1

NASDAQ: MRTX

Targeting the genetic and immunological drivers of cancer

The KRASG12C Inhibitor, MRTX849, Provides Insight Toward Therapeutic Susceptibility of KRAS Mutant Cancer

Presented at AACR-NCI-EORTC International Conference on Molecular Targets

October 28, 2019Jamie Christensen, Ph.D.

2

Targeting KRAS Has Been Historically Challenging

K-Ras(G12C) inhibitors allosterically control GTP affinity and effector interactionsJonathan M. Ostrem,1,* Ulf Peters,1,* Martin L. Sos,1 James A. Wells,2 and Kevan M. Shokat1

5 4 8 | N A T U R E | V O L 5 0 3 | 2 8 N O V E M B E R 2 0 1 3

Direct, Reversible Inhibitors

• Smooth surface

• High affinity for & high intracellular concentrations of GTP/GDP

Downstream Effector Inhibitors

Raf / MEK and PI3K / AKT / mTOR

• Inhibition of WT signaling resulting in low therapeutic index

• Incomplete inhibition of signaling downstream of KRAS mut

Covalent Inhibition of KRAS G12C • Binding in the switch II pocket of GDP KRAS• Covalent bond to cysteine 12 • Locked in the inactive conformation

3

KRAS G12C Background—Reminders

• KRAS G12C mutations prevalent in (MSKCC):

– Lung adenocarcinoma: 14%

– CRC: 4%

– PDAC: 2%

– Other: gastric, uterine/endometroid, CUP. Etc

• KRAS G12C is a transversion mutation—common in smokers

• KRAS G12 mutations impair intrinsic GTPase activity and GTP hydrolysis

• Of KRAS mutations at codon 12, G12C exhibits lower intrinsic GTPase impairment and higher sensitivity to signals that modify extrinsic GTP hydrolysis

4

Drug Discovery Progression Toward MRTX849

• The addition of the C2 substituent significantly improved solubility and cellular potency and demonstrated more rapid modification of the protein compared with compound 1

• The cyanomethyl substituent on the piperazine further improved potency and allowed for the elimination of the naphthyl 3-hydroxyl group improving ADME properties

• The 8-positon of the naphthyl group to filled a hydrophobic pocket and increased potency an additional 5-fold

• Warhead modification and final optimization for reactivity and bioavailability provided MRTX849

13 hour time point in H358 cells; 224 hour time point in MIA PaCa cells

N

NNHO

N

N

O

Compound 1pERK IC50 = 7.6 µM1

N

NNHO

N

N

O

ON

Compound 2pERK IC50 = 70 nM1

N

NN

N

N

O

ON

NC

Compound 3pERK IC50 = 5 nM1

N

NN

N

N

O

ON

NC

CH3

MRTX1257pERK IC50 = 900 pM1

pERK IC50 = 6 nM2

N

NN

N

N

O

ON

NC

Cl

F

MRTX849pERK IC50 = 14 nM1

pERK IC50 = 5 nM2

5

MRTX849 Identified as a Potent, Selective, Orally Bioavailable Inhibitor of KRAS G12C

• MRTX849 only binds to inactive, GDP-bound KRASG12C

• Systematic adjustment of acrylamide reactivity and optimization of the naphthyl 8-substituent led to MRTX849, which shows greater stability in whole blood and hepatocytes

• Oral PK properties of MRTX849 improved with 50% oral bioavailability and >20 hour half-life projected in humans. Extensive tissue distribution observed with Vdss of > 10 L/kg projected in humans.

MW / clogP / tPSA 604/5.8/87

Kinact/KI 35 +/-0.3mM-1s-1

Cellular Potency (24h, MIA PaCa) 5 nM

5 min/3 µM Protein Modification 66%

Hepatocyte ER% (m/h) 61 / 50

Plasma Protein Binding % (m/h) 99.0 / 98.3

F% (m/h) 63 / 50*

* Human projected F% from PBPK modeling

6

MRTX849 Achieves Near Complete KRAS Modification and Inhibition In Tumors In Vivo

Normalized pERK and pS6 Inhibition6 h post administration/single dose

pERK Veh pERK 100 mpk pS6 Veh pS6 100 mpk

• Near complete inhibition of KRAS observed in tumor cells between 30 & 100 mg/kg—based on IHC

• No additional activity at higher doses• Unmodifiable pool of KRAS G12C and stromal cell

signaling impact magnitude of PD effect

KRAS G12C Protein Modification after a single dose

Veh0.5

hr2h

r6h

rVeh

0.5hr

2hr6h

rVeh

0.5hr

2hr6h

r-20

0

20

40

60

80

100

0.01

0.1

1

10

Plasma C

oncentration (ug/mL)

KR

AS

Enga

gem

ent (

% m

odifi

ed)

**10mg/kg 30mg/kg 100mg/kg

**Veh 10 30 100

0

1

2

3

Nor

mal

ized

Exp

ress

ion

MRTX849 (mg/kg)Veh 10 30 100

0.0

0.5

1.0

1.5

Nor

mal

ized

Exp

ress

ion

MRTX849 (mg/kg)

7

MRTX849 Dose-dependent Anti-Tumor EfficacyMaximally Effective Dose Confirmed Between 30-100 mg/kg

• The maximally effective dose of MRTX849 was identified as 100 mg/kg QD• Doses of 200 mg/kg or greater were well-tolerated and did not improve antitumor activity• Near complete target modification/inhibition correlates with maximal antitumor activity

0 10 20 30 40 50 60 70 800

200

400

600

800

1000

1200

Study Day

Tum

or v

olum

e (m

m3 ) 3 mg/kg

10 mg/kg30 mg/kg100 mg/kg

1 mg/kgVehicle

MIA PaCa-2 Xenograft Model

0 5 10 15 20 25 30 35 40 450

200

400

600

800

1000

1200

Day

Tum

or V

olum

e (m

m3 )

Vehicle1 mg/kg3 mg/kg10 mg/kg30 mg/kg100 mg/kg

H358 Xenograft Model

8

• AUC and Cave most closely correlated with antitumor activity based on schedule dependence and infusion studies

• AUC0-24 and Cave at 30 & 100 mg/kg, which demonstrated maximum antitumor activity in sensitive & partially sensitive models; respectively, were used for human efficacious exposure projections

• Free-fraction adjusted target AUCs0-24 were calculated as 14.3 ug*h/mL and 37.1 ug*h/mL

• PBPK Link modeling approaches (PK-SimTM or GastroPlus 9.5TM) were applied to project a human efficacious target dose and exposure dose and fit human data well

Derivation of Target Plasma Levels for MRTX849 Clinical Trials

Model Dose (mg/kg)

AUC0-24 (ug*h/mL)

FF adj AUC0-24 (ug*h/mL)

% Regression (day)

Projected Efficacious Total/FF adj AUC (human, ug*h/mL)

Projected Efficacious Total/FF adj Cave (human, ng/ml)

MIA PaCa-2 10 7 0.07 -52% (13)MIA PaCa-2 30 24 0.24 -96% (13) 14.3 600

HCC-44 100 63 0.63 -61% (13) 37.1 1450

9

MRTX849 Anti-Tumor Efficacy Across Models Studies Designed to Identify Response/Resistance Correlates

• A panel of in vivo tumor models was utilized for response correlations (obviate in vitro disconnect)• 100 mg/kg (max efficacious dose)• No significant activity in non-G12C models

• >30% tumor regression observed in ~65% of all models (17/26) suggesting potential for single agent development

• A 75% ORR In NSCLC, while CRC models were moderately responsive (BRAFi in CRC?)

• No significant correlation with co-occurring genetic alterations

• HER family score & cell cycle defects show trend

• Incomplete modification of KRAS in selected models potentially due to differences in extrinsic modulation of GTP hydrolysis

A549

HCT

116

H12

99

KYSE

-410

LU11

693

LU64

05SW

1573

CR62

56H

2122

LU25

12H

2030

PA13

83CR

6243

LU11

692

SW83

7SW

756

LU11

47LU

1155

4LU

99H

CC44

LU51

91LU

5212

H35

8Ca

lu-1

LU52

45H

1373

LU65

LU25

29M

iaPa

ca-2

-100

-50

0

50

100

non-G12C G12C

EsophagusLung PDX Colon PDXCervixPancreasLung Colon

Pancreas PDX

CDX X X X X X X X X X X X X X

PDX X X X X X X X X X X X X X

KRAS G12C MAF (%) 39 56 97 100 36 100 63 100 55 62 70 57 NA 54 45 54 100 100 100 67 84 93 99 92 100 98

KRAS CNV 3 NA 13 2 2 2 5 2 4 3 NA 2 NA 3 NA 5 2 NA NA 3 4 6 3 6 4 3

STK11 mut N Y N N N Y N Y N N N N Y Y N N N N Y N N N N Y N N

KEAP1 mut N Y N N N Y N Y N N N N N N Y N Y Y Y N N N N N N N

HER family Y N Y N Y Y Y N Y Y Y Y NA Y N N N N Y Y N Y Y N N NCDKN2A homodel N Y Y Y N Y N N Y N N* N NA N Y Y N N N N N N N N N Y

10

MIA PaCa-2 H1373

Veh QD x 1 QD x 524 6 24 246

Veh

6 24 6 24QD x 1 QD x 5

24Veh

6 24 6 24QD x 1 QD x 5

24

Complete Response

Veh6 24 6 24QD x 1 QD x 7

24Veh

6 24 6 24QD x 1 QD x 7

24Veh

6 24 6 24QD x 1 QD x 7

24

H358 H2122Partial response/Refractory

Mechanisms of Response and Resistance to MRTX849Durable Inhibition of ERK (but not S6) tracks with tumor response

11

Mechanisms of Intrinsic Resistance to MRTX849Feedback Signaling and Bypass Pathways---NCI-H358

VehiclepS6

30m

pk Q

D x

130

mpk

QD

x 3

30m

pk Q

D x

3

Vehicle

pERK6hr 48hr24hr

30m

pk Q

D x

1

2 0 2 5 3 0 3 5 4 0 4 5 5 00

2 5 0

5 0 0

7 5 0

1 0 0 0

D a y

Tum

or V

olum

e (m

m3 )

V e h ic le 1 m g /k g 3 m g /k g 1 0 m g /k g 3 0 m g /k g

1 0 0 m g /k g

Dose–dependent Anti-Tumor Efficacy in H358

12

Pharmacogenomic Screens to Identify Combination Targets and Resistance Mechanisms

• Screened ~90 agents in combination with MRTX849 across 8 lung cell lines in vitro that were partially MRTX849-resistant in vivo.

• MRTX849-anchored CRISPR screens: ~1,000 genes, 4 KRASG12C cell lines, in vitro/in vivo

• Top combination targets: EGFR family, FGFR1, SOS1, SHP2, mTOR, CDK4/6.

• Top resistance genes: CDKN2A, KEAP1, RB1, PTEN, NRAS.

GTP-KRASG12C

RSK

MEK

ERK

cRAF

PTEN

AKT

PI3K

mTOR

S6

S6K1

SOSSHP2

EGFR

MYC

CCND1

CDK4/6 RB1 E2F

Grb2

GDP-KRASG12C

NF1RAS GAPs

RTKs FGFR1

KEAP1NRAS

Class HER CDK mTOR/PI3K SRCMRTX KRAS

G12CiCalu-1H1373H1792H2030H2122

HCC1171HCC44

LU99SW1573

SW837

Small Mol Combo Screen MRTX849 Combination Targets and Resistance Gene Map

KRASMYC

mTORSHP2 CDK6

NRASKEAP1

RB

CCND1

Log FC Veh D14 / plasmid

Log

FC 8

49 D

14 /

plas

mid

SOS1CRAF

PTEN

S6

LOF Confers Resistance

MRTX849 Combination

Targets

CRISPR Screens

H2122 in vivo

13

Conclusions

• MRTX849 is a novel small molecule KRASG12C inhibitor in clinical trials

• Maximal and durable inhibition of KRAS liked to defined PK parameters maximizes response

• Tumors harboring KRAS G12C are broadly dependent on KRAS for growth and survival…….

• ……However, complex signaling circuitry in some KRAS-dependent tumors can result in partial bypass of dependence

• Mechanisms are heterogeneous and relate to feedback or bypass signaling, enhanced nucleotide cycling, and KRAS-independent cell cycle transition

• Rational combination approaches provide a practical solution to address heterogeneity

• Target plasma derivation and corelative science will aid in rational development of MRTX849

14

AcknowledgmentsMirati Tx Array CollaboratorsMatt Marx J. Brad Fell Channing Der (UNC)Jill Hallin John Fischer Adrienne Cox (UNC)Peter Olson Brian Baer Piro Lito (MSK)Ruth Aranda Nick Saccamano Pasi Janne (DFCI)Jeff Winkelman Mike Burkard Chip Petricoin (GMU)Andrew Calinisan Dylan HartleyLars Engstrom James BlakeDavid Briere Josh BallardHarrah Chiang Larry BurgessLauren Hargis Mark ChicarelliNiranjan Sudhakar John J. GaudinoVickie Bowcut Lauren Hanson

Tony TangPavel Savechenkov

Monceros Biosystems Guy VigersAdam Pavlicek Karyn BouhanaJulio Fernandez-Banet Ron HinklinSole Gatto Eric Hicken