Winter Clinical Dermatology Conference, Hawaii (January 13–18, 2023) and Miami (February 17–20, 2023) Email: Gerogina.long@sydney.edu.au Conclusions • NIVO significantly reduced the risk of recurrence by 58% compared with PBO in patients with resected stage IIB or IIC melanoma – HR, 0.42 (95% CI, 0.30-0.59; stratified P < 0.0001) – Higher 12-month RFS rates of 89% vs 79% • Benefit of NIVO vs PBO was observed across prespecified subgroups, including T category and disease stage • Clinically meaningful improvement in DMFS was observed with NIVO vs PBO at this initial assessment (HR, 0.47; 95% CI, 0.30-0.72) • Safety profile of NIVO was consistent with previous reports and manageable using well-established treatment algorithms • These results support NIVO as an effective adjuvant treatment option in resected stage IIB or IIC melanoma Background • Approximately one-third of patients with stage IIB and half with stage IIC disease experience recurrence within 5 years (Figure 1)1 – Prognosis of patients with stage IIB and IIC disease is similar to those with stage IIIA and IIIB melanoma, respectively,1-3 highlighting the need for effective and tolerable adjuvant therapies in this population • Checkpoint inhibitors have transformed the adjuvant treatment of resected stage III or IV melanoma4-8 • In KEYNOTE-716, adjuvant pembrolizumab vs placebo (PBO) in resected stage IIB or IIC melanoma improved recurrence-free survival (RFS): hazard ratio (HR), 0.65 (95% CI, 0.46–0.92)9 References 1. Garbe C, et al. J Clin Oncol 2022;40:3741–3749. 2. Garbe C, et al. J Clin Oncol 2020;38:2543–2551. 3. Gershenwald JE, et al. CA Cancer J Clin 2017;67:472–492. 4. Eggermont AMM, et al. Lancet Oncol 2015;16:522–530. 5. Weber J, et al. N Engl J Med 2017;377:1824–1835. Acknowledgments • The patients and families who have made this trial possible • The clinical study teams who participated • Bristol Myers Squibb (Princeton, NJ) and Ono Pharmaceutical Company, Ltd. (Osaka, Japan) • This study was supported by Bristol Myers Squibb • All authors contributed to and approved the presentation; writing and editorial assistance were provided by Melissa Kirk, PhD, and Michele Salernitano of Ashfield MedComms, an Inizio company, funded by Bristol Myers Squibb Figure 1. Recurrence risk of stage IIB and IIC disease aConfirmatory cohort. AJCCv8, American Joint Committee on Cancer, Cancer Staging Manual, version 8; CMMR, Central Malignant Melanoma Registry; MSS, melanoma-specific survival. Adjuvant therapy with nivolumab versus placebo in patients with resected stage IIB/C melanoma (CheckMate 76K) Georgina V. Long,1 Michele Del Vecchio,2 Jeffrey Weber,3 Christoph Hoeller,4 Jean-Jacques Grob,5 Peter Mohr,6 Stephan Grabbe,7 Caroline Dutriaux,8 Vanna Chiarion-Sileni,9 Jacek Mackiewicz,10 Piotr Rutkowski,11 Petr Arenberger,12 Gaëlle Quéreux,13 Tarek Meniawy,14 Paolo A. Ascierto,15 Piyush Durani,16 Maurice Lobo,16 Federico Campigotto,16 Brian Gastman,17* John M. Kirkwood18* 1Melanoma Institute Australia, The University of Sydney, and Royal North Shore and Mater Hospitals, Sydney, New South Wales, Australia; 2IRCCS Istituto Nazionale dei Tumori, Milan, Italy; 3NYU Langone Medical Center, New York, NY, USA; 4Medizinische Universität Wien, Vienna, Austria; 5Hôpital de laTimone, Marseille, France; 6Elbe Klinikum Buxtehude, Buxtehude, Germany; 7University of Mainz Medical Center, Mainz, Germany; 8Hôpital Saint André, Bordeaux, France; 9Istituto Oncologico Veneto IOV-IRCCS, Padova, Italy; 10Institute of Oncology, Poznan University of Medical Sciences, Poznan, Poland; 11Maria Skłodowska-Curie National Research Institute of Oncology, Warsaw, Poland; 12Charles University Third Faculty of Medicine and University Hospital of Královské Vinonrady, Prague, Czech Republic; 13Nantes University Hospital, Nantes, France; 14Sir Charles Gairdner Hospital, Perth, Australia; 15Istituto Nazionale Tumori IRCCS Fondazione Pascale, Naples, Italy; 16Bristol Myers Squibb, Princeton, NJ, USA; 17The Cleveland Clinic Foundation, Cleveland, OH, USA; 18UPMC Hillman Cancer Center, Pittsburgh, PA, USA *Contributed equally Objective • The purpose of this report was to present the primary results of CheckMate 76K evaluating nivolumab (NIVO) vs PBO as adjuvant treatment in patients with resected stage IIB or IIC melanoma 87% 93%82% 83% 0 20 40 60 80 100 C 5-year risk of disease recurrencea (CMMR)1 Stage IIB Stage IIC 35% 43% 86% 80%77% 74% 0 20 40 60 80 100 IIB IIC IIIA IIIB IIB IIC IIIA IIIB M SS ( % ) M SS ( % ) 5-year MSS ratea (CMMR)1,2 5-year MSS rate (AJCCv8)3 Methods • In CheckMate 76K (NCT04099251), treatment-naive patients ≥ 12 years with completely resected stage IIB or IIC melanoma were randomized 2:1 to receive NIVO or PBO as shown in Figure 2 • The primary endpoint was investigator-assessed RFS (time between randomization and first recurrence) – Recurrence events included the following: local, regional, or distant recurrence; new primary melanoma and melanoma in situ; death (due to any cause) – Imaging assessments occurred every 26 weeks during years 1–3 and every 52 weeks in years 4 and 5 • Key secondary endpoints were distant metastasis-free survival (DMFS; time between randomization and first distant recurrence or death) and safety • Recurrence was investigator-assessed with imaging assessments approximately every 6 months in the first 3 years and annually in years 4 and 5 – Tumor assessments were performed per contrast-enhanced computed tomography (CT) of the chest, abdomen, pelvis, and all other known and suspected sites of disease, unless known contraindications for CT intravenous contrast • The design of CheckMate 76K includes an optional on-protocol NIVO open-label portion following recurrence on either NIVO (at ≥ 6 months from treatment) or PBO (at any time after recurrence) – The results presented here are from the initial blinded phase portion of the study • The interim RFS analysis was planned when approximately 123 RFS events occurred (80% information fraction); a final analysis is planned at approximately 154 events • The clinical cutoff for the interim analysis was June 28, 2022 – 135 RFS events (88% information fraction); 76.8% power and 0.678 critical HR • OS is event driven and follow-up is ongoing Figure 2. Study design FFR, freedom from relapse (with censoring patients who died from causes other than disease); OS, overall survival; PFS2, progression-free survival through next-line therapy. Results Patient population • A total of 790 patients were randomized (Figure 3) – 39% of patients treated with NIVO and 25% of PBO-treated patients discontinued treatment, most commonly due to study drug toxicity for NIVO at 18% of patients and due to disease recurrence for PBO at 16% of patients • At a minimum overall study follow-up of 7.8 months, patients had a median follow-up of 15.8 months in the NIVO arm and 15.9 months in the PBO arm • Median number of doses received was 12 in the NIVO arm and 13 in the PBO arm, with a mean (range) duration of treatment of 8.8 (0-12.1) and 9.9 (0-12.7) months, respectively • Patient demographic and disease characteristics were well balanced, with 60% of the population being male and 40% at stage IIC disease (Table 1) – One-half of the patients had nodular melanoma subtype across arms Blinded NIVO/PBO treatment Optional NIVO open- label (within ≤ 3 y) after recurrence ≥ 6 months posttreatment NIVO or any time PBO NIVO IV 480 mg Q4W per patient eligibility and choice Optional on-protocol open-label NIVO treatment after first recurrence NIVO IV 480 mg Q4W for 12 months n = 526 PBO IV Q4W for 12 months n = 264 Stratify by T category R 2:1 Primary endpoint • RFS by investigator Secondary endpoints • OS • Safety • DMFS • PFS2 Exploratory endpoints • FFR • Treatment-free interval • Quality of life Treatment naive patients ≥ 12 y with • Completely resected stage IIB/C melanoma with standard wide local excision • Negative sentinel lymph node biopsy N = 790 Figure 3. Patient treatment disposition aTwo patients were not treated: 1 patient in ITT no longer met study criteria and 1 classified as “other”. bSeven patient discontinuations were related to COVID: patient request (n = 1), death (n = 1), study drug toxicity (n = 2), unrelated AE (n = 3). cTwo patient discontinuations were related to COVID: withdrew consent (n = 1) and death (n = 1). AE, adverse event; COVID, coronavirus disease; ITT, intention-to-treat. NIVO • 526 ITT • 524 safetya PBO • 264 ITT • 264 safety 64 (12%) ongoing 257 (49%) completed 203 (39%) discontinuedb • 94 (18%) study drug toxicity • 29 (6%) patient request • 26 (5%) disease recurrence • 18 (3%) withdrew consent • 16 (3%) other • 11 (2%) unrelated AE • 6 (1%) death • 1 (< 1%) lost to follow-up • 1 (< 1%) maximum clinical benefit • 1 (< 1%) no longer met criteria 790 patients randomized 39 (15%) ongoing 158 (60%) completed 67 (25%) discontinuedc • 7 (3%) study drug toxicity • 0 patient request • 41 (16%) disease recurrence • 7 (3%) withdrew consent • 7 (3%) other • 1 (< 1%) unrelated AE • 2 (1%) death • 0 lost to follow-up • 2 (1%) maximum clinical benefit • 0 no longer met criteria Table 1. Patient baseline characteristics NIVO (n = 526) PBO (n = 264) Mean age, years (SD) 59.9 (13.9) 59.3 (13.6) Male, n (%) 322 (61%) 161 (61%) ECOG PS 0, n (%) 495 (94%) 245 (93%) Stage, n (%) IIB IIC 316 (60%) 210 (40%) 163 (62%)a 101 (38%) T category, n (%) T3b T4a T4b 204 (39%) 112 (21%) 210 (40%) 104 (39%) 58 (22%) 102 (39%) Melanoma subtype, n (%) Nodular Superficial spreading Acral lentiginous Other/Not reported 266 (51%) 151 (29%) 28 (5%) 81 (15%)b 133 (50%) 82 (31%) 15 (6%) 34 (13%)c Region, n (%) Western Europe US and Canada Australia Eastern Europe 303 (58%) 97 (18%) 68 (13%) 58 (11%) 160 (61%) 46 (17%) 30 (11%) 28 (11%) aOne of these patients was incorrectly categorized as stage IIB instead of IIC. bCategorized as: desmoplastic melanoma (n = 21, 4%), lentigo maligna (n = 13, 2%), “other” (n = 44, 8%), and not reported (n = 3, 1%). cCategorized as: desmoplastic melanoma (n = 8, 3%), lentigo maligna (n = 3, 1%), “other” (n = 22, 8%), and not reported (n = 1, < 1%). • NIVO significantly reduced the risk of recurrence vs PBO, with a stratified HR of 0.42 (95% CI, 0.30-0.59) and 12-month RFS rates of 89% vs 79% (Figure 4) – Overall, 13% of patients treated with NIVO and 26% of PBO-treated patients experienced an RFS event (Table 2) • RFS benefit was driven by reductions in the incidence of both distant recurrences (5% for NIVO vs 12% for PBO) and regional recurrences (2% for NIVO vs 8% for PBO) • New melanoma lesions occurred in 2% vs 3% of patients, respectively, and likely did not have an impact on the observed RFS benefit • The RFS forest plot shows that the RFS benefit with NIVO was consistent across prespecified patient subgroups, which is illustrated by the clustering of HRs around the overall ITT unstratified HR of 0.43 (Table 3) – Of particular interest, RFS benefit was consistent across T categories with an expected slightly overall worse prognosis for patients with T4b disease (Figure 5) Figure 4. Primary endpoint: RFS NA, not available; NR, not reached. 100 90 80 70 60 50 40 30 20 10 0 R F S (% ) 0 3 6 9 12 15 18 21 24 27 30 33 Months 526 492 444 364 261 185 116 54 19 6 2 0 No. at risk NIVO PBO 264 243 205 161 119 77 40 20 11 3 2 0 89% (95% CI, 86–92) 79% (95% CI, 74–84) NIVO PBO NIVO PBO Events, n/N 66/526 69/264 Median, mo (95% CI) NR (28.5–NA) NR (21.6–NA) Stratified HR (95% CI) 0.42 (0.30–0.59) Stratified, log rank P < 0.0001 Table 2. Patterns of first RFS events NIVO (n = 526) PBO (n = 264) RFS events, n (%) Recurrencea Distant recurrence Regional node recurrence Local recurrence In-transit metastases New melanoma lesions New primary invasive melanoma Melanoma in situ Deaths prior to recurrence 66 (13%) 45 (9%) 26 (5%) 11 (2%) 8 (2%) 0 11 (2%) 4 (1%) 7 (1%) 10 (2%) 69 (26%) 58 (22%) 31 (12%) 20 (8%) 7 (3%) 0 8 (3%) 3 (1%) 5 (2%) 3 (1%) aFor patients who had multiple recurrences that were identified on the same day, the most serious type is tabulated according to the displayed prespecified hierarchy. Subgroup NIVO n/N NIVO 12-mo RFS rate, (95% CI) PBO n/N PBO 12-mo RFS rate, (95% CI) Unstratified HR (95% CI)a Unstratified HRa (95% CI) Overallb Overall 66 (526) 89.0 (85.6–91.6) 69 (264) 79.4 (73.5–84.1) 0.43 (0.31–0.61) Age category I < 65 years 33 (305) 91.5 (87.4–94.4) 39 (155) 81.2 (73.5–86.9) 0.40 (0.25–0.64) ≥ 65 years 33 (221) 85.4 (79.3–89.8) 30 (109) 76.8 (66.8–84.2) 0.48 (0.29–0.78) Age category II ≥ 18–< 65 33 (305) 91.5 (87.4–94.4) 39 (155) 81.2 (73.5–86.9) 0.40 (0.25–0.64) ≥ 64–< 75 16 (140) 89.2 (81.9–93.6) 18 (77) 79.7 (67.9–87.5) 0.45 (0.23–0.88) ≥ 75–< 85 17 (77) 78.2 (65.6–86.6) 11 (30) 68.4 (46.1–83.0) 0.46 (0.21–0.99) ≥ 85 0 (4) NA 1 (2) 100.0 (100.0–100.0) — Sex Male 39 (322) 89.9 (85.6–93.0) 51 (161) 76.5 (68.6–82.7) 0.33 (0.22–0.51) Female 27 (204) 87.6 (81.6–91.7) 18 (103) 83.9 (74.1–90.2) 0.71 (0.39–1.29) Disease stage IIB 26 (316) 92.6 (88.6–95.2) 36 (163) 84.1 (76.8–89.3) 0.34 (0.20–0.56) IIC 40 (210) 83.8 (77.5–88.4) 33 (101) 72.0 (61.6–80.0) 0.51 (0.32–0.81) T category T3b 16 (204) 92.6 (87.2–95.7) 22 (104) 83.4 (73.8–89.7) 0.36 (0.19–0.68) T4a 10 (112) 92.6 (85.1–96.4) 14 (58) 85.2 (70.7–92.8) 0.27 (0.12–0.63) T4b 40 (210) 83.8 (77.5–88.4) 33 (102) 72.3 (61.9–80.2) 0.52 (0.33–0.82) Region US and Canada 8 (97) 92.7 (84.2–96.7) 8 (46) 84.2 (67.8–92.7) — Western Europe 41 (303) 89.0 (84.5–92.3) 46 (160) 78.0 (70.0–84.0) 0.40 (0.26–0.61) Eastern Europe 8 (58) 84.5 (71.3–92.0) 9 (28) 80.2 (58.6–91.3) — Australia 9 (68) 87.9 (76.2–94.1) 6 (30) 78.8 (58.7–89.9) — Table 3. RFS by subgroup with unstratified HR aPer statistical analysis plan, HR is not computed for subset categories with less than 10 events per treatment group. bStratified RFS is 0.42 (0.30–0.59). Favors NIVO Favors PBO 0.125 1 80.25 0.5 2 4 Figure 6. Secondary endpoint: DMFS • In a descriptive analysis, the benefit of NIVO for reducing the risk of distant metastases or death was similar to the overall RFS benefit (Figure 6) – HR was 0.47 (95% CI, 0.30-0.72) with 12-month DMFS rates of 92% for NIVO and 87% for PBO • The safety profile of NIVO was similar to the known anti–PD-1 monotherapy profile observed across many trials • Grade 3-4 treatment-related adverse events (TRAEs) were 10% in the NIVO group and 2% in the placebo group and any grade TRAE leading to discontinuation was 15% vs 3% (Table 4) – There was 1 treatment-related death (0.2% of patients) with NIVO due to heart failure and acute kidney injury that was not related to myocarditis • As expected, the most common TRAE was fatigue, which occurred at a similar incidence in the NIVO and PBO arms (Figure 7) – The most frequent grade 3-4 TRAEs in the NIVO arm were diarrhea, rash, and increased alanine aminotransferase (ALT), aspartate aminotransferase (AST), and blood creatine phosphokinase, all at 1% each • The only grade 3-4 immune-mediated AE in more than 1% of patients treated with NIVO was hepatitis at 3% (Table 5) Figure 5. Subgroup analysis of RFS: T category aUnstratified. Table 4. Safety summary AE, n (%) NIVO (n = 524) PBO (n = 264) Any grade Grade 3-4 Any grade Grade 3-4 Any AE 502 (96%) 115 (22%)a 229 (87%) 32 (12%)a TRAE 433 (83%) 54 (10%) 142 (54%) 6 (2%) Immune-mediated AE 213 (41%) 41 (8%) 45 (17%) 3 (1%) Any AE leading to discontinuation 91 (17%) 37 (7%) 9 (3%) 2 (1%) TRAE leading to discontinuation 77 (15%) 29 (6%) 7 (3%) 2 (1%) aIn addition, there was 1 grade 5 event in each treatment group, considered unrelated to study treatment, myocardial ischemia for NIVO and “sudden death” for PBO. Figure 7. TRAEs in ≥ 5% patients in the NIVO group 6. Weber J, et al. Presented at the Society for Melanoma Research (SMR) International Congress; October 28–31, 2021; virtual. 7. Eggermont, AMM, et al. N Engl J Med 2018;378:1789–1801. 8. Eggermont AMM, et al. Lancet Oncol 2021;22:643–554. 9. Luke JJ, et al. Lancet 2022;399:1718–1729. Table 5. Immune-mediated AEs by categorya AE, n (%) NIVO (n = 524) PBO (n = 264) Any grade Grade 3-4 Any grade Grade 3-4 Non-endocrine immune-mediated AEs where immune-modulating medication was initiated Rash 45 (9%) 4 (1%) 4 (2%) 0 Diarrhea/colitis 24 (5%) 6 (1%) 2 (1%) 1 (< 1%) Hepatitis 22 (4%) 14 (3%) 1 (< 1%) 0 Hypersensitivity 7 (1%) 0 0 0 Pneumonitis 4 (1%) 1 (< 1%) 2 (1%) 0 Nephritis and renal dysfunction 3 (1%) 2 (< 1%) 1 (< 1%) 0 Endocrine immune-mediated AEs regardless of immune-modulating medication initiation Hypothyroidism 60 (11%) 0 0 0 Adrenal insufficiency 12 (2%) 3 (1%) 3 (1%) 0 Hyperthyroidism 40 (8%) 1 (< 1%) 4 (2%) 0 Hypophysitis 9 (2%) 5 (1%) 2 (1%) 0 Thyroiditis 6 (1%) 0 0 0 Diabetes (Type I) 3 (1%) 3 (1%) 0 0 aImmune-mediated AEs, reported between the first dose and 100 days after the last dose of study therapy, included both non-endocrine events requiring immune-modulating medication and endocrine events, regardless of treatment and not requiring specific laboratory criteria. Nivolumab Placebo Grade 3-4 1-2 25 20 15 10 5 0 Fatigue Pruritus Diarrhea Rash Arthralgia H ypo- thyroidism Asthenia Dry m outh H yper- thyroidism Nausea Increased ALT Increased AST Increased blood creatine phosphokinase Infusion- related M aculo- papular rash M yalgia 106 53 97 25n = 80 25 57 18 54 15 54 0 38 18 36 7 36 3 39 7 33 13 30 6 30 13 27 2 25 5 28 14 20% 20% 19% 9% 15% 9% 11% 7% 10% 6% 10% 0 7% 7% 7% 3% 7% 1% 7% 3% 6% 5% 6% 2% 6% 5% 5% 1% 5% 2% 5% 5% In ci d e n ce ( % ) TRAEs, n (%) NIVO (n = 524) PBO (n = 264) Any 433 (83%) 142 (54%) Grade 3–4 54 (10%) 6 (2%) 100 90 80 70 60 50 40 30 20 10 0 D M F S (% ) NIVO PBO 0 3 6 9 12 15 18 21 24 27 30 33 Months 92% (95% CI, 89–94) 87% (95% CI, 81–90) 526 506 461 381 273 194 122 55 20 7 2 0 No. at risk NIVO PBO 264 252 215 177 130 89 49 26 15 3 2 0 NIVO PBO Events, n/N 42/526 41/264 Median, mo (95% CI) NR (28.5–NA) NR Stratified HR (95% CI) 0.47 (0.30–0.72) NIVO PBO Events, n/N 10/112 14/58 Median, mo (95% CI) 28.5 (28.5–NA) 23.6 (15.7–NA) HR (95% CI)a 0.27 (0.12–0.63) 0 3 6 9 12 15 112 100 90 80 70 60 50 40 30 20 10 0 R F S (% ) Months No. at risk NIVO PBO 3318 58 21 24 27 30 109 53 98 44 82 35 57 23 41 15 26 8 14 5 5 2 2 0 1 0 0 0 93% (95% CI, 85–96) 85% (95% CI, 71–93) NIVO PBO T4a NIVO PBO Events, n/N 16/204 22/104 Median, mo (95% CI) NR NR (18.4–NA) HR (95% CI)a 0.36 (0.19–0.68) 0 3 6 9 12 15 204 93% (95% CI, 87–96)100 90 80 70 60 50 40 30 20 10 0 R F S (% ) Months No. at risk NIVO PBO 3318 104 83% (95% CI, 74–90) 21 24 27 30 187 101 166 84 136 69 99 54 79 36 47 21 17 6 10 4 2 2 1 1 0 0 NIVO PBO T3b 0 3 6 9 12 15 210 100 90 80 70 60 50 40 30 20 10 0 R F S (% ) Months No. at risk NIVO PBO 3318 102 21 24 27 30 196 89 180 77 146 57 105 42 65 26 43 11 23 9 4 5 2 1 0 1 0 0 84% (95% CI, 78–88) 72% (95% CI, 62–80) NIVO PBO T4b NIVO PBO Events, n/N 40/210 33/102 Median, mo (95% CI) NR (23.7–NA) NR (16.0–NA) HR (95% CI)a 0.52 (0.33–0.82) These data were previously presented at the Society for Melanoma Research (SMR) International Congress; October 17–20, 2022; Edinburgh, United Kingdom. mailto:Gerogina.long@sydney.edu.au