Andrew X. Zhu, MD, PhD: For hepatocellular carcinoma, we’ve been using the alpha-fetoprotein, also known as AFP, for decades. I think that marker definitely serves its purpose. We definitely see that it gives you some prognostic significance. But also for diagnosis, the marker definitely is elevated enough—to, say, 70% of HCC. But keep in mind, for 30% of the patients, even though they have true histological confirmation, they can have normal AFP. So for that reason, you rely on the AFP. But on the other hand, I agree with Katie. I think the histological confirmation, if needed, is still the gold standard. Just to follow the comments that Richard and Katie made, I think the tissue diagnosis is important to differentiate HCC from cholangiocarcinoma and potentially other rare, other histological variants in very, very rare instances. I think that’s the first purpose.

I think the second purpose is because really, we’re having a tissue diagnosis. We have the tissue, and that will definitely allow us to explore the molecular signature. I think Richard’s point is that today we don’t have the useful marker so that we can actually really match the right targeted agent to make dramatic therapeutic impact. But I think this may change with time. A lot of things that we’re thinking about today are really not actionable right now. That may change with time. If you have the molecular signature in hand, I think that will definitely inform our patients. We’ll be on the horizon for the potential treatment option down the road. So personally, I do support that kind of effort.

Ghassan K. Abou-Alfa, MD: I like what you said. If anything, what you heard here from everybody is 1) As you can see, interestingly, we all think the same way, and we all treat the same way. And if anything, this is based on ample experience that we all have received and have practiced. Again, the biopsy is critical for HCC, and the reason is because, as we heard, you can very much sometimes understand that the disease not only is actually pure HCC but could be a combination of HCC and cholangiocarcinoma. 2) I like very much what Andrew said, which is that at the end of the day, maybe yesterday, we don’t have a specific targeted therapy based on a specific genetic makeup, exactly as Rich also suggested, but in the future, we don’t know where this might go.

This information will be extremely valuable to make sure that we follow accordingly. And of course, a point that I understand is that in regard to the AFP, yes, it can be understood as being a marker for HCC. But of course it has its caveats, and as we said, yes, a patient with HCC doesn’t necessarily have an AFP, and we have to be very careful in that setting, per se. With this said, we’ll carry on to very important practical points that come to the disease, per se. And if anything, Rich, I’ll go back to you. HCC—we hear this all the time, and we say it all the time—is 2 diseases in 1. Tell us, what are those 2 diseases?

Richard S. Finn, MD: Well, this disease occurs 90% of the time in the setting of chronic liver disease. That etiology of chronic liver disease is very variable; it’s regional. In the United States, historically it’s been driven by hepatitis C, and globally by hepatitis B. We’ve improved on both of those. Certainly, with the treatment of hepatitis C, we expect the relationship between liver cancer and hep C to fall over time. But then there’s increasing awareness of nonalcoholic steatohepatitis or fatty liver disease as a potential driver. But at the end of the day, the common denominator is cirrhosis. And as I like to tell my fellows and students, a cirrhotic liver is a premalignant state. And as cancer develops in that organ and patients come to clinical presentation, how we approach them will be dictated by not only their tumor burden but also their underlying liver disease. And that’s why, when we approach this disease, probably very uniquely among all the tumors we treat, the outcomes will be dictated not only by their tumor burden, or TNM stage, or however we capture tumor burden, but also by the extent of the liver disease.

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

Richard S. Finn, MD: You can have someone with a 2-cm liver cancer, but if their bilirubin is 10 from cirrhosis, that cancer is less important.

Ghassan K. Abou-Alfa, MD: No doubt that we know, Andrew, that there are many ways of assessing the liver functionality as well as the disease itself and different staging systems. But understandably, especially a lot of the clinical trials that we see around us are mainly based on the Child-Pugh. Tell us about the Child-Pugh. What is it, and what are your thoughts on it?

Andrew X. Zhu, MD, PhD: Child-Pugh is actually a scoring system that hepatologists use very routinely. We take the different biochemical parameters, as well as some clinical parameters—ie, the presence or absence of ascites, encephalopathy—and then come up with a score. And the scoring system is actually from 5 up to 15. And then based on the different numbers, you actually categorize them into different categories: the Child-Pugh A, B, or C. Is this the best, most reliable way to really assess the underlying liver function? I think our colleagues would challenge us.

As we all know, there are other scoring systems. For example, our hepatology colleagues like to use the MELD [model for end-stage liver disease] score. But for us, I think the Child-Pugh gives you some framework of the underlying liver function of the patient that you’re treating, and that will definitely give you some starting point, and whether the patient should be aggressively thinking about treating the HCC or actually, perhaps even more, thinking about how we can really make sure that we’re not doing more harm to our patients.

I think if you have patients who already have advanced underlying cirrhosis in the category of Child-Pugh C, I totally agree with Richard: having a newly diagnosed HCC, particularly at the early stage, I’m not quite sure it’s relevant for the overall prognosis of this particular patient, and short of, obviously, if the patient is a potential liver transplant candidate. So I think having this is actually incredibly valuable in my opinion. And also, as we all know, in the last few years, I think that also gives us a good instrument to categorize patients who are actually the better candidates for clinical trials. We can actually speak the same language. Clearly Child-Pugh is not the only one, but I think we found that it’s incredibly practical.

Ghassan K. Abou-Alfa, MD: Fair. No, by all means. If anything, we hear that 2 diseases in 1 is truly about the disease, or the cancer, per se, but also the liver cirrhosis component. And exactly as Rich said, this can sometimes get in the driver seat while the liver cancer is sitting in the back seat. And this kind of assessment, despite the limitation, and of course, as Andrew said, there could be 2 or 3 different ways of reading things or even more. But at the end of the day, at least we need to do something. And we like the Child-Pugh for no reason except that all the data have been driven by Child-Pugh. Why is that? Just simply by default. So this is really something that’s almost 2 decades, and this is what we agreed to, and this is really where we are today. Make sure, please, to check on the Child-Pugh scoring.

Transcript Edited for Clarity