Hepatocellular · HCC

Primary analysis: mOS NR vs 13.2 mo (HR 0.58, 95% CI 0.42-0.79, p<0.001); mPFS 6.8 vs 4.3 mo (HR 0.59, p<0.001); 12-mo OS 67.2% vs 54.6%; updated analysis (J Hepatol 2022, PMID 34902530): mOS 19.2 vs 13.4 mo (HR 0.66); etiology reported: HBV ~49%, HCV ~21%, non-viral ~30%

Population: Unresectable hepatocellular carcinoma, no prior systemic therapy

Interim mOS NR vs 10.4 mo (HR 0.57, 95% CI 0.43-0.75, p<0.0001); mPFS 4.6 vs 2.8 mo (HR 0.56, p<0.0001); etiology: **HBV >93%** (China-only cohort, design specified HBV-associated HCC); 6 vs 2 treatment-related deaths

Population: HBV-associated unresectable or metastatic hepatocellular carcinoma in China, no prior systemic therapy

mPFS 6.8 vs 4.2 mo (HR 0.63, p=0.0012) — PFS endpoint met; interim mOS 15.4 vs 15.5 mo (HR 0.90, p=0.44) — **OS endpoint NOT met**; illustrates PFS/OS disconnect in HCC; etiology HBV ~31%, HCV ~29%, non-viral ~40%

Population: Advanced hepatocellular carcinoma, no prior systemic therapy

STRIDE mOS 16.43 vs sorafenib 13.77 mo (HR 0.78, 96.02% CI 0.65-0.93, p=0.0035); 36-mo OS 30.7% vs 20.2%; durva mono OS 16.56 mo (HR 0.86, noninferior); PFS not significantly different among arms; etiology HBV 31%, HCV 27%, non-viral 41% (approximately equal across regions)

Population: Unresectable hepatocellular carcinoma, no prior systemic therapy

Updated mOS 19.2 vs 13.4 mo (HR 0.66, 95% CI 0.52-0.85); mPFS 6.9 vs 4.3 mo (HR 0.65, 95% CI 0.53-0.81); confirms durability of benefit at ~15.6 mo median follow-up

Population: Unresectable hepatocellular carcinoma, no prior systemic therapy (updated analysis, +12 mo follow-up)

mOS 22.1 vs 15.2 mo (HR 0.62, 95% CI 0.49-0.80, p<0.0001) — longest mOS reported in any 1L HCC phase 3 to date; mPFS 5.6 vs 3.7 mo (HR 0.52); etiology HBV ~77% (Asian-dominant enrollment); FDA approved March 2024

Population: Unresectable or metastatic hepatocellular carcinoma, no prior systemic therapy

mOS 21.2 vs 19.0 mo (HR 0.84, 95% CI 0.71-1.00, p=0.023 — but prespecified threshold was p≤0.019); mPFS 8.2 vs 8.0 mo (HR 0.87, p=0.047 — threshold p≤0.002); **both primary endpoints failed** despite numerical benefit; etiology: HBV 47%, HCV 20%, non-viral 33%

Population: Unresectable hepatocellular carcinoma, Child-Pugh A, no prior systemic therapy

mOS 15.9 vs 14.1 mo (HR 0.85, 95.003% CI 0.71-1.02); **noninferiority met, superiority NOT met**; ORR 14.3% vs 5.4%; DoR 36.1 vs 11.0 mo — durable responses striking; mPFS 2.1 vs 3.4 mo (HR 1.11 — tisle numerically worse PFS); etiology HBV ~60% (global multiregional)

Population: Systemic therapy-naive, histologically confirmed hepatocellular carcinoma, BCLC B/C, Child-Pugh A

mOS 23.7 vs 20.6 mo (HR 0.79, 95% CI 0.65-0.96, p=0.018); early hazard crossing — first 6 mo HR 1.65 (worse) then HR 0.61 (better); 24-mo OS 49% vs 39%, 36-mo 38% vs 24%; 12 TRAE deaths vs 3 — **doubled treatment-related mortality** warrants careful patient selection; etiology HBV ~35%, HCV ~26%, non-viral ~39%

Population: Unresectable hepatocellular carcinoma, Child-Pugh A, ECOG 0/1, no prior systemic therapy

mOS 10.7 vs 7.9 mo (HR 0.69, 95% CI 0.55-0.87, p<0.001); mTTP radiologic 5.5 vs 2.8 mo; no difference in time to symptomatic progression; **~3-month OS improvement was first ever positive systemic Rx in HCC**; predominantly European/Western cohort (HCV ~28%, alcohol ~26%, HBV ~19%)

Population: Advanced hepatocellular carcinoma, no prior systemic therapy

mOS 6.5 vs 4.2 mo (HR 0.68, 95% CI 0.50-0.93, p=0.014); mTTP 2.8 vs 1.4 mo (HR 0.57, p=0.0005); **absolute OS shorter than SHARP (6.5 vs 10.7 mo)** due to Asia-Pacific HBV-dominant, later-stage cohort; etiology: HBV ~73% (vs ~19% in SHARP)

Population: Advanced hepatocellular carcinoma in Asia-Pacific (predominantly HBV), Child-Pugh A, no prior systemic therapy

mOS 9.5 brivanib vs 9.9 sora (HR 1.06, 95.8% CI 0.93-1.22); **noninferiority NOT met** (margin ≤1.08 exceeded); brivanib failed as 1L option despite promising phase 2

Population: Unresectable advanced hepatocellular carcinoma, no prior systemic therapy

mOS 9.1 linifanib vs 9.8 sorafenib (HR 1.046, 95% CI 0.896-1.221); superiority/noninferiority both NOT met; linifanib higher toxicity burden (grade 3/4 AEs, SAEs, discontinuation all P<0.001 worse)

Population: Advanced hepatocellular carcinoma, no prior systemic therapy

mOS 10.6 vs 7.8 mo (HR 0.63, 95% CI 0.50-0.79, p<0.0001); **first positive post-sorafenib 2L phase 3 in HCC**; ITT included only sorafenib-tolerant patients (strict selection); etiology HBV 38%, HCV 21%, alcohol 24%

Population: Hepatocellular carcinoma who tolerated sorafenib and progressed on it, Child-Pugh A

mOS 10.2 vs 8.0 mo (HR 0.76, 95% CI 0.63-0.92, p=0.005); mPFS 5.2 vs 1.9 mo (HR 0.44); ORR 4% vs <1%; allowed up to 2 prior lines (broader than REACH-2/RESORCE); etiology HBV 38%, HCV 24%

Population: Previously treated advanced hepatocellular carcinoma after prior sorafenib, up to 2 prior systemic regimens

mOS 13.6 vs 12.3 mo (HR 0.92, 95% CI 0.79-1.06) — **noninferiority met, superiority NOT met**; lenva superior mPFS 7.4 vs 3.7 mo and ORR 24% vs 9% (mRECIST); different toxicity profiles (lenva=HTN/diarrhea; sora=PPE)

Population: Unresectable hepatocellular carcinoma, Child-Pugh A, no prior systemic therapy

mOS 8.5 vs 7.3 mo (HR 0.71, 95% CI 0.53-0.95, p=0.0199); mPFS 2.8 vs 1.6 mo (HR 0.45, p<0.0001); biomarker entry: **AFP ≥400 ng/mL** (after REACH in all-comers was negative, HR 0.87); etiology HBV 36%, HCV 26%

Population: Advanced hepatocellular carcinoma with AFP >=400 ng/mL, BCLC B/C, Child-Pugh A, prior 1L sorafenib

mOS 8.7 vs 6.8 mo (HR 0.785, 95% CI 0.617-0.998, p=0.048); modest but statistically significant; Chinese-only cohort; apatinib/rivoceranib = VEGFR2 TKI later used as 1L backbone in CARES-310; etiology **HBV ~82%**

Population: Advanced hepatocellular carcinoma refractory/intolerant to >=1 prior line of systemic chemotherapy or targeted therapy (HBV-endemic Chinese population)

ORR 20% in dose-expansion (3 mg/kg) including HCV/HBV/non-viral cohorts; durable responses; grade 3/4 TRAEs manageable; **supported FDA accelerated approval Sep 2017** for 2L post-sorafenib (later withdrawn Jul 2021 after CheckMate-459 OS negative and KN-240 did not meet significance)

Population: Advanced hepatocellular carcinoma with or without chronic viral hepatitis (HCV/HBV); sorafenib-untreated/intolerant or sorafenib-progressor; Child-Pugh A (expansion). Single-arm phase 1/2.

ORR 17% (95% CI 11-26); 1% CR, 16% PR, 44% SD; grade 3+ TRAE ~24%; supported **FDA accelerated approval Nov 2018** for 2L post-sorafenib HCC; etiology mixed (HBV 22%, HCV 26%, non-viral ~45%)

Population: Advanced hepatocellular carcinoma previously treated with sorafenib (intolerant or progressed), ECOG 0-1, Child-Pugh A. Single-arm phase 2.

Arm A (nivo 1 + ipi 3 mg/kg) ORR 32% (95% CI 20-47); DoR not reached; treatment-related grade 3/4 AE ~50% in arm A; **FDA accelerated approval March 2020** for 2L post-sorafenib HCC based on arm A; similar 12-mo OS ~60% across arms

Population: Advanced hepatocellular carcinoma previously treated with sorafenib. Randomized phase 1/2, three dosing arms (no non-IO comparator).

mOS 13.9 vs 10.6 mo (HR 0.781, 95% CI 0.611-0.998, p=0.0238) — **did NOT meet prespecified threshold p≤0.0174**; mPFS 3.0 vs 2.8 mo (HR 0.718, p=0.0022 vs threshold p≤0.002 — also missed); despite numerical benefit, trial technically negative; global cohort including Western patients

Population: Advanced hepatocellular carcinoma previously treated with sorafenib

1L ORR 34.3%, mPFS 5.7 mo, 12-mo OS 74.7%; 2L ORR 22.5%, mPFS 5.5 mo, 12-mo OS 68.2%; 77.4% grade 3+ TRAE (high toxicity); etiology **HBV ~82%** (China-only); provided biology rationale for CARES-310 phase 3

Population: Advanced hepatocellular carcinoma, treatment-naive (1L cohort) or refractory/intolerant to prior 1L targeted therapy (2L cohort). Single-arm phase 2.

mOS 16.4 vs 14.7 mo (HR 0.85, 95% CI 0.72-1.02, p=0.075) — **did NOT meet prespecified threshold p=0.0419**; numerical benefit but statistically negative; triggered July 2021 FDA accelerated approval withdrawal for 2L nivo mono in HCC

Population: Advanced hepatocellular carcinoma, no prior systemic therapy, Child-Pugh A, ECOG 0/1, regardless of viral hepatitis status

mOS 14.6 vs 13.0 mo (HR 0.79, 95% CI 0.63-0.99, p=0.0180) — **MET prespecified threshold** (p<0.0193); mPFS 2.6 vs 2.3 mo (HR 0.74, p=0.0032); ORR 12.7% vs 1.3%; **Asia-only cohort** (HBV ~77%), contrast with negative KN-240 global ITT

Population: Asian patients with advanced hepatocellular carcinoma, progression or intolerance to sorafenib or oxaliplatin-based chemotherapy

mTTP 169 vs 166 d (HR 0.80, P=0.072, NS). DEB-TACE + sorafenib failed to meaningfully improve time-to-progression over DEB-TACE alone. Negative legacy — ended first wave of TKI+TACE enthusiasm.

Population: Intermediate-stage multinodular hepatocellular carcinoma without macrovascular invasion or extrahepatic spread

mOS 17.8 vs 11.5 mo (HR 0.45, P<0.001); mPFS 10.6 vs 6.4 mo; ORR 54% vs 25%. Proof-of-concept for TKI + TACE before the full IO+TKI+TACE era; largely BCLC-C population (most had MVI).

Population: Advanced hepatocellular carcinoma (BCLC B/C), mostly with portal vein/macrovascular invasion or extrahepatic spread

mPFS 15.0 vs 8.2 mo (HR 0.77, P=0.032) for durva+bev+TACE vs TACE alone. Durva-alone+TACE arm did not beat placebo. First phase 3 to validate IO+anti-VEGF + TACE triplet in intermediate-stage HCC.

Population: Unresectable, embolisation-eligible hepatocellular carcinoma

mPFS 14.6 vs 10.0 mo (HR 0.66, P=0.0002); 24-mo OS 75% vs 69% (HR 0.80, NS at first interim). First positive global phase 3 of TACE + IO + TKI in intermediate-stage HCC. Asian population 72%.

Population: Unresectable, non-metastatic hepatocellular carcinoma

4y OS 67.9% (RFA/PLAT) vs 64.0% (resection), NS. First Asian RCT showing RFA non-inferior to resection for small HCC. Chinese center.

Population: Solitary hepatocellular carcinoma <=5 cm

5y OS 54.78% (RFA) vs 75.65% (resection), P=0.001 — resection SUPERIOR to RFA in Milan-criteria HCC. Opposite conclusion from Chen 2006.

Population: Hepatocellular carcinoma meeting Milan criteria

Protocol paper for SIRveNIB — the companion Asia-Pacific trial to SARAH. Final readout (Chow 2018 JCO, PMID 29498924): mOS 8.8 vs 10.0 mo, HR 1.1, NS. Also negative for Y-90 superiority over sorafenib.

Population: Locally advanced hepatocellular carcinoma without extrahepatic disease, not suitable for surgery/transplant/ablation (Asia-Pacific)

mOS 8.0 vs 9.9 mo (HR 1.15, P=0.18) — Y-90 SIRT NOT superior to sorafenib in locally advanced HCC. Negative trial for SIRT as systemic alternative.

Population: Locally advanced or inoperable hepatocellular carcinoma (French multicentre)

mOS 13.37 vs 7.13 mo (HR 0.35, P<0.001); ORR 40.8% vs 2.46%. Transforms HAIC from fringe to mainstream locoregional option in China for HCC with PVI.

Population: Hepatocellular carcinoma with portal vein invasion (Chinese multicentre)

mOS 13.9 vs 8.2 mo (HR 0.41, P<0.001) — HAIC-FOLFOX beats sorafenib head-to-head in advanced HCC. Biomarkers ATP7A/B identified.

Population: Advanced hepatocellular carcinoma, predominantly with macrovascular invasion or extrahepatic spread (Chinese multicentre)

Median RFS 3.5y (surgery) vs 3.0y (RFA), NS. Japanese Milan-subset RCT agrees with Chen 2006, disagrees with Huang 2010 on primary endpoint.

Population: Small hepatocellular carcinoma: <=3 nodules, each <=3 cm (Japanese multicentre)

Median RFS 33.3 vs 33.7 mo (HR 0.940, P=0.26, NS). Sorafenib adds no adjuvant benefit after curative-intent resection/ablation. Clear negative trial.

Population: Hepatocellular carcinoma after curative resection or ablation

ORR 23% (nivo-mono) vs 0% (nivo+ipi) by RECIST; MPR 23% vs 21%. Combo had higher G3-4 AE. Small phase 2 — signal-generating for perioperative IO in resectable HCC.

Population: Resectable hepatocellular carcinoma

4/20 (20%) achieved >70% pathological tumour necrosis with 2 cycles neoadjuvant cemiplimab; 15% radiographic PR. First published phase 2 of single-agent PD-1 neoadjuvant in resectable HCC.

Population: Resectable hepatocellular carcinoma. Single-arm phase 2 (Mount Sinai).

RFS HR 0.72 (95% CI 0.53-0.98, P=0.012) at interim — first positive phase 3 of adjuvant therapy in HCC. Grade 3-4 AE 41% vs 13%.

Population: High-risk hepatocellular carcinoma after curative-intent resection or ablation

MPR responders had ↑ intratumoral CD8 (+26.9%), granzyme B (+15.6%), PD-1 (+20.2%) post-treatment. Baseline tumour size and dynamic immune infiltration as candidate predictive biomarkers for perioperative IO response.

Population: Biopsy-proven resectable hepatocellular carcinoma (N=18 biomarker substudy of Kaseb 2022 Lancet Gastroenterol Hepatol trial)

Updated RFS HR 0.90 (95% CI 0.72-1.12) — benefit NOT sustained; OS HR 1.26 (95% CI 0.85-1.87) trending adverse. Initial positive readout (HR 0.72) reverted with maturity. Indication withdrawn 2024.

Population: High-risk hepatocellular carcinoma after curative-intent resection or ablation (updated analysis)