Activity Transcript

Hussein Tawbi, MD, PhD: Hello, my name is Hussein Tawbi. I'm professor and deputy chair of the Department Of Melanoma Medical Oncology and co-director of the Brain Metastasis Clinic at the University of Texas MD Anderson Cancer Center in Houston, Texas. And I would like to welcome you to this program titled "Depicting LAG3: Exploring a New Target in Metastatic Melanoma."

We're going to discuss the tumor microenvironment for metastatic melanoma, the characteristics, role, and importantly, function of lymphocyte activation gene 3 (LAG3) in metastatic melanoma and compare it to other immune checkpoint inhibitors. And then discuss how this is being taken up in the treatment armamentarium of metastatic melanoma.

We're all familiar with the 2 checkpoint inhibitors that have been most useful in the management of cancer in general and melanoma specifically. And those are cytotoxic T-lymphocyte associated protein 4 (CTLA-4) and programmed cell death protein 1 (PD-1). We know that CTLA-4 is closely associated with T-cell activation and it's a signal that comes up soon after recognition of the antigen by the T-cell receptor and leads to activation of T-cells. And we know that blocking CTLA-4 is capable of increasing antigen recognition and generating an immune host response. We also know that PD-1 is expressed on the surface of effector T-cells, specifically as a marker of activation and later as a marker of exhaustion. And blocking PD-1 can allow for a more efficient recognition of the antigen on tumor cells and more killing of tumor cells.

We know that those drugs are independently active together when we block PD-1 and block CTLA-4 and we know that the combination is more effective than either drug alone. And if you ever wonder how important these 2 checkpoints are, you only need to ask these 2 guys who won a Nobel Prize in 2018 in physiology and medicine for the discovery of CTLA-4 and PD-1. Obviously, discoveries that have changed the landscape of cancer in general, but more specifically metastatic melanoma.

In the last 10 years, we've been really looking consistently at the results of randomized phase 3 trials looking at a single agent PD-1, single agent ipilimumab, and then CheckMate 067 was the landmark phase 3 clinical trial that compared the combination of anti-CTLA-4 and anti-PD-1, nivolumab and ipilimumab to either drug alone. And we now know with 7.5 years of follow-up and kind of longer for ipilimumab, we know that the overall survival (OS) has now a median of 72 months for patients with metastatic melanoma when they're treated with the combination of ipilimumab and nivolumab. We know that each drug alone can cure those patients, but to a lesser degree. So nivolumab, 42% long-term survival, and ipilimumab about 23%.

We also know that when we have to use these drugs in combination, we actually have to contend with a much higher level of toxicity. And you can see on the slide the organs that get involved with immune-mediated toxicity. We have an idea of the pattern of it and when each of the toxicities occur. But in terms of incidence, again, single-agent PD-1 has generally a toxicity rate of about 15 to 20%. Single-agent ipilimumab probably closer to 25% to 30%. But when you put them both together, the toxicity can go up to 59%. So, it's really high.

What we also know about the combination there is that there's no statistically improved survival over single-agent PD-1. So, we know that the combination of checkpoint inhibitors does have improved survival, but it's not statistically significantly better than single agent PD-1. And we also know that the combination has activity in very specific settings and is preferred as a treatment strategy, like in brain metastases and neoadjuvant settings or in rare melanoma subtypes. And we are really pleased to see that other malignancies are starting to benefit from this, including non-small cell lung cancer, renal cell carcinoma, microsatellite instability (MSI)-high colorectal cancer. Again, as I mentioned, it all comes at the cost of increased toxicity.

So, the search has been ongoing now for over a decade looking at all the potential checkpoints and other ways that we can modulate the immune response. There have been many that have been described and multiple agents that are being considered to be brought to the clinic and to try to improve the outcomes of patients with metastatic melanoma. And this is a non-exhaustive list that you see here. That is both the checkpoints on the 1 side and their ligands on the other. And what we're going to focus on today is LAG3.

LAG3 specifically is a marker of T-cell exhaustion. We know that it is expressed and usually co-expressed on T-cells as they get more exhausted: first PD-1, and then with chronic antigen stimulation, then you get LAG3 and other markers. And as you get more and more checkpoints expressed on these T-cells, they become less effective, less able to generate cytokine, less able to kill tumor cells as well.

When you look at LAG3 there is a significant sequence homology with CD4 and what is known is that the ligand for LAG3 is major histocompatibility complex (MHC) Class II. We also know that there's a proline rich loop at the end of LAG3 that actually allows it to have more affinity towards MHC Class II. And that has been considered the primary mechanism of action.

However, we started to learn in the last few years that there are other ligands that are important for LAG3 activity and function, including FGL-1, galectin-3, LSECtin, and even α-synuclein. And these don't necessarily associate with the proline-rich loop and some of them are in different locations of the LAG3 protein.

What we also have learned is that LAG3 dimerization is required for optimal binding to MHC Class II and more importantly the intracellular signaling for LAG3 seems to be very different from other inhibitory receptors. It doesn't have an ITIM motif. It has a very unique KIEELE motif.

And what we've learned over time is what happens is it actually co-localizes with the T cell receptor (TCR)-CD3 complex and reacts stoichiometrically up to 13 to 1. And what that does is that ends up basically taking away the zinc, which is a very important ligand and an important co-effector with ZAP70. That's what actually leads to TCR activation and TCR signaling intracellular. So, as LAG3 sequesters zinc, TCR signaling is significantly inhibited.

So, to summarize, we really understand LAG3 function now as being able to modulate the TCR signaling. And that's very important because that actually is the basis of why we think that LAG3 as a single agent may not be able to be very effective. That was born out in preclinical models where we saw that single-agent LAG3 was not capable of inducing too much tumor regression. But when you combine it with PD-1 you get a much more effective combination both in preclinical models and we've seen it in the clinic as well.

In the first phase 1 trials of relatlimab as an anti-LAG3 antibody it became quite clear that single-agent LAG3 doesn't have clinical activity. Even translational studies showed that it wasn't reactivating the immune system. It wasn't leading to increased CD8 effector memory cells or increased cytokine production.

Then, when nivolumab was added to it, the anti-PD-1, then you started seeing all of these things happening where you have more activation of CD8-positive T cells, more cytokine production, and eventually more activity. So, we started seeing combination activity in the second-line setting. RELATIVITY-020 was the study that did that, where patients with PD-1 refractory melanoma were treated with a combination of nivolumab and relatlimab. And there were some really consistent responses that were deep, durable, very consistent with immune-mediated responses. But they were limited in number. They were in the range of about 13%. If you were LAG3 positive, there was a higher rate of about 18%. But, again, relatively modest activity.

When nivolumab and relatlimab as a fixed-dose infusion was brought to the first-line setting in RELATIVITY-047, this was a randomized controlled trial that compared that combination to single agent PD-1 in treatment-naive patients with metastatic melanoma. And after 19 months of follow-up, what we've seen is a significant improvement in the progression-free survival (PFS) with a hazard ratio (HR) of 0.78 in favor of nivolumab and relatlimab, a median PFS of 10 months vs 4.6 months. And that was done with a blinded independent radiology review.

And when we looked at the OS, which was a secondary endpoint, we saw a decrease in the rate of progression or death by 20%. So, an HR of 0.8, but it was not statistically significant. But you see that there is separation going on all the way to 2 years. And then the response rates were 43% compared to 32% for nivolumab alone.

Importantly, we mentioned the toxicity patterns in the combination of ipilimumab and nivolumab. The toxicity pattern for nivolumab and relatlimab was consistent with single agent PD-1 in terms of profile, although it was a higher incidence, about 21% compared to 11% for single-agent nivolumab.

And in terms of the specific toxicities, there was not a single toxicity that emerged as more significant, more common, or more concerning than single-agent PD-1. So, very similar toxicity profile with a modestly higher incidence.

So, this begs the question: why does the combination of nivolumab and relatlimab have much more significant activity in the first line, whereas we saw it not being able to effect too many responses in the second line? How can it beat single agent PD-1 head-to-head in a phase 3 trial in the first line?

The answer may be related to the mechanism of action. As we discussed, LAG3 eventually works on modulating TCR signaling. If there's not a lot of TCR signaling happening, the amplification of a small signal is going to result in a small signal anyway. But when you're in less immunosuppressed conditions, when you're in a place where TCR signaling is still relatively strong, then when you amplify that, you're likely to have a much higher impact with LAG3 modulation. And I think that's what resulted in us seeing the activity in the first line.

And if that is true, then we would expect that if we take this into even earlier settings for melanoma, then we might actually be able to see even better outcomes. And that was true enough as we showed in a phase 2 trial of neoadjuvant therapy of nivolumab and relatlimab in patients with only local regionally advanced melanoma: so resectable patients, again, usually less immune suppressed patients, it's not already widely metastatic. And what we saw is a pathologic complete response (pCR) rate of 57% -- almost exactly what we see with combination of ipilimumab and nivolumab.

And in this particular study that was published in Nature, we also didn't see any toxicity in the first 2 cycles of treatment preceding surgery. Again, highlighting the fact that this combination has the value of increased activity and decreased toxicity.

Now, I must say this is a really exciting target, LAG3. And there are so many different ways to target it from a LAG3 immunoglobulin agonist that we've seen some results at the Society for Immunotherapy in Cancer (SITC) to different antibodies to the LAG3 receptor and even bispecific antibodies that are being considered.

And one is the cemiplimab and fianlimab combination, which we've seen results from recently at the European Society for Medical Oncology (ESMO). When you look at it in patients that were PD-1 naive, the response rates were quite high, were durable, and you see the median PFS was up to 24 months. It's really hard to say if this is going to be necessarily a better efficacy than the currently existing LAG-3. But it's really interesting to see what that will show in a phase 3 trial. When they looked at patients that are PD-1 refractory, it seemed that the results were very consistent with what we've seen with relatlimab and nivolumab. So, it'll be really interesting to see what those results are going to look like.

In summary, LAG3 is a novel checkpoint. It has a unique mechanism of immune suppression and its blockade already is validated as a therapeutic target in combination for metastatic melanoma. It's FDA approved, but I think it starts an entirely new era of investigation.

We have to understand how LAG3 works. What is its structure? What is its function? How does it work in different settings like in brain metastases or even with other cancers? How else can we use it in combination with other checkpoints or other immune modulators? And importantly, are we going to be able to develop biomarkers that will help us predict response and resistance?

So, it really highlights to me the importance and the value of really a mechanism-driven evaluation of these drugs in combination with each other. And again, we're really excited for our patients because they now have an option that can improve their outcome without adding too much toxicity.

Thank you very much for participating in this activity. Please continue on to answer the questions that follow and complete the evaluation. Thank you.

This transcript has not been copyedited.