Ramucirumab as second-line treatment in patients with advanced hepatocellular carcinoma (HCC) and elevated alpha-fetoprotein (AFP) following first-line sorafenib: Pooled efficacy and safety across two global randomized Phase 3 studies (REACH-2 and REACH)

Background: Ramucirumab (RAM), a human IgG1 monoclonal antibody, inhibits ligand activation of VEGFR2. REACH and REACH-2 were two global, randomized, double-blind, placebo (PBO)-controlled multicenter, phase 3 studies of RAM vs PBO in patients with HCC after prior sorafenib. REACH-2 was designed to confirm the ramucirumab treatment benefit for patients with baseline AFP ≥400 ng/mL first observed in the prespecified subgroup of patients in REACH with AFP ≥400 ng/mL. The primary endpoint of REACH-2 was met demonstrating an improved overall survival (OS) compared to PBO, a result consistent with that in patients with AFP ≥400 ng/mL in REACH. Pooled analyses of patients from REACH-2 and REACH with baseline AFP ≥400 ng/mL was performed. Methods: With the exception of AFP levels, eligibility for both REACH-2 and REACH were similar, and included patients who had advanced HCC, Child-Pugh <7, ECOG PS 0 or 1; BCLC stage C or B refractory/not amenable to locoregional therapy, ≥1 measurable lesion; adequate hematologic and biochemical parameters, and disease progression during or following, or intolerance to sorafenib. All patients received RAM (8 mg/kg) I.V. or PBO every 14 days. In both studies, patients were treated until disease progression/unacceptable toxicity/withdrawal criterion was met. Pooled analyses include assessment of OS, progression-free survival (PFS), objective response rate (ORR), and safety. Pooled efficacy analysis were stratified by study to account for any potential difference in the two studies. Results: 542 patients were pooled from REACH (250) and REACH-2 (292) (N = 316 RAM vs N = 226 PBO). Pooled baseline patient characteristics were balanced between arms including baseline AFP which had been imbalanced in REACH-2. Similar to the individual studies, RAM treatment significantly improved OS (median OS 8.1 mo vs 5.0 mo PBO; HR 0.694; 95% CI 0.571, 0.842; p=.0002). Improvements in PFS (median PFS 2.8 mo vs 1.5 mo PBO; HR 0.572; 95% CI 0.472, 0.694; p<.0001), ORR (5.4% RAM vs 0.9% PBO (p=.0040), and disease control rate (ORR + stable disease = 56.3% RAM vs 37.2% PBO [p<.0001]) were observed. The relative dose intensity was 98.3% RAM vs 99.6% PBO. 9.5% of patients on the RAM arm vs 3.6% on PBO discontinued study treatment due to adverse events (AEs) related to study treatment. Hypertension (12.0% RAM vs 3.6% PBO) and hyponatremia (5.1% RAM vs 2.2% PBO) were the only Grade (G) ≥3 treatment-emergent adverse events (TEAEs) occurring in ≥ 5% in RAM. Conclusions: A pooled analysis of two phase 3 trials assessing ramucirumab as second-line treatment in patients with HCC following first-line sorafenib (REACH-2 and REACH) demonstrates a significant and clinically meaningful benefit with a favorable safety profile in HCC patients with baseline AFP ≥400 ng/mL.