ADRIATIC Trial

[Slide 1] Q = AstraZeneca Imfinzi significantly improved overall survival and progression-free survival for patients with limited-stage small cell lung cancer in ADRIATIC Phase III trial PUBLISHED 5 April 2024

[Slide 1] 2024ASCO ANNUAL MEETING --- [Slide 2] ADRIATIC study design Phase 3, randomized, double-blind, placebo-controlled, multicenter, international study (NCT03703297) Stage I-III LS SCLC Durvalumab (stage I/II inoperable) Dual primary endpoints: 1500 mg Q4W WHO PS 0 or 1 N=264 Durvalumab vs placebo - OS Had not progressed following cCRT* N=730 - PFS (by BICR, per RECIST v1.1) PCI* permitted before Placebo Key secondary endpoints: randomization R* Q4W Durvalumab + tremelimumab vs N=265 placebo cCRT components - OS Stratified by: Four cycles of platinum and Disease stage Durvalumab + tremelimumab - PFS (by BICR, per RECIST v1.1) eloposide (three permitted) (i/D V3 III) D 1500 mg Q4W + T 75 mg Q4W for 4 doses, RT: 60-66 Gy QD over 6 weeks PCI (yes vs no) followed by D 1500 mg Q4W Other secondary endpoints: or 45 Gy BID over 3 weeks N=200 OS/PFS landmarks RT must commence no later Safety than end of cycle 2 of CT Treatment until investigator-determined progression or intolerable toxicity, or for a maximum of 24 months "CRT and PCI treatment, if received per local standard of care, must have been completed within 1-42 days prior to randomization. TII disease control was achieved and no additional benefit was expected with an additional cycle of chemotherapy, in the opinion of the investigator. The first 600 patients were randomized in a 111 cato to the 3 treatment arms, subsequent patients were randomized 1:1 to either durvalumab or placebo. BOX (inded interested - aves BID - taly CT. chematherapy D and/or 2024 ASCO PRESENTED - Dr David R. Spigel PO prophyers - PS. performance KA every 8 weeks, #ASCO24 DD - day REDIST Regurst Education Otera . Sold Tunurs ASCO METCH SOCIETY OF DUNICAL ONCOLOGY ANNUAL MEETING - - of - - - - - - for - - 1 RT address T. terminumal WHO, World Health Organization KNOWLEDGE CONQUERS CANCER --- [Slide 3] Overall survival (dual primary endpoint) Median duration of follow up in censored patients: 37.2 months (range 0.1-60.9) 1.0 Durvalumab Placebo (n=264) (n=266) Events, n (%) 115(43.6) 146 (54.9) 0.8 mOS, months (95% CI) 55.9 (37.3-NE) 33.4 (25.5-39.9) 68.0% HR (95% CI) 0.73 (0.57-0.93) 56.5% p-value 0.0104 Probability of os 0.6 58.5% 0.4 47.6% 0.2 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 Time from randomization (months) No. at risk Durvalumab 264 261 248 236 223 207 189 183 172 162 141 110 90 68 51 39 27 19 11 5 1 0 Placebo 266 250 247 231 214 195 175 164 151 143 123 97 80 62 44 31 23 19 8 5 1 0 os was analyzed using a stratified bank test adjusted for receipt of PCI (yes vs no). The significance level for testing os at this interim analysis was 0.01679 (2-sided) at the overall 4.5% level, allowing for strong alpha control across interim and final analysis timepoints 2024 ASCO #ASCO24 ME STATED BY Dr David R. Spigel ASCO AMERICAN IDENTY OF CUNICAL ONCOLDER ANNUAL MEETING - . - of - who - - - - - - - 1 a confidence - m05, medium CS K - - KNOWLEDGE CONQUERS CANCER --- [Slide 4] Progression-free survival* (dual primary endpoint) Median duration of follow up in censored patients: 27.6 months (range 0.0-55.8) 1.0 Durvalumab Placebo (n=264) (n=266) Events, n (%) 139 (52.7) 169 (63.5) 0.8 mPFS, months (95% CI) 16.6 (10.2-28.2) 9.2 (7.4-12.9) HR (95% CI) 0.76 (0.61-0.95) Probability of PFS p-value 0.0161 0.6 48.8% 46.2% 0.4 36.1% 34.2% 0.2 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 No. at risk Time from randomization (months) Durvalumab 264 212 161 135 113 105 101 98 84 78 51 51 33 21 19 10 10 4 4 0 0 0 Placebo 266 208 146 122 100 88 79 76 71 69 47 47 34 23 22 15 14 5 5 0 0 0 "By BICR per RECIST v1.1. PFS - analyzed using a stratted by-ank test adjusted for dhease stage (1/8 vs (II) and receipt of PCI (yes vs no). The significance level for testing PFS - this interim analysis was 0.00184 (2-sided) at the 0.5% level, and ... (2-scled) . the overall 5% level. Statestical signature b PFS - achieved through the recycling multiple testing procedure framework and testing at the 5% (2-sided) alpha level (adjusted for an interim and final analysis). 2024 ASCO #ASCO24 MILIENTED - Dr David R. Spigal ASCO AMERICAN SOCIETY OF CLINICAL ONCOLGOY ANNUAL MEETING - a - - - - - - - - - - mPFS, median PFS KNOWLEDGE CONQUERS CANCER

[Slide 1] BARCELONA congress 2024 ESMO BID and QD RT subgroups - OS BID RT QD RT ITT D (n = 69) P (n = 79) D (n = 195) P (n = 187) D (n = 264) P (п = 266) Median os (95% CI), months NR (NE-NE) 44.8 (29.4-NE) 41.9 (32.0-NE) 26.1 (21.7-36.8) 55.9 (37.3-NE) 33.4 (25.5-39.9) 3-year OS, % 65.8 57.4 53.1 43.3 56.5 47.6 HR (95% CI) 0.68 (0.40-1.14)* 0.72 (0.55-0.96)* 0.73 (0.57-0.93) Multivariable HR (95% CI) 0.71 (0.42-1.18)± 0.73 (0.55-0.96) - BID RT QD RT 1.0 1.0 0.8 0.8 65.8% Probability of os 0.6 57.4% 0.4 Probability of os 0.6 53.1% 0.4 43.3% 0.2 0.2 0 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 Time from randomisation (months) Time from randomisation (months) No. at risk: No. at risk: D, BID 69 68 63 61 59 56 54 53 51 48 42 35 27 18 13 10 5 5 3 2 0 0 D,QD 195 193 185 175 164 151 135 130 121 114 99 75 63 50 38 29 22 14 8 3 1 0 P, BID 79 79 76 73 69 61 57 56 54 53 45 37 32 27 22 14 9 8 4 3 1 0 P,QD 187 181 171 158 145 134 118 108 97 90 78 60 48 35 22 17 14 11 4 2 0 0 *Subgroup HRs and Cls calculated using an unstratified Cox proportional hazards model. ITT HR and Cls calculated using a Cox proportional hazards model stratified by receipt of PCI. Multivariable analysis interaction p-value 0.95.

[Slide 1] BARCELONA ESMO congress 2024 Phase 3 ADRIATIC trial Ongoing, randomised, double-blind, placebo-controlled, multicentre, international study Durvalumab Dual primary endpoints: Stage I-III LS-SCLC N=264 D VS P N=730 (stage I/II inoperable) - PFS, OS WHO PS 0 or 1 Placebo R⁺ Key secondary endpoints: N=266 D+T VS P Had not progressed following cCRT* - PFS,+ OS Stratified by: PCI* permitted before randomisation Durvalumab + tremelimumab Disease stage Other secondary endpoints: (I/II VS III) N=200 PFS/OS landmarks, safety PCI (yes VS no) Treatment until investigator-determined PD or intolerable toxicity, or for a maximum of 24 months At the first interim analysis:¹ Consolidation durvalumab significantly improved the dual primary endpoints of OS and PFS versus placebo; generally consistent treatment benefit across predefined patient subgroups Treatment well tolerated; safety consistent with known safety profile of durvalumab in the post-cCRT setting Durvalumab + tremelimumab arm remained blinded BICR, blinded independent central review; cCRT, concurrent chemoradiotherapy; D, durvalumab; LS-SCLC, 1. Spigel D, et al. J Clin Oncol 2024;42(17 suppl):LBA5. limited-stage small-cell lung cancer, OS, overall survival; P, placebo; PCI, prophylactic cranial irradiation; 'cCRT and PCI treatment, if received per local standard of care, must have been completed within 1-42 days PD, progressive disease; PFS, progression-free survival; R, randomisation; RECIST, Response Evaluation prior to randomisation. The first 600 patients were randomised in a 1:1:1 ratio to the 3 arms; subsequent Criteria in Solid Tumors; T, tremelimumab; WHO PS, World Health Organization performance status. patients were randomised 1:1 to either durvalumab or placebo. PFS assessed by BICR, per RECIST v1.1. --- [Slide 2] BARCELONA congress 2024 ESMO PCI-yes and PCI-no subgroups - OS PCI-yes PCl-no ITT D (n = 142) P (n = 143) D (n = 122) P (n = 123) D (n = 264) P (n = 266) Median os (95% CI), months NR (43.9-NE) 42.5 (33.4-NE) 37.3 (24.3-NE) 24.1 (18.8-31.1) 55.9 (37.3-NE) 33.4 (25.5-39.9) 3-year OS, % 62.1 56.5 50.2 37.3 56.5 47.6 HR (95% CI) 0.75 (0.52-1.07)* 0.71 (0.51-0.99)* 0.73 (0.57-0.93) Multivariable HR (95% CI) 0.72 (0.50-1.03) 0.73 (0.52-1.02) - PCl-yes PCl-no 1.0 1.0 0.8 0.8 62.1% Probability of os 0.6 56.5% Probability of os 0.6 50.2% 0.4 0.4 37.3% 0.2 0.2 0 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 Time from randomisation (months) Time from randomisation (months) No. at risk: No. at risk: D, PCl-yes 142 139 132 127 124 118 110 105 100 93 82 63 51 40 29 23 19 15 8 4 1 0 D, PCI-no 122 122 116 109 99 89 79 78 72 69 59 47 39 28 22 16 8 4 3 1 0 0 P, PCI-yes 143 140 133 129 122 110 100 95 91 89 77 61 48 37 26 20 14 13 5 3 1 0 P, PCI-no 123 120 114 102 92 85 75 69 60 54 46 36 32 25 18 11 9 6 3 2 0 0 *Subgroup HRs and Cls calculated using an unstratified Cox proportional hazards model. TITT HR and Cls calculated using a Cox proportional hazards model stratified by receipt of PCI. CI, confidence interval; NE, not estimable; NR, not reached; yr, year. Multivariable analysis interaction p-value 0.96. --- [Slide 3] BARCELONA ESMO congress 2024 Carboplatin and cisplatin CT subgroups - OS Carboplatin CT Cisplatin CT ITT D (n = 91) P (n = 88) D (n = 173) P (n = 178) D (n = 264) P (n = 266) Median OS (95% CI), months NR (42.5-NE) 33.4 (21.7-NE) 41.9 (27.7-NE) 34.3 (25.4-40.7) 55.9 (37.3-NE) 33.4 (25.5-39.9) 3-year OS, % 65.3 46.7 52.1 48.1 56.5 47.6 HR (95% CI) 0.56 (0.35-0.89)* 0.82 (0.61-1.10)* 0.73 (0.57-0.93) Multivariable HR (95% CI) 0.55 (0.35-0.87)I 0.81 (0.60-1.08): - 1.0 Carboplatin CT 1.0 Cisplatin CT 0.8 0.8 65.3% Probability of os 0.6 Probability of os 0.6 52.1% 0.4 46.7% 0.4 48.1% 0.2 0.2 0 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 Time from randomisation (months) Time from randomisation (months) No. at risk: No. at risk: D, carboplatin 91 90 84 81 77 71 68 66 65 63 55 40 32 23 17 11 8 4 2 1 1 0 D, cisplatin 173 171 164 155 146 136 121 117 107 99 86 70 58 45 34 28 19 15 9 4 0 0 P, carboplatin 88 86 84 77 69 63 57 52 47 45 41 28 22 16 11 8 6 3 1 1 0 0 P, cisplatin 178 174 163 154 145 132 118 112 104 98 82 69 58 46 33 23 17 16 7 4 1 0 *Subgroup HRs and Cls calculated using an unstratified Cox proportional hazards model. TITT HR and Cls calculated using a Cox proportional hazards model stratified by receipt of PCI. Multivanable analysis interaction p-value 0.17. --- [Slide 4] BARCELONA congress 2024 ESMO BID and QD RT subgroups - OS BID RT QD RT ITT D (n = 69) P (n = 79) D (n = 195) P (n = 187) D (n = 264) P (n = 266) Median os (95% CI), months NR (NE-NE) 44.8 (29.4-NE) 41.9 (32.0-NE) 26.1 (21.7-36.8) 55.9 (37.3-NE) 33.4 (25.5-39.9) 3-year OS, % 65.8 57.4 53.1 43.3 56.5 47.6 HR (95% CI) 0.68 (0.40-1.14)* 0.72 (0.55-0.96)* 0.73 (0.57-0.93) Multivariable HR (95% CI) 0.71 (0.42-1.18) 0.73 (0.55-0.96): - BID RT QD RT 1.0 1.0 0.8 0.8 65.8% Probability of os 0.6 57.4% Probability of os 0.6 53.1% 0.4 0.4 43.3% 0.2 0.2 0 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 Time from randomisation (months) Time from randomisation (months) No. at risk: No. at risk: D, BID 69 68 63 61 59 56 54 53 51 48 42 35 27 18 13 10 5 5 3 2 0 0 D, QD 195 193 185 175 164 151 135 130 121 114 99 75 63 50 38 29 22 14 8 3 1 0 79 79 76 73 69 61 57 56 54 53 45 37 32 27 22 14 9 8 4 3 1 0 P, QD 187 181 171 158 145 134 118 108 97 90 78 60 48 35 22 17 14 11 4 2 0 0 P, BID *Subgroup HRs and Cls calculated using an unstratified Cox proportional hazards model. TITT HR and Cls calculated using a Cox proportional hazards model stratified by receipt of PCI. Multivariable analysis interaction p-value 0.95.

[Slide 1] congress BARCELONA 2024 ESMO PCI-yes and PCI-no subgroups - OS PCI-yes PCI-no ITT D (n = 142) P (n = 143) D (n = 122) P (n = 123) D (n = 264) P (n = 266) Median OS (95% CI), months NR (43.9-NE) 42.5 (33.4-NE) 37.3 (24.3-NE) 24.1 (18.8-31.1) 55.9 (37.3-NE) 33.4 (25.5-39.9) 3-year OS, % 62.1 56.5 50.2 37.3 56.5 47.6 HR (95% CI) 0.75 (0.52-1.07)* 0.71 (0.51-0.99)* 0.73 (0.57-0.93) Multivariable HR (95% CI) 0.72 (0.50-1.03) 0.73 (0.52-1.02) - PCl-yes PCl-no 1.0 1.0 0.8 0.8 62.1% Probability of os 0.6 56.5% Probability of os 0.6 50.2% 0.4 0.4 37.3% 0.2 0.2 0 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 Time from randomisation (months) Time from randomisation (months) No. at risk: No. at risk: D, PCl-yes 142 139 132 127 124 118 110 105 100 93 82 63 51 40 29 23 19 15 8 4 1 0 D, PCl-no 122 122 116 109 99 89 79 78 72 69 59 47 39 28 22 16 8 4 3 1 0 0 P. PCl-yes 143 140 133 129 122 110 100 95 91 89 77 61 48 37 26 20 14 13 5 3 1 0 P, PCI-no 123 120 114 102 92 85 75 69 60 54 46 36 32 25 18 11 9 6 3 2 0 0 "Subgroup HRs and Cls calculated using an unstratified Cox proportional hazards model. ITT HR and Cls calculated using a Cox proportional hazards model stratified by receipt of PCI. CI, confidence interval; NE, not estimable; NR, not reached; yr, year. Multivariable analysis interaction p-value 0.96. --- [Slide 2] congress BARCELONA 2024 ESMO Carboplatin and cisplatin CT subgroups - OS Carboplatin CT Cisplatin CT ITT D (n 91) P (n = 88) D (n = 173) P (n = 178) D (n = 264) P (n = 266) Median os (95% CI), months NR (42.5-NE) 33.4 (21.7-NE) 41.9 (27.7-NE) 34.3 (25.4-40.7) 55.9 (37.3-NE) 33.4 (25.5-39.9) 3-year OS, % 65.3 46.7 52.1 48.1 56.5 47.6 HR (95% CI) 0.56 (0.35-0.89)* 0.82 (0.61-1.10)* 0.73 (0.57-0.93)1 Multivariable HR (95% CI) 0.55 (0.35-0.87)1 0.81 (0.60-1.08)I - 1.0 Carboplatin CT 1.0 Cisplatin CT 0.8 0.8 65.3% Probability of os 0.6 Probability of os 0.6 52.1% 0.4 46.7% 0.4 48.1% 0.2 0.2 0 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 Time from randomisation (months) Time from randomisation (months) No. at risk: No. at risk: D. carboplatin 91 90 84 81 77 71 68 66 65 63 55 40 32 23 17 11 8 4 2 1 1 0 D. cisplatin 173 171 164 155 145 136 121 117 107 99 86 70 58 45 34 28 19 15 9 4 0 0 P, carboplatin 68 86 84 77 69 63 57 52 47 45 41 28 22 16 11 8 6 3 1 1 0 0 P, cisplatin 178 174 163 154 145 132 118 112 104 98 82 69 58 46 33 23 17 16 7 4 1 0 "Subgroup HRs and Cls calculated using an unstratified Cox proportional hazards model. ITT HR and Cls calculated using a Cox proportional hazards model stratified by receipt of PCI. Multivariable analysis interaction p-value 0.17. --- [Slide 3] BARCELONA congress 2024 ESMO BID and QD RT subgroups - OS BID RT QD RT ITT D (n = 69) P (n = 79) D (n = 195) P (n = 187) D (n = 264) P (n = 266) Median OS (95% CI), months NR (NE-NE) 44.8 (29.4-NE) 41.9 (32.0-NE) 26.1 (21.7-36.8) 55.9 (37.3-NE) 33.4 (25.5-39.9) 3-year OS, % 65.8 57.4 53.1 43.3 56.5 47.6 HR (95% CI) 0.68 (0.40-1.14)* 0.72 (0.55-0.96)* 0.73 (0.57-0.93) Multivariable HR (95% CI) 0.71 (0.42-1.18) 0.73 (0.55-0.96) - BID RT QD RT 1.0 1.0 0.8 0.8 65.8% Probability of os 0.6 57.4% Probability of os 0.6 53.1% 0.4 0.4 43.3% 0.2 0.2 0 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 Time from randomisation (months) Time from randomisation (months) No. at risk: No. at risk: D, BID 69 68 63 61 59 56 54 53 51 48 42 35 27 18 13 10 5 5 3 2 0 0 D,QD 195 193 185 175 164 151 135 130 121 114 99 75 63 50 38 29 22 14 8 3 1 0 P, BID 79 79 76 73 69 61 57 56 54 53 45 37 32 27 22 14 9 8 4 3 1 0 P,QD 187 181 171 158 145 134 118 108 97 90 78 60 48 35 22 17 14 11 4 2 0 0 "Subgroup HRs and Cls calculated using an unstratified Cox proportional hazards model. ITT HR and Cls calculated using a Cox proportional hazards model stratified by receipt of PCI. Multivariable analysis interaction p-value 0.95

[Slide 1] BARCELONA congress 2024 ESMO Phase 3 ADRIATIC trial subgroup analyses Post-hoc analyses of durvalumab versus placebo in prespecified subgroups defined by PCI use and prior cCRT-related variables PCI/cCRT components (in line with standards of care)* ITT population Durvalumab (п = 264) Placebo (n = 266) PCI delivered before randomisation, as clinically indicated Received PCI, % 54 54 Four cycles of platinum (cisplatin or carboplatin) and etoposide (three permitted*) Carboplatin / cisplatin CT,$ % 34/66 33/67 RT: 60-66 Gy QD over 6 weeks or 45 Gy BID over 3 weeks BID / QD thoracic RT, % 26/74 30/70 Analyses of OS, PFS, and safety with durvalumab VS placebo in subgroups of patients who received: - PCI or no PCI — Carboplatin- or cisplatin-based CT - BID or QD RT Multivariable analyses: for each subgroup, HRs for durvalumab VS placebo were calculated from an unstratified multivariable Cox proportional hazards model with a treatment-by-subgroup (PCI, CT, or RT) interaction term that was adjusted for PCI, CT, RT, time from cCRT to randomisation, response to cCRT, age, sex, WHO PS, and disease stage *The components and delivery of standard of care may vary based on patient characteristics and region. BID, twice daily; CT, chemotherapy; Gy, gray; HR, hazard ratio; TIf disease control was achieved and no additional benefit was expected with an additional cycle of CT, in the opinion of the investigator. ITT, intention-to-treat; QD, once daily; RT, radiotherapy. RT must commence no later than end of cycle 2 of CT. $Based on the first cycle of CT. --- [Slide 2] BARCELONA congress 2024 ESMO PCI-yes and PCl-no subgroups - OS PCI-yes PCI-no ITT D (n = 142) P (n = 143) D (n = 122) P (n = 123) D (n = 264) P (n = 266) Median OS (95% CI), months NR (43.9-NE) 42.5 (33.4-NE) 37.3 (24.3-NE) 24.1 (18.8-31.1) 55.9 (37.3-NE) 33.4 (25.5-39.9) 3-year OS, % 62.1 56.5 50.2 37.3 56.5 47.6 HR (95% CI) 0.75 (0.52-1.07)* 0.71 (0.51-0.99)* 0.73 (0.57-0.93) Multivariable HR (95% CI) 0.72 (0.50-1.03)± 0.73 (0.52-1.02) - PCl-no 1.0 PCl-yes 1.0 0.8 0.8 62.1% Probability of os 0.6 56.5% Probability of os 0.6 50.2% 0.4 0.4 37.3% 0.2 0.2 0 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 Time from randomisation (months) Time from randomisation (months) No. at risk: No. at risk: D, PCI-yes 142 139 132 127 124 118 110 105 100 93 82 63 51 40 29 23 19 15 8 4 1 0 D, PCl-no 122 122 116 109 99 89 79 78 72 69 59 47 39 28 22 16 8 4 3 1 0 0 P, PCI-yes 143 140 133 129 122 110 100 95 91 89 77 61 48 37 26 20 14 13 5 3 1 0 P, PCI-no 123 120 114 102 92 85 75 69 60 54 46 36 32 25 18 11 9 6 3 2 0 0 "Subgroup HRs and Cls calculated using an unstratified Cox proportional hazards model. TITT HR and Cls calculated using a Cox proportional hazards model stratified by receipt of PCI. CI, confidence interval; NE, not estimable; NR, not reached; yr, year. Multivariable analysis interaction p-value 0.96. --- [Slide 3] BARCELONA congress 2024 ESMO Carboplatin and cisplatin CT subgroups - OS Carboplatin CT Cisplatin CT ITT D (n = 91) P (n = 88) D (n = 173) P (n = 178) D (n = 264) P (n = 266) Median OS (95% CI), months NR (42.5-NE) 33.4 (21.7-NE) 41.9 (27.7-NE) 34.3 (25.4-40.7) 55.9 (37.3-NE) 33.4 (25.5-39.9) 3-year OS, % 65.3 46.7 52.1 48.1 56.5 47.6 HR (95% CI) 0.56 (0.35-0.89)* 0.82 (0.61-1.10)* 0.73 (0.57-0.93) Multivariable HR (95% CI) 0.55 (0.35-0.87)± 0.81 (0.60-1.08) - Carboplatin CT Cisplatin CT 1.0 1.0 0.8 0.8 65.3% Probability of os 0.6 Probability of os 0.6 52.1% 0.4 46.7% 0.4 48.1% 0.2 0.2 0 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 Time from randomisation (months) Time from randomisation (months) No. at risk: No. at risk: D. carboplatin 91 90 84 81 77 71 68 65 65 63 55 40 32 23 17 11 8 4 2 1 1 0 D. cisplatin 173 171 164 155 146 136 121 117 107 99 86 70 58 45 34 28 19 15 9 4 0 0 P, carboplatin 68 $6 84 77 69 53 57 52 47 45 41 28 22 16 11 8 6 3 1 1 0 0 P, cisplatin 178 174 163 154 145 132 118 112 104 98 82 69 58 46 33 23 17 16 7 4 1 0 "Subgroup HRs and Cls calculated using an unstratified Cox proportional hazards model. TITT HR and Cls calculated using a Cox proportional hazards model stratified by receipt of PCI. Multivariable analysis interaction p-value 0.17. --- [Slide 4] BARCELONA congress 2024 ESMO BID and QD RT subgroups - OS BID RT QD RT ITT D (n = 69) P (n = 79) D (n = 195) P (n = 187) D (n = 264) P (n = 266) Median os (95% CI), months NR (NE-NE) 44.8 (29.4-NE) 41.9 (32.0-NE) 26.1 (21.7-36.8) 55.9 (37.3-NE) 33.4 (25.5-39.9) 3-year OS, % 65.8 57.4 53.1 43.3 56.5 47.6 HR (95% CI) 0.68 (0.40-1.14)* 0.72 (0.55-0.96)* 0.73 (0.57-0.93) Multivariable HR (95% CI) 0.71 (0.42-1.18)± 0.73 (0.55-0.96) - BID RT QD RT 1.0 1.0 0.8 0.8 65.8% Probability of os 0.6 57.4% 0.4 Probability of os 0.6 53.1% 0.4 43.3% 0.2 0.2 0 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 Time from randomisation (months) Time from randomisation (months) No. at risk: No. at risk: D, BID 69 68 63 61 59 56 54 53 51 48 42 35 27 18 13 10 5 5 3 2 0 0 D,QD 195 193 185 175 164 151 135 130 121 114 99 75 63 50 38 29 22 14 8 3 1 0 P, BID 79 79 76 73 69 61 57 56 54 53 45 37 32 27 22 14 9 8 4 3 1 0 P,QD 187 181 171 158 145 134 118 108 97 90 78 60 48 35 22 17 14 11 4 2 0 0 "Subgroup HRs and Cls calculated using an unstratified Cox proportional hazards model. ITT HR and Cls calculated using a Cox proportional hazards model stratified by receipt of PCI. Multivariable analysis interaction p-value 0.95.

[Slide 1] ADRIATIC study design Phase 3, randomized, double-blind, placebo-controlled, multicenter, international study (NCT03703297) Stage I-III LS-SCLC Durvalumab Dual primary endpoints: (stage I/II inoperable) 1500 mg Q4W WHO PS 0 or 1 Durvalumab VS placebo N=264 - OS Had not progressed following cCRT* N=730 - PFS (by BICR, per RECIST v1.1) PCI* permitted before Placebo Key secondary endpoints: randomization R+ Q4W Durvalumab + tremelimumab VS N=266 placebo cCRT components - OS Stratified by: Four cycles of platinum and Disease stage Durvalumab + tremelimumab - PFS (by BICR, per RECIST v1.1) etoposide (three permitted1) (1/11 VS III) D 1500 mg Q4W + T 75 mg Q4W for 4 doses, RT: 60-66 Gy QD over 6 weeks PCI (yes VS no) followed by D 1500 mg Q4W Other secondary endpoints: or 45 Gy BID over 3 weeks N=200 OS/PFS landmarks RT must commence no later Safety than end of cycle 2 of CT Treatment until investigator-determined progression or intolerable toxicity, or for a maximum of 24 months *cCRT and PCI treatment, if received per local standard of care, must have been completed within 1-42 days prior to randomization. TIf disease control was achieved and no additional benefit was expected with an additional cycle of chemotherapy, in the opinion of the investigator. The first 600 patients were randomized in a 1:1:1 ratio to the 3 treatment arms; subsequent patients were randomized 1:1 to either durvalumab or placebo. BICR, blinded independent central review, BID, twice daily; CT. chemotherapy, D, durvalumab; 2024 ASCO PRESENTED BY: Dr David R. Spigel PCI, prophylactic cranial irradiation; PS, performance status; Q4W, every 4 weeks; ASCO AMERICAN SOCIETY OF #ASCO24 QD, once daily; RECIST, Response Evaluation Criteria in Solid Tumors; CLINICAL ONCOLOGY ANNUAL MEETING Presentation is property of the author and ASCO Permission required for reuse, contact permissions@asco.org RT, radiotherapy; T. tremelimumab; WHO, World Health Organization KNOWLEDGE CONQUERS CANCER --- [Slide 2] Overall survival (dual primary endpoint) Median duration of follow up in censored patients: 37.2 months (range 0.1-60.9) 1.0 Durvalumab Placebo (n=264) (n=266) Events, n (%) 115 (43.6) 146 (54.9) 0.8 mOS, months (95% CI) 55.9 (37.3-NE) 33.4 (25.5-39.9) 68.0% HR (95% CI) 0.73 (0.57-0.93) 56.5% p-value 0.0104 Probability of OS 0.6 58.5% 0.4 47.6% 0.2 0 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 No. at risk Time from randomization (months) Durvalumab 264 261 248 236 223 207 189 183 172 162 141 110 90 68 51 39 27 19 11 5 1 0 Placebo 266 260 247 231 214 195 175 164 151 143 123 97 80 62 44 31 23 19 8 5 1 0 OS was analyzed using a stratified log-rank test adjusted for receipt of PCI (yes vs no). The significance level for testing OS at this interim analysis was 0.01679 (2-sided) at the overall 4.5% level, allowing for strong alpha control across interim and final analysis timepoints. 2024 ASCO #ASCO24 PRESENTED BY: Dr David R. Spigel ASCO AMERICAN SOCIETY OF CLINICAL ONCOLOGY ANNUAL MEETING Presentation is property of the author and ASCO Permission required for reuse, contact permissions@asco.org CI, confidence interval mos, median OS; NE, not estimable KNOWLEDGE CONQUERS CANCER --- [Slide 3] Pneumonitis/radiation pneumonitis Pneumonitis or radiation Durvalumab Placebo pneumonitis (grouped terms*), n (%) (n=262) (n=265) Any grade 100 (38.2) 80 (30.2) Maximum grade 3/4 8 (3.1) 7 (2.6) Leading to death 1 (0.4) 0 Leading to treatment discontinuation 23 (8.8) 8 (3.0) "Includes the preferred terms of immune-mediated lung disease, interstitial lung disease, pneumonitis, radiation fibrosis - lung, and radiation pneumonitis. Events are included irrespective of etiology and AE management 2024 ASCO #ASCO24 PRESENTED-BY: Dr David R. Spigel ASCO AMERICAN SOCIETY OF CLINICAL ONCOLOGY ANNUAL MEETING Presentation . property of THE author and ASCO Permission - for - contact permissions@ess.org KNOWLEDGE CONQUERS CANCER --- [Slide 4] Anti-PDL1 in Locally Advanced Lung Cancer (vs. ES-SCLC) Global approvals of Durvalumab PACIFIC — 71 independent regions CASPIAN - 69 independent regions* Limited Stage-SCLC PACIFIC (NSCLC) Extensive Stage-SCLC ~23 mo improvement in ~18 mo improvement in ~2 mo improvement in Overall Survival Overall Survival Overall Survival ADRIATIC (Interim) PACIFIC (5-year followup) CASPIAN IMpower133 (cCRT durvalumab vs Placebo) (cCRT -> durvalumab VS Placebo) (EP +/- Durva) (EP-Atezo vs. EP-Placebo) mOS -- 55.9 VS. 33.4 mOS -- 47.5 VS. 29.1 mOS - 13.0mo VS. 10.3mo mOS — 12.3mo vs 10.3mo (HR 0.73) (HR 0.72) (HR 0.73) (HR 0.77) Spigel et al, JCO 2022 Paz-Ares et al, Lancet 2019 Horn et al, NEJM 2018 Global map Rachel Davidowitz, Ph.D., MD. Anderson Cancer Center "CASPIAN approved in all 71 regions except Costa Rica and Cuba, EP (etoposide-platinum) 2024 ASCO #ASCO24 PRESENTED BY: Lauren Averett Byers, MD, MSc @LaurenByersMD ASCO AMERICAN SOCIETY OF CLINICAL ONCOLOGY ANNUAL MEETING Presentation is property of the author and ASCO Permission required for reuse, contact permissions@asco.org KNOWLEDGE CONQUERS CANCER