Immunosynthen:STING-Agonist ADC Platform

November 16, 2020

Legal Disclaimer

This presentation contains “forward-looking” statements within the meaning of federal securities laws. These forward-looking statements are notstatements of historical facts and are based on management’s beliefs and assumptions and on information currently available to management.Forward-looking statements include information concerning Mersana Therapeutics, Inc.’s (the “Company’s”) business strategy and the design,progression and timing of its clinical trials and expectations regarding future clinical results based on data achieved to date.

Forward-looking statements generally can be identified by terms such as “aims,” “anticipates,” “believes,” “contemplates,” “continues,” “could,”“estimates,” “expects,” “goal,” “intends,” “may,” “on track,” “plans,” “possible,” “potential,” “predicts,” “projects,” “seeks,” “should,” “target,” “will,”“would” or similar expressions and the negatives of those terms. Forward-looking statements represent management’s beliefs and assumptions onlyas of the date of this presentation. The Company’s operations involve risks and uncertainties, many of which are outside its control, and any one ofwhich, or combination of which, could materially affect its results of operations and whether the forward-looking statements ultimately prove to becorrect. Factors that may materially affect the Company’s results of operations and whether these forward-looking statements prove to be correctinclude, among other things, that preclinical testing or early clinical results may not be predictive of the results or success of ongoing or later clinicaltrials, and that the development and testing of the Company’s product candidates will take longer and/or cost more than planned, as well as thoselisted in the Company’s Annual Report on Form 10-K filed with the Securities and Exchange Commission (“SEC”) on February 28, 2020, theCompany’s Quarterly Report on Form 10-Q filed with the SEC on May 8, 2020 and subsequent SEC filings. In addition, while we expect that theCOVID-19 pandemic might adversely affect the Company’s preclinical and clinical development efforts, business operations and financial results, theextent of the impact on the Company’s operations and the value of and market for the Company’s common stock will depend on future developmentsthat are highly uncertain and cannot be predicted with confidence at this time, such as the ultimate duration of the pandemic, travel restrictions,quarantines, physical distancing and business closure requirements in the U.S. and in other countries, and the effectiveness of actions taken globallyto contain and treat the disease. Except as required by law, the Company assumes no obligation to update these forward-looking statements publicly,or to update the reasons actual results could differ materially from those anticipated in the forward-looking statements, even if new informationbecomes available in the future.

Copies of the Company’s Annual Report on Form 10-K and our other SEC filings are available by visiting EDGAR on the SEC website athttp://www.sec.gov.

2

Agenda

3

• Anna Protopapas, President & Chief Executive Officer– Opening remarks & introductions

• Tim Lowinger, PhD, Chief Science & Technology Officer– Therapeutic rationale for a STING-agonist ADC– Development & optimization of Immunosynthen platform

• Marc Damelin, PhD, Executive Director & Head of Biology– Preclinical data supporting Immunosynthen ADC pipeline

• Tim Lowinger, PhD, Chief Science & Technology Officer– Potential of the Immunosynthen ADC pipeline and next steps

• Q&A

We Intend to Answer Key Questions Today

4

• Why activate innate immunity with a STING-Agonist ADC?

• How is Immunosynthen different than other approaches to stimulate innate immunity (SITC 2020, Abstract #620)?

• How did Mersana build and optimize the Immunosynthen platform?

• How deep is the pipeline of Immunosynthen ADCs, and which indications might be addressable?

• When will we reach the clinic?

Targeted Stimulation of Innate Immunity has the Potential to Deliver Breakthroughs

Nature Reviews Cancer 4, 11–22 (2004) 5

Innate Immunity• Includes STING• “Step on the gas”

Adaptive Immunity• Includes CTLA4, PD1/PD-L1• “Release the brakes”

• The immunotherapy revolution has focused on adaptive immunity and serves only a fraction of patients

• Innate immune stimulation could address unmet medical needs in– Checkpoint refractory tumors– Checkpoint relapsed tumors– Tumor types where checkpoints

have minimal activity

STING Pathway Activation has Shown Intriguing Signs of Activity

6

• In clinical trials, intratumoral injection of STING agonists has induced:– Tumor shrinkage (top panel)

– Immune cell infiltrates (bottom panel)

• In preclinical studies, compelling genetic and pharmacological evidence for anti-tumor potential of STING activation

• STING activation leads to a potent Type I interferon response without general inflammation

Top panel: Harrington et al. (2018) ESMO 2018 Congress. LBA15.Bottom panel: Meric-Bernstam et al. (2018) SITC 33rd Annual Meeting. P10763.

Immune Cell

Infiltrates

RECIST Clinical

Response

Selected preclinical references:Woo et al. (2014) Immunity; Corrales et al. (2015) Cell Reports; Barber (2015) Nature Reviews Immunology.

Hypothesis: An ADC Approach Could Address Administration Issues, Systemic Tolerability, and Activity

7

• Systemic administration with targeted delivery to the tumor

• Improved anti-tumor activity compared to free agonist

• Improved tolerability compared to free agonist

Intratumoral STING Agonist

Systemic Free STING Agonist

STING-Agonist ADC

Tumor with STING-Mediated Innate Immune Activation

Tumor, no immune activation

Systemic immune activation

8

• STING pathway is silenced in tumor cells

• The STING mechanism of action is limited to immune cells

What We Thought

• Tumor cells activate STING in the presence of immune cues– Field has been misled by

standard monoculture conditions

• Our studies employed:– CRISPR-mediated STING

knockout cancer cells– Fc mutant ADCs– Co-cultures of tumor cells

and immune cells– Conditioned media from

immune cells

What We Now Know

• Immunosynthen ADCs enable target-dependent activation of STING in tumor-resident immune cells and tumor cells– Active internalization into

both cell types in the tumor (via FcγR and via tumor antigen)

• Immunosynthen ADCs avoid STING activation in undesired cell types (e.g., T and B cells)

The ADC Modality Is Ideally Suited

Immunosynthen ADCs Activate STING in Immune and Tumor CellsPresented at SITC 2020

9

1

2

Hit the tumor cells

Hit the tumor-resident immune cells

Target-Dependent Internalization

Target-Dependent FcγR-Mediated Internalization

Tumor cell

Myeloid cell

No internalization into immune cells No stimulation of immune cells

Tumor

Periphery

STING Activated in Tumor Cell

STING Activated in Myeloid Cell

StimulatedT Cells

No internalization into undesired immune cells (T, B)

STING: The One-Two PunchPresented at SITC 2020

Free Payload

0.1 1 10 1000

10000

20000

30000

Payload (nM)

Luci

fera

se A

ctiv

ity

Punch One: Fc-Mediated Delivery to Immune Cells is Target Dependent

• Specific tumor antigen binding provides high local concentration, which promotes FcγR binding on tumor-resident immune cells

• FcγR binding results in internalization into tumor-resident immune cells and STING activation

• STING activation by ADC is ~40-100x more potent than free agonist

~40x

1

TargetedAntigen and FcγR

binding

Fc mutantNo FcγR binding

X

ControlNo antigen binding

XAntibody variants

10

FcγR

Assay plate

Tumor Antigen

Human immune cellSTING pathway activation

(Luciferase reporter)

11

Killing

Growth

Tumor Cells STING Knockout Tumor Cells

Tumor Cell Killing in Immune Cell (PBMC) Co-Culture

Can

cer c

ell c

onflu

ency

at 8

0 ho

urs

(nor

m. t

o t=

0)• Surprising observation that Fc-

mutant ADCs are active in co-culture – which indicates tumor intrinsic STING activation

• Previously believed that STING pathway is silenced in tumor cells

• Co-cultures with STING knockout cancer cells demonstrate the direct contribution of tumor intrinsic STING

TargetedAntigen and FcγR

binding

Fc mutantNo FcγR binding

X

ControlNo antigen binding

X

Antibody Variants

-20

-15

-10

-5

0

5

10

15

20

-20

-15

-10

-5

0

5

10

15

20

2

Payload dose Payload dose5 nM 25 nM 5 nM 25 nM

Punch Two: STING Activation in the Tumor Cell Also Contributes to Efficacy

The One-Two Punch: In Vivo Data Consistent and Differentiated

12

TargetedAntigen and FcγR

binding

Fc mutantNo FcγR binding

X

Control ADCNo antigen binding

X

Vehicle

• Significant anti-tumor activity is maintained by the Fc-mutant ADC, which cannot internalize into the immune cells

• Demonstrates the contribution of tumor cell STING to anti-tumor activity

• Demonstrates the contribution of immune cell STING to activity

• The One-Two Punch

0 7 14 21 28 35 42 49 560

200

400

600

800

1000

Days on study

Mean

Tum

or V

olume

±SEM

(mm

3 )

1

0 7 14 21 28 35 42 49 56 630

300

600

900

1200

1500

Days on study

Mean

Tum

or V

olume

±SEM

(mm

3 )

1

Tumor Antigen A / Tumor Model 1

Tumor Antigen B / Tumor Model 2

2.7 mg/kgsingle dose IV

3 mg/kgsingle dose IV

Holistic Approach to Building a STING-Agonist ADC

13

Design novel STING agonist for

use in ADC

Optimize linker and scaffold for

the novel agonist

Validate multiple targets for

Immunosynthen

Select optimal antibody for each

target

Build pipeline of Development Candidates

Mersana’s ADC Expertise Drives Platform Optimization

ADC Optimization via Modular Approach

Proprietary STING agonist

Drug load per scaffold

Charge balance

Aqueous solubilityBioconjugationAntibody

0 7 14 21 28 35 42 49 56 63 70 77

0

200

400

600

800

1000

Days on study

Mea

n Tu

mor

Vol

ume±S

EM (m

m3 )

1

VehiclC3575 C3525 C3526 C3532

0 5 10 1510

100

1000

10000

100000

Time (Days)C

once

ntra

tion

[ng/

ml]

Mea

n +/

- SD

Total AntibodyConjugated Drug

High Stability and Extended Exposure of Immunosynthen ADC

Optimizing Activity With the Same Antibody and Same STING agonist(single, equivalent IV dose for all ADCs)

14

Total AntibodyConjugated STING-Agonist

Space between curves reflects higher molecular weight of antibody relative to STING-agonist payload

Parallel curves reflect stability of Immunosynthen ADC

Defined Success Criteria Have Been Achieved

15

• Antigen-dependent targeted delivery to the tumor

• Sustained tumor regressions• Consistent results across tumor models and mouse strains

• Proof of mechanism• Induction of STING pathway cytokines in the tumor• Induction of STING genes in the cancer cell• Immune memory

• Well-tolerated; minimal systemic inflammation; favorable NHP

• Compatible with many antigens • Enables a portfolio to address many clinical indications

Comprehensive Approach to Target Selection

16

Image based on DOI: 10.5772/intechopen.72648

• Innate immune activation with STING enables many target classes

• Immune cells

• Tumor cell antigens

• Tumor-associated antigens

• Potential broad target space encompasses multiple indications of high unmet medical need

Single, Low Dose of Immunosynthen ADC Dramatically Outperforms Benchmark

17

0 7 14 21 28 35 42 49 56 630

250

500

750

1000Tumor Growth Inhibition

Days on study

Mea

n Tu

mor

Vol

ume

(mm

3 )±SE

M

*Benchmark IV agonist (5 mg/kg)

Control ADC (1 mg/kg)Control ADC (3 mg/kg)

Targeted ADC (1 mg/kg)#

Targeted ADC (3 mg/kg)#

Unconjugated Antibody (3 mg/kg)

• All groups dosed IV• ADC Doses by mAb [mg/kg]

Vehicle

Legend

*Benchmark described in J.M. Ramanjulu et al. (2018) Nature#1 mg/kg by mAb = 0.03 mg/kg by STING-agonist payload

Single dose administration

Target-Dependent Immune Cell Infiltration into Tumor

18

Vehicle

12 hrs.

72 hrs.

Targeted ADC Control ADC

CD45 Immunohistochemistry

Brown staining indicates immune cells (CD45+)

Punch One: Immunosynthen ADC Induces STING Pathway Cytokines in the Tumor-Resident Mouse Immune Cells In Vivo

19

Vehicle

Targeted ADC

Control ADC

Mouse CXCL10 Mouse IL-6Mouse IFNβ

• Mouse cytokines in tumor microenvironment measured by qPCR

• ADC targets the human tumor cells

0

5

10

15

20

25

0

0.4

0.8

1.2

1.6

0

2

4

6

mR

NA

Expr

essi

on

(rela

tive

to G

APD

H)

Mouse cytokines induced in the tumor microenvironment

1

Punch Two: Immunosynthen ADC Induces STING Pathway in Human Tumor Cells In Vivo

20

0

2000

4000

6000

0

200

400

600

800

1000

0

300

600

900

1200

Human CXCL10 Human IL6Human IFNβ

Vehicle

Targeted ADC

Control ADC

• Human cytokines in tumor microenvironment measured by Nanostring

Nor

mal

ized

mR

NA

Cou

ntHuman cytokines induced in the tumor cells

2

Down for the Count: Immunosynthen ADC Triggers Tumor-Specific Immunological Memory

21

0 7 14 21 28 350

200

400

600

800

1000

Days on Study

Mea

n Tu

mor

Vol

ume±S

EM (m

m3 ) Vehicle

Control ADC 1 mg/kgTargeted ADC 0.9 mg/kg

1

• Tumor free mice re-implanted with targeted tumor on one flank (blue) and a non targeted tumor on the other flank (red).

• Untreated age matched mice also implanted as a control (black line).

Tumor Growth Inhibition Study Tumor Rechallenge Study (Dual Flank)

0 7 14 21 280

50

100

150

Targeted-tumor

Days on Study

Mea

n Tu

mor

Vol

ume±S

EM (m

m3 )

Untreated Control Mouse (age matched)Previously Treated with Targeted ADC

0 7 14 21 280

500

1000

1500

2000

Non-targeted tumor

Days on Study

Mea

n Tu

mor

Vol

ume±S

EM (m

m3 )

Untreated Control Mouse (age matched)Previously Treated with Targeted ADC

6/9 tumor-freeanimals

single IV dose

Dramatically Lower Systemic Cytokine Levels After Immunosynthen ADC Compared to Benchmark IV Agonist

22

CXCL10 IL-6 CXCL1

pg/m

L

0 20 40 60 800

1000

2000

3000

4000

Hours

0 20 40 60 800

5000

10000

15000

20000

25000

Hours0 20 40 60 80

0

2000

4000

6000

Hours0 20 40 60 80

0

2000

4000

6000

Hours

0 20 40 60 800

5000

10000

15000

Hours

0 20 40 60 800

50

100

150

200

Hours

MIP-1a (CCL3)TNF-α

pg/m

L

Serum cytokines measured with Luminex assay

Vehicle control

Targeted ADC (0.09 mg/kg STING agonist)

Control ADC (0.09 mg/kg STING agonist)

Benchmark IV agonist (5 mg/kg ~ maximum tolerated dose)

Systemic Free STING Agonist

STING-Agonist ADC

IFNγ

Tumor with STING-Mediated Innate Immune Activation

Tumor, no immune activation

Systemic immune activation

• NHP studies with Immunosynthen ADCs based on 5 antibodies• 4 fully cross-reactive antibodies for 4 therapeutic targets• 1 non-reactive antibody to assess platform safety profile

• Intravenous administration; single-dose and repeat-dose studies

• Pharmacokinetics• High exposure; dose dependent; overall profile similar to non-STING ADCs• ADC highly stable in circulation; minimal free payload in plasma

• Serum Cytokines• Transient, modest elevation of 5 cytokines out of 24 tested; similar to results in mouse

• No adverse changes in hematology or clinical chemistry

• No adverse findings in histopathology to date

Immunosynthen ADCs Well-Tolerated in Non-Human Primate Studies After Repeat IV Dosing

23

0 7 14 21 28 35 42 49 56 63

0

200

400

600

800

1000

Days on Study

Mea

n Tu

mor

Vol

ume±S

EM (m

m3 )

0 7 14 21 28 35 42 49 56 63 70 77 84

0

500

1000

1500

2000

Days on Study

Mea

n Tu

mor

Vol

ume±S

EM (m

m3 )

Immunosynthen ADCs Active Against Diverse Tumor Antigens and Tumor-Associated Antigens in Multiple Models After Single, Low IV Dose

24

Tumor Antigen A Tumor Antigen B

0 7 14 21 28 35 42 49 56

0

400

800

1200

1600

Days on study

Mea

n Tu

mor

Vol

ume±S

EM (m

m3 )

Tumor Antigen C Tumor Antigen D

0 7 14 21 28 35 42 490

200

400

600

800

1000

1200

1400

Days on Study

Mea

n Tu

mor

Vol

ume±S

EM (m

m3 )

0 7 14 21 28 35

0

200

400

600

800

1000

1200

Days on study

Mea

n Tu

mor

Vol

ume±S

EM (m

m3 )

0 7 14 21 28 35

0

200

400

600

800

1000

1200

1400

Days on Study

Mea

n Tu

mor

Vol

ume±S

EM (m

m3 )

Tumor-Associated Antigen A

Tumor-Associated Antigen B

3.0 mg/kg

2.8 mg/kg

3.0 mg/kg 2.2 mg/kg

2.5 mg/kg 0.88 mg/kg

VehicleControl ADCTargeted ADC

Legend

Opportunity to Build a Robust Pipeline to Treat a Broad Range of Cancers

25

• Bladder Cancer• Breast cancer• Colorectal cancer• Endometrial Cancer• Gastric cancer• Head & Neck Squamous Carcinoma• Lung cancer• Melanoma• Ovarian cancer• Pancreatic cancer

Summary of DataXMT-2056: First Immunosynthen Development Candidate

26

0.01 0.1 1 10 100 10000.0

0.5

1.0

1.5

2.0

2.5

3.0

Payload [nM]O

.D. 4

50nm

0.001 0.01 0.1 1 10 1000

10000

20000

30000

Payload [nM]

RLU

EC50 = 0.08 nM EC50 = 0.11 nM

0 7 14 21 28 35 42 49 56 63 70 77

0

200

400

600

800

1000

Days on study

Mea

n Tu

mor

Vol

ume±S

EM (m

m3 )

Fc-mediated uptake and THP1 cell activationIRF3 Reporter (THP1)

Tumor cells with PBMCsCXCL10 ELISA

In vivo Activity0.96 mg/kg antibody / 0.033 mg/kg STING

Single dose IV

NHP ResultsSingle-dose and repeat-dose studies at 9 mg/kg antibodyIntravenous administration

• No clinical signs, no mortality• High exposure, high ADC stability in circulation• Transient elevation of 5 cytokines out of 24 tested• No adverse changes in clinical pathology• No adverse findings in histopathology

Freepayload

> 1,000x

XMT-2056 Shows Excellent PK after Repeat IV Dosing and a Wide Therapeutic Index

27

Plasma Concentrations in Non-Human Primate vs. Mouse(Conjugated STING agonist)

Exposure of XMT-2056 at well-tolerated dose in non-human primate is ~10-fold higher than the exposure required for sustained tumor regression in mouse

• High stability as indicated by parallel curves of antibody and conjugated drug

• Space between curves reflects higher molecular weight of antibody relative to STING-agonist payload

• Comparable PK profiles after 1st and 2nd dose

1 8 15 22 291

10

100

1000

10000

100000

1000000

Time (Days)

Plas

ma

Conc

entra

tion

(ng/

mL)

1 2 3 4 5 6 7 810

100

1000

10000

100000

Time (Days)

Plas

ma

Conc

entra

tion

(ng/

mL)

Quantitation Limits

Plasma Concentrations in Non-Human Primate(Total Antibody and Conjugated STING agonist )

Non-Human Primate,9 mg/kg

Mouse,1 mg/kg

The Immunosynthen Platform is Already Delivering Multiple Product Candidates

28

Program Target Target Validation DiscoveryIND-Enabling Studies

P1 Dose Escalation

XMT-2056 Tumor Antigen A

To Be Named Tumor Antigen B

To Be Named Tumor Antigen C

To Be Named Tumor Antigen D

To Be Named Tumor-Associated Antigen A

To Be Named Tumor-Associated Antigen B

DevelopmentCandidate Submit IND

Summary

29

• Immunosynthen ADCs have the potential to address the limitations of current approaches to activate innate immunity / STING

• Immunosynthen ADCs deliver One-Two knockout punch from STING activation in tumor cells and tumor-resident immune cells (SITC 2020)

• We have optimized the platform using our ADC expertise

• We are building a deep pipeline of Immunosynthen ADCs with a broad range of clinical indications and potential for value-creating partnerships

• XMT-2056 has been selected as the first Immunosynthen ADC with initiation of Phase I Dose Escalation expected in Q1 2022

Q&A