NNEXEXT-TGGENENERERAATITOIONNOONNCCOOLYLYTITCIC IMIMMMUUNNOOTHTHERERAAPPYY SITC Investor Event Presentation November 8th 2019

Safe harbor 2 Any statements contained herein that are not statements of historical facts may be deemed to be forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, including statements regarding our expectations about our use of cash, our advancement of our clinical trials, our goals to develop and commercialize our product candidates, our plans to establish our own in-house manufacturing capabilities, our proposed scientific presentations, and other statements identified by words such as “could,” “expects,” “intends,” “may,” “plans,” “potential,” “should,” “will,” “would,” or similar expressions and the negatives of those terms. Forward-looking statements are not promises or guarantees of future performance, and are subject to a variety of risks and uncertainties, many of which are beyond our control, and which could cause actual results to differ materially from those contemplated in such forward-looking statements. These factors include risks related to our limited operating history, our ability to generate positive clinical trial results for our product candidates, the costs and timing of establishing, equipping, and operating our planned in-house manufacturing facility, the timing and scope of regulatory approvals, changes in laws and regulations to which we are subject, competitive pressures, our ability to identify additional product candidates, and other risks set forth under the heading “Risk Factors” of our Annual Report on Form 10-K for the year ended March 31, 2019. Our actual results could differ materially from the results described in or implied by such forward-looking statements. Forward-looking statements speak only as of the date hereof, and, except as required by law, we undertake no obligation to update or revise these forward-looking statements.

Schedule of events 3 Introduction to the company and platform overview Clinical data & conclusions Q&A

mission statement 4 Replimune aims to transform the lives of cancer patients through the development of highly effective but low toxicity therapies that provide durable benefit across a broad range of tumor types

5 Founded by the former BioVex management team that developed T-VEC Proprietary ‘Immulytic’ oncolytic immuno-gene therapy platform Intended to maximally activate the immune system against a patient’s cancer Establish Replimune’s products as the second cornerstone of immuno-oncology RP1 – in multiple clinical trials predominantly targeting skin cancers Skin cancer clinical trials underway & planned include Ongoing Phase 2 trial including non-melanoma skin cancers in combination with nivolumab Ongoing registration directed Phase 2 clinical trial in CSCC in combination with cemiplimab Clinical trial of single agent RP1 in CSCC in organ transplant recipients (announced in October) Clinical trial in anti-PD1 refractory melanoma patients (announced today) RP2 & RP3 - intended to treat less immuno-responsive tumors, outside skin cancers Ongoing Phase 1 clinical trial of RP2 alone & combined with nivolumab RP3 intended to enter the clinic in 2020 Commercial scale manufacturing capability expected to be operational H1 2020

Summary of data & conclusions to be presented 6 RP1 alone & combined with nivolumab is well tolerated MOA of RP1 alone & in combination with nivolumab was confirmed Strong support was provided for Replimune’s programs in melanoma & CSCC Includes two of the three anti-PD1 refractory cutaneous melanoma patients enrolled in Phase 1 & the first ipi/nivo refractory Phase 2 patient with follow up scans treated in combination with nivolumab responding (see separate melanoma-specific deck also released today) CSCC data previously disclosed & updated with further improvement today (see separate CSCC-specific deck also released today) Biomarker data also indicated robust virus replication & immune activation As a direct result of this clinical data, Replimune’s clinical programs are being expanded in CSCC & melanoma New clinical trial in CSCC in solid organ transplant patients with RP1 alone New clinical trial in anti-PD1 refractory melanoma (RP1 combined with anti-PD1) Phase 2 data with RP1 combined with nivolumab in bladder cancer & MSI-H tumors is pending

Oncolytic immuno-gene therapy 7 The use of viruses that selectively replicate in & kill tumors to treat cancer Directly kills tumors Highly inflammatory Activates both innate and adaptive immunity Releases the full array of tumor antigens into an inflamed environment Systemically activates the immune system against the tumor & neo-antigens released Armed with additional genes to increase efficacy Single agent T-VEC is FDA approved for the treatment of advanced melanoma

Replimune’s therapies are designed to be best in class 8 Designed to maximize local tumor destruction Clinically beneficial in its own right Provides abundant release of tumor antigens to provide the systemic vaccination effect Designed to maximize the immunogenicity of cell death Increases the potency of the immune response generated to tumor antigens as they are released Deliver potent immune stimulating proteins to the tumor to further amplify the immune response Focus on proteins which act at the site and time of immune response initiation In particular, such proteins often require ongoing antigen presentation (i.e. provision of Signal 1) Virus mediated immunogenic killing provides a potent source of Signal 1, which is not usually present on an ongoing basis in cancer patients Thereby intended to convert immunologically ‘cold’ tumors to immunologically ‘hot’ Intended to maximize synergy with PD1 blockade

Most patients don’t respond to immune checkpoint blockade 9 ORR for tumor types where immune checkpoint inhibitors are FDA approved Source: re-drawn from FDA presentation at AAADV Workshop, 5/8/2019. Data as of 11/2018.

Intended to become the fourth cornerstone of cancer treatment ( Immune activation: Currently missing from Overcome defenses to the host immune response - dominated the cancer paradigm treatment Oncolytic immuno­ gene therapy by anti-PDl Immuno­ therapy Surgery/RT Chemo­ therapy Targeted therapies Traditional c_ytotoxics/ADCs TKis, PARPs, KRAS,various antibodies, ett \ Cell-based therapy l. _

A pipeline designed to address specific needs in cancer therapy 11 RP1 Need 1: Improve outcomes in immune sensitive cancers Tumors that already have a level of responsiveness to anti-PD1 blockade In all cases there still remains a substantial unmet need Intend to establish a franchise for the treatment of skin cancers, with initial clinical data-driven focus on Cutaneous squamous cell carcinoma (CSCC) Melanoma Potential to expand to additional immune sensitive tumor types as clinical data warrants Bladder cancer and MSI-H cancer

A pipeline designed to address specific needs in cancer therapy 12 Need 2: Extend utility of immunotherapy in more immune resistant cancers RP2 RP3 Tumors where anti-PD1 therapy is less effective, e.g. head & neck cancer, TNBC Tumor types where anti-PD1 is not effective, e.g. colorectal cancer Target tumor types for late stage development to be data-driven from earlier clinical trials

Replimune’s Immulytic platform 13 1. A potent underlying HSV-1 strain anti-CTLA-4 CTLA-4, CD40L & RP1 RP2 RP3 cold tumors There is great diversity among clinical HSV strains 2. Increased tumor killing & spread Armed with GM-CSF & a potent fusogenic protein (GALV-GP R-) 3. Delivery of potent immune stimulatory proteins Focus on pathways where systemic engagement is sub-optimal 29 new clinical strains were tested & the most effective selected & engineered for oncolytic use Provides a substantial increase in direct & immunogenic tumor killing potency Intended for immune responsive tumor types, with focus on skin cancers Further armed with anti-CTLA-4 & immune-costimulatory pathway activators Our product candidates were then armed with two to four genes to augment tumor killing & the potency of immune activation Intended for less & non-immune responsive tumor types RP1 additionally expressing anti-4-1BBL Immunologically RP1 additionally expressing Immunologically cold tumors Enhanced potency oncolytic immunotherapy backbone Immunologically warm/hot tumors

Replimune has a robust & steadily advancing pipeline 14 Expresses GALV-GP R-& GM-CSF Phase 2 underway in 4 tumor types + nivolumab Melanoma + nivolumab Enrolling Non-melanoma skin cancers + nivolumab Enrolling Bladder cancer + nivolumab Enrolling MSI high cancer + nivolumab Enrolling RP1+cemiplimab vs.cemiplimab in CSCC Enrolling RP1+antiPD1 in anti-PD1 refractory melanoma Initiates 2020 RP1alone & with nivolumab Enrollment complete REGISTRATION DIRECTED 1. Randomized, controlled clinical RP1 trial in 240 patients with underway 2. Study in organ transplant CSCC RP1aloneinorgantransplantptswithCSCC InitiatesQ12020* recipients with CSCC to Q1 2020 3. Study in PD1 refractory melanoma to initiate 2020 initiate Additionally expresses an anti-CTLA-4 antibody. Phase 1 underway RP2 alone & with anti-PD1 Enrolling RP2 Expresses GALV-GP R-, anti-CTLA-4, CD40L & 4-1BBL. Phase 1initiates 2020 RP3 GXP activities underway *Trial expansion or additional trial expected to be needed for filing PRE-CLINICALPHASE 1PHASE 2

15 Phase 1/2 clinical trial of RP1 alone & in combination with nivolumab

Key objectives of the RP1 Phase 1 part of the clinical trial 16 Demonstrate safety of RP1, alone & combined with nivolumab Via both superficial and deep injection routes Determine the recommended dose of RP1 for further development alone & combined with nivolumab Confirm the MOA of RP1 alone & in combination with nivolumab Provide support for Replimune’s programs in the target tumor types for RP1 Initially skin cancers – melanoma & CSCC With developing Phase 2 data, potentially bladder cancer & MSI-H tumors

Clinical trial design 17 First in human two stage trial of RP1 alone and in combination with nivolumab Phase 1 part 1: Dose escalation of RP1 alone given up to 5 injections into a single tumor in three dose level cohorts each by either direct or imaging guided injection Dose level 1: 1x104 pfu/ml, 1x105 pfu/ml, 1x106 pfu/ml x 3 Dose level 2: 1x105 pfu/ml, 1x106 pfu/ml, 1x107 pfu/ml x 3 Dose level 3: 1x106 pfu/ml, 1x107 pfu/ml, 1x108 pfu/ml x 3 Samples taken for biodistribution & shedding CT scan at baseline & 30 days post last dose Phase 1 part 2: RP1 given up to 8 times into multiple tumors at the recommended dose in combination with nivolumab starting from the second RP1 dose for up to two years Serial biopsies taken for biomarker analysis & as clinically indicated Data being presented today Phase 2: Four cohorts of 30 patients each with melanoma, non-melanoma skin cancers, bladder cancer and MSI-H tumors – recruitment ongoing

Phase 1 key inclusion criteria (dose rising & combined with nivolumab) 18 Key inclusion criteria Advanced or metastatic non-neurological solid tumors, which have progressed on standard therapy or cannot tolerate standard therapy, or for which there is no standard therapy preferred to enrollment in a clinical trial At least one measurable and injectable (including use of image-guided injection) tumor of > 1 cm in longest diameter Adequate hematologic, hepatic and renal function ECOG performance status 0 – 1 No prior treatment with an oncolytic therapy No active CNS metastases

Phase 1 primary & secondary objectives (dose rising & combined with nivolumab) 19 Primary objectives To determine the maximum tolerated dose (MTD) and/or the recommended Phase 2 dose of RP1 To assess the safety and tolerability of RP1 alone and in combination with nivolumab Secondary objectives To assess biological activity by changes in tumor size, inflammation, necrosis and erythema To assess RP1 biodistribution and shedding To assess the changes in levels of anti-HSV-1 antibodies

Phase 1 data summary 20 The side effect profile was as expected alone & in combination Dose rising monotherapy phase The recommended dose for further development was established Tumor destruction was demonstrated, including delayed systemic tumor reductions without further therapy Kinetics of detection of RP1 suggested robust virus replication • • • • Combination with nivolumab phase Clinical activity seen in multiple patients with various tumor types Two of the three anti-CTLA-4 + anti-PD-1 refractory cutaneous melanoma patients responding • • The first ipilimumab/nivolumab refractory cutaneous melanoma patient in Phase 2 has also responded • Clear clinical activity seen in CSCC, including CR • Indications of activity in other tumor types also seen • Rapid tumor reduction seen before nivolumab, which is given from the second dose of RP1 • • • Abscopal effects were observed Increases in CD8 T cells, PD-L1 & inflammatory gene expression seen across tumor types Including reversal of T cell exclusion •

21 The patients enrolled

Patients enrolled - dose escalation 22 * The inclusion criteria were narrowed to exclude PS2 soon after the trial start Total Melanoma Colorectal Head and neck Breast Esophage al Pancreatic CSCC Cholangiocarcin oma Number 22 8 5 2 2 2 1 1 1 Age: Range 22-81 22-71 30-59 61-78 55-65 66-81 58 58 68 ECOG performance status: 0, 1, 2 (%) 0: 13 1: 8 2: 1* 0: 4 1: 3 2: 1* 0: 4 1: 1 1: 2 0: 1 1: 1 0: 2 0: 1 1: 1 0: 1 Number of prior therapies: Range 1-13 1 – 5 2-5 3-5 8,13 1,2 5 2 1 Prior anti-PD1 therapy: Number (%) 8 (36.4) 7 (87.5) 0 1 (50) 0 0 0 0 0 Prior anti-PD1 + anti-CTLA-4 therapy: Number (%) 7 (31.8) 7 (87.5) 0 0 0 0 0 0 0 Baseline HSV serostatus: +ve, -ve (#) 14, 8 4, 4 3, 2 1, 1 1, 1 2, 0 1, 0 1, 0 1, 0

Patients enrolled - in combination with nivolumab 23 Total Cutaneous melanoma Uveal melanoma Mucosal melanoma CRC CSCC BCC Breast Bladder Adeno. of cecum MSI-H (Colon) Esepha geal Number 14 3 2 1 1 1 1 1 1 1 1 1 Age: Range 28-74 28-67 53-64 74 70 61 56 59 51 41 50 70 ECOG performance status: 0, 1 (#) 0: 8 1: 6 0: 2 1: 1 0: 1 1: 1 0: 1 0: 1 1: 1 0: 1 0: 1 1: 1 1: 1 1: 1 0: 1 # of prior therapies: Range 1-8 2-3 1, 2 2 5 1 4 5 2 8 1 6 Prior anti-PD1 therapy: Number (%) 6 (42.9) 3 (100) 1 (50.0) 1 (100) 0 0 0 1 (100) 0 0 0 0 Prior anti-PD1 + anti-CTLA-4 therapy: Number (%) 5 (35.7) 3 (100) 1 (50.0) 1 (100) 0 0 0 0 0 0 0 0 Baseline HSV serostatus: +ve, -ve (#) 11, 3 1, 2 2, 0 1, 0 1, 0 1, 0 1, 0 0, 1 1, 0 1, 0 1, 0 1, 0

Patients enrolled - in combination with nivolumab 24 Current focus for registration directed clinical development Total Cutaneous melanoma Uveal melanoma Mucosal melanoma CRC CSCC BCC Breast Bladder Adeno. of cecum MSI-H (Colon) Esepha geal Number 14 3 2 1 1 1 1 1 1 1 1 1 Age: Range 28-74 28-67 53-64 74 70 61 56 59 51 41 50 70 ECOG performance status: 0, 1 (#) 0: 8 1: 6 0: 2 1: 1 0: 1 1: 1 0: 1 0: 1 1: 1 0: 1 0: 1 1: 1 1: 1 1: 1 0: 1 # of prior therapies: Range 1-8 2-3 1, 2 2 5 1 4 5 2 8 1 6 Prior anti-PD1 therapy: Number (%) 6 (42.9) 3 (100) 1 (50.0) 1 (100) 0 0 0 1 (100) 0 0 0 0 Prior anti-PD1 + anti-CTLA-4 therapy: Number (%) 5 (35.7) 3 (100) 1 (50.0) 1 (100) 0 0 0 0 0 0 0 0 Baseline HSV serostatus: +ve, -ve (#) 11, 3 1, 2 2, 0 1, 0 1, 0 1, 0 1, 0 0, 1 1, 0 1, 0 1, 0 1, 0

Advanced patients were enrolled 25 4401-1003 4401-1006 4401-1007 4401-1016 4402-1006 4401-1019 4401-1020 4401-1018 4401-1005 4401-1008

26 Safety & tolerability

Treatment-related adverse events – dose escalation 27 Side effects as expected for an oncolytic immunotherapy ‘Flu-like constitutional symptoms, chills & rigors were the main side effects observed; self resolving within 72hrs of injection No obvious differences between deep & superficial dosing Modest increase in Grade 1-2 events with dose One DLT (elevated lipase) in the deep low dose cohort led to dose expansion to N=6 No procedure-related AEs 32 SAEs reported, 8 related to RP1 5 pyrexia, 2 vomiting, 1 tachycardia N=22 Preferred term Grade 1-2 (>15%) # (%) Grade 3 (all) # (%) Grade 4 (all) # (%) Grade 5 (all) # (%) Pyrexia 16 (72.7) Fatigue 9 (40.9) Chills 7 (31.8) Vomiting 4 (18.2) Influenza like symptoms 4 (18.2) Headache 4 (18.2) Lipase increased 1 (4.5) Total 19 (86.4) 1 (4.5) Patients who discontinued due to TEAE 0

Treatment-related adverse events – with nivolumab 28 No evidence of increased side effects as compared to that expected for either drug alone Two procedure-related AEs were seen (pneumothorax, n=2, self resolved) 10 SAEs seen, 0 related to RP1 or nivolumab  Procedure-related SAE: Pneumothorax PD-related SAEs: 6 Co-morbid SAEs: 3 If anything, nivolumab related side effects appeared reduced No immune-related adverse events seen N=14 Preferred term Grade 1-2 (>15%) # (%) Grade 3 (all) # (%) Grade 4 (all) # (%) Grade 5 (all) # (%) Pyrexia 5 (35.7) Chills 5 (35.7) Nausea 4 (28.6) Tumour pain 4 (28.6) Influenza like illness 3 (21.4) Vomiting 3 (21.4) Fatigue 3 (21.4) Injection site pain 3 (21.4) Injection site necrosis 1 (7.1) Total 11 (78.6) 1 (7.1) 0 0 Patients who discontinued due to TEAE 0

Injection site erythema (24-48hrs post injection) 29 4401-1002 Melanoma 4401-1011 Colorectal cancer 4401-1013 Breast cancer 4402-1001 CSCC

Safety & tolerability conclusions 30 RP1 is well tolerated alone and in combination with nivolumab, with side effects as expected for each agent alone Both direct injection of superficial & nodal tumors, & imaging guided injection of deep/visceral tumors were well tolerated and practical The recommended Phase 2 dose by both dosing routes was: A first dose of 1x106 pfu/ml followed by multiple doses of 1x107 pfu/ml Up to 10mLs/injection day; Q2W or otherwise in line with cycles of anti-PD1 therapy

31 RP1 replication & seroconversion

Detection of RP1 in swab & blood samples: Dose Escalation Blood Deep cohort Blood: Deep Blood Superfical cohorts Blood: Superficial 32 C44o0h1o-1rt0:013b 1B 4402-1006 109 109 4401-1008 Cohort: 2a 4401-1013 2 A * C44o0h1o-1rt0:129b 4401-1020 C44o0h1o-1rt0:231b 2B 4401-1011 108 3 A108 Cohort: 3a 4402-1004 3B 107 107 Tumor swabs Superficial cohort Injection site: Superficial Tumor swabs 106 106 Deep cohort 107 106 105 104 103 102 101 100 107 Injection site: Deep 4401-1002 Cohort: 1a 1A 4401-1004 4401-1014 106 105 Cohort: 1b 1 B 4401-1015 C4o40h1o-r1t0: 129b 4401-1005 C44o0h1o-r1t0:028a 2 A 2 B 104 103 4401-1013 C4o40h1o-r1t0: 137b 3 B 4401-1011 C44o0h2o-r1t0:031a 4402-1004 3 A102 101 100 * assumes 15ug DNA/ml, 5L blood volume copies/swab of injected tumor surface Viral copies in total copies/swab of injected tumor surface Viral copies in total

Detection of RP1 in swab & blood samples: Dose Escalation Blood Deep cohort Blood: Deep Blood Superfical cohorts Blood: Superficial 33 C44o0h1o-1rt0:013b 1B 4402-1006 109 109 4401-1008 Cohort: 2a 4401-1013 2 A * C44o0h1o-1rt0:129b 4401-1020 C44o0h1o-1rt0:231b 2B 4401-1011 108 3 A108 Cohort: 3a 4402-1004 3B 107 107 Tumor swabs Superficial cohort Injection site: Superficial Tumor swabs 106 106 Deep cohort 107 106 105 104 103 102 101 100 107 Injection site: Deep 4401-1002 Cohort: 1a 1A 4401-1004 4401-1014 106 105 Cohort: 1b 1 B 4401-1015 C4o40h1o-r1t0: 129b 4401-1005 C44o0h1o-r1t0:028a 2 A 2 B 104 103 4401-1013 C4o40h1o-r1t0: 137b 3 B 4401-1011 C44o0h2o-r1t0:031a 4402-1004 3 A102 101 100 Substantial amounts of RP1 are still present two weeks after dosing (i.e. until next dose) in some patients * assumes 15ug DNA/ml, 5L blood volume copies/swab of injected tumor surface Viral copies in total copies/swab of injected tumor surface Viral copies in total

RP1 detection & seroconversion conclusions 34 RP1 was detected in the blood and on the injected tumor surface for up to two weeks (time of next dose) in some patients These kinetics are suggestive of robust virus replication* All HSV seronegative patients seroconverted by the third RP1 dose, and antibody titres increased in seropositive patients (see Appendix) * In the T-VEC Phase 1 clinical trial, at the standard dose of 1x106 pfu/ml followed by 1x108 pfu/ml given twice, only low level virus was detected on the tumor surface of one patient (7.5pfu/swab), & never in the blood beyond 8 hours (Hu et al, CCR 2006 12: 6737-6747)

35 Biomarkers

Biomarkers 36 Tumor biopsies were taken at baseline & day 43 in the expansion cohort Assessed for the presence of tumor Stained for PD-L1 and CD8 T cells Subjected to Nanostring analysis Inflammatory gene signature Bespoke oncolytic virus specific panel of genes Blood samples were assessed for the generation of patients B cell responses in a subset of

PD-L1 & CD8 T cell staining & pathology summary 37 Most post baseline biopsies showed extensive necrosis and/or were tumor free (see Appendix for summary table) Increases in both PD-L1 and CD8 T cells were seen across tumor types (see Appendix for summary table) Lack of tumor and/or necrosis in 50% of patients prevented assessment or quantification

CD8 T cell & PD-L1 staining is increased across tumor types 38 4401-1024 Esophageal cancer PD-L1 4403-1002 CSCC PD-L1 CD8 CD8 Baseline D43

CD8 T cell & PD-L1 staining is increased across tumor types 39 1105-1001 Mucosal melanoma (Ipilimumab/nivolumab refractory) 1119-1002 MSI-H CRC PD-L1 PD-L1 CD8 CD8 Baseline D43

Reversal of T cell exclusion with RP 1 combined with nivolumab .. .' ·'•. ;;. " .. CDS Baseline CDS D43 <!}-Replimune 4403-1003 Cutaneous melanoma (ipilimumab/nivolumab refractory) . ·,·

..Reversal PD-L 1 Baseline ofT cell exclusion with RPl combined with nivolumab PD-L 1 D43 <!}-Replimune 4403-1003 Cutaneous melanoma (ipilimumab/nivolumab refractory)

Nanostring analysis shows increased immune activation 42 * 0.0229 9 1105-1C0X01C-LM-9ucosal melanoma1105-100C1D-M27ucosal melanoma1105-10C0X1C-MRu6cosal melanoma1105-10C0D12-M76ucosal melanoma1105-1001-Mucos CCL-5 0 2.0 2 4 0 ancer4401-1024-Esophageal cancer4401-1024-Esophageal cancer4401-1024-Esoph 4401-1024-Esophageal can 1 4403-1002-CSCC -1 4403-1002-CSCC 1.5 4403-1002-CSCC 8 2 -2 4402-1008-Uveal melanoma 4402-1008-Uveal 4403-1001-Uveal melanoma 4403-1001-Uveal 4401-1022-Melan 4402-1008-Uveal melanoma 4402-1008-Uveal melanoma 4402-1008-Uveal melanoma 1.0 0 -2 4403-1001-Uveal melanoma 4403-1001-Uveal melanom-4a 4403-1001-Uveal melanoma 0 7 -1 -3 0.5 4401-1022-Melanoma 4401-1022-Melanoma -6 -2 0.0 -2 -4 6 -3 -4 -5 -8 -0.5 5 1105-1S00T1A-TM1ucosal melanom1a105-100T1IG-MITucosal melanoma1105-100C1D-M8aucosal melanoma1105-10L0A1G-M-3ucosal melanoma1105-1001-Mucos IDO1 1 3 0 2 0 eal can al canc cancer4401-1024-Esophageal cancer4401-1024-Esoph 0 4403-1002-CSCC 4403-1002-CSCC 1105-1001-Muco 4401-1024-Eso 4403-1002-CSCC 1105-1001-Mucosal melanoma 0 anoma 9 om-2a oma 4402-1008-Uveal melanoma 4402-1008-Uveal 4403-1001-Uveal melanoma 4403-1001-Uveal -1 2 elano-m2 elanom anoma -2 -2 ma a 4401-1022-Melanoma 4401-1022-Melan -3 -4 8 1 -4 9 -4 44041-01022-41-E0so0ph8a-gUeavl ceanacel r 4403-1002-CSCC 44042-01030-81-U0ve0al1m-eUlanvoemaa -4 8 -5 0 -6 -6 -6 7 7 44043-01010-11-U0ve2al2m-eMlaneoml a 1105T-1L0R0-17-/M8ucosal melanom1a105-1T0L0R1--9Mucosal melanom11a05-H1L0A01-E-Mucosal melan1o1m05a-10N0K1G-M7ucosal melanom1a105-1001-Muco 4401-1022-Melanoma TLR3 2 0 1 hagea3l canc4e4r01-1024-Esophage1al ca4n4c0e1r-1024-Esophageal cance4r 401-1024-Esop 6 6 phageal 0 -2 C 4403-1002-CSCC CC 4403-1002-CSCC 4403-1002-CSC 0 2 0 5 melanoma 4402-1008-Uveal melanoma al melan -1 4402-1008-Uveal melanoma 4402-1008-Uvea 5 -4 al melan -1 1 melanoma4403-1001-Uveal me-l2anom4a403-1001-Uveal melanoma 4403-1001-Uvea -2 Time anoma-6 -2 oma0 4401-1022-Melanoma -3 4401-1022-Melanoma 4401-1022-Mela -4 -8 -3 -1 Time Increased tumor -4 -10 -4 -6 -2 -5 inflammatory score 18 gene panel known to be associated with response to anti-PD1/L1 (Haddad R. Abstract 5009; ASCO 2017, Ayers et al 2017 JCI 127.8) Selected genes showing increa TIS score TIS score TIS score log2 gene expression log2 gene expression log2 gene expression log2 gene expression log2 gene expression log2 gene expression log2 gene expression log2 gene expression log2 gene expression log2 gene expression log2 gene expression log2 gene expression log2 gene expression log2 gene expression log2 gene expression sed expressi on canc4e4r 01-1024-Esop 4403-1002-CSC oma 4402-1008-Uveal oma 4403-1001-Uveal 4401-1022-Melan 4401-1024-Eso 4403-1002-CS 4402-1008-Uve 4403-1001-Uve 4401-1022-Mel er4401-1024-Esophageal 4403-1002-CSCC 4402-1008-Uveal melan 4403-1001-Uveal melan 4401-1022-Melanoma ce4r401-1024-Esophage 4403-1002-CSCC a 4402-1008-Uveal mel a 4403-1001-Uveal mel 4401-1022-Melanom 4401-1024-Esophag 4403-1002-CSCC 4402-1008-Uveal m 4403-1001-Uveal m 4401-1022-Melano 4403-1002-CSCC 4401-1022-Melanoma ce4r 401-1024-Esophageal c 4403-1002-CSCC 4401-1022-Melanoma

RP1+nivolumab increases autoimmune B cell responses 43 Oncimmune have developed a high throughput approach to detecting autoimmune B cell responses which may be useful biomarkers for immuno-oncology Oncimmune used this platform to assess pre-and post-blood samples (29 days) from 4 RP1+nivolumab treated patients (4401-1022, 4401-1023, 4401-1024, 4403-1001) Increased reactivity was seen to 72 proteins: While there was some overlap, most were unique to each patient It was concluded that RP1+nivolumab treatment induces a broad autoantibody response in cancer patients Pre Post Pre Post Pre Post 4401-1022 Cutaneous melanoma 4403-1001 Uveal melanoma 4401-1024 Esophageal cancer Oncimmune high-content cancer immunotherapy array Clustered heatmap of top antigens, pre & post treatment

RP1 combined with nivolumab induces vitiligo (Phase 2 patient) 44 Baseline D57 D113 Left foot Vitiligo D113 Right heel (injected once with RP1; subsequent injections into inguinal nodes) Left chest 1119-2001 Cutaneous melanoma

Biomarker conclusions 45 Increases in CD8 T cells & PD-L1 were seen across tumor types Reversal of T cell exclusion was observed Most biopsies showed extensive necrosis and/or were tumor free Increases in autoimmune B cell responses were seen, suggestive of activation broad immune Nanostring data demonstrates increases in the tumor inflammatory score & changes in expression levels of genes in the bespoke oncolytic virus gene panel Data suggests that RP1 is providing broad anti-tumor immune activation

46 Anti-tumor activity

Anti-tumor activity - dose rising cohorts (single agent RP1) 47 CT scans per protocol were performed at baseline and at 30 days post the last RP1 dose for the dose rising cohort patients treated with single agent RP1 Tumor shrinkage was seen in injected and uninjected tumors, including in the three patients with visible tumors Delayed systemic tumor reduction (post initial disease progression and beyond the 30 day cut off) without other therapy was seen in two patients One patient previously refractory to anti-PD1 therapy responded to further anti-PD1 therapy

Tumor necrosis in the visible tumors in the dose rising cohorts 48 4402-1004 Salivary gland cancer 4401-1008 Breast cancer 4401-1001 Melanoma The only other cutaneous tumor in the Only two patients in the dose rising cohorts had a superficially visible (i.e. cutaneous) tumor which was injected dose rising cohort patients (non-injected) became necrotic (injection site: axilla)

Shrinkage of injected tumors 49 Baseline 30 day follow up 4402-1001 CSCC Central necrosis of a large injected tumor in the neck. The patient also had lung metastases 4401-1005 Colorectal cancer Destruction of the injected tumor nodule. The patient also had extensive additional abdominal disease 4401-1015 Melanoma The injected lung lesion reduced by 30%. The patient also had extensive additional disease in the lungs and pleura

Delayed systemic anti-tumor effects in the dose rising cohort 50 Patients were followed per protocol in the dose rising for 30 days post the last RP1 dose to collect safety data, followed by a CT scan Delayed, post study (CT scans not available), anti-tumor effects (systemic reduction in disease burden) were seen in 2 patients following single agent RP1 injected into a single tumor: 4401-1003 Melanoma, prior ipilimumab & nivolumab. Baseline disease: lung, pleura, lymph node & subcutaneous metastases. Five injections into the apical mass. Progression at the 30 day scan, then with no further therapy had reductions in the apical mass & hilar nodes, maintained for 8 months prior to PD 4401-1018 Cholangiocarcinoma, prior chemotherapy. Baseline disease: multiple liver & lung metastases. Five injections into one liver lesion from September 2018. Progression at the 30 day scan, then with no further therapy had reductions in liver lesions, bone metastases & lymph nodes A pembrolizumab refractory melanoma patient responded to subsequent nivolumab: 4401-1004 Melanoma, prior ipilumumab, pembrolizumab & clinical trial. Baseline disease: adrenal gland, liver, gluteal mass, subcutaneous deposits. Five injections into the 6.5cm gluteal mass. The injected lesion stabilized, others progressed, following which the patient received additional nivolumab followed by reduction in liver & subcutaneous metastases 30 day follow up is likely too short for inflammatory & other effects mediated by RP1 to have resolved

Anti-tumor activity – combination with nivolumab 51 Two of the three patients enrolled with cutaneous melanoma, all of whom were ipilimumab & pembrolizumab or ipilimumab/nivolumab refractory, are responding The first ipilimumab/nivolumab patient in Phase 2 with follow up scans has also responded (see melanoma-specific deck also released today) CR in the patient with chemotherapy refractory CSCC enrolled into Phase 1 Further activity in CSCC seen in Phase 2 (see CSCC-specific deck also released today) Clear biologic activity seen in additional tumor types (including uveal melanoma) with two patients with esophageal and MSI-H cancer still ongoing with the opportunity to achieve response The majority of biopsies taken at Day 43 showed extensive necrosis and/or no remaining viable tumor 

Patient #: 4403-1002 (chemotherapy refractory CSCC) 52 3rd Jan 2019 7th October 2019 13th May 2019 3rd Jan 2019 21st Jan 2019 19th August 2019 Patient with extensive recurrent CSCC previously treated with surgery (including skin grafts), radiotherapy, cisplatin/5FU, then electrochemotherapy Now CR with residual areas tumor free by multiple biopsy & continuing to heal In addition to the complete tumor response, the patients’ quality of life has been dramatically improved

Patient #: 4401-1022 (ipilimumab/pembrolizumab refractory melanoma) 53  Disease sites: Breast, lung, mediastinal and peritoneal anterior to the spleen RP1 injection site: Lesion behind the left ear Confirmed progression on prior immune checkpoint blockade, where two sequential PET scans demonstrated new lesions while also concurrently being treated with local therapy for the lesion behind the ear, then entry into the RP1 clinical trial PR with reduction in the breast, lung, mediastinum & anterior to the spleen Patient remains on nivolumab at 11 months Lung Anterior mediastinum Baseline 10 months

Patient #: 4403-1003 (ipilimumab/nivolumab refractory melanoma) 54 10th June 2019 24th June 2019 (pre nivolumab) 2nd September 2019 Patient history: Metastatic disease initially treated with ipili/nivo with best response of SD, then clear progression in the groin & thigh when radiotherapy followed by electrochemotherapy was added to continued nivolumab; following further clear progression, enrolled into the current trial All tumors flattened after the first dose of RP1 (1x106 pfu/ml) & extensive oedema rapidly reduced

Patient #: 4403-1003 (ipilimumab/nivolumab refractory melanoma) 55 Patient also had nodes in the groin which increased and are now reducing and lung metastases which following no change for prior 18 months are now reducing Patient quality of life has also greatly improved, from being essentially immobile at baseline to now able to go on long country walks Patient remains on treatment at >4 months May 2019 (Baseline) August 2019

Patient #: 1119-2003 (ipilimumab/nivolumab refractory melanoma) 56 26th August 2019 2nd July 2019 (screening) 17th Sept 2019 1st July 2019 (screening) 24th October 2019 The first ipilimumab/nivolumab refractory melanoma patient enrolled into the Phase 2 part of the study. Other such patients do not have follow up scans. Current status: Ongoing partial response

Patient #: 4403-1001 (ipilimumab/nivolumab refractory uveal melanoma) 57 Baseline (2nd Jan 2019) 24th April 2019 Patient has numerous additional subcutaneous lesions - remains on nivolumab at >9 months

Patient #: 1119-1002 (chemotherapy refractory MSI-H colorectal cancer) 58 Chemotherapy refractory MSI-H colorectal cancer 5cm liver metastasis and other liver metastases of around 1cm The liver lesions showed initial substantial increase (presumed inflammation), then reduction A biopsy taken at C4 demonstrated “A few viable glands of adenocarcinoma with extensive necrosis, fibrosis & chronic inflammation” Increases in PD-L1 and CD8 T cells were also seen Patient continues on therapy at 4 months 9th July 2019 (5.4x4.3cm) 26th August 2019 23rd September 2019 (5x5cm) (Biopsy taken 9th Sept showed extensive necrosis & inflammation) (8x6cm)

Patient #: 4401-1024 (esophageal cancer) 59 Heavily pre-treated esophageal cancer (8 prior therapies) Lung lesions and lesions around the esophagus. Largest lung lesion injected 6 times after which too soft for further injections. Further smaller lung lesion injected once, but discomfort and self resolved pneumothorax. RP1 injections stopped at C7 CT scan on 8th October 2019 (C11) showed overall disease reduction of 23% Patient continues on nivolumab treatment at 7 months Baseline C5 C11 Baseline C11

Tumor shrinkage after one RP1 dose (1105-2003; pembrolizumab refractory melanoma) 60 23rd October2019 (pre-nivolumab) 10th October 2019 16th October 2019 Increased drainage noted by D6 prior to tumors reducing by D13 (pre-nivolumab) Patient remains on treatment at 2-3 weeks

Tumor shrinkage after one RP1 dose (4402-2001; CSCC) 61 1st July 2019 (post one dose of RP1, no nivolumab) 16th July 2019 (post 2 doses of RP1 & 1 dose of nivolumab) 16th June 2019 (baseline) Patient with recurrent CSCC of the neck (bilateral), previously treated with cisplatin-based chemoradiation & six cycles of carboplatin/5-FU, prior to entering the clinical trial Both the large injected tumor & the smaller contralateral tumor in the neck reduced considerably before the first nivolumab dose, i.e. after the first dose of RP1

Patient #: 4402-2001 (CSCC) 62 Right neck (injected) 8 weeks Left neck (not injected) 8 weeks Baseline 16 weeks Baseline 16 weeks The protocol mandated biopsy of the injected tumor taken at day 43 was tumor free No tumor was found to remain when a biopsy was attempted from the left neck In addition to the tumor response, this patient has had a dramatic improvement in quality of life & is now off morphine which was previously necessary for substantial tumor pain

Patient #: 4402-2001 (CSCC) 63 Baseline 16 weeks The patient also had baseline retroperitoneal tumors which have completely resolved The only remaining disease are a number of non-measurable bone metastases, which were the main source of the cancer pain which has now resolved: These are now sclerotic, also suggestive of a treatment response, with Zometa also now having been withdrawn

Anti-tumor activity conclusions 64 Activity was seen with both RP1 alone and in combination with nivolumab Including the rapid tumor reductions seen before the introduction of nivolumab In particular we saw clear clinical activity in CSCC and in anti-PD1 refractory melanoma Provides strong support for our clinical programs in these tumor types Clinical activity in both tumor types has been further confirmed in initial patients enrolled in Phase 2 Indications of clinical activity in additional tumor types was also seen 

Overall conclusions from the clinical data so far 65 RP1 is well tolerated alone & in combination with nivolumab Both direct injection & imaging guided injection were well tolerated and practical RP1 provides potent oncolytic activity & abscopal effects Clinical activity was seen for RP1 alone & in combination with nivolumab CD8 T cell levels and PD-L1 were increased across tumor types The kinetics of detection of RP1 in blood suggests robust virus replication The clear clinical activity in CSCC and immune checkpoint blockade refractory melanoma provides strong support for expanding Replimune’s clinical programs with RP1 in these tumor types New study intended in organ transplant recipients with CSCC (announced in October) New study intended in anti-PD1 refractory melanoma (announced today) Anti-tumor effects were also seen in other tumor types Data with RP1 combined with nivolumab in bladder cancer & MSI-H tumors is pending

66 Q&A

67 Appendix

Administration site, volume & method used for each deep dose rising cohort patient 68 used (CT or Patient # Tumor type Injection site Volume/ injection (range in mL, given up to 5 times) Imaging guidance ultrasound) 4401-1003 Melanoma Lung 5.0-10.0 Ultrasound 4401-1006 Colorectal Liver 3.0 Ultrasound 4401-1007 Melanoma Liver 6.0 Ultrasound 4401-1014 Esophageal Liver 3.0 – 10.0 Ultrasound 4401-1015 Melanoma Pleural deposit 2.5 – 3.0 Ultrasound 4401-1016 Head and Neck Liver 8.0 – 10.0 Ultrasound 4401-1012 Melanoma Lung 5.0 – 6.0 Ultrasound 4402-1006 Head and Neck Liver 10.0 Ultrasound 4401-1019 Melanoma Stomach 10.0 CT 4401-1020 Pancreatic Liver 0.5 Ultrasound 4401-1017 Colorectal Liver 3.0 Ultrasound 4401-1018 Cholangeocarcinoma Liver 0.5 – 1.0 Ultrasound 4401-1021 Colorectal Liver 4.0 Ultrasound

Administration site, volume & method used for each deep expansion cohort patient 69 used (CT or Patient # Tumor type Injection site Volume/ injection (range in mL, given up to 8 times) Imaging guidance ultrasound) 1105-1002 Melanoma Liver 0.5-2.7 CT/Ultrasound 1110-1003 Colorectal Liver 3.1 CT 1110-1004 Colorectal Abdomen wall 3.1 CT 1119-1001 Bladder Lung 0.1-6.0 CT 1119-1002 MSI-H colorectal Liver 4.0-6.0 CT 4401-1024 Head and Neck Lung 0.5-3.0 CT 4402-1008 Uveal Melanoma Liver 10 Ultrasound

HSV antibody titres – dose rising cohorts 70 Superficial Deep HSV-1 Seronegative patients HSV-1 Seronegative patients HSV-1 Seropositive patients HSV-1 Seropositive patients

HSV antibody titres – expansion cohorts 71 Deep Superficial HSV-1 Seronegative patients (no seronegative patients at baseline) HSV-1 Seronegative patients HSV-1 Seropositive patients HSV-1 Seropositive patients

Most biopsies showed extensive necrosis and/or were tumor free 72 SCC. July 2019: 2 biopsies from last palpable area tumor free (confirmed CR) rare pigmented foamy macrophages inflammation melanocytes, moderate to heavy chronic inflammation * After three doses of RP1 & two doses of nivolumab ** Verbatim from site pathology reports Patient # Tumor type Site of tumor for D43* biopsy Baseline size of tumor for D43 biopsy Biopsy results** 4403-1001 Uveal melanoma Skin D43: No evidence of melanoma. C6: No evidence of melanoma. 4403-1002 CSCC Scalp 6.3cm D43: Heavy chronic inflammatory infiltrate. Small foci of atypical epithelium suspicious for 4401-1022 Melanoma Ear, left, behind 2.5cm Information awaited 1105-1001 Mucosal melanoma Lymph, right, inguinal 3.0cm D43: Melanoma with extensive necrosis 4401-1024 Esophageal Lung 3.1cm Lesion too soft to inject by C6 1119-1001 Bladder cancer Lymph node D43: Tumor present 1110-1003 CRC Liver 3.2cm D43: Local pathology not performed 1105-1002 Melanoma Liver, right, lobe 2.9cm D43: No evidence of melanoma; Benign liver parenchyma with mild portal inflammation, inc 4402-1008 Uveal melanoma Liver 13.4cm D43: Malignant melanoma 1110-1002 Breast cancer Supraclavicul ar node No biopsy taken (patient discontinued) 1110-1004 Adenocarcinom a of the cecum Peritoneal mass 3.3cm Patient refused biopsy 1119-1002 MSI-H CRC 4.6cm D43: A few viable glands of adenocarcinoma with extensive necrosis, fibrosis & chronic 4403-1003 Melanoma Thigh D43: Necrosis with no obviously viable/intact melanocytes, junctional component with viable

PD-L1 & CD8 T cell staining summary 73 (two biopsies) IHC & assessment performed by DCL Pathology *After two doses of RP1 & one of nivolumab **Combined peritumoral (0-3+) & intra-tumoral (0-3+) CD8 staining intensity score. UNK = unknown (archival biopsy without tumor). Data from remaining patients awaited Quantifiable increases in both PD-L1 and CD8 T cells were seen in three of the ten*** patients for whom IHC results are currently available (although can be seen visually in other patients; see earlier slides) Lack of tumor and/or necrosis in 50% of patients prevented assessment and/or quantification *** Excludes the patient with no baseline data Patient # Tumor type PD-L1 at baseline PD-L1 at D43* Peri and intra-tumoral CD8+ at baseline Peri and intra-tumoral CD8+** at D43 4403-1001 Uveal melanoma >5-10% Skin without defined tumor 2+ Skin without defined tumor, skin without evident tumor, skin without evident tumor (three biopsies) 4403-1002 CSCC >5-10%, >5-10% (two biopsies) Skin without definitive tumor 1+, 4+ (two biopsies) Skin without definitive tumor 4401-1022 Melanoma 1-5% >10% 4+ 4+ 4401-1023 BCC <1% <1% 1+ 3+ 1105-1001 Mucosal melanoma 1-5% Tumor is necrotic; limits interpretation 2+ Tumor is necrotic; limits interpretation 4401-1024 Esophageal cancer 1-5% 1-5%; >10% (two biopsies) 4+ 4+; 6+ 1119-1001 Bladder cancer >10% >10% 4+ 2+ 1110-1003 Colorectal cancer UNK <1% UNK 0 1105-1002 Melanoma >10% No tumor identified 1+ No tumor identified 4402-1008 Uveal melanoma <1% 1-5% 3+ 6+

RP1 induces abscopal anti-tumor effects 74 Multiple patients showed abscopal (i.e. uninjected) tumor reduction, including: Reductions at multiple tumor sites (hilar nodes, liver tumors, bone lesions, lymph nodes) following injection of RP1 into a either a single lung or a single liver tumor up to 5 times (patients 4401-1003, 4401-1018) The three patients with visible disease in the dose rising phase saw anti-tumor activity in uninjected lesions (patients 4401-1001, 4401-1008, 4402-1004) All the lesions for patient 4403-1002 (CSCC) flattened after only the first dose of RP1, even though only some of the lesions were injected, before ultimately achieving an overall CR Patient 4401-1022 (ipilimumab/pembrolizumab refractory cutaneous melanoma) had systemic reduction in lesions (including in lung, mediastinum, peritoneum) following injections to only a lesion behind the ear, leading to PR Patient 4402-2001 (chemotherapy refractory CSCC) had reduction in an uninjected tumor in the neck, prior to the addition of nivolumab

Randomized controlled Phase 2 study in CSCC (CERPASS) 75 RP1 IT Q3W x 8 doses† (1x106 PFU/mL for one dose followed by 1x107 PFU/mL for 7 doses) + Cemiplimab 350mg Q3W IV Key Eligibility Criteria: • • • • Locally-advanced/metastatic CSCC ECOG PS 0 or 1 No active autoimmune disease No prior treatment with a PD-1/PD-L1 inhibitor No prior treatment with other immune modulating agents (incl CTLA-4) No untreated brain metastases 2:1 N=240 • Cemiplimab 350mg Q3W IV • 57 weeks treatment‡ †First dose of RP1 to be given as monotherapy with cemiplimab to be given with second dose of RP1 ‡57 weeks treatment for the combination arm; treatment duration for cemiplimab-only arm is 54 weeks Key Endpoints Primary: ORR (RECIST v1.1) Secondary: DOR, PFS, OS, Disease-Specific Survival, safety/tolerability 3-year survival follow up

Phase 1b clinical trial in solid organ transplant recipients with CSCC 76 Key Eligibility Criteria: • Locally-advanced/metastatic CSCC ECOG PS 0 or 1 Renal or hepatic organ allograft recipients on stable immunosuppressive regimen for >12 mos No prior systemic anti-cancer treatment for CSCC No transplant-related viral infections (such as BK, EBV, CMV) within 3 months No untreated brain metastases • • RP1 IT Q2W x 26 doses (1x106 PFU/mL for one dose followed by 1x107 PFU/mL) • • 50 weeks treatment • Key Endpoints Primary: Safety and tolerability Secondary: ORR (RECIST v1.1), DOR, Disease-Free Survival, incidence/severity of graft rejection 3-year survival follow up

Replimune’s key target milestones 77 RP1 Randomized 2 data in CSCC 2019 2020 2021 with RP2 +/-RP2 RP3 Initiated Phase 1 nivolumab H1 2020: Manufacturing facility operational 2021: RP2 Phase 2 data 2020: RP2 Phase 1 data 2020: Initiate clinical development of RP3 H2 2021: controlled Phase Q1 2020: Initiate Phase 1b with RP1 in organ transplant patient with CSCC H2 2020: RP1+nivolumab Phase 2 data 2020: Initiate registration directed trial with RP1 in anti-PD1 refractory melanoma Initiated Phase 2 in 4 tumor types in combination with nivolumab Initiated 240 patient registration directed Phase 2 in CSCC Release Phase 1 data RP1 alone & RP1+ nivolumab

Replimune’s critical focus on manufacturing 78 Product candidates currently manufactured by a CMO for ongoing clinical development For later stage development & commercialization, in-house manufacturing is preferable The team has extensive manufacturing experience 63,000 ft2 manufacturing facility leased in July 2018 in Framingham, MA, appropriate for multi-product production – intended to include translational biomarker lab Of sufficient scale to cover full global commercialization of Replimune’s products Expected to be on-line to produce clinical product in H1 2020 

Reception,offices & utilities

Production areas