Date:



Date: December 10, 2005

To: Maryland Pao, MD, Chair, IRB, NIMH

Recommended by: Robert B. Innis MD, PhD, Chief, Molecular Imaging Branch, NIMH

Protocol Title: PET imaging of brain peripheral type benzodiazepine receptors

Identifying Words: Pharmacokinetics, compartmental analysis, distribution volume, identifiability

Principal Investigator: Masahiro Fujita, MD, PhD, NIMH

Associate Investigators: Masao Imaizumi, MD, PhD, NIMH

Janet Sangare, MSN, C-RNP; NIMH

Yong Hoon Ryu, MD, PhD; NIMH

Robert B. Innis, MD, PhD; NIMH

Estimated Duration of Study: two years

Study Subjects Number Sex Age Range

Healthy controls 15 M & F 18-40

Off Site Project: NO

Project uses ionizing radiation: YES

Project uses Durable Power of Attorney: NO

Table of Contents

I. PRECIS 4

II. INTRODUCTION 4

A. Type of Protocol 4

B. Background 4

C. Research question 9

D. Background of Approach 9

E. Qualifications of investigators 9

III. STUDY DESIGN AND METHODS 9

A. Study design 9

B. Overview 9

C. Study phases 10

D. Sample stratification 10

E. Sample size justification 10

F. Data analysis 10

G. Justification for the use of placebo, medication washout, or provocative stimuli 11

IV. SUBJECT ENROLLMENT 11

A. Recruitment - sample composition and characteristics 11

B. Inclusion criteria 11

C. Exclusion criteria 11

D. Study initiation and screening methods 11

V. PROCEDURES 11

A. Details of method 11

B. Details of assessment by study phase 13

C. Details of secondary procedures 13

D. Relationship to other studies proposed 13

VI. PROVISION OF CARE TO RESEARCH SUBJECTS 13

A. Concomitant clinical care 13

B. After care 13

C. Reasons for discontinuation from study 13

D. Toxicity criteria 14

VII. HUMAN SUBJECT RISKS AND PROTECTIONS 14

A. Consent and assent procedures 14

B. Risks of study participation and minimization of risks 14

C. Benefits of study participation 16

D. Investigator conflicts of interest 16

E. Privacy and confidentiality provisions 16

F. Adverse event reporting 16

G. Data and safety monitoring processes 16

H. Subject compensation 16

VIII. PHARMACEUTICAL, BIOLOGIC AND/OR DEVICE INFORMATION 16

A. Source 16

B. Relevant pharmacology 17

C. Toxicity 17

D. Formulation and preparation 17

E. Stability and storage 17

F. Incompatibilities 17

G. Administration procedures 17

IX: REFERENCES 18

X. APPENDIX: REIMBURSEMENT SCHEDULE 22

I. PRECIS

The peripheral benzodiazepine receptor (PBR) is distinct from central benzodiazepine receptors associated with GABAA receptors. Although PBR was initially identified in peripheral organs such as kidneys, endocrine glands and lungs, later studies identified PBR in the central nervous system. In normal conditions, PBR is expressed in low levels in some neurons and glial cells. PBR can be a clinically useful marker to detect neuroinflammation because activated microglial cells in inflammatory areas express much greater levels of PBR than in microglial cells in resting conditions.

PBR has been imaged with positron emission tomography (PET) using [11C]1-(2-chlorophenyl-N-methylpropyl)-3-isoquinoline carboxamide (PK11195). However, this classical ligand provides only low levels of specific signals and is not sensitive to detect changes occurred in vivo. Recently we developed a new ligand, N-acetyl-N-(2-methoxybenzyl)-2-phenoxy-5-pyridinamine [11C]PBR28, which showed much greater specific signals than [11C]PK11195 in non-human primates. In the present protocol, we plan to perform a kinetic brain imaging study in healthy human subjects to measure PBR in brain regions with [11C]PBR28. Successful development of a PET ligand to image PBR will have a strong impact on clinical management of brain disorders with inflammation such as multiple sclerosis and ischemia and neurodegenerative disorders such as Alzheimer’s and Parkinson’s disease where inflammation is involved in the disease progression.

II. INTRODUCTION

A. Type of Protocol

Healthy subjects will be studied to measure PBR in brain by performing brain PET imaging studies with [11C]PBR28. This study will be performed with an intravenous injection of up to 20 mCi of [11C]PBR28 and imaging for 2 – 3 h.

B. Background

Since their discovery in the late 1950’s, benzodiazepines have been widely used as anxiolytics, anticonvulsants and sedative medications (Sternbach 1983). In the last decades, two pharmacologically distinct subclasses of benzodiazepine binding site have been demonstrated. One class, the central benzodiazepine receptor (CBR) is mainly localized on the extracellular domain of the γ-aminobutyric acid (GABA)A receptor and regulates the chloride channel of GABAA receptors in the central nervous system (CNS) (Tallman et al 1978).

The second class of benzodiazepine receptor was initially identified in peripheral tissue and was called the peripheral benzodiazepine receptor (PBR). PBR is located on the mitochondrial outer membrane in several organs including the kidney, nasal epithelium, lung, heart and endocrine organs such as the adrenal gland, testis and pituitary gland (Anholt et al 1985; Anholt et al 1986; Braestrup et al 1977; Gavish et al 1999). Contrary to the original nomenclature of “peripheral” benzodiazepine receptors, later studies demonstrated the presence of PBR in the CNS (Schoemaker et al 1981; Weissman et al 1984; Zisterer and Williams 1997). In mitochondria, PBR belongs to the mitochondrial permeability transition pore (Anholt et al 1985; Bernassau et al 1993; McEnery et al 1992), where it is intimately associated in trimeric complex with adenosine nucleotide translocase and the voltage-dependent anion channel. This complex is a megachannel located on the inner and outer mitochondrial membrane contact sites. Although the physiological role of PBR in still unclear, PBR has been implicated in various functions such as neurosteroid synthesis (Culty et al 1999; Papadopoulos et al 1997), immunomodulation (Zavala 1997), cell proliferation (Okuyama et al 1999; Schlichter et al 2000; Verma et al 1998) and apoptosis (Bono et al 1999). In the normal brain, PBRs are mainly found in glial cells and are especially highly localized to the ependyma lining of the ventricles, choroids plexus, olfactory bulb (Anholt et al 1984; Benavides et al 1983; Cymerman et al 1986; Schoemaker et al 1983). The expression of PBR in vivo is reported to be increased in microglia activated by brain injury (Banati 2002; Banati 2003), and this increased has been used as an indicator of neuronal injury and neurodegenerative disease (Benavides et al 1987; Cagnin et al 2002).

Several compounds are able to cross the blood brain barrier and specifically bind to PBR in vivo. The most widely used selective ligand is 1-(2-Chlorophenyl-N-methylpropyl)-3-isoquinoline carboxamide (PK11195), which has been labeled with [11C] for positron emission tomography (PET) studies. Various neurologic disorders have been clinically investigated by using [11C]PK11195such as Alzheimer’s disease (Cagnin et al 2001a; Groom et al 1995), multiple sclerosis (Banati et al 2000; Debruyne et al 2003), stroke (Pappata et al 2000), Rasmussen’s encephalitis (Banati et al 1999), herpes encephalitis (Cagnin et al 2001b), amyotrophic lateral sclerosis (Turner et al 2004) and multiple system atrophy (Gerhard et al 2003). However, the brain uptake is very low (only about the same as the average activity in the entire body) with low ratios of specific binding to nondisplaceable radioactivity (less than 20%), which is not high enough for stable quantitative analysis (Pappata et al 1991; Pappata et al 2000; Sauvageau et al 2002).

In the last several years, a new class of high affinity PBR ligands has been radio-synthesized based on aryloxyanilides (Okuyama et al 1999), and some promising PET radioligands have already been developed from this class (Maeda et al 2004; Zhang et al 2003). We have also sought to develop PET ligands with high brain uptake and high ratios of specific binding to nondisplaceable radioactivity based on aryloxyanilides. We successfully developed a new PET ligand ([11C]PBR28) with high affinity ([11C]PBR28; IC50 = 0.6 nM measured with [3H]PK11195) and selectivity. In addition, in monkeys, these PET ligands showed high brain uptake and high ratios of specific binding to nondisplaceable radioactivity (Briard et al 2005).

The objective of this study was to fully characterize pharmacokinetics of this PET ligand by performing compartmental analysis with arterial input function and also by analyzing the composition of radioactive chemicals in the rat brain with high performance liquid chromatography (HPLC).

Studies in Nonhuman Primates.

Two rhesus monkeys (Macacca mulatta, body weight) were used (Table 1). Anesthesia was initiated with i.m. injection of ketamine (10 mg/kg) and then maintained under anesthesia with 1.6% isoflurane and 98.4% O2. The electrocardiograph (ECG), body temperature, heart and respiration rates were measured throughout the experiment. Body temperature was maintained at 37.0-37.5°C.

All of PET scans were performed on a GE Advance scanner (General Electric Medical Systems, Waukesha, WI), with reconstructed resolution of 6 mm full-width half-maximum in all directions in 3D mode by applying scatter correction. The high resolution research tomograph (HRRT, Siemens/CPS, Knoxville, TN, USA) scanner (Schmand et al 1998; Wienhard 2002). All scans of HRRT were acquired in 64-bit list mode format. Data were reconstructed into a 256x256x207 image matrix with pre-determined frame schedule using a list mode OSEM algorithm (Carson et al 2003), resulting in an image resolution of 2.5 mm FWHM. No scatter correction was applied. After a transmission scan, the radiopharmaceutical (dose: 4.17±1.36 mCi, specific activity: 1530±580 mCi/μmol) was intravenously injected. Coronal slices covering the whole brain were obtained. PET scans were acquired for 120-180 min (33-45 frames with longer scan duration at later time points). To measure plasma concentration of [11C]PBR28 and the metabolites, a second line, intra-arterial, in the contralateral limb was used to obtain 14 blood samples. Eight samples (0.5mL each) were drawn at 15 s intervals until 2 min, followed with 1-mL samples at 3, 5, 10, 30, 60, 90, 120min in heparin-treated syringes. Each blood sample was separated into plasma and blood cell fraction by centrifugation.

The tomographic images were analyzed with PMOD 2.65 (pixelwise modeling computer software; PMOD Group, Zurich, Switzerland) (Burger et al 1998). All frames of the original reconstructed PET data were summed, and this summed image was coregistered to a T1- weighted magnetic resonance (MR) image acquired separately on a GE 1.5 T Signa MR scanner (SPGR, TR/TE/flip angle = 13.1 ms/5.8 ms/45°, 0.4 x 0.4 x 1.5 mm and coronal acquisition on a 256 x 256 x 60 matrix) (GE Medical Systems,Waukesha, WI) using SPM2 (Wellcome Department of Cognitive Neurology, London, U.K.), and regions of interest were defined on the frontal, temporal , parietal and occipital cortices, cerebellum, putamen, thalamus, 3rd ventricle and 4th ventricle of the MRI. To normalize brain uptake relative to the injection dose and the body weight, standardized uptake values (SUVs) were determined as (% Injected activity/ g brain) × g body weight.

Estimation of distribution volume with arterial input function:

One-tissue (1C) and unconstrained two-tissue compartment (2C) models were applied. Rate constants (K1, k2, k2', k3, and k4) were defined as described previously (Laruelle et al 1994).

In the 1C,

VT=K1 / k2'f1

where VT is the distribution volume for the single tissue compartment.

In the 2C, VT is described separately by the distribution volumes in nondisplaceable (VN) and specific binding compartments (VS).

[pic]

VN = K1/k2'f1

VS = K1k3/k2k4f1 = Bmax'/Kd

VT = K1(1+ k3/k4) k2f1

where Bmax' is unoccupied binding site density. Under tracer conditions, Bmax' = Bmax. In these two models, although the definition of K1 is the same, that of k2 and k2'is different. That is, k2 is transfer rate to the vascular compartment from the nondisplaceable compartment in the 2C, and k2'refers to the transfer from the total tissue compartment (Laruelle et al 1994).

Vi' is defined as

Vi' = f1Vi

These definitions indicate that Vi values are expressed relative to the free fraction of radioligand in plasma and that VT' values are expressed relative to the total (free plus protein-bound) concentration of radioligand in plasma. Non-linear least-squares analysis was performed on the VOI-generated time-activity data using PMOD 2.65. Parameters were estimated using the Marquardt algorithm (Bevington and Robinson 2003) with constraints restricting parameters to positive values.

Statistical analysis

Goodness-of-fit by nonlinear least squares analysis was evaluated using the model selection criterion (MSC), which is a modification of the Akaike information criterion (AIC) (Akaike 1974). MSC gives greater values for better fitting. Goodness-of-fit by 1C and 2C was compared with F statistics (Hawkins et al 1986). The standard errors (SEs) of non-linear least squares estimation for rate constants were given by the diagonal of the covariance matrix (Carson 1986) and expressed as a percentage of the rate constants (coefficient of variation, %COV). In addition, %COV of VT' was calculated from the covariance matrix using the generalized form of error propagation equation (Bevington and Robinson 2003), where correlations among parameters were taken into account. A value of P ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download