Key Takeaways in the Treatment of HCC - Episode 6
Richard Finn, MD: Initially, CheckMate 040 looked at nivolumab in a second-line population, and this was initially a relatively small phase I/II study that grew in size as the data were very supportive. We saw there that about 14% to 15% of patients had significant responses, and the duration of response was very long. At the same time, there was KEYNOTE-224, which was a single-arm study with pembrolizumab in a very similar population of about 100 patients. It had a very similar response rate of around 15% to 18%, and the patients who responded had a very long duration of response—well over a year.
Both of these drugs had very tolerable safety profiles. We see immune-induced phenomena—essentially anything, including hypothyroidism, hyper-colitis, and things we’ve seen in other diseases. There was not really an increased liver toxicity, which was obviously a concern for patients who have cirrhosis. Based on this favorable disease control profile and favorable safety profile, both nivolumab and pembrolizumab got accelerated approval in the second-line setting. After that, there were 2 phase III studies done. First was the CheckMate-459 study for which there was a press release recently that suggested it did not meet its statistical endpoint in frontline versus sorafenib.
Then, there’s KEYNOTE-240, which I presented at this year’s ASCO [American Society of Clinical Oncology] meeting. That study also did not reach the statistical endpoints, but this was a very high bar for statistical significance. There was a clear treatment benefit with this drug, with improvement of survival of over 3 months. The placebo arm in this population lived 10 months, which is longer than any other study in the second-line setting. With pembrolizumab, they lived over 13 months. This was a hazard ratio of 0.78, which is a 22% decrease in the risk of death, and the P value here was .027. That is well below the commonly used .05, but because of the statistical design, it had to be .015. Therefore, we can’t say that it met its statistical endpoint, but there was a clear treatment effect.
Still, we saw the response rate in this study of 18%, and the duration of response was over 14 months. It confirmed all the data that we saw in the second-line setting. There are still other phase III studies pending with pembrolizumab whose results we’re waiting for to try to boost the statistical confidence that this drug is important in liver cancer.
Transcript Edited for Clarity