Ghassan K. Abou-Alfa, MD: Peter, a second ago you mentioned regorafenib. Understandably, regorafenib is now approved. The story of regorafenib is rather intriguing. Tell us the story; how did we end up with this? What’s the story here?

Peter Galle, MD, PhD: It tells us 2 things. The first thing is, this was a smart trial. It was very well done.

Ghassan K. Abou-Alfa, MD: What was the trial?

Peter Galle, MD, PhD: It was the investigation of patients who had been progressing on sorafenib, and, therefore, switched to second-line therapy and got treated with regorafenib.

Ghassan K. Abou-Alfa, MD: So they did tolerate sorafenib, but they ultimately progressed. There was some requirement specific, I believe…

Peter Galle, MD, PhD: About 4 weeks.

Ghassan K. Abou-Alfa, MD: Exactly. So they got a really good exposure of sorafenib.

Peter Galle, MD, PhD: That was the first smart move, because you could predict that those patients who were tolerating sorafenib would probably also tolerate regorafenib. Regorafenib was considered, after the experience we had in colorectal cancer, to probably be too toxic for liver patients. That sort of fine-tuned the trial to those who already tolerated sorafenib. This was very smart. And then, it still concentrated on those patients with good liver function. That’s an unusual subgroup; one has to say that. This is a small proportion, but for those patients who had good liver function and were tolerating sorafenib, the added overall survival came down to 26 months. This was very meaningful. So this was a smart trial. It was only true for a subset of patients, but it was effective. We saw clearly separating Kaplan-Meier curves, and again, there was an unexpected improvement in the second-line setting—roughly 3 months.

Ghassan K. Abou-Alfa, MD: Riad, if anything, the median survival for the regorafenib versus placebo arm was close to between 10 and 11 months. But interestingly, you remember that these patients were already exposed to sorafenib. And then, if I do recall, some retrospective work shows that if you add sorafenib followed by regorafenib, the median survival with the combination is 26 months.

Riad Salem, MD: Well, I think it’s very important that when you do these types of analyses that you recognize that it’s a highly biased analysis. It’s OK to look for outlier outcomes, which is what I think we do with a lot of immunotherapies. It’s OK to do that. It’s very difficult to predict that up front, right? When you are a patient who is going to be initiated on sorafenib, you want to be the one who is going to tolerate it. And when you progress, you still want to be in good shape. Then you can benefit from that. The problem is, we can’t predict that. And so, I think we have to exercise respectful caution to promise that second curve of survival when we’re not sure how the first one is going to go. I think we have to be fair to patients while we do that, but that’s basically what that is. This is a form of immortal time bias. You have to survive long enough, and you have to do well long enough to even have that conversation to get the benefit from that second-line therapy. So it’s fair. It’s an outlier argument, but it’s fantastic. Now you have people with advanced disease who are living for over 2 years. That’s an interesting observation, and I just hope there’s a way that we can predict that. That’s where you’re going to provide a benefit for the patient.

Ghassan K. Abou-Alfa, MD: I totally agree. If anything, Amit mentioned the 2 drugs earlier, and there’s a history with regorafenib with regard to colorectal cancer. These drugs are really very similar—sorafenib and regorafenib. There are certain differences between them. Preclinically, based on some laboratory work, we know that for the tumors that get prior exposure to sorafenib and progress on it, it seems that there is certain activation for the FGF [fibroblast growth factor] and IGF [insulin-like growth factor] pathways, among other potential pathways. And, if anything, regorafenib might be a great drug to target those pathways. Clearly, there is something in the story. Would it really matter, if the patient received prior sorafenib, to get regorafenib? Or would it just be regorafenib, and who cares what the patient got before? I think the sequence is very critical. We’re going to come to that sequence question in a second.

Transcript Edited for Clarity