Transcript:Jeffrey Weber, MD, PhD: There will be some data, I think, at the SMR meeting from Jen Wargo, where she talks about doing biopsies on patients who get brief periods of BRAF/MEK. And the rumor is that it’s going to turn out that somewhere between 6 and 12 weeks is an optimal amount of time. You folks know the data, I assume, better than I. What happens when you develop resistance to BRAF/MEK? Isn’t the gene pattern somewhat like the resistance pattern to PD-1 therapy? Reinhard?

Reinhard Dummer, MD: Yes. Well, this is a very interesting topic. So, what we can say is that if you have a long-time exposure to kinase inhibitors, there is a transcription adaptation, and this includes an upregulation, for example, of BRAF. But it also includes a change in the differentiation from a more melanocytic phenotype into a mesenchymal phenotype. And with this, the dependency of the tumor cells to the MAP kinase pathway is reduced. Therefore, the treatment is less efficacious.

And a similar mechanism can be seen in immunotherapy. Obviously, the mesenchymal stemness-like phenotype presents less antigen and is less visible for immune cells. So, there we see a common theme in resistance mechanisms. And also the keratinocytic phenotype of melanoma that was discussed yesterday is a phenotype that is associated with resistance for targeted therapy and maybe also for immunotherapy.

Axel Hauschild, MD: I think that’s a very interesting concept, which has been getting a lot of profile with the Cell papers. But I think one has to be a little bit careful. The patient number is very low. In fact, there was some overlap in the samples, which were analyzed. I found the presentation by Antoni Ribas on the patient treated with vemurafenib or the combination of vemurafenib and cobimetinib to be interesting, and the genetic analysis of complete responders and early progressive patients. Looking at the profiles of those, these resistance mechanisms you described, Reinhard, were not discriminating between CR and...

Reinhard Dummer, MD: This was not acquired resistance.

Jeffrey Weber, MD, PhD: This was baseline resistance. Caroline Robert, MD, PhD: Primary resistance.

Reinhard Dummer, MD: None of the patients had ever responded during the targeted therapy.

Dirk Schadendorf, MD: That’s true.

Reinhard Dummer, MD: So, this is another population, and this shows us how complex it is.

Jeffrey Weber, MD, PhD: So, you’re differentiating between primary resistance at the beginning of therapy, showing that it doesn’t have a relationship in the BRAF/MEK population to immune resistance. But then the suggestion is, when you acquire resistance after an initial response, you might become resistant to immunotherapy. So, is this the tumor striking back?

Caroline Robert, MD, PhD: But the thing is, we are missing data. We don’t want to be in this very difficult situation where we stop the targeted agents because the patient is progressing. Then we are afraid when we stop, when we interrupt the treatment, that the progression is going to be so fast that nothing is going to work after all. So, that’s why we would like maybe to initiate another treatment before that point, but we have no data of what would this do for that very patient if we continue. Maybe they will do well. And is it really smart to stop and to give something else? We need to do that in a controlled way, in a smart way, with biomarkers.

Jeffrey Weber, MD, PhD: Yes. Would you agree such a trial needs to be done, perhaps by using digital droplet, either PCR or a single-cell RNA assessment? Only with the single cell RNA assessment could you get this mesenchymal transition, which could be an early warning that, “Uh-oh, you better stop the targeted therapy, time to switch to immunotherapy.”

Axel Hauschild, MD: You are absolutely right. This trial needs to be done. But 90% of the patients have no access to clinical trials. Also, the physicians who have no center of excellence nearby, they need to do something. And I have a good friend of mine my age. She was suffering from stage IV melanoma with multiple lymph nodes—auxiliary, supraclavicular, head and neck—and she wanted to have a response as early as possible. She was BRAF-mutated, and she was on the BRAF plus MEK combination. And after 1 year, she got a 90% response. We switched her to a PD-1 antibody and now she is, for 9 months, in a complete response.

So, we have done this because, mechanistically, we hope that there was no resistance developed yet because there were no signs of resistance. And this has been discussed here at ESMO and will be discussed here at ESMO again. We can expose her with a rechallenge toward the same drugs again, in spite of progressive disease, and then we hope that she will have another response. But it’s a gut feeling.

Dirk Schadendorf, MD: I think this is a very controversial issue. Because your patient is a perfect example of how beautifully BRAF/MEK is working.

Axel Hauschild, MD: You’re right, you’re right.

Dirk Schadendorf, MD: And we have now patients for years actually on the drug. Looking at the COMBI-v and the COMBI-d data, we are seeing that after 3 years, more than 25% of patients have maintained tumor control and are still on drug. And why should you have any benefit and take any risk to a checkpoint blockade, which has a response rate of only 40%, taking a certain risk of relapse? I also have experience with a patient who was not responsive after interrupting BRAF/MEK inhibition, so I think we need to test that in clinical trials.

Transcript Edited for Clarity