AEGEAN Trial

[Slide 1] AACR ANNUAL Schema of the AEGEAN Trial American Association MEETING for Cancer Research" 2023 APRIL 14-19 #AACR23 Study population Durvalumab 1500 mg IV + Treatment-naive platinum-based CT Surgery 5 Durvalumab 1500 mg IV Q3W for 4 cycles Q4W for 12 cycles ECOG PS 0 or 1 Resectable NSCLC* Randomization stratified by: (stage IIA-IIIB[N2]; AJCC 8th ed) R Disease stage (II vs III) 1:1 PD-L1 expression (≥1% vs <1%) Lobectomy, sleeve resection, or bilobectomy as planned surgery* Placebo IV + Confirmed PD-L1 status Q3W for 4 cycles Surgery Placebo IV platinum-based CT Q4W for 12 cycles No documented EGFR/ALK N=802 aberrations* randomized Endpoints: All efficacy analyses performed on a modified population that excludes patients with documented EGFR/ALK aberrations Primary: Key secondary: pCR by central lab (per IASLC 2020') MPR by central lab (per IASLC 2020') EFS using BICR (per RECIST v1.1) DFS using BICR (per RECIST v1.1) OS --- [Slide 2] Does the AEGEAN Define A New Standard of Care for AACR ANNUAL American Association MEETING for Cancer Research' the Treatment of NSCLC 2023 APRIL 14-19 #AACR23 YES This represents a second and even larger randomized phase III trial that now demonstrates the value of immunotherapy in the neoadjuvant setting The Incorporation of Adjuvant Therapy Is shown to be feasible and generally safe. The data are early and additional trial maturity is needed to better understand the benefit (if any) of the extra adjuvant therapy We await the overall survival results

[Slide 1] AACR ANNUAL American Association MEETING for Cancer Research 2023 APRIL 14-19 #AACR23 AEGEAN: A Phase 3 Trial of Neoadjuvant Durvalumab + Chemotherapy Followed by Adjuvant Durvalumab in Patients with Resectable NSCLC John V. Heymach¹, David Harpole², Tetsuya Mitsudomi³, Janis M. Taube⁴, Gabriella Galffy⁵, Maximilian Hochmair⁶, Thomas Winder⁷, Ruslan Zukov⁸, Gabriel Garbaos⁹, Shugeng Gao¹⁰, Hiroaki Kuroda¹¹, Jian You¹², Kang-Yun Lee¹³, Lorenzo Antonuzzo¹⁴, Mike Aperghis¹⁵, Gary J. Doherty¹⁵, Helen Mann¹⁵, Tamer M. Fouad¹⁶, Martin Reck¹⁷ 1Department of Thoracic/Head and Neck Medical Oncology, The University of Texas, M.D. Anderson Cancer Center, Houston, Texas, USA; 2Department of Surgery, Duke University Medical Center, Durham, North Carolina, USA; Division of Thoracic Surgery, Department of Surgery, Kindai University Faculty of Medicine, Osaka-Sayama, Japan; "Bloomberg-Kimmel Institute for Cancer Immunotherapy, Johns Hopkins Kimmel Cancer Center, Baltimore, Maryland, USA; Pest County Pulmonology Hospital, Törökbálint, Hungary; 6Department of Respiratory and Critical Care Medicine, Karl Landsteiner Institute of Lung Research and Pulmonary Oncology, Klinik Floridsdorf, Vienna, Austria; Department of Hematology, Oncology, Gastroenterology and Infectiology, Landeskrankenhaus Feldkirch, Feldkirch, Austria; 8Krasnoyarsk State Medical University, Krasnoyarsk, Russia; Fundacion Estudios Clinicos, Santa Fe, Argentina; 10Thoracic Surgery Department, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China; "Department of Thoracic Surgery, Aichi Cancer Center Hospital, Aichi, Japan; 12Department of Lung Cancer, Tianjin Medical University Cancer Institute and Hospital, National Clinical Research Center for Cancer, Tianjin's Clinical Research Center for Cancer, Key Laboratory of Cancer Prevention and Therapy, Tianjin, China; 13Department of Internal Medicine, Shuang Ho Hospital, Taipei Medical University, New Taipei City, Taiwan; 14Clinical Oncology Unit, Careggi University Hospital, Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy; 1⁵AstraZeneca, Cambridge, UK; 16AstraZeneca, New York, NY, USA; 17Lung Clinic Grosshansdorf, Airway Research Center North, German Center for Lung Research, Grosshansdorf, Germany

[Slide 1] EFS by Pathologic Response KEYNOTE-671 CheckMate 816 100 Chemotherapy (pCR) With PCR HR 0.33 (95% CI, 0.09-1.22) 100 80 90 Pembro, with pCR Nivolumab chemotherapy (pCR) 80 60 70 Placebo, with pCR Nivolumab + 60 chemotherapy (no pCR) 50 Pembro, without pCR 40 40 Chemotherapy (no pCR) 30 Placebo, without pCR Without pCR 20 20 HR 0.69 (95% CI, 0.55-0.85) 10 0 o 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 0 6 12 18 24 30 36 42 48 54 Months Months No. Risk No risk Nivolumab chemotherapy (pCR) 43 43 41 40 40 40 40 35 32 19 14 6 3 2 0 72 59 46 33 15 Chemotherapy (UCP) 4 4 4 4 4 4 4 4 4 3 2 2 2 1 o 258 177 126 Nivolumab chemotherapy (no pCR) 136 108 95 84 78 67 62 52 42 22 20 7 3 0 12 10 0 280 171 114 Chemotherapy (no pCR) 175 140 122 105 90 79 71 57 48 23 22 11 9 3 0 Wakelee H. et al, ASCO Annual Meeting, 2023; Forde PM, et al, N Engl J Med. 2022 May 26;386(21):1973-1985 2023 ASCO #ASCO23 PRESENTED BY: Mark M Awad, MD, PhD. Dana-Farber Cancer Institute, Boston, MA ASCO AMERICAN SOCIETY OF CLINICAL ONCOLOGY ANNUAL MEETING Presentation - property of the author ASCO KNOWLEDGE CONQUERS CANCER

[Slide 1] MEDICINE OF 1928 1812 NEW ENGLAND The NEW ENGLAND JOURNAL JOURNAL of MEDICINE Free full text is available with an account for a limited time. Create a free account now. Already have an account? Sign in. ORIGINAL ARTICLE FREE PREVIEW Perioperative Durvalumab for Resectable Non-Small- Cell Lung Cancer John V. Heymach, M.D., Ph.D., David Harpole, M.D., Tetsuya Mitsudomi, M.D., Ph.D., Janis M. Taube, M.D., et al., for the AEGEAN Investigators* October 23, 2023 DOI: 10.1056/NEJMoa2304875

[Slide 1] CheckMate 816: 3-y efficacy/safety update and biomarker analyses CheckMate 816 study designa Key eligibility criteria Newly diagnosed, resectable, stage IB (> 4 cm)-IIIA NSCLC NIVO 360 mg Q3W (per AJCC TNM 7th edition) N = 358 + Surgery ECOG PS 0-1 chemod Q3W (3 cycles) Radiologic (within Optional R No known sensitizing EGFR restaging 6 weeks adjuvant Follow-up 1:1 mutations or ALK alterations post- chemo treatment) and/or RT Stratified by Chemoe Q3W (3 cycles) stage (IB-II vs IIIA), PD-L1b (≥ 1% vs < 1%c), and sex Primary endpoints Secondary endpoints Exploratory analyses pCR by BIPR MPR by BIPR EFS by surgical outcomes EFS by BICR os pCR and EFS by 4-gene TTDM inflammatory signature score Database lock date: October 14, 2022. Minimum/median follow-up: 32.9/41.4 months. From The New England Journal of Medicine, Forde PM, et al, Neoadjuvant nivolumab plus chemotherapy in resectable lung cancer, 2022;386:1973-1985. Copyright © 2022 Massachusetts Medical Society. Adapted with permission from Massachusetts Medical Society.*NCT02998528. Determined by the PD-L1 IHC 28-8 pharmDx assay (Dako). Included patients with PD-L1 expression status not evaluable and indeterminate. Nonsquamous: pemetrexed + cisplatin or paclitaxel + carboplatin; squamous: gemcitabine + cisplatin or paclitaxel + carboplatin. <Vinorelbine + cisplatin, docetaxel + cisplatin, gemcitabine + cisplatin (squamous only), pemetrexed + cisplatin (nonsquamous only), or paclitaxel + carboplatin. --- [Slide 2] CheckMate 816: 3-y efficacy/safety update and biomarker analyses NIVO + chemo Chemo (n = 179) (n = 179) 100 Median EFS, mo NR 21.1 (95% CI) (31.6-NR) (14.8-42.1) 77% HR (95% CI) 0.68 (0.49-0.93) 80 65% 60 57%b 64% EFS (%) NIVO + chemo 40 47% 43%c Chemo 20 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 No. at risk Months from randomization NIVO + chemo 179 152 136 125 119 108 104 100 97 94 88 69 57 38 20 13 6 5 0 Chemo 179 146 128 110 95 84 79 72 67 62 60 48 39 27 15 13 4 4 0 Minimum/median follow-up: 32.9/41.4 months. "Exploratory analysis. Time from randomization to any disease progression precluding surgery, disease progression/recurrence after surgery, progression in patients without surgery, or death due to any cause per BICR. Patients who received subsequent therapy were censored at the last evaluable tumor assessment on or prior to the date of subsequent therapy. 95% Cls for 3-year EFS rates: 48-64; c35-51. --- [Slide 3] Stratification: Pathological evaluation of Disease stage (II VS III) surgical specimen by PD-L1 TC expression status (<1% VS >1%) Central Review Primary endpoints Q3W X 4 cycles pCR Q4W X 12 cycles EFS Resectable NSCLC Durvalumab + Stage IIA-select IIIB Surgery Durvalumab Secondary endpoints EGFR wt / ALK wt platinum-based chemotherapy mPR Planned for lobectomy, R DFS bilobectomy, or sleeve OS resection 1:1 Placebo + pCR, mPR, EFS, DFS, OS (PD-L1 Surgery Placebo (N = 800) platinum-based chemotherapy TC >1% group) HRQoL/PRO Pharmacokinetics Immunogenicity --- [Slide 4] AACR ANNUAL EFS using RECIST v1.1 (BICR) (mITT) Amer can Association MEETING for Cancer Research First planned interim analysis of EFS 2023 APRIL 14-19 #AACR23 D arm PBO arm 1.0 No. events / no. patients (%) 98/366 (26.8) 138/374 (36.9) mEFS, months (95% CI) NR (31.9-NR) 0.9 25.9 (18.9-NR) Stratified HR* (95% CI) 0.68 (0.53-0.88) 0.8 73.4% Stratified log-rank P-value 0.003902 0.7 63.3% Probability of EFS 0.6 64.5% 0.5 52.4% 0.4 0.3 Median follow-up (range) in censored 0.2 patients: 11.7 months (0.0-46.1) 0.1 EFS maturity: 31.9% + Censored 0.0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 Time from randomization (months) No. at risk: D arm 366 336 271 194 140 90 78 50 49 31 30 14 11 3 1 1 0 PBO arm 374 339 257 184 136 82 74 53 50 30 25 16 13 1 1 0 0 NT Nov 10,2022.EFS or (D) death from is defined any cause. as time "HR from <1 nandomization Incrs the D am to the versus earliest the of PBO (A) am progressive Median and disease (PO) that precludes surgery; (B) PO discovered and reported by the investigator upon attempting surgery that prevents completion of surgery, local/distant recurrence BICR per fication factors disease stage (11 vs B) and POLI expression Status va landmark estimates calculated using the Kaplan-Meier method HR calculated using a stratified Cox proportional hazards model, and value calculated (C) a using rank lest (<1% 2111). Significance boundary 0.009899 (based on total 5% alpha), calculated using a Lan DeMets alpha spending function with O'Brien Floming boundary mEF8, median EFS; using NR, stratified not reached log

[Slide 1] A 100 90 Median 80 76.1 No. of Event-free Survival 70 63.8 Patients Event-free Survival (%) (95% CI) Nivolumab plus 60 mo 63.4 chemotherapy 50 Nivolumab plus 179 31.6 (30.2-NR) Chemotherapy 40 45.3 Chemotherapy 179 20.8 (14.0-26.7) 30 Chemotherapy alone Alone 20 Hazard ratio for disease progression, disease recurrence, or death, 0.63 10 (97.38% CI, 0.43-0.91) 0 P=0.005 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 Months No. at Risk Nivolumab plus chemotherapy 179 151 136 124 118 107 102 87 74 41 34 13 6 3 0 Chemotherapy alone 179 144 126 109 94 83 75 61 52 26 24 13 11 4 0 A Event-free Survival 100 90 80 Event-free Survival (%) 70 60 50 Pembrolizumab group 40 30 Placebo group 20 10 0 0 6 12 18 24 30 36 42 48 54 Months No. at Risk Pembrolizumab group 397 330 236 172 117 72 42 11 0 0 Placebo group 400 294 183 124 74 38 24 9 1 0 AACR ANNUAL EFS using RECIST v1.1 (BICR) (mITT) American Association MEETING Cancer Research 2023 First planned interim analysis of EFS APRIL 14-19 #AACR23 D arm PBO arm No. events / no. patients (%) 98/366 (26.8) 138/374 (36.9) 1.0 mEFS, months (95% CI) NR (31.9-NR) 25.9 (18.9-NR) 0.9 Stratified HR* (95% CI) 0.68 (0.53-0.88) 0.8 73.4% Stratified log-rank P-value 0.003902 0.7 63.3% Probability of EFS 0.6 64.5% 0.5 52.4% 0.4 0.3 Median follow-up (range) in censored patients: 11.7 months (0.0-46.1) 0.2 EFS maturity: 31.9% 0.1 Censored 0.0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 Time from randomization (months) No. at risk: D arm 366 336 271 194 140 90 78 50 49 31 30 14 11 3 1 1 0 PBO arm 374 339 257 184 136 82 74 53 50 30 25 16 13 1 1 0 0 DCO Nov 10. 2022 EFS is defined as time from randomization b be earliestot (A) progressive disease (PO) hat precludes surgery, (8) Discovered and reported by the investigator upon attempting surgey hat prevents completion d surgery: (C) local/distant recurrence using BICRper RECIST death from any cause "HR tavors am versus the PBO am Median and landmark estimates calculated using the Kaplan-Meier method HR calculated using strattled Cox proportional hazards mode and Pealue calculated using strattied log rank test Stratification factors: disease stage and POLI ex presson status (<1% $21%) Significance boundary *0 009899 (based on total 5% alpha), calculated using Lan-DeMes alpha spending function with D'Brien Reming boundary mEFS, median EFS NR, notreached

[Slide 1] 14 Summary of Surgical Procedure Toripalimab +Chemo Placebo + Chemo (N=202) (N=202) No surgery performed n(%) 36 (17.8) 54 (26.7) Patient underwent surgery n(%) 166 (82.2) 148 (73.3) R0 resection n(%*) 159 (95.8) 137 (92.6) 95% CI 91.5, 98.3 87.1, 96.2 Differences between arms 3.2 95% CI -2.0, 8.4 *percent of R0 resection is based on numbers of patients underwent surgery

[Slide 1] AACR ANNUAL AEGEAN: a phase 3, global, randomized, double-blind, American Association MEETING for Cancer Research 2023 placebo-controlled study APRIL 14-19 #AACR23 Study population Durvalumab 1500 mg IV + Treatment-naive platinum-based CT NOT Surgery Durvalumab 1500 mg IV Q3W for 4 cycles Q4W for 12 cycles ECOG PS 0 or 1 Resectable NSCLC* Randomization stratified by: (stage IIA-IIIB[N2]: AJCC 8th ed) R Disease stage (II vs III) 1:1 PD-L1 expression (>1% vs <1%) Lobectomy, sleeve resection, or bilobectomy as planned surgery* Placebo IV + Confirmed PD-L1 status platinum-based CT ADM Surgery Placebo IV Q4W for 12 cycles No documented EGFR/ALK N=802 Q3W for 4 cycles aberrations* randomized Endpoints: All efficacy analyses performed on a modified population that excludes patients with documented EGFR/ALK aberrations¹ Primary: Key secondary: pCR by central lab (per IASLC 2020') MPR by central lab (per IASLC 2020') EFS using BICR (per RECIST v1.1) DFS using BICR (per RECIST v1.1) OS ? Support inded while enrollment was ongoing to exclude (1) potients with tumors dissified 33 T4 for any reason other than NOW (2) patients with planned pneumonectomies, and (3) patients with documented EGFR/ALK aberrations - or amunohistochenistry assay Choice of CT regimen determined by histology and at the investigator's discretion For non-squancus cisplatin permetresed or carboplatin permetrexed For squarious carboplatin pacitared or cesplatin gemotabine (or carboplatin gemoitabine for patients who have comorbidities or who are unable to tolerate displatin per the investigator's judgment) Post-operative radiotherapy (PORT) AG5 permitted where indicated per local guidance All efficacy analyses reported in he presentation were performed on the mill population which includes Mi randomized patients who did not have documented EGFR/ALK aberrations AJCC. American Joint Committee on Cancer BICR blinded independent central review, DFS deease thee survival EFS. event tree survival miTT. modified intert-to-treat MPR. major pethologic response pCR pathologic complete response Travis WD. et at J Thoras Oncol 2020,15,709-40 --- [Slide 2] AACR ANNUAL American Association MEETING for Cancer Research 2023 Baseline characteristics and planned treatment (mITT) APRIL 14-19 . #AACR23 Baseline characteristics were largely D arm PBO arm balanced between the study arms Characteristics* (N=366) (N=374) Age Median (range). years 65.0 (30-88) 65.0 (39-85) >75 years, % 12.0 9.6 Male 68.9 74.3 Sex, % The planned neoadjuvant CT doublet Female 31.1 25.7 0 68.6 68.2 regimen was carboplatin-based for ECOG PS, % 1 31.4 31.8 >70% of patients Asian 39.1 43.9 Race', % White 56.3 51.1 Other 4.6 5.1 Asia 38.8 43.6 Europe 38.5 37.4 Region, % North America 11.7 11.5 D arm PBO arm South America 10.9 7.5 TNM classification+ (N=366) (N=374) Current 26.0 25.4 T1 12.0 11.5 Smoking status, % Former 60.1 59.6 Primary T2 26.5 28.9 Never 13.9 15.0 tumor, % T3 35.0 34.5 II 28.4 29.4 Disease stage T4 26.5 25.1 IIIA 47.3 44.1 (AJCC 8th ed.), % NO 30.1 27.3 IIIB 24.0 26.2 Regional lymph N1 20.5 23.3 Squamous 46.2 51.1 nodes, % N2 49.5 49.5 Histology, % Non-squamous 53.6 47.9 TC <1% 33.3 33.4 PD-L1 expression, % TC 1-49% 36.9 38.0 TC >50% 29.8 28.6 DCO Nov 10. 2022 "Characteristics with missing/other responses any histology (0.3% in the D arm and 1 1%in P80 Planned neoadjuvant Cisplatin 27.3 25.7 am had other histology) and deease stage 0.3% in D am had stage M disease and 0 3% in the PBO am had stage III [NOS] disease, 33 reported per the electronic case report form (eCRF) TAII patients wayng M) except one patient in platinum agent, % Carboplatin 72.7 74.3 the D am who was classified as M1 (NOS) Race was reported per the eCRF NOS not otherwise specified TC. tumor cells --- [Slide 3] AACR ANNUAL EFS using RECIST v1.1 (BICR) (mITT) Amorication MEETING for Concer Research 2023 First planned interim analysis of EFS APRIL 14-19 #AACR23 D arm PBO arm No. events 1 no. patients (%) 96/366 (26.8) 138/374 (36.9) 1.0 mEFS. months (95% CI) NR (31.9-NR) 25.9 (18.9-NR) 0.9 Stratified HR* (95% CI) 0.68 (0.53-0.88) 0.8 Stratified log-rank Pivalue 0.003902 73.4% 0.7 63.3% Probability of EFS 0.6 64.5% 0.5 52.4% 0.4 0.3 Median follow-up (range) in censored patients: 11.7 months (0.0-46.1) 0.2 EFS maturity 31.9% 0.1 Censored 0.0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 Time from randomization (months) No. at risk: D arm 366 336 271 194 140 90 78 50 49 31 30 14 11 3 1 1 0 PBO arm 374 339 257 184 136 82 74 53 50 30 25 16 13 1 1 0 0 000 - Nov 10 2022 FFR a defined M the tom nandomaration - the extest of (*) progremie classes PD) that prociedes surgery (R)PD discovers and inported by to aveatigator apon lurger) that prevents completion of surgery, A incaliduations recurrence wing MOR per RECENT #1 R or (7) death In any cause HR at has the D IT - the PERO - Medium and intrant esteriates calculated useng the Saplan- Moter method HR calculated using a strattled Dex properieral harands model, and Pivalue calculated using . strattled leg - test - Nation Bricase stage 1 vs 060 PD41 expression KTN with Significance Doundary 0.009899 (bled on NGA 9% and - wing a an-DeMents are spending fanetion will O'Bren boardary WEFR Fredian EF8 NR not reached --- [Slide 4] AACR Pathologic response per IASLC 2020 methodology* (mITT) American Association MEETING for Cancer Research 2023 Final analysis APRIL 14-19 #AACR23 pCR (central lab) MPR (central lab) Difference = 21.0% 40 40 (95% CI: 15.1-26.9) 30 30 33.3 Difference = 13.0% pCR rate (%) (95% Cl: 8.7-17.6)* MPR rate (%) P-value = 0.000002 based on interim analysis (n=402)* 20 20 P-value = 0.000036 17.2 based on interim analysis (n=402) 10 10 12.3 4.3 0 0 D arm PBO arm D arm PBO arm (N=366) (N=374) (N=366) (N=374) "Using ASLC recommendations for pathologic assessment of response to therapy including gross assessment and processing of tumor bed (Travis WD et al J Thorac Oncol 2020; 709-40) pCR a lack of any viable tunor cells after complete evaluation of the resected lung cancer specimen and all sampled regional lymph nodes MPR = less than or equal to 10% viable tumor cells in ung primary tumor after complete evaluation of the resected lung cancer specimen To be eligible for pathologic assessment patients needed to have received three cycles of neceduvant study Tx per protocol Patients who were not evaluable were classified 3) non-responders TCls calculated by stratified Mettinen and Nummer method No formal statistical Issting was performed at the pCR final analysis (DCO Nov 10. 2022 n=740 data shown] Statistical significance was achieved at the interim pCR analysis (DCO Jan 14. 2022: n=402 P-value for pCR/MPR calculated using 3 stratified Cochran-Mantel Haenszel test with a significance boundary 0 000082 calculated using a Lan-DeMets alpha spending function with O'Brien Fleming boundary)