`Monika Semmrich [1], Jean-Baptiste Marchand [2], Matilda Rehn [1], Laetitia Fend [2],Christelle Remy [2], Petra Holmkvist [1], Nathalie Silvestre [2], Carolin Svensson [1], Patricia Kleinpeter [2], Jules Deforges [2],
` Fred Junghus [1], Linda Mårtensson [1], Johann Foloppe [2], Ingrid Teige [1], Eric Quéméneur [2], Björn Frendéus [1]
`[1] BioInvent International AB, Lund, Sweden, [2] Transgene S.A., Illkirch-Graffenstaden, France
`Abstract #746
`
`Non Injected
`VVGMαCTLA4
`VVGMαCTLA4 + αCD4
`VVGMαCTLA4 + αCD8
`
`100
`
`0 20
`
`20 25 30
`35
`40
`45 50
`Days post cell injection
`
`55
`
`60
`
`80
`
`60
`
`40
`
`20
`
`0
`
`Percent survival
`
`3. Anti-tumor activity in vivo
`Figure 6. Vectorization in oVV allows intratumoral accumulation of transgenes with low systemic exposure
`Amount of transgenes in tumor vs blood was measured for both anti-CTLA-4 mAb and GM-CSF (not shown) in syngeneic
`CT26 model.
`a)
`Tumor
`
`100
`
`systemic
`
`b)
`
`40
`
`Tumor
`
`systemic
`
`25
`
`10
`
`1
`
`0.1
`
`0.01
`
`EC90
`EC10
`
`30
`
`20
`
`10
`
`20
`
`15
`
`10
`
`αCTLA4 μg/mL serum
`
`100
`
`10
`
`1
`
`0.1
`
`0.01
`
`αCTLA4 μg/g tumor
`
`4. Mode-of-action characterization
`Figure 10. Anti-tumor response is CD8 T cell-dependent
`a)
`b)
`
`****
`****
`
`injected
`Tumor
`
`Spleen
`
`30
`
`20
`
`10
`
`0
`
`15
`
`% IFNγ+ TNFα+ (of CD8+ cells)
`
`c)
`
`10
`
`*
`
`02468
`
`0.6
`
`% AH-1+ CD8+ (of live cells)
`
`b, c) i.t. VVGM-αCTLA4
`induced tumor-specific
`CD8+ T cells both in in-
`jected tumors and in
`peripheral
`compart-
`ments as assessed by
`c) ex vivo restimulation
`of splenocytes with
`CT26
`(AH-1)-specific
`peptide
`and
`the
`percentage of IFN-γ+
`and TNFα+ CD8+ T
`cells or b) dextramer
`staining of AH1- speci-
`fic CD8+ T cells.
`
`***
`
`*
`
`*
`
`**
`
`contralateral
`Tumor
`
`0123455
`
`30
`10
`
`% IFNγ+ TNFα+ (of CD8+ cells)
`
`**
`
`****
`
`VVGM
`αCTLA4
`
`VVGM
`
`αCTLA4
`
`PBS
`
`% AH-1+ CD8+ (of live cells)
`
`0.4
`
`0.2
`
`0.0
`
`Depletion of CD8+
`and CD4+ T cells
`prior to treament of
`CT26-bearing mice with VVGM-αCTLA4 demons-
`trated a crucial role for CD8+ T cells in the anti-
`tumor response.
`d)
`
`adaptive immune responses
`cytolysis
`granzyme-mediated apoptotic signaling pathways
`neutrophil chemotaxis
`neutrophil migration
`
`size
`
`0.001
`
`0
`
`VVGM αCTLA4
`αCTLA4 i.p.
`8
`4
`1
`Days post treatment start
`
`10
`
`0.001
`
`0
`
`1
`
`8
`4
`Days post treatment start
`
`10
`
`1. Background
`Treatment with checkpoint inhibitor antibodies results in long-lasting antitumor responses in a variety of cancers [1].
`However, only a small fraction of patients responds to the treatment, probably due to inadequate tumor infiltration with
`immune cells. While combination therapy with anti-CTLA-4 and anti-PD-1 antibodies significantly improves efficacy, con-
`cerns with tolerability has limited wide-spread clinical use [2].
`Here we present a potentially safe and more efficacious strategy to combine anti-CTLA-4 and anti- PD-1/PDL1 checkpoint
`inhibition in the context of oncolytic virotherapy. A Treg-depleting anti-CTLA-4 antibody has been vectorized alongside
`GM-CSF into the Invir.IO™ oncolytic Vaccinia virus (oVV) based platform. This product named BT-001 (VVGM-αhCTLA4)
`consists of the Copenhagen oVV strain - deleted in J2R and I4L viral genes allowing restricted replication in proliferating
`cells - and the human CTLA-4-specific antibody 4-E03 IgG1, which shows improved Treg-depletion compared with ipilimu-
`mab.
`
`500
`
`400
`
`300
`
`MFI
`
`200
`
`100
`
`CD8 T cell
`
`3.
`
`
`
`CTLA-4αCTLA-4
`
`GM-CSF
`
`2.
`
`1.
`
`Figure 7. Improved survival after treatment with VVGM-αCTLA4 in several syngeneic tumor models
`CT26
`A20
`EMT6
`MC38
`B16
`
`a) Concentration of anti-mouse CTLA-4
`mAb 5-B07 was monitored in syngeneic
`CT26 tumor model. CT26 colorectal tumor
`cells were implanted subcutaneously to
`Balb/c mice and virus was injected 3 times
`i.t. One injection intraperitoneally of 5-B07
`at 3 mg/kg was used as benchmark.
`b) Intratumoral administration of VVGM-αCTLA4 was associated with sustained intratumoral antibody exposure and intratu-
`moral Treg depletion, systemic exposure was orders of magnitude lower and not associated with peripheral Treg depletion.
`
`VVGM
`αCTLA4
`
`PBS
`
`05
`
`% FoxP3+ cells (of CD4+)
`
`0
`
`VVGM
`αCTLA4
`
`PBS
`
`% FoxP3+ cells (of CD4+)
`
`VVGM
`αCTLA4
`
`VVGM
`
`αCTLA4
`
`PBS
`
`01234
`
`% IFNγ+ TNFα+ (of CD8+ cells)
`
`7 1
`
`4
`23
`
`29
`
`2000
`
`1000
`
`0
`
`2000
`
`1000
`
`8/10
`
`60
`
`0
`
`0
`
`8/10
`
`20
`
`40
`
`60
`
`2000
`
`1000
`
`0
`
`3/10
`
`0
`
`20
`
`40
`
`60
`
`80
`
`100
`
`2000
`
`1000
`
`VVGM αCTLA4
`control VV
`
`2000
`
`1000
`
`10/10
`
`10/10
`
`0
`
`0
`
`20
`
`40
`
`60
`
`0
`
`0
`
`20
`
`40
`
`60
`
`80
`
`Tumor volume (mm3)
`
`Figure 2. In primary patient
`material, CTLA-4 is highest
`expressed on
`intratumoral
`Treg cells which makes them
`a good target for Treg deple-
`ting antibodies.
`
`Treg
`CD8+
`CD4+ non-Treg
`
`CD8+ tumor
`CD8+ ascites
`tumor
`ascites
`Patient blood
`healthy PBMC
`tumor
`ascites
`Patient blood
`healthy PBMC
`
`0
`
`fold change
`3.0
`2.5
`2.0
`1.5
`
`d) Gene expression analyses of
`tumors from VVGM-αCTLA4-injected
`compared to VV-injected CT26 tu-
`mor-bearing mice supported the
`CD8+ T cell-dependent (and potenti-
`ally NK cell-mediated granzyme-de-
`pendent) anti-tumor immunity.
`
`*p < 0.05, **p < 0.01, ***p < 0.005, ****p < 0.001
`by one-way ANOVA
`
`Figure 11. Intratumorally induced CD8+ T cell anti-tumor immunity is cDC1-dependent
`
`WT/ PBS
`Batf3 -/-/ PBS
`WT/ VVGM αCTLA4
`Batf3 -/-/ VVGM αCTLA4
`
`100
`
`80
`
`60
`
`40
`
`20
`
`Percent survival
`
`2000
`
`1500
`
`1000
`
`500
`
`0
`
`20
`
`40
`
`100
`80
`
`60
`40
`20
`
`0
`
`0
`
`20
`40
`60
`Days post tumor cell injection
`
`100
`80
`
`60
`40
`20
`
`0
`
`0
`
`20
`40
`Days post tumor cell injection
`
`60
`
`100
`80
`
`60
`40
`20
`
`0
`
`0
`
`20
`40
`60
`Days post tumor cell injection
`
`100
`80
`60
`40
`20
`
`0
`
`0
`
`100
`80
`60
`40
`20
`Days post tumor cell injection
`
`0
`
`60
`40
`20
`Days post tumor cell injection
`
`80
`
`100
`80
`60
`40
`
`20
`
`0
`
`Percent survival
`
`Figure 1. Local administration of BT-001 combines on-
`colytic activities with high intratumoral concentrations of
`anti-CTLA-4 antibody and GM-CSF eliciting a stronger
`antitumor response with improved safety profile.
`
`CD4+ non-Treg
`
`Treg
`
`NOG mice
`
`a)
`
`8.0
`
`hCD80
`
`hCD86
`
`2.0
`
`1.5
`
`Naive CD8+ T cells
`Foxp3+ CD8+ T cells
`Intermediate CD8+ T cells
`CD8+ effector T cells (Klrg1+, T3)
`Activated CD8+ T cells
`Act./Exhausted CD8+ T cells (PD1+ TIM3+, T2)
`Exhausted CD8+ T cells (PD1+ TIM3+, T2)
`Resting CD4+ T cells
`Intermediate CD4+ T cells
`Activated CD4+ T cells
`Treg cells (FoxP3+ Klrg1-)
`Activated Treg cells (FoxP3+ Klrg1+, T1)
`
`Control VVGMαCTLA4
` treated
`Spleen
`
`VVGMαCTLA4
` contralateral
`
`Naive CD8+ T cells
`Intermediate CD8+ T cells
`Activated CD8+ T cells I
`Activated CD8+ T cells II (GzmBhi, Klrg+, S1)
`Naive CD4+ T cells
`Intermediate CD4+ T cells
`Activated CD4+ T cells
`Resting Treg cells
`Treg cells
`Activated Treg cells
`
`20
`
`0
`
`S1
`
`100
`
`80
`
`60
`
`40
`
`20
`
`CD4+
`
`CD8+
`
`Treg
`
`CD8+
`
`UMAP2
`
`500
`
`0
`
`60
`
`Tumor volume (mm3)
`
`0
`
`4-E03
`ipilimumab
`
`2-C06
`
`2-F09
`
`Control
`
`CD8+ T cells (% of control)
`
`0
`
`4-E03
`ipilimumab
`
`2-C06
`
`2-F09
`
`Control
`
`Treg cells (% of control)
`
`2. Generation of BT-001 and mouse surrogate VVGM-αCTLA4
`Figure 3. Treg depleting and blocking anti-CTLA-4 mAb 4-E03
`BioInvent‘s F.I.R.S.T. discovery platform [3] was used to isolate
`scFv antibody fragments recognizing human or mouse CTLA-4.
`Target-specific antibody clones were classified as actives, transfer-
`red to full-length IgG format, and further characterized biochemi-
`cally and functionally. The antibody clone 4-E03 was chosen as
`candidate for vectorization in Copenhagen oVV. It blocks CD80/
`CTLA-4 and CD86/CTLA-4 interactions with the same potency as
`ipilimumab (a) but the Treg depleting activity is improved compared
`to ipilimumab as demonstrated in a PBMC-NOG/SCID transfer
`model in vivo (b). An anti-mouse CTLA-4 antibody (clone 5-B07,
`mouse IgG2a) with similar functional effects in ligand blocking and
`Treg depletion (data not shown) was selected for the generation of
`the murine surrogate virus VVGM-αCTLA4 .
`
`Local, intratumoral admi-
`nistration of VVGM-αCTLA4
`induced complete tumor
`regression in contralateral
`uninjected CT26 tumors
`in 7/9 mice demonstrating the induction of systemic antitumor responses, or the abscopal effect. There was no evidence
`of viral particle dissemination to uninjected tumors as assessed by plaque assay (data not shown).
`
`Tumor volume (mm3)
`
`05
`
`00
`
`60
`
`7/9
`
`40
`20
`Days post treatment
`
`0
`
`20
`40
`Days post treatment
`
`500
`
`0
`
`60
`
`Tumor volume (mm3)
`
`Tumor volume (mm3)
`
`05
`
`60
`
`00
`
`1/10
`
`40
`20
`Days post treatment
`
`0
`
`20
`40
`Days post treatment
`
`MC38 tumor-bearing WT and Batf3-/- C57BL/6
`mice received i.t. injections of VVGM-αCTLA4 or
`PBS. Graphs show tumor volume (left and
`center panels) and mouse survival (right panel).
`Vertical lines indicate end of the treatment.
`
`0
`
`0
`
`40
`30
`20
`10
`Days post treatment start
`
`50
`
`0
`
`0
`
`40
`20
`Days post treatment start
`
`60
`
`Tumor volume (mm3)
`
`4-E03
`2-C06
`ipilimumab
`
`Antitumoral activity of Copenhagen oVV encoding an anti-mouse CTLA-4 and murine GM-CSF (VVGM-αCTLA4) was
`assessed in different syngeneic tumor models after 3 i.t. administrations of 107 PFU. Results demonstrate broad and
`potent antitumor activity of VVGM-αCTLA4 in immune inflamed and immune excluded models.
`
`0.01 0.1
`1
`mAb conc. (nM)
`
`10 100
`
`CD8+ T cells
`
`1500
`
`Figure 8. Intratumoral injection of VVGM-αCTLA4 induces systemic antitumor immunity (abscopal effect)
`PBS
`PBS
`VVGM αCTLA4
`VVGM αCTLA4
`i.t. treated tumor
`contralateral tumor
`i.t. treated tumor
`contralateral tumor
`
`1500
`
`1500
`
`1500
`
`1000
`
`1000
`
`1000
`
`1000
`
`Figure 12. I.t. VVGM-αCTLA4 expands effector CD8+ T cells and reduces Treg and exhausted CD8+ T cells
`Control
`Tumor
`VVGMαCTLA4
`T2
`
`Treg
`
`T1
`
`treated
`
`100
`
`T3
`
`80
`
`60
`
`40
`
`untreated
`
`1.0
`
`0.5
`
`0.0
`0.0001
`
`RLU (luminescence; x105)
`
`150
`
`100
`
`50
`
`1
`0.01 0.1
`mAb conc. (nM)
`
`10
`
`100
`
`Treg cells
`
`*
`
`6.0
`
`4.0
`
`2.0
`
`0.0
`0.0001
`
`RLU (luminescence; x106)
`
`b)
`
`150
`
`100
`
`50
`
`CD4+
`UMAP1
`
`0
`
`Control VVGMαCTLA4
`
`Spleens (lower panels), injected
`and contralateral tumors (upper
`panels) were collected on day 10
`post treatment and stained with a
`high-dimensional panel designed
`to identify T cell populations.
`
`Vehicle
`αPD-1
`VVGM αCTLA4
`VVGM αCTLA4 + αPD-1
`
`*
`
`*
`
`80
`60
`40
`20
`Days post cell inoculation
`
`100
`
`100
`
`80
`
`60
`
`40
`
`20
`
`Probability of Survival
`
`0
`
`0
`
`treated
`
`0/10
`
`60
`40
`20
`Days post cell inoculation
`
`80
`
`1500
`
`Tumor volume (mm3)
`
`1000
`
`500
`
`0
`
`0
`
`Figure 4. Oncolytic Vaccinia virus Copenhagen strain
`Copenhagen oVV strain (Invir.IOTM platform, Transgene)
`- is deleted in J2R (TK locus) and I4L (RR locus) viral genes involved
`in nucleotide synthesis to restrict replication in proliferating cells
`- allows large DNA insertions with successful vectorization of various
`expression cassettes
`- has the best oncolytic activity among VACV strains
`- induces Immune Cell Death
`
`Vaccinia virus (VV)
`Copenhagen strain →
`Oncolytic virotherapy
`
`VV / TK -RR - / TG
`
`(1)
`
`TRANSGENES: Enzyme,
`TRANSGENES:
`cytokines, mAb,
`enzymes, cytokines, mAb
`ligands….
`
`Double deletion of genes to
`enhance tumor-selectivity
`
`Figure 5. oVV expressing 4-E03 and GM-CSF (VVGM-αhCTLA4, BT-001) or 5-B07 and mGM-CSF (VVGM-αCTLA4)
`a)
`b)
`Hepatocytes
`HepG2
`
`Figure 9. Combination of VVGM-αCTLA4 with anti-PD-1 is benificial
`a)
`b)
`αPD-1
`VVGM αCTLA4
`VVGM αCTLA4 + αPD-1
`
`1500
`
`1000
`
`500
`
`0
`
`0
`
`1500
`
`10/10
`
`60
`40
`20
`Days post cell inoculation
`
`80
`
`Tumor volume (mm3)
`
`1500
`
`Tumor volume (mm3)
`
`1000
`
`500
`
`0
`
`0
`
`6/10
`
`60
`40
`20
`Days post cell inoculation
`
`80
`
`1500
`
`5. Conclusions
`• BT-001 is a multifunctional oVV co-developed by Transgene and BioInvent that encodes a Treg-depleting αCTLA-4
`antibody as well as the cytokine GM-CSF.
`Intratumoral delivery of a Vaccinia-Virus encoded anti-CTLA4 antibody achieved tumor-restricted exposure and
`•
`Treg depletion.
`• The murine surrogate VVGM-αCTLA4 has demonstrated a robust antitumoral activity in several syngeneic tumor
`models. This antitumoral activity is CD8+ T cell-dependent and synergized with αPD-1 treatment to reject cold tumors.
`• A clinical study investigating i.t. VVGM-αhCTLA4 (BT-001) alone and in combination with αPD-1 in metastatic or ad-
`vanced solid tumors has been commenced (NCT04725331).
`
`6. References
`1. Sharma, P. and J.P. Allison, Science, 2015. 348(6230): p. 56-61.
`2. Postow, M.A., et al., N Engl J Med, 2015. 372(21): p. 2006-17.
`3. Soderlind, E., et al., Nat Biotechnol, 2000. 18(8): p. 852-6.
`
`4. Arce Vargas, F., et al., Cancer Cell, 2018.
`5. Kleinpeter, P., et al., Oncoimmunology, 2016. 5(10): p. e1220467.
`6. Fend, L., et al., Cancer Res, 2017. 77(15): p. 4146-4157.
`7. www.transgene.fr, www.bioinvent.se
`
`c) The synergizing effects of VVGM-αCTLA4 and
`αPD-1 were confirmed in the A20 model where a
`sub-optimal dose of VVGM-αCTLA4 (105 PFU)
`cured the majority of animals in combination with
`αPD-1.
`c)
`
`100
`
`VVGM αCTLA4 + αPD-1
`VVGM αCTLA4 + isotype
`control VV + αPD-1
`αPD-1
`Vehicle
`control VV
`
`0
`
`20
`40
`Days post cell inoculation
`
`60
`
`80
`
`60
`
`40
`
`20
`
`0
`
`Probability of Survival
`
`2/10
`
`20
`40
`60
`Days post cell inoculation
`
`80
`
`1000
`
`500
`
`0
`
`0
`
`Tumor volume (mm3)
`
`Tumor volume (mm3)
`
`contralateral
`
`105
`
`104
`
`103
`
`102
`
`101
`
`105
`
`104
`
`103
`
`102
`
`101
`
`Replication rate
`
`100
`
`100
`
`p7.5
`
`Light chain
`
`pSE/L
`
`GM-CSF
`
`p7.5
`
`Heavy chain
`
`4-E03 produced by BT-001-
`infected tumor cells compa-
`red to recombinant 4-E03
`
`TG6002
`BT-001
`VV W T
`VV W T
`BT-001
`TG6002
`a) Anti-CTLA-4 mAb and GM-CSF were vectorized in Copenhagen oVV. Thereby deleted J2R and I4L genes were replaced
`by heavy and light chain of anti-CTLA-4, respectively. The expression cassette encoding GM-CSF was also placed at the
`I4L locus. b) BT-001 selectively replicated in tumor cells and not in normal cells, similar to the clinically validated oVV
`TG6002 (Transgene).
`
`1000
`
`500
`
`0
`
`0
`
`0/10
`
`20
`40
`60
`Days post cell inoculation
`
`80
`
`a, b) In the ICB re-
`sistant B16 model,
`treatment with
`VVGM-αCTLA4
`significantly
`reduced tumor growth of the primary injected tumor. However, i.t.
`VVGM-αCTLA4 only induced a slight delay of uninjected tumor’s
`growth, which did not translate into animal survival. Combined
`treatment with i.t. VVGM-αCTLA4 and systemic αPD-1 significant-
`ly inhibited injected and uninjected tumor growth (b), resulting in
`~20 % of animals being cured in this “cold” cancer model.
`
`Replimune Limited Ex. 2010 - Page 1
`Transgene and Bioinvent International AB v. Replimune Limited
`PGR2022-00014 - U.S. Patent No. 10,947,513
`
`