Transcript:Jeffrey S. Weber, MD, PhD: Georgina, we heard 2 presentations—1 from you and 1 from Hussein Tawbi—about the use of combination immunotherapy with brain metastases, which looked impressive. Does the presence or absence of brain metastases play into your decision to recommend either combination or single-agent therapy?

Georgina Long, MD, PhD: I think now, after the data presented today, it does. Admittedly, we need more data and larger numbers of patients. What we saw today from Hussein was a United States study of the combination of ipilimumab and nivolumab in patients with asymptomatic brain metastases, with a response rate of around 55% in the brain and a 6-month progression-free survival rate of 67%. It’s still early. We don’t have a long follow-up: It’s a median follow-up of just over 9 months. They were impressive curves with a flattening of progression-free survival; however, it is still in its early days.

What was great was that this was followed by a completely separate Australian trial, designed with combination ipilimumab and nivolumab in patients who were asymptomatic, had no local therapy to the brain, and were randomized between the combination and nivolumab alone. And, in that trial, when you took the drug treatment—naive patients—patients who did not get BRAF or MEK inhibitors, so very similar to the United States study—we saw a very similar response rate with ipilimumab/nivolumab of 50% and a progression-free survival of 6 months of 47%. These are more mature data with a median follow-up of 16 months, but, again, there is the flattening of that progression-free survival curve. The difference between those 2 studies, however, was that in the Australian study, patients had a greater burden of brain disease. In fact, 50% of patients had more than 4 brain metastases, whereas in the United States study, 20% had more than 3 brain metastases. So, that may explain the slight difference in results, but they’re really confirming one another that there is good activity.

What’s more, there are 2 facts. First, low activity in patients who had had previous BRAF and MEK inhibitors—small numbers, though, but still a hint there, with only a response rate of 16% in the 12 patients who had had no previous BRAF/MEK inhibitors. Second, the activity in nivolumab—it was not powered to compare the 2 arms, but monotherapy did not seem to have as high a response rate, with only 20% of patients responding in the brain. The third thing is that there was a small cohort in the Australian study of patients who had previous treatment, leptomeningeal disease, and they did incredibly poorly, with only a 6% response rate in the brain. So, all of those data together speak to using immunotherapy firstly upfront, particularly the combination if ipilimumab and nivolumab, and then we can come in with our local treatments if needed.

I just want to make 1 other comment, if I can, and that’s about the deaths that occurred in the first 12 weeks in this patient population. So, in the Australian study, we showed on our waterfalls that there were quite a few deaths of patients in the first 12 weeks. That’s not uncommon. That is expected in this patient population. You just never see it: People don’t show you they’re non-evaluable patients, because they don’t get to scan.

Jeffrey S. Weber, MD, PhD: That’s a good point.

Georgina Long, MD, PhD: So, this is not uncommon in a population with a heavy burden of disease in the brain, with 50% having more than 4 targets—that’s just target brain metastases. All in all, an immunotherapy combination of nivolumab/ipilimumab works well upfront. If it works, the responses seem to be quite durable.

Jeffrey S. Weber, MD, PhD: Mike, you’re an expert in the field and very experienced. Would you forgo stereotactic radiosurgery and put such a patient who’s BRAF wild-type with multiple brain metastases directly on to ipilimumab/nivolumab?

Michael A. Davies, MD, PhD: Absolutely. I think the data were quite compelling. The other thing I also thought was quite striking about the 2 studies was actually the safety profile. It was one of the concerns with an agent that works by generating inflammation: Would that actually cause problems in the brain? And there really wasn’t much in the way of brain specific toxicities that we saw with the therapy in either trial, and that, again, was one of the concerns about what the experience would be. So, the fact that it was actually very well tolerated was another important result of both of the studies.

I think that, in terms of patients with the multiple brain metastases, we know that the Gamma Knife, or stereotactic radiosurgery, can be effective at treating the brain metastases that we target. We know that those patients also have a very high risk of developing other brain metastases. Certainly, this is generally happening in the setting of patients who have multiple metastases in other organs in the body, as well. And so, the idea of having a highly effective therapy that can work quite quickly in the whole body looks quite compelling. This question of, if it isn’t working, is there any chance that radiation therapy could rescue the patient? Would you actually have some type of abscopal effect that could help you if you needed it after the initial immunotherapy?—I think that’s another rationale behind starting with the immunotherapy.

Jeffrey S. Weber, MD, PhD: Jason?

Jason J. Luke, MD: I wanted to get Mike’s opinion. Mike presented these data about using BRAF, showing an outstanding response rate in BRAF-mutant patients. So, if you start now merging these together, you really start to get to sequences of therapy that really should be optimized for patients. Because, if we’re seeing these great responses with BRAF in a multi-metastatic, intracranial situation, I think there really is a clear role there for debulking from a systemic therapy and maybe then transitioning to a therapy that has long-term benefit.

Georgina Long, MD, PhD: You mean debulking with dabrafenib/trametinib.

Jason J. Luke, MD: Yes, exactly, not with radiation but, rather, starting with BRAF/MEK inhibition, and then thinking about a transition point where you could try to get that long-term benefit.

Jeffrey S. Weber, MD, PhD: That’s a great segue to another important question, which is, can you use BRAF/MEK inhibition to turn a cold tumor into a hot one? So, let’s hear from Jason and then Robert, because this is also a surgically important question.

Georgina Long, MD, PhD: Can I just make 1 point before you answer? Just to make it clear, do not wait until progression with BRAF/MEK inhibition. That’s what our trial in Australia showed. What Jason is talking about is a short course to debulk, so I just want to make that clear. Because if you start that trend and wait until progression, then go with immunotherapy, the data from Australia suggest that’s not the strategy. So, we’re talking about a short course.

Jeffrey S. Weber, MD, PhD: That’s the dirty little secret. Robert?

Robert H.I. Andtbacka, MD, CM: I think also we have to distinguish this. So, we’re talking about extracranial or intracranial disease here. I think, because of your data that you presented, that in the patients who had been treated with BRAF/MEK combination, when you then give them nivolumab and ipilimumab, the response rate was actually very poor. To me that would mean that, for the brain metastases, going with nivolumab and ipilimumab first, even in the BRAF-mutant patients, is probably a better method. And if you don’t get a response with them, then add the BRAF/MEK combination. But we don’t have the data. I’m concerned for the brain metastases that, if you went with a BRAF/MEK inhibitor first, you may not get the control you wish to have.

Georgina Long, MD, PhD: Right.

Transcript Edited for Clarity