From CARS to armored CARS

CARs and armored CARs Lymphoma and Myeloma 2014 New York NY October 23, 2014 Renier Brentjens MD PhD Associate Member Leukemia Service Chief, Cellular Therapeutics Center Department of Medicine Memorial Sloan Kettering Cancer Center

Conflict of Interest Disclosure Renier Brentjens MD PhD •Stockholder: Juno Therapeutics (scientific cofounder) •Royalties: Juno Therapeutics •Honoraria: none •Reserch Funding: Juno Therapeutics •Consultant fees: Juno Therapeutics •Discussion of off-label drug use: Tocilizumab

Generation of a tumor targeted chimeric antigen receptor (CAR) α-TAA mAb

TCR complex α-TAA scFv—CD8-ζ

ψ

5’ LTR SD

α-tumor scFv VH

VL

CD8 ζ chain

SA

CAR retroviral vector

3’ LTR

Generation of TAA-targeted T cells for treatment of Cancer TAA CAR Native TCR

αTAA scFv

CD8

1. Construct a chimeric antigen receptor (CAR)

CD3 ζ 3. Transduce and expand patient T cells ex vivo

TAA CAR

4. Infuse transduced T cells to eradicate TAA+ tumor cells

2. Subclone CAR gene into a retroviral vector (SFG)

ψ

5’ LTR SD

αTAA scFv VH

VL

SA

SFG-CAR

CD8

ζ chain

3’ LTR

TAA

Advantages of CAR T cell therapy • HLA-independent antigen recognition, therefore universal application • Active in both CD4+ and CD8+ T cells • Target antigens include proteins, carbohydrates and glycolipids • Rapid generation of tumor specific T cells • Minimal risk of autoimmunity or GvHD • A living drug, single infusion

Expression of CD19 and other B cell markers on B lineage cells B cell lymphomas and leukemias

preB-ALL

Stem Cell

pro B

pre B

CD19 CD22 CD20

immature B

mature B

myelomas

plasma cell

Evolution in CAR design

Clinical trials using CD19 targeted T cells in relapsed or refractory B-ALL

Patient characteristics and treatment outcomes

Sci Transl Med. 2013 Mar 20;5(177):177ra38

Clinical Efficacy Outcomes (n=22) Characteristics No. of Patients Overall Complete Response to 8 salvage chemotherapy

% 36

Overall Complete Response to 19-28z CAR T cells (n=21)

18

86

10

83

Complete Remission (CR)

14

64

Complete Remission with incomplete count recovery (CRi)

4

18

Molecular Complete Remission (CRm) (n=21)*

15

72

In patients with morphologic residual leukemia (n=12)

Time to CR/CRi (days)

Post CAR T Allo-SCT** (n=12 eligible patients)

22.1

8

67

* mCR or MRD- as determined by flow cytometry and/or deep sequencing for the index IgH clonotype and/or qPCR for the bcr-abl transcript ** Four patients had medical contraindication to allo-SCT, two patients in CR have declined potential allo-SCT, one is inevaluable and one is being evaluated for an allo-SCT

UPenn studies of relapsed B-ALL • 25 pediatric and 5 adult relapsed or refractory B-ALL patients treated • 19-4-1BBz CAR design • 90% CR • 6 month EFS 67% • 6 month OSR 78%

Maude et al N Engl J Med. 2014, 371:1507-17

NCI studies of relapsed B-ALL • 20 pediatric and young adult relapsed or refractory B-ALL patients treated. • 19-28z CAR design • 70% CR (14/20) • 60% MRD- CR • 5 month EFS 78% in MRD- patients

Lee et al Lancet. Published online October 13, 2014

The Problem

Clinical trials using CD19 targeted T cells in low grade B cell malignancies

UPenn clinical trial results Patient

Prior Chemotherapy

Conditioning Chemotherapy

Response

1

Fludarabine, Rituximab, Alemtuzumab, R-CVP, Lenolidomide, PCR

Bendamustine

CR (3+ years)

2

Alemtuzumab

Bendamustine/Rituximab

PR (7 months)

3

Rituximab/Fludarabine, Rituximab/Bendamustine. Alemtuzumab

Pentostatin/ Cyclophosphamide

CR (3+ years)

Kalos et al Sci Trans Med 2011

Updated UPenn Trials in CLL (ASH 2013) • Abstract 4162 – CD19 CAR T cells treating relapsed/refractory CLL • Utilizing a 4-1BBz CAR construct, 14 CLL patients treated • 3/14 patients obtained CR (21%), 5/14 patients obtained PR (36%), 6/14 patients with no response (43%) • 6/14 patients with persistent detectable CAR T cells (5-35 months) • No CR patients with reported relapsed disease • No dose response reported

• Abstract 873 – Dose randomized dose optimization trial of CLL patients with either high or low dose CAR T cell infusions • Utilizing a 4-1BBz CAR construct, 27 CLL patients treated • Patients randomized to either low dose (5 x 107 CAR T cells) or high dose (5 x 108 CAR T cells) • No dose response benefit seen in these treated patients • Overall response rate (CR + PR) was 40% • No correlation with CRS and RR was observed

MSKCC clinical trial results: CLL

No Cyclophosphamide Cyclophosphamide

Brentjens et al Blood. 2011 Nov 3;118(18):4817-28

CAR questions • What is the etiology of differential responses by CAR T cell therapy to relapsed B-ALL versus CLL? • What is the role of bulky disease and CAR T cell anti-tumor efficacy? • The role of the hostile tumor microenvironment and CAR T cell function • How to build a better T cell?

The hostile tumor microenvironment The tumor microenvironment contains multiple inhibitory factors designed to potentially suppress effector T cells. – – – – –

CD4+ CD25hi FoxP3+ regulatory T cells (Tregs) MDSCs TAMs Expression of inhibitory ligands by tumor (PD-L1) Tumor secretion of T cell suppressive cytokines (TGF-β and IL-10)

The solution? Armored CAR T cells

Moving Forward: Armored CARs

IL-12 • A heterodimeric cytokine secreted by activated APCs, neutrophils and macrophages. • Induces Th1 CD4+ T cell response enhancing IL-2 and IFN-γ secretion • Enhances T cell clonal expansion and effector function in concert with TCR signaling (signal 1) and CD28 costimulation (signal 2), serving as a signal 3. • Avoids/reverses T cell anergy • May overcome Treg mediated effector T cell inhibition • Recruits and activates NK cells • Clinical trials in cancer using systemic IL-12 therapy has been limited by severe inflammatory side effects

Syngeneic EL4(hCD19) tumor model EL4(hCD19)

53%

IV injection

Assess T cell eradication of tumor

mCD19-/- hCD19+/IV injection

Assess T cell homing to tumor Assess longterm survival of T cells Assess memory T cell response to rechallenge with tumor

Assess T cell proliferation

Harvest splenocytes

in vivo

Assess the efficacy of suicide vectors Determine the side effects of therapy

mCD19-/- hCD19+/Retroviral transduction with chimeric receptor

Lymphodepletion enhances anti-tumor efficacy of 19z1+ T cells 6Élive cells

80

10

3

2.07

10

2

10

1

100

Cytoxan + 19z1 No Cytoxan + 19z1 Cytoxan Pz1

60

1Élive cells

40

10

4

10

3

20

24.3

10

2

10

1

100

0.11

24.3%

75.5 100

0.063 101

102 103 FL1-H: FL1-Height

104

0 0

10

20

30 Time (d)

40

50

Days since tumor cell injection

60

0.02

2.07%

97.6 100

FL2-H: hCD19 PE

% Survival

4

FL2-H: hCD19

100

Percent Survival

10

0.29 101

102 FL1-H: mCD19

103

104

19z1IRESIL-12 modified T cells secrete biologically active IL-12 and exhibit enhanced targeted cytotoxic function and resistance to Tregs

A B

C

D

Specific Lysis (%)

40 30

*

* 20 10

*

* 19mzIRESIL-12 19mz

0

1:1

2.5:1

5:1 E:T ratio

10:1

Pegram et al Blood 2012

Syngeneic IL-12 secreting CD19 targeted T cells induce B cell aplasias and tumor eradication A

B

Pegram et al Blood 2012

IL-12 secreting CAR T cells in vivo efficacy

IL-12 genetically modified T cells: Armored CAR T cells

NK cell

NK cell Recruitment and activation

Targeted tumor cytotoxicity

Tumor cell

IL-12 secretion

Targeted tumor cytotoxicity

CAR-IRES IL-12

IL-12 secretion Targeted tumor cytotoxicity

Activated TIL

Reversal of anergy

Anergic TIL

IL-12 secretion Enhanced CM phenotype, enhanced cytotoxicity, enhanced persistence Resistance to Treg and TGFβ inhibition

Conclusions • Autologous CD19 targeted CAR modified T cells have demonstrated very promising anti-tumor efficacy in B cell ALL with more modest responses in patients with low grade B cell malignancies. • Etiologies of CAR T cell resistance may be related to the hostile tumor microenvironment. • Application of CAR T cell therapy for low grade B cell malignancies as well as moving forward towards application to solid tumor malignancies requires “armored” CAR T cells designed to both overcome the hostile tumor microenvironment and exhibit enhanced anti-tumor efficacy and long term persistence. • “Armored” CAR T cells appear to have enhanced anti-tumor efficacy based on pre-clinical tumor models. • Future studies using “armored” CAR T cell technology will focus on translation of these armored CAR T cells to the clinical setting both in the context hematological as well as solid tumor malignancies.

Renier Brentjens Hollie Pegram Mythili Koneru Sarwish Rafiq Swati Pendharkar Eric Smith James Lee Yan Nikhamin Jae Park Kevin Curran Peter Chang

Cell Therapy and Cell Engineering Facility (Isabelle Riviere, Director)

Michel Sadelain Marco Davila Michael Gong Jean Baptiste Latouche

QA/QC Shirley Bartido Yongzeng Wang (Mark Przybylowski) James Hosey Domenick Pirraglia (Vanessa Capacio)

Leukemia Service David Scheinberg Jae Park Martin Tallman Mark Frattini Peter Maslak Mark Heaney Joe Jurcic Nicole Lamanna Marco Davila Dan Douer

R&D, Manufacturing Xiuyan Wang Jolanta Stefanski Malgorzata Olszewska Oriana Borquez-Ojeda Teresa Wasielewska Jinrong Qu

Clinical Research Yvette Bernal

Lymphoma Service Craig Moskowitz Ariela Noy GYN service Samith Sandadi Roisin O’Clearbhail Adult BMT Service Sergio Geralt Craig Sauter Department of Clinical Laboratories Lillian Reich David Wuest Kathy Smith Biostatistics Glenn Heller

Funding CA59350 (MS) ; P30 CA-008748 (CT); 3RO1CA138738-02S1(RJB); Alliance for Cancer Gene Therapy ; Terry Fox Run for Cancer Research; William H. Goodwin and Alice Goodwin, and the Commonwealth Cancer Foundation for Research and the ETC of MSKCC; Damon Runyon Clinical Investigator Award (RJB); William Lawrence & Blanche Hughes Foundation (RJB); CLL-Global Research Foundation (RJB)

Brentjens’ Lab Members