Transcript Slide 1

Dose response curve
Response
100%
50%
0%
IC50, EC50
Dose
Growth of Tumor Xenografts in Mice after Drug Treatment
22 Days MEK Inhibitor
22 Days Vehicle Control
Primary tumor
sections:
DAPI
P-H3
DAPI
P-ERK
L3.6pl
w/o MEK-inhibitor
CK19
CK19
L3.6pl
DAPI
P-H3
DAPI
P-ERK
with MEK-inhibitor
CK19
CK19
Drug Sensitivities:
•IC50 values of CI-1040 for 22 pancreatic cancer cell lines were measured.
•The cell lines rank from sensitive to resistant: low IC50  sensitive; high IC50  resistant
•Why is that important to know?
Patients show often various degrees of sensitivities to drugs.
It would not be smart to treat a patient with a drug if he is not sensitive enough (side
effects).
It would be good to have a knowledge about drug sensivities of a patient in advance.
How can we get such a knowledge?
 measuring biomarkers for drug sensitivities (gene expression
profiling, comparative genomic hybridization, proteomics)
Binding of a strand of DNA to a
complementary strand of DNA or
RNA:
Adenine binds to Thymine
Guanine binds to Cytosine
In RNA Thymine is replaced by
Uracil.
The binding or annealing of
complementary strands is highly
sequence specific.
Microarray technology:
Examples
Gene expression profiling
Array based comparative genomic
hybridization (aCGH)
Array CGH (aCGH): a microarray based technology
Human chromosomes become
fragmentized
Reference DNA
isolated from
control
(Cy5)
Test DNA
isolated from
cancer cells
(Cy3)
gain in the test DNA
loss in the test DNA
balanced in the test
DNA
DNA fragments become spotted on
glass slide
Image (CCD camera)
and data analysis
using software
Example of an aCGH profile: Array contains roughly 2500 printed DNA
fragments covering the human genome at 1.4 Mbp resolution.
Log2 Mean Raw Ratio
1
2.5
3
5
7
9
11
13
15
17 19
21X
Panc - 4.21
1.5
0.5
-0.5
-1.5
Ratio > 0  Tumor DNA is overrepresented: gain in copy number (amplifications)
Ratio < 0  Tumor DNA is underrepresented: loss in copy number (deletions)
Combining drug sensitivities with array CGH:
•We have measured copy number changes in 22 cell lines.
•We also determined the drug sensitivities of those cell lines to CI-1040.
Question: Can we find copy number changes that correlate with sensitivity or resistance
to CI-1040?
More precise: Can we find signatures of copy number changes that characterize
sensitive or resistant cell lines?
Answer: Student T-test; linear regressions
Acknowledgements:
Martin McMahon (my boss)
McMahon Lab
Helen Diller Family Comprehensive Cancer Center UCSF