Synergism - Telenet Service
Download
Report
Transcript Synergism - Telenet Service
The Search for Synergism
A Data Analytic Approach
L. Wouters, J. Van Dun, L. Bijnens
May 2003
Three Country Corner
Royal Statistical Society
Overview
Combined action of drugs
Screening for synergism
Experimental Design
Fitting concentration response curves, estimation of
IC50
Graphical analysis of combined action
– isobolograms
– fraction plots
– combination index
2
Drug Combinations
Additive
Sub-additive: antagonism
fight against one another
Super-additive: synergism
work together
3
Drug Combinations:
Antagonism - Synergism
Major therapeutic areas:
– Oncology
– Infectious disease
Ideal combination:
– Synergistic for therapeutic activity
– Antagonistic for toxicity
4
Non-additivity and Statistical
Interaction
Drug A f(x), drug B g(x)
100
Combination: a + b, h(a,b)
80
f(a) = 50 %, g(b) = 60 %
additivity h(a,b) = 110 % ?
Drug can be antagonistic
with itself
% Effect
60
40
20
0
0.0
0.2
0.4
0.6
0.8
1.0
f(a) = 0%, g(b)=0%
additivity h(a,b) = 0% ?
Drug can be synergistic
with itself
Concentration
5
Problems with Synergism Antagonism
Synergism is controversial issue
Literature large but confusing
Different definitions
Different methods and experimental designs
Pharmacological - biostatistical approaches
Greco (1995) Pharmacol Rev 47: 331-385
6
Sarriselkä agreement (1992)
Combined
effect
Both agents
active (Loewe
model)
Both agents
active (Bliss
model)
Only one agent Neither agent
active
active
> predicted
Loewe
synergism
Bliss
synergism
Synergism
Coalism
= predicted
Loewe
additivity
Bliss
independence
Inertism
Inertism
< predicted
Loewe
antagonism
Bliss
antagonism
Antagonism
-
7
Loewe Additivity
ICx,A, ICx,B concentrations required for each drug
A, B individually to obtain a certain effect x (x %
inhibition)
Let Cx,A, Cx,B doses of drug A and drug B in the
combination that jointly yield same effect x
Drug A has lower potency ICx,A > ICx,B
Relative potency of A: ICx,A / ICx,B
8
Loewe Additivity (cont.)
Assume constant relative potency and additivity
Combination can be expressed as equivalent
concentrations of either drug :
ICx , A
ICx , A
Cx , A
C x ,B ICx , A , C A
C x. B ICx ,B
ICx ,B
ICx ,B
C x , A C x ,B
1
ICx , A ICx ,B
9
Methods Based on Loewe Additivity
Isobologram
Interaction index of Berenbaum (1977)
Bivariate spline fitting method of Sühnel (1990)
Hypothesis testing approach of Laska (1994)
Response surface methodology of Greco (1990),
Machado (1994)
10
Isobologram
Cx , A
ICx, A
C x,B
ICx, B
1 Cx, A ICx, A
ICx, A
ICx, B
C x, B
Cx, A
ICx, A
ICx, A
ICx, B
Antagonism
Synergy
ICx, B
Cx,B
11
Bliss Independence
i1, i2, i12 inhibition as a fraction [0; 1] by drug 1, drug 2,
and their combination
from a probabilistic point of view, when fraction i1 is
inhibited by drug 1, only (1 - i1) is available to respond to
drug 2. Assuming independence:
i12 i1 (1 i1 )i2 i1 i2 i1i2
can be reformulated in terms of u. = 1 - i., the fraction
remaining unaffected
u12 1 i12 1 1 u1 1 u2 1 u1 1 u2
u1u2
12
Bliss Independence
Counter-argument
A drug can be synergistic with
itself
75 % of control at 0.9 mg/kg
Assume a dose of 0.9 mg/kg of
the drug is combined with 0.9
mg/kg of the same drug
Total dose = 1.8 mg/kg
Under Bliss independence:
0.75 x 0.75 = 0.56 = 56 % for
combination
1.8 mg/kg yields 15.7 % of
control
13
Screening for Synergism in
Oncology
Screening experiment
– as simple as possible with limited resources
– carried out on a routine basis
– analysis must be automated
Screening experiments on tumor cells grown in
96-well microtiter plates
14
Screening Experiment
Requirements
– Unbiased estimates of responses
– Avoidance of confounding of random error
and drug effects
– Elimination of plate effects and plate
location effects in 96-well plates
15
Plate Location Effects in 96-well
Plates
Microtiter plates
contain a substantial
amount of
unexplainable
systematic error along
their rows & columns
(Faessel, et al. 1999)
Importance of
standardization
experiment (low,
middle, and high
response)
16
Standardization Experiment (n = 3)
Standardization
experiment at high level
of response, n=3
Within assay presence of
systematic differences of
important magnitude (up
to 50 %) in untreated
microtiter plates after
edge removal
Not repeatable between
different runs of assay
17
How to Eliminate Bias &
Confounding ?
Randomization assures:
– Equal probability to attain a
specific response for each
well
– Independence of results
– Absence of confounding
– Proper estimation of
random error
18
Experimental Design
Ray Design
Mixtures are composed based on preliminary
estimates of IC50 of constituents
Assuming additivity: IC50,Mixture fIC50, A 1 f IC50, B
f : mixturefactor
Construct concentration response curve for different
mixture factors:
D
r
u
g
A
Drug B
19
Ray Design
Composition of Mixtures
Tested concentration Ci of
mixture is composed of:
Ci kC fIC50, A 1 f IC50, B
D
r
u
g
f : mixturefactor
k : dilution factor
A
Proportion of constituents
in mixture:
Drug B
A
fIC50 , A
C
B
1 f IC50, B
C
20
Advantages of strategy
Simplified analysis:
– Consider mixture as new drug
– Fit concentration response curve to
different dilutions of mixture
Easy to carry out in laboratory
Limited number of samples
21
Layout of Screening Experiments in
Oncology
Ray design reference compound A, tested compound B
f = 0, 0.125, 0.25, 0.5, 0.75, 1
Experiments carried out in 3 independent 96-well plates
Dilutions (k): 10/1, 10/2, 10/3, 10/4, 1/1, 1/2, 1/4, 1/10
All dilutions tested within single plate
Wells for background and maximum effect
Allocation of different treatment is randomized within
plate by robot
22
Experimental Data
23
Percentages
24
Lessons from EDA
Asymptotes of sigmoidal curve not reached
always
Some part of sigmoidal curve is still present
Computing percentages makes sense (common
system maximum)
Proposed functional model:
100
yi
1 exp log IC50 log Ci
25
Fit of 2 Parameter Logistic
Ignoring Plate
26
Individual Fits of 2 Parameter
Logistic per Plate
27
Studentized Residuals versus Fitted
Values after Individual Model Fitting
28
Normal Quantile Plot of Pooled
Residuals after Individual Model Fits
29
Individual Estimates per Plate-Factor
30
Lessons from EDA for Functional
Model Fitting
Sigmoidal shape as described by 2-parameter
logistic model
Importance of plate effect even after correcting
for background, etc. by calculating
percentages
How to obtain reliable estimate of IC50 and
standard errors ?
31
Nonlinear Mixed Effects
Nonlinear Mixed Effects Model (Pinheiro,
Bates) allows to model individual response
curves within plates and provides reliable
estimate of standard error
Result = estimates and standard errors of
model parameters as fixed effects
32
Isobologram
• Decompose IC50,M of mixture
into IC50 of constituents C50,A
and C50,B :
C50, A A IC 50, M
Antagonism
C50, B B IC 50, B
• Plot of drug B versus
drug A and line of additivity
C50, A
IC50, A
C 50, B
IC50, B
Synergism
1
33
Fraction Plot
Based upon refined
estimates of IC50 of Drug A
and B recalculate the correct
fraction f :
A IC50, B
f
IC50, A A IC50, A IC50, B
Plot of IC50 of mixture
versus recalculated fraction
34
Combination Index
Chou and Talalay (1984)
CI
C 50 , A
IC50 , A
C 50 , B
IC50 , B
1 : synergism
1 : additivity
1 : antagonism
95% Confidence intervals by
parametric bootstrap (n = 10000)
based upon estimates and
standard errors from nlme fit
35
Conclusions
Present graphical approach appealing to
scientists
Still a lot to be done
– T. O’Brien’s approach (TOB)
– Incorporating design issues in TOB
– Alternative distributions (e.g. gamma)
– Optimal design
36