HCC clinical issues and imaging presentation.pptx

Best hcc

1. Agenda Role of immunotherapy in BCLC stage of HCC Approved 1 st and 2 nd line of immunotherapy for HCC as per NCCN and ESMO guidelines Pathology of Immunosuppressive Mechanisms and Immune Escape in HCC Clinical Trials of various ICI Trials combining ICI’s with TKI’s Trials for 2 nd line ICI therapies Newer ICI combinations

2. Newer Immunotherapy agents 2 nd line ICI’s post progression ICI ‘s with TACE combinations Ongoing adjuvant ICI trials Rationale for neoadjuvant ICI Role of immunotherapy post transplant Summary

3. BCLC Staging and Treatment Strategy for HCC Reig. J Hepatol. 2022;76:681. HCC Prognosis Patient Characterization First Treatment Option Ablation Resection Ablation Transplant TACE Systemic treatment BSC Based on tumor burden, liver function and physical status Refined by AFP, ALBI score, Child-Pugh, MELD To decide individualized treatment approach Very early stage (0) Single ≤2 cm Preserved liver function, PS 0 Advanced stage (C) Portal invasion and/or extrahepatic spread Preserved liver function, PS 1-2 Terminal stage (D) Any tumor burden End stage liver function, PS 3-4 Potential candidate for liver transplantation Single ≤3 nodules, each ≤3 cm Extended liver transplant criteria (size, AFP) Well defined nodules, preserved portal flow, selective access Diffuse, infiltrative, extensive bilobar liver involvement No No Yes Yes Portal pressure, bilirubin Normal Increased Contraindications to LT Intermediate stage (B) Multinodular Preserved liver function, PS 0 Early stage (A) Single, or ≤3 cm nodules each ≤3 cm Preserved liver function, PS 0 Clinical Decision-Making Treatment Stage Migration primes lower priority options due to nonliver related clinical profile (age, comorbidities, patient values, and availability) Expected Survival >2 yr 3 months TACE Not feasible or failure Successful downstaging >5 yr Not feasible or failure >2.5 yr Atezo/Bev or Durva/Tremelimumab If not feasible, sorafenib or lenvatinib or durva (or clinical trial) After sorafenib : Regorafenib, cabozantinib, or ramucirumab (if AFP ≥400 ng/ml) After atezo/bev, durva/treme, lenvatinib, or durva : Clinical Trial Cabozantinib (or clinical trial) Alternate sequencing may be considered but has not been proven 1L 3L 2 L

4. BCLC Staging and Treatment Strategy for HCC Reig. J Hepatol. 2022;76:681. HCC Prognosis Patient Characterization First Treatment Option Ablation Resection Ablation Transplant TACE Systemic treatment BSC Based on tumor burden, liver function and physical status Refined by AFP, ALBI score, Child-Pugh, MELD To decide individualized treatment approach Very early stage (0) Single ≤2 cm Preserved liver function, PS 0 Advanced stage (C) Portal invasion and/or extrahepatic spread Preserved liver function, PS 1-2 Terminal stage (D) Any tumor burden End stage liver function, PS 3-4 Potential candidate for liver transplantation Single ≤3 nodules, each ≤3 cm Extended liver transplant criteria (size, AFP) Well defined nodules, preserved portal flow, selective access Diffuse, infiltrative, extensive bilobar liver involvement No No Yes Yes Portal pressure, bilirubin Normal Increased Contraindications to LT Intermediate stage (B) Multinodular Preserved liver function, PS 0 Early stage (A) Single, or ≤3 cm nodules each ≤3 cm Preserved liver function, PS 0 Clinical Decision-Making Treatment Stage Migration primes lower priority options due to nonliver related clinical profile (age, comorbidities, patient values, and availability) Expected Survival >2 yr 3 months TACE Not feasible or failure Successful downstaging >5 yr Not feasible or failure >2.5 yr Atezo/Bev or Durva/Tremelimumab If not feasible, sorafenib or lenvatinib or durva (or clinical trial) After sorafenib : Regorafenib, cabozantinib, or ramucirumab (if AFP ≥400 ng/ml) After atezo/bev, durva/treme, lenvatinib, or durva : Clinical Trial Cabozantinib (or clinical trial) Alternate sequencing may be considered but has not been proven 1L 3L 2 L

5. NCCN.org Substantial progress since 2017! Reproduced with permission from the NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines®) for Hepatocellular Carcinoma V2.2023. ©National Comprehensive Cancer Network, Inc 2023. All rights reserved. Accessed November 14, 2023. To view the most recent and complete version of the guideline, go online to nccn.org.

6. Europe: ESMO guidelines for HCC Prognostic stage BCLC stage 0 or A BCLC stage B BCLC stage C BCLC stage D Staging Treatment choices Resection LTX [A] Systemic therapy [A] Atezolizumab + bevacizumab [A](MCBS-5*) Sorafenib [A], lenvatinib [A ](MCBS-4) BSC [A] SBRT Brachytherapy SIRT [C] 1L 2L Ablation [A] TACE [B] LTX Resection [A] TACE [A] SIRT [C] After Sorafenib - Regorafenib [I A ] Cabozantinib [I A ] Ramucirumab ‡ [I A ] After Atezo+Beva - Sorafenib [V,C ] Lenvatinib [V,C ] Regorafenib [V,C] Cabozantinib [V,C] Ramucirumab ‡ [V,C]

7. Immune Checkpoint Inhibition (ICI): Combination Strategies in Advanced HCC ICI + anti-VEGF mAb ICI + antiangiogenic TKI ICI + ICI Aref . Lab Chip. 2018;18:312.9. Yi. Mol Cancer. 2019;18:60. Foerster. Cancers (Basel). 2021;13:1962. Figures used under terms and conditions of the Creative Commons Attribution 4.0 International license (CC BY 4.0: https://creativecommons.org/licenses/by/4.0/). CD8+ T cell Tumor cell Antigen-presenting cell MHC TCR PD-L1 PD-1 CD28 B7 B7.1/2 (CD80/86) CTLA-4 Anti-CTLA-4 Anti-PD-(L)1 Infiltration FasL VEGF Infiltration Adhesion molecule Pretreatment: Angiogenesis Post-treatment: Vessel normalization ICI + antiangiogenesis Adhesion molecule FasL VEGF PD-1 PD-L1 ICI Antiangiogenesis IFN- ⲩ Tumor cell TAM CTL M1-like phenotype Mature DC Imm ature DC EC Treg M2-like phenotype

8. Cancer immunotherapy can target the multiple immune checkpoints involved in evading the immune response CTLA-4, cytotoxic T-lymphocyte-associated protein 4; MHC, major histocompatibility complex PD-1, programmed cell death protein 1; PD-L1/2, programmed death ligand-1/2 Adapted from 1. Li et al. J Hematol Oncol 2018 Copyright © 2018, Springer Nature Promoting T-cell activation Antigen-presenting cell or cancer cell B7-1 / B7-2 GI TR L O X 40L I C OSL CD 28 GI T R OX 40 I C OS T - cell B 7 -1 PD - 1 PD-L1 CT L A - 4 LAG -3 TIM-3 B T LA P D-L1 PD -L2 B 7 - 1 B 7 - 2 MHC class II GAL -9 H VEM Inhibiting T-cell activation Antigen-presenting cell or cancer cell Anti-CTLA4 (ipilimumab, trem e l imum a b ) Anti-PDL1 (atez o l izu ma b , durvalumab, avelumab ▼ ) Anti-PD1 (nivolumab, p em b r o liz u m a b )

9. Immunosuppressive Mechanisms and Immune Escape in HCC The liver has a complex immune microenvironment. It is continuously exposed to various antigens passing through the portal vein, especially those from intestinal tract. Therefore , the liver microenvironment continues to show immune tolerance, which is to inhibit inappropriate inflammatory reaction and prevent autoimmune liver injury The specific immune system of HCC and tumor cells constitute a special immune tolerance microenvironment, which can protect tumor cells from the attack of their own immune system and promote the immune escape of tumor cells 

10. Mechanisms of immunosuppression in HCC 1) chronic inflammation (e.g. viral hepatitis B and C) and chronic disease (e.g. liver cirrhosis )- many inhibitory cytokines (e.g. IL-10, IL-35 and TGF-β) are constantly produced, and a large number of immunosuppressive cells, such as regulatory T cells ( Tregs ), M2 macrophages, and myeloid-derived suppressor cells (MDSCs), are recruited into the liver  2) Immunosuppressive cells in tumor tissue can promote HCC tolerance . Tumor-associated monocytes, for example, can significantly increase the glycolysis level in the area around the tumor. Activation of glycolysis induced these cells to express PD-L1 (through NF- κB signaling pathway) and decreased the function of cytotoxic T lymphocyte

11. HCC cells can release some cytokines, such as 14-3-3 ζ, which can destroy the activation, proliferation and anti- tumor function of tumor -infiltrating T lymphocytes (TILs) ( 16 ). Beyond that, overexpression of 14-3-3 ζ can also differentiate naive T cells from effector T cells to Tregs Tumor-associated macrophages (TAMs), as one of the key components constituting the immunosuppressive microenvironment of HCC, not only cannot eliminate tumor cells, instead will promote tumor growth The expression of immune checkpoints in HCC tissues is increased ( 5 ,  17 ,  18 ). The combination of immune checkpoints and their respective ligands will inhibit the activation and proliferation of T cells  

12. Activation or alteration of some genes and signaling pathways may promote immune escape in HCC ( 19 ). For example, activation of β-Catenin ( 20 ) or mutation of  CTNNB1  ( 21 ) may promote immune escape in HCC Epithelial-to-mesenchymal-transition (EMT) can induce the up-regulation of PD-L1, PD-L2, CD73 and B7-H3; and reversing EMT can inhibit the expression of these markers ( 22 ).

13. First line Immunotherapy as per NCCN Guidelines Atezolizumab + Bevacizumab Durvalumumab + Tremelimumab

14. IMbrave150 : First-line Atezolizumab + Bevacizumab vs Sorafenib for Advanced HCC Multicenter, randomized trial Finn. NEJM. 2020;382:1894. Treatment until loss of clinical benefit or intolerable toxicity Patients with locally advanced or metastatic and/or unresectable HCC with no previous systemic therapy; Child-Pugh A; EGD within past 6 mo; ECOG PS ≤1* † (N = 501) Atezolizumab 1200 mg Q3W + Bevacizumab 15 mg/kg Q3W (n = 336) Sorafenib 400 mg BID (n = 165) Primary endpoints: OS and PFS *Trial included subgroups of high-risk patients excluded from other contemporary phase III trials: ≈40% had macrovascular invasion; specifically included patients with 50% hepatic involvement or main portal vein invasion or invasion of the portal vein branch contralateral to the primarily involved lobe. † Excluded patients with untreated or incompletely treated varices with bleeding or high risk for bleeding. Key secondary efficacy endpoints: ORR and DoR per RECIST 1.1 Stratified by r egion (Asia, excluding Japan/rest of world), ECOG PS (0/1), presence or absence of macrovascular invasion and/or extrahepatic spread, baseline a-fetoprotein (<400/≥400 ng/mL)

15. IMbrave150: Updated Results Finn. NEJM. 2020;382:1894. Cheng. J Hepatol 2022;76:862. Updated Analysis RECIST 1.1 Atezo + Bev (n = 326) Sorafenib (n = 159) Confirmed ORR, % (95% CI) 30 (25-35) 11 (7-17) CR, n (%) 25 (8) 1 (<1) PR, n (%) 72(22) 17 (11) SD, n (%) 144 (44) 69 (43) DCR, n (%) 241 (74) 87 (55) PD, n (%) 63 (19) 40 (25) Ongoing response, n (%) 54 (56) 5 (28) Median DoR, mo (95% CI) 18.1 (14.6-NE) 14.9 (4.9-17.0) Established new benchmarks for efficacy in first-line: Median OS: 19.2 vs 13.4 mo (HR: 0.66) Median PFS: 6.9 vs 4.3 mo (HR: 0.65) Durable responses in 30% (CR in 8%) Acceptable safety in carefully screened population: Treatment-related grade 3/4 AE: 43% vs 46% Discontinuation for AE: 15.5% vs 10.3% m OS, mo (95% CI) Atezo + Bev (n = 336) 19.2 (17.0-23.7) Sorafenib (n = 165) 13.4 (11.4-16.9) HR: 0.66 (95% CI: 0.52-0.85) P = .0009 mPF S, mo (95% CI) Atezo + Bev (n = 336) 6.9 ( 5.7 -8.6) Sorafenib (n = 165) 4.3 ( 4.0 -5.6) HR: 0.65 (95% CI: 0.53-0.81) P = .0001

16. HIMALAYA: Tremelimumab + Durvalumab vs Sorafenib as First-line Treatment for HCC Randomized, open-label, multicenter study Adults with confirmed uHCC, Child-Pugh; BCLC (not eligible for locoregional therapy), No prior systemic therapy for HCC; ECOG PS 0/1; no main portal vein thrombosis (N = 1171) STRIDE: Tremelimumab 300 mg (1 dose) + Durvalumab 1500 mg Q4W (N = 393) Sorafenib 400 mg BID (N = 389) Durvalumab 1500 mg Q4W (N = 389) Stratified by aetiology of liver disease (HBV/HCV/nonviral); Macrovascular invasion (yes/no); ECOG PS (0/1) Primary objective: OS superiority (STRIDE vs sorafenib) S econdary objectives: OS noninferiority (durvalumab vs sorafenib), 36-month OS rate, PFS, ORR and DCR, s afety Abou-Alfa. NEJM Evid. 2022;1:EVIDoa2100070.

17. 6 12 18 24 30 36 42 48 54 60 HIMALAYA: Survival By Response Long-term OS benefit was observed for patients treated with STRIDE, regardless of response 3-yr and 4-yr OS nearly 45% and 36% (respectively) in patients who achieved disease control with STRIDE Patients with CR or PR: 4-yr OS ~67% Patients with SD: 4-yr OS ~25% BOR, n (%) ITT Long-term Survivors (≥3 Yr ) STRIDE (n = 393) Sorafenib (n = 389) STRIDE (n = 103) Sorafenib (n = 64) CR 12 (3.1) 12 (11.7) PR 67 (17.0) 20 (5.1) 41 (39.8) 10 (15.6) SD 157 (39.9) 216 (55.5) 39 (37.9) 45 (70.3) PD 141 (35.9) 118 (30.3) 10 (9.7) 6 (9.4) NE 16 (4.1) 35 (9.0) 1 (1.0) 3 (4.7) DCR 236 (60.1) 236 (60.7) 92 (89.3) 55 (85.9) Sangro. World Congress GI Cancer. 2023. Abstr SO-15. OS, % Disease Control No Disease Control STRIDE (n = 236) Sorafenib (n = 236) STRIDE (n = 157) Sorafenib (n = 153) 36 mo 44.6 27.9 9.7 6.8 48 mo 36.2 20.3 8.7 6.8 36-mo OS 48-mo OS Sorafenib (DC) 1.0 0.8 0.6 0.4 0.2 0.0 Probability Mo OS STRIDE (no DC) Sorafenib (no DC) STRIDE (DC)

18. Summary: First-line Treatment Efficacy Trial IMbrave150 HIMALAYA Agent(s) Atezo + Bev Sor Sor Durva + Trem Durva Sor Median OS, mo (95% CI) 19.2 (17.0-23.7) 13.4 (11.4-16.9) 13.4 (11.4-16.9) 16.4 (14.2-19.6) 16.6 (14.1-19.1) 13.8 (12.3-16.1) 18 mo, % 24 mo, % 36 mo, % 52 -- -- 40 -- -- 40 -- -- 49 41 31 47 40 25 42 33 20 HR for death (95% CI) 0.66 (0.52-0.85) 0.78 (0.65-0.92*) 0.86 (0.73-1.03 † ) P value .0009 .0035 NR Median PFS, mo (95% CI) 6.8 (5.7-8.3) 4.3 (4.0-5.6) 3.78 (3.68-5.32) 3.65 (3.19-3.75) 4.07 (3.75-5.49) ORR (RECIST 1.1), % (95% CI) 30 (25-35) 11 (7-17) 20.1 (–) 17.0 (–) 5.1 (–) Cheng. J Hepatol. 2022;76:862. Abou-Alfa. NEJM Evid. 2022;1. *96.02% CI. † 95.67% CI.

19. Safety With Immunotherapy Regimens for Advanced HCC AE, % 1,2 Atezo + Bev (n = 329) Sorafenib (n = 156) Durva + Trem (n = 388) Sorafenib (n = 374) Any Grade Grade 3/4 Any Grade Grade 3/4 Any Grade Grade 3/4 Any Grade Grade 3/4 TRAE 3,5 86 43 95 46 75.8 25.8 84.8 36.9 AE leading to d/c 9 14 Grade 5 AE 4.6 8 irAEs req steroids 1,4 12 20.1 1.9 Most common AEs Hypertension 1,4 * 30 15 24 12 5.9 1.8 18.2 6.1 Fatigue 26 2 32 6 26 3.9 30 6 Proteinuria 20 3 7 0.6 NR NR Pruritus 19 10 23 6 0.3 Diarrhea 19 1.8 49 5 27 6 45 4.3 Rash 12 17 2.6 32 2.8 57 12 Musculoskeletal pain NR NR 22 2.6 17 0.8 Hemorrhage events 3,4 30 9 18 6 11.3 3.9 15.0 4.3 1. Atezolizumab PI. 2. Durvalumab PI. 3. Cheng. J Hepatol. 2022;76:862. 4. Abou-Alfa. NEJM Evid. 2022;1. 5. Abou-Alfa. ASCO GI 2022. Abstr 379.

20. ICI + TKI Combinations in HCC: COSMIC-312: cabozantinib + atezolizumab vs sorafenib as first-line treatment for advanced HCC Improved PFS but no difference in OS Serious TRAEs: 18% (combination) vs 8% ( sorafenib ) LEAP-002: lenvatinib + pembrolizumab vs lenvatinib as first-line therapy for advanced HCC No difference in OS or PFS Serious TRAEs: 25% ( len + pembro ) vs 17% ( len + placebo ) Cares-310 - Phase III study comparing the efficacy and safety of camrelizumab (anti-PD-1) + rivoceranib (VEGFR2 inhibitor) vs sorafenib in previously untreated HCC OS: Median 22.1 vs 15.2 mo (HR: 0.62) PFS: Median 5.6 vs 3.7 mo (HR: 0.52) ORR: 25.4% vs 5.9% by RECIST 1.1

21. Second-line ICI Therapies With US Accelerated Approval Pembrolizumab KEYNOTE-240 : did not improve OS but showed better ORR 18.3% Nivolumab + ipilimumab CheckMate 040 : ~30% deep, durable responses Steroids required in ~25% to 50% Finn. JCO. 2020;38:193. Qin. GI ASCO 2022. Abstr 383. Roessler. J Cancer Res Clin Oncol. 2023;149:3065. Yau . JAMA Oncol. 2020;6:e204564 . KEYNOTE-394 : improved OS, PFS, and ORR compared to placebo 20 40 60 80 100 OS (%) 4 8 12 16 20 24 28 32 36 40 44 48 Mo Pembrolizumab Placebo Median OS, Mo (95% CI) 13.9 (11.6-16.0) 10.6 (8.3-13.5) HR (95% CI) 0.77 (0.62-0.96) Nominal P = .0112 Small, retrospective series suggests potential for benefit after atezo + bev Nivo 1 mg/kg Q3W Ipi 3 mg/kg Q3W Nivo 3 mg/kg Q2W Ipi 1 mg/kg Q6W Nivo 3 mg/kg Q3W Ipi 1 mg/kg Q3W Change From Baseline (%) Change From Baseline (%) Change From Baseline (%) 100 80 60 40 20 -20 -40 -60 -80 -100 100 80 60 40 20 -20 -40 -60 -80 -100 -100 -80 -60 -40 -20 20 40 60 80 100

22. Summary: Current IO Approaches in Advanced HCC Atezo + bev and durva + treme (STRIDE) are category 1 first-line systemic treatment options Choice based on individual patient comorbidity/risk factors (eg, varices, cardiovascular disease, autoimmune disease, patient preferences) TKIs such as lenvatinib and sorafenib or single-agent durvalumab are first-line options in patients not eligible for ICI or combination therapy Second-line treatment options include TKI, pembrolizumab, nivo + ipi Choice guided by first-line treatment and response, toxicity No prospective second-line data after ICI-based first-line combinations to date Only a subset of patients (≤~30%) experience deep and durable responses with frontline ICI-based combinations

23. New ICI Combination: CheckMate 9DW Multicenter, randomized, open-label phase III trial of nivolumab + ipilimumab as first-line NCT04039607. Patients with advanced HCC; no previous systemic therapy; Child-Pugh A; ECOG PS ≤1 (N = 732) Nivolumab 1 mg/kg + Ipilimumab 3 mg/kg Q3W (4 doses) followed by Nivolumab monotherapy Sorafenib or Lenvatinib Primary endpoints: OS Secondary endpoints: ORR, DoR, TTSD

24. NCT04524871. Finn. ASCO 2023. Abstr 4010. New ICI Combination: Atezo + Bev + Tiragolumab MORPHEUS-Liver Open-label, randomized phase Ib/II umbrella study Patients with locally advanced/metastatic and/or unresectable HCC; no prior systemic therapy ; underwent biopsy (target N = 400) Atezo + Bev + Tiragolumab (n = 40) Atezo + Bev + Tocilizumab Atezo + Bev (n = 18) Atezo + Bev + Other Agents Atezo + Bev + TPST-1120 RO7247669 + Bev Primary endpoint: ORR Secondary endpoints: PFS, OS, DoR, DCR, safety ORR: 42.5% Median PFS: 11.1 mo Similar safety to atezo + bev Led to launch of phase III IMbrave152 trial for first-line advanced HCC

25. New ICI Combination: Atezo + Bev + Tiragolumab IMbrave152/SKYSCRAPER-14 Double-blind, placebo-controlled, randomized phase III study Patients with unresectable HCC; no previous systemic therapy ( systemic adjuvant therapy permitted if recurrence within 6 mo); Child-Pugh A; ECOG PS ≤1 (Target N = 650) Tiragolumab 600 mg IV + Atezolizumab 1200 mg IV + Bevacizumab 15 mg/kg IV Q3W Atezolizumab 1200 mg IV + Bevacizumab 15 mg/kg IV Q3W + Placebo NCT05904886 . Finn. ASCO 2023. Abstr 4010. Primary endpoints: PFS by investigator, OS Secondary/exploratory endpoints: ORR, DoR, landmark PFS/OS, safety, QoL/PRO, biomarker analyses Until loss of clinical benefit or unacceptable toxicity

26. Emerging immunotherapies in HCC

27. IMbrave251: randomized, 2-arm phase III study Primary endpoint: OS Secondary endpoints: PFS, ORR, DoR, TTP, TTD in PROs, safety Second-line ICI Post Progression Patients with HCC ineligible for resection or local ablation that progressed following atezolizumab + bevacizumab (estimated N = 554) Atezolizumab 1200 mg IV Q 21D + Lenvatinib 8 or 12 mg QD or Sorafenib 800 mg QD Lenvatinib 8 or 12 mg QD or Sorafenib 800 mg QD Stratified by selected TKI (lenvatinib vs sorafenib); disease etiology (HBV/HCV vs nonviral); ALBI score at baseline (grade 1 vs 2/3); AFP Until loss of clinical benefit or unacceptable toxicity NCT04770896.

28. Ongoing IO Studies in Child-Pugh B Selected prospective, multicenter, systemic therapy trials requiring Child-Pugh B for eligibility Trial Title Treatment Arms Design NCT A Study Evaluating Atezolizumab, With or Without Bevacizumab, in Patients With Unresectable Hepatocellular Carcinoma and Child-Pugh B7 and B8 Cirrhosis Atezo + bev vs atezo Parallel-cohort phase II NCT06096779 Testing Immunotherapy for Patients With Liver Cancer and Moderately Altered Liver Functions Tislelizumab Phase II NCT05622071 Durvalumab and Tremelimumab as First Line Treatment in Participants With Advanced HCC (SIERRA) Durva + treme (STRIDE) Phase IIIb NCT05883644

29. Ongoing Phase III HCC Clinical Trials: TACE Combinations 1. NCT03778957. 2. NCT04246177. 3. NCT04340193. 4. NCT04803994. 5. NCT04777851. Trial Treatment Arms Primary Endpoint(s) Current Status* EMERALD-1 1 Durvalumab + TACE vs durvalumab + bevacizumab + TACE vs TACE + placebo PFS Active, not recruiting LEAP-012 2 Lenvatinib + pembrolizumab + TACE vs TACE + placebo PFS, OS Active, not recruiting CheckMate 74W 3 Nivolumab + ipilimumab + TACE vs nivolumab + TACE vs TACE Time to TACE progression, OS Active, not recruiting ABC-HCC 4 Atezolizumab + bevacizumab vs TACE Time to failure of treatment strategy Recruiting RENOTACE 5 Regorafenib + nivolumab vs TACE PFS Not yet recruiting *As of May 2023.

30. EMERALD-1: TACE ± Durvalumab and/or Bevacizumab for Locoregional HCC Multicenter, randomized, double-blind, placebo-controlled phase III study Primary endpoint: PFS (arm B vs C) Secondary endpoint: PFS (arm A vs C), OS, QoL and symptoms (EORTC) NCT03778957. Sangro. EASL Liver Cancer Summit 2020. Abstr P-347. Patients with intermediate-stage HCC; ineligible for curative therapy; no prior TACE or systemic therapy; Child-Pugh Score class A-B7; ECOG PS ≤1; no evidence of extrahepatic disease (N = 600) Arm A: TACE* + Durvalumab (n = 200) Arm C: TACE* + Placebo (n = 200) Arm B: TACE* + Durvalumab (n = 200) Durvalumab + Placebo Placebo + Placebo Durvalumab + Bevacizumab 5-yr follow-up *Drug-eluting bead or conventional TACE.

31. Ongoing Adjuvant ICI Studies Adjuvant Therapy Control Arm Design Patient Number Primary Endpoints NCT Durvalumab ± bevacizumab Placebo Randomized phase III 888 RFS NCT03847428 (EMERALD-2) Nivolumab Placebo Randomized phase III 530 RFS NCT03383458 (CheckMate-9DX) Atezolizumab + bevacizumab Active surveillance Randomized phase III 662 RFS NCT04102098 (IMbrave050) Pembrolizumab Placebo Randomized phase III 950 RFS, OS NCT03867084 (Keynote 937) Enrolling patients with HCC: At h igh risk for recurrence after resection or ablation Child-Pugh A www.clinicaltrials.gov

32. IMbrave050: Atezolizumab + Bevacizumab vs Active Surveillance for Resected or Ablated High-Risk HCC Randomized, open-label, multicenter, phase III trial Primary endpoint: RFS by independent review Key secondary endpoints: OS, RFS by investigator, time to recurrence Qin. Lancet. 2023;402:1835. NCT04102098. Patients with n ewly diagnosed HCC; curative resection or ablation within prior 4-12 wk; high risk of recurrence*; Child-Pugh A ; ECOG PS ≤1; 1 cycle of TACE permitted (N = 668) Atezolizumab 1200mg IV Q3W + Bevacizumab 15mg/kg IV Q3W (n = 334) Active Surveillance (n = 334) 12 mo or 17 cycles of active treatment, d isease recurrence or unacceptable toxicity Stratified by r egion, composite factor including high-risk features, ablation vs resection, and use of TACE *Based on composite criteria including tumor size >5 cm; tumor number >3; and presence of macrovascular or segmental portal vein invasion and/or grade 3/4 pathology. Crossover permitted

33. Imbrave 050 Overall survival is better with surveillance group,compared to Atezo + Bevaci group TRAE are more with Atezo + evaci group compared to surveillance group

34. IMbrave050: Safety Summary *Esophageal varices hemorrhage and ischemic stroke; 1 related to atezo + bev, 1 related to bev only. Atezo + Bev (n = 332) Active Surveillance (n = 330) IMbrave150 (n = 329) Treatment duration, median, mo Atezo: 11.1 Bev: 11.0 NA Atezo: 7.4 Bev: 6.9 Patients with ≥1 AE, n (%) 326 (98.2) 205 (62.1) 323 (98.2)  Treatment-related AE 293 (88.3) NA 276 (83.9) Grade 3/4 AE , n (%) 136 (41.0) 44 (13.3) 186 (56.5) Treatment-related grade 3/4 AE 116 (34.9) NA 117 (35.6) Serious AE , n (%) 80 (24.1) 34 (10.3)    125 (38.0) Treatment-related serious AE 44 (13.3) NA 56 (17.0) Grade 5 AE , n (%) 6 (1.8) 1 (0.3) 15 (4.6) Treatment-related grade 5 AE 2 (0.6) * NA 6 (1.8) AE leading to interruption of any study treatment , n (%)   155 (46.7) NA 163 (49.5) AE leading to withdrawal from any study treatment , n (%) 63 (19.0) NA 51 (15.5) Qin. Lancet. 2023;402:1835. Chow. AACR 2023. Plenary. Finn NEJM. 2020;382:1894.

35. Adjuvant Trials in HCC Agent Study Design Sample Size NCT Nivolumab Randomized phase III 545 NCT03383458 (CheckMate 9DX) Pembrolizumab Randomized phase III 950 NCT03867084 (KEYNOTE-937) Camrelizumab + apatinib Randomized phase III 687 NCT04639180 Durvalumab + bevacizumab Randomized phase III 908 NCT03847428 (EMERALD-2)

36. Liver Resection or Ablation and Adjuvant Therapy: Current Guidance Tumor recurrence complicates 70% of cases at 5 yr Singal . Hepatology. 2023;78:1922. Galle. J Hepatol. 2018;69:182. NCCN. Clinical practice guidelines in oncology: h epatocellular carcinoma. v.2.2023. nccn.org. Organization Recommendation AASLD (Dec 2023) “AASLD recommends use of adjuvant immune checkpoint inhibitor-based systemic therapy in patients at high risk* of recurrence after liver resection or local ablation.” EASL (Apr 2018) “Neoadjuvant or adjuvant therapies are not recommended because they have not been proven to improve the outcome of patients treated with resection.” NCCN (Sept 2023) “To date, no adjuvant therapies have been shown to have benefit but there are ongoing clinical trials.” “Currently, no adjuvant therapies have been shown to have added value post-ablation.” “Data are currently too preliminary for the panel to provide specific recommendations regarding immunotherapy treatment in an adjuvant setting.” *High-risk features include tumor size >5 cm, >3 tumors, microvascular or macrovascular invasion, and poor tumor differentiation.

37. Rationale for Neoadjuvant Immunotherapy 38 O’Donnell. Clin Can Res. 2019;25:5743. Neoadjuvant immunotherapy further seeks to use primary tumor as source of antigens (“in situ vaccination”) Also may improve resectability Reducing tumor burden Testing disease biology

38. Neoadjuvant ICI Therapy Has Demonstrated Benefit in Melanoma Phase II trial evaluated whether neoadjuvant and adjuvant pembrolizumab compared with adjuvant administration only would affect outcomes in 313 patients with stage III/IV melanoma Patel. NEJM. 2023;388:813. 0.2 0.4 0.6 0.8 1.0 Mo Since Randomization Probability of EFS (%) 12 24 36 6 18 30 154 159 69 67 25 22 1 2 96 98 46 40 17 10 Patients at Risk, n Neoadjuvant/adjuvant group Adjuvant only group P = .004 by log-rank test Adjuvant only group Neoadjuvant/adjuvant group

39. Perioperative Nivolumab ± Ipilimumab for Resectable HCC Randomized phase II study of perioperative nivolumab ± ipilimumab* for patients with HCC eligible for surgical resection (N = 27) Kaseb. ASCO 2020. Abstr 4599. *Preoperative nivolumab 240 mg Q2W x 3 doses ± ipilimumab 1 mg/kg Day 1; postoperative nivolumab 480 mg Q4W for 2 yr ± ipilimumab 1 mg/kg Q6W x 4 doses. Outcome, n (%) Total (N = 27) Nivolumab (n = 13) Nivolumab + Ipilimumab (n = 14) Pathologic complete response 5 (19) 2 (15) 3 (21) Major pathologic response 3 (11) 1 (8) 2 (14) Grade ≥3 AEs, pre-op 2 (7) 2 (14) Grade ≥3 AEs, post-op 4 (17) 2 (15) 2 (14)

40. Neoadjuvant Cemiplimab for Resectable HCC Single-arm, open-label phase II trial in 21 patients with resectable HCC Neoadjuvant cemiplimab 350 mg IV Q3W x 2 prior to surgical resection Marron. Lancet Gastroenterol Hepatol. 2022;7:219. Pathologic Tumor Necrosis at Resection, n (%) Patients (n = 20) Significant tumor necrosis (>70%) 4 (20) Complete tumor necrosis (100%) 3 (15) Tumor necrosis ≥50% 7 (35) Tumor necrosis <50% 13 (65)

41. Perioperative Strategies in HCC 42 Mark Yarchoan Strategy Overview and Current Status Neoadjuvant radioembolization Downstaging potential (ORR: 72.2% [mRECIST] in LEGACY study) Not anticipated to reduce micrometastasis No randomized data vs upfront surgery Neoadjuvant immunotherapy – based regimens Some downstaging potential (ORR: ~35% [mRECIST] with modern systemic therapy) Anticipated to reduce micrometastasis Multiple exploratory phase I studies demonstrating feasibility; larger studies needed Adjuvant immunotherapy Multiple phase III studies ongoing (all vs placebo): EMERALD-2: durvalumab ± bevacizumab (NCT03847428) KEYNOTE-937: pembrolizumab (NCT03867084) CheckMate-9DX: nivolumab (NCT03383458) Salem. Hepatology. 2021;74:2342.

42. Role of Immuno therapy in Patients With Liver Transplant

43. Posttransplant I mmunotherapy H as H igh R isk of ACR and G raft L oss Checkpoint Inhibitor Liver Allograft Rejection, n (%) Time to Rejection, Days (Range) Ipilimumab 1/3 (33) 13 Nivolumab 2/4 (50) 12.5 (7-18) Pembrolizumab 1/3 (33) 7 Ipilimumab plus pembrolizumab 0/1 (0) - All 4/11 (36) 10 (7-18) Abdel-Wahab. J Immunother Cancer. 2019;7:106. OS in Patients W ith and Without Allograft Rejection Overall Survival 1.0 0.8 0.6 0.4 0.2 0.0 5 15 10 25 20 30 Mo Allograft rejection No allograft rejection P = .03 Median OS: 12 mo (95% CI: 8-16) Median OS: 5 mo (95% CI: 1-9)

44. VEGF-Targeted Therapies Have Activity After Sorafenib: Phase III Data Regorafenib : multitargeted TKI Bruix. Lancet. 2017;389:56. Abou-Alfa. NEJM. 2018;379:54. Zhu. Lancet Oncol. 2019;20:282. RESORCE CELESTIAL REACH-2 Regorafenib vs placebo Cabozantinib vs placebo (N = 707) Ramucirumab vs placebo 2L, sorafenib-tolerating patients only (N = 573) 2L or 3L (N = 707) 2L, AFP ≥400 ng/mL (N = 292) Median OS: 10.6 vs 8.0 mo Median OS: 10.2 vs 8.0 mo Median OS: 8.5 vs 7.3 mo HR: 0.63 ( P <.0001) HR: 0.76 ( P = .005) HR: 0.71 ( P = .0199) Cabozantinib : multitargeted TKI Ramucirumab: anti-VEGFR2 Ab 33 Placebo 100 80 60 40 20 Mo 3 6 12 15 18 21 24 27 30 9 Regorafenib OS (%) Probability of OS 1.0 Mo 42 3 6 9 12 15 21 24 27 30 33 36 39 18 0.8 0.6 0.4 0.2 Cabozantinib Placebo Mo 3 6 9 12 15 18 21 24 27 OS (%) 100 80 60 40 20 Ramucirumab Placebo

45. 46 Progress has been made in 1L systemic treatment of HCC in recent years Positive superiority phase III trial Positive non-inferiority phase III trial Phase II/III positive trial Negative phase III trial a Sorafenib ( S H A R P ) 1 ¶ Sunitinib 2 Erlotinib ( SEA RCH) 3 Lenvatinib (R E FL E CT ) 5 Lenvatinib + pe m b r o li z u m a b (LEAP-002) 12 2 1 9 Atezolizumab + b e v aciz u ma b * ¶ § (IMbrave150) 7 2020 Sintilimab + b e v -b i o similar** (ORIENT-32) 8 2 2 1 Nivolumab (CheckMate 459) 6 Donafenib** 9 2 7 2 1 8 2 022 Durvalumab + tremelimumab (HIMALAYA) 11 Atezolizumab + cabozantinib (C OS M I C- 312 ) 1 Tislelizumab (RATIONALE-301) 13 Rivoceranib + ca m r e li z u m a b 1 4

46. Atezolizumab MOA- A humanized monoclonal anti-programmed death-ligand 1 (PD-L1) antibody that inhibits PD-L1–programmed death 1 (PD-1) PD-L1–B7-1 signaling , thereby resulting in tumor -specific cytotoxic T-cell immunity Atezolizumab is FDA-approved for intravenous use only Dosing is generally 1200 mg IV every three weeks AE- Anaphylaxis and hypersensitivity Immune-mediated colitis is characterized by signs and symptoms of diarrhea or increased ostomy output, colitis, and or perforation Immune-mediated cutaneous adverse reactions are common and manifest as itching, maculopapular rash, lichenoid reactions, vitiligo, and pruritus.

47. Immune-mediated endocrinopathies can present with signs and symptoms of adrenal insufficiency, thyroiditis, hypothyroidism, hyperthyroidism, hypophysitis , and diabetes Immune-Mediated Hepatitis typically manifests with elevation in liver function tests. Immune-Mediated pneumonitis is characterized by nonproductive cough, shortness of breath, and radiological abnormalities of the lung. Immune-mediated Renal Dysfunction and Nephritis The incidence of immune-mediated neurological toxicity is 1 % and is characterized by clinical symptoms of polyneuropathy, facial nerve palsy, aseptic meningitis, transverse myelitis, myasthenia gravis, Guillain Barre syndrome(GBS)

48. Bevacizumab

49. Summary: Future Directions for IO in HCC New IO approaches being studied in advanced HCC include: New ICI combinations, eg, IMbrave152 (tiragolumab + atezo + bev vs atezo + bev + placebo) Novel IO strategies such as cell therapy, vaccines, ADCs, others Expanding advanced HCC contexts including second-line, Child-Pugh B Emerging roles for ICI in earlier stages include adjuvant and combinations with liver-directed therapy: Adjuvant post resection or ablation: interim analysis of IMbrave050 showed improved RFS with atezo + bev; await longer follow up data for mature RFS and OS outcomes BCLC B: EMERALD-1 reported improved PFS with durvalumab + bev combined with TACE; await full presentation of data Multiple other ICI studies are ongoing in adjuvant and intermediate HCC