Transcript Receptors/receptor tracers/kinetic models

concepts for discussion
1.
2.
synopses – don’t overdo it
how comfortable were you with Wagner paper, Mintun, Farde?
3.
4.
5.
6.
7.
8.
9.
10.
non-specific vs specific binding
how to measure non-specific binding (displacement)
can we do displacement in humans? maybe…
qualitative imaging vs quantitative imaging
“tracer” vs “drug” … vs “transmitter”
how to create a kinetic (compartmental) model for a PET tracer
what is a compartment?
what is a mass-balance? Conservation of mass (not conservation of radioactivity; not conservation of
concentration…)
11.
12.
13.
14.
what do we measure with PET? But what quantity do we need to model?
why arterial sampling (need input function)
are all parameters estimatable? consider y = m*n*x + b
what are simulations good for?
15. binding potential – what does it represent?
16. Binding potential may be an equilibrium measure, but do we need to reach equilibrium to measure it?
(how ‘bout your adult height?)
17. drug occupancy – how do we measure it with PET?
18. how did Farde do it? What assumptions did he make? Do you accept them?
ENAS 880 – Sept 13, 2011
Recall from last time…
Tracer in each state has a unique temporal trajectory
early
late
which is which ?
striatum
tracer concentration
What the
PET
scanner
sees
Bound
cerebellum
Time (minutes)
Free
Plasma
why is most of the brain blue?
are these “early” or “late”
images?
is this structural or functional?
is this a ‘detection’ or a
‘characterization’ experiment?
Fowler et al., Science & Practice Perspectives, April 2007, 4-16
what is the point of excess cold spiperone?
what do we learn from it re the caudate?
can we compare these numbers across
scans?
why divide by cerebellum value?
could we do this exp’t in a human?
‘late’ or ‘early’ image?
whose brain is this?
why a CT scan first?
why increasing contrast w/ time?
why increasing noise w/time?
0.8
1.8
3.4
t
(B+F)/F
B/F
Main points/questions/criticisms – Wagner et al.
1. so who really did the “first in vivo scan” of DA receptors in a primate?
2. how did this get into Science (and not Mintun)?
3. can we assume that cerebellum is exactly the same as the ‘free’ in the
striatum?
4. whose brain was it?
First PET Image Of D2 Receptors
in Brain
First PET Image Of D2 Receptors
in Brain
understand this?
got equilibrium?
•where do error bars come from?
•(if so) is the Y axis correct?
•where does “specific putamen”
come from – special brain area
known only in Sweden?
•what is plotted? (B vs F) – how
did they get these values?
•how did they get 5 data points
•what is the slope near origin?
•what is the assymptotic value for
Y axis?
where does “expected spec binding” come
from?
Is this a safe comparison? what factors are
ignored?
raclo in cereb (assume = free) --> plug into Bmax*Free/[Kd + Free] using Bmax
and Kd from Healthy controls --> predicts B at equilib for schizo patients.
Compare to actual B measured for patients (via [striat-cereb] per injected dose –
I hope)
receptor occup = DB/B (fractional change in binding due to presence of bound
drug)
= (Bpredicted – Bactual)/Bpredicted
first ‘drug occupancy’ study with PET?
claim: uniformity of occupancy of D2 by different drugs suggests that
occupancy level is related to efficacy.
What’s all this about models in PET?
Dopaminergic synapse
Free
raclopride
Bound
raclopride
‘stuck’
raclopride
Plasma
raclopride
The modeling process
Imagine a
living brain
Consider the
relevant
neurochemistry
(a)
(b)
dF ( t )
dt
P
Abstract the
neurochemistry
into a series of
connected
pools
F
B
dB ( t )
dt
N
dN ( t )
dt
(c)
 K 1 P ( t )  k 2 F ( t ) ...
 k on ( B max ...
'
 k 5 F ( t ) ...
Describe
exchange
between
pools by a
series of
differential
equations
(d)
Morris, E.D., C.J. Endres, K. Schmidt, B.T. Christian, R.F. Muzic, R.E. Fisher. Kinetic Modeling in PET. In Emission Tomography: The
Fundamentals of PET and SPECT, ed. M. Wernick and N. Aarsvold. Academic Press. Amsterdam, Netherlands, 2004, pp. 499-540.
what’s all this about?
what is F?
conservation of mass
typically written as…
P
F
B
under what condition?
what is the blue? what is the fraction of P that is
blue – according to Mintun? according to us?
parameter estimation – what
is it? why is it iterative?
simulations? what are they
good for? what are they used
for here?
sensitivity analysis – huh?
kon*Bmax =
koff
Bmax
KD
BP is an equilibrium ratio (has no
units of time or rates) but we can
get it from a TRANSIENT
experiment.
anything tricky about using an in
vitro sample for anatomy?
what’s the difference between expt
1 and expt 2 in same baboon?
why didn’t they estimate BP in 2nd
study?
Schematic Diagram of Ligand Binding
“Rest” condition
BP = B/F
at steady state
endogenous NT
unlabeled tracer
radiolabeled tracer
Schematic Diagram of Ligand Binding
“Rest” condition
loss of receptors
BP ↓
endogenous NT
unlabeled tracer
radiolabeled tracer
Schematic Diagram of Ligand Binding
DA-release condition
DA ↑
BP ↓
endogenous NT
unlabeled tracer
radiolabeled tracer
DBP is the (fractional) difference in BP
between conditions
DBP= (BP1-BP2)/BP1
terms for next week: Sept 20
displacement
release
endogenous
simulation
what is Logan proving by her simulations?
where they get their parameter vals
what things might be causing DA release in
Koepp expt? what might have been a better
control condition(s)? what do (+) changes in
binding potential (Fig 2) mean?
what is the ‘proof’ that DA release is being
measured?
what is being optimized in Morris paper?
where do they get their parameter vals?