International, open-label phase 2 study of regorafenib plus pembrolizumab in patients with advanced hepatocellular carcinoma (HCC) previously treated with immune checkpoint inhibitors (ICI).

4007 Background: The optimal second-line treatment for HCC after ICI is not established. Regorafenib is approved for treatment of advanced HCC after sorafenib. The primary aim of this study was to evaluate the activity of regorafenib plus pembrolizumab in patients with advanced HCC who progressed on only one prior ICI regimen. Methods: Patients aged ≥18 years, CP Class A, BCLC stage B or C, and ECOG PS of 0 or 1 received oral regorafenib 90 mg once daily 3 weeks (wk) on/1 wk off plus i.v. pembrolizumab 400 mg every 6 wk. The daily regorafenib dose could be escalated to 120 mg after the first 4-wk cycle of regorafenib if tolerated. Pembrolizumab dose reductions were not allowed. Cohorts were defined by prior treatment: Cohort 1 = atezolizumab + bevacizumab; Cohort 2 = any other ICI regimen (alone or combination). Primary endpoint was overall response rate (ORR) by independent central review (RECIST 1.1). Results: 95 patients were treated in 8 countries; Cohort 2 appeared to have more favorable disease characteristics (Table). Most common prior ICIs in Cohort 2 were durvalumab (30%), nivolumab (30%), ipilimumab (22%), and pembrolizumab (19%). Overall, median follow up was 7.1 months (mo). Median duration of regorafenib/pembrolizumab treatment (including interruptions, delays) was shorter in Cohort 1 vs Cohort 2 (11.0/9.4 wk vs 21.4/24.1 wk). ORR was 5.9% in Cohort 1 and 11.1% in Cohort 2; stable disease rates were 48.5% and 63.0%, respectively. Median PFS was 2.8 mo in Cohort 1 and 4.2 mo in Cohort 2. Overall, treatment-emergent adverse events (TEAE) were grade 3 in 56% of patients and grade 4 in 5%; drug-related grade 3 and 4 TEAE occurred in 37% and 3%, respectively. One patient (Cohort 1) had a grade 5 drug-related TEAE (cardiac arrest). Most common drug-related TEAE were palmar-plantar erythrodysesthesia syndrome (39%), asthenia (33%), decreased appetite (32%), diarrhea (28%), and hypertension (20%). Biomarkers assessed in paired biopsies (baseline and wk 6/day 1) showed on-treatment reductions in macrophages and angiogenesis RNA signatures. Conclusions: To our knowledge, this is the first prospective evaluation of a kinase inhibitor plus an ICI following first-line ICI therapy for HCC. Regorafenib plus pembrolizumab had modest activity in second line after first-line ICI regimens. The safety profile of the combination was consistent with those of either drug. Optimal treatment for patients with advanced HCC who progress after an ICI regimen remains an unmet need. Biomarker findings will be detailed in presentation. Clinical trial information: NCT04696055 . [Table: see text]