Transcript:Jeffrey Weber, MD, PhD: Dirk, if you see a patient either with very locally or regionally advanced or metastatic disease, how do you stratify the patients in your mind, and how do you explain this to the patient? That is, what biomarkers or what criteria do you use to classify the patients in your own mind into risk groups?

Dirk Schadendorf, MD: So, it’s localized disease you’re talking about?

Jeffrey Weber, MD, PhD: Well, no. Advanced localized disease, as Caroline is discussing. Her opinion is that these are patients who are metastatic, and I think we all agree. So, it’s a metastatic patient. How do you classify them in your mind? What biomarkers, what criteria do you use?

Dirk Schadendorf, MD: Usually these patients who have some limited tumor growth or tumor load, they would be normal LDH (lactate dehydrogenase), and usually they would be in good performance status. So, I think these are those patients—whatever you give them: targeted therapy, immunotherapy—who have the best chance to get a complete response. And also, we obviously have, in these cases, surgery and radiation therapy, which is helping us possibly to achieve a complete response. I think the treatment goal nowadays would be to convert this patient into a complete responder by using systemic treatment, possibly in combination with surgery and radiation therapy.

Jeffrey Weber, MD, PhD: Interestingly, you’re all European dermato oncologists. Many of you have surgical training. Do you think there will actually be more of a place for surgeons or less of a place for surgeons in the future?

Dirk Schadendorf, MD: I think not at the very beginning. I think surgery might be the option. But if you are able to shrink the tumor significantly, there may be remaining nodules not responding and nodules you could either follow-up with radiation or with surgery.

Caroline Robert, MD, PhD: Yes, I think we will use surgery more. Not in the aim of acting alone and getting rid of the tumor, but more with helping other treatments. Surgery, radiotherapy, electrochemotherapy, cryotherapy, all these things are used more because we have effective treatments. So, the colleagues who are doing, for example, radiotherapy, now they expect to do more for us because they know that we have hope.

Jeffrey Weber, MD, PhD: Now, actually, we’ll give more work to the surgeons and the radiotherapists.

Caroline Robert, MD, PhD: I think so.

Jeffrey Weber, MD, PhD: Well, that’s probably a good thing for them. But, tell me, when you see the patients with metastatic disease, what tests will you do? That is, what biomarkers are you going to assess on the tumor, in the blood, etc, when you see them up front to help you stratify for risk and to choose a treatment?

Dirk Schadendorf, MD: I think the first thing that you need to do is test for a BRAF mutation. I think that’s the first thing that is helping you to decide whether the patient would also be eligible for BRAF-targeted therapy. BRAF/MEK is a new standard for this.

Jeffrey Weber, MD, PhD: And will you PD-L1 test them routinely, Axel?

Axel Hauschild, MD: I think it’s not a routine test in Germany simply because the PD-L1—negative patients are doing quite well—maybe not as good as the PD-L1–expressing patients or high expressers. But, therefore, it’s not part of the routine. This might change over time, but it’s not a selection criterion right now, for yes or no, for PD-1 antibodies.

Jeffrey Weber, MD, PhD: But, Reinhard, you’ve done some work in this field. Do you think PD-L1 testing is simply a surrogate for a “hot tumor” with a T-cell—inflamed environment?

Reinhard Dummer, MD: As an academic center, I think we have to think forward and get new information. Therefore, we do much more than just BRAF testing. So, we have a gene panel that is applied to every patient, and we are doing immunohistochemistry on all tumors that are resected, including PD-ligand 1 staining and CD-8 counts. And with this we hope we can develop criteria that help us to assign patients to certain therapies. But this is an academic setting. I still think that melanoma is a disease that should be treated in an academic setting because we can expect so much progress if we collect the patients and follow them carefully, and do, in addition, a high-profile workup. Therefore, we do it, the utility, and it’s also written in the ESMO guidelines. We think that PD-ligand 1 might be the first marker used for treatment decisions. Not alone, but certainly in the context of other markers. And the decision is only anti—PD-1 monotherapy versus combination therapy.

Dirk Schadendorf, MD: I completely disagree.

Jeffrey Weber, MD, PhD: How so?

Dirk Schadendorf, MD: I think PD-1 ligand marker expression is absolutely not ready for primetime. There are so many differences in the test systems we have. We have no alignment of the tests. We have tests that produce 80% of tumors that are positive, and other tests from other companies have only a 20% positivity rate. We have so many other factors introduced in the testing by handling the tissue, not only the antibodies we are using. I think we have to come to other methods measuring these markers. Whether this is NanoString or whatever, we have to find out. I completely agree this is a hot topic of research. I completely agree that testing PD-1 ligand expression has prognostic value for the patient. But I think, if you look, for example—on the response rates with a combination of nivolumab and ipilimumab—it’s much higher than monotherapy and also in PD-1 ligand positive patients. So, I think that’s absolutely not ready for primetime.

Reinhard Dummer, MD: Dirk, I have to tell you though, I just had a patient 2 weeks ago, an elderly patient who progressed off the targeted therapy and had metastases, and it was clear that he will switch to immunotherapy. Together with the pathologist, we saw the histology of his tumor and it was absolutely negative for PD-ligand-1. Actually, I would have loved to put this patient on anti—PD-1 monotherapy because he has a number of comorbidities where I think he might suffer a lot if he has the combination. But with this histology in front of me, I could only say, “If you switch to immunotherapy, we have to use the combination.”

Jeffrey Weber, MD, PhD: So, what was the final decision? What do you decide to do?

Reinhard Dummer, MD: He went on an ipilimumab and nivolumab combination therapy. And actually he had a very deep complete response. We also had the tumor load of this patient and the mutational load of this patient.

Axel Hauschild, MD: What is a very deep complete response?

Jeffrey Weber, MD, PhD: Well, 80% by 12 weeks I guess was the original criteria in the Wolchok paper. This is arbitrary, of course.

Dirk Schadendorf, MD: Yes, we have also a number of patients who responded on the bone.

Caroline Robert, MD, PhD: I think we all agree that this test might be interesting in the future when it’s optimized. Right now, it is not, and we cannot choose it routinely.

Jeffrey Weber, MD, PhD: I know. It sounds like we all agree it was a great concept, but it’s not quite ready from primetime.

Caroline Robert, MD, PhD: Optimize and use with other markers, probably.

Jeffrey Weber, MD, PhD: But, in risk-stratifying your patients, how often do you use the criterion just of clinical pace of disease? Reinhard, do you look at it that way?

Reinhard Dummer, MD: No. We look at this, we monitor this, and we try to find out the consistency over time. So, if we have multiple metastases from one patient, we are working on this, we are doing the correlations with other markers, like IDO expression, like CD8 counts. But I think it’s too early to draw clinical conclusions out of these data.

Transcript Edited for Clarity