Second-Line Pembrolizumab Has Maintained Survival Benefit in Advanced HCC

January 17, 2021 - Pembrolizumab demonstrated maintained numerical improvements in overall survival and progression-free survival compared with placebo in patients with previously treated advanced hepatocellular carcinoma.

Pembrolizumab (Keytruda) demonstrated maintained numerical improvements in overall survival (OS) and progression-free survival (PFS) compared with placebo in patients with previously treated advanced hepatocellular carcinoma (HCC), according to updated findings from the randomized phase 3 KEYNOTE-240 trial that were virtually presented during the 2021 Gastrointestinal Cancers Symposium.1

At a median follow up of 39.6 months for pembrolizumab (n = 278) and 39.8 months for placebo (n = 135), the median OS was 13.9 months (95% CI, 11.6-16.0) with pembrolizumab versus 10.6 months (95% CI, 8.3-13.5) with placebo (HR, 0.77; 95% CI, 0.62-0.96; P = .0112). Moreover, the estimated OS rates at 24 months were 28.8% and 20.4% for pembrolizumab and placebo, respectively. At 36 months, the estimated OS rates were 17.7% and 11.7%, respectively.

The median PFS was 3.3 months (95% CI, 2.8-4.1) with pembrolizumab compared with 2.8 months (95% CI, 1.6-3.0) with placebo (HR, 0.70; 95% CI, 0.56-0.89; P = .0011). The estimated 24-month PFS rates were 11.8% and 4.8%, respectively. The estimated 36-month PFS rates were 9% and 0%, respectively.

“Sorafenib-treated patients saw improvement in OS and PFS that was maintained over time with pembrolizumab compared with placebo,” said Philippe Merle, MD, PhD, a professor of hepato-gastroenterology and digestive oncology at the Centre of Research in Cancerology at Lyon University and Croix-Rousse Hospital in Lyon, France, during a virtual presentation of the data. “The safety profile of pembrolizumab remained consistent over time with no new or unexpected adverse events. These data support the benefit/risk profile of pembrolizumab.”

The study enrolled adult patients with confirmed diagnoses of advanced HCC who experienced progression after or intolerance to sorafenib (Nexavar). Patients were randomized 2:1 to receive 200 mg of intravenous pembrolizumab every 3 weeks plus best supportive care versus placebo plus best supportive care for 35 cycles or until progression, unacceptable toxicity, patient withdrawal of consent, or investigator decision.

Patients were stratified by geographic region (Asia without Japan versus non-Asia with Japan), macrovascular invasion (yes versus no), and alpha-fetoprotein level (AFP; < 200 ng/mL versus ≥ 200 ng/mL).

To be eligible for enrollment, patients had to have measurable disease by RECIST v1.1 criteria, Child-Pugh liver class A disease, Barcelona Clinic Liver Cancer (BCLC) stage C disease, or BCLC stage B disease not amenable to locoregional therapy or refractory to locoregional therapy, and not amenable to a curative treatment approach, and an ECOG performance status (PS) of 0 or 1.

Prior findings of the KEYNOTE-240 trial revealed the coprimary end points of OS and PFS were numerically improved but were not statistically significant with pembrolizumab plus best supportive care versus placebo plus best supportive care.2

KEYNOTE-240 was a confirmatory trial for pembrolizumab, which was granted an accelerated approval by the FDA in November 2018 for the treatment of patients with HCC who were previously treated with sorafenib. The regulatory decision was based on data from the phase 2 KEYNOTE-224 trial.

The numerical improvements in OS and PFS were consistent and maintained, favoring pembrolizumab, with longer follow up across predefined subgroups, including age, sex, geographic region, ECOG PS, macrovascular invasion, AFP, reason for sorafenib discontinuation, extrahepatic spread, and etiology.

Further findings from the updated analysis showed that the objective response rate was 18.3% (95% CI, 14.0%-23.4%) with pembrolizumab compared with 4.4% (95% CI, 1.6%-9.4%) with placebo. The median time to response was 2.7 months (95% CI, 1.2-16.9) and 2.9 months (95% CI, 1.1-6.9), respectively. The median duration of response (DOR) was 13.9 months (range, 1.5+ - 41.9+) versus 15.2 months (range, 2.8-21.9), respectively. Moreover, 53.7% of responders in the pembrolizumab arm had a DOR of at least 12 months versus 50% of responders in the placebo arm.

Additionally, the disease control rate (DCR) was 61.9% with pembrolizumab versus 53.3% with placebo.

Best overall responses among patients who received pembrolizumab included 10 complete responses (CRs), 41 partial responses (PRs), and 121 instances of stable disease (SD). No patients achieved a CR with placebo, 6 had PRs, and 66 had SD.

Additionally, 85 patients in the pembrolizumab arm and 54 patients in the placebo arm experienced progressive disease. The median time-to-progression was 4.0 months (95% CI, 2.8-5.3) and 2.8 months (95% CI, 1.6-3.0), respectively.

Regarding safety, 61.3% (n = 171) of patients who received pembrolizumab and 48.5% (n = 65) of patients who received placebo developed any-grade treatment-related adverse effects (TRAEs). Grade 3/4 TRAEs occurred in 19% (n = 53) and 7.5% (n = 10) of patients, respectively. In the pembrolizumab arm, 6.8% (n = 19) of patients experienced TRAEs that led to treatment discontinuation, and 1 TRAE led to death. The death was attributed to malignant neoplasm progression. In the placebo arm, 0.7% of patients experienced TRAEs that led to treatment discontinuation.

AEs of special interest were observed in 17.9% (n = 50) of patients with pembrolizumab and 8.2% (n = 11) of patients with placebo. Of these, 7.2% (n = 20) and 0.7% (n = 1), respectively were grade 3/4. In the pembrolizumab group, 3.6% (n = 10) of AEs of special interest led to treatment discontinuation.

Immune-mediated hepatitis occurred in 3.6% (n = 10) of patients who received pembrolizumab and no patients who received placebo. Of these 10 patients, 8 (2.9%) received systemic corticosteroids. Additionally, no patients in either arm experienced a flare event of viral hepatitis.

References

1. Merle P, Edeline J, Bouattour M, et al. Pembrolizumab versus placebo in patients with advanced hepatocellular carcinoma previously treated with sorafenib: updated data from the randomized, phase 3 KEYNOTE-240 study. J Clin Oncol. 2021;39(suppl 3):268. http://bit.ly/39DGqU9.

2. Finn RS, Ryoo BY, Merle P, et al. Pembrolizumab as second-line therapy in patients with advanced hepatocellular carcinoma in KEYNOTE-240: a randomized, double-blind, phase III trial. J Clin Oncol. 2020;38(3):193-202. doi:10.1200/JCO.19.01307