Slides
Page 1
gsk.com 27 October 2026 ESMO Analyst briefing
Page 2
Cautionary statement regarding forward-looking statements 2 This presentation may contain forward-looking statements. Forward-looking statements give the Group’s current expectations or forecasts of future events. An investor can identify these statements by the fact that they do not relate strictly to historical or current facts. They use words such as ‘anticipate’, ‘estimate’, ‘expect’, ‘intend’, ‘will’, ‘project’, ‘plan’, ‘believe’, ‘target’, ‘potential’ and other words and terms of similar meaning in connection with any discussion of future operating or financial performance. In particular, these include statements relating to future actions, prospective products or product approvals, future performance or results of current and anticipated products, sales efforts, expenses, the outcome of contingencies such as legal proceedings, dividend payments and financial results. Other than in accordance with its legal or regulatory obligations (including under the Market Abuse Regulations, UK Listing Rules and the Disclosure Guidance and Transparency Rules of the Financial Conduct Authority), the Group undertakes no obligation to update any forward-looking statements, whether as a result of new information, future events or otherwise. Investors should, however, consult any additional disclosures that the Group may make in any documents which it publishes and/or files with the US Securities and Exchange Commission (SEC). All investors, wherever located, should take note of these disclosures. Accordingly, no assurance can be given that any particular expectation will be met and investors are cautioned not to place undue reliance on the forward-looking statements. Forward-looking statements are subject to assumptions, inherent risks and uncertainties, many of which relate to factors that are beyond the Group’s control or precise estimate. The Group cautions investors that a number of important factors, including those in this presentation, could cause actual results to differ materially from those expressed or implied in any forward-looking statement. Such factors include, but are not limited to, those discussed under ‘Risk factors’ in the Group’s Annual Report on Form 20-F for the full year (FY) 2025 and as described on pages 48 and 49 of the Group's Q2 2026 results press release
Page 3
Speakers 3 Ahmed Rabie SVP Global Product Strategy, Oncology Hesham Abdullah SVP, Global Head Oncology R&D Nina Mojas President, Global Product Strategy
Page 4
Agenda 4 ESMO review Hesham Abdullah, Global Head, Oncology R&D Closing remarks Nina Mojas, President, Global Product Strategy Q&A Hesham, Nina and Ahmed Rabie, Head Oncology Global Product Strategy
Page 5
gsk.com ESMO review Hesham Abdullah, Global Head, Oncology R&D 5
Page 6
6 2L+ SCLC: Emerging global data appear consistent with data generated in China n End points EMBOLD Pan-Tumour 101 2L+ SCLC: 8mg/kg q3w 18 ORR: 61% n End points ARTEMIS-008 China phase III pivotal 2L+ SCLC 461 - ORR 58.3% - OS HR 0.46; mOS 18.5m
Page 7
Presenter: Jie Wang Risvutatug Rezetecan vs Topotecan in Relapsed SCLC: Primary Results of Phase 3 ARTEMIS-008 1National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences Peki ng Union Medical College, Beijing/CN, 2Shandong Cancer Hospital and Institute, Shandong First Medical University, Shandong Academy of Medical Sciences, Jinan/CN, 3Jilin Cancer Hospital, Changchun/CN, 4The Affiliated Cancer Hospital of Zhengzhou University & Henan Cancer Hospital, Institute of Cancer Research, Henan Academy of Innovations in Medical Science, Zhengzhou/CN, 5The First Affiliated Hospital of Nanchang University, Nanchang/CN, 6The First Affiliated Hospital of Zhengzhou University, Zhengzhou/CN, 7Fujian Cancer Hospital, Fuzhou/CN, 8Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan/CN, 9Shengjing Hospital of China Medical University, Shenyang/CN, 10Shanghai Chest Hospital, Shanghai Jiao Tong University, School of Medicine, Shanghai/CN, 11Zhejiang Cancer Hospital, Hangzhou/CN, 12Cancer Institute & Hospital, Tianjin Medical University, Tianjin/CN, 13Shanghai Pulmonary Hospital Affiliated to Tongji University, Shanghai/CN, 14Shanxi Provincial Cancer Hospital, Taiyuan/CN, 15Harbin Medical University Cancer Hospital, Harbin/CN, 16Sichuan Clinical Research Center for Cancer, Sichuan Cancer Hospital & Institute, Sichuan Cancer Center, University of Electr onic Science and Technology of China, Chengdu/CN,17Xiangyang Center Hospital, Affiliated Hospital of Hubei University Arts and Science, Xiangyang/CN, 18Shanghai Hansoh BioMedical Co., Ltd, Shanghai/CN J. Wang1, L. Wang2, J. Duan1, H. Liu3, Q. Wang4, H. Wang2, L. Sun5, S. Huang6, Y . Huang7, F . Tong8, W. Zheng9, T. Chu10, Y . Fan11, D. Huang12, P . Zhang13, W. Guo14, B. Liu15, J. Zhou16, Q. Li17, Y . Dong18, Q. Liu18, M. Zhang18, X. Zhang18, J. Yu2 7
Page 8
8 Abbreviation: CI, confidence interval; CTFI, chemotherapy-free interval; IA, interim analysis; OS, overall survival; VALG, Veterans Administration Lung Study Group Hazard ratio and P value were estimated using a Cox proportional hazards model and Log-rank test, stratified by baseline brain metastasis status, CTFI, and VALG stage at study entry. Censoring rule for OS: patients were censored at their last known survival time; for the 15 untreated patients who withdrew consent in topotecan group, censoring occurred at the withdrawal date. Ris-Rez N=230 Topotecan N=231 Median OS, months 18.5 10.3 Hazard Ratio (95% CI) 0.46 (0.35, 0.62) P value <0.0001 Primary Endpoint: OS Ris-Rez Topotecan ➢ At this IA (DCO June 6, 2026), 77 (33.5%) OS events in Ris -Rez arm and 121 (52.4%) in Topotecan arm. ➢ With a median follow-up of 12.2 months, Ris-Rez demonstrated statistically significant and clinical meaningful survival benefit over topotecan.
Page 9
9 9 Abbreviation: CI, confidence interval; ECOG PS, Eastern Cooperative Oncology Group Performance Status;N, number of participants; PD-1, programmed death receptor 1; PD-L1, programmed cell death ligand 1; VALG, Veterans Administration Lung Study Group Overall Survival – Benefit Across Pre-defined Subgroups
Page 10
10 Secondary Endpoints: PFS by BICR and investigator Abbreviation: BICR, blinded independent central review; CI, confidence interval; HR, hazard ratio; m, median; PFS, progression-free survival BICR-assessed PFS Investigator-assessed PFS 26.1% 59.9% 5.3% 28.5% 61.5% 25.5%30.5% 7.5% Ris-Rez N=230 Topotecan N=231 mPFS, months 7.8 4.0 HR (95% CI) 0.35 (0.28, 0.45) Ris-Rez N=230 Topotecan N=231 mPFS, months 7.2 3.0 HR (95% CI) 0.33 (0.25, 0.42)
Page 11
11 Secondary Endpoints: ORR, DCR and DoR by BICR • Ris-Rez demonstrated consistent benefit across secondary efficacy endpoints 0 10 20 30 40 50 60 70 Ris-Rez Topotecan PR CR PR ORR 58.3% ORR 12.6% Best overall Response (%) Ris-Rez (N= 230) Topotecan (N= 231) ORR, % (95% CI) 58.3 (51.6, 64.7) 12.6 (8.6, 17.5) DCR, % (95% CI) 90.4 (85.9, 93.9) 60.2 (53.5, 66.5) Median DoR, months (95% CI) 6.9 (5.6, 8.2) 5.6 (3.0, 8.3) Abbreviation: BICR, blinded independent central review; CI, confidence interval; CR,complete response; DCR, disease control rate; DoR, duration of response; ORR,objective response rate; PR, partial response 1.7% (4)
Page 12
Notes: †versus investigators choice chemotherapy., mOS: median overall survival, Sources: 1. Hansoh WCLC presentation 2. Medilink TAISHAN-302 Presentation at WCLC 3. IMDELLTRA FDA label Recent phase III SCLC datasets 12 18.5 mo. mOS vs 10.2 for topotecan1 HR:0.46, ∆8.3 mos, 12 mos follow-up 13.3 mo. mOS vs 9.4 for topotecan2 HR:0.46, ∆3.9 mos, 9 mos follow-up 13.6 mo. mOS vs 8.3 for plt-based chemo3 HR:0.6, ∆5.3 mo.† Ris-Rez ARTEMIS-008 (China only) Tam-peli TAISHAN-302 (China only) tarlatamab USPI Global phase III EMBOLD SCLC 301 2L+ SCLC Recruiting EMBOLD SCLC 303 1L SCLC Q4 ‘26 start
Page 13
Ris-Rez global pooled analysis: ILD rates ca4% with most events <Grade 3 13 0% 5% 10% 15% 20% NSCLC + SCLC Other solid tumours mCRPC SCLC Osteosarcoma NSCLC n=44 SCLC n=152 n=362 n=50 n=230 n=48 0.9% 0% 4% 0% EMBOLD global data ARTEMIS 3 ARTEMIS 8 ARTEMIS 2 ILD incidence with Ris-Rez in patients with advanced solid tumours Data from Frail et al. ESMO 2026 1068P NSCLC: Non-small cell lung cancer; SCLC; small cell lung cancer, mCRPC; metastatic castration resistant prostate cancer, ILD interstitial lung disease Tumour type Number of patients Grade >=3 ILD
Page 14
14
Page 15
Risvutatug Rezetecan Versus Gemcitabine and Docetaxel in Heavily Pretreated Patients With Relapsed or Refractory Osteosarcoma: Primary Results From the Phase 3 ARTEMIS-011 Study L. Xie1; W. Yao2; Y. Hua3; W. Luo4; C. Tu5; J. Wang6; J. Chen7; B. Li8; D. Zhu9; X. Li10; Y. Xiao11; J. Ma12; H. Hu13; M. Fang14; H. Wang15; Q. Liu15; Y. Jin15; X. Zhang15; X. Niu16; X. Tang1 Presented by: Lu Xie, MD Musculoskeletal Tumor Center, Peking University People's Hospital (PKUPH), Beijing, China Madrid, Spain; October 26, 2026 LBA9 1Musculoskeletal Tumor Centre, Peking University People‘s Hospital, Beijing, China; 2Department of Orthopedics and Soft Tissue, Henan Cancer Hospital, Affiliated Cancer Hospital of Zhengzhou University, Zhengzhou China; 3Department of Bone Tumor, Shanghai General Hospital, Shanghai, China; 4Department of Orthopaedics, Xiangya Hospital, Central South University, Changsha, China; 5Department of Orthopedics, Orthopedic Research Institute, West China Hospital, Sichuan University, Chengdu, China; 6Department of Bone and soft tissue tumors, Cancer Center of Sun Yat-Sen University, Guangzhou, China; 7Cancer Center, Hubei Key Laboratory of Precision Radiation Oncology，Institute of Radiation Oncology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China; 8Department of Orthopaedics, The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China; 9Department of Oncology, Shandong Cancer Hospital and Institute, Shandong First Medical University, Shandong Academy of Medical Sciences, Jinan, China; 10Department of Bone and soft tissue, Hunan Cancer Hospital, Changsha, China; 11Department of Bone Surgery, Yunnan Cancer Hospital, Kunming, China; 12Department of Oncology, The First Affiliated Hospital of Guangxi Medical University, Nanning, China; 13Clinical research ward, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China; 14Department of Rare Diseases and Head and Neck, Zhejiang Cancer Hospital, Hangzhou, China; 15Clinical Development Center, Shanghai Hansoh BioMedical Co., Ltd, Shanghai, China; 16Department of Orthopaedic Oncology Surgery, National Center for Orthopaedics, Beijing Jishuitan Hospital, Capital Medical University, Beijing, China 15
Page 16
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. ARTEMIS-011: A Multi-Centre, Open-Label, Randomised, Controlled Phase 3 Study Presented by: Lu Xie, MD, PKUPH BICR, blinded independent central review; DCR, disease control rate; DOR, duration of response ECOG PS, Eastern Cooperative Oncology Group performance status; IV, intravenous; ORR, objective response rate; OS, overall survival; PD, progressive disease; PFS, progression-free survival; Q3W, every 3 weeks; R, randomisation; RECIST v1.1, Response Evaluation Criteria in Solid Tumors version 1.1; Ris-Rez, risvutatug rezetecan • PFS by BICR per RECIST v1.1 • OS • PFS by investigator per RECIST v1.1 • ORR, DCR, DOR by BICR/investigator per RECIST v1.1 • Safety Primary Endpoint Secondary Endpoints Study Treatment Stratification: • Age (12 to <18 years or ≥ 18 years) • Lung metastases (Yes or No) R 2: 1 • Age ≥12 years • Pathologically confirmed diagnosis of osteosarcoma • Progression or intolerance to at least two previous lines of therapy • ≥1 measurable target lesion per RECIST v1.1 • ECOG PS 0–1 Gemcitabine 1000 mg/m2 IV on Day 1 and 8 + Docetaxel 75 mg/m2 IV on Day 8, Q3W (n = 42) Treatment until PD, toxicity, or withdrawal of consentRis-Rez 12.0 mg/kg IV on Day 1, Q3W (n = 80) Key Eligibility Criteria Crossover allowed* * Crossover to Ris-Rez following objective PD in patients meeting the crossover criteria 16
Page 17
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. Patient Demographics and Disease Characteristics • A total of 122 patients were randomised and received ≥1 dose of study drug • Median follow-up time was 7.1 months Presented by: Lu Xie, MD, PKUPH Ris-Rez (n = 80) Gemcitabine + Docetaxel (n = 42) Age Median (IQR), years 22.0 (17.5–33.0) 21.0 (17.0–37.0) 12 to <18 years, n (%) 20 (25) 11 (26) ≥18 years, n (%) 60 (75) 31 (74) Male, n (%) 55 (69) 24 (57) BMI, median (range), kg/m2 22.3 (12.2–36.8) 22.8 (15.1–31.5) Time from initial diagnosis to randomisation, median (range), mo 23.1 (5.7–116.9) 23.6 (7.0–107.6) ECOG PS, n (%) 0 26 (33) 9 (21) 1 54 (68) 33 (79) Patients with metastasis, n (%) Lung metastasis 79 (99) 40 (95) Liver metastasis 3 (4) 3 (7) Sum of diameters of target lesions at baseline, mean (SD), mm 74.4 (59) 81.1 (66) Prior lines of systemic anticancer therapy Number of prior lines, median (range) 2.5 (2, 6) 3.0 (2, 5) Chemotherapy, n (%) 80 (100) 42 (100) Immunotherapy, n (%) 21 (26) 11 (26) Targeted therapy, n (%) 55 (69) 33 (79) BMI, body-mass index; ECOG PS, Eastern Cooperative Oncology Group performance status; IQR, interquartile range; mo, months; Ris-Rez, risvutatug rezetecan; SD, standard deviation 17
Page 18
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. Primary Endpoint: BICR-assessed PFS with Ris-Rez vs Gemcitabine + Docetaxel • At data cutoff date, there were 41 PFS events per BICR (41/80; 51%) in the Ris-Rez arm and 31 (31/42; 74%) in the Gemcitabine + Docetaxel arm • Ris-Rez demonstrated a statistically significant and clinically meaningful improvement in BICR-PFS vs Gemcitabine + Docetaxel Presented by: Lu Xie, MD, PKUPH DCO: 17 June 2026. BICR, blinded independent central review; CI, confidence interval; DCO, data cutoff; HR, hazard ratio; mo, months; NR, not reached; PD, progressive disease; PFS, progression-free survival; Ris-Rez, risvutatug rezetecan 80 100 60 40 20 0 0 1 2 3 4 5 6 9 12 80 65 59 43 37 26 13 1 0Ris-Rez 42 24 17 12 9 7 0 0 0Gemcitabine + Docetaxel Number at risk Time since randomisation (mo) PFS (%) 3-mo rate 6-mo rate 45% 70% 33% NR Ris-Rez Gemcitabine + Docetaxel Ris-Rez (n = 80) Gemcitabine + Docetaxel (n = 42) Median, mo (95% CI) 5.5 (4.1–7.4) 1.4 (0.8–2.7) HR (95% CI) 0.33 (0.20–0.55) P-value <0.0001 18
Page 19
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. BICR-assessed ORR Tumour Response Assessed by BICR Presented by: Lu Xie, MD, PKUPH DCO: 17 June 2026. No patient had a CR BICR, blinded independent central review; CI, confidence interval; CR, complete response; DCO, data cutoff; DCR, disease control rate; DOR, duration of response; ORR, objective response rate; PR, partial response; Ris- Rez, risvutatug rezetecan; SD, stable disease. 0 5 10 15 20 Ris- Rez (n = 80) Gemcitabi ne + Docetaxel (n = 42) Patients (%) PR PR ORR 15% ORR 2% Odds Ratio (95% CI): 7.1 (0.9–56.1) • BICR-assessed ORR and DCR were markedly higher with Ris-Rez vs Gemcitabine + Docetaxel • Median DOR by BICR was 7.6 months with Ris-Rez and was not reached in the Gemcitabine + Docetaxel arm (only one patient achieved PR) BICR-assessed DCR 0 20 40 60 80 100 Ris-Rez (n = 80) Gemcitabi ne + Docetaxel (n = 42) Patients (%) PR DCR 76% DCR 33% Odds Ratio (95% CI): 6.2 (2.8–14.0) SD P R SD 19
Page 20
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. OS in R/R Osteosarcoma with Ris-Rez vs Gemcitabine + Docetaxel • OS data were immature but suggested a favourable trend towards an OS benefit with Ris-Rez • 10 events had occurred in each arm: 13% with Ris-Rez and 24% with Gemcitabine + Docetaxel • 17 patients* in the Gemcitabine + Docetaxel arm crossed over to Ris-Rez following PD Presented by: Lu Xie, MD, PKUPH DCO: 17 June 2026. Symbols show censoring. *Crossover criteria were met. BICR, blinded independent central review; CI, confidence interval; DCO, data cutoff; HR, hazard ratio; mo, month; OS, overall survival; PD, progressive disease; Ris-Rez, risvutatug rezetecan; R/R, relapsed/refractory Ris-Rez (n = 80) Gemcitabine + Docetaxel (n = 42) Median, mo Not reached Not reached HR (95% CI) 0.51 (0.21–1.23) 80 10 0 60 40 20 0 0 3 4 6 9 12 Number at risk Time since randomisation (mo) OS (%) 15 80 78 72 45 16 1Ris-Rez 0 42 37 32 26 10 2Gemcitabine + Docetaxel 0 Ris-RezGemcitabine + Docetaxel 20
Page 21
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. Safety Profile TRAEs reported in ≥20% of patients (in either treatment arm) Presented by: Lu Xie, MD, PKUPH ALT, alanine aminotransferase; AST; aspartate aminotransferase; GGT, gamma glutamyltransferase; LDH, lactate dehydrogenase; Ris-Rez, risvutatug rezetecan; R/R, relapsed/refractory; TRAE, treatment-related adverse event; WBC, white blood cell Ris-Rez (n = 80) Grades 1–2 Grade ≥3 Gemcitabine + Docetaxel (n = 42) Grades 1–2 Grade ≥3 Incidence, % Percentage of patients Percentage of patients 1 6 341 1415979716380 6112019212115242524253134363440149203014 2 3343 602940 21247 195107 3121217 331924710 33 31 295243 100 90 80 70 60 50 40 30 20 10 0 10 20 30 40 50 60 70 80 90 100 Rash Constipation Epistaxis Hypokalaemia Infusion-related reaction Decreased appetite Hypoalbuminaemia Vomiting Lymphocyte count decreased Platelet count decreased WBC count decreased 21
Page 22
Mocertatug Rezetecan (Mo-Rez) in Patients With Platinum-Resistant Ovarian Cancer (PROC): Updated Results With Longer Follow-Up From the Global BEHOLD-1 Study Vanda Salutari1, Stephanie Lheureux2, Myong Cheol Lim3, Isabelle Ray-Coquard4, Marloes G van Dongen5, William McKean6, Jung-Yun Lee7, Victor Moreno8, Rowan E Miller9,10, Toon Van Gorp11, Jean-Sébastien Frenel12, Joo Ern Ang13, Lisa Liu Burström14, Katriina Jalkanen15, Shruti D Shah16, Evi Bakirtzi17, Sybil Zildjian18, Morrys C Kaisermann19, Ana Oaknin20 1Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy; 2University Health Network, Princess Margaret Cancer Centre, Toronto, ON, Canada; 3Center for Gynecologic Cancer, National Cancer Center, Goyang-si, Republic of Korea; 4Centre Léon Bérard, Université Claude Bernard Lyon 1, Lyon, France; 5The Netherlands Cancer Institute, Amsterdam, The Netherlands; 6The START Center for Cancer Research, Mountain Region, West Valley City, UT, USA; 7Yonsei University College of Medicine, Seoul, Republic of Korea; 8START Madrid-FJD, Hospital Fundación Jiménez Díaz, Madrid, Spain; 9University College London Hospitals NHS Foundation Trust, London, UK; 10St Bartholomew’s Hospital, London, UK; 11Division of Gynaecological Oncology, University Hospital Leuven and KU Leuven, Leuven Cancer Institute, Leuven, Belgium; 12Institut de Cancérologie de L'ouest, Site René Gauducheau, Saint-Herblain, France; 13Department of Oncology, Cambridge University Hospitals NHS Foundation Trust, Cambridge, UK; 14Theme Cancer, Karolinska University Hospital, Stockholm, Sweden; 15Helsinki University Hospital, Comprehensive Cancer Center, Helsinki, Finland; 16CMPS Oncology, GSK, Waltham, MA, USA; 17Oncology Statistics, GSK, London, UK; 18Oncology Clinical Development, GSK, Waltham, MA, USA; 19Oncology Clinical Development, GSK, Philadelphia, PA, USA; 20Medical Oncology Department, Hospital Universitario Puerta de Hierro Majadahonda, Madrid, Spain 24 October 2026; Publication #LBA41 22
Page 23
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. BEHOLD-1 is a two-part, open-label Phase 1 study *For patients with advanced high-grade serous/endometrioid ovarian, primary peritoneal, or fallopian tube cancer. †Mo-Rez 5.8 mg/kg Q3W selected as the recommended dose for both PROC and advanced/recurrent EC cohorts. ‡Current protocol amendment allows for 1–4 prior LOT. B7-H4, B7 homologue 4 protein; cORR, confirmed objective response rate; DoR, duration of response; EC, endometrial cancer; ECOG PS, Eastern Cooperative Oncology Group Performance Status; INV, investigator; LOT, lines of therapy; Mo-Rez, mocertatug rezetecan; PFS, progression-free survival; PROC, platinum-resistant ovarian cancer; Q3W, every 3 weeks; R, randomisation; RECIST v1.1, Response Evaluation Criteria in Solid Tumors version 1.1; TOP1i, topoisomerase 1 inhibitor. • Aged ≥18 years • Histologically confirmed advanced solid tumour • ECOG PS 0–2 • ≥1 measurable target lesion per RECIST v1.1 • No prior B7-H4 therapy Phase 1b only: • 1–3 prior LOT‡ • No prior TOP1i Eligibility Criteria Phase 1a Dose Escalation 2.8 mg/kg Q3W 4.8 mg/kg Q3W 5.8 mg/kg Q3W 7.2 mg/kg Q3W Advanced solid tumours Expansion phase at Mo-Rez 5.8 mg/kg Q3W is ongoing† Phase 1b Dose Optimisation/Expansion Key Endpoints Primary: cORR assessed by INV according to RECIST v1.1 Secondary: DoR and PFS assessed by INV according to RECIST v1.1, safety 2.8 mg/kg Q3W, (n~30–50) 4.8 mg/kg Q3W, (n~30–50) 5.8 mg/kg Q3W, (n~30–50) PROC * R 1:1:1 2.8 mg/kg Q3W, (n~30–50) 4.8 mg/kg Q3W, (n~30–50) Advanced/recurrent EC R 1:1 23
Page 24
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. Baseline characteristics PROC Total (N=136) Age, median (range), years 60.5 (38, 79) Race, n (%) White / Asian 112 (82) / 23 (17) Black or African American 1 (<1) BMI, median (range), kg/m2 25.8 (16, 50) ECOG PS, n (%) 0 / 1 / 2 68 (50) / 67 (49) / 1 (<1) Histology at diagnosis, n (%) High-grade serous carcinoma / serous adenocarcinoma* 134 (99) / 1 (<1) Endometrioid 1 (<1) Number of prior LOT, median (range) 2 (1, 5) Prior LOT, n (%) 1 / 2 / ≥3† 28 (21) / 58 (43) / 49 (36) Prior therapies, n (%) Bevacizumab 105 (77) PARP inhibitor 79 (58) FRα ADC 6 (4) B7-H4-evaluable tumour sample, n 122 B7-H4 tumour membrane positivity, n (%)‡ 118 (97) DCO: 30 June 2026. Baseline characteristics reported for all randomised patients. *Confirmed as high-grade serous at time of study enrolment. †Two patients were enrolled despite having more than three LOT and were recorded as protocol deviations. One patient had missing data. ‡B7-H4-positive was defined as ≥1% of tumour cells with membrane staining at any intensity determined by B7-H4 IHC assay. ADC, antibody-drug conjugate; B7-H4, B7 homologue 4 protein; BMI, body mass index; DCO, data cut-off; ECOG PS, Eastern Cooperative Oncology Group Performance Status; FRα, folate receptor alpha; IHC, immunohistochemistry; LOT, lines of therapy; PARP(i), poly (adenosine diphosphate-ribose) polymerase (inhibitor); PROC, platinum-resistant ovarian cancer. 36% of patients received >2 prior LOT Most patients received prior bevacizumab or PARPi 97% of evaluable patients showed B7-H4 tumour membrane positivity 24
Page 25
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. Mo-Rez was associated with rapid and deep responses at the 5.8 mg/kg Q3W dose Mo-Rez 2.8 mg/kg Q3W (N=45) Best Percent Change from Baseline in Sum of Target Lesion Diameters Mo-Rez 4.8 mg/kg Q3W (N=46) Best Percent Change from Baseline in Sum of Target Lesion Diameters Mo-Rez 5.8 mg/kg Q3W (N=45) Best Percent Change from Baseline in Sum of Target Lesion Diameters 30 20 10 0 -10 -20 -30 -40 -50 -60 -70 -80 -90 -100 P D P D P D P D S D S D S D S D S D S D S D S D S D S D S D S D P D S D S D S D S D S D S D S D P R S D P R P R S D P R S D P R P R P R P R P R P R P R P R P R CR CR P R 30 20 10 0 -10 -20 -30 -40 -50 -60 -70 -80 -90 -100 S D S D S D S D S D S D S D S D S D S D S D S D S D S D S D S D P R P R P R P R P R P R P R S D P R P R P R P R P R P R P R P R P R P R P R P R CR CR P R P R CR CR P R 30 20 10 0 -10 -20 -30 -40 -50 -60 -70 -80 -90 -100 S D S D S D P D S D S D S D S D S D S D S D P R P R P R S D P R P R S D P R P R P R P R P R P R P R P D P R CR S D P R P R P R CR P R P R P R P R P R P R CR PR CR CR DCO: 30 June 2026. Efficacy analysed in all randomised patients. Waterfall plots show confirmed best overall response for patients with at least one post-baseline tumour assessment. *Two out of four patients who received prior mirvetuximab soravtansine had a confirmed objective response to Mo-Rez (all dose levels). †DpR refers to best change from baseline in SLD of target lesions. B7-H4, B7 homologue 4 protein; CI, confidence interval; cORR, confirmed objective response rate; CR, complete response; DCO, data cut- off; IQR, interquartile range; mDpR, median depth of response; Mo-Rez, mocertatug rezetecan; mTTR, median time to response; PD, progressive disease; PR, partial response; Q3W, every 3 weeks; SD, stable disease; SLD, sum of lesion diameters. Responses were observed across a range of B7-H4 expression levels (scan QR code) CR: 2 (4%) patients PR: 14 (31%) patients CR: 4 (9%) patients PR: 22 (48%) patients CR: 5 (11%) patients PR: 23 (51%) patients Mo-Rez dose, mg/kg Q3W 2.8 4.8 5.8 mTTR, days (IQR) 80.0 (42.5, 85.5) 80.0 (43.0, 121.0) 42.0 (41.5, 50.5) mDpR†, % (IQR) −28.6 (−51.1, −11.1) −38.2 (−60.0, −21.4) −45.5 (−78.3, −27.3) cORR* (95% CI): 36% (21.9, 51.2) cORR* (95% CI): 57% (41.1, 71.1) cORR* (95% CI): 62% (46.5, 76.2) 25
Page 26
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. Mo-Rez was associated with durable responses at 5.8 mg/kg Q3W Median (IQR) follow-up: 13.4 (10.4, 13.9) months Kaplan-Meier plot of DoR 80 100 50 40 0 0 16 26 28 2 16 24 27 4 14 20 25 6 8 15 20 8 5 8 14 10 1 3 8 12 0 1 7 2.8 mg/kg 4.8 mg/kg 5.8 mg/kg Number at risk Time (months) Mo-Rez 5.8 mg/kg Mo-Rez 4.8 mg/kg Mo-Rez 2.8 mg/kg Duration of Response (%) 14 0 0 0 90 70 60 30 20 10 90% (70, 100) mDoR (IQR): 12.2 months (7.0, 12.7) Mo-Rez dose, mg/kg Q3W 2.8 (N=45) 4.8 (N=46) 5.8 (N=45) Events, n (%) 10 (63%) 16 (62%) 11 (39%) Median DoR, months (IQR)* 7.8 (4.4, NE) 6.9 (5.7, 9.7) 12.2 (7.0, 12.7) Probability of DoR at 6 months, % (95% CI)† 60% (30, 80) 70% (50, 90) 90% (70, 100) DCO: 30 June 2026. Efficacy analysed in all randomised patients. *Estimation was conducted using the Kaplan-Meier method and CIs were estimated using the Brookmeyer-Crowley method. †Estimation was conducted using the Kaplan-Meier method and CIs were calculated using Greenwood’s formula for the standard errors, with a log-log transformation to derive the bounds. CI, confidence interval; DCO, data cut-off; DoR, duration of response; IQR, interquartile range; mDoR, median duration of response; Mo-Rez, mocertatug rezetecan; NE, not estimable; Q3W, every 3 weeks. 26
Page 27
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. PFS was longest at the 5.8 mg/kg Q3W Mo-Rez dose DCO: 30 June 2026. Efficacy analysed in all randomised patients. *Estimation was conducted by using the Kaplan-Meier method and CIs estimated using the Brookmeyer-Crowley method. †Estimation was conducted by using the Kaplan-Meier method and CIs calculated using Greenwood’s formula for the standard errors, with a log-log transformation to derive the bounds. ‡HR and 95% CIs were estimated using a Cox proportional hazards model with Efron’s method for tie handling and randomised treatment arm (dose level) as the single covariate. CI, confidence interval; DCO, data cut-off; HR, hazard ratio; IQR, interquartile range; Mo-Rez, mocertatug rezetecan; PFS, progression-free survival; Q3W, every 3 weeks. Kaplan-Meier plot of PFS Median (IQR) follow-up: 13.4 (10.4, 13.9) months 80 100 50 40 0 0 45 46 45 2 38 43 41 4 28 39 39 6 12 28 30 8 11 23 28 10 7 14 16 12 2 4 9 2.8 mg/kg 4.8 mg/kg 5.8 mg/kg Number at risk Time (months) Progression-Free Survival (%) 14 0 1 2 90 70 60 30 20 10 16 0 0 1 18 0 0 0 mPFS (IQR): 11.1 months (8.3, 13.9) 80% (60, 90) 5.8 vs 4.8 mg/kg HR: 0.66 (95% CI: 0.4, 1.1)‡ Mo-Rez 5.8 mg/kg Mo-Rez 4.8 mg/kg Mo-Rez 2.8 mg/kg Mo-Rez dose, mg/kg Q3W 2.8 (N=45) 4.8 (N=46) 5.8 (N=45) Events, n (%) 35 (78%) 30 (65%) 23 (51%) Median PFS, months (IQR)* 5.6 (4.0, 5.8) 8.3 (7.4, 10.9) 11.1 (8.3, 13.9) Probability of PFS at 6 months, % (95% CI)† 30% (20, 50) 70% (50, 80) 80% (60, 90) 27
Page 28
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. B7-H4 is widely expressed in PROC, enabling tumour selective payload delivery *Baseline TC B7-H4 expression levels in patients with PROC enrolled in BEHOLD Phase 1b and with evaluable tissue samples. †Waterfall plot shows best change for patients with baseline, and at least one post-baseline, tumour assessment receiving Mo-Rez 5.8 mg/kg by the percentage of B7-H4 TC membrane+. ‡TC membrane+ ≥1+ intensity of <25%. B7-H4, B7 homologue 4 protein; CI, confidence interval; cORR, confirmed objective response rate; CR, complete response; IHC, immunohistochemistry; Mo-Rez, mocertatug rezetecan; PD, progressive disease; PR, partial response; PROC, platinum-resistant ovarian cancer; SD, stable disease; TC, tumour cell; TL, target lesion. 96% of patients with PROC had B7-H4+ tumours (n=107/111)* Responses were observed across a range of B7-H4 expression levels at the 5.8 mg/kg dose (n=34; data cut-off Nov 2025)† 100 20 60 80 40 0 % cases (95% CI) 19% (12–28) 22% (15–31) 22% (15–32) 33% (25–43) % TC membrane+ ≥1+ intensity 1–24% 25–49% 50–74% ≥75% 30 20 10 0 -10 -20 -30 -40 -50 -60 -70 -80 -90 -100 SD‡ SD SD SD SD SD SD SD PR SD PR PRPR PR PRPR PR PR PR PR PR PR CR CRPR PR PD PR PR PR PR SD Best % change from baseline in sum of TL diameters % TC membrane+ ≥1+ intensity 1–24% 25–49% 50–74% ≥75% Baseline B7-H4 expression was assessed via IHC across all dose levels 28
Page 29
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. The most common treatment-related AEs were gastrointestinal and haematological DCO: 30 June 2026. Scan QR code for data on Mo-Rez 2.8 mg/kg dose. Safety analysis set included patients who received ≥1 dose of study treatment. AEs assessed using MedDRA 29.0 and CTCAE 5.0. *TRAEs occurring in ≥15% of patients in the PROC cohort are shown. †These AESI were investigator-assessed, not adjudicated, and based on a search of the Interstitial Lung Disease Standardised Medical Dictionary for Regulatory Activities Query (Narrow). ‡Including PROC and EC patients receiving Mo-Rez 5.8 mg/kg Q3W. Primary prophylactic therapy with G-CSF defined as that administered within four days of the first Mo-Rez dose during Cycle 1 (GSK data on file). Febrile neutropenia occurred in three patients (7%) at 5.8 mg/kg dose (all Grade 3). AE, adverse event; AESI, AE of special interest; CTCAE, Common Terminology Criteria for Adverse Events; DCO, data cut-off; EC, endometrial cancer; G-CSF, granulocyte colony-stimulating factor; ILD, interstitial lung disease; MedDRA, Medical Dictionary for Regulatory Activities; Mo-Rez, mocertatug rezetecan; PROC, platinum-resistant ovarian cancer; Q3W, every 3 weeks; TRAE, treatment-related AE. Mo-Rez 5.8 mg/kg Q3W (N=45)Mo-Rez 4.8 mg/kg Q3W (N=45) Patients*, % Patients*, % 11 49 49 29 13 20 13 31 40 44 69 24 11 2 1 2 4 4 0 50 100 Thrombocytopenia Neutropenia Anaemia Constipation Asthenia Decreased appetite Diarrhoea Vomiting Alopecia Fatigue Nausea Grade ≥3 Any… 38 87 62 16 33 27 42 33 58 53 87 16 60 9 9 4 7 4 0 50 100 Thrombocytopenia Neutropenia Anaemia Constipation Asthenia Decreased appetite Diarrhoea Vomiting Alopecia Fatigue Nausea Grade ≥3 Any grade9 Non-haematological Haematological ILD/pneumonitis occurred in 9 patients (7%): 8 at 4.8 mg/kg and 1 (2%) at the 5.8 mg/kg dose† • 8 of 9 events were Grade 1–2 Primary G-CSF prophylaxis was used in 8% of patients at the 5.8 mg/kg dose‡ G-CSF prophylaxis is now strongly recommended across the BEHOLD study programme 29
Page 30
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. Mo-Rez is being investigated in five global Phase 3 trials spanning ovarian and endometrial cancer 1L, first-line; 2L+, second-line and beyond; BEV, bevacizumab; CT, chemotherapy; EC, endometrial cancer; MMRp, mismatch repair proficient; Mo-Rez, mocertatug rezetecan; non-HRd, non-homologous recombinant deficient; OC, ovarian cancer; PD-1, programmed cell death protein 1; PROC, platinum-resistant ovarian cancer; PSOC, platinum-sensitive ovarian cancer; SoC, standard of care. Mo-Rez vs SoC in 2L+ PROC NCT07286266 | GOG-3132 | ENGOT-ov98 | APGOT-OV24 Mo-Rez + BEV vs CT + BEV in PSOC NCT07684612 | GOG-3136 | ENGOT-ov99 | APGOT-OV19 Mo-Rez ± BEV vs active observation ± BEV as 1L maintenance in non-HRd OC NCT07694427 | ENGOT-ov105 | GOG-3135 | APGOT-OV20 Mo-Rez + anti-PD-1 vs anti-PD-1 as 1L maintenance in MMRp EC NCT07684599 | ENGOT-en33 | GOG-3130 | APGOT-EN5 Mo-Rez vs CT in 2L+ advanced or recurrent EC NCT07286331 | ENGOT-en34 | GOG-3131 | APGOT-EN4 30
Page 31
Clinical Complete Response with Dostarlimab Monotherapy in dMMR/MSI-H Locally Advanced Rectal Cancer: First Results from the Global Registrational AZUR-1 Trial Andrea Cercek1, Jean-Baptiste Bachet2, Gunnar Folprecht3, Hideaki Bando4, Jenny Seligmann5, Naureen Starling6, María del Carmen Riesco Martínez7, Pilar Garcia Alfonso8, Denis Smith9, Lisa Salvatore10,11, Encarna Gonzalez Flores12, Emmanuel Mitry13, Eric Chen14, Jaume Capdevila15, Federica Morano16, Susana Roselló Keränen17, Neda Stjepanovic18, Zsolt Szijgyarto19, Luis A. Diaz, Jr1, Jeanine M.L. Roodhart20 1Memorial Sloan Kettering Cancer Center, New York, NY, USA; 2Sorbonne University, Pitié Salpêtrière Hospital, Paris, France; 3Technical University Dresden/Medical Faculty Carl Gustav Carus, Dresden, Germany; 4National Cancer Center Hospital East, Kashiwa-shi, Chiba, Japan; 5University of Leeds, Leeds, UK; 6Royal Marsden NHS Foundation, London, UK; 7Hospital Universitario 12 de Octubre, Madrid, Spain; 8Hospital General Universitario Gregorio Marañón, Faculty of Medicine, Complutense University of Madrid, Instituto de Investigación Sanitaria Gregorio Marañón, Madrid, Spain; 9CHU HAUT LÉVÊQUE, Pessac, France; 10Università Cattolica del Sacro Cuore, Rome, Italy; 11Comprehensive Cancer Center, Fondazione Policlinico Universitario Agostino Gemelli, IRCCS Rome, Rome, Italy; 12Hospital Universitario Virgen de las Nieves, Instituto de Investigación Biosanitaria IBS, Granada, Spain; 13Institut Paoli-Calmettes, Marseille, France; 14Princess Margaret Cancer Centre, Toronto, Canada; 15Vall Hebron University Hospital and Vall Hebron Institute of Oncology (VHIO), Barcelona, Spain; 16Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy; 17Hospital Clínico Universitario, INCLIVA Biomedical Research Institute, University of Valencia, Valencia, Spain; 18GSK, Baar, Switzerland; 19GSK, London, UK; 20University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands 31
Page 32
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. AZUR-1 (NCT05723562) is the first global registrational study to evaluate the clinical benefit of dostarlimab monotherapy in patients with dMMR/MSI-H locally advanced rectal cancer11 Dostarlimab, a PD-1 inhibitor, has demonstrated compelling durable responses as monotherapy in patients with dMMR/MSI-H locally advanced rectal cancer in a single-centre study, supporting a non-operative management (NOM) approach that may also allow patients to avoid chemotherapy and radiation10 Locally advanced rectal cancer (LARC) has historically been treated with multimodal SOC1,2 • CRT and surgical approaches are associated with substantial toxicity, morbidity, and QoL burden3,4 • There has been growing interest in organ preservation strategies5,6 • While data are limited, cCR rates post-SOC in dMMR/MSI-H locally advanced rectal cancer are approximately 20–25%7–9 We report interim results from the AZUR-1 trial, focusing on clinical complete response (cCR) and safety outcomes Background Dr Andrea Cercek CRT, chemoradiotherapy; dMMR, mismatch repair deficient; MSI-H, microsatellite instability high; PD-1, programmed cell death protein 1; QoL, quality of life; SOC, standard of care; TNT, total neoadjuvant therapy. 1. Conroy T, et al. Lancet. 2021;22(5):702-15; 2. Bahadoer RR, et al. Lancet Oncol. 2021;22(1):29-42; 3. Conces ML, et al. Curr Oncol. 2024;31(1):366–382; 4. Wyatt J, et al. Clin Oncol. 2023;35(2):117–123; 5. Williams H, et al. Front Oncol. 2024;12:1477510; 6. Cercek A, et al. NEJM. 2022;386(25):2362–2378; 7. Audisio A, et al. Jama Oncol. 2025;11(9):1045–1054; 8. Habr-Gama A, et al. Ann Surg. 2004;240(4):711–717; 9. Lunenberg RA, et al. Int J Cancer. 2025;158(4):963–976; 10. Cercek A, et al. NEJM. 2025;392(23):2297–2308; 11. Cercek A, et al. Clin Colorectal Cancer. 2025;24(2):325-330. 32
Page 33
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. Baseline characteristics Dr Andrea Cercek DCO 20 Apr 2026. *Data de-identified for demographic variables if n≤10. †Tumour location was determined by the distance from interior border primary tumour to anal verge (cm) based on imaging. High, middle, and low are >10–15 cm, >5–10 cm, and 0–5 cm from anal margin; ‡Mucinous tumour defined as >50% mucin within the tumour, determined by imaging. dMMR, mismatch repair deficient; ECOG, Eastern Cooperative Oncology Group; MSI-H, microsatellite instability-high Dostarlimab monotherapy (N=154) Age, median (range), years 61.5 (23−92) ≥65, n (%) 65 (42) Sex, n (%) Male 95 (62) Female 59 (38) ECOG performance status, n (%) 0 109 (71) 1 45 (29) dMMR/MSI-H status, n (%) dMMR 138 (90) MSI-H 16 (10) Lynch syndrome, n (%) Yes 35 (23) No 18 (12) Not tested 101 (66) Dostarlimab monotherapy (N=154) Disease stage at screening, n (%) Stage II 22 (14) Stage III 132 (86) Clinical tumour stage, n (%) T1–T2 14 (9) T3 100 (65) T4 40 (26) Clinical node stage positive, n (%) 132 (86) Tumour location†, n (%) High 24 (16) Middle 67 (44) Low 62 (40) Unknown 1 (<1) Mucinous tumour‡, n (%) Yes 23 (15) No 131 (85) 33
Page 34
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. High rate of cCR12 observed with dostarlimab alone Dr Andrea Cercek *Sufficient follow-up for cCR12 analyses required 12 months of follow-up and adequate disease assessments according to the scheduled assessment plan (i.e., a patient with a cCR may have completed 12 months of follow-up but would still not be cCR12-evaluable if the required disease assessments were not performed); †Patients who were not evaluable (n=4) have been pooled with those who were N/A (n=6): four patients did not reach end of treatment response assessment and did not initiate SoC treatment; six patients did not undergo end of treatment response assessment but were recorded as continuing on study and received SoC. cCR, clinical complete response; N/A, not available; nCR, near complete response; pCR, pathologic complete response; PD, progressive disease; PR, partial response; SD, stable disease. cCR/pCR nCR PR SD PD N/A† cCR with ≥12 months follow-up* (n=98) pCR post dostar (n=12) Non-cCR/pCR (n=34) cCR with <12 mo follow-up (n=10) Patients evaluable for cCR12 = 144 pCR after dostarlimab monotherapy (n=12) nCR (n=3) Late cCR (n=8) cCR with ≥12 mo follow-up (n=98) cCR with <12 mo follow-up (n=10) Non-cCR/pCR (n=34) (98/144) cCR12 by INV 68% 34
Page 35
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. Durable responses were observed in patients who achieved cCR Dr Andrea Cercek Median follow-up was 26.3 months (range 2–36). cCR, clinical complete response; EOT, end of treatment Patients achieving cCR12 (INV) after dostarlimab alone (n=98) 98% of cCRs were durable with no recurrences (2 patients with cCR elected for surgery and had pCR) Late cCR (n=6) cCR at EOT response assessment (n=92) cCR reported prior to EOT EOTX Time from enrolment to start of cCR12 follow-up In follow-up with sustained cCR for >12 months Ongoing follow-up (further response assessment pending) Months 35
Page 36
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. cCR12 + pCR shows high overall benefit of dostarlimab alone Dr Andrea Cercek *No subsequent systemic treatment or radiotherapy after completing dostarlimab; †Patients who were not evaluable (n=4) have been pooled with those who were N/A (n=6): four patients did not reach end of treatment response assessment and did not initiate SoC treatment; six patients did not undergo end of treatment response assessment but were recorded as continuing on study and received SoC. cCR, clinical complete response; iCR, incomplete response; nCR, near complete response; NE, not evaluable; ORR, overall response rate; pCR, pathologic complete response; PD, progressive disease; PR, partial response; SD, stable disease. cCR with <12 mo follow-up (n=10) cCR12 (n=98) + pCR (n=12) cCR/pCR nCR PR SD PD N/A† cCR with ≥12 months follow-up* (n=98) pCR post dostar (n=12) Non-cCR/pCR (n=34) Patients evaluable for cCR12 by INV = 144 pCR after dostarlimab monotherapy (n=12) nCR (n=3) Late cCR (n=8) cCR with ≥12 mo follow-up (n=98) cCR with <12 mo follow-up (n=10) Non-cCR/pCR (n=34) cCR12 + pCR (110/144) 76% 36
Page 37
Image of complete response after Dostarlimab monotherapy in LA dMMR/MSI-H rectal cancer Before Dostarlimab Baseline evaluation after Dostarlimab completion 2nd follow-up after Dostarlimab completion 37
Page 38
Content of this presentation is copyright and responsibility of the author. Permission is required for re-use. Safety profile consistent with prior studies of dostarlimab Dr Andrea Cercek DCO 20 April 2026 *Results in death, is life-threatening, requires or prolongs hospitalization, causes persistent disability, results in a birth defect, or other medically important event; †AEs considered related to dostarlimab were at the discretion of the investigator; ‡In 2 patients with response to dostarlimab the obstruction was at primary tumor site, one of whom died due to sepsis with complications post surgery; §Includes large intestinal perforation (n=1) and procedural intestinal perforation (n=1); ‖Includes Grade 3 adrenal insufficiency (n=1), Grade 4 diabetic ketoacidosis (n=1), Grade 3 lichen planus (n=1) and Grade 3 hepatitis; ¶Hepatitis (n=1), amylase increase (n=1), and lichen planus (n=1). AE, adverse event; SAEs, serious adverse event. Dostarlimab median number of cycles 9 (range 2–9) N=154 Among adverse events related to dostarlimab: • 3 (2%) patients discontinued treatment¶ • No fatal AEs SAEs* Any causality Dostarlimab related† Any treatment emergent SAE, n (%) 31 (20) 8 (5) SAEs of clinical interest Intestinal obstruction‡ 3 (2) 0 Rectal stenosis 2 (1) 0 Intestinal perforation§ 2 (1) 0 AEs (≥10% of patients) Gr1‒2 ≥Gr3 Any dostarlimab-related AE, n (%) 98 (64) 15 (10) Pruritus 21 (14) 0 Hypothyroidism 19 (12) 0 Fatigue 16 (10) 1 (<1) Immune-mediated AEs (≥2% of patients) Gr1‒2 ≥Gr3 Any immune-mediated AE, n (%) 32 (21) 4 (3)‖ Hypothyroidism 12 (8) 0 Arthralgia 3 (2) 0 Hyperthyroidism 3 (2) 0 38
Page 39
39 Locally advanced rectal cancer (LARC) is typically treated with multi-modal standard of care Surgery Current Standard of Care AZUR-1 data • Single agent dostarlimab for 9 cycles • cCR rate 68% per investigator • 76% cCR12 + pCR • 98% of cCRs were durable, with no recurrences Adverse events • 5% SAE considered dostar related • No fatal AEs • 2% of patients discontinued treatment Chemotherapy Radiotherapy Outcomes Adverse events • cCR rate 20-25% • Death due to disease ca30% • Surgery related complications (bleeding, infection, injury to neighbouring organs) • Bowel, urinary sexual dysfunction • Colostomy bags • Chemo/radio related toxicities
Page 40
Alexander E. Drilon, Adrianus Johannes de Langen, Benjamin Besse, Aurélie Swalduz, Christina S. Baik, Elaine Shum, Tatsuya Yoshida, Geoffrey Liu, Stephen V . Liu, Julien Mazieres, Joel W. Neal, Benjamin J. Solomon, Daniel Tan, A. J. van der Wekken, Ryo Ariyasu, Jessica R. Bauman, Ji-Youn Han, Dong-Wan Kim, Lorenza Landi, Jessica J. Lin, Dwight H. Owen, Hiroaki Akamatsu, Rossana Berardi, Antonio Calles, Gilberto de Lima Lopes, Enriqueta Felip, Marina Chiara Garassino, Gee-Chen Chang, Hidetoshi Hayashi, Melissa Johnson, Steven Kao, Chia-Chi Lin, Chien-Chung Lin, Kiichiro Ninomiya, Sehhoon Park, Giulia Pasello, Elvire Pons-Tostivint, Alice T. Shaw, Ross Andrew Soo, Shuhei Teranishi, Saiama N. Waqar, Manoj Samant, Junwu Shen, Vivek A. Upadhyay, Byoung Chul Cho Zidesamtinib in TKI-naive patients with advanced/metastatic ROS1+ NSCLC: ARROS-1 efficacy and safety data Memorial Sloan Kettering Cancer Center and Weill Cornell Medical Center, New York, NY , USA 40
Page 41
ARROS-1: Patient Population a Includes TKI-naïve patients enrolled across any cohort. b Includes patients with ≤1 prior line of chemotherapy with or without immunotherapy from the registrational intent ROS1 TKI-naïve cohort. c Measurable or non-measurable lesions by BICR. d 16/25 patients with prior chemotherapy also received prior immunotherapy. BICR, blinded independent central review; CNS, central nervous system; DOR, duration of response; ECOG PS, Eastern Cooperative Oncology Group Performance Status; QD, once daily; NSCLC, non-small cell lung cancer; ROS1, c-ros oncogene 1; TKI, tyrosine kinase inhibitor. Patient Characteristic All data shown as n (%) unless otherwise specified ROS1 TKI-Naive Efficacy Population n = 94 Age, median (range), years 59 (26, 87) Female 55 (59%) Never smoker 55 (59%) Geographic region Asia Pacific 29 (31%) Europe 27 (29%) North America 38 (40%) ECOG PS 0 56 (60%) 1 38 (40%) Baseline CNS metastases c 16 (17%) Tumor stage at study entry III 7 (7%) IV 87 (93%) Treatment History Prior platinum-based chemotherapy ± immunotherapy d 25 (27%) ROS1 TKI-Naive Efficacy Population: b n = 94 with measurable disease by BICR Treated by 15 June 2025 (≥ 9 months DOR follow up) ROS1 TKI-Naive a n = 183 Safety Population: N = 532 Advanced ROS1+ NSCLC Zidesamtinib 100 mg QD (any line of therapy) Data cut-off: 16 April 2026 Total Enrolled: N = 629 Any ROS1+ solid tumor, any dose Phase 1 + Phase 2 pooled Median duration follow-up: 15.2 months (range 1.1– 30.5) 41
Page 42
ARROS-1: Objective Response in ROS1 TKI-Naive Patients with NSCLC 42 c CR with <100% shrinkage due to residual non-pathologic lymph node. BICR, blinded independent central review; CR, complete response; NSCLC, non-small cell lung cancer; ORR, objective response rate; PD, progressive disease; PR, partial response; RECIST, Response Evaluation Criteria in Solid Tumor; ROS1, c-ros oncogene 1; SD, stable disease; TKI, tyrosine kinase inhibitor; uCR, unconfirmed CR (single-timepoint CR pending confirmation in ongoing patient with prior confirmed PR); uPR, unconfirmed PR (single-timepoint PR pending confirmation in ongoing patient). -100 -80 -60 -40 -20 0 20 40 60 Best % change in target lesions + Prior chemotherapyPD SD uPR PR uCR CR ROS1 TKI-Naive ± Chemotherapy Advanced ROS1+ NSCLC RECIST v1.1 by BICR ROS1 TKI-Naive ± Prior Chemotherapy (n = 94) ORR, % (n/N) a [95% CI] 94% (88/94) [87, 98] CR, % (n/N) b 15% (14/94) a Includes 1 single-timepoint PR pending confirmation in ongoing patient. b Includes 1 single-timepoint CR pending confirmation in ongoing patient with prior confirmed PR. c c 42
Page 43
ARROS-1: Duration of Response in ROS1 TKI-Naive Patients with NSCLC 43 DOR, duration of response; NE, not estimable; NR, not reached; NSCLC, non-small cell lung cancer; ROS1, c-ros oncogene 1; TKI, tyrosine kinase inhibitor. Advanced ROS1+ NSCLC Kaplan-Meier Estimate ROS1 TKI-Naive ± Prior Chemotherapy n = 87 % DOR ≥ 6 months [95% CI] 96% [89, 99] % DOR ≥ 9 months [95% CI] 94% [86, 97] % DOR ≥ 12 months [95% CI] 86% [75, 92] Median DOR, months [95% CI] NR [20, NE] 0 3 6 9 12 15 18 21 24 27 0 25 50 75 100 Months Patients in response (%) No. at risk: 87 85 82 72 41 31 20 7 4 0 Kaplan-Meier Plot of DOR ROS1 TKI-Naive ± Chemotherapy 96% 94% 86% 43
Page 44
ARROS-1: Progression-Free Survival in ROS1 TKI-Naive Patients with NSCLC 44 NE, not estimable; NR, not reached; NSCLC, non-small cell lung cancer; PFS, progression-free survival; ROS1, c-ros oncogene 1; TKI, tyrosine kinase inhibitor. Advanced ROS1+ NSCLC Kaplan-Meier Estimate ROS1 TKI-Naive ± Prior Chemotherapy n = 94 % PFS ≥ 6 months [95% CI] 95% [87, 98] % PFS ≥ 9 months [95% CI] 95% [87, 98] % PFS ≥ 12 months [95% CI] 90% [81, 95] Median PFS, months [95% CI] NR [NE, NE] 0 3 6 9 12 15 18 21 24 27 30 0 25 50 75 100 Months Event-free probability (%) No. at risk: 94 92 86 85 63 36 28 14 5 1 0 Kaplan-Meier Plot of PFS ROS1 TKI-Naive ± Chemotherapy 95% 95% 90% 44
Page 45
-100 -75 -50 -25 0 25 50 Best % change in CNS target lesions + Prior chemotherapyIC-PR IC-CR ARROS-1: CNS Activity in ROS1 TKI-Naive Patients with NSCLC BICR, blinded independent central review; CNS, central nervous system; IC-CR, intracranial complete response; IC-DOR, intracranial duration of response; IC-ORR, intracranial objective response rate; IC-PR, intracranial partial response; NE, not estimable; NR, not reached; NSCLC, non-small cell lung cancer; RECIST, Response Evaluation Criteria in Solid Tumor; ROS1, c-ros oncogene 1; TKI, tyrosine kinase inhibitor. CNS Response-Evaluable Population a RECIST v1.1 by BICR ROS1 TKI-Naive ± Prior Chemotherapy n = 10 IC-ORR, % (n/N) [95% CI] 100% (10/10) [69, 100] IC-CR, % (n/N) 70% (7/10) % IC-DOR ≥ 6 months b [95% CI] 100% [100, 100] % IC-DOR ≥ 9 months b [95% CI] 100% [100, 100] % IC-DOR ≥ 12 months b [95% CI] 78% [36, 94] Median IC-DOR, months b [95% CI] NR [9, NE] a Includes patients without brain radiation within 2 months of the first dose of zidesamtinib with measurable (≥5 mm) CNS lesions at baseline by BICR. b Analyses of DOR based on Kaplan-Meier estimates. 0 3 6 9 12 15 18 21 24 27 0 25 50 75 100 Months Patients in response (%) Kaplan-Meier Plot of IC-DOR ROS1 TKI-Naive CNS Response-Evaluable ROS1 TKI-Naive CNS Response-Evaluable No. at risk: 10 10 10 9 6 5 3 1 1 0 100% 100% 78% No CNS progression events observed among the 78 ROS1 TKI- naive patients who entered the study without brain metastases at baseline per BICR 45
Page 46
ARROS-1: Safety in Advanced ROS1+ NSCLC 46 ALT, alanine aminotransferase; AST, aspartate aminotransferase; CPK, creatine phosphokinase; NSCLC, non-small cell lung cancer; QD, once-daily; ROS1, c-ros oncogene 1; TEAE, treatment emergent adverse event; TKI, tyrosine kinase inhibitor. • Treatment-related AEs in ≥15% of patients (any grade; grade ≥3): ‒ Peripheral edema (34%; 0.6%) ‒ Weight increased (18%; 4%) ‒ Blood CPK increased (18%; 3%) ‒ Dysgeusia (17%; 0%) ‒ AST increased (15%; 1%) • Dose reduction due to TEAE: 12%; TRAE: 11% ‒ TEAEs leading to dose reduction in ≥1% of patients: peripheral edema (2.4%), peripheral sensory neuropathy (1.5%), blood CPK increased (1.1%), weight increased (1.1%) • Discontinuation due to TEAE: 4%; TRAE: 1% ‒ TEAEs leading to discontinuation in ≥2 patients: pneumonia (n=3), pneumonitis (n=3), weight increased (n=2) TEAEs in ≥ 15% of TKI-Naive or TKI-Pretreated Patients with Advanced ROS1+ NSCLC Receiving Zidesamtinib 100 mg QD (N = 532) Preferred Term Any Grade Grade ≥3 Peripheral edema 42% 1% Blood CPK increased 23% 5% Constipation 23% <1% Weight increased 22% 7% AST increased 19% 1% ALT increased 18% 2% Dysgeusia 18% 0 Dyspnea 17% 4% Arthralgia 17% 1% Peripheral sensory neuropathy 17% 1% Hypertriglyceridemia 16% 3% Headache 15% <1% 46
Page 47
gsk.com Closing remarks Nina Mojas, President Global Product Strategy 47
Page 48
ORR: objective response rate; cORR: confirmed ORR; (c)CR: (clinical) complete response; m(P)FS/mOS: median (progression-free)/overall survival; DoR: duration of response; HR: hazard ratio; SoC: standard of care; m: months; (N)SCLC: (non-)small cell lung cancer; r/r relapsed/refractory; TKI: tyrosine kinase inhibitor; LARC: Locally advanced rectal cancer; dMMR: mismatch repair deficient; 2L+: second line or later. Data: EMBOLD Pan-tumour 101, ARTEMIS-011, BEHOLD Ovarian 01, AZUR-1 (ESMO 2026); ARROS-1, ARTEMIS-008 (WCLC 2026). Context: published literature (Nature Cancer 2025; Sci Rep 2015; Acta Oncol 2005), topotecan data in sensitive relapse. Jideytro. is not currently approved for TKI naïve NSCLC Ris-Rez: 2L+ SCLC EMBOLD pan-tumour 101 61% ORR ARTEMIS-008 ORR 58% Ris-Rez: r/r osteosarcoma ARTEMIS-011 0.33 PFS HR mPFS 5.5m vs 1.4m for current standard of care Jemperli: dMMR LARC AZUR-1 68% cCR; 98% durable at 26m SoC achieves cCR in only 20-25% of cases Jideytro: TKI-naive ROS-1+ NSCLC ARROS-1 94% ORR 90% progression free at 12mo IC-ORR= 100% (n=10) ESMO and WCLC 2026 data summary 48 Mo-Rez: PROC BEHOLD Ovarian 01 62% cORR Median duration of response 12.2m 11.1month mPFS Ris-Rez: 2L+ SCLC ARTEMIS-008 18.5m mOS vs 10.3m topotecan (HR 0.46) 54% reduction in risk of death
Page 49
Q&A Nina Mojas President, Global Product Strategy Ahmed Rabie SVP Global Product Strategy, Oncology Hesham Abdullah SVP, Global Head Oncology R&D Eric Richards SVP Oncology Product Leadership 49 Klaas Bakker MDL Ris-Rez Richard Davidson MDL Mo-Rez