Case (Dr Sullivan): A woman in her late teens presents with a nodule on her chin and is diagnosed with Stage IIIA melanoma

DR LOVE: Welcome to Oncology Today — Melanoma Edition, a special audio program focused on current management and emerging strategies in the treatment of melanoma. This is medical oncologist Dr Neil Love. For this program, I met jointly with Drs Jason Luke, Michael Postow and Ryan Sullivan. To begin, Dr Sullivan presented a patient from his practice.

DR SULLIVAN: This young woman came into my office, had seen a dermatologist and had a — she thought was a pimple, and it persisted. And they then looked at her graduation picture, which was taken 6 months before in anticipation of her graduating from high school, and it wasn’t there. And so her parents were concerned. And so went to the dermatologist. The dermatologist said yes, it certainly doesn’t look like a typical major problem, but let’s do a biopsy. And a biopsy was indeed performed and showed a 1.4-mm thick nonulcerated melanoma with 4 mitoses. And so that was how she got to our dermatology practice and our surgical practice.

DR LOVE: Can you comment a little bit about, Jason, about how a patient like this would be managed in your setting?

DR LUKE: I think I’d highlight the cosmetic versus the therapeutic aspect of the surgical resection. And I think this an area as medical oncologists is sometimes we don’t focus on enough, which is, when we get a consult, did the adequate margins, were they performed? Was the actual correct surgery done? This is a very tough case, because you’re going to have the added — it’s not as simple as the board’s answer of it should be 1 or 2 centimeters here. You’re going to have to take into account where exactly the lesion is. But that is really an important discussion point with the patient and their family about what would we do in a normal circumstance relative to what would we have to do in this scenario to get as much margin as we could get.

DR LOVE: Any comments, Mike, in terms of these — also, the surgical procedure. And what actually is done in the community?

DR POSTOW: Yes, I think the major concern in the facial area is whether a skin graft is going to be needed, depending on how much extra skin that the surgeon has geographically to kind of make a kind of primary closure without a graft.

Usually for forehead lesions or scalp primary lesions, the skin graft is almost always required. But maybe some areas you may be able to get away without doing a graft, maybe down lower on the cheeks, for example. But I think that’s a big consideration, at least in terms of the healing. And especially for younger patients, this 18-year-old girl, I think it sounded like her pimple was somewhere on her face, but a lot of people with scalp primary lesions will have a skin graft that goes on top, so there’s some hair loss. And that’s a big cosmetic issue.

Anyone diagnosed with melanoma anywhere, but especially on the face, very important to understand what kind of oncologic surgery we need to make sure that the melanoma is cured and not compromise that by any extent, but important to consider the cosmesis of these kinds of approaches.

Association between the use of tanning beds and risk of melanoma

DR LOVE: What was her history of her sun exposure?

DR SULLIVAN: Pretty limited, actually.

DR LOVE: And what was her pigmentation status?

DR SULLIVAN: She’s pretty light pigment. The Fitzpatrick scale, it’s probably a 1 or a 2. Fairly lightly pigmented. Fair hair.

But hadn’t had a lot of sunburns and hadn’t used tanning booths. Which is a very important thing to ask, because unfortunately, a lot of young women, some young men too, but a lot of young women use tanning booths and get a little too much extra UV exposure in an artificial way.

DR LOVE: That’s interesting about tanning booths. I hadn’t heard that. What do we know about that?

DR SULLIVAN: We know that really any use of tanning is associated with a high risk of melanoma. And the more use, the higher the risk still.

We know that there is virtually an epidemic amongst teenage and probably women use more than men, but young men also use tanning booths as well. The industry is not particularly well regulated, and oftentimes the tanning booths can be in places that are easy to access, like a gym or sometimes an apartment building. And so then it’s even less regulated than if you actually pay somebody at a regular interval to go to a place.

But ultimately, the problem is, it’s very intense UV radiation, and that does damage. And it’s probably not a whole lot worse than the old days when people put baby oil on and had their mirrors out, but it’s bad.

DR LOVE: Wow. Is the incidence of melanoma increasing?

DR POSTOW: I think there’s an increase in the incidence of primary melanomas. The metastatic disease incidence is about the same to my understanding, the statistics. But I think we’re seeing a lot of young patients with this disease, and that’s an important consideration. And building on the tanning bed issue, a lot of people think, I’m going to go to a tanning bed to get a base tan so that I won’t burn on my vacation to Miami, Florida or to the Caribbean. And that’s kind of the wrong philosophy to take with you And by kind of having a base tan, it’s not that you’re protecting yourself in any way from damage from ultraviolet irradiation. In fact, you’re doing damage to yourself by having that tan from even, obviously, artificial tanning from tanning beds. I think that’s a little bit of a misconception to get the base tan, to protect yourself from sunburn, which then in and of itself is harmful.

It’s best just to avoid it. You can use spray tanning, or just kind of be smart in the sun in general.

DR LOVE: Wow, really interesting.

Clinical care of patients with Stage IIIA melanoma

DR LOVE: Why don’t we continue on with this case?

DR SULLIVAN: The plan ultimately was for her to have an appropriate excision and the sentinel lymph node biopsy. Sentinel lymph node biopsies have been a standard practice for about 20 years in melanoma. And so there’s an injection of radioactive colloid, and a blue dye traces to the draining lymph nodes of that area of the skin. From an excision standpoint, a less than 1 mm-thick melanoma can be appropriately excised with a 1-cm margin. A greater than 2-mm thick melanoma should be excised with a 2-cm margin, and in the 1- to 2-mm range it’s a bit gray. The preference is to do a wider margin, if at all possible. I believe she had a 2-cm margin. Actually has a good cosmetic effect or outcome rather and did have involvement of her sentinel lymph node. She had 3 sentinel lymph nodes that were identified in the right parafacial node distribution.

Importantly, the involvement of lymph node was individual cells and some small clusters but no kind of big, either bulky — certainly wasn’t a bulky lymph node, and the amount of involvement would have been considered less than 1 millimeter, which is important when we begin to think about where is the data for adjuvant therapy and what’s the amount of lymph node involvement that may be associated with higher risk of recurrence?

DR LOVE: At this point, what would you be thinking?

DR POSTOW: I think the number 1 consideration is if she’s had all of the adequate surgery that she has needed and complete excision of the primary. It sounds like yes, that was the case. Sentinel lymph node biopsy, that has been performed.

Historically, there would be something called a completion lymph node dissection, which would be considered in any patients with a positive lymph node. Her nodal burden sounds very, very minimal. But regardless, the traditional, historical way of dealing with melanoma was to remove additional lymph nodes when there was lymph node involvement. However, there have been recent results from randomized trials that suggest there’s no melanoma-specific survival advantage to doing completion lymph node dissection following a positive sentinel lymph node. And, therefore, in most practices, especially with a patient like this with very minimal lymph node burden, there would be no more additional surgical planning. And then likely this patient would meet with a medical oncologist to discuss pros and cons of adjuvant therapy.

But I think it’s really important to recognize surgical practices are changing. Completion lymph node dissection is not routinely being performed anymore. Positive lymph node does not equal completion lymph node dissection any more. And I think a lot of these patients are just being observed with ultrasounds in their lymph node basin.

Risk of recurrence for patients with Stage III melanoma

DR LOVE: Jason, anything you want to add to that? Is that your basic approach? And what would you estimate this patient’s chance of recurrence would be? And what stage is she?

DR LUKE: There’s a bunch of really good questions and actually all inter-related, and that’s why it’s an interesting case, I think. You have 2 separate issues for this patient: One was the primary resection and the likelihood of possibility of recurrence locally. And then there’s the possibility of distant recurrence.

And so I have seen patients like this where you get into a weird figure 8 logic pattern where they want to do less surgery for the thing that we have more data that needs to be removed and then they want to do more surgery for the lymph nodes. But as was just mentioned, we don’t really do that anymore, based on lack of data.

It’s really important to emphasize again — an adequate surgical resection of the primary is the key in this case so far — with a positive node, historically, we’ve talked about doing more surgery. And there are clinical trials now that would advocate that for any patient who has even a single lymph node that’s positive, there’s an approval by the FDA to give adjuvant therapy. But the catch is, that’s a complicated question now. Because the clinical trials that were done to justify that benefit used now are the outdated AJCC7 staging system. And with AJCC8, we now see that patients like this, to have a very, very good outcome, meaning a less than 10% risk of death by 5 to 10 years from melanoma, and so that is not the same population as was studied in the adjuvant clinical trials.

And it really becomes an art of medicine question about should we be actually giving adjuvant therapy for patients with very tiny microscopic involvement of lymph nodes? Because when you start to get in the granularity of it, the possibility of side effects from, say, immunotherapy end up approximating what the potential benefit might be as well.

Perspective on the benefits and risks of adjuvant therapy for patients with melanoma

DR LOVE: What about in terms of estimating benefits? In breast cancer and other solid tumors, you have the concept of relative risk reduction. You just apply it on the absolute risk. Same concept here? Is that the way you see it?

DR SULLIVAN: Yes, that’s how we see it. We read out our trials in hazard ratio, which is essentially the statistical version of relative risk, and appropriately applied to that type of data set.

For patients like this, we’re not entirely sure whether or not the relative risk reduction of this type of — or adjuvant therapy, which we haven't talked about yet, applies, but assuming it applies, assuming that there’s a 50% reduction or so of recurrence and maybe a 40% reduction or improvement in the cure rate, which isn’t always the same thing.

If she has a 10% chance of recurring in a way where she would die if she didn’t get effective therapy at the time of recurrence, then we would be in a scenario where we’d say your 10% is reduced to 5% or 6%, and in some groups that’s enough of a risk reduction to put people through certain types of therapy.

In the melanoma community and in the community of providers that have been treating patients with adjuvant therapy, we have generally taken the approach that you really need to factor in the toxicity, and we just don’t have enough data about the long-term outcomes in this low-risk population.

In breast cancer, which you mentioned as a potential scenario, there are so many patients that enrolled on to the adjuvant trials in breast cancer that you really can have a much better approximation, not just what’s the chance of your reducing their risk of the disease coming back, but are they more likely to be alive 5 or 10 years from now? We don’t have that data.

And we definitely know that our therapies are at least as toxic in this patient population. And not just toxic in a way where you stop the therapy and people get better. But when we’re thinking specifically about immunotherapy, we can cause thyroid dysfunction that’s lifelong. We can cause pituitary dysfunction that’s lifelong. We can cause Type 1 diabetes that’s associated with a brittle outcome and likely in an 18-year-old would reduce her life expectancy.

There are a lot of intricacies to the concept of adjuvant therapy, particularly in a low-risk population — obviously, in a really high-risk population or in the situation where somebody has metastatic disease, nobody cares if they’re on levothyroxine for the rest of their life, if they’re alive. But it becomes a bit more of an issue when you’re talking about potential permanent toxicity, not just severe, but permanent toxicity in patients.

DR LOVE: Just to clarify though, what was her BRAF status?

DR SULLIVAN: We didn’t check. And the reason we didn’t check is, she had very low amount of disease. We would never be able to identify enough — and we don’t do microdissection of a lymph node to try and understand whether or not there’s a BRAF mutation. And the amount of disease in the primary is minimal.

I’m not sure I believe a negative test. And if she had a mole associated with her primary melanoma, I’m not sure I’d believe a positive test.

DR LOVE: Wow. That’s really interesting. If she did get adjuvant therapy, you’d give her immunotherapy?

DR SULLIVAN: Yes.

DR LOVE: Wow. Any comments?

DR LUKE: I wanted to double down, and especially this case, as being that perfect risk of recurrence versus toxicity problem. In a young woman like this, you have a disproportional impact potentially on toxicities that we generally act like aren’t a big deal in the metastatic setting. In a young woman who is 18 who may or may not even have considered whether or not she’s going to have children, giving lifelong adrenal insufficiency or thyroid dysfunction has a major potential to complicate her life choices down the road.

And so that’s something we really need to communicate with people. And we in melanoma land have not yet had to think about fertility preservation or even really these kinds of questions about long-term impact, because, unfortunately, melanoma was a disease where we didn’t think that way until just a couple of years ago, so that’s even more here. And I would have to say in my practice, I would be very reticent to treat this patient with adjuvant therapy. Because I think the absolute benefit of how much we could reduce her risk is not greater than the absolute possibility of giving her a lifelong toxicity. And to me, that’s not a great tradeoff.

DR LOVE: I was flashing on the fact, though, the idea that even if she does recur you can hopefully successfully treat her. And I don’t know if you know, as far as I know the only adjuvant trial that compared adjuvant treatment on relapse was the tamoxifen study. It was done by the Scottish group, like, a million years ago — and they literally randomized between giving tamoxifen after surgery to giving it on relapse, and there was a survival benefit. But since then, we just assume that patients are going to be treated, et cetera, when they develop recurrent disease. But again, you think you would treat this patient?

DR POSTOW: I would not treat this patient. I think that, just in agreement with Dr Luke, the risk of toxicities outweighs the benefit in terms of improving her long-term survival. Which, actually, it’s important to know, we actually don’t know if some of our newer adjuvant treatments, namely PD-1 inhibitors and BRAF and MEK inhibitors, whether they improve overall survival at all, regardless of substage of Stage III melanoma.

These are the same drugs this patient could have in the future should she ever have a recurrence. I would reserve them for that time period. And we’re really still trying to understand if the people we benefit in the Stage III setting are the same that would have that equally durable, wonderful benefit in the Stage IV setting, such to try to avoid overtreating patients who are already cured from surgery.

Case (Dr Luke): A man in his early 50s with locally advanced melanoma with a BRAF V600E tumor mutation receives adjuvant dabrafenib and trametinib

DR LOVE: I think this is a great way to start out. But let’s start getting into actual adjuvant treatment. And Jason, you have this 52-year-old man I’d like to hear about.

DR LUKE: Yes. My patient is a 52-year-old man, really, generally speaking, healthy. Had some blood pressure. But eventually noticed a lesion on his upper back. And being a guy really didn’t say anything about it until his wife caught him with his shirt off once and said, like, “What is that? I don’t think that was there previously.”

And so was referred to a dermatologist and got in the system. Had a wide local excision and sentinel nodal biopsy that showed a microscopic deposit, actually, in 2 of his nodes — so now we’re not talking about a single node being involved but more than 1, which actually is still an area that we don’t have definitive understanding of but is enough that most of us say, like, “Okay, that’s a positive node.” In that patient we were able to send off BRAF mutational analysis, and it came back as a BRAF V600e, which we expect in about 50% of our patients.

BRAF tumor mutation testing for patients with melanoma

DR LOVE: Could I just interrupt and ask, because I noticed that in your write-up up put next-gen sequencing of nodal mets. It shows this BRAF. It kind of surprised me to see that. How do you typically work a patient up?

DR LUKE: The standard of care should be the surgery, and the standard of care aspect of Stage III or Stage IV melanoma is to get the BRAF status in some fashion. And there are some companion diagnostic assays that are approved, and yet many in the community kind of have moved to the next place to say, “Should I be looking more broadly at other things that maybe aren’t FDA approved?”

In melanoma land, this is not a hard and fast rule. People do it different ways. We all come from academic institutions where we have kind of an internal bias towards doing large-scale sequencing, because we aggregate all those results for research, but even in the community setting there are interesting things that you can learn beyond, again, that don’t necessarily immediately impact the standard of care, though, like tumor mutational burden — if it was high versus if it was low. There are other mutations that can also show up.

We tend to be somewhat more interested in other things. But for standard treatment we, generally speaking, can use even just a PCR test to tell us the BRAF status.

DR LOVE: That is really interesting. You’re saying NGS, like Foundation One, for example?

DR LUKE: That’s one example of it.

DR LOVE: And again, do you think that that is a logical thing for a community-based oncologist to be ordering?

DR POSTOW: I think it’s reasonable, but I think —

DR LOVE: On a primary?

DR POSTOW: I think, to echo the main point, you only need to get a BRAF test for patients with Stage III resected melanoma. That’s the standard approach. All the additional mutations that are tested from those types of platforms are really investigational at this point in the Stage III resected population. Tumor mutation burden, PD-L1 testing, all of that is still a research question in my mind in Stage III resected melanoma.

If you’re in community practice, you have a Stage III patient and you’re looking over what kind of pathology do I have, if you want to just make sure you get that BRAF test, you are more than fine with just doing that.

Therapeutic options in the adjuvant setting for patients with melanoma

DR LOVE: I see this man had a completion node dissection with no nodes out of 9. Ryan, what would you be thinking about? Can you envision your conversation with this man?

DR SULLIVAN: Absolutely. By the way, a classic example of how men present with melanoma, which is, somebody says, “What’s that thing on your back that’s bleeding?”

DR LOVE: Really?

DR SULLIVAN: Yes. That’s just men. For whatever reason, we’re programmed either to deny things or to just kind of endure.

DR POSTOW: How often do you look at your back, right? That’s the other thing.

DR SULLIVAN: That’s true.

This is a pretty typical example of a middle-age gentleman who presents with melanoma. And risk wise, so 3 millimeters, ulcerated, a couple of nodes involved, this guy’s risk is probably in the 30% to 40% range that he’s going to die if he doesn’t have some intervention, either in the adjuvant setting or in the metastatic setting, from his cancer.

That’s certainly high enough to have a conversation about adjuvant therapy. And it’s certainly high enough to consider that even though there are risks of adjuvant therapy that they may be worth it.

For just the landscape of options that are available, interferon’s longstanding FDA-approved medicine for patients who have Stage II or Stage III melanoma — it’s gone out of favor in some places just because of the toxicity and the general malaise that everybody involved with interferon from the patients who were receiving it to the doctors and nurses who were caring for patients who were receiving it. And then actually since then, there’s been some data that’s shown it’s less effective.

Ipilimumab, which is an anti-CTLA4 antibody, FDA approved for Stage IV melanoma in 2011, became FDA approved in the adjuvant setting for Stage III melanoma in 2015.

High dose was the dose that was tested in those trials. Lots of toxicity, more toxicity than we see in patients at the lower dose and probably more toxicity even at the high dose in the Stage III setting and in the Stage IV setting.

Next came the PD-1 antibodies, and nivolumab was approved in either late 2017, early 2018 for adjuvant therapy, based on a trial that demonstrated superiority of nivolumab to ipilimumab in terms of relapsed-free survival and also was less toxic. And then pembrolizumab, based on a trial of pembrolizumab versus placebo. It was FDA approved just earlier this year. And then BRAF targeted therapy in the form of dabrafenib, a BRAF inhibitor, trametinib, a MEK inhibitor.

For a patient who’s middle aged, otherwise healthy, has reasonable risk to discuss adjuvant therapy, we typically — and has a BRAF mutation — we’re going to talk about BRAF-targeted therapy, and we’re going to talk about anti-PD-1 single-agent therapy.

Efficacy and tolerability of immunotherapy in the adjuvant setting

DR SULLIVAN: As a general rule, oncologists and patients love the concept of immunotherapy. Immunotherapy is kind of inspiring our bodies to fight our cancer better. It’s empowering to offer it. It’s empowering for a patient to receive it.

And the challenge is that it’s the immune system, when it’s activated, can do some really bad things. And so all of us have seen just the worst outcomes you could possibly imagine from anti-PD-1 antibody therapy, even though severe life-threatening toxicities only happen 15% to 20% of the time.

DR LOVE: Only.

DR SULLIVAN: Right.

DR LOVE: That’s still a lot.

DR SULLIVAN: That’s still a lot.

DR LOVE: Is that right?

DR LUKE: That might be a bit high.

DR LOVE: That’s high.

DR SULLIVAN: Severe, so Grade III/IV/V toxicity is 15% to 20%. If you talk about fatal toxicity, it’s probably 1 in 200.

DR LUKE: I would just quick note that the ones I worry about in the adjuvant setting are the irreversible ones.

DR SULLIVAN: It’s fair enough.

DR LUKE: The endocrinopathies tend to be more like about 5-ish-percent to 10%.

DR SULLIVAN: Eh, thyroid dysfunction is about 20%.

DR POSTOW: But I think if you look at the Grade III to IV toxicities in the trials, I think that’s where most of us get the statistics that we’re speaking of. It’s about 15%, 20%, that’s Grade III or IV toxicities. Some are not so bad at all. You have a patient come in, they feel great, you check their liver tests and they have Grade III liver toxicity. They do need to take steroids and hold their treatment, and there is some comorbidity, of course, with being on prednisone in that kind of respect, and that counts as a moderate to severe toxicity because it’s a Grade III or IV toxicity. But those patients usually resolve. They feel better. They get off steroids eventually, and they’re fine.

DR SULLIVAN: True.

DR POSTOW: I think it depends on kind of how bad we’re really talking about are the really bad things.

DR SULLIVAN: I think what I’d counsel my patients, about 20% chance of a permanent endocrinopathy. It’s about 20% for thyroid. It’s about 1% or 2% for pituitary. It’s about 1% for diabetes, in that there’s a 1% or 2% chance of these terrible toxicities that can happen: myocarditis, neuropathies, that sort of thing.

And then I also tell my patients about immunotherapy, that in the adjuvant setting, anti-PD-1 antibody, the nivolumab or pembrolizumab for melanoma, is unique in adjuvant therapies because it is a therapy associated with cure in the metastatic setting, and it’s associated with presumably cure in the adjuvant setting. There’s no other therapy like that. Chemotherapy for breast cancer, if you give AC followed by paclitaxel in a woman with metastatic breast cancer, she’s not cured.

DR LOVE: Same thing with tamoxifen.

DR SULLIVAN: Same thing with tamoxifen. Same thing with FOLFOX, FOLFIRINOX. Think about the adjuvant chemotherapies that we use. They’re not curative in the metastatic setting. They help, but they’re not curative. We’re potentially offering a curative therapy early, but we could offer that same curative therapy or intensify it with, say, a combination of immune checkpoint inhibitor therapy down the road.

Choice of dabrafenib/trametinib versus immune checkpoint inhibition as adjuvant therapy for melanoma with BRAF mutation

DR SULLIVAN: I do make sure that patients understand that there’s permanent toxicity, that there is a low risk of severe toxicity that could be life threatening, but a not so low risk of needing to be on steroids, 15%, 20%. And that they may be cured anyways if their disease comes back and they receive that type of therapy.

For BRAF-targeted therapy, it’s much more like offering AC followed by paclitaxel for breast cancer. When patients have side effects, you stop the medicine. The side effects go away. You don’t have to put patients on 6 weeks of prednisone and/or intensify to infliximab or some other secondary immunosuppression when the side effect doesn’t go away. It’s completely consistent with the way we practice in the adjuvant setting for 20, 30 years.

It’s probably associated with a similar risk reduction. And it’s also probably curative in the adjuvant setting but not necessarily curative in the metastatic setting. And so in some ways when patients have BRAF-mutated melanoma and have Stage III disease that’s been resected, BRAF-targeted therapy is — it’s a way to offer somebody the best version of that therapy in the course of their care, because when they’re metastatic, it’s not going to be the best therapy for them necessarily.

DR LOVE: Would you be thinking about BRAF/MEK for a patient like this?

DR SULLIVAN: In our practice, we tend to offer much more BRAF-targeted therapy in BRAF-mutant melanoma patients for all those reasons.

DR LOVE: And do patients accept that?

DR SULLIVAN: Yes.

DR LOVE: Interesting. It’s kind of surprising, to be honest with you. Mike?

DR POSTOW: Yes. For Stage III resected populations of patients with the BRAF mutation, I’ve actually been thinking more strongly recently about using BRAF and MEK inhibitors, mostly because we believe that the efficacy is similar between adjuvant PD-1 and adjuvant dabrafenib and trametinib.

Tolerability of and quality of life with BRAF/MEK inhibitor combinations versus immunotherapy

DR POSTOW: The big difference is on side effects. And when we think about the side effects, the BRAF MEK inhibitors are more likely to cause some degree of side effects than PD-1 in terms of total numbers of patients having some kind of side effect. But I think the nice thing about the BRAF and MEK inhibitors from a toxicity standpoint is, if you stop the drugs, those side effects will kind of resolve. And it’s very different with immune therapy. When you stop the immune therapy, sometimes the side effects linger. Sometimes they can be permanent.

I think you can do more damage with immune therapy, although it’s rarer in incidence. And I like the idea of less irreversible side effects with BRAF and MEK inhibitors in the adjuvant setting.

DR LOVE: That’s interesting, because I’ve got to say that I kind of had the feeling when the data first came out — I think it was an ESMO meeting when both of them came out — I had the feeling initially that people are much more favoring immunotherapy for BRAF, because of that feeling that somehow it’s curative or something. Is that kind of, you think, dissipating a little bit now?

DR LUKE: I think that there was, honestly, a little, and so frankly, many of the investigators who worked in melanoma for a long time are immunotherapists. And so we have an inherent bias that we're going to go that way. And I think that all of this data came out, and I think things have shifted around — it’s the same data but people’s interpretations back then.

DR LOVE: Right.

DR LUKE: And I would emphasize that point once again, which is, the data in terms of recurrences, it looks almost the same. The slopes of the Kaplan-Meier plots look different, but when you get out to years after treatment in the adjuvant setting, the BRAF/MEK combo looks just as effective as the PD-1 combo. It really becomes, what therapy do you think is going to be the right — what to give for this patient.

DR LOVE: Do you buy into Ryan’s statement of immunotherapy is curative in the adjuvant and metastatic but BRAF only in the adjuvant? Because, I mean, don’t you see long-term disease-free survival with BRAF/MEK?

DR LUKE: That’s a nuanced question, and that’s a tough one. And that gets us to the edge of the true data that we have. I mean, we have had this recent analysis of 5-year outcomes in metastatic setting, which certainly would suggest that some patients are deriving that long-term benefit. The caveat is that even in those patients who are still on those trials, we don’t believe we can stop the BRAF inhibitor.

Your question is, really, how many of those patients could we take off drug? And none of us are actually willing to do that. And so that’s where this comes into play a little bit as well.

DR LOVE: It’s interesting. When I was listening to what you were saying about checkpoint inhibitors, I was flashing on the fact that the other major situation that I’ve seen, at least in CME, where you see adjuvant therapy in solid tumors — I call it adjuvant — is in Stage III lung cancer, where they now give durvalumab consolidation for a year. And it’s funny, because you don’t hear anybody talking about toxicity there, because they have Stage III. They’ve just got chemoradiation. They had a 75% chance they’re going to recur. But I guess the same stuff’s going on with them that you just described, right? They’ve got COPD, whatever, but they’re not maybe seeing it, but it should be there, right?

DR SULLIVAN: I presume that they’re going to be seeing a fair amount of toxicity. Although I’m not sure I’m willing to make the same conclusions about lung cancer patients as melanoma patients for all the — meaning, like, it’s entirely possible that they’ll be more effective in the adjuvant setting than in the metastatic setting in lung cancer patients who have comorbidities and maybe right for treatment now but might not be right for treatment down the road.

DR LOVE: Absolutely.

DR SULLIVAN: And so I’ll just want to make sure my statements stay with melanoma, but the concept is, of course, the same.

DR LUKE: For this patient additionally, there was one other factor that’s, like, the most obvious, but we didn’t really talk about it, which is to say that the BRAF and MEK inhibitors are pills that you take at home. And the infusions are a monoclonal antibody, which honestly is pretty easy. It’s either every 3 or 4 weeks at this point. But for the patient that I was discussing, that’s obviously why he chose what he chose is, he’s a busy guy. He didn’t want to come in. He said, I’m going to take these pills, and I’m going to go to work and I’m going to do my stuff, right? I think even on some level that simple stuff is an important thing to think about in patients in terms of what their perspectives are in terms of all this treatment.

DR LOVE: Yes, I was going to ask you about him as a patient and what he valued — I thought it was very interesting what you just said in terms of, I guess, the transportation or coming into clinic thing.

Monitoring and management of side effects associated with BRAF/MEK inhibitors

DR LOVE: The other thing I want to hear a little bit about — I’ve been hearing for years about the issue of BRAF/MEK from you all in terms of toxicity issues, but now you have it in the adjuvant setting. I’m curious, for example, what happened with this man.

DR LUKE: He started on dabrafenib and trametinib, based on the idea it was oral therapy and he actually, I think, worked in some sort of construction-related thing, and he was an independent kind of guy, again, why he never found the thing in the first place. But he started on the pills. And actually, as is relatively common in this patient population, about a month-ish into treatment, he started getting sweats and feeling chills. And, fortunately, I had told his wife that they should be on the lookout for this, because we find that in upwards of half, maybe more of patients will get sort of this pyrexia syndrome taking dabrafenib and trametinib, really any BRAF/MEK inhibitors, but definitely with this combination. And they tend to escalate over time, meaning that they’ll start at something and get worse and worse and worse. And the worst case scenario for us is with the really tough guys who never tell anybody. They end up getting hospitalized for dehydration, et cetera, et cetera.

But as Dr Sullivan had mentioned, it’s actually usually pretty easy with this, which is, you just take a break from the drugs, and all the symptoms go away. I counsel the patients and their families that when they start on this they should be on the lookout, so as soon as they feel chills or sweats or anything like that, just stop and then give our office a call. We’ll talk it through. Usually about 48 hours to 72 hours off the drugs, the guy’s feeling fine again and then we can restart.

And that’s what happened with this guy. It was around 30 days he started getting that, and again his wife noticed he wasn’t feeling well. She asked him, and they called, and we said, “Just hold off on the drugs.” I think he held them for 4 days, actually, and he was able to restart and really didn’t have any issues after that. And it’s an idiosyncrasy with these drugs. We don’t really understand why this happens, but this transient pyrexia that we can just dose delay and then it doesn’t recur. It certainly does in some patients, but the majority of patients are able to actually just take a break and then go right back to it.

DR LOVE: What about corticosteroids?

DR LUKE: There are all manner of algorithms that people have proposed on how to properly do this. In the melanoma community, again due to immunotherapy constraints, we are pretty corticosteroid-averse. And I tend to see a fair amount of toxicity just from the corticosteroids. And given our experience, all you have to do is basically just stop then you don’t need to do any of that stuff. Sometimes we tell patients, “Stay well hydrated. Maybe take some NSAIDS.” Every once in a while we’ll get one of these patients that really this just keeps coming back. After, like, the third time of this sometimes, then we’ll do a low-dose 5 mg of prednisone or something like that to try to help them through, but that’s definitely an art of medicine sort of thing.

The big picture is just take a treatment holiday. Usually it will go away.

Counseling patients who are receiving BRAF/MEK inhibitors about sun exposure

DR LOVE: You said he was in construction. What did you advise him in terms of sun?

DR LUKE: The sun recommendations are really the same for everybody whether they’re in construction or not, but really its once you’ve had this you should be cognizant that you will have some sensitivity to the sun beyond what you normally would.

With this combination it’s not as bad, but certainly it can be that patients can be more likely to get burned. We do counsel them that try to avoid being out in the sun for continued long periods of time. But we tell our patients they should be slamming the sunblock on and wearing a hat anyway. I didn’t tell him he couldn’t go to work, but I do think he eventually transitioned more into an administrative role as opposed to being out there every day.

DR LOVE: I mean, I’ve heard stories about people. They put their arm out the window for 15 minutes, or whatever. That’s extreme, but maybe it was a different combination or…

DR POSTOW: Vemurafenib and cobimetinib. Yes, there’s a little bit more sun sensitivity with that combination as distinct from the dabrafenib and trametinib combination.

DR LOVE: Because there, people were saying, we, like, tell people, do not — is that...?

DR LUKE: It can be very minimal exposure leading to very significant burns.

DR LOVE: Any other skin issues? There was the issue of second cancers that was out there for a while. Is that still there?

DR LUKE: Honestly, I don't want to say it was a hypothetical, because it truly was defined as second cancers, but this was something I think that was, to some degree, overblown in the first place.

On monotherapy BRAF inhibitors, we would get this paradoxical activation of the mitogenic protein kinase pathway, too complicated to talk about. That pretty much goes away with the combo. It’s not entirely gone. We do see patients get some keratoacanthomas and sometimes very early squamous lesions. Which is why if patients are on these treatments, it’s not that you should forget about them and tell them they never need to come in. You do need to see them regularly, and you should be doing a skin exam, because you can find these things, but they’re almost always pretty easy to just take off.

DR POSTOW: The other thing I think to mention about sun, as we’re thinking about putting patients on treatment that have had Stage III melanoma diagnosed, I think it’s really important a lot of patients think, sun, I could be at risk of the melanoma coming back. And I think an important distinction is, we don’t know that sun exposure increases their risk of recurrence of their Stage III melanoma.

Of course, these patients are at risk of an independent primary melanoma on their skin because they’ve already had a cutaneous primary, as in Dr Luke’s example. But I think that’s a helpful thing for patients to think about, because they’re worried about their Stage III melanoma recurring, and that’s an important principle for them to absorb.

DR LOVE: Wow. Interesting. Makes a lot of sense.

DR POSTOW: That you don’t need to be in a cave all the time to avoid this melanoma coming back. And I think that helps patients. And that they usually see dermatologists so frequently, so any little funny mole is going to come off of them very quickly.

Unfortunately, at that point with Stage III melanoma the diagnosis has already been made. My concern as a medical oncologist is making sure that that doesn’t come back to be Stage IV at some point. And I don’t think sun exposure in that specific context is maybe as much of a concern. Be smart in the sun, sure, but you don’t have to be crazy.

Selection of anti-PD-1 antibody therapy for patients with locally advanced melanoma

DR LOVE: Any comment, it was mentioned in terms of both pembro and nivo being approved, choosing between the two.

DR POSTOW: I think they’re the same in terms of efficacy and side effects, and across so many different studies that’s been the conclusion, although they’ve never been compared directly head to head. It’s a dosing interval question mostly that nivolumab is either given every 2 weeks or every 4 weeks, depending on the dose. And pembrolizumab at this time is given every 3 weeks. There may be changes to these dosing intervals, but it’s really just a dosing interval question.

DR LUKE: I would just tag on that quick. To be aware that there is a plan and there is an application to the FDA to move pembrolizumab dosing to every 6 weeks. And some of us have been in meetings and arguing with each other about what the appropriate amount of time is for this. And I think there is no right or wrong answer. I think the important point would be that if we can believe that the dosing interval is less important, then it’s really going to be about individual patients. How do we take care of them?

One could imagine that if you’re only dosing every 6 weeks, that might be a long time to see somebody, given that a lot of the side effects that are going to happen are going to happen within that first 6-week window. Maybe even if we’re dosing every 6 weeks you might want to see patients more frequently than every 6 weeks, but that’s definitely going to be something that evolves over time.

Role of vemurafenib in the adjuvant setting

DR LOVE: What about the other BRAF/MEK combinations we’re going to talk about that in metastatic disease, are they being looked at in the adjuvant setting?

DR SULLIVAN: I don’t think there’s active open trials.

DR POSTOW: Not with BRAF/MEK in the adjuvant, to my knowledge either, I agree. I do think it’s worth just mentioning that vemurafenib as a monotherapy for BRAF inhibitor was looked at in the adjuvant setting in patients with resected Stage III melanoma, but they also included very high-risk Stage IIc melanoma. And there, a lot of complicated conclusions from that study that one could make, but I think one takeaway I have is that even BRAF inhibitor monotherapy had some degree of activity in Stage III resected melanoma, but because the BRAF/MEK combo has demonstrated superiority in Stage IV melanoma over BRAF monotherapy, at this time there’s really no BRAF monotherapy use in the adjuvant setting and no additional exploration of other BRAF/MEK combos in the adjuvant setting.

DR SULLIVAN: And I think you might imagine that there will be trials of the alternative combinations in various different studies, whether they’re industry sponsored or whether they are cooperative group studies. At this point, there’s no active BRAF/MEK combination trials in adjuvant.

DR LUKE: Here’s one other point that we haven’t touched on, which is, we’ve talked about Stage III, which is certainly the on-label use of all of these adjuvant therapies. However, the field is starting to consider Stage II, and there, actually, there are some rumblings about using BRAF and MEK inhibitors.

There’s a somewhat odd idiosyncrasy in melanoma, which is that high-risk Stage II disease is actually more risky than low-risk Stage III disease, which seems odd on a first pass, but it has just to do with the biology of melanoma. Many of us believe that for Stage IIc melanoma and maybe Stage IIb that those patients are actually at higher risk than Stage IIIa, and, therefore, if we’re going to treat Stage IIIa and b, then we ought to consider treating these patients as well. And there are clinical trials ongoing to look at the efficacy of these therapies in that population.

DR LOVE: A lot of it comes down to what the actual risk is. And I’m kind of curious — it kind of sounds like 10% was getting too low for you all. But what’s your bar, so to speak, 15, 20?

DR LUKE: It’s about right. I mean, there are machinations that each doc will give them an hour and they’ll pull data from every study and make it right handed, left handed and try to figure it out. But yes, it’s, broadly speaking, from my purposes, what I like to think, which we started with, which is, is the absolute risk reduction greater than the potential for severe toxicity? And if the answer is yes, then I’m probably on board after discussion with the patient and everything.

For example, Stage IIb and IIc, you land right around there. So IIb ends up around 15, and IIc can be 20, 25. And so then that’s a little bit higher than the IIIa population we were talking about.

DR LOVE: Again, kind of reflecting back on breast cancer, because it kind of is an intuitive — I know it’s different people, different disease, different situation, but still, once you get into 5% absolute benefit or less, you start thinking about acute leukemia, cardiomyopathy from doxorubicin, anything. And it kind of sounds like it’s following a similar — I see you all kind of shaking your heads — paradigm here.

Case (Dr Postow): A woman in her early 30s with Stage III melanoma with a BRAF V600E tumor mutation receives adjuvant dabrafenib and trametinib

DR LOVE: I want to learn a little bit more, also, about how taking care of patients in the adjuvant setting and issues that come up. You have also, Mike, had a patient who got adjuvant dabrafenib/trametinib, a 32-year-old woman. Can you just briefly go through her case? And I’m curious what happened when she was treated.

DR POSTOW: Sure. This is a young 32-year-old woman, no other medical problems, 2.4-mm nonulcerated melanoma on the left flank. And this patient had a sentinel lymph node involved in the left axilla. It was 1 lymph node with 2 millimeters of melanoma. And that’s an important point, I think, as we’ve been touching upon, to really push the pathologist to know how much melanoma is in the lymph node. The fact that this was 2 millimeters, that, of course, passed the 1-mm threshold for which we have adjuvant therapy data in these kinds of patients with nonulcerated primary melanoma.

And this patient did not have a completion lymph node dissection, as is usually the case to avoid that in these patients, and started treatment with dabrafenib and trametinib. And this is a great patient — no other medical problems, very reliable, would have been a great candidate for PD-1-based adjuvant treatment and also BRAF/MEK. And I think the decision here is really again on the wish to avoid the very rare but not zero possibility of long-term permanent toxicity with immune therapy.

DR LOVE: What was her life situation and kind of her thinking about this?

DR POSTOW: She was married. She had a family, a young family, and didn’t want to have any chance of something that would kind of render her in any kind of permanent disability. Now, I told her that those are very, very rare events to happen with immune therapy, mainly the neuropathy or other severe endocrinopathies or the diabetes that we had spoken about. But I think when you introduce the possibility of another permanent medical problem in a young patient with immune therapy, that’s a particularly tough issue. And the longer I’ve been doing this, I think one could say, you get more experience. You get more comfortable with immune therapy.

I actually feel the opposite of seeing so many more things now having done this so long that I’ve actually been astonished with some of the really tough side effects that some people can go through, including some permanent ones. And there is a different calculus with using immune therapy in the metastatic setting than in the adjuvant setting. And in patients with a BRAF mutation, I think very strongly about a BRAF and MEK inhibitor in those patients, knowing that I’m more likely to make them feel that — in some case with the fever and chills episode or upset stomach or these other issues — but to say, “Then just stop your pills. You’re going to be fine in the long run.”

I can’t quite as comfortably say that with adjuvant PD-1-based immune therapy. I use adjuvant PD-1-based immune therapy all the time for certain patients, especially a different context and certainly for BRAF wild-type patients. But I think when you have a choice, and we don’t even know really what the overall survival benefits of any of these treatments are, so I think that’s making me even a little bit more cautious.

Reduction in the risk of recurrence with adjuvant therapy for patients with Stage III melanoma

DR LOVE: I’m kind of curious in general with these patients, and I’m curious about this 32-year-old woman, how often do you see patients who really want to hear numbers. Like, “Tell me your best estimate on what my risk of recurrence is and how it’s going to change with treatment.”

DR POSTOW: A lot of times people want to understand the numbers, but if you get bogged down in the details of the statistics, it’s very difficult. I think some of the key points to know, Stage III melanoma has a high risk of recurrence of anywhere from 30-ish-percent in some settings up to 70%, 80% in very high risk Stage III.

In terms of melanoma-specific survival, one of the issues that I caution patients about is that the staging system, and when you look at all those Kaplan-Meier curves on the graphs, those are overall survival curves and not recurrence rate curves. And, in fact, if you look at Stage IIIa, IIIb, IIIc, whatever, on the staging criteria, the risk of recurrence is going to be higher than what you see for the overall survival statistics. Decent risk of recurrence.

And then I tell patients about 40% or so, or maybe up to about 50% in terms of reducing the risk of recurrence with adjuvant treatment. And I think that is enough of a number that most people are comfortable with.

DR LOVE: That is very useful information. It sounds like you have some pretty interesting numeracy issues as you get in the lower end.

DR LUKE: For me, the most important number is the hazard. In almost all the trials, the hazards are basically 0.5, and it makes it actually pretty simple to explain. If you take the treatment, will cut the risk in half —

DR LOVE: Although technically it’s not exactly half, because, right, it’s, like, a little bit less. That’s probably why you said 40 to 50.

DR POSTOW: Depending on what trial you look at, what the comparator is.

DR LOVE: Also, relative risk is per year, so it’s not exactly, I guess — theoretically...

DR POSTOW: Agree.

DR SULLIVAN: True, but it’s close enough.

DR LOVE: Yes, no. Close enough.

DR SULLIVAN: It’s close enough where it’s definitely easier to grasp than to say your risk is 48%.

DR LOVE: Oh, right. Yes. No.

Recurrence-free survival benefit with adjuvant nivolumab, alone or in combination with ipilimumab, compared to placebo for patients with Stage IV melanoma and no evidence of disease in the Phase II IMMUNED study

DR LOVE: What about Stage IV NED?

DR SULLIVAN: It’s a wonderful question. And there is data with 1 drug — actually, with 2 drugs. I think there was data with ipilimumab and then the nivolumab versus ipilimumab CheckMate 238.

DR POSTOW: And then also data from ESMO. Recent data from ESMO looked at this population of patients, Stage IV, either surgically resected or completely irradiated and randomized between placebo/nivolumab monotherapy and ipi/nivo combo at the regular dose of ipi, 3 mg/kg. It’s the IMMUNED study for NED at the end, IMMUNED, no evidence of disease, and it was relapse-free survival only, randomized Phase II, approximately 50, 60 patients in each arm, showing really, really impressive relapse-free survival for the ipi/nivo combo and nivo, even monotherapy, over placebo, just follow-up.

You can certainly improve recurrence-free survival in Stage IV NED patients by giving them checkpoint inhibitors, either single agent or combination immune checkpoint inhibitors, but we don’t know overall.

DR SULLIVAN: And practically speaking, though, the only drug that’s on label to offer right now would be nivolumab.

DR POSTOW: Right.

DR LUKE: I think the other weird thing, though, is that I don’t think — this trial didn’t quite take into account what we more commonly see, which is, say you gave nivolumab and the patient recurred in 1 site. You resect it. Then what do you do, right? Then you give nivo again. Do you give them ipi/nivo? Do you give them BRAF? There’s no data for any of this. And I think this is where you might actually get 3 different answers in terms of what do you do. I tend to be more cautious in that scenario, because I worry that I’m just as likely to cause toxicity as benefit, because patients have already failed on 1 thing. But I’ve heard other investigators say all kinds of stuff.

I tend to be rather conservative. If somebody has a recurrence, we take it out and we usually just watch that person. But I don't know what you guys do in that scenario.

DR SULLIVAN: Recurrence on...?

DR LUKE: Like, for example, patients got a high risk Stage IIIb, they recur with a lung met on nivo —

DR SULLIVAN: Got it.

DR LUKE: — on adjuvant, we take the lung met out. Then what do you do?

DR SULLIVAN: Yes, I usually would stop.

DR LOVE: Or are you thinking about switching?

DR SULLIVAN: I mean, I think if the data from the study that Mike just talked about were to lead to an approval of ipi/nivo in that setting, then maybe I would intensify therapy. But it’s a really challenging conceptual problem. We know that there are heterogeneous responses to checkpoint inhibitors in patients, that some lesions go away. Some lesions grow. You’re probably having a benefit in a patient who has 1 thing growing, and everything else is responding, and then taking that out, you probably would continue.

If the patient had metastatic disease, everything was responding, 1 thing was growing, you’d probably deal with the thing growing, and you might continue on therapy, because you have proof that you were helping that patient prior to that. In that scenario, where a patient’s on adjuvant therapy and then recurs with metastatic disease, that therapy was a failure for that patient, and you have to imagine the potential benefit that that patient had.

DR SULLIVAN: That’s why I typically would think about stopping.

DR LUKE: I’ve heard some other people say, if you’ve progressed on PD-1, then you’d give ipilimumab after.

DR LOVE: Add it?

DR LUKE: There’s even another level of confusion, right, but I guess my problem with that is that I think the benefit, for the most part, overlaps between PD-1 and CTLA-4. Not entirely, but a lot. But the toxicity does not necessarily. If you then challenge the patient who failed on PD-1 with CTLA-4 in the adjuvant setting, from my perspective, you’re, like, asking for toxicity, and I think there’s an unknown benefit in that scenario. But I’ve seen many cases of patients referred having gone through such a treatment. And again, it’s not right or wrong, per se, but I don’t feel confident that that’s what I would want to do.

DR LOVE: Just to be clear, though, because I think it’s very different when you have somebody who develops a single met while they’re on a therapy than somebody who’s never had therapy.

DR POSTOW: Yes.

DR LOVE: Just to be clear about that. Stage IV NED in somebody who’s never had systemic therapy, treat or not?

DR POSTOW: Those are the patients that this study I mentioned looked at, so those are all de novo Stage IV that had not had prior checkpoint inhibitor. The questions that Drs Sullivan and Luke are bringing up are really relevant to clinical practice, because that’s a different situation. But this showed relapse-free survival advantage to both combo ipi/nivo and nivo over just following those patients without anything.

Until something has an overall survival benefit, though, I don’t think it’s required that you have to give treatment for relapse-free survival only. Because we don’t know if those patients would do just as well if you introduced these drugs if they ever had a recurrence thereafter. And you could buy someone, depending on their biology, what led to their Stage IV NED diagnosis, you could buy a lot of time, like if someone had Stage III melanoma 8 years ago and a single solitary lung metastasis that’s removed, I mean, you may get a decent amount of time.

DR SULLIVAN: But the risk is certainly about the threshold, the way we’d consider therapy for Stage III patients, and even wanting to do trials in Stage II patients. I think the risk is probably closer to the Stage IIIb, 70%, 80% chance that that pace may be higher, that that’s going to come back. I think in a patient who’s never been treated before, oligometastatic disease, resected, we typically would offer therapy for that patient.

Tumor mutation burden and other potential biomarkers of response to targeted therapy or immune checkpoint inhibition in the adjuvant setting

DR LOVE: Actually, I was going to ask about that presentation at ESMO, because one of the things that I’ve been trying to figure out for a long time from you all, I’m still not clear about, I’m hoping you’re going to straighten me out soon, is where biomarkers fit in. You mentioned TMB, PD-L1. Did this study look at any biomarkers in terms of prediction of benefit or anything?

DR POSTOW: Not familiar about the details of this study and biomarker specifically, but I believe that the conclusion is, we don’t have any biomarkers yet to direct either immune therapy or targeted therapy or even which immune therapy combination to do.

They still remain an area of research, as far as I’m concerned. And we always kind of get this information, and it’s interesting, like, if they have a high TMB, I’m kind of secretly hopeful that a patient will respond better to immune therapy, or if you have a lot of CD8 T cells on the tumor, you may feel kind of good that the patient is more likely to benefit. And all that’s true, but these are kind of shades of gray, and nothing is so black and white and nothing, at least to my impression of the data, is yet ready to say, “Because you have this marker, we’re going to do drug X,” or “Because you have that marker, we’re going to do drug Y.” Only something like a BRAF mutation is that kind of high level.

DR LUKE: I would follow in on this to also differentiate biomarkers as we think about them relative to metastatic therapy versus adjuvant, and those are not the same thing. It’s very clear now that PD-1 expression, mutational burden, interferon signature, these are all biomarkers of treatment response in the metastatic setting. How we use them is a different issue. But that’s actually not what we’re worried about necessarily in the adjuvant setting, because there what we want to know is, are they going to recur? And those two things are not necessarily the same, and the field has not yet come around to this to sort this out.

There are tests that are available in clinical practice not evaluated by the FDA that can purport to say whether or not a patient is going to recur. There’s a few of them out there now.

DR LOVE: Measuring what?

DR LUKE: Generally speaking, they measure gene expression profiling and retrospectively, hundreds to thousands of patients have said, “If you have this sort of a signature, then you’re very likely to recur and pass away” based on samples that they’ve historically analyzed. These tests remain somewhat controversial in the academic community. They’re starting to get pretty widespread use, actually, in the community. It’s very common, actually — I’ll get a referral about a patient who had Stage I melanoma who has this test and it says that they have high risk.

DR LOVE: And it kind of reminds of the Oncotype in breast cancer.

DR LUKE: The concept is the same. The difference was that the Oncotype was validated in a prospective randomized clinical trial. We don’t have that data in melanoma yet.

There are murmurings amongst the community that we need to launch such efforts, and there are discussions at all the meetings about this. And probably some day we will know, but right now it’s important to just really highlight the fact that we don’t know the answer. We have biomarkers for metastatic disease. They’re not the same thing as the biomarkers for adjuvant. And we don’t know how these tests that are available overlay with whether or not the patient is going to respond to a treatment. Even if they are high risk, you would think, oh, if they’re high risk, then we better treat them, right? But we don’t know if it means if you’re high risk that you should get PD-1 or BRAF or anything like that. We don’t know any of those things.

DR LOVE: It’s not just Oncotype. There are a bunch of breast markers, MammaPrint^®, et cetera, but there they kind of focus more on the receptor expression and proliferation. Which kind of makes sense, intuitively. What are they looking at here?

DR LUKE: It’s similar kinds of things.

DR LOVE: Proliferation?

DR LUKE: Proliferation genes that are associated with high proliferation are in there. Immune genes are in there.

DR LOVE: Huh. Wow.

DR LUKE: There’s a handful of things. And a couple of companies have done it different ways. Some of them have taken unbiased discovery approaches just to say, what gene expression patterns associate with recurrence? Another one prespecifies you’re alluding to say, like, what elements would we be looking for? The point is that we just don’t know how to use these things yet, but hopefully over time, we’ll figure it out.

Optimal duration of adjuvant therapy with dabrafenib/trametinib and immune checkpoint inhibitor therapy

DR LOVE: I want to move on in a second, but just want to follow up with this 32-year-old lady. Where is she right now in her treatment, and how’s she doing?

DR POSTOW: She’s on adjuvant dabrafenib and trametinib. She’s actually tolerated it well with no fevers or chills. I’m keeping my fingers crossed, because I think that’s one of the main problems that people do have. About 6 or 7 months into treatment, kind of counting down the days until that magical 1 year of adjuvant treatment is finished, which we could talk a long time about, how long do we think we need to treat people with adjuvant treatment? But the studies say 1 year and then stop. And I tell patients I’ll treat them for up to 1 year. If they have bad side effects, again, because there’s no overall survival benefit, I’ll stop early. But we’re going to a year, and then we’ll stop and then we’ll go into monitor after that. She’s doing well.

DR LOVE: Do you have any sense yourself of what you think an optimal therapy theoretically would be?

DR POSTOW: Duration?

DR LOVE: Yes, duration.

DR POSTOW: It’s an interesting question, because I do wonder with dabrafenib and trametinib if longer courses would actually be better. If you look at the Kaplan-Meier curves for relapse-free survival, very few patients will relapse during that year on adjuvant dabrafenib and trametinib. And then after that year is finished, there is some relapse events that start to happen over time. Still better than the placebo arm, but it really kind of seems like the benefit is greatest on that year when you’re on the treatment, actually.

Like the example in resected GIST where patients took adjuvant imatinib and it was shown that longer courses of adjuvant imatinib were actually better, I would wonder if there were trials in the future that tested the 1 year of adjuvant BRAF/MEK versus longer than 1 year of adjuvant BRAF/MEK, if the longer course would actually be better. I don’t have any data to say that yet, but it’s an interesting thought, and I do wonder if that would actually be better for patients to be on longer than a year.

DR LOVE: Are there trials looking at that?

DR POSTOW: I’m actually not aware of any.

DR LOVE: Those are tough.

DR SULLIVAN: We’ve had discussions about trying to do trials to shorten or lengthen, and I think again, it’s the cooperative group type of study that would be done. A company might not be so disappointed with doubling the amount of time somebody is on therapy, but they certainly don’t want to have it.

DR LUKE: The other thing I’ll just add on that, actually, to flip that over and think about immunotherapy along the same —

DR LOVE: That’s what I was going to ask, actually.

DR LUKE: Yes. It’s actually the reverse.

DR LOVE: Right. Sure.

DR LUKE: We now have neoadjuvant data, presurgical data, suggesting that as few as 1 dose might be enough to generate the kind of immunotherapy effect that we’re looking for. We don’t know the answer. But I think this goes back to how long do you treat people in the adjuvant setting? If my patients get sick in the adjuvant setting, we’re done. We’re not, like, going back to it. In 3 months they get colitis or whatever it is, 6 weeks, we’re not then going back to it at 6 months, because we have clear data now to say that even just a few doses is enough. And so again, given all the potential long-term problems and when we get that immunological effect, especially in the adjuvant setting, that’s enough.

If you have metastatic disease that’s continuing to progression, then obviously it’s a different ballgame. But in the adjuvant setting, we want to maximize the benefit and limit the toxicity.

DR SULLIVAN: In that scenario I’ve had a lot of patients say, “I know, but I want to go back on therapy. I haven’t had enough therapy.” And I think a technique I’ve used for patients who are on adjuvant PD-1 inhibitor therapy is to say, “You’ve had toxicity” — that exact same logic you just used — that it might be enough. There’s definitely data in the metastatic setting you’d get a few doses and the toxicity happens. The patients get long-term benefit. And then furthermore, if we keep treating you and we drive you into the ground with toxicity, if your disease does come back, it’s going to be really, really hard to treat. And so that sometimes can sway a patient. I mean, it doesn’t have to — we’re not going to offer it, but just to have a patient kind of be harmonized with our recommendations sometimes, that level of logic can work.

Ongoing investigation of neoadjuvant therapy for locally advanced melanoma

DR LOVE: What do we know about neoadjuvant therapy in melanoma? And are there situations outside a clinical trial right now where you use it?

DR POSTOW: There have been a number of recent prospective trials in the neoadjuvant space, both with single-agent immune checkpoint inhibitors, namely PD-1, although ipilimumab monotherapy also has been tested historically. But the most promising data I think are for neoadjuvant ipi/nivo combinations at different doses and schedules and also for neoadjuvant dabrafenib and trametinib.

There are really great reasons to treat someone in the neoadjuvant setting. And it is, I would still say, kind of mostly a realm for clinical trials, not routine standard of care practice to always give neoadjuvant treatment, despite the promising data that are coming forward. But if you really worry about the micrometastatic disease possibility in patients and the patient has a palpable lymph node, for example, I would consider that patient for neoadjuvant treatment with either checkpoint inhibitors or dabrafenib and trametinib. That’s the data that we have at the moment. But it’s an area that’s changing dramatically in the melanoma field.

And I think what we’re starting to see is very high rates of pathologic complete response, meaning you can kill these tumors completely with some of these treatments. And the question that we’re still trying to answer is, does that really matter for long-term outcomes? Because you could remove the tumor surgically and also get rid of all of the melanoma. What is treating these patients with macroscopic palpable tumors doing for the efficacy against the micrometastatic disease that’s really the bigger threat to life? That’s kind of what the field is trying to grapple with a little bit. And we’re seeing some really promising results so far. But it’s still realm for clinical trials mostly, I would say.

DR LUKE: And I would definitely fall in on that. These are tough questions, and the upside on some of these questions is really high, and we’re really excited about some of them, but the complexity here is quite high. I think it’s generally the case that we would not advocate using a neoadjuvant PD-1 monotherapy. That data actually doesn’t look that good. And so then would you use the ipi/nivo combo? Would you use BRAF? The answer should be here that it’s only the rarer circumstance where you would want to do this without a clinical trial, because we don’t understand a number of these issues.

And so referring patients for trials, we’re learning a lot. We’re getting to a place. There actually is now a cooperative group study of neoadjuvant therapy. Some of us, we argue with each other about how it was designed and whether or not it’s the perfect design, but we’re moving towards this space, but this is not something we want to advocate people doing in clinical practice.

DR SULLIVAN: I think I can only imagine 1 scenario where standard of care neoadjuvant study, and that’s a patient who really is very borderline resectable, has a BRAF mutation and you’re able to get substantial reduction of disease burden prior to doing some sort of surgery. I think that’s a scenario where we probably all have utilized neoadjuvant therapy as a standard of care, but otherwise, yes, I think it’s a clinical trial question.

DR LOVE: I’m kind of curious. In some of the neoadjuvant trials — actually, I was thinking about a lung study that was done at your place — where they saw what they called major path responses, but there was still stuff there on imaging. And they go in there, and the cells were all dead. And it was interesting also in terms of just imaging in general with immunotherapy. Anything that’s been learned with melanoma in that regard?

DR POSTOW: It’s very interesting that when you take the tumors out, usually they’re more killed than you would actually anticipate from the scan themselves. Sometimes you’ll see the tumor shrink on the scan, but they’re not gone completely. But then when they’re removed surgically in these neoadjuvant trials, there’s no active melanoma.

I think that the rates of pathologic response are exceeding the rates of radiographic response in general in most of the trials, so that’s 1 thing. Just because there’s something there on a CAT scan doesn’t mean it’s still alive. And we need to develop new imaging modalities and ways to understand how these tumors are biologically after treatment. Because just looking at the size of them on a CT scan is insufficient. I think that’s where the neoadjuvant trials are really helping us understand what’s going on in the tumor.

The other point I’ll make on that, though, is, a lot of times if you look at the pathology reports, it’ll say “tumoral necrosis,” and it will say “complete pathologic response.” That is true, but we have to keep in mind there’s a lot of necrosis. It’s part of these tumors anyway, so it’s hard to attribute all of the necrosis of the tumor exclusively to the treatment. Because we have really no idea at the outset how much of these tumors were at least partially necrotic.

Now, you can kind of infer that from CT scan results, but it is just a little bit of a word of caution to make sure we understand that just because the tumors are dead doesn’t mean that all of that happened due to the treatment. But I think presumably we believe that the treatment certainly has a positive effect on those cases.

DR LUKE: Yes, in terms of immunotherapy response, this is an area I think actually which is going to be one of the next big movements in the field, which is the idea of integrating, say, PET scans that are targeting PD-L1 or CD8 to try to look at the immune response in the tumor. And these sorts of neoadjuvant studies are almost the perfect venue to investigate this kind of technology, but again, goes back to the fact that these are investigational studies. This is not a standard of care thing to be doing.

And over time, could we learn that there’s some kind of really tight correlation between one of these scans and path CR? And yes, we could do that. We’re trying to do those clinical trials now, but we can’t take it on faith that we even know what’s going on in the tumor at this point, as witnessed by the fact that the tumor’s either are more dead or not more dead than what we thought based on the scans. For now, this is still an important research question.

Effect of BRAF/MEK inhibitors on the tumor microenvironment

DR LOVE: I’m also curious about predictors of response. And again, when I think about something like dabrafenib or trametinib in the adjuvant setting, it’s a little bit like endocrine therapy in breast cancer. And there, they’ve seen when you give neoadjuvant therapy, you see a decrease in proliferation immediately. Do you see that when you give BRAF/MEK?

DR SULLIVAN: You see a lot of changes in the neoadjuvant setting in those trials and then also when we’ve looked at serial biopsies in patients who had metastatic disease. Interestingly, one of the key things that you see are changes in the tumor immune microenvironment with BRAF targeted therapy.

DR LOVE: Really?

DR SULLIVAN: Yes. And you can see that in neoadjuvant, and you can see that in the metastatic setting. And the only real data that’s been biomarker-y in the setting of the COMBI-AD study, which was dabrafenib/trametinib versus double placebo in patients with Stage III melanoma, really did show that patients who had the best outcomes had the most inflamed tumor microenvironments before starting therapy. It was actually the immune score, so to speak.

DR LOVE: Wow. Hmmm.

DR SULLIVAN: It wasn’t like the immuno score, which is something that’s slightly different. But the amount of inflammatory cells that were there seemed to correlate best with outcome. And in neoadjuvant studies, we saw some patients who had some immune microenvironmental inflammation went up, and in some patients it went down, but mostly you tend to see inflammation in tumors. And that’s probably the most reliable biomarker. I’m not sure it’s on target, but it’s definitely associated within adding MAP kinase where you do see that. And I’m not sure that that’ll ultimately be associated with benefit or not if you’re looking in the neoadjuvant setting. It’s only a small number of patients that have been published so far, but it probably is a pretty good sign that you’re doing something right if you’re getting the immunologic benefits of targeted therapy for —

DR LUKE: Let me just make that point too, is that the few patients who we do take off of targeted therapy in the metastatic setting are, generally speaking, those who have had really severe side effects where we just had to stop, and in my patients, those are almost all people who had pseudoimmune therapy-like toxicities while taking targeted therapy. They’ve got iritis. They got some kind of inflammatory process, and then it turns out they don’t need to go back on the targeted therapy, and it’s almost all certainly because of this. You basically cause an immune response in the context of the targeted therapy.

Challenges with identifying immune-related adverse events

DR LOVE: I’ve got a great slide, it’s my favorite slide now, it just came out, a trial of a checkpoint inhibitor. It was actually atezolizumab in small cell with chemotherapy, but it was placebo controlled, okay? The incidence of immune-related toxicity in the placebo arm was 25%. Have you seen that with melanoma? I was like, wow! That really tells you the story, man.

DR SULLIVAN: In the adjuvant studies, the placebo rate is not that high. It’s maybe a couple of percentage points.

DR LOVE: Really? That’s interesting.

DR POSTOW: But you find — I mean, I remember there were patients unblinded, ipilimumab/placebo trials in the past. Everyone has diarrhea.

DR LOVE: Exactly.

DR POSTOW: From time to time over the course of a year, right?

DR LOVE: Right.

DR POSTOW: And so if you’re on an IV placebo and you have diarrhea, it’s colitis. And you report it that way.

It is just to show you how hard it really is to tell that you’ve activated an immune system and how hard it is to identify and say this is an immune therapy toxicity from all these other issues. I think in the adjuvant setting it’s a little easier, because you don’t have the metastatic disease process itself that could be causing symptoms of some sort. But it’s very, very difficult.

DR LOVE: That is interesting, though, in terms of adjuvant melanoma, but again, lung cancer, metastatic, they’ve got problems from their disease. They’ve got COPD. They get short of breath. Every time they’re short of breath, they’re going to think about that. I just thought from the point of view of the clinicians, the nurses out there trying to figure out what’s going on.

DR SULLIVAN: And I would say, obviously it has nothing to do with melanoma, but caring for patients who get hospitalized with maybe pneumonitis in that lung cancer patient who has had radiation and surgery and smoking damage, and immune checkpoint inhibition is one of the most challenging clinical scenarios that I think oncologists face in this day and age. And it’s not an uncommon one, because there’s a lot of patients with lung cancer getting immune checkpoint inhibitors.

Management of immune-related adverse events in patients receiving immune checkpoint inhibitor therapy

DR LOVE: I don’t want to get too far off the topic, but you all have the longest history with these drugs. And I am kind of curious what advice you would give to general medical oncology practices.

Any pearls or recommendations about how a general medical oncology office can get tuned up for immunotherapy in general, not just melanoma? Any thoughts about that, Jason?

DR LUKE: Yes, I think really the key is broad differential diagnosis, but that’s a little too vague probably for community practice. Maybe something along the lines of the checklists. And so if you have some sense of what it is you’re looking for, and if you’re having people screened through, what are kind of the big questions you go through?

For example, in the diarrhea one, in somebody you want to ask them, “Have you been in the hospital in the last few weeks?” You’re going to make sure it’s no C diff, right, and get that test, right? There’s obviously the outlier things about travel, et cetera. But even simple stuff like that, just those history elements to take to make sure we know what’s going on.

I would just note that in my practice, the number 1 thing I have my folks do is an ambulatory O2 sat. Anybody who shows up short of breath, we check their oxygen and make them walk around a little while, because that is one of the real strong differentiators for pneumonitis versus everything else that could be causing shortness of breath. Because if the number gets dropped low, something bad is happening, and we need to figure it out fast.

DR LOVE: Julie Brahmer was the author on the 80-page paper in JCO, but one of the things that came out there that I thought was really interesting in that paper, and again, with your perspective I’m kind of curious, was the grading of toxicity and the kind of uniform — again, you all are used to this, but general medical oncology, et cetera, it’s maybe not quite the same. Any thoughts about the more global approach that was outlined in that paper?

DR POSTOW: I think that grading is important in general, not so much that you know the difference between Grade II and III, but the grading is this kind of quantifiable way to say, how bad is it? And you can break it down. You can get the app on your phone, the CTCA app, and put in how many bowel movements a day over baseline someone’s having, and it’ll give you a grade. And then you can follow these different algorithms, whether it’s the JCO paper, you can go to the NCCN guidelines, and they have an immunotherapy toxicity algorithm.

I think the grading is important if you want to follow the algorithms in terms of where to put your patient and what to do. And you can just use the app if you really want to get quantifiable. But there is this kind of qualitative nature to this too, that kind of it’s just, how bad is it? When it’s really bad, then you should do more with immunosuppression. If it’s not that bad, maybe you can get away with monitoring.

It’s a good idea to try to quantify it if you can, especially if you’re looking at the algorithms, but I don’t know that you absolutely have to. And I wouldn’t totally bog down too much in the numbers, because it can make the overall clinical gestalt go away when you get too pinned down in the numbers. Certain grades are not completely reflective of how bad certain things can be, for example.

Neurologic toxicities associated with the use of immune checkpoint inhibitors

DR POSTOW: I guess I could give an example. Neurologic toxicity, a lot of people in trials, you blow off the Grade 1 and 2 toxicity as just kind of minimal toxicities, and you only talk about Grade 3 and 4 as being really bad. But neurologic toxicity Grade 2 is neurologic toxicity that interferes with instrumental activities of daily living. I think if someone is having neuropathy that’s affecting their life in instrumental activities of daily living, like shopping and grooming themselves and taking care of things around the house, I think that’s a major issue, actually.

DR LOVE: Are you talking about peripheral neuropathy from checkpoint inhibitors?

DR POSTOW: Mm-hmm. Mm-hmm. Yep.

DR LOVE: Can you talk more about that?

DR POSTOW: Absolutely can happen.

DR LOVE: The only thing I think about is Guillain-Barré or something. I don’t know. What —

DR POSTOW: There can be all kinds of neurologic toxicities from checkpoint inhibitors, ranging from Guillain-Barré syndrome to myasthenia gravis, so people with double vision that comes from diplopia that arises from antibodies produced that cause myasthenia gravis. That can happen. Peripheral neuropathy, demyelinating disorders and such, that can happen where people have numbness and tingling in the hands and the feet.

DR LOVE: But a demyelinating peripheral neuropathy?

DR POSTOW: Sometimes — we don’t know exactly the mechanism of why those sensory changes or motor changes are happening —

DR LOVE: But, like, regular peripheral neuropathy

DR POSTOW: — but there have been some biopsies, and it can be very disabling, especially —

DR LOVE: Really?

DR POSTOW: — in someone — I mean, everyone uses their hands for work, whether you’re typing, a desk job.

DR LOVE: And it’s immune based?

DR LUKE: Presumably. In our patients, we’ve more commonly gotten able to evaluate it on a clean level, which is to say that many of our patients are a little younger. They don’t have a lot of comorbid illness. Go back to your case of the patient getting lung cancer. Now they’re getting chemo plus PD-1. Now becomes super complicated, because maybe they also have diabetes, and there’s all kinds of other stuff. Absolutely you have to be aware of this.

And I wanted to follow up about the grading just real quick, which is to say, understanding on some flavor the grading is important because it can also help you ask the right questions. Sometimes that’s the other piece of it. Because the grading does give cutoffs, and if your staff who are taking the phone calls have some general sense of what the cutoffs ought to look like, so how severe is it, that can be actually very helpful just to being able to raise the red flag early instead of letting things go on too long.

Multidisciplinary team approach in the management of immune-related adverse events

DR SULLIVAN: And I would have one final thing for community practitioners of what you can do to be more comfortable with taking care of these patients. Find a gastroenterologist that you trust that can get patients in quickly. We found in our practice that if we get a scope in the first day or two, we can identify which patients have mucosal inflammation, which patients don’t. Of those who don’t have mucosal inflammation, you can probably wait until the biopsy. If the biopsy doesn’t show inflammation, they probably have another cause of their diarrhea. Diarrhea is a pretty common problem. If they have lots of inflammation, you know that’s a patient you want to give either oral or sometimes IV steroids to.

If a patient has no mucosal inflammation and has evidence of the inflammation on the biopsy, those patients respond to budesonide. And how did I find this out? Because I have a great gastroenterologist who knows all about inflammatory bowel disease and says, “You know what we give for microscopic colitis? Budesonide. Let’s see if it works.” It turns out, it does.

Find yourself a good rheumatologist that you can trust and get your patients into being seen. The arthralgias and arthritis are one of those true chronic conditions that we treat through but definitely impact the quality of life of our patients.

Find a neurologist who knows something about Guillain-Barré and myasthenia, because, though you're not going to see it very often, when you do, you’re going to wish you knew a neurologist who knows how to treat that condition.

I think the bottom line is that the team that your patients have access to is certainly your nurses and your advanced practitioners and your colleagues, but it also should be that extended team of medical specialists that have had training in how to deal with information. The nuances of how to treat it may be slightly different, but they ought to at least be able to read the literature from the perspective of a gastroenterologist or a neurologist, rheumatologist, et cetera, and be able to apply what their practice has been for these inflammatory conditions and then learning and adopting the data that’s coming in.

DR LUKE: One thing that I see a lot, especially coming from community practice, is maybe having not as much familiarity with this or previously not having had it, it’s not aggressively enough treating these conditions. Which is to say that when a patient develops one of these things, we don’t half seize it. This isn’t COPD, where you give steroids for a week, right? You need to treat it as a primary problem and then return to treatment thereafter.

And so I actually, unfortunately, I think commonly have seen referrals of patients who had colitis and got 40 mg of prednisone and their colitis never really went away for, like, 3 months. And just think how terrible that would be — you get diarrhea, like, 4 or 5 bowel movements, liquidy diarrhea for 3 months. And you just realize that what they should have done is given 1 mg/kg of prednisone for, like, 4 days and then tapered, right? But they never quite got to that threshold to really get rid of this problem.

And so if you’re going to treat it, treat it. And don’t worry that you’re going to block the therapeutic effect. There’s now lots of data to suggest that that is not an issue. Treat the patient. Make the patient better. And then re-engage about what do we think the most appropriate therapy needs to be as the next step for that person?

Educating patients and emergency room personnel about immunotherapy and the potential for immune-related adverse events

DR POSTOW: The other thing I just want to add to, it’s a very — I agree with both Drs Luke and Sullivan on all of this — and I think the one other piece, just to mention very simply, is, make sure your patient understands that they’re on immune therapy. And I think that’s especially relevant in this era where immune therapy is often combined with chemotherapy for triple-negative breast cancer, lung cancer.

DR LOVE: Everything now.

DR POSTOW: For PD-L1 not as high patients. And a lot of times, chemo — when patients are on treatment, are they going to the chemo suite, which we sometimes, I think, mislead patients — go get your treatment in the chemo suite. That’s already starting to sound like chemo, and that’s just our typical language sometimes. But for patients, it’s really important to know that you’re on immune therapy, so that if you go to an emergency room or an urgent care center or somewhere that’s away from the oncology center, which is usually what happens when you’re being cared for in different settings, that you can tell the doctors, “I’m on immune therapy. Yes, I may also be getting chemotherapy, but I’m on immune therapy.”

If you can have a little card that says what drug you’re on, then the providers who are emergency department physicians that might be dealing with all kinds of different things and may have an aortic dissection next door, an MI and a car accident in the other bay, they’re going to think cancer patient, treatment: chemo. And you have to really make sure that there’s some nuance. I think the patient can be a real advocate for that.

DR LOVE: I was flashing, when you said find a gastroenterologist — we just had this case of this oncologist in a little town in Indiana. He’s the only oncologist in the hospital, and he had this patient who had diarrhea. And the GI guy refused to scope the patient.

DR POSTOW: I realize that the volume of these problems presenting to most emergency departments across the country is still not high, but it’s getting higher, and it’s important that —

DR LOVE: For sure.

DR POSTOW: I think that the patients can play a role in the education, especially do — very infrequently do practices have an ability to communicate directly with these emergency departments where patients are being seen to say, “My patient on immune therapy is coming. Look out for immune-related neuropathy.”

I think that the patient, even if they have a little card that can have the name of the drug, what the drug is doing and a reference to some of the published work, like Dr Brahmer’s paper, I think could be helpful.

DR LOVE: That’s a great idea. Yes, I was just flashing on the fact too that every year at ASCO we do these satellites in this big room in the Hilton Hotel, and every Friday night for the last 12 years we’ve done lung cancer. This year I showed the syllabus of our meeting from 5 years ago, 2014, and the word “checkpoint inhibitor” was not on the syllabus. Five years ago. Now every patient with lung cancer gets a checkpoint inhibitor. Probably there are a lot of people coming into emergency departments.

Response to immune checkpoint inhibition in noninvasive solid tumors

DR LOVE: Another question I just want to ask. I’m curious, Jason, just biologically, we saw some data at ESMO using a checkpoint inhibitor in people with noninvasive bladder cancer, the BCG failures. And I don't know, like, if that really is playing out or not, but it seemed like it was effective. What would you expect in terms of noninvasive — solid tumors in general, how they would respond to a checkpoint inhibitor?

DR LUKE: Noninvasive bladder cancer is a little bit of an idiosyncratic clinical entity, but to your point, I’ll get to in a second, those data I think for pembrolizumab were the first data that came forward and it suggested a benefit. Like 40% of the patients in the BCG-refractory setting could derive a benefit.

DR LOVE: Right.

DR LUKE: The extent to which that’s going to become a real clinical utility I think is not totally clear. Most of those patients will go on to develop muscle invasive disease, but is a strong degree of heterogeneity there in that population? And so I don’t know whether or not that’s going to get a big uptake. You could imagine insurance companies not being excited about that either, because I think it starts to be a big number of patients treated.

I think one of your questions, just to back that up though, and make it a little bit broader, would be maybe unresectable neoplasms. In not melanoma, but in skin cancer we deal with cutaneous squamous cell carcinomas. And, actually, there’s already an approval for anti-PD-1 in that scenario when they’re unresectable. And then the question becomes, what about adjuvant therapy?

And so I think the big picture on this is that there is going to be an expanding universe around the use of immunotherapy and a lot of these other clinical entities. And I think we’re going to have to come to it to get the randomized data and then figure out, how does that sit? Because some of these same questions that we talk about in adjuvant for toxicity are going to become very relevant as well. What are you willing to expose the patient to before you do what we would consider our standard therapy?

DR LOVE: Yes. That same doc in Indiana had a patient with squamous cell that he gave — what’s it called, cemip…?

DR LUKE: Cemiplimab.

DR LOVE: Yes. Imagine, here you are in the middle of Indiana getting a response to that therapy.

Incidence and management of nonmelanoma skin cancer

DR LOVE: It also brings up a question I was going to ask you in general — I know we’re talking about melanoma here today — but who’s treating squamous cell? How much is out there? What about basal cell? Merkel cell? Who treats these?

DR SULLIVAN: Medical oncologists theoretically treat these, but we actually have to see them first. And it’s a disease that has always been the purview of the surgeon, because they tend not to spread to — occasionally squamous cell spreads to regional lymph nodes. They rarely spread outside of the regional lymph nodes.

DR LOVE: But, I mean, there are deaths from squamous cell of the skin, right?

DR SULLIVAN: There are definitely deaths. And we actually recently hired a junior investigator, and he’s also trained in dermatology. He’s like, “I want to do nonmelanoma skin cancer.” I was like, “Alright, great, we don’t have that many patients, but see if you can start a multidisciplinary clinic and maybe you’ll see some patients,” and within 3 or 4 months, his clinics are packed with patients who have really nasty squams, really nasty Merkel cells. And so I think this may be —

DR LOVE: What about basal cell?

DR SULLIVAN: — basals as well. But the squams I think tend to be more regionally involved, and they’re also, the basal cells are less responsive to immune checkpoint inhibition. And I think a lot of dermatologists are seemingly comfortable giving the —

DR LOVE: Hedgehog.

DR SULLIVAN: — Smoothen inhibitor, the hedgehog inhibitor.

DR LOVE: I don’t think they like them too much.

DR SULLIVAN: Yes. Nobody likes them. But I guess my point is, if you build it, people come. And these patients are out there.

DR LOVE: Wow. Interesting.

DR SULLIVAN: And it’s really been eye opening to me. I never — I mean, I figured he would have some patients to see, but I didn’t know he was going to build an actual program within 6 or 8 months.

DR LUKE: I think part of this, too, is that there was a bias in that. A lot of folks, say, with Merkel cell, the older folks, and if they were to progress, it’s kind of like, why would you do this to an old person? That kind of thing. But now with how effective some of these therapies are, so for Merkel and for cutaneous squam, the response rates are upwards of 50% to 60%. And generally speaking, again, we’ve been speaking of the ills of immunotherapy here, but generally speaking in the context of other cancer therapies, these are highly effective and generally safe therapies. And so it’s just to that point.

I think there’s a lot of patients that they weren’t coming in because there was nothing to do for them. But now that we do have active therapies, we can really improve quality of life. And we’re seeing a lot of these kinds of patients now.

DR LOVE: To what extent are patients with immune issues like transplants? Is that a big source of squamous, or…?

DR SULLIVAN: It is, and squamous cancers or just nonmelanoma skins cancers in general is one of the leading causes of mortality in patients who have a solid organ transplant.

It is a problem. Now, that also leads to the bigger question of, what the heck are you going to do for somebody? Are you going to give that patient a checkpoint inhibitor or not? And there is certainly a building literature on using immune checkpoint inhibitors in the transplant setting. Probably most of the transplant is actually in, say, Hodgkin disease, where patients have allogenic stem cell transplants and then they have refractory disease, and Hodgkin’s lymphoma responds relatively well to checkpoint inhibitors.

But it’s about a third of patients lose their graft and — or not lose their graft. Their graft is perfectly functioning. The patients die of graft-versus-host-disease, and I’ve had a patient like that, unfortunately. I think like anything, you say, “Alright, there’s a possibility that the worst thing that can happen, and in that setting, the possibility is probably higher than the possibility of having a severe side effect if you don’t have one of those scenarios.”

Perspective on the utility of immune checkpoint inhibitors for patients with melanoma who have undergone transplants

DR LOVE: Getting back to melanoma, though, I am kind of curious — I was going to bring this up, and we can get back into the issue, first of all, of patients who have had transplants, for example, renal transplants. Do you give adjuvant checkpoint inhibitors?

DR LUKE: Adjuvant no.

DR LOVE: Metastatic yes.

DR LUKE: Metastatic becomes a harder question, because it’s really — so what are the alternatives here? If the person is going to pass away from metastatic melanoma, then if you have a discussion around —

DR LOVE: What’s the likelihood that you lose your kidney transplant?

DR LUKE: I don’t know that it’s entirely clear, but it’s definitely on the order of 20% to 30%, somewhere around there. And so that’s a real issue. And that’s where you really need a multidisciplinary team to attack this kind of a problem. Because I think there’s probably a sweet spot for many of these patients that when you’re doing by yourself you may or may not be sensitive to.

There is a clinical trial now launched by the NCI to try to study this question to try to see about giving these and try to better understand it. And in that trial, as soon as you see the creatinine change by some really tiny amount, like 0.1 or 0.2, the patient immediately goes to renal to get a biopsy. Because they’re trying to see, could we actually titrate this? And we were talking before about how many doses do you need, what side effects do you tolerate? And here is a population where you tolerate no side effects. Because if you do, you’re going to lose the graft.

But again, if you take a step back, you say, the patient was going to pass away from metastatic melanoma. Then are we willing to try it? Probably — and just needs to be done right away.

DR SULLIVAN: And that’s definitely the kidney transplant setting. Obviously the patient is the most important part of the decision-making. But this is a scenario where I’ve had a woman who was in her 80s, and I said, “We could give you this. It’s a very good chance you could lose your graft.” She said, “I’m not going on dialysis. I’d rather die.”

There, people who have been through a certain thing in their life may not be willing to go back to it. And so I think the kidney transplant is one of those where most people have been on dialysis at some point, and many patients are like, “No problem. I’d rather be alive and on dialysis than not alive.” But I have had a few patients that have strongly opposed the concept of immune checkpoint inhibition because of the fear of dialysis.

DR POSTOW: But that’s where surgery for Stage IV melanoma, if possible, comes up and immune therapy really becomes a last resort. If you have a BRAF mutation, these are clearly patients you’d want to treat with BRAF and MEK first.

Risks and benefits of immunotherapy for patients with metastatic melanoma who have received solid-organ transplants

DR LOVE: Just to understand, though, the biology. When you think about the different strategies that are used in different types of transplant, for example, liver transplant, heart transplant, kidney transplant, you mention allotransplant, and then you think about the way, let’s say, PD-1 antibodies work, what would you expect when you put them together?

DR POSTOW: I would imagine it depends on the type of antirejection strategy that you’re using, although most antirejection strategies are T-cell-based inhibitory, which, therefore, would be not good if we’re giving T-cell stimulatory drugs like checkpoint blockade. I’m not familiar enough with the nuances and kind of what immune suppression strategy is used for what type of transplant patients may have, other than most are T-cell inhibitory and therefore would be at risk of organ compromise in the presence of checkpoint inhibitors. I think it’s a risk-benefit type of issue. And one might still even think, despite how terrible being on dialysis may be for some patients, that is a livable condition in contrast to some patients with heart transplant, lung transplant or other issues.

DR LOVE: I mean, they have data on liver transplants. I think they had 8 patients or 10 patients, and, I mean, half of them lost their livers, but half of them didn’t. If they have a lethal condition, maybe you think about it.

DR LUKE: One important clinical point on this is that if you get a patient, even with metastatic disease, the first thing you want to do is decrease the immunosuppression as far as you can. Now, sometimes these patients end up a little ways out from the transplant, and it’s become a little bit ho-hum monitoring, because they otherwise seem to be doing fine, but they might actually be on more than they have to be. And if you can go back with a team and say, “Can we just pull this down a little?” I’ve actually seen patients where you can see actually disease regression just from doing that.

DR LOVE: Wow.

DR LUKE: And especially people who’ve had primaries, like people who show up who have a melanoma that gets resected, you definitely want to work with a team then to decrease their risk of getting more. Because you don’t want the one that’s going to become metastatic then to pop up. But this is probably the most important thing you can do for a patient like that before you even consider whether or not to give them checkpoint blockade.

Monitoring and care of patients with melanoma and preexisting autoimmune disease receiving immune checkpoint inhibitors

DR LOVE: That actually leads into the issue of prior autoimmune problems, which again goes across oncology. But with the transplant thing, I kind of came away with the idea, yes, it’s not unlikely you’re going to lose it, but you have to, like you said, balance it out. But maybe you can comment a little bit about prior autoimmune disease, Crohn’s, et cetera, practically speaking, how you approach that, adjuvant versus metastatic.

DR SULLIVAN: There’s a building literature on this. There are actually trials in the cooperative group setting looking at studying the use of PD-1 inhibitors in certain populations of patients with autoimmune disease in cancer. I think the general principles are ipilimumab tends to be more likely to reactivate GI malignancies. PD-1 inhibitors seem to be more likely to reactivate rheumatologic conditions. And so if you have Crohn’s disease and you get ipilimumab, there’s a 30% to 50% chance you’re going to reactivate your Crohn’s disease. And if you have really bad Crohn’s when you develop your metastatic melanoma, you probably need to be on something, ideally not high-dose steroids, before you start your immune checkpoint inhibitor therapy.

And so I think — is it burnt out RA? That may be safe to treat somebody in that scenario. Is it florid Crohn’s where somebody just had a small bowel resection 2 years ago or 1 year ago or 6 months ago? I think there’s clear nuance here, and there’s quite a lot of variability. But I think the bottom line is, you can safely treat patients, most patients, with checkpoint inhibitors and you just have to monitor them very closely. And it’s really important to have whatever specialist has been caring for that patient, or if they haven’t had a specialist caring for them, find them one to help comanage that patient during their checkpoint inhibition.

DR LOVE: And, of course, this is really a big problem when you run into a situation where the patient has metastatic disease. And a lot of it comes into looking at the individual situation. What about neurologic autoimmune problems, particularly MS?

DR POSTOW: There’s not a lot of data. There are case reports that are published that immune therapy is okay in MS. Some case reports have said that MS was exacerbated by immune therapy. I think it’s hard within each autoimmune disorder specifically to quote what’s the incidence of recurrent toxicity. I can’t really say X percent of the MS patients would have an exacerbation.

It is, again, one of the situations where I’m a little bit nervous. I think neurologic toxicities in general make me nervous, and it depends on the degree of the MS that the patient is suffering from and the requirements for immunosuppression at the outset. Some people will have very minimal amount of neurologic toxicity based on not on any immunosuppression. I would hope that those patients could be safely treated with immune therapy. And I wouldn’t rule anyone out that’s facing metastatic disease for being a candidate for immune therapy regardless of the underlying autoimmune condition. But something that you would just want to monitor for.

DR LUKE: I would highlight, one of my patients, actually, one of the more striking patients I ever treated, which is just this whole debate about risks and benefits. And she was a young woman who — I think she was in her late 30s — developed really terrible de novo metastatic melanoma in the hospital. We just started BRAF inhibitors because she was tanking and going away. It turned out she was BRAF mutant — she actually recovered her performance status over the course of about 6 weeks and went from, like, completely bed bound to now — but yet we know from the data in the metastatic setting for BRAF that the likelihood of her PFS is only on the order of about 6 months.

I had a frank conversation with her, which was, we can continue BRAF, but I can tell you as an expert in melanoma, it's going to stop working. Should we try ipi/nivo immunotherapy? And she had MS. She had active MS.

DR LOVE: Wow.

DR LUKE: And so we gave her the first dose, and she came back 3 weeks later, and I was like, “How are you doing?” And she said, “I think I’m okay.” Which to me was, like, ring the bell, because I was, like, then she’s okay, right?

We gave her the second dose, and after she came back from the second dose, she said, “Now my finger’s a little tingly,” and I was like, “Okay, that’s it. That’s it.” And best I know she is fine, and that was about a year and a half ago.

DR LOVE: Wow.

DR LUKE: Right. But that’s not to say this is going to go this way with every patient, but all those steps about risk and benefit — because I completely agree. I went into this thinking, like, we might kill you with checkpoint blockade, right, but the alternative was metastatic melanoma, which was really awful, right? How do you make that decision? That’s informed consent.

DR SULLIVAN: And I think patients are very willing and able to tell us what they’re interested in. And there are some patients say, “Just tell me what to do,” and that’s really a hard situation in this scenario. But I would say most patients come in with an “I’m willing to accept that risk. I understand.” And the bottom line is, most often the biggest problem in the room, to quote one of our colleagues Mike Atkins, is the cancer, not the underlying other problem.

Case (Dr Sullivan): A man in his late 50s with Stage IIIC, BRAF wild-type melanoma on his arm experiences intermittent tingling of his mouth during treatment with adjuvant nivolumab

DR LOVE: Let’s finish out with one more case. You have a 57-year-old man who was BRAF-negative in the adjuvant setting. And I was curious what happened to him when he got nivolumab.

DR SULLIVAN: Yes. This is 57-year-old gentleman, on his arm had a 2.4-mm thick ulcerated melanoma, a bunch of mitoses. Had a wide local excision, sentinel lymph node exam, 1 out of 2 nodes had 2-mm focus of melanoma. And imaging showed no evidence of disease. And he ended up having no detectable BRAF mutations.

We started nivo. After about 4 doses he developed migratory, mild transient joint pains that were managed with NSAIDs, and we continued to push and go. He didn’t have inflamed joints.

DR LOVE: No history of arthritis?

DR SULLIVAN: He had no history of arthritis. He may have had a little bit of osteoarthritis. He’s an active guy. But had no history of rheumatologic condition.

And then after 11 doses, he presented, and he said, “My mouth feels a little funny.” And I said, “Alright, tell me about that.” And it wasn’t clear if he was describing dry mouth, which is a common scenario — that Sicca syndrome that we can see, which often can include joint issues and dry mouth — or whether what he was feeling was actually numbness in his mouth.

And so I said, “I’m not sure what you’ve got.” I said, “I’m going to hold a dose.” He said, “No, no. I want to get my dose today.” I said, “I’m going to hold your dose.” And then we got some imaging, because I said the worst 2 things this could be is leptomeningeal disease, which sometimes you see kind of at the lip or numbness in weird places, and the second worse thing this can be is an emerging neuropathy of cranial nerves or vestibular bulbar type of inflammatory syndrome.

And so we did imaging. We did skull-based imaging. Did an LP. And he didn’t end up having any evidence of disease, so there was no leptomeningeal disease. There may or may not have been a little bit of enhancement. I sent him to one of our neurologists and I said, “Can you do an EMG? Actually,” I said, “What do you want to do?” “Let’s get an EMG.” I said, “Great. That sounds great.” And she did. And it wasn’t perfect. And his CSF had a little bit of an elevated protein.

DR LOVE: EMG on his peripheral muscle, not up here?

DR SULLIVAN: No. I think he they did it actually —

DR LOVE: Wow!

DR SULLIVAN: Yes.

DR LOVE: Okay.

DR SULLIVAN: Small needles, I guess.

Cranial neuropathy associated with immune checkpoint blockade

DR SULLIVAN: And so he had neuropathy NOS presumably of the cranial nerves, and I stopped and I said, “I don’t want to do that.” And he said, “I want to keep going.” And this one of those scenarios where I said, “I could keep treating you and this could become either a permanent problem or a problem that’s so severe that we’ll never be able to treat you again with immune checkpoint inhibitor therapy if this disease comes back.”

And then I also highlighted the fact that there was a recent publication that characterized the fatal events from immune checkpoint inhibitors, and on the list of most fatal immune checkpoint inhibitor-related disease is neuropathies. And I’m not sure cranial neuropathies are the ones that are fatal, but they certainly don’t make me —

DR LOVE: They don’t sound good at all.

DR SULLIVAN: — feel comfortable about continuing to manage somebody with ongoing.

DR LOVE: How much trouble was he having swallowing?

DR SULLIVAN: It was partial trouble swallowing, like, he never aspirated. He had trouble swallowing. It didn’t feel normal the way that he was swallowing. And his mouth was numb. And he definitely had just tingling sensations. It was clearly, as we teased it out, just clinically talking to him, it seemed more neurologic than it did dry mouth. But again, this concept of the Sicca syndrome and dry mouth can feel really weird. And sure, you could feel tingling and you could have trouble swallowing. It wasn’t entirely clear what it was other than it was a syndrome of symptoms that I did not want to pursue ongoing PD-1 inhibitor therapy.

DR LUKE: Sullivan sign.

DR SULLIVAN: Sullivan sign. When I want to stop your therapy —

DR LOVE: What I’m trying to find out is, what happened when you stopped it?

DR SULLIVAN: It got better. We actually put him on gabapentin, because gabapentin’s apparently good for every condition, but at the recommendation of the neurologist, and the symptoms did ease up. And then we ultimately stopped the gabapentin, and his symptoms are mostly gone. There may be occasional times where he feels a little numbness or tingling in his mouth, but he’s recovered by all accounts, and, fortunately, also, he’s had no evidence of recurrent disease.

Counseling patients about the benefits and risks of immune checkpoint inhibitors in the adjuvant setting

DR LUKE: This reminds me of all of our conversations about the risk and the benefit. And we just talked a little bit before about how we don’t know how long people really need immunotherapy in the adjuvant setting. This is the thing that I always say to my patients is, we’ll never know if immunotherapy worked. We’ll only know if it didn’t work. Right? And so you think about that in the context of somebody who’s having active symptoms, that puts me in a position of saying, “If you’ve already got the treatment, I’m not going to keep pushing the envelope in the adjuvant setting, because I don’t know how many doses you needed in the first place.” And so we would hate to do long-term damage when we really don’t know whether or not we need to keep giving the drugs.

DR POSTOW: And telling a Stage III patient that’s been resected, the first thing is, you may be cured right now. And the best thing I could ever do for you is leave you alone. That’s one kind of issue. And then we’re trying to talk about reducing risk of recurrence.

I think we’re grappling in an adjuvant era where we’re really remiss for overall survival data, which we need to really know, does treatment now as preventative treatment to prevent the possibility of Stage IV? Does that really make a difference long term with outcomes versus just treating the patients who have Stage IV disease? If those patients are the same ones that would benefit from adjuvant treatment, Stage III, then I would say we have to be really careful about avoiding overtreatment in Stage III.

DR LOVE: I’m curious whether either of you have seen anything like what he just described and what you think’s going on.

DR POSTOW: We’ve had a lot of strange neurologic toxicities. The cranial neuropathies, I’m very glad this patient did not have leptomeningeal disease, because that, unfortunately, is usually what happens. We have a lot of patients with dry mouth, and I’ve frankly not really considered the possibility of a neurologic dry mouth, because often patients will have physical exam, findings of dry mouth. And I’ve actually had people with dental problems because they don’t have salivary production. One little plug to make sure you go see the dentist, which is something that we often don’t make a big deal about, but especially in these patients with oral toxicity.

But I’ve seen a lot of these other strange neurologic toxicities. And it’s another reason that the threshold to stop is clearly passed. That’s when I blow the whistle and say, “We’re done.” But certainly you have to be really mindful of this neurologic stuff, because there’s so many different symptoms someone can have.

DR LUKE: And I haven’t seen this exactly, but I have definitely seen mucositis. Mucositis happens, and we see that.

DR LOVE: Mucositis.

DR LUKE: Again, it’s an inflammatory process in the glands in the throat. It can be down, further back in their mouth. It can be painful. It almost looks like thrush sometimes.

DR LOVE: Huh.

DR LUKE: And so we can see things like that. That’s different than these neurological toxicities.

I have seen a patient who I think had immune-related Bell’s palsy. I mean, it’s basically what it looked like, because it looked like the postviral infection, but it was induced by a checkpoint inhibitor. In thinking about cranial nerves and everything, I mean, you definitely can see — kind of anything you’ve read about it can happen in the context of immune checkpoint blockade.

DR LOVE: Is Bell’s palsy immune related?

DR LUKE: It’s residual inflammatory due to attacking the virus, right? The immune system attacks the cranial nerve 7 distribution. It’s the overattacking of the immune system against the virus. The virus is cleared, and then you get residual toxicity related to that.

DR LOVE: When people recover from Bell’s, can it be reinduced with a checkpoint inhibitor?

DR LUKE: I haven’t had that many cases to know that question.

DR LOVE: Sure, not that many, but…

DR LUKE: The patient I had didn’t really get better, unfortunately.

DR SULLIVAN: I have a patient who — she had bilateral cranial nerve palsy.

DR LOVE: Wow.

DR SULLIVAN: And one came back.

DR POSTOW: I had a patient on BRAF/MEK inhibitors that had Bell’s palsy. Also, and that’s been published too, but recurrent after time period of recovery of the Bell’s palsy and then later on had to go back on BRAF/MEK inhibitors, and the Bell’s palsy recurred.

DR LOVE: Wow.

DR POSTOW: There’s probably an immunologic mechanism underlying that, because otherwise, I’m not too familiar with how inhibition of the MAP kinase pathway would lead to Bell’s palsy and cranial nerve 7 impairment. But I otherwise don’t really know mechanistically what happened in that case.

Correlation between toxicity and benefit with immune checkpoint inhibitors

DR LOVE: We talked about some of the toxicities associated with checkpoint inhibitors. Any thoughts or questions before we get into the metastatic setting?

DR POSTOW: Yes, I think one of the questions that comes to mind, I think, is what we know about side effects and immune therapy efficacy. And I know there’s a lot of studies in the metastatic setting, some of which had said skin toxicity is good for benefit of immune therapy in the metastatic setting. I guess one of my questions is related to what we know about side effects in the adjuvant setting, and could that be related to likelihood of benefit in the adjuvant setting, because it’s a hard situation to know. You can’t follow tumors over time to see if something is working or not working. And so it’s really hard to know if something is working or not. I mean, you know if it’s not working, as Dr Luke said earlier, if the tumors come back. But what is your thought on side effects in the adjuvant setting and expected efficacy?

DR SULLIVAN: I think even in the metastatic setting, it probably matters what type of side effect you’re having. I don’t think colitis is associated with better benefit. I don’t think pneumonitis is associated with better benefit. I don’t think neuritis is necessarily associated with better benefit. I think vitiligo is associated with better benefit. I think the rash that people get before they have vitiligo is associated with better benefit. Endocrine toxicity may be associated with better benefit. I think if the patient is on adjuvant PD-1 inhibitor therapy and they develop vitiligo, I’m encouraged. If they develop wicked pneumonitis, I’m discouraged. But not necessarily related, in that latter situation, that their disease is going to come back but that we’re going to have to manage their pneumonitis.

I think practically thinking, toxicity in the adjuvant setting, I don’t know of any really strong data to support that it’s associated with improvements in relapse-free survival in that setting. And I also don’t know whether or not managing the toxicity in ways that either spare steroids or are overly aggressive with steroids will influence relapse-free survival. But I think as more and more data comes out and more and more patients are treated with these agents that hopeful we’ll have an answer to those questions.

DR LOVE: Jason, any questions you want to ask him?

DR LUKE: In the context of now having FDA approval for all patients with Stage III disease, I was wondering if you just wanted to give a short commentary with some discussion on who would you actually not treat, because I think if you just look on the label, it seems like, wouldn’t we treat everybody? And yet we had some discussion before about how actually we wouldn’t do that. How do you think about that?

DR SULLIVAN: I think patients with the least risk, Stage IIIA, in the AJCC 8^th edition, mostly weren’t included in those studies. Some, if you tease out, some patients may have been. But I think IIIA patients generally are patients that I think long and hard about offering single-agent PD-1 inhibitor or even combination BRAF/MEK. If I’m going to choose a therapy, if the patient’s BRAF mutated in the IIIA population, I’ll probably more likely choose BRAF-targeted therapy because I’m less concerned about permanence of toxicity. And even fatal toxicity, even though it’s rare.

I think the IIIA population is — the line’s clearly between IIIA and IIIB in terms of which patients would you routinely feel more comfortable about offering therapy to and which patients you feel uncomfortable about offering therapy to.

Selection among the BRAF/MEK inhibitor combinations dabrafenib/trametinib, vemurafenib/cobimetinib and encorafenib/binimetinib for patients with metastatic melanoma with BRAF tumor mutations

DR LOVE: So let’s talk a little bit about metastatic disease, beginning with the choice of now 3 approved BRAF-MEK combinations. Mike?

DR POSTOW: I have a question about thoughts about choosing among the BRAF/MEK inhibitor combinations that are out there. We have 3: vemurafenib/cobimetinib, we have dabrafenib and trametinib and we have encorafenib/binimetinib. We have no data to suggest head to head exactly 1 of these BRAF/MEK combinations is better than the other. How do you work with the patient to decide which among these 3 BRAF/MEK combinations is best for them?

DR LUKE: It’s a really good point, which is to say of these approved therapies, which is truly better? I think there’s a combination, and I’m sure you’ve seen what’s the convenience factor of how do you take some of the medicines. For trametinib, especially the dabrafenib/trametinib combination, you have to keep the medicines cold before you take them. There’re all kinds of issues about how many hours before and after you can eat.

Advantages of the encorafenib/binimetinib combination for patients with melanoma with BRAF tumor mutations

DR LUKE: And then beyond that we have some hints from clinical trials that the most recently approved combination might have slightly more efficacy than the other combinations due to the randomization strategy and the dose escalation of the BRAF inhibitor to a higher dose. There’s no right —

DR LOVE: You’re talking about the encorafenib?

DR LUKE: Encorafenib and binimetinib.

DR LOVE: Binimetinib combination. What about tolerability of that combination versus the other two?

DR LUKE: It’s a little hard to say. Again, because these are different clinical trials. I think our anecdotal experience is that it might be better tolerated, although people argue about that point. I generally am also of the opinion that the efficacy might be slightly higher. And the reason is that dose for dose, the encorafenib dose is actually higher than the dabrafenib or vemurafenib dose, and so you actually suppress the pathway even more. But these are tough questions to know with any kind of authority which one is the best.

Use of encorafenib/binimetinib versus dabrafenib/trametinib versus vemurafenib/cobimetinib in clinical practice

DR LOVE: But just in terms of practice patterns, what do you generally utilize?

DR LUKE: Really we’ve been using much more of the encorafenib and binimetinib combination because of these data about it having a stronger potential to suppress that pathway.

DR POSTOW: Yes, I like the encorafenib/binimetinib also. I think no one would argue that dabrafenib and trametinib has the longest follow-up data and the most number of patients on randomized big Phase III studies, so there is some strength to that. But I think that if you look at vemurafenib as a control arm, in a number of these randomized Phase III studies the encorafenib/binimetinib combination really did well against vemurafenib, and obviously this is cross-trial comparisons — impossible to make any kind of definitive statement. But encorafenib/binimetinib looks really, really effective, so I’d go with that one first, and especially not having to keep the pills in the refrigerator like you have to do with trametinib. And cobimetinib you have to take for a few weeks on, and then you take a week break. It’s a little confusing for patients sometimes. I think the ease of administration and the really profound efficacy with hopefully pretty good tolerability for enco/bini. I like that one.

DR LOVE: And Ryan?

DR SULLIVAN: I usually do the same. Interestingly, it’s the most tablets per day, but it’s probably the easiest way to get the tablets in, too, because there’s no issues with eating. And I think the efficacy maybe, maybe not, but I think it is better tolerated. It’s just easier to get in patients, and it’s also, just like the vemuraf/cobimetinib, it’s easier to dose modify than it is with, say, dabrafenib/trametinib.

COLUMBUS: Results of a Phase III trial comparing encorafenib/binimetinib, vemurafenib and encorafenib for patients with melanoma with BRAF tumor mutations

DR LOVE: It’s interesting you’re talking about the encorafenib/binimetinib combination. It was approved based on the randomized Phase III COLUMBUS trial that compared encorafenib/binimetinib versus vemurafenib or encorafenib alone for patients with metastatic BRAF-mutant melanoma. The median PFS was 14.9 months for patients receiving encorafenib/binimetinib versus 7.3 months for vemurafenib alone, with a hazard ratio of 0.5. The hazard ratio of encorafenib/binimetinib versus encorafenib alone was 0.77. With respect to overall survival, the results showed a median OS of 33.6 months for the encorafenib/binimetinib group versus 23.5 months for encorafenib alone and 16.9 for vemurafenib.

Overall, the combination was well tolerated as you all have noted, and the most frequent adverse events with the combination were GI, including nausea/diarrhea and vomiting. The incidence of pyrexia and photosensitivity with encorafenib/binimetinib seemed to be less, comparing indirectly to trials of other BRAF-MEK inhibitor combinations.

The authors of the COLUMBUS trial concluded that the combination was “a new benchmark against which new BRAF-MEK inhibitor therapies for BRAF mutant melanoma can be measured.”

So, Jason, that’s a transition to your BRAF-mutant metastatic case.

Case (Dr Luke): A man in his mid-50s with metastatic melanoma with a BRAF V600E tumor mutation experiences a complete resolution of CNS lesions after receiving ipilimumab with nivolumab

DR LUKE: Absolutely. A patient of mine, 56-year-old guy, was diagnosed with a lymph node involved Stage III melanoma. It was originally on the right shoulder and involved the axilla, and underwent at that time, which was considered standard therapy, with surgery for sentinel node and then completion nodal dissection — this is dated back a few years now — but at the time we did test the tumor for BRAF and found that it was BRAF mutant, and at that time, actually, the standard consideration was to proceed in the route, and the patient was amenable.

We gave him dabrafenib and trametinib. And he did have a little bit of fever syndrome but actually tolerated the full year of treatment. And actually seemingly did very well.

I think the thing we’ve seen with some of these patients who are getting targeted therapy, however, was within a few months after completing the therapy, he started to get headaches and generally feel unwell. And we got an MRI and unfortunately found that he had already developed small-volume brain metastases.

DR LOVE: Before you go on, I’m curious, that comment you just made in terms of dabrafenib/trametinib and brain mets, does that mean it’s not as effective in the brain?

DR LUKE: It doesn’t mean that per se but rather that we’ve observed across the BRAF inhibitor trials, most in the metastatic setting, but there’s some suggestion of this in the adjuvant as well, which is that sort of a site of escape tends to be the CNS. And we can see that patients will continue to maintain a good treatment effect throughout the body, they can go on to develop brain metastases. Certainly you want to be able to look out if a patient starts to develop new headaches with melanoma anywhere on their chart. You want to get an MRI as quick as you can.

Therapeutic approach for patients with melanoma who develop brain metastases

DR LOVE: Can you talk a little bit about how you might think through this situation? This is occurring 4 months after stopping the dabrafenib/trametinib.

DR LUKE: Yes. The patient was seen for 1 follow-up visit at 3 months and seemed to be okay and then started developing these headaches and called in.

DR POSTOW: And did this patient have extracranial metastases that were responding, progressing?

DR LUKE: No, the patient was NED and got adjuvant BRAF/MEK, and then this was sort of right after having completed —

DR POSTOW: And brain-only recurrence then?

DR LUKE: Yes.

DR POSTOW: One of my major considerations if you have a brain-only recurrence is, make sure that you understand the full burden of disease throughout the entire body. Not unusual to have brain-only disease in melanoma in general.

One of the other questions then would be, what is the degree of burden of brain metastasis involvement? If this patient was having headaches, presumably they were sizeable metastases or maybe, potentially, a number of them. Because I think one of the questions is, should this brain metastasis event be considered for locoregional treatment? If this patient is having headaches or they’re having other neurologic symptoms, should we be thinking of the possibility of surgery? If it’s 1 isolated metastasis in someone who, after a year of treatment, otherwise pretty high risk, may have benefited from surgical intervention? I think kind of understanding what are we talking about with the brain metastasis is going to be really important here.

DR LOVE: Before we go farther, can you clarify that? What was the situation?

DR LUKE: Yes. The patient didn’t have any neurological deficit, just these headaches that would kind of come and go. They weren’t even persistently debilitated or anything by the headache, they just noticed this as a new symptom. And when we got the MRI — first the CT scan showed no other systemic disease, so to that point. And then got the MRI. There were actually 3 different isolated metastases that showed up, I think 2 in the frontal lobe and 1 in the parietal, I think. But there was no associated edema. That was one of our other important considerations with just how much of a clinical entity was this?

DR LOVE: In terms of edema, how often do you see patients who have edema as opposed to not having edema?

DR SULLIVAN: I think it may be more of the degree of edema. Oftentimes we see edema with tumors, and it may or may not be clinically significant. I think you see a tumor and there’s a lot of edema — you can sometimes in a very small tumor have a ton of edema, and you get a very large tumor, minimal edema. And in each of those scenarios may be slightly different. But I’d say, by and large, clinically significant edema that we have to do something about is common enough that we consider as Mike said to take out tumors that are edematous and causing problems.

DR LUKE: For people in the community, you’re treating a patient, and they show up with brain edema, so you pretty much can’t start, because they’re on steroids. You start them on BRAF inhibitor. What are the landmarks? How do you think about, should you give it forever? Should you switch? How would you think about that with a patient? Because I think this is a pressing issue that comes up in community practice.

DR POSTOW: If I had to start with BRAF/MEK in the brain for a BRAF-mutant patient, I would probably switch them within 2 months, I would say. Three months maximum, because we see a lot of those progression-free events happening very early. The median time of initial response is only 6 months. I would switch them 2 to 3 months into treatment to make sure that they had enough time to hopefully benefit from ipi/nivo if we could get them off of steroids with the BRAF/MEK initially.

DR SULLIVAN: And then in the BRAF wild-type population, how do you try and push to get people off steroids? For example, will you take people to surgery? What’s your approach?

DR POSTOW: I think you have to ask the question of, why is this patient on steroids? A lot of times steroids just get thrown on patients because you hear brain mets and may see edema on a scan, and you just start steroids. But the patient might not really need them. If they absolutely need them for symptoms, I do see if a patient’s a surgical candidate first, because if you could remove a big brain metastasis, especially if it’s a solitary brain metastasis, it’s usually in the patient’s best interest to do so.

If a patient really needs stereotactic radiosurgery up front to get off of steroids, then I do think about that up front. I think it remains a question of how early do you get the systemic therapy going in those patients? And I don’t think you’re wrong to get started with ipi/nivo if the patient’s also getting SRS to hopefully get off steroids, if there’s absolutely no way to get off steroids otherwise. I wouldn’t delay systemic therapy for a very long time to wrestle with getting them off of steroids in the absence of a BRAF mutation, for example. But I do think of local treatments in those patients.

Locoregional versus systemic therapy for patients with brain metastases

DR LOVE: What are you thinking at this point?

DR POSTOW: Three tumors is a lot for surgery in someone who’s only a year-plus after initial diagnosis. It’s a concerning biology that now we’ve already had metastatic disease after completing a year of BRAF/MEK treatment. One would say, is it really resistant to BRAF and MEK? Because this arose now 4 months after the completion of treatment, so technically these diseases in the brain arose off of BRAF/MEK.

Hard to really define what resistance is going on there. But you could certainly say that the BRAF/MEK didn’t kill all the micrometastatic disease for this to have happened. I’m worried about RAF resistance, even though this event happened within 4 months after the discontinuation of BRAF/MEK in the adjuvant setting.

And then the question really is, we’re trying to get our hands around, is this a patient for whom we need to think about immunotherapy? And I think getting to the issues of how much edema there is. How much, if any, steroid we need to use to control the headache symptom in relationship to the edema. That’s important. Because what we know is that immune therapy is less effective in the presence of steroid use at baseline in patients with brain metastases.

Trying to get a handle on what are the different ways we can treat this patient? Because in the symptomatic patient with CNS metastases, it’s not so obvious that immune therapy is for sure the best way to go. And so I’d like to really understand what locoregional options there are on top of whatever systemic therapy choice I might be thinking about switching to, including stereotactic radiosurgery.

DR LOVE: Just out of curiosity, we were talking about Stage IV NED before, but I didn’t ask about specifically this scenario with CNS Stage IV NED. If there were 1 lesion, for example, and it was taken out, what kind of systemic therapy, if any, would you use in that situation?

DR POSTOW: You would have to decide how much you believe that the BRAF/MEK was resistant. And should you go back on BRAF/MEK, because this maybe was an aberration? But with 3 lesions in the brain, I’d be very hesitant to believe that.

The most aggressive approach in that situation would be switching to ipi/nivo, even after resection. We don’t have data specifically for that, for resected brain metastasis at least, but what we do know is ipi/nivo, more so than single-agent PD-1, is more effective in established CNS metastases that are not surgically resected. And so I believe that when you have 3 new lesions that pop up on a patient who’s likely RAF resistant that it’s probably in the patient’s best interest to be on ipi/nivo thereafter, protecting against microscopic CNS metastasis that, unfortunately, are very likely to pop up over time.

Activity and tolerability of nivolumab/ipilimumab in patients with CNS metastases

DR LOVE: Let’s talk about what happened.

DR LUKE: Yes. For this patient, we were aware about using ipi/nivo in that context and actually then recently data has now come forward for prospective Phase II trials treating patients who are not on steroids and have good performance status with that combination at the higher dose of ipilimumab, 3 mg/kg. That’s what we did for this patient, and, actually, the patient had a complete response in the brain.

And that’s actually consistent with the data set that has been presented by 2 different groups in the US and in Australia, which I really think is important to emphasize, because it’s a rather unique — it’s not the way we historically thought about treating brain metastases. And these patients, they did not get radiation, rather went straight to immunotherapy and its long-term disease control rate of 50%.

You asked a real good question earlier, which is, how many patients actually don’t have edema and aren’t on steroids? It turns out that actually is the minority of patients, because many of them do have edema or do have some sort of problem where you can’t really just give immunotherapy up front.

This isn’t to say that in every patient who shows up you can transition immediately to ipi and nivo, but I think for patients who have low enough volume disease that they’re asymptomatic, this really should be the priority approach as opposed to doing radiation initially.

DR LOVE: And how did he do in terms of tolerability of the ipi/nivo?

DR LUKE: Oh, he got colitis.

DR LOVE: Really?

DR LUKE: It was the fourth dose. It was, like, 2 days after the fourth dose. He called in and said, I have this… But that’s not uncommon. That’s really what we see is, half the patients will develop Grade 3, 4 toxicities. And so we weren’t surprised at all to get that call. And so we immediately put him on immunosuppression to block the effect, and he got better within a week or two.

Management of colitis associated with immune checkpoint inhibitors

DR LOVE: What specifically did you put him on?

DR LUKE: We gave him steroids, first as an outpatient. We gave him — I think it was 80 mg of prednisone. It was about 1 mg/kg. He did not get immediately better, so we did put him in the hospital and gave him 2 mg/kg, but within just a couple of days he was okay.

DR LOVE: Do you ever use infliximab?

DR LUKE: We do. And I actually am interested in what others’ perspectives are. I tend to be a pretty heavy user of infliximab, in fact. Just harkening back to the good old days of ipi monotherapy, when we first started doing this, we used to leave patients in hospital for days and weeks, giving them steroids. And you’d go to the bedside and sit there with the family and they’d say, “How long is this going to go on?” And you’d say, “I don’t know,” right?

But in this day and age, we don’t do that at all any more. Now I give the patient 1 mg/kg for a day, maybe 48 hours, then I give them 2 mg/kg, again for a day. And if he isn’t better by, like, day 3 or 4, I give them infliximab, 5 mg/kg. Because it’s just — the patients are miserable. There’s now tons of data to suggest that we should not be belaboring this and the outcomes are not diminished for these patients.

DR LOVE: Anything you want to add to that?

DR SULLIVAN: No. I again would say that we often will use the findings at colonoscopy to help guide. If you go in and the mucosa looks like pizza, we’re much more likely to use infliximab even as we’re loading up with high-dose steroids. We’ll probably consider it if they’re not better in 24 hours to give infliximab. Whereas if we see that microscopic colitis picture where there’s minimal amounts, we’re much more likely to go to oral steroids and ride out the steroids a little bit longer.

DR POSTOW: I think it’s an important point. I think you benefit from being a center where you can get a scope quickly —

DR SULLIVAN: For sure.

DR POSTOW: — and use that information to help manage. And sometimes in community practices trying to get a GI referral, then booking a scope, all these things can take some time. I think it’s also important to mention you’re not wrong to treat with steroids and/or infliximab on top of that without having histologic proof or even colonoscopies to visualize the mucosa to really know. Sometimes you just have to empirically treat with immunosuppressants.

DR LUKE: I would also add that there’s some pretty good data now for radiologic evaluation of colitis, which is the correlation between CT scans in the ER and eventual colonoscopy findings, is very tight, like, more than 0.9 for a Spearman correlation.

Basically, if you pick up edema in the colon, you can tell sometimes, if the colon looks wrong on the CT scan, those patients almost always have active colitis. And some centers have even moved to that. And if they see anything on CT, they just give steroids and they don't even do the colonoscopies any more.

DR LOVE: Wow. This is the wall of the bowel?

DR LUKE: Mm-hmm. It’s just like an IDD or anything else, right, and the patient comes in, and sometimes you can worry there could be a diverticulitis or something like that, but the correlation is really high. If your pretest probability is there for colitis, then that’s probably what the patient has.

Efficacy of radiation therapy for brain metastases in melanoma

DR LOVE: Just to follow up, though, in terms of the brain mets. Mike, any comments about radiosensitivity of melanoma to radiation therapy? It kind of has a reputation of being radio resistant. Is that the case?

DR POSTOW: We’ve seen radiation work really well in melanoma. I think a lot of that comes from comparing melanomas to lymphomas or other really classically really radiosensitive tumors, but melanoma can respond very nicely to radiotherapy. And when you radiate small brain metastases, especially in the presence of immune checkpoint blockade, we believe that can be very, very effective. And I think that as a field, we need to acknowledge that radiotherapy can have a really important role in treating melanoma, especially brain metastases. It’s just a matter of where to integrate that into the timing of checkpoint blockade and radiation if it’s still open.

DR LOVE: Are you an abscopal believer?

DR POSTOW: I will say that the abscopal effect can happen in individual patients, but it happens rarely enough that you can have an exciting case report on it. And Dr Sullivan, and he’s seen them also in Boston, but I don’t think you should radiate melanoma tumors with the idea that you’re going to induce an abscopal effect. It’s a little bit like Santa Claus, like you want to believe that it’s true, but it usually doesn’t happen, unfortunately.

DR LOVE: I mean, abscopal is a clinical observation. But the way you phrased it earlier, do you think that radiation in some way primes the patient for immunotherapy?

DR POSTOW: Yes. In terms of interactions with radiation and checkpoint blockade, there are a couple that I think are important to keep in mind. One of them is, does the abscopal effect exist? Which is, if you radiate a tumor, does that globally enhance the ability of checkpoint inhibition to be more effective in all tumors, even the nonirradiated ones? And that we don’t have proof of.

The second is, if a patient’s on checkpoint inhibition already, like with pembrolizumab, and then you irradiate a tumor, does the tumor that you irradiate respond remarkably well? Does checkpoint inhibitor treatment increase the efficacy of the irradiated lesion? It’s kind of like, does the checkpoint inhibitor help the radiation work better, or does radiation help the immune therapy work better? And I think what we know is a little bit more about the former. That when you irradiate a tumor in someone who’s on checkpoint inhibition, the radiation of that specific tumor seems to be pretty successful.

DR LOVE: Hmm. That’s interesting.

Optimal type of radiation therapy for patients with CNS metastases

DR LUKE: I wanted to make a comment quickly about radiotherapy in melanoma, more broadly, however. Because I think one thing that’s a little different that probably needs to be teased out was that for many years when we had generally ineffective systemic therapies, we used to be very aggressive with stereotactic radiosurgery in the brain. And the reason I’m highlighting that is that as compared with some other cancer types, whole brain radiation had very little efficacy in melanoma, to the point that it’s actually broadly recommended not to pursue it.

There was an important study at ASCO last year looking at patients to get whole brain versus not get whole brain after resection of mets and actually showed that for some clinical outcomes, the patients who got the whole brain radiation did worse, and that was a cognitive impact. And there was definitely no improvement in survival outcomes.

DR LOVE: You don’t use it?

DR LUKE: We basically don’t use it. That isn’t to say you can’t ever do it. I mean, there are patients who, unfortunately, have progressed on everything, and you’re willing to try anything. But generally speaking, I actually would go that far out to say we don’t do it unless we really have completely run out of other options.

DR LOVE: I’m thinking about some of your colleagues in lung, Geoff Oxnard, et cetera, who nowadays when they see people with multiple bilateral brain lesions who are asymptomatic who have the targetable lesion, like EGFR, they’ll use systemic therapy even in the presence of something that you could use stereotactic for.

It almost sounds like this patient fits into that category, you could have used stereotactic, but you didn’t. How do you approach that question?

DR SULLIVAN: I think brain metastases deserve systemic therapy. And if our general practice is to, in a patient like this, we would commence ipi/nivo. Instead of waiting 12 weeks, which is our normal interval to repeat scans, we would probably repeat the brain imaging in 6 weeks and just see what’s happening. If things are starting to grow and we’re getting a little nervous that we’re not going to see a response, then we tend to do radiation.

And I think it’s important to do this in a multidisciplinary way. Every community oncologist has a radiation oncologist they work with and to have a conversation before starting this therapy and saying, it’s a great data with immune checkpoint inhibitors in patients with brain metastases in melanoma. I would like to start that, but I’d like you to be available to see the patient, and at 6 weeks we’ll see them again together, and then we can make a decision about whether we need to do stereotactic radiation or not.

Radiation necrosis with the combination of stereotactic radiation and immune checkpoint blockade

DR SULLIVAN: And the argument could be made, what’s the downside of doing stereotactic radiation? The downside is, there’s probably more profound radionecrosis in the brain, specifically when you give stereotactic radiation and immune checkpoint inhibitors together, and I’m sure we’ve all had cases of patients who’ve had very wicked radionecrosis.

DR LOVE: I’ve never heard that. Is it an immune issue?

DR SULLIVAN: I don’t know if it’s an immune issue or if it’s just — I’m sure it’s somehow immune related, because we’re giving immunotherapy. But there does seem to be this robust signal. And it’s hard to know whether the robust signal is being seen because patients are living 3 and 4 and 5 years after they receive stereotactic radiation and that might be the natural history of stereotactic radiation in patients with melanoma, and now that patients are living that long we can see it.

DR LOVE: Hmm. This is an imaging finding usually?

DR SULLIVAN: And oftentimes symptomatic.

DR LOVE: Symptomatic?

DR LUKE: The patients often have to go back for resection of that area of the brain to get that tissue out. Because they can start to seize, and they can start to have all manner of neurological problems.

DR LOVE: I mean, when I think about that, maybe it’s wrong, I think about, like, a vascular issue. Is that what happens with radiation necrosis?

DR POSTOW: I think that we’re still trying to figure out the mechanisms, but there’s got to be a component of vascularity. There have been some —

DR LOVE: It’s necrosis.

DR POSTOW: — data sets that suggest bevacizumab could be helpful in some of these contexts.

DR LOVE: Really? Wow.

DR POSTOW: Or antichromatoids with steroids sometimes can be helpful. But I think — Dr Sullivan mentioned that this may also be because patients are living so much longer with their disease in general. Because they’ve done well with checkpoint inhibitors, we’re seeing some of these longer-term late effects like radionecrosis in patients for whom we’ve radiated in the brain.

Case (Dr Postow): A man in his late 60s with low-volume metastatic mucosal melanoma and 1 incidental brain metastasis receives nivolumab/ipilimumab

DR LOVE: This is the good side of ipi/nivo for brain mets, but I’ll just say, the more interesting side would be your case 3, in terms of maybe not being quite so straightforward. Just to maybe balance it out a little, can you tell us about this 67-year-old man?

DR POSTOW: Sure. Yes. This is a 67-year-old man. He had low-volume metastatic actually mucosal melanoma, so not cutaneous melanoma. And this patient had 1 incidental brain metastasis that was discovered at time of initial staging, and it was 0.6 cm, so 6 millimeters in general, which is relatively on the smaller end of things but still measurable and would have been included in any other clinical trials.

And so one of the questions that we had with this patient with metastatic mucosal melanoma is, as we always try to decide, do we always go with ipi/nivo when we see a little brain metastasis? Does the mucosal melanoma factor into this at all? And this 6-mm lesion in the brain was really very ripe for picking off with stereotactic radiosurgery, so what’s the role of stereotactic radiosurgery in the patient? All the discussion that we’ve been having kind of comes up right in this type of a patient.

If he had way too many brain metastases, then stereotactic radiosurgery really isn’t an option, but this was just one. Those are some of our initial treatment considerations that we had.

DR LOVE: Any thoughts from what you heard so far?

DR LUKE: I’d go along the lines of what we referred to, and then really the question becomes, what do we think the upside benefit of this sort of an approach, and what’s the input even from the patient in addition to the different doctors about what do they want to do. I think in mucosal, an especially difficult one where our treatment option is actually less robust than in other forms of melanoma, and so there, we’re going to be even more concerned that we’re going to have issues with primary resistance.

I don’t know what you guys chose, but I think I could probably go either way. These are patients that don’t do as well, and so going for the house right up front could be reasonable, and taking a more nuanced approach could be as well.

Response to nivolumab/ipilimumab in patients with brain metastases

DR LOVE: What happened?

DR POSTOW: We gave this patient ipi/nivo as well. And I think it’s important — there are a couple of different ipi/nivo doses that are out there right now. There’s a lower-dose ipi 1 mg/kg plus the higher dose of nivolumab 3 mg/kg. And then there’s the higher dose of ipi 3 mg/kg with the lower dose of nivo 1 mg/kg. And we went with the higher-dose ipi with the lower-dose nivo, which is the dosing regimen that is the standard FDA-approved dose and the dose that we have brain metastasis efficacy results. I think that’s a really important point to make if you’re going with ipi/nivo for brain mets. I would advocate for using that ipi 3 mg/kg dose with nivo 1 mg/kg. That’s what we gave this patient.

DR LOVE: I’m just kind of curious, when you had the discussion with the patient, did he get involved in the decision, or he just left it up to you?

DR POSTOW: Yes. I mean, I think patients are always part of — this is a dialogue back and forth and trying to understand what’s best, but it’s really complicated, I think, to ask a patient, “What do you want to do?” I think they’re looking for our guidance. I think we can give them a suggestion. And a lot of patients don’t like the idea of radiation treatment, just the thought of it and having to go for a mask and the simulation and some of the ideas of it patients don’t like.

I think the challenging thing is trying to know — you wish you knew who was going to respond outright to ipi/nivo, especially in the brain. And I felt more comfortable going with systemic therapy alone for this patient, because this was a relatively small lesion. It was asymptomatic and picked up on routine brain MRI as a pretreatment standard test, and I think that’s important.

As we think about response rates in the brain to ipi/nivo, it’s not 100%, and I think we’ve had this discussion about starting with systemic therapy following up closely with the 6-week brain MRI, which I think is great. But it is important to think about, if you have a brain metastasis in an eloquent area, like the brain stem or the midbrain or potentially in the motor strip, even if it’s not causing any symptoms, I’d be a little bit more inclined to be a little bit more aggressive with up-front stereotactic radiosurgery in that patient, because with immune therapy alone, you’re still running a risk of 40% to 50% of progression there.

And so if the progression is going to cause a significant neurologic toxicity from the brain metastasis disease itself, I’d be a little bit more inclined to do stereotactic radiosurgery. But this patient did not have it in an eloquent area. It was relatively small. This patient was not on steroids. No symptoms. I felt more comfortable with immune therapy alone.

Side effects with nivolumab/ipilimumab

DR LOVE: What happened?

DR POSTOW: After dose 1 of immune therapy, kind of a classic first early toxicity, which is a rash, which is easily managed with triamcinolone cream. And then after the second dose had a lot of side effects. And this was a whole spectrum of kind of nonspecific side effects that you’re trying to figure out what “itis” it is, what organs are involved, but this was a febrile toxicity, tachycardia and fatigue. And so those are sometimes some of the nonspecific symptoms that are on us now to tease out what organs are inflamed and what’s going on, because this patient was having fever, high heart rate and just feeling terrible. And went to an emergency room with these symptoms for evaluation.

And when he was in the emergency room, we noticed that as often happens with fevers, his transaminases were a little bit elevated, so AST and ALT were elevated. Thankfully bilirubin and alkaline phosphatase were normal. Kind of an AST/ALT problem initially. CBC was normal. We’re starting to think about this being some kind of an immune therapy-related problem, especially after 2 doses of immune therapy. And what was also interesting is, the EKG that was done, that the patient’s heart rate was going right around 140, and the patient had new onset of atrial fibrillation.

We were trying to tease this apart: hepatitis, fever, A-fib. Is this A-fib because of dehydration and sensible losses from fever? Could something else be going on?

And what we ended up finding out is that his thyroid tests were totally out of whack. And sometimes those thyroid tests don’t come back right away, but it’s really important to think about when you have 1 immune therapy side effect like hepatitis and there could be other things involved, like some atrial tachyarrhythmias and fevers, to make sure that you think about endocrinopathy as underlying some of this. And we sent TSH and a free T4, and free T4 was very elevated and TSH was low, suggestive of thyroiditis and primary hyperthyroidism as a result.

Management of hyperthyroidism associated with immunotherapy

DR LOVE: What did you do?

DR POSTOW: It’s an interesting problem when you have this thyroiditis. First, we had to give the patient hydration. And we were thinking about giving this patient steroids anyway for the liver problem. But it’s a really open-ended question when you have someone with acute thyroiditis, what is the role of giving steroids in that type of a context? And I think unsuccessfully I’ve tried a few times to try to save the thyroid gland and give steroids to try to see if I can squash that inflammation early and prevent long-term hypothyroidism from developing. And I’ve never been successful in that approach. And Dr Sullivan and others have suggested perhaps high doses of steroids and certain endocrinopathies, mainly pituitary inflammation, might actually be detrimental to patients. And you should just replace the missing hormones that they have rather than giving high doses of steroids.

DR LOVE: Detrimental immunologically, or in what way?

DR SULLIVAN: You live less long.

DR LOVE: Do you know why?

DR SULLIVAN: The specific scenario that Mike is talking about is hypophysitis. In hypophysitis you have to give steroids, but you can give low dose or you can give high dose, high dose with the idea of trying to kind of heroically save the pituitary gland and low dose meaning you’re just saying we give up on the pituitary gland ever being able to make enough cortisol, and so we’re just going to replace cortisol.

And in that study, what we found, and this is work led by one of our neuroendocrinologists, that the survival — so symptomatically patients had the same headaches and double vision and others, fatigue, that patients get with hypophysitis, that resolved quickly whether you gave high dose or low dose. But patients who got high dose lived less long than patients who got low dose. And we didn’t see any specific difference in their tumors. This is all patients with melanoma who developed hypophysitis.

It’s not proof. It certainly could be confounded with other variables that we didn’t see or didn’t have access to, but it’s also possible that the high-dose steroids blunt the initial antitumor response.

DR LOVE: What’s the patient’s current situation? What happened?

DR POSTOW: This patient’s thyroiditis and initial hyperthyroidism eventually, as almost always is the case, kind of burns out to become hypothyroidism over time.

DR LOVE: How long did it take?

DR POSTOW: Probably about a month.

DR LOVE: And what about the A-fib?

DR POSTOW: A-fib resolved over time. We gave this patient beta-blockers to control his rate. And I think that’s one of the keys to think about. When you have hyperthyroidism from thyroiditis, it’s best to just treat these patients symptomatically, in my mind, rather than giving them a lot of steroids up front.

This patient was on a beta-blocker for a little while, felt okay with that. And then eventually the thyroid function went down, and now the patient’s on levothyroxine to supplement what is kind of expected to be more permanent hypothyroidism.

DR LOVE: And in terms of the disease?

DR POSTOW: The antitumor effects were fantastic, actually. That little brain metastasis responded. The patient’s without evidence of disease. It was a rough go with ipi/nivo, but, thankfully, this patient has done well from a melanoma standpoint.

DR LOVE: How long has it been now?

DR POSTOW: That was probably the fall of 2017, so we’re going on almost 2 years or about 2 years.

DR LOVE: Interesting.

DR POSTOW: Which is great. And what we know from our studies is that those responders in the brain can have durable responses, which is really fantastic with ipi/nivo.

Monitoring for endocrinopathy in patients receiving immune checkpoint inhibitors

DR LUKE: One follow-up point, and I think it’s more general than just this case, but if you’re treating patients with checkpoint blockade, and definitely if you’re treating with CTLA-4-based checkpoint blockade like ipilimumab, if weird stuff is going on, check the hormone levels. Just because you can have all kinds of weird presentations for patients that are undergoing some kind of abnormality, adrenal. And I got laughed at when I started doing this, but I sometimes even check a random cortisol in clinic at 2:00 in the afternoon. And the endocrinologists, when they see this pop up in the computer, will call you and tell you “Hey, didn’t you go to medical school? You can’t do that.” But it’s actually twice in my career I’ve picked up people who had zero for their cortisol level in clinic.

DR LOVE: Wow.

DR LUKE: Which, basically, meant they were going to go home and fall down, right? And we managed to catch that. And so because if you don’t think about it, you won’t find it. And so some of these things are just very odd in terms of presentation.

DR POSTOW: Yes. And the other thing, I mean, you can mask hypophysitis. If you’re on steroids for something else like colitis or pneumonitis or some other problem and you’re giving high doses of steroids and that patient had had pituitary dysfunction, as you’re tapering off the steroids, everyone gets tired, and then you might stop steroids, and people feel terrible. And you realize that now this patient also needs chronic replacement hydrocortisone or some other low dose of prednisone. Because when they’re on high doses of steroids for something else, you were actually masking the hypopituitarism. These are really tricky things.

And depression, that’s another thing I’ve caught hypopituitarism on, or even hypothyroidism. Patients come in, “I’m just feeling down. I’m tired. And I just feel depressed,” and people think that they’re kind of lethargic and apathetic, and we wonder if it’s a psychiatric problem. And then you check the hormones, and they’re hypothyroid or low cortisol, and then you can fix it right away.

DR LOVE: Yes, I remember this case Jeff Weber presented to me of this man who was very cantankerous. And then he came in, and his wife goes, “He’s so nice. He’s, like, not the same person.” And he had hypophysitis. And then he went back to being obnoxious.

BRAF tumor mutation status and benefit from nivolumab/ipilimumab

DR LOVE: I wanted to really focus on CNS mets in terms of metastatic disease, because I know how important that message is. But I have to ask you an old message, which is the approach to patients with BRAF-mutant disease, first-line therapy. And particularly what I want to get — because over the years what we’ve been hearing from you all is generally, I think, a preference for immunotherapy. But I just want to get an update on — we were talking before about biomarkers. Because when I first saw the data coming out, I think most people were like, “Oh, this is going to make a lot of sense. Look at PD-L1 levels.” And yet as I started to talking to you, I was hearing questions about the validity of the PD-L1 assay and the actual meaning of it. And here we are now in lung and all these other things, and we’re looking at it all the time, and we hear this from you. Can you kind of, like, go through that story a little bit in terms of where it is today?

DR SULLIVAN: In terms of the PD-L1 staining and the other general biomarkers and how we would work up a patient with metastatic disease, make decisions?

DR LOVE: And what we know about biomarkers to try to inform that decision. And particularly the issue of ipi/nivo, to me.

DR SULLIVAN: We could start at BRAF. BRAF mutation means that BRAF-targeted therapy is an option. No BRAF V600 mutation means BRAF targeted therapy, as we know it are the 3 combinations, is not an option, but it doesn’t mean that should be the option if you’re BRAF mutated.

There’s data now that’s looking fairly robust. It’s hard to say anything about subset analyses, but the subset analyses of the CheckMate 067 study, which is comparing ipilimumab plus nivolumab versus nivolumab versus ipilimumab in the patients who are BRAF mutated, and it was a stratification factor, so it’s fair to look at in subset analysis to at least say, we at least controlled for this.

It looks like almost all of the benefit is in the patients who are BRAF mutated, in terms of combination checkpoint inhibitor therapy versus PD-1 inhibitor therapy. And I’m not sure why that is. And when we first saw it, or at least when I first saw it, I wasn’t sure what to make of it. I said, “It’s just a subset analysis and it’s BS, because you can’t just look at a subset analysis and say it’s real.” But the data is pretty strong, and those curves continue to be very separate.

And so one could very logically look at that data and say for BRAF-mutated patients, all things being equal, you might want to consider a combination therapy. Again, that has nothing to do with PD-L1 status. It has nothing to do with tumor mutation burden. It has nothing to do with whether there’s T cells there, but just the presence of a BRAF mutation seems to be associated with a higher benefit with combination therapy versus single-agent PD-1 inhibitor therapy.

Correlation of PD-L1 expression and benefit from immune checkpoint blockade in patients with BRAF wild-type melanoma

DR LOVE: What about BRAF wild type?

DR SULLIVAN: BRAF wild-type patients in general, when I’m deciding to give combination immune checkpoint inhibitor therapy versus single-agent therapy, say in a BRAF wild-type patient, but in every patient, do they have a brain metastasis? The data in brain mets suggests we should give combination.

Is this their 1 shot at therapy? Would I logically predict that they’re going to die if they don’t respond to whatever I offer them as front-line therapy? If the answer is yes, they get combination therapy.

Is their LDH very high, or do they have very widespread disease? Is it symptomatic? I would be favoring combination therapy. Those are biomarkers on some level, but they’re more clinical biomarkers. They’re not a test that we’re doing.

DR LOVE: But then PD-L1 level in BRAF wild type?

DR SULLIVAN: Yes. PD-L1 level in BRAF wild-type. I have no idea what to make of it. The data suggests that it’s a little better than a coin flip if you’re PD-L1-positive in terms of helping you decide whether you should give combination or single agent.

And so ultimately I think that the challenge is, if you need a response, you can look at every single cutoff for PD-L1. You can look at 1%, 5%, 10%. If you’re PD-L1-positive, you do better from a response standpoint with combination therapy than you do with single agent. And so if you fall into one of those clinical categories where it’s really important that you respond, you should get combination no matter what the PD-L1 status.

DR LOVE: And if it’s not important?

DR SULLIVAN: And if it’s not important, it’s probably still not important.

DR LUKE: It’s also then a risk-benefit consideration around the toxicity profile, because they’re so different when you’re engaging with patients around which therapy to choose. On the one hand, you have a therapy with a combo, which has a high upside, but it’s got a 50% chance of giving you toxicity. On the other hand, you have fairly high upside, but you only have a 10% chance of getting severe toxicity, right? That’s the other piece that ends up playing into this.

DR SULLIVAN: And the other thing is, if it’s somebody who doesn’t fall into that clinical category that they’re going to die within 3 months if they don’t respond, you have the chance to offer that patient ipilimumab 3 months down the road when you do their scan and you see that they’re progressing. And so you’ve probably taken into account many of the patients that don’t benefit from single agent if you’re able to directly give them the ipilimumab post-PD-1 or combination post-PD-1.

Role of tumor mutation burden and PD-L1 expression as predictive markers of benefit with immune checkpoint blockade

DR LOVE: TMB was mentioned before. What do you know about that?

DR SULLIVAN: We know that high TMB and also high UV signature — so if you look at the specific DNA changes associated with UV, it tends to be C to T transitions. Those are associated with increased response rate. But there’s no cutoff that you can make to say if you’re above this you’re definitely going to respond, if you’re below this you’re not going to respond. The data is, like, they do these grafts, and you see all the dots, and, yes, the responders have higher TMB and the nonresponders have lower TMB, but there’s no line you can cross to say this is the cutoff where we give single agent or we give combination.

DR LOVE: And also, another question is the validity of the PD-L1 assay. I think it was Keith Flaherty who brought this up to me, like, again, you never hear that in these other cancers.

DR SULLIVAN: Yes.

DR LOVE: Is it an issue?

DR SULLIVAN: I think it’s valid now for sure. I think when the assays were being developed it seemed like some labs could get really good data and some labs couldn’t get really good data. I think any lab that’s offering it has figured out the intricacies of doing the test, and I’m not worried about the validity of the test any further.

I think one thing is, it’s an inducible protein. What happened before might influence whether it goes up or down. And it’s heterogeneously expressed. If you stick a needle here and the PD-L1 expression is here, it’s going to be negative, but you missed it.

I think the challenge is, it’s not a perfect assay mainly because it’s not homogeneously expressed throughout the tumor. And in melanoma we’re usually getting core biopsies. In lung cancer you take the whole tumor out, and you get a chance to look at the whole tumor. Because that patient had their initial —

DR LOVE: Core biopsies usually in metastatic disease.

DR SULLIVAN: In metastatic disease you’re doing a core biopsy.

DR LOVE: Again, just kind of thinking about the predictive value of it, what do you see first of all in terms of this spectrum of PD-L1 levels in melanoma, like, what — and again, we're used to thinking about 50% using the lung assay.

DR SULLIVAN: Yes.

DR LOVE: What do you see here?

DR SULLIVAN: Most of the data now is, like, doing 1% or less than 1%. And the majority of our patients have expression, but I’m not sure I know the exact number in my head. Maybe you guys know.

DR LUKE: There were previously data from 2016 around these questions, but the field in melanoma has really noticed what Ryan had mentioned earlier, which is that the effect is the same no matter what cutoff you use. The reports now generally report out 1%, because it doesn’t really matter. It tracks linearly over time, so even if you did 10%, you end up somewhere nearby.

There are some differences in terms of what percentage of patients get called positive, but again, because of these other factors, we don’t end up differentiating all of that. And when we think about the clinical utility of should we give a combination immunotherapy versus a monotherapy.

DR POSTOW: That being said, I mean, I think there is some interest in the heterogeneity question, and there are clinical trials that are testing noninvasive whole body imaging of PD-L1 with nuclear tracers. Whereas you think you go for a PET scan to look at —

DR LOVE: Wow.

DR POSTOW: — FDG activity among different tumors, there are new tracers in clinical trials that are able to look at PD-L1 expression within tumors to try to give the whole-picture score of a patient’s PD-L1-positive disease burden. And I’m hopeful that over time, although we don’t have a lot of data on this at this point, just getting to kind of future directions, hopefully this will be more accurate as a biomarker than one specific tumor location where you happen to put 1 needle from 1 core biopsy and reflecting a whole patient’s PD-L1 positivity.

DR LOVE: I mean, does it make sense to you that a patient who has a tumor that’s expressing a lot of PD-L1 would benefit as much from PD-L1 alone as with the combination?

DR POSTOW: I think it’s hard to say that they would benefit more from PD-L1.

DR LOVE: No, the same.

DR POSTOW: The same.

DR LOVE: The same. That you don’t get an extra advantage if you’re PD — wasn’t that sort of the initial story?

DR POSTOW: That was the initial — yes, the initial CheckMate 067 data was, maybe ipi/nivo is best in PD-L1-negative patients. Because PD-L1-positive patients do just fine with PD-1 monotherapy.

DR LOVE: Right. Right.

DR POSTOW: But I think as the data have emerged we’ve seen that there’s that slight advantage, at least in terms of response rate and some of these other clinical endpoints for combination, albeit with much higher toxicity, but kind of regardless of the PD-L1-positive or -negative status.

We don’t really use it as a distinguishing measure of combo immune therapy or single agent. Unfortunately, it’s not that simple. I think it would have been nice if it worked out that way.

DR LOVE: This concludes our program. Special thanks to our faculty, and thank you for listening. This is Dr Neil Love for Oncology Today — Melanoma Edition.