Dr Michael Morse on the Etiologies of HCC and Treatment Options for Patients With Worsening Liver Function

Michael Morse, MD, FACP, MHS, professor of medicine at Duke University School of Medicine and medical oncologist at Duke Cancer Center, discusses the etiologies of hepatocellular carcinoma (HCC) and how clinicians can tailor treatment for patients with varied levels of liver function.

Transcript

HCC is known to be driven by various etiologies—what are the various risk factors and drivers, and what are their implications for prognosis and treatment?

We know that hepatocellular carcinoma is the common final pathway for many different stimuli that damage the liver. Most people with hepatocellular carcinoma will have cirrhosis. You can get cirrhosis by having hepatitis C infection, hepatitis B—although hepatitis B can sometimes cause hepatocellular carcinoma without cirrhosis—alcohol misuse, some inherited syndromes, and of course the one that's becoming quite a bit more common now: metabolic-associated steatohepatitis, MASH. People used to know it is NASH [nonalcoholic steatohepatitis]. Of course, there are many people that have more than one of these risk factors. But the reality is that they are the causes of most of the hepatocellular carcinoma really worldwide, if you look at those different causes.

There's a small percentage of people that don't appear to have an obvious etiology. We really don't know why those people have hepatocellular carcinoma. But if you look at those that cause cirrhosis, they all essentially cause damage to hepatocytes; there's an attempt at regeneration, there's inflammation, and that environment leads to more damage. And eventually, you get outgrowth of dysplastic nodules, and then eventually invasive hepatocellular carcinoma.

Now, the question that we've had now that we have more effective therapies is, is there a therapy that fits a particular etiology? And this is a very controversial area. My take on it is, there really isn't a particular therapy for a particular etiology. Most of the data comes from subgroup analyses from the individual trials. And although you can come up with biologic explanation why a particular etiology might lead to a therapy that's more effective for that, the reality is, it's not been consistent.

We do see some interesting observations. Hepatitis B patients tend to do better in general with the immune therapies. But really, you can look across the board and we really choose therapies more for what seems more appropriate based on other clinical considerations; certain other risk factors for bleeding, for example; risk of autoimmune events; whether people have had transplant before; patient choice; convenience; and so on—but not so much the etiology.

Can you speak to the need for including patients with HCC of viral origin and those with HCC of non-viral origin in clinical trials?
Right now, clinical trials in HCC, pretty much take any etiology of hepatocellular carcinoma and allow those people to participate if they meet the other inclusion criteria, and I think that's still important, because the reality, is we have a worldwide problem with hepatocellular carcinoma. It's obviously even a bigger problem in parts of Asia and Africa, but it really is a worldwide problem. And it's important to have patients that have different etiologies and different demographics included in the studies.

There's obviously been some discussion about when you design those studies, making sure you have adequate representation from different groups and then stratifying, if necessary, based on certain factors. But we're still faced with the fact that the most important prognostic factors are things like stage, which is partly reflected in portal vein invasion or extrahepatic spread, and Child-Pugh function, although most studies enroll only Child-Pugh A liver function. But those are the most important.

Because of different practice patterns in different parts of the world, you will often find stratifications based on where patients come from—Asia vs rest of the world, Japan vs the rest of the world. But I still come back to, as long as the study includes some attempt to enroll patients from different parts of the world with different demographics and different etiologies, that is what's most important.

With most trials enrolling only patients with Child-Pugh A liver function, how do you approach treatment for those with Child-Pugh B liver function?
You're correct that most clinical trials only enroll patients with Child-Pugh A liver function. They certainly have much longer survival across the board, regardless of how you treat them. Prognostically, being Child-Pugh A, or preserved liver function, is better than Child-Pugh B. Child-Pugh C is controversial because those are often patients who die of their underlying liver dysfunction, even before the cancer is necessarily lethal to them.

But the Child-Pugh B population is quite large. In some centers, like mine, it approaches half of the patients, and in some it's maybe a quarter or a third, but there's a lot of them out there. And once again, the clinician is faced with how to manage them. Some drugs have been tested—not in randomized pivotal trials—but they've been tested in patients that have Child-Pugh B liver function, and more of these studies are planned. And we do know that it's true: They don't appear to survive as long as Child-Pugh A liver function patients, but they do seem to get benefit out of the various therapies that have been used.

I think it's an important discussion with a patient: Why are they Child-Pugh B? Is it because they have a failing liver and it's rapidly failing, and they're crossing from A to B, on to C? Those people are better off trying to get anything that can be done to tune up their liver function, or palliative care. But maybe they’re Child-Pugh B because they've got a large burden of tumor, and so they really need the cancer treated and their liver function will improve. Let's take that scenario—we are certainly treating those people, even though they’re Child-Pugh B and even though there's limited data.

The other kind of data, though, that I think has been very instructive is those studies that have only enrolled Child-Pugh A patients, inevitably a fraction of those individuals will have a decline in their liver function. And we can look at that group—those that now have Child-Pugh B liver function, let's say 8 weeks into the study—and ask the question there, "How do those patients fare?"

We also see evidence—now obviously, this is in some cases retrospective evaluation—we see in those patients that they, too, get benefits just like the broader patient, that Child-Pugh A liver function patient, population does. So, I'm comfortable discussing treatment for Child-Pugh B liver function patients, and in most cases offering the same therapies. In some people where you're a little bit concerned, you might choose a different therapy, or instead of a combination, you might choose just one of the therapies. NCCN [National Comprehensive Cancer Network] Guidelines does deal with this. They mention certain drugs for just Child-Pugh A, they mention other drugs that could be for any patient regardless of their Child-Pugh function.