Mario Sznol, MD, discusses adjuvant vs neoadjuvant immunotherapy in melanoma and highlights emerging vaccines, cell therapies, and T-cell engagers.
Anti–PD-1 therapy remains the backbone of adjuvant treatment for high-risk resected melanoma, but emerging neoadjuvant data, most notably from the phase 3 NADINA trial (NCT04949113), have triggered new questions, according to Mario Sznol, MD.
“If, in fact, the treatments that are being studied in the adjuvant and neoadjuvant settings work very effectively, it may be that a very large number of patients never get metastatic disease, which would be great. That would mean that we’re curing them up front, before they ever develop metastatic disease,” Sznol said.
In an interview with OncLive®, Sznol, professor emeritus at Yale School of Medicine in New Haven, Connecticut, and recently appointed faculty at the University of Miami in Florida, unpacked the evolving evidence, the pull between adjuvant and neoadjuvant strategies, and the next wave of investigational therapies that could further shift the survival curve.
If we limit ourselves to cutaneous, mucosal, and acral melanoma, there are only a couple of targeted therapies that really work: BRAF/MEK inhibitors for patients with BRAF mutations, and [that population is only a subset]. Patients who have KIT mutations can respond to KIT inhibitors, but again, that’s a very small subset. The targeted therapies provide some, but not very much, benefit. The long-term, durable responses are related to immune therapy.
Then you go back, and you start asking, what’s the best way to treat a patient who has had their primary tumor removed and is sentinel node positive and is relatively high risk for recurrence? Those patients, if they don’t have a BRAF mutation, would be offered anti–PD-1 therapy. The field has looked at adding anti–CTLA-4 inhibitors to anti–PD-1 inhibitors, and those trials were not better than anti–PD-1 inhibition alone. The field has also looked at nivolumab plus relatlimab-rmbw [Opdualag] vs anti–PD-1 therapy alone, and that [trial] also did not improve outcomes compared with anti–PD-1 therapy alone.
The other complexity about treating adjuvant patients is that it’s not clear that treating patients at that point in time is better than waiting until they progress, because even though you improve progression-free survival [PFS] in the adjuvant setting, so far, none of the data show an improvement in overall survival [OS]. So it’s a very difficult discussion to have with patients when they come [to you] in that setting, [because] sure, we can give you anti–PD-1 therapy as adjuvant treatment, but I’m not sure that it’s going to prolong OS. They could possibly wait, [but] if they take it now, they have that 10% chance of having a severe immune-related adverse effect [AE]. If you have a BRAF mutation, it turns out that you can give BRAF/MEK inhibitors adjuvantly and, strangely enough, the outcome is as good as giving anti–PD-1 therapy as far as we can tell from cross study comparisons. So, if [a patient] has a BRAF mutation, and they’re a candidate for adjuvant therapy, you could offer them BRAF/MEK inhibitors. They’re not as well tolerated, but they’re also not associated with long-term AEs.
There was a small, randomized phase 2 trial that suggested that a neoadjuvant vaccine could add to the activity of pembrolizumab [Keytruda] in the adjuvant setting.1 But we need to wait to see if those data are confirmed in a subsequent phase 3 trial. To add to the complexity, we now go back to the setting where people have active nodal disease. For example, when they come in, either clinically detected by PET or detected by physical exam, those patients are eligible for neoadjuvant therapy. In the neoadjuvant setting, it turns out that probably the combination of ipilimumab [Yervoy] and nivolumab is better than anti–PD-1 therapy alone and probably better than targeted therapies, [but] based on, again, not all completely randomized trials.
The very large investigator-led NADINA study showed that giving 2 cycles of neoadjuvant therapy was better than surgery followed by adjuvant anti–PD-1 therapy, at least in terms of event-free survival, which was the [primary] end point.2 You might say, well, we should wait for survival, and we should, but the differences were so large between neoadjuvant and adjuvant treatment that most people who look at those data would say you should get neoadjuvant therapy in that setting. We may be overtreating a fair number of [individuals], but the results were quite dramatic. So in the neoadjuvant setting, you would choose ipilimumab plus nivolumab, the flip dose ratio for 2 cycles. And it turns out that if you just give those 2 cycles of treatment, and the patient has a near complete or complete pathological response [pCR], they don’t need any additional therapy. Beyond that point, they tend to have very long PFS.
There’s not a right and wrong approach, but there are a lot of investigators who think that we can give neoadjuvant ipilimumab/nivolumab [and] do node resection. In some cases, people only take out the involved node, and if the [patient has] a complete or any complete pathologic response, you can tell the patient the chances that they’ll ever occur from melanoma are very low, and they can go about their normal lives. [From that point,] you just follow them with scans.
If the patient doesn’t have a major response and they have a BRAF mutation, you could then offer them BRAF/MEK inhibitors for a year. We don’t know if that’s better than [observation], but that’s another possibility. In the future, those patients who have a higher risk for relapse could be [randomly assigned] to receive X vs Y in the postneoadjuvant setting to try to reduce their recurrence rates even further. It would be great, for example, if the vaccine worked in that setting; for example, you don’t get a pCR or near pCR, so you have a high risk for relapse. It’d be great if you could just give them a vaccine at that point and reduce their risk for occurrence beyond that point, but we don’t have that data yet.
A number of [approaches] are being studied in the clinic. The randomized phase 2 trial [that evaluated a vaccine] in the adjuvant setting looked very promising, but if I had a nickel for every promising phase 2 trial that didn’t [pan out] in a phase 3 study, I probably would never have to work again. If the [phase 3] study is positive, that will have an enormous impact, because there are a lot of patients who aren’t candidates for neoadjuvant therapy, because they don’t have palpable nodes, but they end up having positive sentinel nodes and would be candidates for adjuvant therapy. Maybe in that setting, [we would see] improved survival, but we don’t know. That could have a huge impact. [In terms of] the oncolytic virus [that’s being evaluated] in the relapsed/refractory setting, [the developers are] trying to get approval on the basis of a large phase 2 cohort. I don’t know whether that would have a great [effect], but if their randomized trial is positive, that would have a huge [effect] in our field, because [it would be used to treat] a group of patients that probably wouldn’t do well after doublet immune therapy.
Then there’s Immatics’ cell therapy, which is a TCR [T-cell receptor] T-cell therapy in which you take T cells out of the body, and you add the TCR for PRAME. It’s only for HLA-A*02:01–positive patients, but if the ongoing randomized trial turns out to be positive, that gives us yet another therapy to treat [patients with] advanced, refractory disease, and that would have a huge [effect] on the field.
Then you could start thinking about moving those therapies up closer, because TCR T-cell therapies, even though they require lymphodepletion, are easier to get than tumor-infiltrating lymphocytes.
The other therapy out there that’s interesting is T-cell engagers. Tebentafusp-tebn [Kimmtrak] has had a major [effect] in uveal melanoma, and there’s a PRAME T-cell engager that’s being studied with anti–PD-1 therapy compared with either anti–PD-1 therapy alone or nivolumab and relatlimab. If that study is positive, that also would have a major impact on the field. A number of things are out there that are cooking that, if they turn out to be positive, would move that survival curve up.
If we’re thinking now that for metastatic disease we’re curing or at least providing very long-term survival for 50% of patients, the goal has always been to move that up to 60%, 70%, or 75%, and if a number of those treatments turn out to be positive in phase 3 trials, we may start to be able to move that survival curve up. The value of neoadjuvant and adjuvant therapies, even though they can have toxicity in that setting, is that those patients have localized disease, so they never develop metastatic disease. When you develop metastatic disease, you can develop it in bad places such as the brain, spine, or other places. There’s a fair bit of morbidity that you would prevent if you could keep those patients from ever developing metastatic disease in the first place.
