Precision Cancer Medicine: a 36-year journey

NCT/DKFZ

October 25, 2016

John Mendelsohn, M.D. Director,

Institute for Personalized Cancer Therapy and Past President,

The University of Texas MD Anderson Cancer Center

2

Disclosure Information John Mendelsohn, M.D.

I have the following relationships to disclose:

• I receive royalty payments from the University of California, which owns the Cetuximab patent.

• I have never prescribed Cetuximab, or treated a patient with Cetuximab.

3

Paul Ehrlich: Birth of Targeted Therapy (1) Antibodies: Nobel Prize for Serum Therapy in 1908 (2) Targeted Chemotherapy: 1901-11

Receptors on Cells A Bacterial Toxin AND A Targeted Chemotherapy

Postulated “side-chains,” or “receptors” specific for external substances (dyes), antigens and nutrients.

Model: bifunctional agent, containing a chemical structure that binds to the “receptor” linked to a toxic molecule.

4

Timeline in Genomic Medicine 1944 The genetic material is DNA 1950s Structure of DNA. Molecular biology 1960s The genetic code and machinery 1970s Manipulating and sequencing DNA Cancer is a genetic disease with Darwinian clonal evolution. 1980s Oncogenes and suppressor genes Therapies targeting products of aberrant genes 1990s Sequencing of the human genome 2000s Genomic medicine in the clinic

5

Hypothesis: 1980 John Mendelsohn

and Gordon H. Sato

Monoclonal antibodies which bind to EGF receptors and block access to EGF or TGF-α may prevent cell proliferation, by inhibiting activation of the EGF receptor tyrosine kinase.

Gordon H. Sato, Ph.D.

6

Rationale 1980

Stanley Cohen, Ph.D

• EGF characterized 19621. EGFR characterized 1975-80.2 (Cohen-Nobel Prize, 1986).

• Autocrine hypothesis: EGF or TGFα can autostimulate the cell’s EGFRs. (Todaro and Sporn).3

• Tyrosine kinase activity first identified in src oncogene and EGFR (Cohen, Hunter, Erickson).2,4,5

• Overexpression of EGFR common in human cancers (Ozanne, many others).6

• Preferential addiction of transformed cells. • “Experiments of nature.” Circulating autoantibodies against receptors can

cause stable physiologic change (disease): myasthenia gravis, thyroid disease and insulin resistance.

• Right technologies: nude mice, monoclonal antibodies. 1. Cohen S. J Biol Chem 1962;237:1555-1562; 2. Chinkers M, Cohen S. Nature 1981;290:516-519;

3. Sporn MB, Todaro GJ. N Engl J Med 1980;303:878-880; 4. Cooper JA, Hunter T. J Cell Biol 1981;91:878-883; 5. Erickson E, et al. J Biol Chem 1981;256:11381-11384; 6. Mendelsohn J, Baselga J. Oncogene 2000;19:6550-6565

7 • control group, o treated group

Cancer Research 44, 1002-1007, March 1984

8

Inhibition of P- Tyrosine by mAb225

A431 cells incubated with 32P, then (1) no addition, (2) EGF, (3) mAb225, (4) EGF + mAb225: immunoprecipitated with mAb528, gel electrophoresis, hydrolysis and 2D-thin layer electrophoresis.

Sunada, J Cell Physiol. 1990

9 Hanahan D and Weinberg RA, Cell 144, 5:646-674, 2011 HALLMARKS OF CANCER

10

Mechanism of Growth Inhibition

Peng, Cancer Res, 1996 (modified)

11

Transitional Cell Carcinoma - Dinney

Dinney

12

Dinney

Inhibition of Metastatic Capacity by C225

13 Masui H et al. Cancer Res 1986

Complement–Mediated Cytotoxicity Against 51Cr-A431 Cells

∆ 528 Ig G1 murine ο 225 Ig G2a murine

14

C225 Mediates ADCC on Melanoma Target Cells

Naramura et al, Cancer Immunol Immunother 1993

∎ Murine mAb

▨ human:murine chimeric mAb

J Natl Cancer Instit 85:1327-1333, 1993 15

Antitumor Effects of Doxorubicin in Combination With Anti-epidermal Growth Factor Receptor Monoclonal Antibodies Jose Baselga, Larry Norton, Hideo Masui, Atanasio Pandiella, Keren Coplan, Wilson H. Miller, Jr., John Mendelsohn

16

Days After Initial Treatment0 8 16 24 32 40 48 56 64 72 80

Tumo

r Size

(mm)

4

6

8

10

12

14

control C225x1 18Gy C225x1+18Gy C225x3 C225x3+18Gy

A431 Xenograft

Targeted Therapy: RT + C225

Days After Initial Treatment (completed by day 10) Milas L et al. Clin Cancer Res. 2000;6:701−708

17

Phase I Trial with 111In-murine 225 (ascites) in NSCLC: Hybritech

• No toxicity • Single dose up to 300 mg • At > 40 mg dose, imaged tumor and all metastases > 1 cm

detected by CT scan • At > 120 mg dose, serum level > 40 µg/ml after

3 days (capable of saturating receptors) • Antibodies against murine 225 within 2 weeks. NCI

subsequently funded production of human:murine chimeric mAb, C225.

• Subsequent FDA approval in 2004 for colon cancer, 2006 for head and neck cancer

Divgi…Mendelsohn, JNCI 1991

J Natl Cancer Inst 83:97-104, 1991 18

Phase I and Imaging Trial of Indium 111-Labeled Anti-Epidermal Growth Factor Receptor Monoclonal Antibody 225 in Patients with Squamous Cell Lung Carcinoma C. R. Divgi, S. Welt, M. Kris, F. X. Real, S. D. J. Yeh, R. Gralla, B. Merchant, S. Schweighart, M. Unger, S. M. Larson, J. Mendelsohn

Tumor uptake of 111In-mAb 225

Slide

19

20

Summary of Accomplishments 1980 - 1990

1. First hypothesis, with Dr. Gordon Sato, that an agent blocking activation of a growth factor receptor could inhibit cell proliferation.

2. First production of an agent that inhibited a receptor tyrosine kinase.

3. First clinical trial in humans with an agent targeting a growth factor receptor and a tyrosine kinase.

4. First clinical trial with a monoclonal antibody specifically designed to alter a biologic function, not to elicit an immunological response. In fact, it can do both.

21

Properties of C225 (Cetuximab)

• IgG1 (chimerized antibody), can bind complement and can mediate ADCC

• Binds with EGF receptor with high affinity (Kd = 0.2 nM)

• Competes with growth factor binding to receptor

• Inhibits activation of receptor tyrosine kinase

• Stimulates receptor internalization

22

Impass (1990-1993) • General skepticism about mAbs • Mechanism as a tyrosine kinase inhibitor

suggested other approaches (soluble inhibitors) • Hybritech (including license for 225) bought by

Lilly – no movement forward

• Solution: license to another company (ImClone 1993), later Merck (Germany) 1998, (Bristol-Myers Squibb 2001)

• Outcome: successful trials leading to regulatory approval (1994-2004).

• ImClone purchased by Lilly (2008)

23

C225 + Cisplatin treatment (patient had progressed on Cisplatin). Courtesy, Dr. W. K. Hong.

Patient reported in Shin et al, Clin Ca Res 2001

The Clinical Journey with C225: “Moment of Truth”

24

Clinical Anti-EGF Receptor Therapies

Tyrosine kinase Inhibitors:

Gefitinib, Erlotinib (others)

Monoclonal Antibodies:

Cetuximab, Panitumumab (others)

Signal Transduction

EG FR

EGF

K K

25

Summary of Clinical Trials Data with EGF Receptor Inhibitors

• Over a dozen different competing anti-EGFR agents are in clinical trials with a wide variety of cancer types. Five are approved by the FDA in the USA.

• Response rates as a single agent are generally 0-13%. Vary with agent and with tumor type.

• Patients with mutated EGF receptors have a higher response rate to the oral TKIs and can achieve prolonged stabilization of disease (years).

26

Lessons with oral EGFR TKIs 1. Drugs failed large Phase III trials, until it was

discovered that the few responders had mutations in EGFR.

2. Trials based on prescreening for mutated EGFR succeeded.

3. Survival with mutated EGFR was greater with TKI therapy, whereas survival with EGFR wild type was greater with standard chemotherapy.

4. Resistance developed, due to (1) another EGFR mutation or (2) activation of a bypass pathway.

27

Clinical and Biological Differences between mAb and TKIs

• Responses in colon cancer to mAb, not to TKIs.

• Gastrointestinal dose-limiting toxicity from TKIs, not mAb.

• Cetuximab is specific for EGFR. TKIs are not.

• Cetuximab downregulates EGFR. TKIs do not.

• Cetuximab reduces glucose uptake (not TK dependent).

• Cetuximab (Ig G1) mediates ADCC. TKIs do not.

• (Cetuximab and TKIs are additive against xenografts.)

EGFR Based Resistance to Targeted Therapies Drug Target Resistance Mechanisms Laboratory

Gefitinib Mutated EGFR (lung)

MET amplification activates ERB-B3

Engelman JA, et al. Science. 2007;316(5827):1039−1043

Cetuximab EGFR (colon) Mutated KRAS Van Cutsem E, et al. J Clin Oncol 2011;29:2011–2019

Vemurafenib BRAF (V600E) (colon)

Activated EGFR Bernards (Prahallad A) et al. Nature. 2012;483(7387):100-103

Vemurafenib

BRAF (V600E) (melanoma)

• Mutation in MEK • Increased copy number

BRAF • Aberrant BRAF • PTEN loss reduces

apoptosis • Increased IGF-RI

Many

Many Experimental Drugs

PIK3CA • Receptor tyrosine kinases

Engelman (Ebi H) et al. J clin Invest. 2011;121:4311-21

Crizotinib ALK rearrangements

• Activated EGFR Yamaguchi N et al. Lung Cancer 2014;83:37−43 28

29

Challenge: Kinetic Considerations

1. PC12 neuronal cell line EGF: rapid downregulation of receptor, short ERK activity: proliferation NGF: slower downregulation of receptor, prolonged ERK activity: differentiation Marshall, Cell, 1995

2. EGF stimulation of DNA synthesis in cultured resting 3T3 cells requires at least a 5-7 hour period of exposure. Brooks, Nature 1976, Das and Fox, PNAS 1978

3. Early and later stimulatory time points required for gene expression. Feedback regulatory loops.

Yarden, Nat. Rev. Molec. Cell Biol. 2006, 2011

30

Challenge: Mechanisms of Cell Cycle Arrest and Cell Death

1. Senesence

2. Autophagy

3. Apoptosis

4. Necrosis

31

Personalized Cancer Therapy – Recent Successes: Importance of Biomarkers

1. Trastuzumab for high-HER2 breast cancer. Slamon, NEJM, 2001 2. Imatinib, first for CML, then for GI stromal tumors with cKit mutations.

Drucker, NEJM, 2001, Demitri, NEJM, 2002. 3. PARP inhibitor olaparib for BRCA 1/2-associated cancers. Fong,

NEJM, 2009. 4. Gefitinib against the EGF receptor as first line therapy for advanced

NSCLC. Mok, NEJM 2009 5. Crizotinib for lung cancers with ALK-EML rearrangements. Kwak,

NEJM, 2010. 6. Vemurafenib for melanomas with BRAF V600E mutations. Flaherty,

NEJM, 2010.

32

Sheikh Khalifa Institute for Personalized Cancer Therapy:

Goals 1. Create the infrastructure and platforms for genetic

analysis of large numbers of clinical cancer specimens. Other “omics” to follow.

2. Support clinical trials bringing therapies to patients that target the genetic aberrations in their cancers.

3. Provide decision support to create personalized cancer treatment plans.

4. Demonstrate the value of this approach so that it will become standard of practice and reimbursed.

5. Educate the next generation of clinical investigators.

Potentially actionable somatic mutations 39% 47% (not including TP53)

Non-actionable somatic mutations 21% Likely germline variants 10% No mutations/variants 30%

Patients Screened for Non-Standard of Care Potentially Actionable Genomic Aberrations: first 2,000 patients,

updated 2016

IPCT: F. Meric-Bernstam, G. Mills, K. Shaw, J. Mendelsohn

Treated on genotype matched trials 11% 24%

50 gene panel

400 gene* panel

*More genes, includes copy number, decision support provided, increased number of trials available.

33

30,8

2

12,9

5

11,3

0

7,00

5,30

5,03

4,15

2,92

1,89

1,40

1,37

1,26

0,97

0,86

0,86

0,69

0,65

0,63

0,63

0,57

0,57

0,51

0,51

0,46

0,40

0,35

0,35

0,34

0,34

0,29

0,23

0,23

0,23

0,17

0,17

0,17

0,17

0,17

0,11

0,11

0,11

0,11

0,11

0,11

0,06

0

5

10

15

20

25

30

35

% P

atie

nts

with

Lik

ely

Som

atic

Mut

atio

ns

Gene Mutated

2000 patients likely to enter trials Hot Spot Mutation: 50 Gene Panel

Potentially actionable 39% TP53 not counted (31%) KRAS counted (11%) Oct 2014

Most targetable aberrations are rare across cancers

34 IPCT: F. Meric-Bernstam, G. Mills, K. Shaw, J. Mendelsohn

Frequency of Potentially Actionable Alterations by Tumor Type on 50 Gene Panel: Varies by Tumor Type

(Not counting p53)

78.9 77.2

0

20 10

30

40

80

% o

f Pts

with

Act

iona

ble

Alte

ratio

ns

66.9

53.8 52.8 50.0

33.1

22.2 21.4 20.9 16.0 15.8

8.3 6.7 6.0 5.0

50

60

70

35 IPCT: F. Meric-Bernstam, G. Mills, K. Shaw, J. Mendelsohn

1. Biomarkers predicting a likely response. Biomarkers in addition to genetic aberrations.

2. Combinations of therapies. Rationale – systems and computational biology Optimal timing and sequencing of therapies Avoiding toxicities

3. Understanding mechanisms of sensitivity and resistance to targeted therapies.

4. Sharing of information by all stakeholders.

Precision Medicine: Major Challenges

36

37