Ghassan K. Abou-Alfa, MD, MBA: I would like to now talk a little more about how we diagnose patients. Dr Kaseb, patients come to you, they have HCC [hepatocellular carcinoma]. Biopsy or no biopsy?

Ahmed Kaseb, MD: Yeah, so that’s a very important question.

Ghassan K. Abou-Alfa, MD, MBA: I always ask only important questions.

Ahmed Kaseb, MD: Right. For us as medical oncologists, we all were trained that way: no meat, no treat. We’re used to getting biopsies. In the old era of 15 to 20 years ago, patients were presenting with Child-Pugh C, coagulopathy, severe ascites.

The biopsy of risk was very high. In my practice—and many of us now do this—if it is safe to obtain a biopsy in this era of immune profiling and molecular profiling, we always require it, especially in cases where you have no evidence of cirrhosis, which is a biopsy in that case that is really mandated.

Ghassan K. Abou-Alfa, MD, MBA: You bring up very important points. I’m going to come back to that. But before we go there, Anthony: imaging, what’s your favorite?

Anthony B. El-Khoueiry, MD: Well, the importance of imaging is that it’s multiphase imaging. Whether it’s CT [computed tomography] scan or MRI [magnetic resonance imaging], the multiple phases are important because the reality is that, especially in the community, we still diagnose HCC very commonly by imaging. We need to see the arterial enhancement and the arterial phase, and then the last shot in the venous or delayed phases. This is a critical component because we see many patients come in with single-phase scans, and arbitrated diagnoses are made that way.

I do agree that to advance the field, obtaining tissue is important to allow us to do the appropriate research, especially because we have more options of systemic therapy. But in reality in the community, radiological diagnosis is certainly an acceptable approach as well.

Ghassan K. Abou-Alfa, MD, MBA: And to that extent, the radiological approach now with the LI-RADS [Liver Imaging Reporting and Data Systems], how much do you use that in your practice, Catherine?

Catherine T. Frenette, MD: We very much use the LI-RADS criteria. When I’m deciding on a biopsy, I really do think about the stage of cancer that the patient is in. You know, needle-tract seeding and bleeding from the biopsy is rare but real. If I have a patient who may potentially be curative through resection or a transplant, I’m a little more reticent to do a biopsy because of that risk, where we don’t want to spread the tumor that could then prevent a curative treatment.

I agree, getting a biopsy to advance the field and help with molecular profiling with more advanced disease might be helpful. But in the early stage disease, LI-RADS is very, very useful. I have a lot of patients who come in from the community for a consult, and I always prefer to discuss them in a multidisciplinary tumor board before we think about a biopsy.

Ghassan K. Abou-Alfa, MD, MBA: Pierre, to go back to the next-generation sequencing [NGS] that Ahmed and Anthony brought up, did you find certain genetic alterations that were interesting?

Pierre Gholam, MD: We have become somewhat enamored with performing next-generation sequencing in many of our patients who present at our multidisciplinary teams at meetings. Of about 220 unique patients we see every year, you very rarely come up with druggable, actionable information from this. I can recall notably a patient who turned out to have MSI, microsatellite instability, who actually benefited tremendously from an I/O [immuno-oncology] treatment when one might not have thought that would be the case. But I can’t name you many other examples of patients where we’ve done NGS and ended up finding a track that we could treat and effectively prolong life in a dramatic way.

Ghassan K. Abou-Alfa, MD, MBA: It’s amazing that we’re hearing about these essential components. To re-summarize, I like the way Dr Kaseb said it: no meat, no treat. We need to look at tissue; it’s very critical. Especially from the perspective of medical oncology. However, I also agree with Dr Frenette. We have to look at the whole spectrum. For patients with very early disease, with a potential for cure—for example, somebody with a disease that is transplantable or surgically resected, or RFA [radiofrequency ablation]—I can totally understand why a biopsy is not necessarily required before we move on to this. Because of the risk of that, Dr Frenette said—even though I agree with her, it’s extremely low, 0.000003 for seeding. Nonetheless, it can still happen. On the other hand, the transplant community would prefer the patient is really not open in regard to any biopsies. That’s totally fine.

However, when we come to patients with metastatic disease, advanced disease—where curability is not the point but rather to ensure that we have controlled the cancer and now with advent of therapies, we’re going to talk about this endlessly—by all means a biopsy is needed. I totally agree with Dr El-Khoueiry that the biopsy is needed not only for the confirmation of the diagnosis but also so it can give the opportunity for further analysis and germline testing, interestingly.

If you ask any of us today if it’s really critical that we have germline testing for HCC patients, based on the therapy we have, I would say it’s probably not necessary. But if we don’t do it, we’ll never know if it’s necessary or not. So it’s very critical that we continue to do this. As we heard from Dr Gholam, interestingly, the patient that he referred to with MSI-high status and HCC—this is the second patient that I hear about with an MSI high with HCC. For example, FGF19 patients with a mutation of FGF19 and HCC definitely exist as well. So you have to make sure that we obtain the tissue, confirm the diagnosis, and understand that especially in the metastatic setting, we give the right therapy, because at least based on some data, 10% to 20% of the patients might have a combined HCC and cholangiocarcinoma.

It’s very important to delineate that. On top of this, as we heard, the next version of sequencing could be very critical to either guide patients who are in need for other therapy besides what we’re going to talk about today. And No. 2, of course, were the research purposes that will help us understand better where we’re moving in regard to that field.

Transcript Edited for Clarity