tissue Top row, 4G8 immunohistochemistry; pseudocolor images show tissue !-amyloid aggre- gation per unit area (0%-30%). Bottom row, florbetapir F-18 autoradiography.12
302 J. Lister-James et al
Dosimetry studies in mice suggested that the upper colon and intestines would be the critical organs in humans.
Two major metabolites were identified from in vivo stud- ies, the desmethyl derivative AV-160 and the acetylated des- methyl derivative AV-267.8 Independent injection of the ra- diolabeled metabolites revealed that both can cross the blood-brain barrier, and both wash out rapidly from the brain. Both metabolites have lower affinity for !-amyloid than does florbetapir (AV-160, Ki $ 54 nmol/L and AV-267, Ki $ 400 nmol/L), and there was no evidence of specific binding to !-amyloid in autoradiography studies of the bind- ing of radiolabeled metabolites and AD patient brain sec- tions. Studies with human liver microsomal preparations re- vealed that florbetapir was rapidly metabolized to AV-160, with a biological half-life of approximately 5 minutes.
Clinical Results of Florbetapir F-18 Florbetapir F-18 has been administered to more than 2000 subjects; 496 of these subjects took part in completed studies included in the New Drug Application. In these studies flor- betapir was well-tolerated. The most commonly reported ad- verse events have been headache, musculoskeletal pain, nausea, and fatigue. Adverse events were typically mild, tran- sient, and usually considered by the investigator to be unre- lated (remote) to florbetapir F-18 treatment.
A preliminary estimate of absorbed radiation dosimetry of florbetapir F-18 in humans11 was consistent with the expo- sure previously predicted by studies in mice and has been confirmed in a larger study conducted by Avid. In humans, florbetapir F-18 rapidly clears from the circulation after in- travenous injection, leaving "5% of the injected radioactivity remaining in blood after 20 minutes. The plasma terminal half-life is between 20 and 90 minutes, and by 90 minutes after injection, radioactivity remaining in the blood (approx- imately 2% ID) is mainly in the form of polar metabolites. Approximately 17% of the ID is excreted in the urine over 200 minutes after injection, and most of the radioactivity in the urine is in the form of polar metabolites. The organs with the greatest exposure are gallbladder wall, small intestines, and liver. The average human whole body effective dose was determined to be 7 mSv for the recommended 370-MBq flor- betapir F-18 dose (exclusive of computed tomography/trans- mission scan), which is comparable to the exposure for ap- proved 18F-labeled compounds such as fluorodeoxyglucose.
Fig. 6 shows the typical pattern of tracer uptake for a presumed !-amyloid positive (A!%) subject with a clinical
Figure 5 Radioactivity concentration (% ID/g) in selected tissues in the mouse after intravenous injection with florbetapir F-18.8
Figure 6 Florbetapir F-18 PET imaging (coronal, axial, and sagittal views). Top left, healthy control (SUVR $ 0.98; visual read score, $ 0); top right: patient with clinically diagnosed AD and interpreted as A!% (SUVR $ 1.68; visual read score $ 3); bottom left, patient with mild cognitive impairment and interpreted as A!& (SUVR $ 1.03; visual read score $ 0); bottom right, patient with mild cognitive impairment and interpreted as A!% (SUVR $ 1.61; visual read score, $ 4).
Florbetapir F-18: a !-amyloid PET imaging agent 303
diagnosis of AD and a presumed !-amyloid negative (A!-) cognitively healthy control subject. Patients with AD selec- tively accumulate tracer in cortical areas expected to be high in amyloid deposition, whereas typical cognitively healthy control subjects show only minimal cortical tracer reten- tion.11,12 However, an A!-pattern of tracer uptake was seen on florbetapir PET scans for some clinically diagnosed AD patients, consistent with literature reports that 10%-20% of clinically diagnosed AD patients do not have amyloid pathol- ogy at autopsy. Conversely, A!% scans were seen in approx- imately 15%-20% of the cognitively normal elderly subjects, consistent with literature findings from other PET amyloid imaging agents and reports that cognitively healthy elderly subjects can have significant !-amyloid pathology at au- topsy. Images from mild cognitive impairment subjects could be either A!% or A!- (Fig. 6).
For quantitative assessment, images were fitted to a PET template in Talairach space, and standard volumes of interest for cortical target regions and cerebellum reference region were applied.9 The mean cortical target to cerebellum ref- erence ratio (standard uptake value ratio [SUVR], where SUV $ [(Volume interest (Bq/cc) # subject weight (g)/in- jected dose (Bq))] increased rapidly from 0 to 30 minutes after administration of florbetapir F-18 and approached as- ymptote thereafter with little change in SUVR between 30 and 90 minutes.13 SUVR at a representative time point, 50-60 minutes after administration, was consistently and signifi- cantly greater for AD subjects than for cognitively healthy controls.9,11,13
The phase III pivotal trial9 was designed to evaluate the correlation between the level of cortical amyloid burden on PET scan and true !-amyloid burden assessed by postmor- tem histopathology. Cortical amyloid burden on florbetapir PET scans was visually assessed on a 0-4 scale (no to high cortical tracer uptake) by 3 independent raters, blinded to clinical information, with the median rater score as the pri- mary outcome variable, and by a semiautomated quantitation of the SUVR in 6 cortical target areas (frontal, temporal, pa- rietal, precuneus, anterior and posterior cingulate) relative to the cerebellum reference region. True !-amyloid burden was assessed at autopsy by quantitative immunohistochemistry (primary outcome variable) and by a direct microscopic as- sessment of the density of neuritic plaques (modified Biel- schowsky’s silver stain) in the 6 cortical target regions. There was a strong, statistically significant correlation between the level of cortical tracer uptake in the PET image, whether assessed by median visual read or SUVR, and true !-amyloid burden, whether assessed post mortem by quantitative im- munohistochemistry or silver stain (# $ 0.71-0.78, P " 0.0001). Results for individual readers were similar to the results for the median score. These results provided strong
evidence that florbetapir PET can identify the presence of pathologically significant !-amyloid in living individuals.
In summary, florbetapir F-18 has been shown to bind with high affinity to !-amyloid in the brain and to localize specif- ically to !-amyloid plaques in brain tissue from patients meeting neuropathologic criteria for AD. It is reproducibly prepared with high specific activity and high radiochemical purity on standard automated synthesis equipment. The pharmacokinetics of florbetapir F-18 make it possible to ob- tain a PET image with a brief (10 minutes) acquisition time within a convenient time window of 30-90 minutes after dose administration. Clinical studies have demonstrated a clear correlation between the in-life assessment of !-amyloid in the brain derived from the evaluation of florbetapir F-18 PET images and histopathologically confirmed !-amyloid plaques in the brain determined at autopsy.
References 1. Alzheimer’s Association. Alzheimer’s disease facts and figures. Alzhei-
mers Dementia 6, 2010 2. Hyman BT, Trojanowski JQ: Consensus recommendations for the post-
mortem diagnosis of Alzheimer disease from the National Institute on Aging and the Reagan Institute Working Group on diagnostic criteria for the neuropathological assessment of Alzheimer disease. J Neuro- pathol Exp Neurol 56:1095-1097, 1997
3. Mirra SS, Heyman A, McKeel D, et al: The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD): part II—standardization of the neuropathologic assessment of Alzheimer’s disease. Neurology 41: 479-486, 1991
4. McKhann G, Drachman D, Folstein M, et al: Clinical diagnosis of Alz- heimer’s disease: 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, 1984
5. Knopman DS, DeKosky ST, Cummings JL, et al: Practice parameter: diagnosis of dementia (an evidence-based review)—report of the Qual- ity Standards Subcommittee of the American Academy of Neurology. Neurology 56:1143-1153, 2001
6. Roe CM, Mintun MM, Ghoshal N, et al: Alzheimer disease identification using amyloid imaging and reserve variables: proof of concept. Neurol- ogy 75:42-48, 2010
7. Zhang W, Kung MP, Oya S, et al: 18F-labeled styrylpyridines as PET agents for amyloid plaque imaging. Nucl Med Biol 34:89-97, 2007
8. Choi SR, Golding G, Zhuang Z, et al: Preclinical properties of 18F-AV- 45: a PET agent for A! plaques in the brain. J Nucl Med 50:1887-1894, 2009
9. Clark CM, Schneider JA, Bedell BJ, et al: Use of florbetapir-PET for imaging !-amyloid pathology. JAMA 305:275-283, 2011
10. Zhang W, Oya S, Kung MP, et al: F-18 polyethylene glycol stilbenes as PET imaging agents targeting aggregates in the brain. Nucl Med Biol 32:799-809, 2005
11. Lin KJ, Hsu WC, Hsiao IT, et al: Whole-body biodistribution and brain PET imaging with [18F]AV-45, a novel amyloid imaging agent: a pilot study. Nucl Med Biol 37:497-508, 2010
12. Choi SR, Schneider JA, Bennett DA, et al: Alzheimer’s Disease & Asso- ciated Disorders (in press)
13. Wong DF, Rosenberg PB, Zhou Y, et al: In vivo imaging of amyloid deposition in Alzheimer disease using the radioligand 18 F-AV-45 (florbetapir F 18). J Nucl Med 51:913-920, 2010
304 J. Lister-James et al