Effect of selective COX-2 inhibition on neuroinflammation in Parkinson's disease.

Local substantia nigra brain stem neuroinflammation mediated by microglial
activation has been suggested to play a pivotal role in the pathogenesis of
PD1-3. Animal studies have strongly suggested the relevance of microglia
activation to the nigral cell death seen in PD patients; however the relation
of microglia activation and PD progression in vivo in human PD patients remains
unclear.

In vivo activated glial cells in the brain can be measured using the
radiotracer
[C-11]PK11195, a peripheral benzodiazepine ligand, and positron emission
tomography (PET)4. Recently, in PD patients increased inflammation was found in
the midbrain related to clinical asymmetry and inversely to the degree of
striatal dopaminergic dysfunction1. Gerhard et al5 found apart from the
mesencephalic increase of [11C]-PK11195 also cortical and basal ganglia regions
with increased uptake. They did not find correlation with clinical severity or
putamen [18F]-dopa uptake, nor longitudinal changes of microglial activation,
suggesting that microglia are activated early in the disease process and levels
then remain relatively static.
We recently studied neuroinflammation in PD patients with [11C]-PK11195 PET and
found increased tracer uptake in basal ganglia, thalamus and mesencephalon in a
pilot group of PD patients, compared to healthy control subjects (see protocol
METc 2006.153)

The role of neuro-inflammation in the pathogenesis and clinical course of
Parkinson*s disease needs to be further investigated. It may be possible to
inhibit further progression of disease with treatment aimed at reducing
neuro-inflammation.

Epidemiological data indicates that anti-inflammatory agents such as
non-steroidal anti-inflammatory drugs (NSAIDs) have a protective effect on
Parkinson*s disease 6. Regular intake of nonaspirin NSAIDs but also high doses
of aspirin have been associated with a 45 % lower risk of PD in two large
cohorts 7.
Increased expression of the enzyme cyclooxygenase-2 (COX-2) and elevated levels
of prostaglandin E2 (PGE2) have been implicated in the cascade of deleterious
effects leading to neurodegeneration 8. Teismann et al found that COX-2
expression is induced specifically within SNpc dopaminergic neurons in
postmortem PD specimens and in the MPTP mouse model of PD during the
destruction of the nigrostriatal pathway. A lack of COX-2 but not of COX-1
decreased MPTP neurotoxicity. The selective COX-2 inhibitor rofecoxib blocked
ventral midbrain PGE2 production in MPTP injected mice and attenuated neuron
and fiber loss, demonstrating the crucial enzymatic function of COX-2 to its
neurotoxic effects on dopaminergic neurons8;9.
Until now, anti-inflammatory treatment has only been investigated in animal
parkinsonian models. Increased expression of cyclooxygenase type 2 (COX-2) and
production of prostaglandin E2 synthesis have been implicated in
neurodegeneration in several parkinsonian animal models. Selective inhibition
of cyclooxygenase (COX-2) by celecoxib remarkably reduced the microglial cell
density in 6-OHDA lesioned parkinsonian rats. At later stages it also prevented
further dopamine cell loss10.

Anti-inflammatory treatment with celecoxib showed promising results in rodent
models for PD and can be studied safely in patient in vivo using [11C]-PK11195
PET. It will be important to investigate the effect of possible treatments
neurochemically in vivo in PD patients and to study whether [11C]-PK11195 PET
provides a sensitive method for evaluation of response in humans.

To measure in vivo effect on neuroinflammation of treatment with celecoxib, a
selective COX-2 inhibitor, in PD using PK111-95 and PET.
Celecoxib showed broad utility in animal models of neurodegeneration.
Neurochemical effect of the treatment on microglia activation in humans will be
measured in vivo comparing the PET scans of 10 early stage PD patients before
and after one month of treatment. An interim analysis will be perfomred after 5
patients have been investigated. If no effects of intervention are seen the
study will be stopped.



This study is meant to be a pilot study which tries to deliver a proof of
principle, namely whether neuroinflammation of the brain can be blocked by
available medication and whether this can be monitored by the applied
radiotracer scans. If a positive result is achieved then a larger study is
warranted to investigate longitudinally whether long-term treatment has a
positive clinical effect.

Synthesis of [11C]-PK111-95
[11C]-PK111-95 will be prepared through a reaction of [11C]-methyliodide with
(R)-desmethyl-PK11195 with potassiumhydroxide as a base (see Cremers J.E. et
al, Nucl.Med.Biol 1992) The product is purified by semi-preparative reversed
phase HPLC and meets the requirements of the responsible hospital pharmacist.

Toxicity of PK11195
PK11195 has been used in a clinical trial in daily doses of 200 to 400 mg
during 2 weeks, without toxic effect11. With the [11C]-labeled (R)-PK11195
tracer, many studies have been done in humans. The amount of labeled tracer,
using 400 MBq with a specific activity of 4000 GBq per mmole, will be 100
nmole, corresponding with 35 microgram of PK11195. Toxic effects of this amount
of tracer, being 1/10.000 of the amount used in the clinical trial, have never
been demonstrated.

Scanning methods
PET scans using [11C]-PK111-95 will be carried out at the dept. of Nuclear
Medicine and Molecular Imaging of the UMCG, University of Groningen.
The subject will be positioned with the head in the centre of the PET camera.
To improve the signal-to-noise ratio, a neuroshield will be applied. This
consists of a lead protective shield, which will be placed at the level of the
shoulders of the subject, to prevent detection of radiation from other body
parts (e.g. the bladder) by the camera. First, a 5 minutes transmission scan
will be performed to measure normal tissue absorption of the radiation. Then a
dose of 400 (minimum dose of 200) MBq [11C]-(R)-PK11195 will be injected in a
bolus followed by 3D dynamic emission scanning during 60 minutes. During the
emission scan, blood samples will be taken from a canula in the radial artery
using a blood monitor system, for measurement of radioactivity in the blood.
Extra blood samples will be taken by hand at 10, 20, 30, 45 and 60 minutes
after injection of the tracer, for analysis of tracer metabolites. With the
data of the emission scan and blood samples, the binding potential of the
tracer to the peripheral benzodiazepine receptor (PBR) in activated microglia
can be calculated on a pixel-by-pixel basis.
The tracer is cleared rapidly via the bladder. After the scan, the subject will
be asked to empty the bladder to minimize radiation dose.
After one month of treatment, the same PET scanning procedure will be repeated.


Radiation dose
In this study, the R-enantiomer of [11C]-(R)-PK11195 will be used, because this
tracer has better affinity for the peripheral benzodiazepine receptor (PBR) in
activated microglia than the racemic PK1119512. Radiation dose is calculated
using a model in which activity spreads homogenously over the body and no
activity is excreted. From the calculation according to the *Formules Shapiro
in Radiation Protection* (1992), radiation load for a 75 kilogram subject will
be 3,37 *10-3 mSv/MBq. Radiation load of a scan with 400 MBq [11C]-PK111-95 is
1,35 mSv, which is 0.8 times the annual radiation load of natural background
radiation (1.7 mSv in the Netherlands). For two PET scans, subjects are exposed
to a radiation load of 2,70 mSv. According to the standards of the
International Commission on Radiological Protection (ICRP62), the amount of
radiation reaches category 2b, minor to intermediate level of risk, 1-10 mSv.

MRI scan
Anatomical T1-weighted MRI scans of the brain will be obtained, for
co-registration with the analysis of the PET scan data. MR scans will be made
using the 1,5 Tesla machine at the department of Radiology of the UMCG.

Treatment
Patients are informed about the possible side effects of the treatment by
celecoxib as listed in the *Farmacotherapeutisch kompas 2005* and asked to
contact the investigators if any effect is experienced during the treatment. If
the drug is not tolerated, treatment will be stopped and the second PET scan
will be canceled. Treatment with celecoxib will start the day after the first
PET scan and will be stopped the night before the second PET scan.
Regular dosage of celecoxib in clinical practice is 100 to 200 mg twice a day.
We will use a dose of 100 mg twice a day. Most common side effects include
gastro-intestinal complaints as stomachache, nausea, diarrhea, dyspepsia and
flatulence, and fluid retention with edema. When mild gastro-intestinal
complaints occur, patients can receive treatment for stomach protection (the
proton-pump inhibitor omeprazol, 20 mg daily during the treatment with
celecoxib).

Twice half a day for preparation and performance of the radiotracer PET scans.
Half an hour for the accompanying MRI scan (once).

The scans do not pose any particular risk.

One month taking of 100mg celecoxib. Possibly unwanted effects like
gastrointestinal sigsn and symptoms may occur.