Dirk Schadendorf, MD, director, Department of Dermatology, West German Cancer Center, University Hospital Essen, Essen, Germany, discusses expected toxicities with the combination of encorafenib (Braftovi), binimetinib (Mektovi), and pembrolizumab (Keytruda) in 

Data from the phase 3 IMspire150 trial (NCT02908672) and the phase 2 KEYNOTE-022 trial (NCT02130466) offer insight into potential safety signals with triplet regimens, Schadendorf explains. Additionally, the results of the phase 1/2 IMMU-TARGET trial (NCT02902042) demonstrated that encorafenib/binimetinib/pembrolizumab have overlapping efficacy and toxicity in patients with 

The ongoing phase 3 STARBOARD trial (NCT04657991) is evaluating the triplet combination of encorafenib, binimetinib, and pembrolizumab vs pembrolizumab plus placebo as a first-line treatment for patients with metastatic or unresectable locally advanced 

Regarding expected toxicities that could be seen in the STARBOARD study, pembrolizumab is the current standard of care for this patient population, so it is unlikely that unexpected safety signals will arise in the control arm, Schadendorf says.

Findings from the IMMU-TARGET trial showed that encorafenib/binimetinib/pembrolizumab was associated with increased inflammatory signals, including pyrexia; however, it is difficult to determine what agent within the combination generates the increase, Schadendorf explains. Notably, because the pyrexia was found to be manageable in this patient population, it is likely that patients will be able to tolerate the triplet therapy, Schadendorf concludes. 

Videos

Dirk Schadendorf, MD, discusses expected toxicities with the combination of encorafenib, binimetinib, and pembrolizumab in BRAF V600E/K-mutated melanoma.  



Dr. Schadendorf on Expected Toxicities With Encorafenib/Binimetinib/Pembrolizumab in BRAF V600E/K-Mutated Melanoma 

Dirk, we want to switch to the last question, and this is a question on our personal approach with immune checkpoint inhibition and dabrafenib-trametinib in the adjuvant setting in particular, but also for stage IV and the COVID-19 [coronavirus disease 2019] crisis. Germany has not been as affected as the United States by the crisis, but still it was a matter of debate. Can you summarize what you are doing right now in metastatic melanoma? Eventually it’s the same for the immune checkpoint inhibition in cutaneous squamous cell carcinomas.

From what we have seen, regarding COVID-19 infection and cancer patients, we have not seen many patients with skin cancer affected, at least in Germany, and there is not an increased mortality for these patients. On the other hand, one has to keep in mind in the peak season, in March and April, in Germany there was a huge reduction in various measures, including prevention and screening. One has to keep in mind that advanced tumors—including melanomas, squamous cell carcinoma, macrocell, and all that—while we have treatment options available, are much more deadly than obviously COVID-19, so these patients need treatment. That’s the No. 1 statement, and I think we agree on that. The question is, if we have multiple options like in melanoma who are 

 mutated, is there any preference? To be honest, we have during the pandemic chosen oral drugs in order to not force patients to come to the hospital for the infusions. We have used a lot of telecommunication and digital communication with our patients to manage adverse effects, to stay in touch with the patients, and that worked out quite nicely, to be honest. Right now we are also using nivolumab-ipilimumab as we did before the pandemic in a routine way.

We are doing the same, and we are behaving the same. This is also in concordance with the NCCN [National Comprehensive Cancer Network] Guidelines from the United States, where they were speaking about a variable to dabrafenib-trametinib in the adjuvant setting, but now it’s no longer an issue. The discussion was that it is interesting to prolong the intervals, and at least for nivolumab and pembrolizumab we have 2 options: nivolumab, to give 2 or 4 doses weekly, and pembrolizumab, 3 or 6 weekly. Is this also a consideration in your center and the other German centers to use during the pandemic in order to reduce the number of visits, the longer intervals?

Yeah, definitely. To reduce the workload for the staff and also for the patients not to get endangered by traveling to the center, we have taken these aspects into account, discussed with the patients, and used quite frequently.

A kind colleague—I’m not sure if it was you, Dirk—was providing me with a paper published in the 

 in 2007, in which MEK inhibitors were described to reduce virus load of coronaviruses. And that was from Chinese, all those working in the United States. I was not even aware that the coronavirus disease existed and discussed over there. Do you think there might also be an inhibitory capacity, not only of BRAF and MEK inhibition but eventually of PD-1 antibodies?

For the PD-1 antibodies, that’s clear. PD-1 antibodies are being developed for viral diseases as well. They are quite capable. The immune system is quite capable to catch and to control viruses. However, I think that the challenge is the lung toxicity and to discriminate inflammation in the lung caused by checkpoint blockade, especially combined checkpoint blockade, and the virus infection by COVID-19. To be able to discriminate that and to treat that accordingly is a challenge. Now we have more guidance. We know, for example, that steroids—especially in the more inflamed phases of COVID-19—are helping to control the inflammation. That’s what we also use for controlling immune-related adverse effects. Still, I believe it’s some sort of experimental setting, and it depends on the experience of the center and the comorbidities of the patients, which is guiding our way.

I have a last question, Dirk, before you summarize our session. Do you think we will see a renaissance of the good old interferons? Because they were described as antiviral agents in the late 1950s, and now that we have the problem of this virus, are they coming back to the adjuvant setting simply because they might also have a protection against influenza and COVID-19?

Simple answer: no. I don’t think that will be an option. We have more effective options to target the immune system. Checkpoint blockade is 1 way, and there are other ways to do that. As you have mentioned, COVID-19 and all the restrictions going with it—interaction in the hospital and social distancing and all that—will probably affect us for quite some time. It will restrict not only our travel time—Axel, you are a frequent traveler—but also the travel of our patients to the center. So we need to think of the toxicity profiles of the drugs we are using, of the comorbidities of our patients, and what can we do to protect our staff and our patients to guide our patients through this pandemic, keeping in mind that adjuvant treatment and treatment in stage IV is potentially lifesaving. It’s possibly endangering a patient’s life much more than the fear of COVID-19. With that, thank you for your lively discussion, my dearest Axel, and I hope you found it interesting and entertaining. See you the next time. Best regards from Germany.

Immune Checkpoint Inhibition in Stage IV Melanoma

Do you think there is a role for triplet therapy?

Because you mentioned the TRILOGY trial twice—but we need to say, for those who are not experts, that the TRILOGY IMspire150 trial was using different drugs. It’s a different company. It was vemurafenib and cobimetinib, plus a PD ligand 1 and not a PD-1 [programmed death-1] antibody, namely atezolizumab. The trial design was a bit different, but this is not of real importance. The difference, by the way, was 5.8 months in PFS [progression-free survival]. It was 10.2 vs 16, so they were ending up with the triplets in the same ballpark with 16 months. Vemurafenib-cobimetinib was a little inferior to dabrafenib-trametinib, but it’s an indirect comparison. Coming back to your question, even if the trial would show positive results and demonstrate a larger difference, the question is still if this could replace PD-1 antibodies alone or ipilimumab and nivolumab. So if these is considered as new standards of care, I would have my doubts because you have only indirect comparisons, but still I would say there is a space for the triplets. These results were a bit disappointing because they were leading to enthusiasm prior to this release, based on the data presented only 3 months ago by Georgina Long at ASCO [American Society of Clinical Oncology Annual Meeting], which was a 42—almost a world’s record complete response rate in COMBI-i trial part 1 and 2. Now we are at part 3, and the complete response rate goes down to 17. How can you explain that, Dirk? Is it patient selection?

There are several interesting caveats from the COMBI-i study. What you see is for the 2-year landmark and overall survival for the triplet is in the same ballpark as nivolumab-ipilimumab, so there is not much of difference from the other triplet from the other company. But the doublet, dabrafenib-trametinib, is overperforming. We have a 2-year survival rate that is more than 7% higher than what was described for COMBI-d, for COMBI-v, and also for vemurafenib and cobimetinib. Usually targeted doublets had a 2-year survival in the range of 50%, 52%. Now you are above 60%, no explanation. That’s another interesting detail that needs explanation: the duration of response. We have learned over the last 5 years that duration of response using BRAF-MET combination is in the range of 13, 14 months. Duration of response in COMBI-i trial is now above 20 months for the doublet. There is no explanation why the duration of response is so high. The question indeed comes back: Is a specific patient selection we have had in phase 1/2 possibly driven by the fact that these biomarker studies required cutaneous lesions, and are therefore driving the patient selection to a certain subset of patients that is particularly favorable? Then these patients should get triplet therapy, because if we expect 40% complete responses, that would be really great. Or what happened in this stage in the phase 3 study—is there a particular patient selection that was driving extremely the positive results for dabrafenib-trametinib as doublet still under investigation?

Thank you very much for your expert opinion. I have a question. Were you thinking about a negative trial early in advance, or were you convinced that this trial flies out easily as a trial that showed significant PFS?

I was convinced after the phase 1/2 data that this study would meet its primary end point and hit PFS. I was having doubt that an overall survival end point would have been positive, but with the PFS end point, I was quite convinced that it would fly. I think everybody was surprised about that.

I just wanted to confirm myself that I was not completely wrong in my believing that this is a positive trial, but it isn’t.

Triple Therapy for the Management of Melanoma 

As the older chairman of the 2 of us, I’d like to switch now to COMBI-i trial and the triplet regimen that has been used. For me, the presentation of the press release was a big disappointment. How do you feel about the results of the triplet, which was spartalizumab—a PD-1 [programmed death-1] antibody—dabrafenib, and trametinib and the idea behind it?

We have heard about the rationale that we are trying to combine all the goods of both worlds to be in paradise, so having quick, deep responses and making them durable and long-lasting in the majority of our patients. That was the concept. There were also some animal models in brackets. You always find animal models that support your rationale because you are not counting all the animal models that are not supporting your model. You have all that together, and it was a big disappointment for the entire community to see that the TRILOGY IMspire150 trial and also the COMBI-i trial, giving up-front, targeted therapy and checkpoint inhibition, did not increase response rate. The effect on PFS [progression-free survival] is marginal, and whether this will translate into any survival benefit is highly questionable in the way that the studies were done. In parallel, you are increasing toxicity substantially, so it brings us back to square 1—rethinking what and how it should be used, what we have achieved over the last 10 years with targeted therapies and a checkpoint blockade, and how we use it best for our patients.

Yes. We need to mention, Dirk, that in the COMBI-i trial the end point was progression-free survival, and the progression-free survival for the triplet was 16 months compared with 12 months—not unexpected for dabrafenib-trametinib—so there is a difference of 4 months. It was almost statistically significant. But in the end it was not statistically significant, and therefore it was considered a negative trial. Do you expect with longer follow-up time that this can change?

No, but it’s interesting that the positive trial, the TRILOGY IMspire150 trial, which led to a registration of the triplet in the United States, has only 2 weeks longer PFS. It’s 4.2 or 4.3 months difference in PFS, so you see how close and small differences in statistical design will determine whether a drug combination at the end will be approved. But I don’t see that the median PFS with prolonged follow-up will change. I think this is caught by the primary end point of the study. This is not true here, so this study will stay a negative. The only thing what we can do is try to learn from negative results in terms of what patient groups seemed to do better. That’s an important thing to discuss. Is there a role of triplet therapy at all for our patients?

COMBI-i Trial Data Updates 

If you don’t have a good explanation, may I ask you a question, Dirk? What role is the 10 mg/kg playing for ipilimumab? Because the trial you mentioned, the CheckMate 064, was using a conventional dose, which is 3 mg. Is ipilimumab high dose better than ipilimumab low dose? We know it’s not approved and there was a trial comparing 3 and 10 mg/kg in the stage IV setting.

The role of ipilimumab will become more intentional over the next years. As you have mentioned, there is an old study that has just reported 5-year overall survival data comparing 3 and 10 mg, and we have seen already a survival difference already in favor of that higher dosage. There is even an almost ancient trial that tested not only 1, 3, and 10 mg, but I think 0.1 mg as well, which is more than 10 years old, also demonstrating a dose-dependent effect of ipilimumab. Obviously, increasing the dose of ipilimumab is not only increasing toxicity—what we are focusing on also in various other tumor diseases—but also inflammation and possibly also clinical effects. This is something we need to take into consideration, not only for the metastatic setting but also in the adjuvant setting. In the neoadjuvant setting, we have 1 mg of ipilimumab, which is probably going to be established and used in the future. We have 1 mg with even longer treatment intervals in CheckMate 915, which is currently being tested. For toxicity reasons, that’s probably the right decision. For efficacy reasons, that’s still an open question.

I completely agree, Dirk. Would you agree with my following conclusion on the adjuvant part? The PD-1 antibodies are equally effective in the adjuvant setting, and they are also equally effective to dabrafenib and trametinib. The toxicity pattern is completely different—different toxicities, different grades—and the treatment discontinuation rates for dabrafenib-trametinib appears to be higher, but both of them showed efficacy of 50% improvement of relapse-free survival, distant metastases-free survival improvement. For dabrafenib-trametinib, there is an overall survival improvement over placebo. We need to say that ipilimumab is not a placebo in the adjuvant setting. It’s an active drug.

Absolutely. There are 2 aspects I would like to add. No. 1 is the potential of the reversibility of adverse effects. There is a small subset of patients who are receiving adjuvant treatment, and this will become even more relevant if we move to stage IV, who will have irreversible, mostly endocrine toxicities that need substitution. If the number of patients to be treated in order to benefit is increasing, then this is becoming relevant in the benefit discussion for the patient. This is No. 1. No. 2, I would like to add, is also the kinetics of relapse, which is different between checkpoint blockade and targeted therapy in stage III adjuvant disease management. With targeted therapy we are confident that as long as the patient is on treatment for this 1 year, the relapses are very small. But as soon as you stop, the majority of the relapses will occur within 6 to 18 months. For the checkpoint blockade, for pembrolizumab and nivolumab in the stage III disease, 25% of the relapses that occur over the first 4 years will occur while on treatment, within this first treatment year. That’s important to keep in mind. To select the right treatment your patients, you have to discuss it.

Ipilimumab Dosing Significance in Adjuvant Setting

This is very good bridge to the question I wanted to address to you to discuss CheckMate 238 on nivolumab vs high-dose ipilimumab, because you are mentioning it already. There is 1 caveat we didn’t discuss, and we can mention it only briefly. In CheckMate 054, in contrast to COMBI-AD, there is in the clinical trial protocol the option for the crossover. This allows for patients who were progressive on placebo and those who had at least 6 months of a stabilization without progressive disease on pembrolizumab when they stopped to cross over to pembrolizumab. This is certainly also diluting the interpretation of the data in the long run, particularly on overall survival.

That’s a very important point because if we see that at 2 years later also there is no overall survival benefit because of the crossover, would that mean that we would not need adjuvant treatment in stage III at all? We just wait? Because we know that roughly one-third of the patients will never relapse if they got surgery done properly. Would we just wait, and those patients who get new treatment get treatment? Would that be the consequence for you?

That’s a very good question. I was hoping that you wouldn’t address the question to me because it’s pure speculation. It’s a secondary end point, more or less, to ask this question, and it’s very fair. I would say no. I have my doubts that we would see overlapping survival curves 1 day. I’m pretty sure that they would still stay separated because there is a very good relapse-free survival in all these clinical trials in our days on the active drugs. This was also going along with overall survival improvement. I personally have no doubt, but this is gut feeling at the moment, and it’s not the data. I want to address the same question for you, because ipilimumab and nivolumab—we’ll say it in advance—showed more or less the same overall survival at the last data cut. How do you interpret this in the context of your question?

That’s what has stimulated our discussion in the last minute, because we have seen at ESMO [European Society for Medical Oncology Congress] as presented by Jeff Weber very nicely the update on the CheckMate 238 data. We see also in that cohort the relapse-free survival difference, the distant metastases-free survival difference, and surprisingly, at the 4-year landmark, the overall survival curves are completely overlapping. No difference at all. This caught everyone by surprise, and obviously it’s a little early to speculate on the exact reasons. If you look for subsequent therapies, one sees that in both groups roughly 70% of the patients are receiving systemic treatment, and in the group of ipilimumab-treated patients in the adjuvant setting, they are getting predominantly treated with PD-1–based regimens. Obviously, if you have been treated in the adjuvant setting with high-dose ipilimumab, there is a certain rescue potential for nivolumab subsequently in the metastatic setting. And the other way around, that in patients who have failed nivolumab, there is a small percentage who are benefiting from subsequent ipilimumab treatment. It is sort of surprising for me to see these survival curves, because in the metastatic setting we have had a similar clinical trial. This trial, CheckMate 064, which was presented by Stephen Hodi a couple of years ago, was testing a run-in phase of 2 or 3 months of nivolumab and then switching to ipilimumab, and the second arm was doing the opposite, started with the 4 doses of ipilimumab and then switching to nivolumab. The overall survival results for that were clearly different in favor of the group that started with nivolumab and then tried to continue with ipilimumab in contrast with the other group, so I was surprised to see that overall survival was overlapping so much. I don’t have any good explanation for that.

KEYNOTE-054 and CheckMate 238 Trial Updates 

We’ll move for a short period now to checkpoint blockade and the role of checkpoint blockade in stage III disease, preventing disease progression. We have seen at ASCO [American Society of Clinical Oncology Annual Meeting] but also at ESMO [European Society for Medical Oncology Congress] some updates on the KEYNOTE-054 study, which is a registration study for pembrolizumab in stage III disease, as well as CheckMate 238. Is that changing anything? What is new? Maybe you can briefly summarize what we have seen, and in the subsequent question we will try to elucidate whether this is changing anything in our view on adjuvant treatment.

Yes, you’re right. First, there is only 1 trial comparable to COMBI-AD. It was not exactly the same but very much the same eligibility criteria. This is the KEYNOTE-O54 study on pembrolizumab. This clinical trial is less mature because the data are a year and a half behind COMBI-AD in the follow-up. The 4-year data were presented by Alex Eggermont in an entertaining way, as always. What happened is that the 4-year data for the patients who are 

 mutated look very much the same, almost on the point to the 4-year data of the COMBI-AD study, so there is no sign that the targeted therapies are inferior to immune checkpoint blockade.

Let’s stop here because that’s a very important statement, comparing both studies, with all the caveats of comparing clinical studies. There is actually, in terms of clinical efficacy, no doubt that when they are treating patients in stage III with targeted therapy or a checkpoint inhibitor, the clinical benefit for the patient seems to be up to year 4, landmark 4, the same. What else makes you decide checkpoint blockade or targeted therapy?

This is the most burning question for all of us these days. I would say it’s only the adverse events, and an oral drug works as an infusion. So it’s the adverse events.

 Mainly are reversible after BRAF and MEK inhibitors, whereas there are some rare irreversible adverse events, particularly caused by autoimmunity, by the immune checkpoint inhibitors. Therefore, if I have the choice in a stage IIIA patient with relatively low tumor burden, I would favor in a patient with 

 mutation, dabrafenib-trametinib. For stage IIID, I would also say that dabrafenib-trametinib is in favor because if you compare the long-term data, they look better for the targeted therapies because there’s a very high risk of an explosion of the metastasis in this very unfortunate patient group, called stage IIID now. Therefore, for all the others I would leave it open. This is 90% of all patients nowadays.

All right. There was another interesting piece of datum which we were given as part of the KEYNOTE-054 update. It was not only the report of the 4-year landmark for relapse-free survival, but there was a surrogate marker presented: distant metastasis-free survival. Do you want to comment on that as well?

Distant metastasis-free survival in the past was a very good surrogate marker for overall survival, but in our days it may not be the same because it is pulling the development of distant metastasis, which is stage IV in melanoma and may contribute to the overall survival 1 day. Therefore, the data are very good. Again, in the same ballpark as for COMBI-AD, so I would say very good data. It’s very likely that we can cure a proportion of patients with stage III melanoma by an adjuvant treatment, and the goal for me in person is to cure patients, not only to prolong disease-free or distant metastases-free survival.

That’s an interesting statement, Axel, because indeed, as part of your answer—it was hidden—in the past distant metastases-free survival was considered a possible surrogate for overall survival with 2 or 3 years of delay showing the same result. That has been shown, actually, in the previous ipilimumab adjuvant trial conducted 8 or 9 years ago by the EORTC [European Organisation for the Research and Treatment of Cancer], which led to the approval of high-dose ipilimumab in the United States. There we see it. Actually a very similar pattern is also seen with the KEYNOTE-054 study, where we see a hazard ratio for distant metastases-free survival that is clearly statistically significant between placebo and the treatment arm, which was pembrolizumab. Why do you believe this is not true anymore? This is based on the overall survival data, which we have seen work for CheckMate 238.

Checkpoint Blockade Trial Data Updates

I would suggest to switch to the adjuvant treatments, and there were some releases at ASCO [American Society of Clinical Oncology Annual Meeting] and ESMO [European Society of Medical Oncology Congress].

Axel, you have been a key author on some of these presentations at ASCO and ESMO and also leading some of the publications on, for example, the COMBI-AD study. For COMBI-AD, maybe you can explain briefly this study design and what we have learned over the last 5 years?

 mutations, a combo of dabrafenib and trametinib, so BRAF and a MEK inhibitor. It was a randomized clinical trial on 870 patients. The randomization was 1:1 toward the combo or matching placebos. The results, in the first place, were very exciting, and everybody was expecting that over the long run the results might be less exciting. The big surprise is that we have 5-year follow-up time now, and still the 5-year data are in the same range regarding the hazard ratios and the relapse-free survival as the 2, 3, and 4-year analyses, and that I guess was a big surprise.

That’s a very important point, the longevity of the responses we are seeing and, as you mentioned, it’s exemplified by the hazard ratio. In the past we have speculated so much about cross between targeted therapy and checkpoint blockade. That’s what we will revisit in a minute, but before we come to comparison, or indirect comparison, of statements about checkpoint blockade in stage III disease, I want to home in on some details and subgroups of the COMBI-AD data set that are also of relevance for the clinical practice, which is the staging. There is stage IIIA, IIIB, and IIIC, and according to the AJCC [American Joint Committee on Cancer] seventh edition, we also have IIID. Is it of any relevance whether the patient is stage IIIA, IIIB, IIIC, or IIID in terms of offering adjuvant treatment? What is your take on that?

In the German guidelines we are saying that we need to offer the adjuvant treatment for all comers of stage III. The clinical trials, COMBI-AD and the trial on pembrolizumab, the KEYNOTE study, were done in patients with more than 1-mm tumor thickness in the cellular node, which means the cellular node micrometastases need to have a certain diameter. It is approved for all comers, but we need to discuss it. There is no question about value of dabrafenib-trametinib for stage IIIB, IIIC, and IIID. The hazard ratios are all the same, and in the old justification, it was also in the range of 0.50, so 50% improvement of relapse-free survival, and there is improvement of distant metastases-free survival, and there is improvement of overall survival. All this has been shown. The only debate that is still ongoing, and it will on go for the next years—I can predict it—is stage IIIA, because the risk to relapse is in general much lower, particularly in the eighth edition of the American Joint Committee on Cancer staging system. That is the only question, because there’s a low number of events.

That’s true, but if we assume it will be the same hazard ratio and that roughly 50% of the patients we can protect from tumor progression, isn’t that the same argument you would take for those patients as well?

I agree, and the risk for relapse on stage IIIA is underestimated in the AJCC classification. There are studies coming from Germany, 1 from your center and the other from the so-called Central Registry of Melanoma in Germany, nicely published in 2020 showing that the real risk is much higher. It’s in the range of 20%, eventually a bit more, for risk of relapse for stage IIIA patients. Therefore, it’s worthwhile to discuss it and to mention the pros and cons.

COMBI-AD Trial 5-Year Analysis Data

 Hello. My name is Axel Hauschild. I’m a dermatology-oncologist from the University of Kiel in northern Germany, and it’s my great pleasure to welcome you to the ESMO [European Society of Medical Oncology Congress] wrap-up on melanoma. It’s also my great pleasure to introduce my cochairman, professor Dirk Schadendorf. He is the chief of the department of dermatology at the University of Essen in Germany, which is approximately 400 miles away from my town. Hi, Dirk. How are you?

 Thank you. We want to discuss in this format for 

®, and we have prepared a number of questions regarding the updates from ASCO [American Society of Clinical Oncology Annual Meeting] and ESMO. My first question goes to Dirk on the molecular testing for stage III melanoma. Dirk, what is the importance of 

 It’s part of the work-up of a patient. If a patient is entering a treatment decision, 

 status is known. It’s standard of care now that patients, starting in stage III at the latest, have the information of 

 status, because that opens an avenue for adjuvant treatment in that setting.

 mutational status for neoadjuvant settings?

That’s a good point. Adjuvant treatment has been improved in Europe and across the globe since 2 years ago. It’s been a standard of care, including dabrafenib and trametinib. In the adjuvant setting it’s stage III disease, but the neoadjuvant setting is still experimental. There is a lot of excitement in the neoadjuvant field, but the most exciting clinical studies in the neoadjuvant field are checkpoint inhibition, a combination of checkpoint inhibitors. In the past there have been also targeted approaches in the neoadjuvant setting in nonoperable disease, but right now the enthusiasm is being mostly focused on nivolumab-ipilimumab in that setting.

 We need to say that at least in Germany, it’s not standard of care to use the neoadjuvant drugs at the moment. For the adjuvant setting, we are very happy in Germany that we have dabrafenib and trametinib approved for the 

 mutations, and for all comers, the PD-1 antibodies, pembrolizumab and nivolumab. And they are reimbursed in Germany, which is a very fortunate situation.

If you rephrase—if you’re a little more flexible with the term 

—it’s also tumor reduction. And targeted agents are perfect tools if a patient is 

 mutant, obviously, to try to get a quick and reliable response and trying to reduce tumor mass. In certain circumstances that could also open the way for surgery or other ablative treatment strategy, so if you would define that as neoadjuvant, it’s part also of the community treatment strategies we could apply nowadays.

 Thank you very much for clarifying, Dirk. Another question is on the routine testing of PD ligand 1 and 2 mutation burden. Is this belonging to your panel of biomarkers? Would you test them as routine, and how is the situation in Germany in general?

PD-1 ligand expression is standard of care in various tumors and disease entities. It’s established usually in the pathology. It’s part of the label, for example, in lung cancer. We also know that the PD-1 ligand status in melanoma is of relevance in timing the prognostic value off checkpoint blockade; however, it’s not part of the label. It’s not required to test, and in most of the centers across the board, not a routine tool to be analyzed as part of the routine work-up to make treatment decisions. Particularly in the academic centers, it’s being frequently used as a research tool for analysis. I don’t know how you do it in your center, Axel. A number of centers are doing that but not making treatment decisions at the final point.

 It’s true because the adjuvant trials are not making a difference at the moment for PD ligand 1 high and low expressers, and their efficacy seems to be the same. There is no significant difference. I completely agree with you, and the tumor mutation burden is still even more experimental than PD ligand 1 testing. It’s more far away from being used in the routine, correct? Is it not?

Yes. We are hunting for biomarkers or selection criteria for the last 5 or 6 years, with the first success of checkpoint blockade, and PD-1 ligand expression was the first marker. Tumor mutational burden was the next. We also have emerging discussion on, for example, inflammatory gene signatures as another tool of probably. And this entire discussion is fueled by the cancer immunogram put forward by Christian Blank and the Amsterdam group trying to explain which patients are really responding to checkpoint blockade. Actually, if we are trying to dig deeper, for example, also for palliative therapy, we see it’s the same group of patients who show these inflammatory gene signatures who are showing also PD-1 ligand and other factors, which obviously identify a patient group that is particularly sensitive to the new drugs we have available, including targeted drugs, surprisingly.

Dirk Schadendorf, MD, Essen University Hospital

Articles