Hepatocellular Carcinoma: Practical Implications of Emerging Data - Episode 15

HCC: Liver Function and Treatment Decisions

, , , , ,


Richard S. Finn, MD: We’ve talked about all this excitement, and we’re big cheerleaders for all the clinical research going on. Some of that is in our own self-interest, but for patients, the options are amazing. But let’s be honest. Are there patients in second-line who aren’t candidates for treatment? Maybe they get frontline, and it’s time to stop. What do those patients look like to you?

Reena Salgia, MD: This conversation has, thankfully, focused a lot on the patients we assume will get to second-line and have that option. We’ll be fortunate enough to maintain their liver function and their performance status, and that would certainly be a good problem to have in that setting. There are going to be patients who first achieved a durable response and then started to progress and developed decompensation of their liver function because of the progression of their tumor burden. Or patients who are intolerant—depending what that intolerance is to the first-line regimen, they may or may not recover enough to progress to second-line therapy. We may see, but hopefully it will be a small percentage of our patients for whom we’re not able to pursue second-line therapy. As we learn more about the options first-line with combination therapies, hopefully we’ll be able to figure out which ones are more likely to be able to go on to second-line options and what that would look like after.

Richard S. Finn, MD: Something that’s been very intriguing to me in practice: We’ve always seen patients who have a very large tumor burden in the liver. They’re borderline decompensating, right? It’s hard to know, is this patient just cirrhotic, or is this big tumor pushing them over? If they had a response, they would feel better. They would regain some liver function. I’ve actually had a few patients more recently with whom the discussion is, “You have a very advanced tumor. You presented very late, unfortunately. We have some drugs that have fairly high response rates, and you might not qualify for a clinical trial, given your physiology. I suspect if your tumor responded, you would get better.” This is a risk-benefit discussion with patients now that we have regimens that have high response rates. Amit, what’s your thought on that? Is there a way we can tell how much is underlying liver dysfunction versus tumor burden? Sometimes, these patients have big portal vein thromboses; sometimes they just have a lot of tumor in the liver. What do we do?

Amit Singal, MD: You know, Rich, it’s an interesting question. We all have these patients who, as you’re saying, presented with large tumors, some degree of liver dysfunction. You talked to the patient, you treated them, you had a response, and the liver function actually improved. The patient went from a Child-Pugh B to an A as the tumor responded. This is, once again, a nice problem to have, because it means our therapies are actually improving, and we’re seeing better responses.

The one thing I use is their natural history coming into this. The patients who are easier are the patients who were already in the system. For example, before they presented with their large tumor, you started to see their liver function decline over time. For those patients, you can say more confidently that this is related to slow deterioration and true liver dysfunction. The ones that are tough are those patients who present up front—so new to your system—as a Child-Pugh B. Let’s say with a large tumor, plus or minus some portal vein tumor thrombus. I have to say that, at least personally, I don’t know of any great way to tell if somebody’s liver dysfunction is related to their tumor versus their underlying liver disease. It’s worth a discussion with that patient, that if you can have a well-tolerated regimen, it’s worth a shot. Then see what their liver function does—the test of time. There are some tests where you can see how well the background liver is “functioning.” But I don’t think those have been well evaluated in this specific setting to use in clinical practice at this time. There are research tools that are being evaluated to give us a sense of “liver function.”

Richard S. Finn, MD: Anthony, should every patient get I/O before we call it a day in liver cancer? What are your thoughts?

Anthony El-Khoueiry, MD: The point you’re making is critically important, but I want to move a little beyond the anecdotal to look at the CheckMate040 Child-Pugh B cohort. This is a unique prospective cohort of Child-Pugh B7 and B8 patients who were treated with single-agent nivolumab. To second what you both have said is that when you look at those patients, the ones who responded, the response rate was 11%, 12%. The majority—80%, 90%—actually moved from Child-Pugh B to Child-Pugh A.

These were carefully selected Child-Pugh B patients. There were only 7 or 8, no paracentesis in the last 3 months, no active encephalopathy in the last 6 months. Maybe using those types of parameters—you know, if the patient has a large infiltrative tumor, portal vein invasion, and a high bilirubin and low albumin, that puts them in Child-Pugh B. But you’re requiring paracentesis. They don’t have hepatic encephalopathy. Those patients might benefit from treatment of their tumor. Irrelevant of what the cause of the decompensation is, if you’re starting to require paracentesis and you have hepatic encephalopathy, the prognosis becomes quite poor. I’m not sure it’s wise to try to treat the cancer at that point from a quality-of-life perspective, from a public health perspective, etc.

Amit Singal, MD: This goes back to your prior point, Rich, of having 2 diseases at once. It’s a disease within a disease. As your liver function starts to continue declining there, your mortality is really being driven at that point by your cirrhosis, not your HCC [hepatocellular carcinoma]. I agree with Anthony that at that point, it doesn’t necessarily make sense to continue treatment if you’re starting to truly have significant liver dysfunction.

Transcript Edited for Clarity