Screening for HCC: Current and Emerging Biomarkers

Transcript:Richard S. Finn, MD: What’s really exciting about liver cancer, as a disease, is it’s one of the few diseases that we know the cause—which is cirrhosis. Given that there’s a variety of causes of cirrhosis, we know who’s at risk, which gives us a huge opportunity to intervene early, potentially. Dr. Vogel, could you comment about how we should try to find liver cancer, early?

Arndt Vogel, MD, PhD: Yes. I think you are right, but not completely right. As you mentioned before, we have a shift in the underlying cause of liver cancer. We know that most patients have viral hepatitis. Most of these patients, specifically, with hepatitis C, have liver cirrhosis once they develop hepatocellular carcinoma. In hepatitis B, we have also seen that some of these patients have HCC in a noncirrhotic liver, which makes it a little bit more difficult. And, specifically with fatty liver and NASH, we see more and more patients who do not really have a chronic history of liver disease, which makes it more difficult to apply a screening program.

In high-risk groups, we would generally recommend to do screening. Screening would include an ultrasound and some testing of tumor markers, like AFP, which also is not perfect. Specifically, AFP has a low sensitivity, so we might miss a lot of patients who have tumors without AFP. We would not see them if we would just rely on AFP testing.

On the other hand, if you do an ultrasound as a screening method, it depends on the physician who’s doing the ultrasound and the treatment he/she has to do the ultrasound with. In general, ultrasound is a good tool to detect liver nodules. But, if we have now a shift from viral hepatitis, or liver cirrhosis, to fatty liver disease, it’s getting more complicated. In patients who have obesity and have a fatty liver, detecting nodules is far more difficult than in patients without fatty liver.

The other point is, not only is it getting more difficult, we are getting, also, more patients that are potentially at risk. On the one hand, we have to acknowledge that, of course, not all patients with fatty liver get liver cancer. But, they are at risk. We would have to screen not really patients, but people with fatty liver who are at risk to develop liver cancer.

Richard S. Finn, MD: What you’re saying confirms a lot of what is in guidelines—you can’t rely on a blood test alone. Imaging is really the standard. Ultrasound has limitations. If you want to check AFP, that is okay, but you can’t rely solely on AFP. Are there other serum markers that people are looking at?

Arndt Vogel, MD, PhD: There are more serum markers that are in development. We have some retrospective data that indicate that we will detect more patients through biomarkers like AFP-L3 and DCP (des-gammacarboxy prothrombin), if you combine them and if you also include gender and age. We might develop scores that indicate who has a higher risk to develop cancer. Maybe this could be a method to, really, do some kind of pre-screening based on that test and have those patients who have an increase in their scores referred to more regular screening by ultrasound or even MRI, if there is a strong increase in their scores based on these biomarkers. I think this will be, probably, in the future, the way we have to go. We really have to combine different markers and, then, identify those patients who really require good imaging.

Richard S. Finn, MD: I think we can say, pretty certainly, there is a group that definitely needs screening. Those are probably patients who have known cirrhosis and known hepatitis. Importantly, we can only find cirrhosis if you look for liver disease, and there’s probably a big underawareness of that in the community.

Arndt Vogel, MD, PhD: Of course. We have to acknowledge that patients who have chronic liver diseases do not really know that they have underlying liver diseases. If they have hepatitis C or B without any symptoms of advanced liver diseases, they might not even know that they have liver disease, which is specifically tragic if you think about hepatitis C. We now have very effective treatment available for hepatitis C, but the people need to know that they have a disease that can be cured.

Richard S. Finn, MD: Sure.

Jordi Bruix, MD: Could I make a comment about the value and efficacy of tumor-marker screening? The endpoint of screening is to detect the disease at an early stage so that the patients may benefit from treatment and have an improved survival. We all know that AFP, and all the tumor markers, increase on our reflection of more advanced or more aggressive stage. So, in a way, the detection through tumor markers, by default, is going to detect tumors that are more advanced—that are more aggressive and can be found in patients who will not benefit, potentially, from the detection and from the treatment because they will be more advanced.

Arndt Vogel, MD, PhD: We don’t know.

Jordi Bruix, MD: When you do the modeling of what the outcome is going to be, this is one of the important considerations. The goal is to detect the disease whether it is tested by tumor size, no tumor markers, no symptoms, or no vascular invasion. AFP, and all the other markers, have always been associated with a higher rate of microscopic vascular invasion. So, when you have positive results, you have more, or less, optimistic outcomes.

Amit Singal, MD, MSCS: One of the limitations of the current study is that they use their outcome as HCC at any stage. Whereas, really, the outcome for these biomarker studies has to be HCC at an early stage, if not at a very early stage, as you said. I think that you’re right—all of these biomarkers perform much better when they’re at an advanced stage. We’re hoping that some of these biomarkers will be able to perform—particularly, when you use it as a panel—at least reasonably well at an early stage. Given the heterogeneity of HCC, I think it’s going to have to be multiple biomarkers. I don’t think any single biomarker is going to work in isolation.

Dr. Vogel brings up a very important point—all of the old ultrasound studies were primarily conducted in a viral hepatitis patient population and had significant limitations in terms of the patient population. These were often very lean individuals, and there was more oversampling of hepatitis B—where ultrasound performs better. There have been several studies that show that ultrasound, even outside of its operator dependency, works much worse in patients with nonalcoholic steatohepatitis. I think biomarkers are going to be the future, and I think that’s going to be one of the biggest things that we have to do: find accurate biomarkers for detection at an early stage if we want to have a chance to do well with this cancer.

Jordi Bruix, MD: It’s an area that needs lots of research, because it’s an unmet need to have a biomarker that detects the disease at an earlier stage.

Amit Singal, MD, MSCS: Agreed.

Jordi Bruix, MD: At the same time, AFP, as I mentioned, is a predictor of poorer prognosis. So, in a way, it always gets redundant in the same direction. We need something better than AFP. And the fact that ultrasound may be difficult does not make the other tools that you may consider antiquated.

Arndt Vogel, MD, PhD: Let me make one point. Maybe it’s better to have a “late diagnosis” than to have a diagnosis “too late.” It’s always a question as to, is the glass half full or empty? In colorectal cancer, we have recommended, for a long time, blood testing and those tools. This is also a very insensitive method, but, still, it helps to decrease the incidence of mortality of colorectal cancer. So, I would really not discourage the use of screening tools in HCC, and I completely agree that AFP is not good, but I think we still can introduce it.

Transcript Edited for Clarity