Imprime PGG triggers a coordinated anti-cancer immune ... · TGI
Transcript of Imprime PGG triggers a coordinated anti-cancer immune ... · TGI
Imprime+ DC101
DC101 Vehicle 0
2
4
6
8
Rel
ativ
e Fo
ld C
hang
e O
ver V
ehic
le C
trls ****
TNFα Expression
0.0
0.5
1.0
1.5
Rel
ativ
e Fo
ld C
hang
e O
ver V
ehic
le C
trls
****CD206 Expression
Imprime+ DC101
DC101 Vehicle 0.0
0.5
1.0
1.5
Rel
ativ
e Fo
ld C
hang
e O
ver V
ehic
le C
trls
******
Bev Vehicle Imprime+ Bev
Tumor associated Neutrophils
1000
0
500
****0.0001
CD
86 G
MFI
Tumor associated Macrophages
200
400
0
600
Bev Vehicle Imprime+ Bev
Abstract #601
Imprime PGG triggers a coordinated anti-cancer immune response in concert with anti-angiogenic antibodies, re-polarizing the immune microenvironment to suppress tumor growth Kathryn Fraser, Nadine Ottoson, Xiahong Qiu, Anissa SH Chan, Steven Leonardo, Adria Jonas, Takashi Kangas, Jeremy Graff, Nandita Bose
Biothera, Eagan, MN, USA 55121; [email protected]
Abstract
Background
Summary
Imprime PGG (Imprime) is a soluble, yeast-derived β-1,3/1,6 glucan in clinical development for the treatment of cancer in combination with other anti-cancer therapies. Imprime acts as a Pathogen Associated Molecular Pattern (PAMP) and can be recognized by cells of the innate immune system. Preclinical data using human whole blood from healthy volunteers show that Imprime binding to innate immune cells triggers a coordinated immune response that includes repolarization of M2 macrophages, activation of neutrophils and maturation of dendritic cells. This response ultimately leads to cross-talk with the adaptive immune system driving T cell expansion and the production of interferon gamma (IFNγ). In a randomized phase 2 clinical study in stage IV non-small cell lung carcinoma (NSCLC) patients, treatment with Imprime plus bevacizumab (bev; anti-VEGF antibody), carboplatin and paclitaxel showed a median overall survival of 16.1 months versus 11.6 months in patients not receiving Imprime. We sought to explore a mechanistic understanding for this promising clinical activity. Angiogenic factors, such as VEGF, not only drive the formation of new leaky vessels but also facilitate the establishment of a suppressive immune microenvironment enabling tumor survival and growth. Recent work has shown that anti-angiogenics not only block neovascularization but may also promote a shift in the immune microenvironment, enabling immune recognition and destruction of the tumor. We therefore sought to evaluate whether Imprime may complement the effect of anti-angiogenics on the immune microenvironment. We tested Imprime in combination with either bev or DC101 (anti-VEGFR2) in distinct NSCLC xenograft models in athymic nude mice. Once tumors reached a mean size of 100mm3, mice were treated with Imprime, bev or DC101. H1299 and H441 tumor-bearing mice were used in the bev and DC101 studies, respectively. In the bev study, Imprime plus bev induced >75% tumor growth inhibition in ~50% of mice vs 20% in the bev alone groups. Both macrophages and neutrophils from spleen and tumor tissue of combination-treated mice showed significant upregulation of the activation marker CD86 compared to tissues from bev alone treated mice. Moreover, splenic MDSCs in combination-treated mice showed significantly increased iNOS2 expression with reduced Arg-1 expression compared to bev alone treated mice. Tumors from the Imprime plus bev groups showed significantly reduced expression of the potent immunosuppressor, TGFβ, when compared to tumors from mice treated only with bev-with the greatest reduction evident in the tumors with the greatest growth inhibition. In the H441 tumor-bearing mice treated with Imprime and DC101, a significant suppression of tumor growth compared with DC101 alone was also observed and additional mechanistic studies in this model are ongoing. These data show for the first time that Imprime-based treatment prompts a shift in the immune microenvironment of a tumor in situ, eliciting enhanced tumor growth inhibition in concert with anti-angiogenic therapy.
• Imprime is a soluble yeast-derived β-1,3/1,6 glucan immunomodulator (Figure 1)
being developed for cancer treatment in combination with anti-tumor antibodies. • In a randomized phase 2 clinical study, stage IV NSCLC patients treated with
Imprime plus the anti-VEGF antibody bevacizumab (bev), carboplatin and paclitaxel showed a median overall survival of 16.1 months versus 11.6 months in patients not receiving Imprime.
• Imprime, a pathogen-associated molecular pattern (PAMP), forms an immune
complex with endogenous anti-β-glucan antibodies, then binds and primes innate immune cells including macrophages, neutrophils, and DCs (Figure 2). Activation of the above innate cells is central to influencing adaptive immune cell responses. Generating functional and long-lived anti-tumor innate and adaptive immune responses is key to providing durable tumor control.
• In addition to blocking neovascularization, anti-angiogenics can promote a shift in the immune microenvironment enabling immune activation.
• OBJECTIVE: To evaluate the ability of Imprime to complement the effect of anti-
angiogenics on the immune microenvironment in in vivo xenograft models of NSCLC.
Results Figure 1: The general structure of yeast-derived Imprime PGG
Imprime enhances macrophage, neutrophil and DC activation and function. (A) M1 or M2 macrophages were prepared by culturing Imprime-bound monocytes isolated from whole blood in the presence of M1 or M2-polarizing conditions (XVivo 10 media supplemented with 10% autologous serum and 50 ng/mL recombinant human GM-CSF or M-CSF respectively) for 6 days. Macrophages were subsequently evaluated for phenotype, enhancing CD4 T cell proliferation by CFSE-dilution assay, and IFN-γ production. (B.; first panel) Whole blood was incubated with +/-Imprime for 2hrs and cells were analyzed by flow cytometry. (B.; second panel) Whole blood was treated as in above and neutrophils were isolated by negative selection. Cells were then mixed with Raji cells +/- Rituximab treatment (1µg/ml) at a 25:1 neutrophil:Raji ratio in the presence of luminol (50µM). Cells monitored for production of reactive oxygen species (ROS) measured by relative light units (RLU). (C) MoDC were prepared by culturing Imprime-bound monocytes isolated from whole blood in the presence of XVivo15 media supplemented with 10% autologous serum and 50 ng/mL recombinant human GM-CSF + 50 ng/mL IL-4 conditions conditions for 6 days and then maturing the cells with LPS and TNF-α (50 ng/mL) for 48 hrs. MoDC were subsequently evaluated for phenotype, enhancing CD4/CD8 T cell proliferation by CFSE-dilution assay and IFN-γ production.
Figure 4: Splenic macrophages isolated from Imprime+Bev treated mice display a M1-like phenotype.
Imprime+ Bev
Imprime+ Bev
Gated on CD11b+ cells
CD
68
F4/80 Bev
400
300
350
200 Vehicle
iNO
S G
MFI
250
*0.0108 1000
600
800
200
400
0
CD
86 G
MFI
Bev Vehicle
*0.0356
Bev Vehicle Imprime+ Bev
Arg
1 G
MFI
400
300
350
250
**0.0027
Splenic macrophages are M1. At day 20 post tumor injection, spleens were harvested and single cell suspensions were stained with mouse antibodies and analyzed by FACS. GMFI was calculated in Flowjo after gating on CD11b+CD68+F4/80+ cells.
Figure 7: Imprime+DC101 treated animals have reduced splenic MDSCs, an increase in activated splenic macrophages, and an increase in Th1-like phenotype on cells within the tumor microenvironment.
Imprime+ Bev
TGFβ
pg/
mL
200
400
0
300
100
**0.0048
***0.0001
Bev Vehicle >50% TGI
<50% TGI
Immune cells are activated in the tumor microenvironment. (A) At day 37 post tumor injection, spleens were harvested and single cell suspensions were analyzed by FACS. Frequency or GMFI was calculated in Flowjo after gating on indicated cells. (B) Whole tumors were digested and mRNA levels of TNFα, CD206, and TGFβ1 were analyzed by qRT-PCR.
Minutes
Figure 2: Imprime PGG triggers a coordinated immune response in human ex vivo studies.
100
80
60
40
20
0
Rea
ctiv
e O
xyge
n S
peci
es (R
OS
) in
RLU
50 37.5 25 12.5 0
PMN Treatment Tumor Treatment
Vehicle
Imprime
Vehicle
Imprime
None
None
Rituximab
Rituximab
CD11b CD62L CD88
Imprime Vehicle
B.
A. CD163 CD86 PD-L1
0 102 103 104 105
Cou
nt
Cou
nt
0 102 103 104 105
Median 58
168 158
Median 86
2434 759
Median 75
2398 2329
Median 88
1102 1253
Median 93
5189 4234
Median 103 439 790
MFI
M1
M2 CD
4 T
cell
prol
ifera
tion
Div
isio
n In
dex
Isotype ctrl staining M1/M2-Imprime M1/M2-Vehicle
1.0
0.8
0.6
0.4
0.2
0.0
NS ****
IFNγ
(pg/
mL)
800
600
400
200
0
NS **
C.
0
5
10
15
20
25
% o
f Pro
lifera
ting
Cells
**p = 0.0038 ***p = 0.0004
Cou
nt
CD80 CD86 CD83 HLA-DR
Mean Fluorescence Intensity (MFI)
Median 151
4286 7049
Median 223 640 744
Median 162 168 466
Median 151 871
10759
Isotype ctrl staining Imprime Vehicle
Allo-CD4 T Cells
0
50
100
150
IFNγ
(pg/
ml)
**p = 0.0026
% P
rolif
erat
ing
Cel
ls
Allo-CD8 T Cells
IFNγ
(pg/
mL)
Figure 3: Anti-tumor efficacy of Imprime PGG with anti-angiogenics in vivo.
Xenograft models. (A) Outline of in vivo studies. (B) Graph depicting H1299 tumor volume of individual mice at day 18 post challenge. (C) %Tumor Growth Inhibition (TGI) of H1299 tumors was calculated by %TGI= (1-individual treated mouse/median control mice)*100. (D) Graph depicting H441 tumor volume of individual mice at day 37 post challenge. (E) %TGI of H441 tumors was calculated by %TGI= (1-individual treated mouse/median control mice)*100.
A.
Figure 6: Imprime+Bev tumors have cells that show a more activated phenotype and a decrease in TGFβ.
C.
Ex-Vivo human studies: • Imprime enhances human macrophage and MoDC activation and function which
leads to crosstalk with the adaptive immune system.
• Imprime elicits the generation of Reactive Oxygen Species (ROS) from neutrophils specifically in response to antibody-coated tumor cells.
In-Vivo studies:
• Imprime PGG treatment in vivo can activate myeloid cells within both the tumor and spleen to orchestrate a profound shift in the immune microenvironment which promotes tumor recognition and suppression.
Treatment Groups (2x/week): 1. Vehicle 2. Imprime (1.2mg/ms) 3. Bevacizumab/DC101 (5/10mg/kg IP) 4. Bevacizumab/DC101 (5/10mg/kg IP)+ Imprime (1.2mg/ms IV)
Tumors meet mean size of ~100mm3
start treatment
Human NSCLC cells (H1299 or H441) Given s.c.
Nude mouse: No T, B cells
Control Tumor Endpoint Animals Sacrificed
Spleen, Tumors analyzed
iNOS GMFI CD86 GMFI Arginase 1 GMFI
A.
Live
/Dea
d
CD11b Bev Vehicle Imprime+ Bev
Bev Vehicle Imprime+ Bev
PD
-L1
GM
FI
200
400
0
100
300
iNO
S G
MFI
Arg
1 G
MFI
1000
1500
500
0
2000
1000
1500
500
0
2000 *0.033 *0.091
*0.029
Bev Vehicle Imprime+ Bev
iNOS GMFI Arginase 1 GMFI PD-L1 GMFI
Bev Vehicle Imprime+ Bev
PD
-L1
RFC
/ V
ehic
le C
ontro
l
10
15
5
0
20 *0.0430
iNO
S R
FC/
Veh
icle
Con
trol
Bev Vehicle Imprime+ Bev
10
15
5
0
*0.0275
Bev Vehicle Imprime+ Bev
Arg
1 R
FC/
Veh
icle
Con
trol
1.0
1.5
0.5
0
*0.0320
PD-L1 Expression iNOS Expression Arginase 1 Expression B.
CD86 GMFI
Myeloid cells from tumors are more activated. Cells were harvested from the tumor using type I collagenase. (A) GMFI was calculated after gating on CD11b+ cells. (B) mRNA levels of PD-L1, iNOS and Arg1 from single cell suspensions were analyzed by qRT-PCR. (C) GMFI was calculated after gating on Ly6g+ CD11b+ (red) or Gr1-CD11b+ (blue) cells. (D) Cells were incubated overnight in XVivo10 media and supernatants were then analyzed for TGFβ concentration by ELISA
Ly6g
CD11b
TGFβ Expression
Figure 5: Imprime+Bev splenic MDSC display an increase in iNOS and decrease in Arginase 1.
MDSCs show a decrease in Arg1. At day 20 post tumor injection, spleens were harvested and single cell suspensions were stained with mouse antibodies and analyzed by FACS. GMFI was calculated in Flowjo after gating on CD11b+Gr1+ cells.
GR
1
CD11b
Gated on Lymphocytes
600
800
200
400
0
iNO
S G
MFI
Bev Vehicle Imprime+ Bev
***0.0007
Bev Vehicle Imprime+ Bev
Arg
1 G
MFI
600
400
500
300
**0.0022
iNOS GMFI Arginase 1 GMFI
Vehicle 0
2500
1500
2000
1000
500
Imprime Bev
Day
18
Tum
or V
olum
e (m
m3 )
Tumor Volume
Bevacizumab 5mg/kg
Bevacizumab 5mg/kg +Im
prime 1.2mg/m
s
Bevacizumab 10mg/kg
Bevacizumab 10mg/kg +Im
prime 1.2mg/m
s0
20
40
60
80
100
%TG
I Day
18D
ay 1
8 %
TGI
0
80
100
60
40
Bev Imprime+ Bev
20
75%
B. C.
Imprime+ Bev
Tumor Growth Inhibition
Imprime+ DC101
0
1500
2000
1000
500
Vehicle Imprime DC101
Day
37
Tum
or V
olum
e (m
m3 )
D. E.
0
80
100
60
40
20
Day
37
%TG
I
*0.0123 *0.0179
75%
Imprime+ DC101
DC101
DC101/H441 Xenograft Model
Tumor Volume Tumor Growth Inhibition
Bevacizumab/H1299 Xenograft Model
A.
B.
CD86 GMFI Macrophages
Imprime+ DC101
DC101 Vehicle Imprime+ DC101
DC101 Vehicle
Vehicl
e Contro
l
DC101
DC101+
Imprim
e0.0
0.1
0.2
0.3
%M
o-M
DS
C in
spl
een
0.3
0.2
0.1
0
%C
D11
b+G
r1+
Ly6c
+ ce
lls
***0.0003
% Mo-MDSC
Vehicl
e Contro
l
DC101
DC101+
Imprim
e0
1
2
3
4
5
% P
MN
-MD
SC
in s
plee
n
%C
D11
b+G
r1+
Ly6c
dim
cel
ls 5
4 3 2
1 0
% PMN-MDSC
Vehicl
e Contro
l
DC101
DC101+
Imprim
e0
20
40
60
80
%C
D45
+CD
11b+
GR
1-C
D68
+F4/
80+
Cel
ls
80
60
40
20
0
****0.0001 % Macrophages
****0.0001
Imprime+ DC101
DC101 Vehicle
*0.0271
%C
D11
b+C
D68
+ F4
/80+
cel
ls
Vehicl
e Contro
l
DC101
DC101+
Imprim
e0
1000
2000
3000
4000
CD
86 G
MFI
CD
45+C
D11
b+G
R1-
CD
68+F
4/80
+ C
ells
4000
3000
2000
0
1000
Imprime+ DC101
DC101 Vehicle
****0.0001 **0.0051 **0.0028
CD
86 G
MFI
CD
11b+
C
D68
+F4/
80+
cells
SPLEEN
TUMOR TGFβ Expression
Imprime+ DC101
DC101 Vehicle