#4982. Anti-mCTLA-4 Treatment Results in Early and Late Immune Response Effects in a Murine Model of Colorectal CarcinomaDavid Draper, Alden Wong, Stacey Roys, Scott Wise, and Maryland Rosenfeld FranklinCovance Laboratories, Inc., 800 Technology Drive, Ann Arbor, MI 48108 USA

Presented at the AACR Annual Meeting 2019, March 31- April 3, 2019 – Atlanta, GA

Introduction and Background▶ During pre-clinical research, analysis of the tumor immune response at a single post-treatment timepoint

is often standard when therapeutic mechanism of action is investigated.

▶ The use of a single sampling point for analysis can limit research on new immune modulation therapies as it provides only a partial view into the dynamic nature of the developing immune response in the tumor.

▶ In this study we examined the kinetics of tumor-directed immune infiltration and activation on day 11 and day 17 post-implantation using the CT26 model for colorectal carcinoma.

▶ The CT26 model is moderately responsive to checkpoint inhibition. To examine early and late effects that checkpoint inhibition has on the immune response, tumor-bearing mice were treated with anti-mCTLA-4. Tumor analysis corresponded to day 4 and day 10 post-treatment.

▶ Endpoints analyzed consisted of:

– 11 distinct tumor-infiltrating subsets

– CD8+ T cell activation marker expression

– T cell cytokine analysis

– Effector/Memory T cell phenotype

▶ We hypothesized that multiple sampling points would uncover unique immune subset specific profiles of infiltration and activation in the CT26 tumor.

Materials and Methods▶ CT26 cells were implanted subcutaneously into the right axilla of BALB/c mice. Mice were randomized

into treatment groups following establishment of tumors. Anti-mCTLA-4 antibody (clone 9D9) was acquired from Bio X Cell and dosed IP. Tumor progression was monitored by caliper measurements.

▶ For immune profiling, tumors were dissociated into single cell suspensions (gentleMACS™, Miltenyi), labeled with fluorescent antibodies, and analyzed by an Attune™ NxT flow cytometer (Thermo Fisher Scientific). The immune subsets were then delineated using FlowJo (FlowJo, LLC). The number of cells/gram of tumor was quantified using Precision Count Beads™ (BioLegend). For effector/memory T cell analysis, a spleen from a naïve BALB/c mouse was analyzed for comparison.

▶ Tumor-infiltrating immune subsets were immunophenotyped using the Covance MI-CompLeukocyte™ package and MI-T Effector/Memory™ panel.

▶ For IFNγ and granzyme B analysis, tumor-derived cells were stimulated ex vivo with PMA and ionomycin for 5 hours in the presence of brefeldin A. Following incubation, non-adherent cells were collected and immuno-stained with fluorescent antibodies targeting cell surface receptors. The cells were then fixed, permeabilized and stained with anti-IFNγ and granzyme B antibodies prior to sample acquisition.

Figure 1. Anti-mCTLA-4 Inhibits CT26 Tumor Growth.(Left) In vivo study design and timeline. (Right ) Mean tumor volumes on days 11 and 17 post implant (n=10/group).

Figure 2. TIL Kinetics are Regulated by Anti-mCTLA-4.A increase in the infiltration of CD8+ T cells, B cells, macrophages and G-MDSCs was observed from day 11 to day 17. Helper T cell, NK, NKT and M-MDSC infiltration peaked at the day 11 timepoint.Anti-mCTLA-4 treatment enhanced infiltration of all T cell subsets, NKT cells and M-MDSCs at day 11 and day 17.Anti-mCTLA-4 treatment triggered marked increases in NK and M1 TAM infiltration at day 11 only.Anti-mCTLA-4 inhibited the infiltration of M2 TAMs at day 17.

Figure 4. Anti-mCTLA-4 Increases IFNγ in CD8+ T Cells and NKT Cells.(Top) Representative plots to illustrate IFNγ and granzyme B responses following treatment with PMA and ionomycin (Gate: Tumor CD3+ cells). (Bottom) Anti-mCTLA-4 treatment enhanced IFNγ (but not granzyme B) production in CD8+ T cells and NKT cells.

Figure 5. Anti-mCTLA-4 Prolongs Effector CD8+ T Cell Infiltration.Anti-mCTLA-4 triggered an increase in effector CD8+ T cells in the tumor and a corresponding decrease in naïve and memory T cells at day 17.

Figure 3. Anti-mCTLA-4 Alters the Kinetics of T Cell Activation.(Left) CD69, PD-1 and Ki-67 expression increased on CD8+ T cells from day 11 to day 17. (Right) Anti-mCTLA-4 delayed CD69 upregulation but upregulated PD-1 and Ki-67 expression at the day 11 timepoint.

Day 11 Day 170

500

1000

1500

Tum

or V

olum

e (m

m3 )

IsotypeAnti-mCTLA-4

Day 0 Day 7 Day 17

Subcutaneous Implant(n=10 Animals/Group)

Tumor Sampling

Anti-mCTLA-4 (Intraperitoneal; 10 mg/kg)

Day 7 – Cohort with established tumors

Study Timeline

77% TGI

Day 11 Day 170

100

200

Cells

/gra

m tu

mor

(X10

4 )

Day 11 Day 170

150

300

Day 11 Day 170

50

100

Day 11 Day 170

25

50

Day 11 Day 170

250

500IsotypeAnti-mCTLA-4

Day 11 Day 170

100

200

Cells

/gra

m o

f tum

or (X

104 )

Day 11 Day 170

300

600

Day 11 Day 170

300

600

Day 11 Day 170

150

300

Day 11 Day 170

400

800

Day 11 Day 170

50

100

Helper T Cells CD8+ T Cells Tregs B Cells NK Cells

NKT Cells M1 TAM M2 TAM G-MDSC M-MDSC Dendritic Cells

* *

*

*

* *

**

*

** *

** *

* p < 0.05 (Student’s T-test)

Results and Conclusions▶ Using the CT26 model, we profiled the kinetics of immune subset infiltration in tumors with or without

checkpoint blockade. The results demonstrate that kinetic analysis can reveal trends that would go unnoticed if analysis was instead performed at a single time point.

▶ Notably, while the tumor infiltration of most subsets increased from day 11 to day 17, Helper T cell, NK, NKT and M-MDSC subsets peaked at day 11. Furthermore, the effects of anti-mCTLA-4 had temporal components, which included NK and TAM infiltration, Ki-67 expression in CD8+ T cells, IFNγ responses in CD8+ T cells and NKT cells and memory T cell formation.

▶ Taken together, these data emphasize that a model-specific understanding of how the immune response develops during tumor progression is important. These characteristics can drive model selection and sampling time decisions when different target endpoints are considered for analysis.

Day 11 Day 170.0

12.5

25.0

%of

CD8+

TCe

lls

Day 11 Day 170

50

100

%of

CD8+

TCe

lls

Day 11 Day 170

25

50

%of

CD8+

TCe

lls

CD62L

CD44

Effector Naïve

Effector Memory

Naïve SplenicCD8+ T Cells

Day 11 Day 170

1

2

%of

CD8+

TCe

lls

IsotypeAnti-mCTLA-4

Central Memory

CT26 Tumor-InfiltratingCD8+ T Cells

*

*

*

*

* p < 0.05 (Student’s T-test)

Negative stainingcontrol

Ki-67 stained

Day 11 Day 170

50

100

CD69

Percen

tPositive

Day 11 Day 170

50

100

PD-1

Percen

tPositive

Day 11 Day 170

2000

4000

6000

8000

Ki-67MFI

Anti-mCTLA-4Isotype

CD69 PD-1 Ki-67

* p < 0.05 (Student’s T-test)*

* * *

*

*

Gate: Tumor-Infiltrating T Cells

CD4

CD8

CD69

PD1 Ki-67

Nor

mal

ized

to M

ode

CD8

IFNγ

Gra

nzym

e B

Day 11 Day 170

50

100

%IFNgPo

sitiv

e

Day 11 Day 170

50

100

%IFNgPo

sitiv

e

Day 11 Day 170

50

100

%Gran

zymeBPo

sitiv

e IsotypeAnti-mCTLA-4

CD49b CD8

IFN

γ

Media Control PMA/Ionomycin Media Control PMA/Ionomycin Media Control PMA/Ionomycin

IFNγ+

CD8+ T CellsGranzyme B+

CD8+ T Cells

**IFNγ+

NKT Cells

* p < 0.05 (Student’s T-test)