KEYNOTE-394 a win for Asians with advanced HCC

Combining pembrolizumab with best supportive care (BSC) improved survival and response when used as second-line therapy in Asians with advanced hepatocellular carcinoma (HCC).

This was the pivotal finding from the phase III randomized, double-blind, KEYNOTE-394 trial presented at ASCO GI 2022.

Pembrolizumab achieved a median overall survival (OS) of 14.6 months vs 13.0 for placebo (hazard ratio [HR], 0.79). The prespecified p-value required for statistical significance was 0.0193, and the trial showed a p-value of 0.0180 for OS. The rates of OS were 34.3 percent and 24.9 percent for pembrolizumab and placebo, respectively, at 24 months. [J Clin Oncol 2022;40(suppl4):383]

Pembro for 2L tx of advanced HCC

“The findings reinforce observation in globally-conducted studies of pembrolizumab for the second-line treatment of advanced HCC,” said study author Dr Shukui Qin from Jinling Hospital, the Nanjing University of Chinese Medicine in Nanjing, China, during his presentation. “These also provide support for the generalizability of our data worldwide.”

KEYNOTE-394 was conducted on the heels of pembrolizumab’s success both in the KEYNOTE-240 and KEYNOTE-224 trials. [J Clin Oncol 2020;38:193-202; Lancet Oncol 2018;19:940-952]

As for the KEYNOTE-394 trial, 453 patients from Asia with previously treated advanced HCC  were randomly assigned 2:1 to pembrolizumab 200 mg every 3 weeks + BSC (n=300) vs placebo every 3 weeks + BSC (n=153). The BSC lasted for up to 35 cycles.

The patients had to be treated previously with sorafenib or oxaliplatin-based chemotherapy to be eligible for enrolment, with Child-Pugh class A disease, Barcelona Clinic Liver Cancer stage B or C, and an ECOG performance status of 0 or 1.

The primary endpoint was OS. Secondary endpoints included progression-free survival (PFS), objective response rate (ORR), and duration of response (DOR).

Qin said 50.7 percent of patients in the pembrolizumab group and 66.7 percent in the placebo group received systemic anticancer therapy after treatment discontinuation. He added this could have attenuated the observed treatment effect in patients who progressed.

PFS, ORR, and DOR outcomes

The 24-month PFS rate was 11 percent with pembrolizumab vs 9 percent with placebo. Median PFS was 2.6 months vs 2.3 months with pembrolizumab and placebo, respectively, (HR, 0.74; p-value 0.0032). This surpassed the prespecified 0.0134 required for statistical significance.

ORR was 12.7 percent with pembrolizumab vs 1.3 percent with placebo (p<0.0001). Six patients achieved complete response with pembrolizumab vs 1 with placebo, whereas partial response was seen in 32 of those on pembrolizumab vs 1 with placebo.

The median DOR was 23.9 months and 5.6 months with pembrolizumab and placebo, respectively.

Bonus win in meta-analysis

When a meta-analysis of KEYNOTE-394 and KEYNOTE-240 (with similar patient population treated with pembrolizumab + BSC) was conducted in the intent-to-treat groups of both studies,  there was a meaningful improvement in the pembrolizumab arm than the placebo arm.

The median OS was 14.2 months with pembrolizumab relative to 12.5 months with placebo (HR, 0.79).

“Pembrolizumab was associated with a manageable adverse event profile, consistent with [the toxicity] profile in previous studies,” Qin reported.

Nearly 70 percent of patients receiving pembrolizumab experienced treatment-related adverse events (TRAEs) of any grade vs 49.7 percent of patients receiving placebo.