ODAC Unanimously Supports Continued Approval of Pembrolizumab for HCC in Sorafenib-Pretreated Patients

In an 8 to 0 vote, the FDA’s Oncologic Drug Advisory Committee opted to continue to accelerated approval of pembrolizumab for the treatment of patients with hepatocellular carcinoma who have been previously treated with sorafenib.

In an 8 to 0 vote, the FDA’s Oncologic Drug Advisory Committee (ODAC) opted to continue the accelerated approval of pembrolizumab (Keytruda) for the treatment of patients with hepatocellular carcinoma (HCC) who have been previously treated with sorafenib (Nexavar).1

“The accelerated approval for pembrolizumab in the second-line setting was appropriate based on a modest response rate seen in KEYNOTE-224 and a clear unmet need in a patient population that needs more effective therapies,” voting ODAC member Christopher Lieu, MD, associate professor, University of Colorado Medicine, told Targeted Oncology, in an interview. “I believe the key issue for this approval is that the landscape for HCC treatment has changed dramatically, and there is now an approval for immunotherapy in an earlier line setting which brings into question the continued need for this approval in the refractory setting. In the refractory setting, patients are now highly likely to have received some form of immune checkpoint therapy previously. However, if we assume 15% to 20% of patients cannot receive bevacizumab in the first-line setting, is it reasonable to continue an indication in an immunotherapy-naïve population in the second-line setting where an overall survival benefit may exist? I believe this answer to be yes, until we have final results from the KEYNOTE-394 study.”

Pembrolizumab was granted accelerated approval for this indication in November 2018 in the phase 3 KEYNOTE-244 clinical trial (NCT02702414) of 104 patients with HCC who had disease progression on or after sorafenib or were intolerant to sorafenib. KEYNOTE-244 was a single arm, multinational trial which assessed objective response rate per central review as its primary end points. Secondary end points in the study were duration of responses (DORs), disease control rate, time to progression, progression-free survival (PFS), and overall survival (OS).2

The study population included 72 patients with Child-Pugh (CP) A5 disease, and 22 with CP A6 disease. Baseline characteristics showed that 21 patients were positive for Hepatitis B virus seropositive, 25 were positive for Hepatitis C virus seropositive, and 9 had both. In addition, 17 patients have a vascular invasion, 64 had extrahepatic disease, and 38 had an alpha-fetoprotein of at least 400.

Pembrolizumab was administered to patients at 200 mg every 3 weeks until unacceptable toxicity or confirmed disease progression. Treatment was continued for up to 24 months and tumor status was assessed every 9 weeks.

An ORR of 17% (95% CI, 11-26) was achieved with pembrolizumab in KEYNOTE-244 per RECIST criteria. Responses lasted greater than 6 months in 89% of patients and greater than 12 months in 56%. The modified RECIST analysis showed an ORR of 15% (95% CI, 9%-24%).

To continue the approval of pembrolizumab for the treatment of patients with HCC who have been previously treated with sorafenib, a confirmatory trial was required. Pembrolizumab was therefore analyzed further in the phase 3 KEYNOTE-240 trial of pembrolizumab 200 mg administered every 3 weeks in combination with best supportive care (BSC) compared with placebo plus BSC in 413 patients. The coprimary end points explored in the confirmatory trial were PFS and OS.1,2

A total of 278 patients received pembrolizumab with BSC, and 135 received a placebo with BSC. Baseline characteristics were comparable between the 2 treatment arms.1

According to data from the final analysis of KEYNOTE-240, the median OS achieved 13.9 months (95% CI, 11.6-16.0) in the pembrolizumab arms compared with 10.6 months (95% CI, 8.3-13.5) in the BSC alone arm (HR, 0.781; 95% CI, 0.611-0.998; P = .0238). The result failed to meet the prespecified P-value required for statistical significance (P = .0174).

Longer-term follow-up of the HCC population showed a median OS of 13.9 months (95% CI, 11.6-16.0) in the pembrolizumab combination arm compared with 10.6 months (95% CI, 8.3-13.5) in the BSC alone arm (HR, 0.77; 95% CI, 0.62-0.96; P = .0112).

The median PFS observed in the primary analysis of KEYNOTE-240 was 3.0 months (95% CI, 2.8-4.1) in the pembrolizumab/BSC arm versus 2.8 months (95% CI, 2.5-4.1) in the placebo/BSC arm (HR, 0.775; 95% CI, 0.609-0.987; P = .01186). The PFS results also failed to meet the threshold for statistical significance (P = .002). Longer-term follow-up for PFS showed a median of 3.3 months (95% CI, 2.8-4.1) with pembrolizumab plus BSC versus 2.8 months (95% CI, 1.6-3.0) in the placebo/BSC arm (HR, 0.70; 95% CI, 0.56-0.89; P = .0011).

In response to these data, Lieu stated: “Though the KEYNOTE-240 study did not reach its prespecified statistical significance, there is a believable separation of the curves, that appeared to be maintained in longer term follow-up. There is also a confirmatory study in KEYNOTE-394 that I believe will either support this data or potentially prove that pembrolizumab is simply not effective enough in this setting. We will know this potentially within 1 month.”

As the treatment paradigm for patients HCC who previously received sorafenib continues to shift, experts question whether the benefit-risk profile of pembrolizumab warrants its continued FDA approval. The topics of the most concern include immune-related adverse events (AEs), the fact that most patients do not derive benefit, and that fact the confirmatory trial did not meet either coprimary end point. The benefits of pembrolizumab in both clinical trials were the long DOR and the fact that the drug offers a different mechanism of action compared with other tyrosine kinase inhibitors (TKIs) on the market. The is also uncertainty about the bleeding risk associated with pembrolizumab in this patient population.

In response to a review of the data during the ODAC meeting, Richard S. Finn, MD, stated: “I and other clinicians feel strongly that single-agent pembrolizumab fulfills an unmet need in clinical practice. At least 15% to 20% of patients will not receive atezolizumab (Tecentriq) and bevacizumab (Avastin) in the front-line setting, and for those patients, first-line TKIs have been shown to improve survival and progression. A decision whether to continue with the TKI or offering them an [immuno-oncology] agent remains an important decision point.”

Another attendee, Cathy Eng, MD, FACP, FASCO, David H. Johnson Chair in Surgical and Medical Oncology; co-leader, Gastrointestinal (GI) Cancer Research Program; professor of Medicine (Hematology and Oncology); co-director, GI Oncology; vice-chair, SWOG GI Committee; and director, VICC Young Adult Cancers Initiative at Vanderbilt-Ingram Cancer Center, explained why she favored the continued approval of pembrolizumab for patients with HCC in the second-line setting.

“HCC patients are sorely in need of treatment options give the associated mortality. They need these options based upon acceptability, convenience of treatment schedule, reduced toxicities, and especially in the era of the pandemic. So, keeping in mind that the need of quality is exemplified by this year's ASCO scheme ‘equitable care for all patients,’ we need to take into account our underserved minority and remote patient population. Decreasing treatment options is not be approached to improve community outreach, and I do not believe that discontinuation of pembrolizumab’s for accelerated approval as a second-line option is in the best interest of the patient,” said Eng.

ODAC voting members were asked whether the indication for the pembrolizumab monotherapy in patients previously treated with sorafenib should be maintained pending conduct or completion of additional trials.

In agreement with the voting of the other committee members, Lieu added, “I support the current indication for pembrolizumab as written.”


1 April 27-29, 2021: Meeting of the Oncologic Drugs Advisory Committee Meeting Announcement. FDA website. April 29, 2021. Accessed April 29, 2021. https://bit.ly/2PyhdV1

2 FDA approves Merck’s KEYTRUDA® (pembrolizumab) for the treatment of patients with hepatocellular carcinoma (HCC) who have been previously treated with sorafenib. News release. Merck. November 9, 2018. Accessed April 29, 2021. https://bit.ly/32XjbS0