IMMUNOTHERAPY STRATEGIES FOR COLORECTAL CANCER › content › download › ... · IMMUNOTHERAPY...
Transcript of IMMUNOTHERAPY STRATEGIES FOR COLORECTAL CANCER › content › download › ... · IMMUNOTHERAPY...
IMMUNOTHERAPY STRATEGIES FOR COLORECTAL CANCER
Guillem Argilés, MD
Clinical Investigator, Gastrointestinal Malignancies
Programme
Early Drug Development Program
Vall d’Hebron University Hospital
Vall d'Hebron Institute of Oncology (VHIO)
◆ Understand the molecular basis of cancer immunotherapy
◆ To gain insight on the immunobiology and different therapeutic strategies for the
different immune subtypes of colorectal cancer
◆ MSI mCRC
◆ MSS mCRC
LEARNING OBJECTIVES OF
THE MODULE
◆ Introduction to cancer immunotherapy
◆ Overview of immunologic characteristics of CRC
◆ (Immunobiology of mCRC)
◆ Therapeutic immune strategies in CRC in MSI and MSS tumours
◆ Immune checkpoint inhibitors
◆ Adoptive cell therapies
◆ Anti-tumour vaccines
◆ Other agents
MODULE OUTLINE
INTRODUCTION TO CANCER IMMUNOTHERAPY
Immunotherapy is directed to the immune system compartment, not to the tumour
INTRODUCTION TO
IMMUNOTHERAPEUTICS
Tumour Progression Immune surveillance
As tumours progress
on their tumourigenic
cascades, the ability of
the immune system to
control tumour growth
decreases
Cancer immunotherapy
seeks to induce the
recovery of this ability
Adapted from Zitvogel L, et al. Nat Rev Immunol. 2006 Oct;6(10):715-27. Reprinted by permission from Springer Nature, Nat Rev Immunol, Cancer despite
immunosurveillance: immunoselection and immunosubversion, Zitvogel L, et al. Copyright 2006
Pre-malignant
lesion
ImmunosurveillanceElimination
Advanced
oncogenesis
ImmunoselectionEquilibrium
Tumour
growth
ImmunosubversionEscape
Cell intrinsic Cell extrinsic (immune system)
ANTI-TUMOUR IMMUNITY
Cell mediated
The anticancer response is cell mediated and either adaptive
(T-cell mediated) or innate (natural killer cell-dependent)
Cell mediated
Reprinted by permission from Springer Nature, Nature Reviews Cancer, Cytokines in cancer pathogenesis and cancer therapy, Dranoff G. Copyright 2004
Innate immunity
(rapid response)
Adaptive immunity
(slow response)
NATURAL KILLERS
Natural killers are responsible for detecting and killing virally infected
cells and cancer cells that do not express MHC class I
Failure of MHC class 1 to bind to the Killing Inhibitory Receptor (KIR) leads to tumour cell lysis
Reprinted by permission from Springer Nature, Nature Reviews Cancer, NK cells and cancer: you can teach innate cells new tricks, Morvan MG, et al.
Vol. 16; 7-19. Copyright 2016
1. Release of cancer
cell antigens
2. Cancer antigen
presentation
3. Priming and
activation
5. Infiltration of T cells
into tumours
6. Recognition of cancer
cells by T cells
7. Killing of cancer cells
4. Trafficking of T cells to tumours
MECHANISM OF ACQUIRED
CELL-MEDIATED IMMUNE RESPONSE
Reprinted from Immunity, 39(1), Chen DS, Mellman I, Oncology Meets Immunology: The Cancer-Immunity Cycle, 1-10, Copyright 2013, with permission from Elsevier.
TUMOUR ANTIGEN SHEDDING AND
PROCESSING ON APC
1. Tumour antigens of dying cells are released to
tumour microenvironment
2. Antigen Presenting Cells (APC) process these
antigens and present them through MHC class 2
3. Some populations of APC can cross-present the
phagocyted antigens through MCH class 1
DAMPs Immature DCDying
tumour cell
Reprinted by permission from Springer Nature, Nature Reviews Immunology, Cross-
presentation in viral immunity and self-tolerance, Heath WR, et al. Copyright 2001
Republished with permission of Annual Reviews, from Annu Rev Immunol,
Kroemer G, et al. 31:51-72, copyright 2013, permission conveyed through
Copyright Clearance Center, Inc.
ANTIGEN PRESENTATION BY APC
Once activated through antigen capturing and incorporation to the MHC, type 2 APCs migrate
to the loco-regional lymph nodes following chemokine gradients. Once in the lymph node, they
interact with the different populations of lymphocytes looking for T cell receptor compatibilities
Sanchez-Paulete AR, et al. Antigen cross-presentation and T-cell cross-priming in cancer immunology and immunotherapy, Ann Oncol. 2017;28(suppl_12):xii44-xii55,
by permission of The European Society for Medical Oncology (ESMO).
PRIMING IMMUNE SYNAPSE
Once the dendritic cells find a T helper cell or cytotoxic T cell with a compatible T cell receptor to the presented
antigen, both cells interact through three levels of signalling:
Signal 1: the so called MHC-TCR binding that depends on TCR specificity for the antigen
Signal 2: the interaction of the different activating and inhibitory immune checkpoint receptors and ligands
Signal 3: consisting of the exchanging of cytokines between both cells. The outcome of this signalling
determines whether there will be an immune response against the presented antigen or not
Pacheco R, et al (October 10th 2012). Cells, Molecules and Mechanisms Involved in the Neuro-Immune Interaction, Cell Interaction, Sivakumar Gowder, IntechOpen,
DOI: 10.5772/48367. Available from: https://www.intechopen.com/books/cell-interaction/cells-molecules-and-mechanisms-involved-in-the-neuro-immune-interaction.
Reproduced under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0).
Activation
Dead
PRIMING IMMUNE SYNAPSE
SIGNAL 2
Signal 2: The agonistic
and inhibitory ligands and
ligand receptors involved
are referred to as immune
checkpoints. Immune
checkpoint inhibitors
directed against PD-1/
PD-L1, CTLA-4, LAG-3,
ICOS, CD-40, OX-40
TIGIT constitute the main
current therapeutic field in
cancer immunotherapy
Reprinted by permission from Springer Nature, Nature Reviews Immunology, Molecular mechanisms of T cell co-stimulation and co-inhibition, Chen L, et al. Copyright 2013
LYMPHOCYTE TRAFFICKING
◆ Activated antigen reactive lymphocytes
abandon the lymph node and enter into the
circulation looking for zones of inflammatory
endothelium
◆ Inflammatory endothelium expresses
proteins that allow activated lymphocytes to
cross the endothelial barrier and enter into
the tissues
◆ The activated lymphocyte subsequently
moves towards the inflammatory focus
through a cytokine gradient that will direct
the lymphocyte to the inflammatory stimuli,
in this case the tumour
◆ This process may be hampered in case of
aberrant tumour vasculature
Gong C, et al. Front Immunol 2014;5:57. Reproduced under the terms of the Creative Commons Attribution License (CC BY 3.0; https://creativecommons.org/licenses/by/3.0/)
Reprinted by permission from Springer Nature, Nature Medicine, 19:1423-1437. Microenvironmental regulation of tumor progression and metastasis, Quail D, et al. Copyright 2013
TUMOUR MICROENVIRONMENT
Within the tumour microenvironment, activated lymphocytes usually have to cope with a myriad of
mechanisms, such as PD-L1 overexpression in tumour cells, paucity of nutrients, low tissue pH and T-helper
cell differentiation into regulatory T cells (Treg), which lead to a decreased anti-tumour immune response
IMMUNOLOGIC CHARACTERISTICS OF COLORECTAL CANCER (CRC)
With regard to treatment with immune checkpoint inhibitors, it is
important to discriminate between:
◆ Microsatellite Instable (MSI) CRC
◆ Microsatellite Stable (MSS) CRC
IMMUNOBIOLOGY OF CRC
◆ Microsatellites are repeats of 1 to 10 nucleotides with a variable
length (5–50 repeats), such as:
◆ A A A A A A A→ 7 poly-A microsatellite
◆ GT GT GT GT→ 4 poly-GT microsatellite
◆ ACGTCC-ACGTCC-ACGTCC→ 3 poly ACGTCC
◆ Microsatellites are localised in coding or non-coding DNA regions
MICROSATELLITES
Microsatellites cause DNA
polymerase slips in the replicative
fork causing DNA mismatches and
ultimately protein mutations
BIOLOGIC IMPORTANCE OF
MICROSATELLITES
◆ Zones of accumulations
of mutations
◆ Commonly frame shifts
Cox M, et al. Molecular Biology: Principles and Practice; 2nd Edition. WH Freeman and Company, New York
Evolution endowed eukaryotic cells with a dedicated system to repair mismatches in DNA
DNA MISMATCH REPAIR SYSTEM
Protein Heterodimer Repair function
MutSα MSH2 ♦ MSH6 Binds base-base
mismatches and insertion-
deletion mismatches
MutSβ MSH2 ♦ MSH3 Binds insertion-deletion
mismatches
MutLα MLH1 ♦ PMS2 Early step before excision
Vilar E, et al. Nat Rev Clin Oncol 2010;7(3):153-62;
Cox M, et al. Molecular Biology: Principles and Practice; 2nd Edition. WH Freeman and Company, New York
Human mismatch repair proteinsBinding of mismatch
proofreading proteins
DNA scanning detects nick in
new DNA strand (before
sealing by DNA ligase)
Repair DNA synthesisUp to 1000 bp can be removed
Strand removal
hMLH1
Error in newly
synthesized strand
hMSH2
◆ Alterations on microsatellite repairing system can be determined in two ways:
◆ By PCR against a group of microsatellites looking for mutations
◆ Given the name of microsatellite instable tumours
◆ By immunohistochemistry looking for the expression of the different mismatch
repair genes
◆ Given the name of mismatch repair deficient tumours
◆ The overlap between both techniques is 90%
◆ The frequency of MSI-H status in metastatic CRC is 5%
MSI-HIGH VS. MISMATCH REPAIR
DEFICIENT TUMOURS
THEORETICAL BASIS
For improved clinical benefit of treatment with checkpoint inhibitors in
case of microsatellite instability
◆ The loss of mismatch repair system causes mutations and ultimately cancer
◆ Colon cancers harbouring a defective mismatch repair system have a higher
number of neoantigens
◆ This increased mutational load increases the chance of a checkpoint inhibitor-
mediated anti-tumour response
TUMOUR RESPONSE TO ANTI-PD1
AXIS AGENTS ACCORDING TO
MICROSATELLITE STATUS
MSI CRC → Higher mutation tumour burden → higher lymphocyte infiltration → Responses to anti-PD-1 inhibitor
MSS CRC → LOW mutation tumour burden → low lymphocyte infiltration → No responses to anti-PD-1 inhibitor
Adapted from The Cancer Genome Atlas Network. Nature 2012;487:330–7. Reproduced under the terms of the Creative Commons Attribution-Non-Commercial-Share
Alike licence (http://creativecommons.org/licenses/by-nc-sa/3.0/).
From N Engl J Med, Le D, et al., Immune-therapy strategies for Colorectal Cancer, 372(26):2509–20. Copyright © 2015, Massachusetts Medical Society.
Reprinted with permission from Massachusetts Medical Society.
OS in cohorts with CRC
WATERFALL PLOT OF MS-STABLE
AND -INSTABLE PATIENTS TREATED WITH
PEMBROLIZUMAB FOR MCRC
200 400 600
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
20
30
40
50
60
70
80
90
100 MMR-proficient CRC
MMR-deficient CRC
MMR-deficient non-CRC
Days
%C
han
ge f
rom
Baseli
ne S
LD
MS-instable patients presented
dramatic responses in contrast
with MS-stable patients
MMR, mismatch repair
Le TD, et al. J Clin Oncol 2015;33(18 Suppl). Abstract LBA100. Presented at ASCO 2015. Courtesy of Dr Le.
Le TD, et al. N Engl J Med 2015;372(26):2509–20
CLINICAL TRIAL DESIGN:
KEYNOTE-164
A Phase 2, single-arm study of pembrolizumab in pretreated patients to
address significant patient unmet needs
Patient population
◆ Locally advanced or
metastatic MSI
CRC
◆ Patients must have
received at least
2 prior treatments
◆ Stage IV disease
Pembrolizumab 200 mg q3w IV
Complete
response
Discontinuation
permitted
Confirmed
progressive
disease
Discontinue
Partial response
or stable disease
Treat for up to 2 years
or until progression or
unacceptable toxicity
ORR by RECIST 1.1
Target enrollment: 60
Assessments: Radiological assessments using RECIST 1.1 and irRC every 9 weeks
Primary Endpoint
Le TD, et al. Ann Oncol 2018;29(suppl_5). abstract O-021 Presented at ESMO 20th World Congress on Gastrointestinal Cancer 20–23 June 2018,
Barcelona, Spain
MSI mCRC patients presented 32% RR regardless of the number of prior therapy lines
BEST PERCENTAGE CHANGE
FROM BASELINE IN TARGET
LESION SIZE*
*RECIST v1.1 per IRC. Data cutoff: September 12, 2017.
Le TD, et al. Ann Oncol 2018;29(suppl_5). abstract O-021 Presented at ESMO 20th World Congress on Gastrointestinal Cancer 20–23 June 2018, Barcelona,
Spain. Courtesy of Dr Le.
76% of patients alive at
1-year follow-up
OVERALL SURVIVAL
Data cutoff: September 12, 2017.
Le DT, et al. Ann Oncol 2018;29(suppl_5) abstract O-021. Presented at ESMO 20th World Congress on Gastrointestinal Cancer 20–23 June 2018, Barcelona,
Spain. Courtesy of Dr Le.
PROGRESSION-FREE SURVIVAL*
*RECIST v1.1 per IRC. Data cutoff: September 12, 2017.
Le D, et al. Ann Oncol 2018;29(suppl_5) abstract O-021. Presented at ESMO 20th World Congress on Gastrointestinal Cancer 20–23 June 2018, Barcelona,
Spain. Courtesy of Dr Le.
…and 41% progression-free
at 1-year follow-up
CHECKMATE 142
Nivolomab provided similar benefits (RR 31%, 48%
free from progressing and 74% alive 12 months
after treatment initiation)
Reprinted from Lancet Oncol, 18(9), Overman MJ, et al. Nivolumab in patients with metastatic DNA mismatch repair-deficient or microsatellite instability-high colorectal
cancer (CheckMate 142): an open-label, multicentre, phase 2 study :1182–91. Copyright 2017, with permission from Elsevier.
Patients, n
(%)
dMMR/MSI-H per Local
Laboratory (N=74)
dMMR/MSI-H per Central
Laboratory (n=53)
Investigator BICR Investigator BICR
ORR, n (%)
95% CI
23 (31.1)
20.8, 42.9
24 (32.4)
22.0, 44.3
19 (35.8)
23.1, 50.2
19 (35.8)
23.1, 50.2
Primary endpoint: ORR per investigator assessment
Secondary endpoint: ORR per blinded independent central review (BICR)
Other endpoints: PFS, OS, biomarkers, safety and tolerability
Histologically
confirmed metastatic/
recurrent CRC
dMMR/MSI-H per
local laboratory
≥1 prior line of
therapy
Nivolumab
3 mg/kg Q2W
Nivolumab
3 mg/kg Q2W
Patients Stage 1 Stage 2
Nivolumab
3 mg/kg +
ipilimumab
1 mg/kg Q3W
for 4 doses
Then
nivolumab
3 mg/kg Q2W
Stage 1
Nivolumab
3 mg/kg +
ipilimumab
1 mg/kg Q3W
for 4 doses
Then
nivolumab
3 mg/kg Q2W
Stage 2
Investigator PFS Investigator OS
Median (95% CI), months 9.6 (4.3, NR) NR (18.0, NE)
12-month rate (95% CI), % 50.4 (38.1, 61.4) 73.4 (61.5, 82.1)
Change in tumour burden
OVERALL SURVIVAL BY BEST
OVERALL RESPONSE
Very interestingly, none of the
responders died during the
trial follow-up, which
demonstrates the immunising
effect of the treatment
CR + PR
SD
PD
0 3 6 9 39
Months
Ove
rall
surv
ival
(%)
100
90
80
70
60
50
40
30
20
10
0
No. at Risk
CR + PR 25 25 25 25 25 25 19 14 14 13 9 5 1 0
SD 23 23 22 20 17 16 9 2 2 2 2 1 0 0
PD 22 15 12 10 9 7 4 1 1 1 1 0 0 0
NE = not estimable; NR = not reached.
Overman MJ, et al. J Clin Oncol 36, 2018 (suppl. 4S; abstr. 554). Presented at ASCO GI 2018. Courtesy of Dr Overman.
CR + PR
n=25
SD
n=23
PD
n=22
Median OS
(95% CI), months
NR
(NE)
NR
(14.3, NE)
10.3
(3.0, NE)
12 15 18 21 24 27 30 33 36
PEMBRO AND NIVO APPROVED IN
US BUT NOT IN EU FOR MSI MCRC
These ground-breaking results merited the
approval of pembrolizumab and nivolumab for
the MSI population in the refractory setting.
The EMA, however, halted the approval until
randomised data are analysed
Available at: https://www.fda.gov/Drugs/InformationOnDrugs/ApprovedDrugs/ucm617370.htm.
Accessed January 2019
Available at: https://www.fda.gov/Drugs/InformationOnDrugs/ApprovedDrugs/ucm560040.htm.
Accessed January 2019
Available at: https://www.ema.europa.eu/documents/withdrawal-letter/withdrawal-letter-
opdivo_en.pdf Accessed January 2019
◆ Grant the approval in Europe
◆ Move the treatment to earlier lines
◆ Maximise the percentage of responding patients
FUTURE GOALS OF
IMMUNOTHERAPY IN ADVANCED
MSI CRC
KEYNOTE 177 IS THE FIRST
RANDOMISED TRIAL EXPLORING
ANTI-PD1 AXIS BLOCKADE IN MSI-H MCRC
A Phase 3 study of pembrolizumab monotherapy vs. standard
chemotherapy in 1L MSI CRC
Keynote 177 will make it clear whether anti-PD1 inhibitors should become common frontline
practice in MSI mCRC and, if positive, could support final EMA approval for Europe
Patient population
◆ Locally advanced or
unresectable or
metastatic CRC
◆ MSIRandomisation 1:1
Pembrolizumab
200 mg q3w
Up to 2 years
PD
mFOLFOX6 + bevacizumab:
Bevacizumab 5 mg/kg
+
Oxaliplatin 85 mg/m2 IV
+
Leucovorin 400 mg/m2 IV
+
5-FU 400 mg/m2 IV bolus on
Day 1, then 1,200 mg/m2/day
for 2-day continuous infusion;
repeat every 2 weeks until
progressive disease
PD
PFS
OS
Target enrollment: 270
Primary Endpoint
Off study
Crossover
PD, disease progression
Diaz Jr LA, et al. J Clin Oncol 2016;34(Suppl 4S): Abstract TPS789
CHECKMATE 142: DOES CLINICAL
BENEFIT IMPROVE IF NIVOLUMAB
AND IPILUMUMAB ARE COMBINED?
Overman MJ, et al. J Clin Oncol 2018;36(8):773–9. Courtesy of Dr Overman.
3 512
26
31
38
51.3
31
3.4
0
20
40
60
80
100
Nivolumab + ipilimumabN=119
NivolumabN=74
Pat
ient
s (%
)
CR
PR
SD
PD
Unknown
ORR (95% CI):
31% (20.8, 42.9)ORR (95% CI):
55% (45.2, 63.8)
◆ DCR was 80% (95% CI: 71.5, 86.6) with combination therapy and 69% (57.1, 79.2)
with monotherapy
◆ Combination therapy provided a numerically higher ORR, including CRs, and DCR
relative to monotherapy during a similar follow-up period
INVESTIGATOR-ASSESSED
RESPONSE AND DISEASE CONTROL RATE
COMBINATION VS. SINGLE AGENT
71. Overman MJ, et al. Lancet Oncol 2017;18:1182–91; 2. Overman MJ, et al. J Clin Oncol 2018;36(8):773–9.
21
PROGRESSION-FREE AND
OVERALL SURVIVAL
COMBINATION VS. SINGLE AGENT
• With similar follow-up, combination therapy provided improved PFS and OS relative to monotherapya,e,f,2
11
Nivolumab +
ipilimumaba,dNivolumab1,e,f
9-month rate (95% CI), % 87 (80.0, 92.2) 78 (66.2, 85.7)
12-month rate (95% CI), % 85 (77.0, 90.2) 73 (61.5, 82.1)
Nivolumab +
ipilimumaba,bNivolumab1,e,f
9-month rate (95% CI), % 76 (67.0, 82.7) 54 (41.5, 64.5)
12-month rate (95% CI), % 71 (61.4, 78.7) 50 (38.1, 61.4)
aMedian follow-up 13.4 months (range, 9–25). bMedian PFS not reached (95% CI, not estimable). cPFS per investigator assessment. dMedian OS not reached (95% CI,
18.0, not estimable). eMedian follow-up 13.4 months (range, 10–32). fCheckMate-142 monotherapy and combination therapy cohorts not randomised or designed for a
formal comparison.
1. Overman MJ, et al. Lancet Oncol 2017;18:1182–1191; 2. Overman MJ, et al. J Clin Oncol 36(8), 2018: 773–9. Reprinted with permission. © 2018 American Society of
Clinical Oncology. All rights reserved. Courtesy of Dr Overman.
◆ Anti-PD-1 axis monoclonal antibodies produce dramatic and
long-lasting response in MSI CRC and are the new standard of care
in US
◆ Same will happen in the EU once randomised data are available
◆ Current efforts aim to move the treatment to earlier lines and
increase the % of patients who respond to treatment
◆ Anti-CTLA4 + anti-PD-1 agents will soon become new standard
of care
CONCLUSIONS
MSS CRC
◆ Immune checkpoint inhibitors and specifically-designed
immunotherapy combinations failed to improve outcome
BACKGROUND
Pembrolizumab1 Nivolumab + Ipilimumab2 Cobimetinib + Atezolizumab3
Nivolumab
1 mg/kg +
ipilimumab
3 mg/kg
(n=10)
Nivolumab
3 mg/kg +
ipilimumab
1 mg/kg
(n=10)
ORR, n (%) 1 (10) 0
Median PFS,
mo (95% CI)
2.28
(0.62, 4.40)
1.31
(0.89, 1.71)
Median OS,
mo (95% CI)
11.53
(0.62, NE)
3.73
(1.22, 5.62)
1. From N Engl J Med, Le D, et al., Immune-therapy strategies for Colorectal Cancer, 372(26):2509–20. Copyright © (notice year) Massachusetts Medical Society. Reprinted
with permission from Massachusetts Medical Society;
2. Overman MJ, et al. J Clin Oncol 34, 2016 (suppl; abstr 3501. Presented a ASCO 2016;
3. Bendell J, et al. Ann Oncol 2018;29(suppl_5) ESMO WGI 2018; abstract LBA-004. Courtesy of Dr Bendell.
Time (months)
Ove
rall
surv
ival
(%
)
100
80
60
40
20
0
0 3 6 129 15 18 21
Atezolizumab + cobimetinib
Atezolizumab
Regorafenib
Tumour mutational burden and ORR correlation with anti–PD-1 axis
agents in 27 tumour types1
CONSIDERING THE MUTATIONAL
BURDEN OF MSS CRC RESPONSE RATE TO
TREATMENT WITH ANTI-PD-1 AXIS AGENTS IS LOW
1. From N Engl J Med, Yarchoan M, et al. Tumor Mutational Burden and Response Rate to PD-1 Inhibition., 377(25 ):2500-2501. Copyright © 2017. Massachusetts
Medical Society. Reprinted with permission from Massachusetts Medical Society.
2. Reprinted from Cancer Discovery, 2018; 8(6):730–49, Grasso CS, et al., Genetic Mechanisms of Immune Evasion in Colorectal Cancer, with permission from AACR.
THIS COULD BE EXPLAINED BY ITS
IMMUNOSUPPRESSIVE BIOLOGY, WHICH IS
CAUSED BY GENOMIC ALTERATIONS
The described stepwise model of CRC carcinogenesis starts with WNT path activation
1. The Cancer Genome Atlas Network, Nature 2012490, pages 61–
70; under the terms of the Creative Commons Attribution-Non-
Commercial-Share Alike licence
(http://creativecommons.org/licenses/by-nc-sa/3.0/),
2. Reprinted by permission from Springer Nature, Nat Rev Cancer,
Walther A, et al. Genetic prognostic and predictive markers in
colorectal cancer. Copyright 2009
3. Fearon ER, et al. Cell 1990;61(5):759–67
WNT PATHWAY ACTIVATION LEADS
TO DENDRITIC CELL EXCLUSION FROM
TUMOUR MICROENVIRONMENT Melanoma1 Localised CRC2 mCRC3
1. Reprinted by permission from Springer Nature, Nature, Melanoma-intrinsic β-catenin signalling prevents anti-tumour immunity, Spranger S, et al. Copyright 2015;
2. Reprinted from Cancer Discovery, 2018; 8(6):730–49, Grasso CS, et al., Genetic Mechanisms of Immune Evasion in Colorectal Cancer, with permission from AACR;
3. Rodón J, et al. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer
Res 2018;78(13 Suppl):Abstract nr CT175.
MSS mCRC HAS AN
IMMUNOSUPPRESSANT BIOLOGY
The second step is the activation of the MAPK-PI3K pathway
1. The Cancer Genome Atlas Network, Nature 2012490, pages 61–
70; under the terms of the Creative Commons Attribution-Non-
Commercial-Share Alike licence
(http://creativecommons.org/licenses/by-nc-sa/3.0/),
2. Reprinted by permission from Springer Nature, Nat Rev Cancer,
Walther A, et al. Genetic prognostic and predictive markers in
colorectal cancer. Copyright 2009
3. Fearon ER, et al. Cell 1990;61(5):759–67
MAPK-PI3K PATHWAY ACTIVATION
IMPAIRS ANTIGEN PRESENTATION BY
DECREASING MHC CLASS 1 EXPRESSIONCD8+ T cells
per tumour cell
Proportion of IFNγ-producing
CD8+ T cells
60
0
20
40
80
MEKiNo drug
IFN
γ+(%
)
Proportion of CD8+ cells
with low PD-1 expression
40
30
0
10
20
50
MEKiNo drugP
D-1
lo/C
D8+
cel
ls (
%)
0.04
0.03
0
0.01
0.02
MEKiNo drug
MHC class 1
1. Reprinted from Immunity, 44(3), Ebert PJR, et al. MAP Kinase Inhibition Promotes T Cell and Anti-tumor Activity in Combination with PD-L1 Checkpoint Blockade, 609–21;
Copyright 2016, with permission from Elsevier; 2. Bendell J, et al. J Clin Oncol, 34(15_suppl) 2016: 3502. Presented at ASCO 2016. Permission granted by author, Dr Bendell.
MSS mCRC HAS AN
IMMUNOSUPPRESSANT BIOLOGY
The third step is the intracellular loss of function of the TGF-beta pathway
1. The Cancer Genome Atlas Network, Nature 2012490, pages 61–
70; under the terms of the Creative Commons Attribution-Non-
Commercial-Share Alike licence
(http://creativecommons.org/licenses/by-nc-sa/3.0/),
2. Reprinted by permission from Springer Nature, Nat Rev Cancer,
Walther A, et al. Genetic prognostic and predictive markers in
colorectal cancer. Copyright 2009
3. Fearon ER, et al. Cell 1990;61(5):759–67
TGFβ blockade in combination with PD-1 axis inhibition can
reverse this process and induce anti-tumour immune responses
TGFβ ACTIVATION
TGFβ activation in the tumour microenvironment is a primary mechanism of
immune evasion that promotes T-cell exclusion and blocks acquisition of the
TH1-effector phenotype
Bladder Cancer
CRC
1. Reprinted by permission from Springer Nature: Nature, TGFβ drives immune evasion in genetically reconstituted colon cancer metastasis,Tauriello DVF, et al. Copyright 2018;
2. Reprinted by permission from Springer Nature: Nature, TGFβ attenuates tumour response to PD-L1 blockade by contributing to exclusion of T cells, Mariathasan S, et al. Copyright 2018
MSS mCRC HAS AN
IMMUNOSUPPRESSANT BIOLOGY
The fourth step is the P53 pathway inactivation
1. The Cancer Genome Atlas Network, Nature 2012490, pages 61–
70; under the terms of the Creative Commons Attribution-Non-
Commercial-Share Alike licence
(http://creativecommons.org/licenses/by-nc-sa/3.0/),
2. Reprinted by permission from Springer Nature, Nat Rev Cancer,
Walther A, et al. Genetic prognostic and predictive markers in
colorectal cancer. Copyright 2009
3. Fearon ER, et al. Cell 1990;61(5):759–67
• Wild type p53 in tumours unleashes the
CTL response via inhibition of PD-L1 and
enhances their effectiveness by
upregulating Fas/APO-1 and MHC I
• Given that p53 is mutated in
approximately 50% of human cancers and
also impacts the immunoreactivity of
cancer cells, a significant number of
patients can be affected by the impaired
CTL response that results from non-
functional p53
• An attenuated CTL response due to p53
mutations could decrease response rates
to immunotherapeutic drugs, leading to
poor patient prognoses
TP-53 PATHWAY INACTIVATION
CONTRIBUTE TO IMMUNE ESCAPE
Braun MW, et al. Transl Cancer Res. 2016 Dec; 5(6): 692–697; Translational cancer research by NANCY INTERNATIONAL LTD SUBSIDIARY AME
PUBLISHING COMPANY. Reproduced with permission of NANCY INTERNATIONAL LTD SUBSIDIARY AME PUBLISHING COMPANY in the format Use in an
e-coursepack via Copyright Clearance Center.
THE COMPLEXITY OF
IMMUNOTHERAPEUTICS
Contrary to normal cancer therapeutics, cancer immunotherapy is based on a bi-compartment
model where tumour and immune system interact constantly
Biologic compartments implied
in normal cancer therapeutics
Biologic compartments implied
in immunotherapeutics1
One-compartment model Two-compartment model
1. Adapted from Immunity, 39(1), Chen DS & Mellman I. Oncology Meets Immunology: The Cancer-Immunity Cycle, 1-10, Copyright 2013, with permission from Elsevier.
TRANSCRIPTOMIC
CLASSIFICATION OF CRC
• Gene signatures can capture the bi-compartmentation of cancer immunotherapy
• The most validated gene-signature for mCRC is the Consensus Molecular Subtype
Classifier that divides CRC in four molecular subtypes.
Guinney J, et al. Nat Med 2015;21(11):1350–6
CMS1
MSI Immune
CMS2
Canonical
CMS3
Metabolic
CMS4
Mesenchymal
14% 37% 13% 23%
MSI, CIMP high
HypermutationSCNA high
Mixed MSI status
SCNA low, CIMP lowSCN high
BRAF mutations KRAS mutations
Immune infiltration
and activation
WNT and MYC
activationMetabolic deregulation
Stromal infiltration
TGF beta activation
Angiogenesis
Worse survival
after relapse
Better survival
after relapse
Worse relapse-free
and overall survival
TRANSCRIPTOMIC
CLASSIFICATION OF CRC
CMS1 behaves like
MSI tumours at an
RNA level
CMS2 depends on
WNT activation
CMS3 harbours MAPK
pathway activation and
metabolic reprogramming
CMS4 is characterised by
stromal TGFβ activation
and diffuse inflammation
Guinney J, et al. Nat Med 2015;21(11):1350–6
CMS1
MSI Immune
CMS2
Canonical
CMS3
Metabolic
CMS4
Mesenchymal
14% 37% 13% 23%
MSI, CIMP high
HypermutationSCNA high
Mixed MSI status
SCNA low, CIMP lowSCN high
BRAF mutations KRAS mutations
Immune infiltration
and activation
WNT and MYC
activationMetabolic deregulation
Stromal infiltration
TGFβ activation
Angiogenesis
Worse survival
after relapse
Better survival
after relapse
Worse relapse-free
and overall survival
IMMUNE LANDSCAPE OF MSS
mCRC ACCORDING TO CMS SUBTYPE
MSI-LIKE WNT pathway activation MAPK / PI3Kactivation
TGFβ
CMS1 behave like MSI tumours. CMS2 and CMS3 are cold tumours, while
CMS4 harbour a non-anti-tumoral inflammatory microenvironment
Adapted from Clinical Cancer Research, © 2016, 22(16), 4057–66, Becht E, et al. Immune and Stromal Classification of Colorectal Cancer Is Associated with
Molecular Subtypes and Relevant for Precision Immunotherapy, with permission from AACR.
T cells chemotaxis
and activation
T-cell-specific inhibitionMyeloid-cells chemotactism
Angiogenesis
Immunosuppression
Complement
Tertiary lymphoid structures
Major Histocompatibility Complex I
(MHCI)
CMS1: MSI-LIKE SUBGROUP WITH
HEALTHY TILS
Strategy:
PD1 blockade
Immune
CMS1
MSI
PD1 blockade
responsive
FDA approval:
Nivolumab
Pembrolizumab5%
10%
Anti-PD-1 therapies could achieve the same results in CMS1, as seen in MSI tumours.
Motricolor CT3 trial (atezolizumab + bevacizumab) is exploring this hypothesis
Adapted from Clinical Cancer Research, © 2016, 22(16), 4057–66, Becht E, et al. Immune and Stromal Classification of Colorectal Cancer Is Associated with
Molecular Subtypes and Relevant for Precision Immunotherapy, with permission from AACR.
CMS1
T cells chemotaxis
and activation
T-cell-specific inhibition
Myeloid-cells chemotactism
Angiogenesis
Immunosuppression
Complement
Tertiary lymphoid structures
MHCI
CMS2 WNT ACTIVATION
Block Wnt pathway
Allow dendritic cell infiltration
Foster T-cell anti-tumour
reactivity
NCT02675946 (Wnt pathway inhibition):
CGX1321 Porcupine inh + Pembrolizumab
for mCRC
CMS2 is characterised by Wnt pathway activation and by having a cold tumour microenvironment.
WNT inhibition may foster anti-tumoral immune response in these tumours
CMS2
Wnt pathway activation
T cells chemotaxis
and activation
T-cell-specific inhibition
Myeloid-cells chemotactism
Angiogenesis
Immunosuppression
Complement
Tertiary lymphoid structures
MHCI
Adapted from Clinical Cancer Research, © 2016, 22(16), 4057–66, Becht E, et al. Immune and Stromal Classification of Colorectal Cancer Is Associated with
Molecular Subtypes and Relevant for Precision Immunotherapy, with permission from AACR.
Wnt pathway inhibitors are, however, often not effective due to downstream APC
mutations
◆ LRP 5/6 Inh
◆ Anti-RSPONDIN MoAb
◆ Porcupine inhibitor
WNT PATHWAY INHIBITORS
Clevers & Nusse. Cell 149, June 8, 2012Mc Donald Clevers & Nusse. Cell 149, June 8, 20121. Reprinted from Developmental Cell, 17, MacDonald BT, et al. Wnt/β-Catenin Signaling: Components, Mechanisms, and Diseases, 9–26; Copyright 2009, with permission
from Elsevier; 2. Reprinted from Cell, 149(6), Clevers H, et al. Wnt/β-Catenin Signaling and Disease 192–205, Copyright 2012, with permission from Elsevier.
Seshagiri S, Nature 2012; Koo BK, Nature 2012
Giannakis M, Nat Genetics 2014
CMS3: DEPENDS ON MAPK
PATHWAY ACTIVATION
MAPK/PI3K
activation1
Rational approach should be MEK + PD-1 axis inhibition2
1. Adapted from Clinical Cancer Research, ©2016, 22(16), 4057–66, Becht E, et al. Immune and Stromal Classification of Colorectal Cancer Is Associated with
Molecular Subtypes and Relevant for Precision Immunotherapy, with permission from AACR;
2. Bendell J, ESMO-WGI 2018. Courtesy of Dr Bendell.
CMS3
Unresectable
mCRC patients
Received at least
2 regimens in
metastatic setting
(not including
maintenance)
2:1:1
N=360
ARM ACobimetinib + atezolizumab
n=180
ARM BAtezolizumab
n=90
ARM CRegorafenib
n=90
Treatment to
continue until
loss of clinical
benefit
COTEZO TRIAL
OVERALL SURVIVAL
Atezo + cobi(n=183)
Atezo(n=90)
Rego(n=90)
Median OS, mo(95% CI)
8.9 (7.00, 10.61)
7.1 (6.05, 10.05)
8.5 (6.41, 10.71)
HR vs. rego(95% CI)
1.00 (0.73, 1.38)
1.19 (0.83, 1.71)
N/A
P-value 0.9871 0.3360a N/A
12-mo OS, % 38.5 27.2 36.6
The MEK inh./ PD-L1 inh-combination did not achieve a better outcome
The COTEZO trial was
negative for the overall
population.
Outcomes of the CMS3
population need to be
determined to validate the
hypothesis
Bendell J, et al. Ann Oncol 2018;29 Suppl 5:v123, abstract LBA-004. Presented at ESMO WGI 2018. Courtesy of Dr Bendell.
Future directions for CMS 3
NCT02876224: Cobimetinib + atezolizumab + bevacizumab for 2nd line MSS mCRC
NCT02060188: Cobimetinib + Ipilimumab + Nivolumab for MSS mCRC
NCT03271047: Binimetinib + Nivolumab +/- Ipilimumab for RAS mut mCRC
CMS3 MAPK PATHWAY ACTIVATION
1. Adapted from Clinical Cancer Research, © 2016, 22(16), 4057–66, Becht E, et al. Immune and Stromal Classification of Colorectal Cancer Is Associated with
Molecular Subtypes and Relevant for Precision Immunotherapy, with permission from AACR; 2. Segal N, et al. J Clin Oncol 36, 2018 (suppl; abstr 3540).
Presented at ASCO 2018. Courtesy of Dr Segal.
MAPK/PI3K
activation1
CMS3
Natural killer cell therapy
Monalizumab
+
Durvalumab
MSS mCRC 3/39 PRs
Natural Killers are the cells
that kill those cells that do
not express MHC class 1
Change in Tumor Size in MSS-CRC Patients in the Expansion Phase2
CMS4: TGFβ ACTIVATION
NCT03436563: M7824 (bi-specific TGFβ/PD-L1 MoAb) for CMS 4 mCRC2
Block TGFβ pathway in CMS 4 mCRC
Change the inflammatory microenvironment
to a Th1 immune response
Foster T-cell infiltration
Increased sensitivity to
Immune checkpoint inhibitors
1. Adapted from Clinical Cancer Research, © 2016, 22(16), 4057–66, Becht E, et al. Immune and Stromal Classification of Colorectal Cancer Is Associated with
Molecular Subtypes and Relevant for Precision Immunotherapy, with permission from AACR;
2. Clinicaltrials.gov. https://clinicaltrials.gov/ct2/show/NCT03436563. Accessed January 2019
CMS4
TGFβ1
CRC IMMUNE CLASSIFICATION AT
TRANSCRIPTOMIC LEVEL
Hypermutation WNT Pathway activation KRAS m TGFβ
TREATMENTS FOR ALL SUBGROUPS
Adapted from Clinical Cancer Research, © 2016, 22(16), 4057–66, Becht E, et al. Immune and Stromal Classification of Colorectal Cancer Is Associated with
Molecular Subtypes and Relevant for Precision Immunotherapy, with permission from AACR.
T cells chemotaxis
and activation
T-cell-specific inhibition
Myeloid-cells chemotactism
Angiogenesis
Immunosuppression
Complement
Tertiary lymphoid structures
MHCI
CHEMOTHERAPY OR
RADIOTHERAPY–IMMUNE CHECKPOINT
INHIBITOR COMBINATIONS Target Therapy Phase Trial Design Trial ID
Anti-PDL1
Atezolizumab (engineered IgG1,
no ADCC)
I Solid tumours NCT01375842
Ib Solid tumours (+ bevacizumab ± FOLFOX) NCT01633970
II mCRC (+ bevacizumab + fluoropyrimidine) NCT02291289
MEDI4736(modified IgG1,
no ADCC)II mCRC NCT02227667
Anti-PD-1
Nivolumab(IgG4)
I/II mCRC (± ipilimumab) (CheckMate 142) NCT02060188
I/II Solid tumours (+ INCB24360) NCT02327078
I/II Solid tumours (+ chemotherapy) NCT02423954
I/II Solid tumours (+ varlilumab) NCT02335918
Pembrolizumab(IgG4, humanised)
I Solid tumours (+ aflibercept) NCT02298959
I/II GI cancers (+mFOLFOX6) NCT02268825
I/II WT mCRC (+ cetuximab) NCT02318901
II mCRC (+ radiotherapy or ablation) NCT02437071
II mCRC (+ chemotherapy) NCT02375672
II mCRC (+ azacitidine and/or romidepsin) NCT02512172
II MSI-positive/-negative CRC NCT01876511
*Recruiting studies
Clinicaltrials-gov
◆ Binds simultaneously with one arm to CD3 on T cells and with two arms to CEA on tumour cells
◆ Flexible 2-to-1 format enables high-avidity binding and selective killing of CEA-overexpressing tumour cells
◆ Longer half-life compared with other TCB formats
◆ Silent Fc results in reduced risk of FcγR-related cytokine release/IRRs
CEA-TCB structure1,2
BI-SPECIFIC T-CELL
ENGAGING ANTIBODIES
Direct T-cell activation skipping antigen
recognition upon binding to CEA protein
◆ Simultaneous binding of TCB to tumour (CEA) and T cells (CD3)
◆ Killing of tumour cells independent of pre-existing immunity
◆ T-cell proliferation at site of activation
Fab, fragment antigen-binding region; IRR, infusion-related reaction.
Adapted from: 1. Bacac M, et al. Clin Cancer Res. 2016; and 2. Bacac M, et al. OncoImmunology. 2016;
Argilés G, et al. CEA-TCB in CRC. ESMO WGI 2017. Courtesy of Dr Argilés.
Source: Juno official website
Study 2: CEA-TCB + atezolizumab (n=11, 80 and 160 mg of CEA-TCB)
p
*p
*
p
-50
0
50
100*p
WithdrawalProgressionOngoingFirst new lesion
a
160 mg80 mg
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
CEA-TCB AT DOSES ≥60 MG +
ATEZOLIZUMAB IN 3L+ PATIENTS
WITH MSS MCRC
Bes
t ch
ang
e in
tar
get
lesi
on
s
fro
m b
asel
ine,
%
Weeks after treatment start
*p
Withdrawal
Progression
Ongoing
First new lesion
160 mg
80 mg
Data reported by investigators, cut-off: March 3, 2017. aRadiological signs of tumour inflammation seen at ≥ 60 mg (safety data cut-off is ≥ 40 mg).
Argilés G, et al. Ann Oncol 2017;28(suppl_3); abstract LBA-004. Presented at ESMO WGI 2017. Courtesy of Dr Argilés.
A PHASE I STUDY OF
ENADENOTUCIREV
An oncolytic Ad11/Ad3 chimeric group B adenovirus, in combination with
nivolumab in tumours of epithelial origin
Harb W, et al. J Clin Oncol 35, 2017 (suppl; abstr TPS3115). Presented at ASCO 2017. By permission from PsiOxus and courtesy of Dr Harb...
A PHASE I STUDY OF
ENADENOTUCIREV
An oncolytic Ad11/Ad3 chimeric group B adenovirus, in combination with
nivolumab in tumours of epithelial origin
Harb W, et al. J Clin Oncol 35, 2017 (suppl; abstr TPS3115). Presented at ASCO 2017. By permission from PsiOxus and courtesy of Dr Harb..
Personalised immune therapies are a good option in expert hands
PERSONALISED IMMUNE
THERAPIES
Personalised peptide vaccines1 Adoptive T-cell therapy2
1. Reprinted from Clin Cancer Res, 2016, 22(4):807-12, Desrichard A, et al., Cancer Neoantigens and Applications for Immunotherapy, with permission from
AACR; 2. Reprinted by permission from Springer Nature, Nature Reviews Cancer, Adoptive cell transfer: a clinical path to effective cancer immunotherapy,
Rosenberg SA, et al. Copyright 2008.
◆ Personalised immune therapies can be good options for patients not
responding to other available therapies
◆ However, their complexity makes these options only available in
highly experienced centres, since their success depends on the
experience of the team
PERSONALISED IMMUNE
THERAPIES
EXAMPLES OF VACCINES FOR
IMMUNOSTIMULATION IN CRC
Therapy Antigen Enhancer Phase Study population Trial ID
Anti-tumour
vaccines
Autologous
tumour cells
BCG
II Adjuvant CRC PMID: 8445413
III Adjuvant CRC PMID: 15755632
III Adjuvant Stage II CRC NCT02448173
Newcastle
Disease Virus II Liver ressected CRC PMID: 18488223
Dendritic cell
vaccines
CEA Dendritic cells I Adjuvant Stage III CRC NCT01890213
MUC 1 Dendritic cells IILiver or lung ressected
CRCNCT00103142
Pubmed - Clinicaltrials.gov
Clinicaltrials.gov
EXAMPLES OF ADOPTIVE CELL
THERAPY FOR IMMUNOSTIMULATION
IN CRC
Therapy Target Enhancer Phase Trial Design Trial ID
TILSAutologous
tumour cells
IL-2
PembrolizumabII GI tumours NCT01174121
CAR T cells
CEA ITumours expressing
CEANCT02349724
CEA Yttirum 90 ITumours expressing
CEANCT02416466
EGFR I/IITumours expressing
EGFRNCT01869166
Cytokine-
induced-
killers cells
Autologous
tumour cellsII
Adjuvant CRC in
combination with
XELOX
NCT01929499
◆ Immune checkpoint inhibitors and MEK-based combinations have
failed in MSS mCRC, potentially reflecting the lack of implementation
of tumour biology knowledge into study design
◆ A new generation of promising compounds and combinations
specifically designed for this disease are currently entering clinical
development
◆ Deeper understanding of tumour biology is crucial to ensure their
success and further implementation in clinical practice
TAKE HOME MESSAGES FOR
MSS MCRC
THANK YOU!