Adaptimmune investor and analyst day 2016

Adaptimmune investor and analyst day 2016 April 22, 2016

disclaimer This presentation contains forward-looking statements, as that term is defined under the Private Securities Litigation Reform Act of 1995 (PSLRA), which statements may be identified by words such as believe, may , will, estimate, continue, anticipate, intend, expect and other words of similar meaning. These forward-looking statements involve certain risks and uncertainties. Such risks and uncertainties could cause our actual results to differ materially from those indicated by such forward-looking statements, and include, without limitation: the success, cost and timing of our product development activities and clinical trials; our ability to submit an IND and successfully advance our technology platform to improve the safety and effectiveness of our existing TCR therapeutic candidates; the rate and degree of market acceptance of T-cell therapy generally and of our TCR therapeutic candidates; government regulation and approval, including, but not limited to, the expected regulatory approval timelines for TCR therapeutic candidates; and our ability to protect our proprietary technology and enforce our intellectual property rights; amongst others. For a further description of the risks and uncertainties that could cause our actual results to differ materially from those expressed in these forward-looking statements, as well as risks relating to our business in general, we refer you to our Annual Report on Form 20-F filed with the Securities and Exchange Commission (SEC) on October 13, 2015 and our other SEC filings. We urge you to consider these factors carefully in evaluating the forward-looking statements herein and are cautioned not to place undue reliance on such forward-looking statements, which are qualified in their entirety by this cautionary statement. The forward-looking statements contained in this presentation speak only as of the date the statements were made and we do not undertake any obligation to update such forward-looking statements to reflect subsequent events or circumstances. We intend that all forward-looking statements be subject to the safe-harbor provisions of the PSLRA. 2

Adaptimmune investor and analyst day 2016 April 22, 2016 James Noble Chief Executive Officer

Today s agenda Helen Tayton-Martin, PhD, MBA; Chief Operating Officer, Adaptimmune The Role of T Cells in the Immuno-Oncology Landscape Bent Jakobsen, PhD; Scientific Founder, Adaptimmune SPEAR T Cells : Adaptimmune s Proprietary Technology Platform Stephan Grupp, MD, PhD; U.Penn Perelman School of Medicine Adoptive T Cell Therapy: Clinical Activity of NY-ESO-1 in a Solid Tumor Aaron Rapoport, MD; U.Md Marlene & Stuart Greenebaum School of Medicine NY-ESO-1 T Cell therapy in Multiple Myeloma: Long Term Efficacy and Persistence Gwen Binder-Scholl, PhD; Chief Technology Officer, Adaptimmune The Adaptimmune Pipeline Engine Manufacturing Excellence and Commercial Delivery Rafael Amado, MD; Chief Medical Officer, Adaptimmune Update on Progress with NY-ESO TCR Accelerating Adaptimmune s Wholly-Owned Clinical Pipeline 4

Adaptimmune: Leading the TCR T cell space Proprietary SPEAR T cell technology uniquely delivers: Correctly identified targets Specificity and optimal affinity Supra-natural TCRs to accelerate programs Enhanced effectiveness of TCRs Generation 2 and 3 TCRs No other company can currently deliver all of these New data on the above are being presented today 5 Clear scientific leadership in the field of T cell engineering

Adaptimmune: Leading the TCR T cell space Multiple clinical responses in synovial sarcoma, a solid tumor Large solid lesions resolved Breakthrough status Pivotal trial planned for around year end 2016 Over 90% response rate in multiple myeloma study in conjunction with ASCT No other company is as far advanced as Adaptimmune in the clinic with a TCR T cell New updates presented on both diseases today 6 Clear scientific leadership in the field of T cell engineering Most compelling clinical data in the field

Adaptimmune: Leading the TCR T cell space Company INDs open for NY-ESO, MAGE-A10 and AFP Next company INDs due 2017 These TCRs all derive from Adaptimmune s proprietary technology No other company has routinely delivered INDs from an in-house TCR platform Today, we will disclose: Our next IND target The pipeline coverage of tumors 7 Clear scientific leadership in the field of T cell engineering Most compelling clinical data in the field Deep pipeline across major cancers

Adaptimmune: Leading the TCR T cell space Current capital can fund the business through mid-2018 8 Clear scientific leadership in the field of T cell engineering Most compelling clinical data in the field Deep pipeline across major cancers Strong financial position

Adaptimmune: Leading the TCR T cell space Three INDs open Manufacturing processes optimized Goal: first TCR T cell therapy to market 9 Clear scientific leadership in the field of T cell engineering Most compelling clinical data in the field Deep pipeline across major cancers Proven ability to execute Strong financial position

Building a leader A history of scientific pre-eminence Origins of TCR technology 2005-2007 Collaboration with Steve Rosenberg, MD, Ph.D. 2008-2011 Collaboration with U. Penn 2016 + Adaptimmune SAB 1999-2006 2006-2008 2008-TODAY 1993-1999 Academic Research Identifying targets Generating soluble TCRs used in R&D Engineering affinity optimized TCRs Cell Manufacturing TCR intellectual property Oxford IP All IP Adaptimmune spun out as new virtual company in 2008 First soluble TCR Perfected stable soluble TCRs and affinity enhancement First high affinity engineered TCR Avidex bought by Medigene First cell therapy data with affinity optimized TCR First solid tumor data First TCR with breakthrough designation 10 Adaptimmune leadership position in:

Adaptimmune investor and analyst day 2016 The ROLE Of T CELLS IN the Immuno-Oncology Landscape April 22, 2016 Helen Tayton-Martin, PhD, MBA Chief Operating Officer

Why immunotherapy? Cancers Primarily derived from changes to self-proteins Contain many mutations Are heterogeneous, even in the same patient Are good at mutating to avoid selective pressure Deploy a range of tactics to avoid immune system detection Re-establishing T cell recognition and catalysing a polyclonal T cell response is key 12 The key to tackling cancer effectively is immune engagement

EARLY BEGINNINGS WITH TUMOR INFILTRATING LYMPHOCYTES 13 IMMUNOTHERAPY emerging evidence TIL Therapy

EARLY BEGINNINGS WITH TUMOR INFILTRATING LYMPHOCYTES 14 2005 IMMUNOTHERAPY emerging evidence TIL therapy can mediate significant tumor regression in patients heavily pre-treated with IL-2 in refractory metastatic melanoma Significant toxicities TIL Therapy

IMMUNOTHERAPY emerging evidence IMMUNe-MODULATIOn CHECKPOINT BLOCKADE ANTI-CTLA-4 15 TIL Therapy Anti-CTLA-4

IMMUNOTHERAPY emerging evidence IMMUNe-MODULATIOn CHECKPOINT BLOCKADE ANTI-CTLA-4 16 Source: Postow, Callahan and Wolchok, J Clin Oncol 2015 33, 1974-1982 Anti-CTLA-4 antibodies block CTLA4 on T cells which disables their brakes TIL Therapy Anti-CTLA-4

17 2010 IMMUNOTHERAPY emerging evidence IMMUNe-MODULATIOn CHECKPOINT BLOCKADE ANTI-CTLA-4 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.0 0.0 0 4 8 12 16 20 24 28 32 36 40 44 48 52 56 PROPORTION ALIVE Ipilimumab can cure patients, response rate is low but significant (1/5 get long term survival) Immune-mediated toxicity and adaptive resistance TIL Therapy Anti-CTLA-4

IMMUNOTHERAPY emerging evidence ANTI-PD-1 NEXT STEP CHECKPOINT BLOCKADE 18 TIL Therapy Anti-CTLA-4 Anti-PD-1

IMMUNOTHERAPY emerging evidence ANTI-PD-1 NEXT STEP CHECKPOINT BLOCKADE 19 Source: Postow, Callahan and Wolchok, J Clin Oncol 2015 33, 1974-1982 Anti-PD-1 and anti-PDL-1 antibodies stop T cells encountering fatigue, especially in the tumor TIL Therapy Anti-CTLA-4 Anti-PD-1

20 IMMUNOTHERAPY emerging evidence ANTI-PD-1 NEXT STEP CHECKPOINT BLOCKADE Keytruda (anti PD-1 - Melanoma) Source: www.Keytruda.com Clear effects on survival in multiple indications Durable responses, lower toxicity 2015 TIL Therapy Anti-CTLA-4 Anti-PD-1

21 IMMUNOTHERAPY emerging evidence CHECKPOINt COMBINATIONS TIL Therapy Anti-CTLA-4 Anti-PD-1 Anti-CTLA-4 + Anti-PD-1

22 IMMUNOTHERAPY emerging evidence CHECKPOINt COMBINATIONS Enhanced effects with multiple immune checkpoint blockade Source: www.Opdivo.com TIL Therapy Anti-CTLA-4 Anti-PD-1 Anti-CTLA-4 + Anti-PD-1

IMMUNOTHERAPY CHALLENGEs Strong evidence for t cells but 23 TIL Therapy Anti-CTLA-4 Anti-PD-1 Anti-CTLA-4 + Anti-PD-1 Check-point blockade/TIL therapy requires effective pre-existing anti-tumor immunity Mutational burden is correlated with effectiveness of immunotherapy The endogenous immune response to self antigens is low affinity Tumors are selected to avoid immune detection a high affinity response is required

IMMUNOTHERAPY CHALLENGEs Strong evidence for t cells but 24 Engineered T cells can overcome many of these challenges TIL Therapy Anti-CTLA-4 Anti-PD-1 Anti-CTLA-4 + Anti-PD-1 Engineered T Cells Check-point blockade/TIL therapy requires effective pre-existing anti-tumor immunity Mutational burden is correlated with effectiveness of immunotherapy The endogenous immune response to self antigens is low affinity Tumors are selected to avoid immune detection a high affinity response is required

The opportunity for engineered t cell therapy A powerful modality 25 Ability to engineer-in effective tumor antigen specificity to T cells Ability to engineer-in alterations to overcome the tumor microenvironment (next generation) Specific therapy: Engineered T cells migrate to antigen / tumor and provide localized responses

There ARE TWO MAIN WAYS TO REDIRECT A T CELL SYNTHETIC RECEPTORS (CAR) AND T CELL RECEPTORS (TCR) 26 *CAR Chimeric Antigen Receptor, ALL Acute Lymphoblastic leukemia, DLBCL Diffuse Large B-Cell Lymphoma

CAR T Cells Evidence in hematological cancers Not easily transferable to solid tumors 27 Anti-CD19 CAR-Ts have demonstrated evidence of high tumor shrinkage and remissions in B cell malignancies *CAR Chimeric Antigen Receptor, ALL Acute Lymphoblastic leukemia, DLBCL Diffuse Large B-Cell Lymphoma, CR - Complete Response, CRM Complete Molecular remission Efficacy of CD19 CAR-Ts in hematological cancers Source: Juno Therapeutics; CITI Car-T / TCR launch initiation report, 2016

CAR T Cells Evidence in hematological cancers Not easily transferable to solid tumors 28 Efficacy of CD19 CAR-Ts in hematological cancers Anti-CD19 CAR-Ts have demonstrated evidence of high tumor shrinkage and remissions in B cell malignancies Two issues: very few targets and little evidence of efficacy in solid tumors *CAR Chimeric Antigen Receptor, ALL Acute Lymphoblastic leukemia, DLBCL Diffuse Large B-Cell Lymphoma, CR - Complete Response, CRM Complete Molecular remission

Optimized affinity TCR T cells Address solid tumors and intracellular targets 29 Optimized affinity TCR T cells demonstrate efficacy in solid tumors Sources: SITC, November 2015

Optimized affinity TCR T cells Address solid tumors and intracellular targets 30 Optimized affinity TCR T cells demonstrate efficacy in solid tumors Vast majority of cancer targets are intracellular and ONLY engaged by T cells via TCRs

DEVELOPING EFFECTIVE affinity optimized TCR T CELLS high barriers to entry 31 Specialized data and expertise required to identify the correct peptide targets Challenging to identify TCRs Essential to optimize the affinity of the TCR Specialized expertise and assays required for establishing TCR specificity Manufacturing expertise for both cell and vector required Clinical expertise for safe administration and effective study design for cell therapy required

ADAPTIMMUNE SPEAR t cell PLATFORM uniquely overcomes these hurdles 32 SPEAR T CELLS Specific Peptide Enhanced Affinity Receptor Target Identification TCR Identification TCR Engineering Optimized Affinity TCR Safety Testing Generation 2 T cells Generation 3 T cells

Adaptimmune investor and analyst day 2016 SPEAR T cells: Adaptimmune s Proprietary Technology Platform April 22, 2016 Bent Jakobsen, Ph.D. Chief Scientific Officer and Co-founder, Immunocore Scientific Founder, Adaptimmune Therapeutics plc Fellow of The Academy of Medical Sciences

ADOPTIVE T CELL - Most powerful unit in Immunotherapy Four components to an effective adoptive therapy: T cell must recognize a cancer cell via a guiding receptor The guiding receptor has two important aspects Affinity Specificity Cancer cell Apoptosis T cell Armory Perforin Granzyme T cell 34 Cancer cell Affinity Specificity T cell 1. 2.

ADOPTIVE T CELL - Most powerful unit in Immunotherapy Four components to an effective adoptive therapy: T cell must recognize a cancer cell via a guiding receptor The guiding receptor must have two important aspects Affinity Specificity The T cell needs to be resistant to suppression T cell Cancer cell Inhibitory mechanisms of cancer cells make T cells insensitive Make T cells resistant to suppression 35

ADOPTIVE T CELL - Most powerful unit in Immunotherapy Four components to an effective adoptive therapy: T cell must recognize a cancer cell via a guiding receptor The guiding receptor must have two important aspects Affinity Specificity The T cell needs to be resistant to suppression The T cell (either alone or via other mechanisms) needs to break cancer immune tolerance T cell Cancer cell Make T cells resistant to suppression Epitope spreading Epitope spreading 1 2 3 4 36

37 Affinity & specificity

ADOPTIVE T CELL - Most powerful unit in Immunotherapy Four components to an effective adoptive therapy: T cell must recognize a cancer cell via a guiding receptor The guiding receptor has two important aspects Affinity Specificity Cancer cell Affinity Specificity T cell 38

TCR Affinity - Determined by Thymic selection The entire peptidome (all peptides) is presented in the thymus T cells undergo positive and negative selection within the cortex and medulla of the thymus High affinity Low affinity Negative selection Positive selection Thymus 39 x

Natural T cells are ill-equipped to clear cancer Due to negative selection virtually all circulating T cells that have self peptide specificity will have low affinity TCRs This mechanism guards the body against autoimmunity However all reasonably prevalent peptide antigens of cancer relevance are of self origin Many of these peptide antigens are derived from proteins for which the encoding gene is silenced (or severely suppressed) in all (or almost all) adult tissues 40

Viral TCRs have higher affinity than natural cancer TCRs Aleksic, M. et al. (2012). Eur J Immunol 10.1002/eji.201242606 Strong Weak Antigen recognition Virus TCRs Naturally-occurring Cancer TCRs MAGE-A4 KD ~ 64 mM MAGE-A10 KD ~ 2 mM Not a natural TCR Identified from proprietary TCR display libraries 41

Affinity optimizing cancer TCRs is pivotal to t cell function Some non-engineered TCRs will recognise antigen well e.g. NY-ESO Even so, engineering improves antigen recognition Some non-engineered TCRs fail to recognise antigen well e.g. MAGE-A4 Engineering enables antigen recognition Affinity engineering is a critical step in TCR optimization 42

NY-ESO: natural Versus engineered TCR functionality IM-9 SKMel137 HEP2 N2 NY-ESO+ NY-ESO_ NY-ESO engineered NY-ESO wt IFNg Spot Numbers GrB Spot Numbers IM-9 SKMel137 HEP2 HA2 NY-ESO+ NY-ESO_ 43 Even for NY-ESO affinity engineering improves T cell function

NY-ESO: natural Versus engineered TCR functionality IM-9 (NY-ESO+) N9 (NY-ESO-) NY-ESO wt NY-ESO Engineered 44 Even for NY-ESO affinity engineering improves cancer killing % cytotoxicity % cytotoxicity E:T ratio E:T ratio E:T ratio -10 0 10 20 30 40 50 60 70 0.625 1.25 2.5 5 -10 0 10 20 30 40 50 60 70 0.625 1.25 2.5 5

MAGE-A4: effect of optimizing tcr affinity 45 TCR Affinity (mM) HCT116 Human colon carcinoma U266 Human myeloma IM9 EBV-transformed B cell A375 Human melanoma SKMel28 Human melanoma Mel624 Human melanoma ntd 208 144 64 40 25 24 7 3 1 ntd 208 144 64 40 25 24 7 3 1 ntd 208 144 64 40 25 24 7 3 1 ntd 208 144 64 40 25 24 7 3 1 ntd 208 144 64 40 25 24 7 3 1 ntd 208 144 64 40 25 24 7 3 1 WT WT WT WT WT WT

Affinity optimizing cancer TCRs is pivotal to t cell function Some non-engineered TCRs will recognise antigen well e.g. NY-ESO Even so, cancer cell killing is dramatically improved by affinity optimization Some non-engineered TCRs fail to recognise antigen well e.g. MAGE-A4 The optimal affinity is crucial for T cell function and the same across all cancer cell lines Affinity engineering is a critical step in TCR optimization 46

47 Specificity & Toxicity

Specificity and non-specificity Off target (specific) Off target (non-specific) alloreactivity Closely related antigen on normal tissue Alternate HLA restriction Target antigen on normal tissue normal tissue On target 48 Promiscuous recognition (unrelated to target)

Target antigen on normal tissue Specificity: target expression On target Optimized target selection process Select targets expressed on cancer cell Low expression (due to HLA downregulation) overcome by high affinity TCRs No / extremely low expression in normal tissue* normal tissue *Expression tolerable in some normal tissues (e.g. prostate, breast, pancreas, immuno- privileged tissues) 49

MAGE-A4 - Expression in Cancer Cancer Type Normalized Expression MAGE-A4 expression from cancer genome atlas (TCGA) lung cervical bladder head & neck breast ovarian melanoma stomach 50

MAGE-A4 - Expression in normal tissue Tissue Type RNAseq Log scale RNAseq (Log scale) GTEx Portal 51 Expression is absent / low in most adult non-reproductive tissues

MAGE-A10 - Expression in Cancer Normalized Expression MAGE-A10 expression from cancer genome atlas (TCGA) 52 lung SgCC bladder head & neck breast ovarian melanoma lung Cancer Type

MAGE-A10 - Expression in normal tissue RNAseq (Log scale) GTEx Portal RNAseq Log scale Tissue Type 53 Expression is absent / low in most adult non-reproductive tissues testis

Specificity and non-specificity Off target (specific) Closely related antigen on normal tissue 54 Potential for off-target specificity can be analyzed because the antigen is short and linear X-Scan Identify potential targets via genome search Test recognition by high affinity TCR

X-SCAN: individual peptide position Specificity Testing T A R G E T H E R E A A R G E T H E R E C A R G E T H E R E D A R G E T H E R E E A R G E T H E R E F A R G E T H E R E G A R G E T H E R E H A R G E T H E R E I A R G E T H E R E K A R G E T H E R E L A R G E T H E R E M A R G E T H E R E N A R G E T H E R E P A R G E T H E R E Q A R G E T H E R E R A R G E T H E R E S A R G E T H E R E V A R G E T H E R E W A R G E T H E R E Y A R G E T H E R E T A R G E T H E R E A I K - - M - - E D C L H K G D V N K E S Q F S G Y H I K L M N P Q R S V W Y TCR tolerance motif Search Human Genome for all possible peptide matches and investigate these via: TCR recognition Peptide presentation Essential to determine which amino acids are critical This is achieved by Single Amino Acid Substitution Mapping (X-scan) 55

MAGE-A10 - TCR GENERATION AND SELECTION Three TCRs selected by specificity testing from an original pool of 21 parentals Affinity enhancement leads to 15 variants plus 3 parents and 1 reverted heavy chain parental for testing Cell based potency and specificity testing to select five candidates from 2 parents Additional efficacy testing resulted in two comparable leads for X scan evaluation clone 796 clone 756 non-specific * * * Point mutation in heavy chain (corrected at level 2) 56

MAGE-A10 TCR: X-scan Specificity Analysis TCR peptide recognition mapping using combinatorial amino acid substitutions POSITION 1 2 3 4 5 6 7 8 9 G L Y D G M E H L X L Y D G M E H L G X Y D G M E H L G L X D G M E H L G L Y X G M E H L G L Y D X M E H L G L Y D G X E H L G L Y D G M X H L G L Y D G M E X L G L Y D G M E H X TOLERANCE MOTIF X ACFGILMNQSTV FWY ACDY AGLPST MP EQ CHLMQR ACFILMTV Position 7 substitution Position 9 substitution Position 8 substitution Position 4 substitution Position 6 substitution Position 5 substitution Position 1 substitution Position 3 substitution Position 2 substitution 57 R e l a t i v e r e s p o n s e A C D E F G H I K L M N P Q R S T V W Y A C D E F G H I K L M N P Q R S T V W Y A C D E F G H I K L M N P Q R S T V W Y 0.0 0.5 1.0 1.5 p1 p2 p3 * * R e l a t i v e r e s p o n s e A C D E F G H I K L M N P Q R S T V W Y A C D E F G H I K L M N P Q R S T V W Y A C D E F G H I K L M N P Q R S T V W Y 0.0 0.5 1.0 1.5 p4 p5 p6 * * R e l a t i v e r e s p o n s e A C D E F G H I K L M N P Q R S T V W Y A C D E F G H I K L M N P Q R S T V W Y A C D E F G H I K L M N P Q R S T V W Y 0.0 0.5 1.0 1.5 p7 p8 p9 * * *

MAGE-A10 TCR: X-scan Specificity Analysis TCR peptide recognition mapping using combinatorial amino acid substitutions G T Y D G M E L L A Q F C A P Q I V R V D N L G F M T K L E M I L G H I C S I A T A I G M Q A K Q P V K A F E C L S S D M P C T M M F H H N V Q P I V H C I P C A W G A K E K T F S M N T P F N V S W L F W G M S S T D H Y C L Q C R Y K S H H E D E A N R I D S I W T C R H F E N V K W H Q N D P P W R F L D E N K Q Y N R W K G Y W G R V Y Y E P V R R T Y Q D W Y >50% >30% >20% >15% >10% NR Tolerated residues at each position cutoff: >10% of native MAGE-A10 peptide response [INDEX PEPTIDE] M 58 SLLMWITQC Spot count 10 -12 10 -11 10 -10 10 -9 10 -8 10 -7 10 -6 0 50 100 150 200 EC 90 10% cut-off

MAGE-A10 TCR SELECTION-MOTIF SEARCH AGAINST PROTEOME 80 hits 5 shared c796 only These 8 peptides were tested with 796 and no responses detected 59 TCR specificity is a key component Tolerated Residues 756 796 3 hits c756 only Peptides contained within X-scan motif (allowing any number of changes)

Specificity and non-specificity Off target (non-specific) Promiscuous recognition (unrelated to target) 60 Potential for off-target non-specific reactions are tested by examining the ability of high-affinity TCRs to react to a panel of normal cell-lines

MAGE-A10 TCR - Screening against normal primary cells MAGE-A10c796T was evaluated in IFN- ELISpot assays against 59 normal primary cells expressing HLA-A2 No increase above background levels with transduced T cells 61 + con n o t a r g e t I M 9 H C P E p i C 3 H C P F 1 H A 1 3 H A 2 H A 5 H M E p i C 2 _ A 2 H M E p i C 3 _ A 2 H M F 1 _ A 2 C M 5 d i f f C M 5 d i f f C M 7 C M 1 2 d i f f C M 1 2 C M 1 3 C M 1 4 C M 1 8 H R C E p 1 H R C E p 5 H R G E C 1 H R G E C 2 R E N 2 R E N 5 H E P 2 H E P 3 H E P 5 H P A S M C 1 H P M E C 2 P F 4 P F 5 H S k M M 2 H S k M M 2 d i f f H S k M M 3 H S k M M 3 d i f f H S k M M 5 H S k M M 5 d i f f S k M C 3 H D F 1 0 H D F 2 H D F 5 N 1 N 9 N H E K 3 N H E K 4 S M C 3 G S M C 1 H V M F 1 H A o 1 5 H A o 1 6 H A o A F 2 H A o A F 4 H C A E C 2 H C A E C 6 H C A E C 7 H C A S M C 7 H D M E C 2 H D M E C 6 H D M E C 7 H O F 2 H O M E C 1 0 100 200 300 I F N g s p o t n u m b e r no T cells ntd T cells (n=3) MAGE-A10 796 T cells (n=3) breast heart kidney liver brain lung skeletal muscle skin intestinal vasculature ovary

Specificity and non-specificity alloreactivity Alternate HLA restriction 62 Potential for alloreactivity are tested by examining the ability of high-affinity TCRs to react to a panel of normal cell-lines with alternate HLA restrictions

MAGE A10 TCR: Alloreactivity assay Panel of EBV-transformed B cells expressing a range of HLA types 67 cell lines expressing a total of 131 different HLA alleles Responses observed in cells expressing HLA-B1501 and HLA-B4601 B1501 B4601 con MAGE-A10 TCR T cells non-transduced control T cells 63 HLA-B1501 and HLA-B4601 become exclusion criteria for clinical trial

64 tcrs with Supra-natural specificity

Tcr from display library has supra-natural affinity Aleksic, M. et al. (2012). Eur J Immunol 10.1002/eji.201242606 Strong Weak Antigen recognition Virus TCRs Naturally-occurring Cancer TCRs 65 MAGE-A10 KD ~ 2 mM Not a natural TCR Identified from proprietary TCR display libraries

Natural tcrs typically can recognize one million different peptides Thymic selection narrows TCR specificity / cross-reactivity spectrum TCR has to recognize approximately 1,000,000 peptides to be positively selected 103 104 105 106 107 108 109 Number of peptides recognized by a TCR Number of TCRs 66 Natural repertoire

tcrs selected from phage libraries can have supra-natural specificity Thymic selection narrows TCR specificity / cross-reactivity spectrum TCR has to recognize approximately 1,000,000 peptides to be positively selected 103 104 105 106 107 108 109 Number of peptides recognized by a TCR Number of TCRs 67 Natural repertoire Supra-natural TCRs e.g. MAGE-A10 TCR

68 Finding the right Targets

The spectrum of potential cancer targets for immunotherapy see http://cancerimmunity.org/peptide/ for a list of tumor antigens reported in the literature Cancer testis antigens expression restricted to immune-privileged tissue e.g. MAGE family / NY-ESO Differentiation antigens tissue restricted expression e.g. Tyrosinase / gp100 Overexpressed antigens overexpressed in tumour cells e.g. WT1 / telomerase Tumour specific antigens not expressed in normal tissues Viral antigens e.g. EBV/HPV Mutated antigens e.g p53 Neo-antigens Ideal Good Depends on tissue Depends on extent of normal tissue expression Ubiquitous antigens expressed in all cells e.g. Her2/neu Unlikely to be suitable 69 Tumor selectivity Prevalence

Peptide target Validation via mass spectrometry Many identified target peptides fail to be presented in vivo Not found by Adaptimmune mass spectrometry Not detected by potent TCRs / T cells Adaptimmune ONLY considers peptides to be validated if detected by mass spectrometry Currently ~ 660,000 unique peptides within our databases 70 PSCA 14-22 (ALQPGTALL) WT1 126-134 (RMFPNAPYL) Telomerase 540-548 (ILAKFLHWL)

recently filed patents on 63 targets 873 peptides Targets / cancer indication 71 0 5 10 15 20 Melanoma AML/B cell malignancies Head and Neck Kidney Ovarian Oesophageal/Gastric Bladder Liver Prostate Pancreas Breast Colon Lung SC Lung Ad Lung Sq

72 Generation 2: making T cells resistant to suppression

ADOPTIVE T CELL : generation 2 Four components to an effective adoptive therapy: T cell must recognize a cancer cell via a guiding receptor The guiding receptor must have two important aspects Affinity Specificity The T cell needs to be resistant to suppression T cell Cancer cell Inhibitory mechanisms of cancer cells make T cells insensitive Make T cells resistant to suppression 73

Overcoming inhibition in the tumour microenvironment Gen2(A) makes T cells insensitive to inhibitory factor x ntd GEN1 TCR GEN2 TCR targets alone INHIBITORY FACTOR X - + - + TARGET 1 TARGET 2 150 100 50 IFNg (pg / mm) 0 74 0 100 200 300 400 I F N - g ( p g / m l ) targets alone ntd TCR Gen2(A)_TCR no target - + - + - + inhibitor X target 1 target 2

Overcoming inhibition in the tumour microenvironment Gen2(A) makes T cells insensitive to inhibitory factor x ntd GEN1 TCR GEN2 TCR targets alone INHIBITORY FACTOR X - + - + TARGET 1 TARGET 2 150 100 50 IL-2 (pg / mm) 0 75 0 50 100 150 I L - 2 ( p g / m l ) targets alone ntd TCR Gen2(A)_TCR inhibitor X no target target 1 target 2 - - - + + +

TCR + FACTOR X TCR alone Gen2(A) TCR maintains enhanced killing in the presence of inhibitors Target cells / mm2 GEN2 TCR alone GEN2 TCR + FACTOR X Overcoming inhibition in the tumour microenvironment Gen2(A) makes T cells insensitive to inhibitory factor x Time (h) 76 0 10 20 30 40 0 200 400 600 800 1000 Time (h) T a r g e t c e l l s / m m 2 TCR + inhibitor X TCR

Overcoming inhibition in the tumour microenvironment Gen2(C) makes T cells insensitive to inhibitory factors Y & Z GEN1 TCR GEN2 TCR 77

Overcoming inhibition in the tumour microenvironment (II) Gen2(C) makes T cells insensitive to inhibitory factors Y & Z Gen2(C) TCR maintains enhanced killing in the presence of inhibitors Target cells / mm2 TCR + inhibitor TCR alone GEN2 TCR inhibitor Time (h) 78

79 Generation 3: enabling T cells to help break cancer immune tolerance

ADOPTIVE T CELL : generation 3 Four components to an effective adoptive therapy: T cell must recognize a cancer cell via a guiding receptor The guiding receptor must have two important aspects Affinity Specificity The T cell needs to be resistant to suppression The T cell (either alone or via other mechanisms) needs to break cancer immune tolerance T cell Cancer cell Epitope spreading Epitope spreading 80

Broadening the immune response Gen3(B) enhances cd40L expression to promote epitope spreading CD40L expression Gen2(B) enhances TCR mediated CD40L upregulation on CD4+ T cells in response to antigen positive targets Normalised events ntd T cells (MFI= 118) T cells + TCR (MFI= 116) T cells + TCR + Gen3(B) (MFI= 313) 81

Generation 2 and generation 3 T cells Several Generation 2 projects that help T cells overcome sensitivity to inhibitory factors in the tumor microenvironment Several Generation 3 projects that enable T cells to promote epitope spreading and therefore have the potential to aid the breaking of tumor immune tolerance First Generation 2 / 3 IND anticipated in 2017 82

Adaptimmune t cell technology TCR affinity optimization crucial for best T cell response to cancer Specificity crucial for lowest toxicity supra-naturally specific TCRs identified from proprietary display libraries Several Generation 2 technologies making T cells resistant to tumour microenvironment inhibitory factors Several Generation 3 technologies enabling T cells to facilitate breaking immune tolerance to tumor 83 summary

Adaptimmune investor and analyst day 2016 Adoptive T cell Therapy: Clinical Activity of NY-ESO-1 in a Solid Tumor April 22, 2016 Stephan Grupp, M.D., Ph. D. Novotny Professor of Pediatrics University of Pennsylvania Perelman School of Medicine

93% CR rate for relapsed/refractory ALL after CTL019 59 r/r pediatric ALL pts: 55 in CR at one mo (93%) median f/u 12 mo Six went to subsequent transplant, one to DLI Six mo RFS: 76% (95%CI:0.65,0.89) 12 mo RFS: 55% (95%CI: 0.42,0.73) No relapses past one year 18 patients in remission beyond one year, 13 without further therapy 85

Two approaches to genetically engineered T cells: CARs and TCRs 86 Sensitive signal amplification derived by evolution Hard to isolate and engineer Low avidity Can targets intracellular proteome (3/4) Requires MHC-I expression and HLA matching on tumor Signal amplification from synthetic biology Easier to make Avidity controllable Targets only surface structures (1/4) MHC independent: off the shelf Toxicity difficult to predict TCR CAR

Ny-eso-1 76% of synovial sarcomas express strong staining, defined as 2-3+ in >50-70% (Lai, Mod Pathol 2012) A TCR recognizing NY-ESO-1 in the context of HLA:A0201 was cloned from a patient with cancer, then modified for higher affinity (Zhao, J Immunol, 2007) A Cancer-testis antigen highly expressed in synovial sarcoma NY-ESO IHC screening on Synovial Sarcomas 87

Two CLINICAL TRIALS OF Adaptimmune s NY-ESO-1 TCR in Synovial Sarcoma Investigator Initiated Trial: The Surgical Branch of the NCI did a study of Adaptimmune s NY-ESO-1-transduced lymphocytes in synovial sarcoma (Cy/Flu + HD IL-2) Partial response in 4 of 6 synovial sarcomas (Robbins et al, JCO 2011) Follow-up report: Objective responses in 11 of 18 synovial cell sarcomas (61%) (Robbins, Clin Can Res 2015) Estimated 3-year OS: 38%; 5-yr OS 14% 88 Adaptimmune Trial: Included changes to improve safety and treatment feasibility Determine response rate without HD IL-2 Use of lentiviral vector Central manufacturing site (GMP) and cryopreserved final product Recent new cohort with cyclophosphamide alone (no fludarabine) Recent additional cohort of NY-ESO-1 low expressors (<2+ in 50%)

Adaptimmune s NY-Eso-1 sarcoma trial schema -3 -4 -2 -1 0 d+2 +4 +7 +10 -5 d-6 Cyclophosphamide 1800mg/m2 IV (+ mesna) Fludarabine 30mg/m2 IV Daily until ANC>1000 Filgrastim 5ucg/kg SQ daily Enrollment Apheresis Manufacturing No IL-2 * 1 x 109/kg (min 1 x 107/kg; max 40x109 total) Supportive care: Infection prophylaxis, observation for CRS, GvHD, f/u safety and efficacy 89 T cell infusion: (Mean=5x109 cells*)

Ny-eso-1 sarcoma study: Cohort 1 Patient NY-ESO Staining (archival tissue) Total Transduced T cells (x109) NY-ESO TCR+ T cells / kg (x106) Best Overall Response 200 2-3+ in >50% 14.4 91.3 SD 201 3+ in 100% 8.3 165.01 CR 202* 3+ in 30% 6.6 69.99 PR 204 2-3+ in 50% 3.8 60.32 PR 205 3+ in ~100% 3.4 62.50 PR 261 3+ >99% 0.72 9.11 SD 206 2+ >50% 0.45 5.51 SD 207 3+ >80% 2.67 25.36 SD 208 3+ >95% 4.84 47.97 PR 209 3+ in ~100% 2.51 27.9 PR 263 3+ >50% 2.51 45.39 PD 230 2-3+ in 100% 7.86 143 PD Mean 4.17 57.4 60% objective response rate in patients treated at target dose 90 *Treated in cohort 1 under a protocol exception

Synovial sarcoma overall survival cohort 1 91 5/12 patients alive 4/2016 1 year survival: 75% 2 year survival: 25% Source: Adaptimmune April 2016 cutoff Median survival: 56 weeks (95% CI: 37, 159)

Synovial Sarcoma Study: All cohorts Incidence (n,%) of SAEs (>1 occurrence) 92 January 2016 cutoff Preferred Term Number of Subjects by Maximum Grade (N=16) All SAEs Related* Fatal Pyrexia 4 (25) 2 (12.5) 0 Cytokine release syndrome 2 (12.5) 2 (12.5) 0 Lymphopenia/Lymphocyte count decreased 2 (12.5) 2 (12.5) 0 Neutropenia 2 (12.5) 2 (12.5) 0 Febrile neutropenia 2 (12.5) 1 (6.3) 0 Thrombocytopenia 2 (12.5) 2 (12.5) 0 Dyspnea 2 (12.5) 1 (6.3) 0 Source: Adaptimmune

PHASE I/II STUDY IN SYNOVIAL SARCOMA Radiographic pseudoprogression and response of lung metastases leading to complete response Baseline: Bilateral miliary metastatic disease C-Reactive Protein Day +2: Pseudoprogression due to immune infiltration Day +101: Complete Response Source: AACR April 2015 93

Clinical response followed by Resection at Progression Mass began to show regrowth ~6 months Surgically resected at 7 months No NY-ESO-1 TCR cells found in tumor Substantial CD4+ T cells 1/22/14 3/3/14 4/7/14 94 No evidence of disease 27 months post NY-ESO-1 T cell infusion; 20 months from surgical resection of metastasis

Near complete response to NY-ESO-1 T Cells of Unresectable Primary Tumor in the knee Complete surgical resection accomplished, no irradiation Local disease remained controlled; patient developed lung metastasis with loss of NY-ESO-1 2 months post NY-ESO TCR T cells 95 Baseline Source: SITC, November 2015

Tumor Shrinkage Over the Course of Several Months Following NY-ESO-1 TCR for Synovial Sarcoma Multiply Recurrent, Unresectable Pulmonary Masses 96 Ongoing PR 400+ days post T cell infusion 2 months Baseline 12 months Source: Adaptimmune

T cells traffic to the site of tumor At resection Remaining tumor was NY-ESO negative 97 NY-ESO expression CD8 Expression Source: Adaptimmune

Clinical responses observed across a spectrum of ny-eso-1 expression 98 Baseline 11-05-13 Month 3 02-18-14 Response in a patient with Low NY-ESO-1 expression Source: Adaptimmune

Clinical responses observed across a spectrum of ny-eso-1 expression 99 Baseline Month 6 Response in a patient with Very High NY-ESO-1 expression Source: Adaptimmune

Ny-eso-1 sarcoma study Durable persistence of NY-ESO-t Days post infusion Days post infusion Vector Copies / g genomic DNA Subjects receiving minimum evaluable dose (>1x109 NY-ESO-1c259T cells) Time of progression LOD LOD Among evaluable subjects, higher peak persistence was observed among responders compared to non responders To date, among responders, NYESO-1 T cells have been detected up to 21 months. 100 Source: November SITC, 2015 10 100 1000 10000 100000 1000000 0 10 19 26 56 90 127 176 265 363 Non - responders 200 207 230 263 10 100 1000 10000 100000 1000000 0 9 16 23 42 59 91 127 162 189 278 540 Responders 201 202 204 205 208 209