1

Jennifer Wu, PhD

ProfessorFeinberg School of MedicineRobert Lurie Comprehensive Cancer Center Northwestern University

NKG2D ligands regulation of immune checkpoint blockade therapy of cancer: efficacy and toxicity

1

Immuno-Oncology Summit, August 2018

2

Disclosures

• Founder, CanCure LLC

• Inventor, B10G5 antibody

Versatile Signaling of NKG2D receptor

Normal

no activation

Immune tolerance

NKG2D

NK, T cell

Anti-tumor Immunity

• Co-stimulates CTL• Activates NK cell• Regulates Th1 and Th17

Autoimmunity

• Regulates ILC activation• Activates pathogenicTh1/Th17• Expands pathogenic NKG2D+CD4+

• Amplifies inflammation

NK, CD8 T cellNKTγδ T

Diseased

NKG2D-L

3

NKG2D and NKG2D-L in tumor immunity

Normal cell

On Guard

NKG2D

NK, CD8TNKT, γδT

Tumor cell

NK CD8, NKT

γδT

Activation Co-stimulation

NKG2D ligand(MIC)

Genomic insult

tumor suppressed

Human:

Mouse:MICA, MICB

NONE

ULBP1,2,3,6

RAE-1α-ε

ULBP4,5

MULT1, H60a,b

Distinct NKG2D-L in Human and Mouse

MIC = MHC I chain related molecule ULBP = UL16 binding protein (Modified From Huergo-Zapico e4tal.)

HealthyProstate

Pre-Cancerous

carcinoma

ATM p-ATM MICA

B

C

D

E

F

G

H

I

NKG2D ligand expression is an early response to DNA damage or oxidative stress

8-oxo-dG

(Wu et al, Translational Oncology 2009)

Shedding of surface MIC as an immune evasion strategy

Progressive

“shedding”Soluble MIC

(sMIC)

IDOMMPsADAMSHypoxiaExosome

NK. T

Immune Dysfunction

Indolent Tumor

NK CD8, NKT

γδT

Activation Co-stimulation

MIC

Genomic insult

Pros

tate

Canc

er

Tamaki et al, AntiCancer Research 2010; Kumar V et al, PLoS One 2012; Vyas et al. Oncoimmunology 2017.

p < 0.05

p < 0.0525

20

15

10

5

0

Pate

nt Su

rviv

al (%

)Se

rum

sMIC

(ng/

ml)

Months after sampling

sMIC low

sMIC high

Ora

lCan

cer

Tumor Type MICA/BCarcinoma

• Ovarian A/B

• Cervical A

• Breast A/B

• Lung A/B

• Hepatocellular A/B

• Colon A

• Prostate A

• Head and Neck A/B

Multiple Myeloma A

Melanoma A

Neuroblastoma A

Cancers with circulating sMIC

8

Wu, JCI 2004; Liu, JCI,2013Wu, Current Opinion Immunoology,2018

Clinical evidence: sMIC and Cancer Prognosis

Ove

rall

surv

ival

Months after sampling

P < 0.008

sMIC low

sMIC high

Live

r Can

cer

Tumor-bound MIC and sMIC in tumor immunity

Groh, et al., Nature, 2002; Wu et al., JCI, 2004; Liu, et al, JCI 2013; Xiao et al, JHO 2015; Zhang et al Science Adv. 2017…>80 in Pubmed

NORMALCELL

“GOOD” TUMOR

“BAD”TUMOR

Impact on Cancer controlledprogression

and metastasis

Membrane MIC (mMIC)

Soluble MIC (sMIC)

NKG2D

-

NK, CD8T NK, CD8T NK, CD8T

Tumor MIC Form - membrane-bound (mMIC)

Tumor-shed (sMIC)

MIC = MHC I Chain-related molecule

Lymphocyte (NK, CD8 T) on guard activated disabled

Suppressor cells (MDSC, TAMS) - thriveLimited (?)

MIC(B)

Probasin (Androgen responsive)

(MIC/B6-Tg)x

TRAMP/MIC mice(Bona Fida Onco-Immune interaction)

(TRAMP)SV40TAg

Probasion

(Liu et al, JCI 2013, Zhang, Sci. Adv., 2017)

Transplantable sMIC+-tumor Host

sMIC+- tumor

(Liu, et al, JCI, 2013; Lu et al., CCR, 2015; Basher, Oncotarget, 2015)

Animal models recapitulate onco-immune dynamics in human

6

Men

Cancer progression is associated with aberrant shedding of MIC: in mouse and men

mou

se

Normal Prostate Pre-cancer Localized PCa Invasive PCa

Poorly-Differentiated

(PD)

TRAMP/MICB

TRAMP Pre-cancerWell-Differentiated

(WD)

Anti-MIC IHC

(Liu et al., JCI 2013)

mAb B10G5 Targeting sMIC Induces Primary Tumor Regression and Eliminates Metastasis

TRAMP/MIC mice

Puberty

Age 6

WD(100%) PD/met (>40%)

16 24 27 35 (wks)

Cancer (100%)

Rx: i.p., twice weekly, 4 mg/Kg

surv

ival

%

0

50

100

25 30age (wks)

35

B10G5 (29)

cIgG (23)

p = 0.002

Rx

1.51.00.50.0

2.0

5432

5

1510

B10G5(35 wk)

cIgG (35 wk)

Pre-Rx (26-27 wk)

Pros

tate

Wei

ght(

g) p <0.01p <0.01

Mic

e%w

ith

Met

asta

sis

none

100806040200

p <0.001

(Lu et al, CCR 2015 , 21:4819-30)

B10G5

Ki67H&E

7

cIgG

B10G

5

CD40

CD80

cIgG

CD86

B10G5

DC in dLN

NKG2D

10.1

B10G5

2.98

cIgG

INF-γ9.051.65

IFNγ

CD44

32.917.4

CD8 T in dLN

NK1.1

2.380.68

BrdU

18.30.92

24.90.90

IFNγ

NK in spleen

cIgG B10G5

B10G5 Clearance of tumor-derived sMIC invigorates innate and adoptive cell immunity

cIgG B10G5

CD

8

spln

dLN

TIL

SV40TAg-tetramer

2.490.17

2.890.29

0.64 5.68

Day 9

Adoptively transfer of SV40TAg-specific CD8 T cells

cIgG B10G5

Day 9

TAg-tetramer+ CD8 T cells

Day 14 2.34

CFSE

10.3

d.

2.226.82

4.696.27

0.98 0.78

B10G5 anti-SMIC overcomes antigen-specific CD8 T cell tolerance

MOA of B10G5Dual action: neutralize sMIC immune suppression and enhances mMIC immune stimulation

mMIC

sMIC

NKG2D

Neutralizes sMIC immune suppression and stabilizes

surface NKG2D

Forms immune cluster to enhance stimulatory

function of mMIC

Non-blocking

B10G5 (mIgG1)

NK, CD8T

NK, CD8T

15

Zhang et. AL., 2017, Science Adv.Lu et al., 2015, Clin. Can. Res.Xiao, 2015, JHO

Dual action Key MOA

Differentiate from other in-development anti-MIC programs

IPH-4301, a blocking mAb Shedding blocking mAb (Dana-Farbe)

Stabilize CD3ζ, a key signaling molecule for TCR/CD3 complex

Eliminate sMIC-induced expansion of MDSC and TAM

Enhance DC antigen presentation of MIC-positive tumors through opsonization

Bystander MOA on TME

H&

E

CD8

NK

CD

8

B10G5cIgG TCR3β clonality0.15

0.10

0.05

0.00

TCR3β complexity300

200

100

0

Clearing sMIC flares up a “cold” tumor

Zhang et al., Science Advances (May 17, 2017) 8

+B10G5MFI: 151

sMIChi

MFI: 96.5

CD3ζ

Therapy:cIgG or αCTLA4

Puberty

Age 6

TRAMP or TRAMP/MICBmice

27 35 (wks)16 24

Tumor initiation metastasis Study

End point

A

C

Pros

tate

weig

ht(g

) (a

tdea

th)

TRAMP

poor survivor

101418

22.0

1.5

1.0

0.5

0.0

6

TRAMP/MICB

ns

D

cIgG αCTLA4

Poor response to anti-CTLA4 therapy in sMIChi mice

(Zhang et al., Science Advances 2017)

B

Surv

ival

%

TRAMP + αCTLA4 TRAMP + cIgGTRAMP/MICB + cIgG TRAMP/MICB + αCTLA4

0

25

50

75

100

0 2 4 6Rx week

8

*

(sMIChi prior to therapy)

05

1015406080

100120

TP/MICB lung micrometastasis

# of

mic

rom

etas

tasi

s

Serendipity: High circulating sMIC exacerbated anti-CTLA4-induced colon inflammation

cIgG

TRAMP (no MIC)

αCTLA4

0.0

0.5

1.0

46

8

serum sMIC prior to anti-CTLA4 monotherapy

ng/m

l

*

0

5

10

15

αCTLA4cIgG

TRAMP/MIC

sMIClo sMIChi sMIClo sMIChi

Col

on le

ngth

(c

m)

P< 0.01

B10G5 anti-sMIC enhanced anti-CTLA4 response and eliminated colitis

Pros

tate

wei

ght(

g) 18106

2.0

1.5

1.00.5

0.0

P < 0.05

P=0.48 P < 0.001

WT (B6)

Poor-survivors sMIChi

DCs inTILS

1500

1000

500

0

CD

80

P < 0.05

1500

1000

500

0

P < 0.05

0

1500

1000

500CD

40

P < 0.05

MFI

CD

86(Zhang et al., Science Advances 2017 May 17)

αCTLA4+B10G5sMIChi

(Zhang et al., Science Adv. 2017 and in submission)

cIgGaPDL1

CuraB10aPDL1+CuraB10

8 9 10 11 12 13 14 15 16 17 180

50

100

days in therapy

Surv

ival

%

*** ***

B16F10-sMIC melanomasyngeneic model

Spontaneous tumor model(prostate tumor)

tum

or in

the

pros

tate

(g)

1810

2.01.51.00.50.0

(sMIC-Hi)non-responder to ICI

Single Agent

Combination

Clearance sMIC enhances immune check point inhibitor (ICI) therapy: both responders and non-responders

Optimal dosing of combination therapy induces CR of established tumors with no colitis

Days since Rx

On Therapy Off Therapy

Transplant Syngeneic

sMIC+-TRAMP-C2

MIC/B6 Tg host13

500

400300

200

1000

0 2 5 8 12 15 18

Control500

400

300

200

100

0

0 2 5 8 12 15 18

500

400300

200

1000

0 2 5 8 12 15 18

anti-CTLA4 anti-sMIC

100

200

300

400

500

0

0 2 5 8 12 15 18 21 24 27 30 33 36 38 41 45 48

anti-sMIC + anti-CTLA4

4/5 CRTu

mor

volu

me

(mm

3 )

Tum

orvo

lum

e (m

m3 )

Clinical evidence: patients who developed anti-sMIC autoantibody elicited favorable outcome during anti-CTLA4 therapy

Soluble NKG2D ligands are biomarkers associated with the clinical outcome to immune checkpoint blockade therapy of metastatic melanoma patientsMaccalli C. et al., Oncoimmunology. 2017; 6(7):e1323618

Soluble NKG2D ligands limit the efficacy of immune checkpoint blockadeLópez-Soto A et al., Oncoimmunology.2017;6(10):e1346766.

Therapy-induced antibodies to MHC class I chain-related protein A antagonize immune suppression and stimulate antitumor cytotoxicityJinushi M, Hodi FS, Dranoff G.Proc Natl Acad Sci U S A.2006;103:9190-5

mCRPCRx: ipilimumab

0

20000

40000

60000

80000

100000

120000*

(Zhang et al., Science Advance, 2017)

*PSA: 192 ng/ml

0.5 ng/ml at cycle 5

Seru

m a

nti-M

IC Ig

G(p

g/m

l)

Adopted from Peters CP et al., Immunol Lett. 172:124-31 (2016)

15

IL-10Th1 NK CTL ILCs

TNFαIL-

22 IL-6 IL-17Th1/Th17 ILCs

αCTLA4

αCTLA4

anti-sMIC

Two Hits Hypothesis for anti-CTLA4-induced colitis

Healthy Gut

Inflammed Gut

24

Take Home Message

NKG2DL, sMIC can skew response to ICI (anti-CTLA4 and PD1/PDL1 blockade) therapy from efficacy to toxicity

Patients may be screened for serum sMIC to improve clinical outcome with ICI therapy

Antibody clearing sMIC could alleviate ICI therapy-induced colon toxicity

MIC/TRAMP model is an unique platform not only to validate targeting human sMIC, but also to study ICI-associated colon toxicity

B10G5 (CuraB10) Development Milestones(CanCure)

25

MILESTONES 2018 2019-2020

Humanization (CuraB10), stable cell line, pre-CMC,Pilot non-GLP tox study

GLP/cGMP Manufacturing

IND-Enabling Toxicology

IND Application

Phase I Clinical Trial (Planned Academic Partnership-NU, NIH)

Humanization, preliminary formulation, stable cell line, and RCB bank completed, in process of pilot non-GLP tox study

seeking Pharma partnership for GLP/GMP and IND-enabling studies

Acknowledgements

Wu LabPayal Dhar, PhD CandidateMahmoud Kalaffalla, PhDJu Wu, MDJiangping Deng, MDYoulin Kuang, MD/PHDPablo Larrocha, PhDSizhe Liu, PhD candidate

Fahmin Basher, PhD CandidateJinyu Zhang, PhDJohn Jarzen, MScWei Sun, PhDWoonki Kim, PhDJohn Lu, PhDJun Wang, Ph.DGavin Liu, Ph.D MDGang Xiao, MD/PHDCatherine Atteridge, MSAshley Lundgren, MS

Northwestern/LCCJeffrey Sosman, MDStephen Hanauer, MDEmanuelle Bellaguarda, MDRonen Sumigen, PhDChad Thaxon, MD/PHD

MUSC/HCCMark RubinsteinMichael Lilly

Mizzou Cancer CenterKevin O’Carroll Guangfu Li

Knights Cancer Center / OHSUJulie Graff

Funding

26

Collaborators

Scanlan Family Endowment