1. Transforming treatment and improving survival for ovarian cancer patients
Applications of genomic profiling for real time decisions in cancer treatment
Laura Shawver, PhD
Founder

2. Overview Ovarian Cancer statistics and treatment
Clearity Foundation: mission and process
Clearity profiling panel and results interpretation
Incorporating profiling in clinical trial designs

3. Ovarian cancer statistics and standard of care
22,280 new cases and 15,500 deaths estimated in the US in 2012
~75% diagnosed at stage III/IV
Most treated with surgical cytoreduction followed by adjuvant platinum-taxane based chemotherapy
Most patients recur within 2 years and receive multiple rounds of subsequent chemotherapy
27% survive >10 years

4. Multiple choices for recurrent disease
Can we inform the treatment decision?
NCCN Guidelines for Epithelial Ovarian Cancer/Fallopian Tube Cancer/Peritoneal Cancer

5. The Clearity Foundation launched as a non-profit organization in 2008 to:
Bring molecular profiling to the forefront of ovarian cancer diagnosis and treatment
Assist doctors in identifying therapy informed by their patient’s tumor molecular profile
Expedite the clinical development of novel targeted agents for ovarian cancer
Increase the probability of success by utilizing molecular profiling to select patients for clinical trials

6. Leading advisors and scientific presentations
Scientific Advisory Board
Beth Karlan, MD, Chair
Cedars Sinai & UCLA Medical Center
Doug Levine, MD
Memorial Sloan Kettering Cancer Center
Johnathan Lancaster, MD
Moffitt Cancer Center
Julie Cherrington, PhD
Pathway Therapeutics
Ursula Matulonis, MD
Dana Farber Cancer Center & Harvard Medical School
Deb Zajchowski, PhD
Clearity Foundation Scientific Director
Mol Cancer Ther; 11(2) February 2012: Treatment-related protein biomarker expression differs between primary and recurrent ovarian carcinomas DA Zajchowski, BY Karlan and LK Shawver
ASCO 2011: Expression Profiles in Matched Primary and Recurrent Ovarian Carcinomas
DA Zajchowski, BY Karlan and LK Shawver,
AACR 2011: Molecular Profiling in Recurrent Ovarian Cancer Patients DA Zajchowski, C Bentley, J Gross, BY Karlan and LK Shawver
AACR 2010: Selecting Patients for Ovarian Cancer Clinical Trials by Profiling Tumors against a Broad Panel of Molecular Markers
DA Zajchowski, J Gross, BY Karlan, K Bloom, D Loesch, A Alarcon and LK. Shawver 6

7. Ovarian cancers are molecularly heterogeneous
Genomic alterations observed in nearly every chromosome
Serous ovarian cancer
Any one genomic alteration is found in only a small fraction of patient tumors
 One drug will not be effective for the population
 Drugs must be developed for specific molecular tumor types
 Profiling of individual tumors is essential
 A broad profiling panel is necessary 7
Data provided by Dr. D. Levine

8. Oncologists and Patients
How Clearity works
Clearity Profiling Services:
Physician and patient education
Testing coordination
Secure data analysis
Results reported to patient
Oncologists and Patients
Patient and medical team using molecular profiles to prioritize therapeutic options 8

9. Chemotherapy Marker Panel Targeted Therapy Marker Panel
Use tumor molecular profiles to inform choice of chemotherapy and/or clinical trial
Chemotherapy Marker Panel
Targeted Therapy Marker Panel
Select chemotherapy for next treatment
Select chemo for combination with targeted agent in clinical trial
Select clinical trial with targeted agent 9

10. Key take away messages
Chemotherapy agents are targeted cytotoxic treatments
Chemotherapy will continue to play an important role but molecular targeted agents, on an individualized basis, will become increasingly important
Molecular profiling is available to help prioritize treatments and will be increasingly utilized
for chemotherapy agents
for molecular targeted agents
for clinical trials
Clinical trials should be considered early in the treatment process

11. Current panel of IHC tests
Growth Factors/ Receptors
EGFR*
Her2*
IGF1R
c-Met*
VEGF
PDGFR
Cytoplasmic Signal Transducers and Apoptosis Regulators
K-ras**
B-raf**
PIK3CA**
PTEN
Bcl-2
Survivin
Cox-2
Nuclear Signaling Proteins
Hormone Receptors/Transcription Factors
Cell Cycle ER AR PR
Ki67
p16 Rb
Chemotherapy Sensitivity Markers
DNA Synthesis/Transcription
ECM
TLE3
Topo1
Top2A
SPARC
Chemotherapy Resistance Markers
Drug Transporters
DNA Repair/ Modification
DNA Synthesis/Cell Division
BCRP
MRP1
MDR1/PGP
ERCC1
MGMT
RRM1 TS
TUBB3 11
* DNA amplification
** DNA mutational analysis

12. Data stored and analyzed in
Diane Barton Database
*200 patients; March 15, 2012 (n=242; Aug 15, 2012)

13. Data collected as histoscores
% tumor stained x Intensity =92 13

14. Marker expression in all patients provides basis for interpretation of individual results
Box, inter-quartile range; line, median; whiskers, maximum and minimum values

15. Clinical research evidence for biomarkers is used to interpret results

16. Chemotherapy selection uses published evidence and expression cut-offs derived from current database
High Topo I  Irinotecan, topotecan
High Topo II  Doxorubicin, etoposide
Low RRM1  Gemzar
Low TS  Fluoropyrimidines
High SPARC nab-Paclitaxel
High PGP  No Taxane, no doxil
High BCRP  No Topotecan
Low: <25th percentile High: >75th percentile

17. Case Study: profile for patient diagnosed in 3/2005 with stage IIIB papillary serous carcinoma17

18. Case Study: profile for patient diagnosed in 3/2005 with stage IIIB papillary serous carcinoma
High EGFR (98th percentile)
EGFR inhibitors have been ineffective in clinical trials although benefit seen in individual patients
Mutations can predict sensitivity and resistance to EGFR inhibitors in lung cancer
Follow up mutation analysis conducted; patient was wt 18

19. Case Study: profile for patient diagnosed in 3/2005 with stage IIIB papillary serous carcinoma
High ER
Anti-estrogens and aromatase inhibitors utilized in ovarian cancer patients but not approved due to lack of efficacy in clinical studies 19

20. Case Study: profile for patient diagnosed in 3/2005 with stage IIIB papillary serous carcinoma
High SPARC
Clinical trial for nab-paclitaxel (none at the time) or off-label use
patient had long history of taxane treatment 20

21. Case Study: profile for patient diagnosed in 3/2005 with stage IIIB papillary serous carcinoma
Low TS
Fluoropyrimidines and pemetrexed considered as a reasonable option 21

22. Case Study: profile for patient diagnosed in 3/2005 with stage IIIB papillary serous carcinoma
Low RRM
High RRM1 associated with resistance to gemcitabine
Gemcitabine is an approved agent in recurrent ovarian cancer 22

23. RRM1 is gemcitabine’s target and efficacy biomarker
For information on each marker, visit 23

24. Case Study: profile for patient diagnosed in 3/2005 with stage IIIB papillary serous carcinoma
Ge,zar
Gemzar (3/2010)
Gemcitabine
9/2011: NED 24

25. Pemetrexed, capecitabine
Often, only one of the commonly used agents to treat recurrent ovarian cancer is prioritized by the profile
Pemetrexed, capecitabine
Gemcitabine
Topo I inhibitors
Topo II inhibitors
Teal, tumor marker expression met quartile cutpoint criterion: RRM1, TS <25th percentile; TOP2A, TOP1 >75th percentile.
150/196 (76%) can be assigned to one of these agents. 25

26. Biopsy of recurrent disease is needed to obtain relevant profiling information
Mol Cancer Ther; 11(2) February 2012
Primary
Recurrence
EGFR
5% 2+; 10
68% 3+; 255 26

27. Marker expression differences in patient-matched primary and recurrent samples
Note that in some patients, recurrences are very similar to the primary (above) yet in others (below), they are very different as indicated by the number of arrows above the markers with altered expression. Cannot know whose recurrence will be different so need to biopsy all. In the second patient the two recurrences were within 5 months of each other. Some markers are more similar for the two recurrences (ERCC1, SPARC) and others more similar to the primary (Ki67, Topo1, MGMT).

28. Our knowledge and ability to rationally target ovarian cancer has evolved since 2008
Molecular characterization of ovarian tumors reveals new means for targeting and stratifying patients
Advances in methods for genomic characterization of tumor samples (next-gen sequencing in CLIA setting)
Increase in the number of molecularly targeted agents entering trials for ovarian cancer 28

29. —> extensive genomic interrogation necessary to characterize tumors
Low frequency of specific genetic aberrations in primary high grade serous ovarian carcinoma
—> extensive genomic interrogation necessary to characterize tumors 29
From TCGA study: Nature 474, (2011)

30. Genomic markers can be used to assign patients to clinical trial agents
PARP inhibitors RB
PI3K/RAS
HR Alterations/BRCAness
Notch
CDK inhibitors
AURK inhibitors
PI3K/AKT/mTOR inhibitors
MEK inhibitors
Notch inhibitors 30
From TCGA study: Nature 474, (2011)

31. New emphasis for profile: clinical trial selection
Chemotherapy Marker Panel
Targeted Therapy Marker Panel (IHC and DNA SEQ)
Select chemotherapy for next treatment
Select chemo for combination with targeted agent in clinical trial
Select clinical trial with targeted agent 31

32. Expanded panel provides more options for Clearity patients
ChemoTx panel
Taxanes
Gemcitabine
Doxil
Topo I inh
(Pemetrexed)
IHC
Clinical trial options
Targeted Tx panel
DNA repair inhibitors
Cell cycle inhibitors
Proliferation/
survival signaling inhibitors
IHC + DNA Seq*
* Next-gen exon sequencing of 182+ genes 32

33. Frequency of hotspot mutations in PIK3CA, KRAS, BRAF, and EGFR is histology-dependent
Clear cell: PIK3CA
Mucinous and carcinosarcoma: KRAS

34. Examples of gene alterations in recurrent
ovarian cancer
AMP
>5-fold
AMP
≤5-fold
SS, splice site mutation; FS, frameshift; TR, truncation

35. Case Study: profile for patient diagnosed in 2009 with stage IIIC clear cell carcinoma
4/2009
CDDP/tax (ip) x 3
Carbo/tax (iv) x3
12/2009
recurrence
Tumor profiled
1/2010
Topotecan +
AMG 386*
(clinical trial)
5/2011
Stable disease
Topotecan

36. 6/2011 Surgery -- residual disease
Case Study: profile for patient diagnosed in 2009 with stage IIIC clear cell carcinoma
4/2009
CDDP/tax (ip) x 3
Carbo/tax (iv) x3
12/2009 recurrence
Tumor profiled
1/2010 Topotecan +
AMG 386*
(clinical trial)
5/2011 Stable disease
IHC Results
Irinotecan
Can profiling and proper drug selection make ovarian cancer a chronic disease? Everolimus already approved and is in clinical trials in ovarian cancer.
6/2011 Surgery -- residual disease
Tumor profiled
DNA Sequencing Results
RxIrinotecan + Everolimus
Everolimus
9/2012 NED

37. Protein and DNA results are integrated into one report that highlights therapy options
SUMMARY of AGENTS
IHC Results
IHC results—Drug correlations
DNA Sequencing Results

38. Clearity report summarizes results from multiple labs and provides consensus interpretation
Summary of relevant patient medical history
Contact information for help with clinical trials, lab reports.
Number of patients whose data are included in Diane Barton Database
Individual profile compared to ovarian cancer population
Summary of agents (approved and in clinical trials) associated with clinical benefit extracted from pg 2
Compilation of data from all labs with interpretation (percentile rank, potential drugs) 38

39. New emphasis for profile: clinical trial selection
Chemotherapy Marker Panel
Targeted Therapy Marker Panel (IHC and DNA SEQ)
Select chemotherapy for next treatment
Select chemo for combination with targeted agent in clinical trial
Select clinical trial with targeted agent 39

40. Vision of an Ovarian Cancer Clinical Trials Coalition (OCCTC)
Increase the probability of drug development success by utilizing molecular profiling to select patients for clinical trials
Facilitate access to eligible patients through patient advocate-driven clinical trials education, outreach, and established web portal for clinical trials identification and matching
Remove barriers for screening large number of patients for enrollment by testing for all drug biomarkers in every patient
Reduce costs of profiling assays and make efficient use of biopsy samples
Make more treatment options available for ovarian cancer patients 40

41. Summary
Ovarian cancer is heterogeneous and a broad profiling panel is needed to capture data relevant to each individual
Commonly utilized agents for treatment of recurrent ovarian cancer can be prioritized using molecular markers
Molecular profiling can help prioritize drugs being tested in clinical trials 41

42. Thank you! To everyone who has participated in the Clearity process
To our physician partners
To the Clearity team
Dr. Deb Zajchowski
Hillary Theakston
Kathleen Zajchowski
And our volunteers! 42

43. 43