Multimodality Management of Advanced Liver Cancer - Episode 2

Implications for I/O Therapy in the Treatment of HCC


Ghassan K. Abou-Alfa, MD, MBA: Riccardo, so you see now we’re heavily involved with regard to the checkpoint inhibitors: anti—PD-1 anti–programmed cell death protein 1], anti–PD-L1 [anti–programmed death-ligand 1], anti-CTLA-4 [cytotoxic T-lymphocyte–associated protein 4]. But also an important perspective in liver cancer still is the etiology: hepatitis B, hepatitis C, NASH [nonalcoholic steatohepatitis], alcohol. Our perspective, we would like to hear from you—forget about the checkpoint inhibitors for a second—is a patient with hepatitis B-related HCC [hepatocellular carcinoma] different from hepatitis C? How do you read them?

Riccardo Lencioni, MD, FSIR, EBIR: Well, definitely these are different patients. Clearly, it’s important to point out that there’s been a major advancement even in the management of the viral hepatitis itself. Until recently, for hepatitis C, this was a very difficult disease to treat, and clearly with the introduction of direct antiviral agents, now most patients can achieve a sustained response, clearly opening a new prospect for their life and in a way changing the prognosis significantly. Of course, these patients remain at risk for development of a tumor. This is why it’s important to maintain surveillance and try to diagnose, as much as we can, tumors early in all these patient subsets. There may be specific considerations when it comes to deciding the treatment strategy for each of these categories.

Ghassan K. Abou-Alfa, MD, MBA: Fair. I hear you very clearly. We’ll take the other view. Mike, in your clinic, regarding a patient with hepatitis B-related HCC versus hepatitis C-related HCC, are they the same disease, different diseases? How do you read them? And I will add here, especially from the perspective of the checkpoint inhibitors.

Michael A. Morse, MD, MHS, FACP: Well, fortunately, we found that the checkpoint inhibitors work fine, and even better, we don’t see flares of the underlying disease or damage to the liver based on an immune attack on the hepatitis. I’m sitting next to a hepatologist and saying, “Well, they’re all the same.” But to a medical oncologist, the main issue is that hepatitis B patients sometimes do present with much larger tumors, but they don’t always have cirrhosis. Whereas, the hepatitis C patients by and large do sometimes have more chronic liver disease that we have to deal with simultaneously.

Ghassan K. Abou-Alfa, MD, MBA: I hear you very clearly. And by the way, there are 2 theories on this. We at Memorial Sloan Kettering Cancer Center are working on this is. Are they really different diseases or can they come to the end of the line, same story? Because with checkpoint inhibitors, as you mentioned, it seems it doesn’t matter. But let’s take a little bit of a transplant perspective. Checkpoint inhibitors from the transplant perspective, do you think there is anything that the suppressive immunity in the transplant patient might trigger things differently, or how does it work? And will we be concerned about that?

Catherine Frenette, MD, FAST, AGAF: We are quite concerned about using the checkpoint inhibitors in transplant. As you know, we need to immune suppress patients so that they don’t reject their allografts. And that’s regardless of what kind of transplant: kidney transplant, liver, heart, lung, etcetera. There have been reports of using checkpoint inhibitors after transplant for various diseases, in HCC, but also things like melanoma. And there are patients who will lose their graft. And when you look across all of the organs, the risk of graft loss really is different depending on which transplant. The liver is a little bit more of a tolerant organ, again because of that chimeric immune system that we see. But it seems that there’s somewhere around a 35% to 40% risk of graft loss with immune checkpoint inhibitors after a transplant.

Ghassan K. Abou-Alfa, MD, MBA: Fair.

Catherine Frenette, MD, FAST, AGAF: With a kidney, they go on dialysis. With a liver, you really can get into big trouble. And there have been reports of very severe rejections that have not been able to be reversed. That’s something that patients have to be aware of if we choose to use those. Interestingly, there have also been some data looking at biopsying the allograft and staining for PD-1 in the allograft rather than the tumor to try to predict whether that patient has a higher or lower risk of rejection. And the patients who do have PD-1 staining in the allograft may actually have a higher risk of rejection with the immune checkpoint inhibitors. So that’s something we need a few more data for.

Ghassan K. Abou-Alfa, MD, MBA: Actually, you bring up another important point. And I’ll go back to Tim. Interestingly, in most cancers, we look at the PD-1 expression, and in HCC it seems like we don’t do that much. You already suggested that a little bit. Again, tell us what’s the connectivity between an anti—CTLA-4 and then anti–PD-1, anti–PD-L1 that makes things look differently for HCC compared to other tumors?

Tim F. Greten, MD: Well, that’s a difficult question. The initial studies in HCC were actually done with CTLA-4, like in any other disease, we started with CTLA-4 in melanoma also. And in those studies, we noticed that one of the major mechanisms that CTLA-4 really has is that it actually brings T-cells at the site of the tumor. Tumors usually are not very heavily infiltrated by T-cells. Although we know from many old studies that the higher the infiltration rate with the cytotoxic T-cells, the better the spontaneous outcome is actually for patients. What we usually find more with CTLA-4 is an influx symptom, acute inflammation in the tumor. Whereas, with PD-1, we see more sustained response. And that’s why there are a number of ongoing studies where those 2 drugs are actually being combined.

Ghassan K. Abou-Alfa, MD, MBA: Yes. Along that same line, we’re hearing about a lot of combinations. And another combination I would like to ask Mike about is, what’s the story with combining an anti—PD-1 and an antiangiogenic, anti-VEGF, or TKI [tyrosine kinase inhibitor]?

Michael A. Morse, MD, MHS, FACP: Well, you can think about this in 2 ways. One is a very simplistic view of they both have efficacy, and you could certainly expect added effects, but what we’re really hoping is synergistic effects. And there are a number of ways this could occur. One that I actually studied a number of years ago was the fact that VEGF seems to cause, among other things, defects in dendritic cell maturation and increases in myeloid-derived suppressor cells. So just inhibiting VEGF in that environment or the signaling of VEGF is likely to give more mature dendritic cells, ultimately leading to fewer regulatory T-cells in the environment and fewer myeloid-derived suppressor cells. There’s really an immune effect from antiangiogenic therapy. Also, we know that there are some changes in the vascularity, some people would say pseudonormalization, but some changes that might actually allow better immune cell trafficking. And there’s also changes in interstitial pressure within the tumor, which also may allow better distribution of the influx of immune cells. I think there’s a real synergy, and clearly preclinical trials and now early clinical trials have demonstrated that.

Ghassan K. Abou-Alfa, MD, MBA: We’re hearing quite a bit of information so far. To summarize what we just heard, understandably, the perception that we all have in regard to the immune system and the liver is not as simple as we might have thought. It’s pretty complex. And no doubt, we’re hearing from many of our colleagues that the interaction between an anti—PD-1 or anti–PD-L1 with the tumor cell in HCC is not necessarily appearing to be sufficient. In other words, an anti–CTLA-4 interaction probably is necessary because of the buildup of the immune system in the liver cells per se. And this is really interesting, what Tim just suggested with regard to the staining, because understandably we don’t necessarily need to stain for a PD-1 expression in the tumor cells. And this tells us in HCC that the mechanism of how the checkpoint inhibition works is not necessarily the same as we are seeing it in regard to other solid tumors.

Now, at the end, Mike brought up a very intriguing point, and I’m sure we are all intrigued by the idea of a combination of a checkpoint inhibitor plus an antiangiogenic. He is right. We would like to see some potential synergistic effect. There is a lot of activity in that regard. We’re going to talk about it a little bit later in the program, but for now the understanding is that yes, there could be, through the cycle of the immune cells, a certain interaction at the VEGF level that actually will trigger further activation and deactivation and as such, inhibiting VEGF would probably make sense. However, of course, this all would be need to be shown in the data to carry on.

Transcript Edited for Clarity