AACR Virtual Annual Meeting 2020 April 27-28, 2020 Clinical Plenary – Abstract #9670

TP53 abnormalities correlate with immune infiltration and are associated with response to flotetuzumab, an investigational immunotherapy, in acute myeloid leukemia

Jayakumar Vadakekolathu1, Catherine Lai2, Stephen Reeder1, Sarah E. Church3, Tressa Hood3, John Muth4, Heidi Altmann5, Marilena Ciciarello6, Antonio Curti6, Peter J.M. Valk7, Bob Löwenberg7, Martin Bornhäuser5, John F. DiPersio8, Jan K. Davidson-Moncada4, Sergio Rutella1,10

1John van Geest Cancer Research Centre, School of Science and Technology, Nottingham Trent University, Nottingham, UK; 2MedStar Georgetown University Hospital’s Lombardi Comprehensive Cancer Center, Washington, DC; 3NanoString Technologies Inc., Seattle, WA; 4MacroGenics Inc., Rockville, MD; 5Department of Internal Medicine I, University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany; 6Institute of Hematology "L. and A. Serágnoli", Department of Hematology and Oncology, University Hospital S. Orsola-Malpighi, Bologna, Italy; 7Department of Hematology, Erasmus University Medical Centre, Rotterdam, The Netherlands; 8Division of Oncology, Department of Medicine, Washington University School of Medicine, St Louis, MO; 10Centre for Health, Ageing and Understanding Disease (CHAUD), School of Science and Technology, Nottingham Trent University, Nottingham, UK.

[email protected] Disclosures

• MacroGenics Inc.; Research funding • NanoString Technologies Inc.; Research funding Background

• Chemotherapy remains the standard of care for most patients with AML, despite recent approvals of novel drugs • We have identified immune subgroups of AML (‘immune-infiltrated’ and ‘immune- depleted’) that predict chemotherapy resistance but also response to flotetuzumab immunotherapy (Vadakekolathu J, et al. Under revision)

• The genetic drivers of immune infiltration in AML are presently unknown • TP53 mutations occur in 8-10% of de novo AML cases and are associated with chemotherapy resistance, high risk of relapse and dismal prognosis even after hematopoietic stem cell transplantation • The functional consequences of TP53 mutation/inactivation on host immune regulation have been largely overlooked in AML

• The TP53 mutants studied thus far in AML do not show any evidence of gain-of-function mechanisms (Boettcher S, et al. Science 2019) Objectives

• To determine whether TP53 abnormalities correlate with the composition and functional orientation of the tumor immunological microenvironment (TME) in AML

• To determine whether TP53 abnormalities identify a subgroup of patients with AML that may benefit from immunotherapy with flotetuzumab, a CD123×CD3 bispecific DART® molecule for redirecting host T cells to AML (Chichili GR, et al. Science Translational Medicine 2015) in the CP-MGD006-01 clinical trial (NCT#02152956) Graphical ‘cohorts and methods’

A In silico analyses (TCGA-AML) B Primary AML blasts C Flotetuzumab cohort (n=35)

147 non-promyelocytic AMLs TP53-mutated (n=42) TP53-wt (n=22) TP53-mutated and/or TP53-wt (n=24) (14 with TP53 mutations) 17p abnormalities with genomic loss of TP53 (n=11)

RNA extraction

RNA extraction Immune cell type and biological activity signatures (computed as in Danaher P, et al. JITC 2017 and 2018) Targeted Immune GEP (PanCancer IO 360™ Panel) - RUO Targeted Immune GEP Correlation with prognostic molecular lesions (TP53 (PanCancer IO 360™ Panel) - RUO mutational status, NPM1 mutational status, FLT3-ITD DE status, CHIP-defining mutations) and clinical outcomes GO ontologies (METASCAPE) (Cox PH) Network analysis Predictors of response TP53 mutations associate with an immune-infiltrated TME in TCGA-AML B A 8 8 KW=53.03 KW=32.55 7 P<0.0001 P<0.0001 Molecular lesion 6 Molecular lesion 6 Molecular Risk Group FLT3−ITD 6 B7−H3 4 NPM1+FLT3−ITDneg Apoptosis Other TGF−beta 5 4

MMR.loss 2 RUNX1 score APM.loss TP53 4 Mast cells TIS score PD1 0 Molecular Risk Group 2 CTLA4 Good 3 Treg

B cells −2 Intermediate Inflammatory chemokine Th1 cells N.D. 2 0 Cytotoxicity Mut. Mut. NPM1 FLT3-ITD Mut. CHIP-defining Mut. Mut. Mut. NPM1 FLT3-ITD Mut. CHIP-defining Mut. Poor NPM1 w FLT3-ITD RUNX1 (TET2, TP53 NPM1 w FLT3-ITD RUNX1 (TET2, TP53 Cytotoxic cells DNMT3A, DNMT3A, NK CD56dim cells ASXL1) ASXL1) Exhausted CD8 8 9.0 TIGIT KW=21.01 KW=40.1 T cells P=0.0008 P<0.0001 Lymphoid cells 6 CD8 T cells 7.5 NK cells Stroma Endothelial.cells 4 Inflammatory.chemokines score 6.0 MHC2 γ Immunoproteasome 2 APM IFN- CD45 4.5 0 score Lymphoid Myeloid cells IL10 -2 3.0 PDL1 Mut. Mut. NPM1 FLT3-ITD Mut. CHIP-defining Mut. Mut. Mut. NPM1 FLT3-ITD Mut. CHIP-defining Mut. PDL2 NPM1 w FLT3-ITD RUNX1 (TET2, TP53 NPM1 w FLT3-ITD RUNX1 (TET2, TP53 TIS DNMT3A, DNMT3A, IFN.downstream ASXL1) ASXL1) IFN−gamma DC IDO1 Glycolytic.activity C D Hypoxia 10 P<0.0001 8 P=0.0002 10 P=0.0002 Myeloid.inflammation 100 ARG1 TP53 mutations Proliferation MAGEs 8 NOS2 8 Other molecular lesions 6 75 Immune-infiltrated Immune-depleted 6 6 18.5 mo. (n=14 with TP53-mut.) (n=0 with TP53 mut.) mRNA 50 4 4 4 TIGIT PD- mRNA FoxP3 mRNA 25 N=118 cases with available 2 Percent survival Log- χ2=20.69; P<0.0001 2 2 information on prognostic 4.5 mo. HR=3.43 (95%CI 1.34-8.74) 0 0 molecular lesions, including TP53-mut. Other TP53-mut. Other TP53-mut. Other 0 12 24 36 48 60 72 84 96 TP53 mutations (n=14) Overall survival time (months) TP53–related immune profiles in primary BMs D 8 12 C P<0.0001 P=0.0042 A TP53 status 6 9 84% TP53 missense p53 status p53 status Mast Cells Mutated

4 mRNA mRNA 12% TP53 no missense ARG1 WT/NA/ND 4 Proliferation 6 4% N.A. JAKSTAT.loss 2 APM.loss

MHC2 LAG3 PD-L1 Immunoproteasome 0 2 TGF−Beta 3 Neutrophils Myeloid −2 Macrophages CTLA4 0 CD8 T Cells −4 TP53-wt TP53-mut TP53-wt TP53-mut Lymphoid ND/NA ND/NA NK Cells T−Cells 8 12 Th1 Cells P<0.0001 P<0.0001 Cytotoxic Cells Cytotoxicity TP53 statusp53 status ● MutatedMutated● WT/NA/ND WT/NA/ND TIGIT 9 CD45 6 B−Cells MMR.loss Apoptosis

Endothelial Cells mRNA

B mRNA

Stroma 6 IL10 4 γ Myeloid Inflammation ● Inflammatory Chemokines

20 IFN- ● Glycolytic Activity FoxP3 3 Hypoxia ● Exhausted CD8 ● 2 ● ● ● IFN Downstream ● ● TIS ● ●● ● ● ● IDO1 0 ● ● ● ● ● IFN Gamma ●●● ● ● ●● TP53-wt TP53-mut ● ● ● NK CD56dim Cells TP53-wt TP53-mut 0 ● ● ● ● PD1 ND/NA ND/NA ● ● ● ● ● ● ● ● NOS2 ● DC 15 ● ● ●● ●● ●● MAGEs P=0.0013 10 P<0.0001 ● ● ● ● APM PC2 (9.9%) ●● ● ● ● Treg ● PDL1 12 ● ● PDL2 8 B7−H3 −20 9

mRNA 6 ● 20/23 TP53-mut. 21/28 TP53-mut. 1/13 TP53-mut. mRNA 87% 75% 8% 6 4 CD8A High Intermediate Low GZMB 2 −40 3 −25 0 25 PC1 (26.4%) 0 TP53-wt TP53-mut TP53-wt TP53-mut Immune infiltration ND/NA ND/NA A unique immune TP53 classifier

A

TP53 status 4 p53 status SLC2A1HLA−A ●IL6 p53 status B ● ● ● TNFRSF10D● IRF1 CXCL1 ● ●BIRC5 OASL● FCAR 4 Mutated ● ● ● NFKBIE CD79A ● DUSP2 CXCR2 WT/NA/ND ● ● ● FAM124B ●PRF1 FOSL1 C PRF1 ●● ● DE immune genes between ●VSIRKIF2C IFNG 2 GIMAP4 ● ●GLS ● GIMAP6 CD79B● ● TP53 mutated and CD79A ● SLC1A5 Hallmark TP53 pathway genes CXCL8 TBXAS1 ● ●SFXN1 3 ● CCNE1 ●SGK1 AREG TP53 wild t AML ● ● ● ● ● CDKN2A●CXCL3 (M5939 MSigDB; n=200) ITGB3 0 ●NFATC2 ● ●FCAR OASL ● ●KLRB1 ●CSF1R●LY9 VCAM1 ● ●ATF3●PC THBS1 ● PTGS2●FCN1 ●● ● ● XCL1/2TNF ● ● ● −2 ● IL33 value ● HAVCR2 ● ● ● GBP1 ●TRAT1 195 0 35

p FCGR3A/B ADORA2A THBS1 CXCL2 ● CD27 ● ●CDC20● ● ● BRCA1 MKI67HLA ● ● ● ● ●ADM (46.1%) (8.3%) ● ● RRM2 (0%) CD79B ● ●CD6 ●NFAM1 ENTPD1 ● 10 2 ● CDKN1A TTC30A ● ● ●ISG15 FAM30A ● ●TNFSF13 ● PTCD2●ZAP70 ● ● ● CD96 ● CCL5 ZC3H12A ● ● ●CDKN1CIL10RA ● RIPK2 ●● ●FBP1 ●IL11RA ● ●ICAM1 ● ● ● HRAS DUSP5 ● ● 0 FAM30A MAML2FCGR1A ● VEGFA − log10(P.Value) ● ● ● ● ● ● IL10● Log ● ● CCL2 ● MFGE8 ●IFIT2 ●CCL2 BCL2 ● SERPINH1 ● ●● BMP2 - ● ● ● ● ● ● (0%) PTGER4 ● ● ●TICAM1 5 0 ● LYZ ●LCK● ● ● ● ● CXCL1 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● (1.2%) (0%) ● ● ● ● ● ●● ● ● ● CCL3/L1 ● ● ● ● ● ● ● ●● ●● ● ● ● ● ● ● ● ● THBD 1 ● ●● ● ● ● ●●●● ● ● ● ● ● ● ● ● ● ●● ● ● ● CCR2 ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ●● ● 188 ● ● ●● ● IFNG ● ● ● ●● ● ● ● ● ● ● ● ●●●● ● ●● ●● ● ● ● ● ● ● ●● ● ●● ●● ● MS4A1 ● ● ● ● ● (44.4%) ● ● ● ● ● ● ● ● ● ● ● ● ●●●● ● MYCT1 ●● ●●● ● ● ● ● ● ● ●●● ● ●● ●●● ●● ●●●● ● ● ● ● ●● ● ● ● CD19 ● ●●●●●● ●● ●● ●●●●●●● ● ●●●●●●● ●● ● ●●●● ● ●●● ● ● ● ● CX3CR1 ● ●●●● ●●● ●●●●●● ●●●●● ● ●● ●●● ●●●● ●● ●● ● ●●●●● ●●● CSF1 ●●●●●●● VTCN1 0 ●●● TP53 pathway genes in NCI-60 cell lines CD2 MARCO −2 −1 0 1 2 (M2698 MSigDB; n=193) IL6 log2(Fold−change) NT5E Log2 fold change BBC3 Up in TP53-wt AML Up in TP53-mut. AML (n=22 cases) (n=42 cases) TP53–related immune genes stratify survival

A In silico prognostic power in TCGA-AML cases (18 upregulated genes in TP53 mutated AML) C 100 Altered (median RFS=11.4 mo.) Not altered (median RFS=24.1 mo.) 75

50 STRING Interaction Network 25 P=0.0068 Percent survival

Log rank χ2=7.32; HR=1.86 (95%CI 1.12-3.1) “Altered”: 0 0 24 48 72 96 120 144 mRNA up-regulation Relapse-free survival time (months) amplification deep deletion

100 Altered (median OS=11.4 mo.) mis-sense mutations B Not altered (median OS=27.0 mo.) 75 KEGG Pathway Description Count in gene set FDR

-9 hsa04060 -cytokine interaction 9 of 263 5.56×10 50 hsa05323 Rheumatoid arthritis 6 of 84 6.37×10-8

-8 25 P=0.0009

hsa04657 IL-17 signaling pathway 6 of 92 8.02×10 Percent survival

hsa04621 NOD-like receptor signaling pathway 6 of 166 1.56×10-6 Log rank χ2=10.94; HR=1.88 (95%CI 1.24-2.85) 0 hsa04668 TNF signaling pathway 5 of 108 4.16×10-6 0 24 48 72 96 120 144 Overall survival time (months) Flotetuzumab immunotherapy cohort

Characteristic NCT#02152956 Patients (n=35)* Age (median and range) 54 years (27-74) Male 16 (46%) Gender Anti-CD3 Anti-CD123 Female 19 (54%) Late relapse (CR with initial duration >6 months) 7 (20%) Refractory to HMA 2 (5.7%) Disease status at study entry Primary induction failure (PIF; ≥2 induction attempts) 20 (57.1%) Refractory Early relapse (CR with initial duration <6 months) 6 (17.2%) Favorable 3 (8.6%) 2017 ELN risk stratification Intermediate 8 (22.9%) Flotetuzumab Adverse 24 (68.6%) Secondary AML 11 (31.4%) Number of prior lines of 3 (1-9) therapy (median and range)

*Subgroup of 35/50 patients treated at the RP2D for whom BM samples were available Response assessment criteria employed in analysis: Anti-leukemic activity (ALA): CR/CRh, PR, “other benefit” (>30% decrease in BM blasts compared to baseline) Non-responders (NR): treatment failure, stable disease, progressive disease Flotetuzumab cohort – TP53 mutations associate with an immune-infiltrated TME

A p53 status 3 p53 status CD45 p53 mut./17p abn. IL2RA 2 WT APM.loss Exhausted CD8 T cells 1 T cells B CD8 T cells Lymphoid cells 0 Cytotoxicity US005-0004 CR Cytotoxic cells −1 Neutrophils Myeloid cells −2 US011-0015 SD Macrophages * BATF3 IDO1 −3 US011-0006 PD PDL2 NOS2 US011-0020 CR B−cells * Treg cells NK cells US003-0012 MLFS TIGIT CTLA4 Th1 cells US011-0014 CRh CD56dim NK cells Hypoxia US007-0009 OB PD1 Stroma Time on treatment MAGEs US011-0018 PD PDL1 Time to HSCT DCs US007-0002 PD Inflammatory.chemokines IL10 Time to death/last FU Endothelial cells US006-0002 SD B7−H3 MMR.loss Alive at last FU TIS US010-0001 PD * APM IFN.downstream IFN.gamma Immunoproteasome 0 3 6 9 12 15 18 21 24 27 30 MHC2 Glycolytic.activity Mast cells Months from study entry Myeloid.inflammation TGF.beta Apoptosis Proliferation ARG1 JAK/STAT.loss ALA in 45.5% (5/11) evaluable patients with TP53 High Intermediate Low mutations and/or 17p abnormalities (3 TP53/17p) (4 TP53/17p) (2 TP53/17p) (2 CR, 1 CRh, 1 morphologic leukemia-free state [MLFS], and 1 OB)

Immune infiltration int.-to-high in 7/9 patients Response to flotetuzumab in TP53 mutated patients

A B 100 BM blasts (TP53 mut./17p abn.) P=0.016 8 8 8 P=0.032 6 P=0.016 P=0.016 5 50 7 6 6

BM N.A. 4

0 4 4 score

6 γ * 3 TIS score Treg score Treg IFN- -50 5 2 2 42% blast reduction 2 on average

-100 4 Inflammatory chemokine score 0 0 1 Best change from baseline (%) ALA NR ALA NR ALA NR ALA NR US011-0015 US011-0006 US011-0014 US011-0020 US011-0018 US010-0001 US007-0002 US006-0002 US007-0009 US003-0012 US005-0004 D HOVON (TP53 mutated) E HOVON cohort C 100 100 100 PIF (n=6) TP53 mutated (n=13; median OS=3.58 mo.) CR (n=7) PIF (n=125; median OS=3.78 mo.) 75 75 75 N=11 with TP53 mut./17p abn. Log-rank χ2=6.77; P=0.0093 Median=4.0 months (n=11) HR=7.1 (95%CI 1.62-30.96) 50 50 50

25 25

25 Percent survival Percent survival Percent survival 6.76 mo. 1.16 mo. 0 0 0 0 2 4 6 8 10 12 14 16 18 20 22 24 0 4 8 12 16 20 24 0 48 96 144 192 240 Survival from study enrollment (months) Overall survival time (months) Overall survival time (months) Conclusions

• Immune transcriptomic analyses of in silico and wet-lab cohorts of TP53 mutated AML suggest the presence of high T-cell infiltration and high expression of immune checkpoints and IFN-g signaling molecules compared with AML subgroups with other risk-defining molecular lesions • Immunotherapy with flotetuzumab may be efficacious in individuals with altered TP53 status, with an overall reduction of BM blasts averaging 42% and with evidence of ALA in 45.5% (5/11) of the patients • The overall response rate observed in TP53-mutated patients treated with flotetuzumab encourages further study of this immunotherapeutic approach HEFCE business plan 2011-2015

Principles, priorities and practices

Acknowledgements

Co-authors and Collaborators Funding Sources

JVGCRC, NTU Stephen Reeder Jayakumar Vadakekolathu Tung On Yau Leonido Luznik Sidney Kimmel Comprehensive PhD Students Cancer Centre Barbara Seliger Jenny Ashforth Baltimore, MD National Priorities Research Programme, Claudia Wickenhauser Melissa Courtney 2016-2020 Martin Luther University 2011/34 Halle – Wittenberg, Germany November 2011

Joseph M. Beechem Mark D. Minden Sarah E. Church Toronto, Canada Sarah E. Warren Mainstream QR Funding, 2017-2020 Heidi Altmann Seattle, WA Martin Bornhäuser Jan Moritz Middeke Marc Schmitz Friedrich Stölzel SAL Studienallianz Leukämie Jan K. Davidson-Moncada Dresden, Germany Tasleema Patel Francesco M. Marincola John Muth Sarah K. Tasian Menlo Park, CA Rockville, MD Philadelphia, PA