Interpreting Key Trial Data in the Era of IO in Advanced HCC

Immunotherapy-based combinations are reshaping the frontline treatment landscape for advanced unresectable hepatocellular carcinoma (HCC), with multiple phase 3 trials reporting improved survival and response outcomes over prior standards. In a virtual Case-Based Roundtable event, Mark Yarchoan, MD, associate professor of oncology at Johns Hopkins Medicine, led a conversation with oncologists in the Great Lakes/Great Plains region discussing emerging clinical trial data and real-world considerations influencing regimen selection and outcomes in this treatment setting.

Register today to join a Case-Based Roundtable near you.

This article is part 2 of a 2-part series from a Case-Based Roundtable event.

DISCUSSION QUESTIONS

• What are your impressions of the efficacy outcomes from CheckMate-9DW (NCT04039607)?¹
  • How do you interpret the magnitude of overall survival (OS) improvement in the context of clinical practice?
  • How does the objective response rate (ORR) or duration of response (DOR) impact your perception of the regimen’s benefit?
• When comparing nivolumab (Opdivo) plus ipilimumab (Yervoy) with lenvatinib (Lenvima) or sorafenib (Nexavar), what stands out to you?
  • Most patients in the lenvatinib/sorafenib arm received lenvatinib (85% vs sorafenib 15%); how does that impact your interpretation of the efficacy findings?

Mark Yarchoan, MD: These are the new data for HCC. Are people impressed by this or not impressed? Trial comparisons are obviously always fraught, but how does this stack up in your mind vs other things?

Yan Ji, MD: I think the data are impressive, although I have to say that I'm a little bit worried about the first 12 months. The nivolumab-ipilimumab curve looks like we lost some patients. Like you said, is that because of toxicity, or is it because of other [factors]? But I think in response rate and DOR, that's what we have seen in immuno-oncology [IO]-IO regimens in other types of cancer, so that's always what convinces us and our patients that it's worth it, because we want to find those long-term survivors. I'm curious to learn what kind of toxicity we're going to face, because in melanoma, the reason why we change to ipilimumab 1 mg/kg plus nivolumab 3 mg/kg is because the ipilimumab 3 mg/kg is too toxic. And there are some patients who may get better with one cancer, but then they develop lifelong toxicity, either type 1 diabetes or hypophysitis. I'm just curious how toxic this regimen is compared with others.

Yarchoan: The rate of grade 3/4 adverse events [AEs] was 28%.¹

Vijay Rao, MD: Toxicity is definitely a concern. Also, could you comment on the [finding that the] comparator arm has done much better on the nivolumab-ipilimumab trial than the other ones? Why would that be?

Yarchoan: All the other trials were compared with just sorafenib. This trial was compared with lenvatinib because most of the patients got lenvatinib. And even though lenvatinib is not inferior to sorafenib, I think anybody who looks at lenvatinib vs sorafenib, the hazard ratio [HR for OS] is 0.85.²

Rao: Lenvatinib seems to have done better than its approving trial where the HR is 10.8 and here it's…significantly better than the original lenvatinib trials.

Yarchoan: I think this is always the case. Sorafenib was the comparator for a decade, right? You have trials going back to 2007 and every single time sorafenib is used as the control arm, it does a little bit better. I don't know if we're getting patients on systemic therapy earlier or if we have better subsequent second-line therapy, but I think it's a great point.

Priya Kumar, MD: Overall, in my clinical practice, I can never replicate the results of clinical trials. I think it's rare to find Child-Pugh A patients. It's rare to find the same kind of patients that go into your clinical trials, and therefore the results are less impressive than what I've seen generally discussed.

Yarchoan: I feel that also as someone who sees patients.

DISCUSSION QUESTIONS

• How do you interpret the efficacy data from IMbrave150 (NCT03434379)³ and HIMALAYA (NCT03298451)⁴ in the context of real-world use?
  • Are there specific patient populations where you see a clear preference for one regimen over the other?

Yarchoan: Everyone answered that they had a lot of experience with either atezolizumab [Tecentriq]-bevacizumab [Avastin] or durvalumab [Imfinzi]-tremelimumab [Imjudo]. So, we've talked a little bit about this; all our patients are maybe not Child-Pugh A, not performance status 0 to 1. How has your experience been compared with the trials?

Rao: Honestly, not very good. Like Dr Kumar just said, I rarely see patients who are [performance status] 0 or 1, and it's a different population than a clinical trial population, so I have rarely seen the efficacy that is shown in the clinical trials, unfortunately.

Yarchoan: Is that because patients are sicker, they decompensate, and they end up in the hospital?

Rao: That's right. To begin with, they come in with…a performance status of 2, at least the last few that I've seen. They're probably [performance status] 3, they’re Child-Pugh B; they're a different patient population, I would say.

Bipinkumar Amin, MD: I have only treated 1 patient a few years ago, and at that time, Mayo Clinic recommended atezolizumab-bevacizumab, and [my patient] didn't have any AEs. He hung on for more than about 9 months. So that's my only experience; I don't have much IO-IO experience in HCC.

Kumar: Why do you think the nivolumab-ipilimumab arm did better—in general, when you do cross-trial comparisons—than the durvalumab-tremelimumab?

Yarchoan: Certainly, the response rate was higher with nivolumab-ipilimumab. The OS looks favorable, but I think it's hard to compare across studies. The caution that I would give is [to consider the response rates for] sorafenib [across studies]… Is it [2]%, which was in the SHARP trial [NCT00105443],⁵ or is it 15%, which was in the IMbrave study?³ There are very big differences in the way the same drug can be evaluated in different studies with different populations.

What do I think, having used these regimens? I do think that nivolumab-ipilimumab may have a functionally higher CTLA-4 dose, and I think that's supported by the higher response rate, the longer DOR, but also the marginally higher toxicity [in Checkmate-9DW]: 12% vs 28% for grade 3/4.¹ I think there's a regimen for every patient, and it's nice to have options. I think if I had a patient who I really thought was only going to get 1 line of therapy and had bulky disease, that's a patient [whom] I probably wouldn't go with the STRIDE [single tremelimumab regular interval durvalumab] regimen, just because of the 20% response rate.

For a patient who's going to have 3 lines of therapy, starting with something that's not going to cause a lethal immune-related event is also very reasonable. Does anybody else have thoughts?

Madhu Midathada, MD: I think we see both extremes. There's this very sick population that you just have that one chance, and who may not even make it through that first cycle or second cycle and are done with treatments. And there's that one sector who do extremely well and are living a long time. I have a 95-year-old on atezolizumab-bevacizumab and [has an] amazing response. I sent her to interventional radiology [IR] to see if they can do any local treatment, and they said, it's all gone with whatever you've done, and there's nothing left. So, you see both extremes, and it's just hard to gauge at the get-go, what's the right treatment for which patient? …I've used all 3 regimens, and I can't say this regimen is the best or that regimen was better, maybe because I don't see as many patients as you see.

Yarchoan: Yeah, it is really wonderful when you get these complete responses. And then the IR folks are sort of wondering if there is going to be a role for local therapies, but I think there probably will be for most patients for a long time.

Register today to join a Case-Based Roundtable near you.

DISCLOSURES: Yarchoan receives grant/research support (to Johns Hopkins) from Bristol-Myers Squibb, Exelixis, Incyte and Genentech; receives honoraria from Exelixis, AstraZeneca, Replimune, Hepion, Lantheus, Genentech and Incyte; is the co-inventor of a patent related to neoantigen vaccines; and is a co-founder with equity of Adventris Pharmaceuticals.

REFERENCES

1‌. Galle PR, Decaens T, Kudo M, et al. Nivolumab (NIVO) plus ipilimumab (IPI) vs lenvatinib (LEN) or sorafenib (SOR) as first-line treatment for unresectable hepatocellular carcinoma (uHCC): First results from CheckMate 9DW. J Clin Oncol. 2024;42(17_suppl):LBA4008-LBA4008. doi:10.1200/jco.2024.42.17_suppl.lba4008

2. Dipasquale A, Marinello A, Santoro A. A comparison of lenvatinib versus sorafenib in the first-line treatment of unresectable hepatocellular carcinoma: Selection criteria to guide physician's choice in a new therapeutic scenario. J Hepatocell Carcinoma. 2021;8:241-251. Published 2021 Apr 15. doi:10.2147/JHC.S270532

3. Cheng AL, Qin S, Ikeda M, et al. Updated efficacy and safety data from IMbrave150: Atezolizumab plus bevacizumab vs. sorafenib for unresectable hepatocellular carcinoma. J Hepatol. 2022;76(4):862-873. doi:10.1016/j.jhep.2021.11.030

4. Sangro B, Chan SL, Kelley RK, et al. Four-year overall survival update from the phase III HIMALAYA study of tremelimumab plus durvalumab in unresectable hepatocellular carcinoma. Ann Oncol. 2024;35(5):448-457. doi:10.1016/j.annonc.2024.02.005

5. Llovet JM, Ricci S, Mazzaferro V, et al. Sorafenib in advanced hepatocellular carcinoma. N Eng J Med. 2008;359(4):378-390. doi:10.1056/nejmoa0708857