Dr. Harding:
Welcome to First-Line HCC in Focus: Survival, Safety, and Real World Evidence, an ASCO Post Roundtable. I'm Dr. James Harding from Memorial Sloan Kettering Cancer Center. Joining me today are two of my dear colleagues, Drs. Katie Kelley and Anthony El-Khoueiry. Please allow you to introduce yourself.
Dr. Kelley:
Thanks so much, Dr. Harding. It's a pleasure to be here. I'm Katie Kelley. I'm a Professor of Clinical Medicine at University of California, San Francisco. Thank you.
Dr. El-Khoueiry:
I'm Anthony El-Khoueiry. I'm a GI medical oncologist at University of Southern California, Los Angeles, focused on hepatobiliary cancers and phase I drug development.
Dr. Harding:
Thank you both so much. So in our third and final segment, we'll discuss the management of a patient with metastatic hepatocellular carcinoma after failure of first-line immunotherapy. And so let's get right into the case. And in our case series so far, just to remind the audience, this is a 64-year-old woman who was diagnosed with liver-limited hepatocellular carcinoma. She was initially treated with local regional therapy with an intended downstage for either surgical resection or transplant. Unfortunately, although tolerating local regional therapy very well at 3 months post-procedure, her CT scan showed multiple new liver lesions, new pulmonary metastasis, an increase in AFP, as well as a portal vein PV1 tumor thrombus. EGD was conducted at that time, no evidence of varices. And at the time the only available regimen was atezolizumab and bevacizumab, and so she was treated accordingly with that. And in the next segment of the case, she tolerated the atezolizumab and bevacizumab very well.
She did develop some hypothyroidism over the course of treatment with serial TFTs indicating that, that was treated with levothyroxine with improvement that was asymptomatic. On her first scan, she actually had a minor partial response to treatment that continued for about 10 months. There was a nadir of AFP of about 200. And you can see the nadir scan, which shows a minimal burden of disease in the lung at this time. Unfortunately though, at a 12-month scan, there was clear evidence of progression of disease. There was increase in all of the pulmonary metastasis, liver, lesions. AFP increased actually to 10,000. And at that time, patient was clinically well, still with intact liver functions, Child-Pugh A6, ECOG performance status was 0 to 1.
So now here we have the patient well into a disease course and unfortunately now progression of disease, but fit for more treatment. So I now look to my colleagues to help guide the discussion. Dr. El-Khoueiry, what do you think about when you see a case like this?
Dr. El-Khoueiry:
First, Dr. Harding, I want to emphasize the importance of how this patient was managed and followed with regular re-staging scans every 2 months or so, because it's very important to monitor the disease carefully. This is a disease where the PFS is still less than 1 year. So regular imaging is important to allow to switch the therapy to a second-line treatment in a timely fashion before deterioration of liver function and performance status. So that's number one.
Number two, this is a space where we actually struggle a bit. Even though we have multiple options, we do not have level 1 evidence randomized data as to what the best therapy is after failure of combination immunotherapy. So, this is an area where I prioritize clinical trials, as this is an area ripe for drug development. You mentioned the Wnt beta-catenin alteration in this patient previously, so that could be a potential clinical trial option there.
Outside of a clinical trial, we tend to use data that's generated in the post-sorafenib era for the different TKIs, or we use phase II data as well as real-world evidence to also use the TKIs. So we don't really have a particular order in which they should be given. So briefly, lenvatinib can be used, cabozantinib can be used, certainly regorafenib, ramucirumab, but certainly, the evidence as far as to the best sequence, is really lacking. So I'll stop here just to allow Dr. Kelley to also comment.
Dr. Kelley:
No, thank you, Dr. El-Khoueiry. I fully agree with those points. I guess I would add that not knowing exactly what year this patient was originally treated and where she was at this in time point at recurrence or progression, I should say, there is some data from case series about activity of a CTLA-4 inhibitor post-atezolizumab/bevacizumab progression. And so in patients that are quite fit and still eligible for nivolumab and ipilimumab in combination. I would also counsel that case series suggests maybe a 10% to 20% response rate in patients who had progressed on atezolizumab/bevacizumab previously.
Dr. Harding:
I think that's a great point. So from my perspective, I do as well prioritize clinical trials. We see in addition to molecular targets, there are a number of GPC3-targeted approaches, immunotherapy, ADCs, radioligands. So there's very active development. Unfortunately for this patient, we didn't have that option at the time. But one thing that I've often heard about is, should we continue the backbone of atezolizumab? There's this hypothesis there that you continue to treat with an immune backbone and then alter your anti-angiogenic.
Dr. Kelley, what are your thoughts about that approach? And do we have any recent evidence that would help guide us?
Dr. Kelley:
Yeah, definitely the right question to be asking, and we've all been scratching our heads about this for a while now. But as of last weekend or earlier this week, I should say, we have data from this year ASCO-2026 in the form of the IMbrave251 trial, which asked that very question as Dr. Harding posed, what if a patient who had previously benefited from, then progressed on the combination of atezolizumab plus bevacizumab, then continued the atezolizumab with the addition of a different partner drug, in this case, a tyrosine kinase inhibitor. The trial allowed physician's choice of either lenvatinib or sorafenib, continued with or added to the continuation of atezolizumab in the experimental arm of the study.
The control arm was the physician's choice of tyrosine kinase inhibitor, again, either lenvatinib or sorafenib alone as monotherapy. And interestingly, as you allude to, this trial did answer the question. It unfortunately did not show any benefit from continuation of atezolizumab, and showed very similar outcomes, both PFS-wise and OS-wise with a slight favoring of the OS in the control arm.
For predominantly a lenvatinib-based study, the physician's choice was lenvatinib, and I believe over 90% of the time. So what we learned was that lenvatinib is an appropriate second-line choice. The overall survival, I believe, is 4.8 months.
Sorry, the progression-free survival was 4.8 months, excuse me. But again, no benefit from the combination of continuation of atezolizumab.
Dr. Harding:
Yeah. So I think that it was actually an important question, because you have that in the minds of oncologists and not clear what to do. So without that approach, I think we're now really limited to trials, the use of TKIs and potentially CTLA-4–based approaches. I did want to ask just partly, because the discussant, Dr. Chan in the ASCO session spoke about the continuation of immunotherapy in the setting of an oligometastatic progression.
Dr. El-Khoueiry, is there ever a role to continue immunotherapy in a patient that's responding with one or two lesions that might be escaping? Have you encountered that before? Do you do anything related to that?
Dr. El-Khoueiry:
Yeah, I mean this is where the art of medicine comes up and some single-center experiences reported where if the patient has had an excellent response, let's say, with a significant burden of disease, but now they're only escaping in one of two lesions. One liver lesion is behaving differently or one extrahepatic site, and those are accessible to a local therapy. I think that is a reasonable approach. Again, we don't have level 1 evidence that this is superior to switching to a different systemic regimen, but it makes sense if the patient is tolerating the current regimen well, the disease has responded well to tackle those one or two areas. And the modalities include ablative therapies, liver-directed therapies, such embolization-type therapies. And then lastly, SBRT, which has gained momentum in this disease as well.
Dr. Harding:
Great. Thank you so much. One of the aspects of the case, obviously the patient developed hypothyroidism probably as a result of the immunotherapy. How do you both think about immune-related toxicities? How do you manage and how do you monitor? Dr. Kelley?
Dr. Kelley:
I think we have increasing data both within HCC as well as other tumor types that immune-related toxicity can be associated with a greater likelihood of benefit from immunotherapy, not just because of guarantee time bias or patients are on treatment longer and then get them. We actually see if you can find your time analysis to just the first few months that's homogenous amongst the responders and the non-responders. Turns out the responders do tend to have a higher rate of immune-related toxicity, even in a confined window of time, which suggests immune activation is underlying both the response and the toxicity.
Obviously, our goal would be to have response without toxicity, but some degree of toxicity, immune-related toxicity does seem to correlate with better immune response. The key point is monitoring for these, and especially with the CTLA-4–based combinations being aware of when the likelihood of these events is, and counseling patients accordingly for both nivolumab and ipilimumab, as well as durvalumab and tremelimumab. The CTLA-4 exposure is limited to the first part of the regimen.
And thankfully, that also tells us, and we see in the data that the high-grade immune-related toxicity is also limited early in durvalumab/tremelimumab, it's in the first 2 to 3 months really that we see the majority of immune toxicities. For nivolumab/ipilimumab it's a little bit longer, but those are times when I tell patients, "Please don't travel outside the country, please talk to me before you go anywhere." And we keep a really close eye on their symptoms.
Dr. Harding:
Yeah, that's absolutely in line with Mayo practice as well. I think we are coming actually now to the end of the session. So I think we'll close with our key clinical takeaways.
I first would thank Dr. El-Khoueiry and Kelley for their very thoughtful discussion on all three of the cases today. The key clinical takeaways for this are immune-related adverse events occur on first-line systemic therapies, require close monitoring at the disease progression. Our level of evidence for further treatment is less robust, but would include clinical trials, tyrosine kinase inhibitors, and the potential of anti–CTLA-4 blocking antibodies. The key important data from ASCO 2026 was the IMbrave251, which explored the continuation of atezolizumab in those patients that benefited from atezolizumab/bevacizumab into the second line with lenvatinib, with a TKI vs a dealer's choice TKI, ultimately showing that this did not improve outcomes.
The references are noted below and for further reading, and I want to again thank my colleagues for joining me this early hour for this session, and we look forward to seeing you again in the future.
Thank you so much.
