Michael A. Postow, MD
Combination therapy continues to propel the treatment landscape for patients with melanoma in a positive direction, according to Michael A. Postow, MD, who adds that an area of interest is the combination of BRAF and MEK inhibitors with PD-1 therapy as well as combinations of LAG-3 inhibition plus PD-1 agents.
“The world is moving to 2- and 3-drug combinations and, if those show benefit in randomized studies, we will all be happy about that,” said Postow. “We need to keep in mind that we should be conservative with our thinking until we have the data to support what could be more expensive and toxic regimens.”
In an interview during the 2017 OncLive®
State of the Science SummitTM
on Melanoma, Postow, a medical oncologist at Memorial Sloan Kettering Cancer Center, discusses the evolution of combination therapies for patients with melanoma.
OncLive®: Please provide an overview of your presentation on combination therapies for patients with melanoma.
: There are many new combination therapies in patients with melanoma. For patients with BRAF-mutant melanoma, we are seeing combinations of BRAF and MEK inhibitors with immunotherapy drugs that are moving into randomized phase III studies. We will know if something improves when you add immunotherapy to BRAF and MEK inhibitors. Outside of that targeted therapy world for patients with BRAF-mutant melanoma, we are seeing new immuno-oncology combination agents.
I discussed a few of those combinations that are moving forward in clinical trials, such as IDO blockers plus PD-1 agents. Also, a new immune checkpoint called LAG-3 has some data in combination with PD-1 blockade in patients had progressed on prior PD-1 therapies. That is interesting to know—that you can block LAG-3 after someone has progressed on PD-1 agents and there can be efficacy, at least in combination, with ongoing PD-1 therapy in some of those patients.
We are seeing efficacy now in a number of earlier-phase studies with different types of agents. We are still waiting on randomized data to be certain that one of those approaches is better than another. We are hopeful that some of these approaches will be.
Can you go into more detail with the LAG-3 trial and how it is seen to be tolerated?
The LAG-3 agent now has a name called relatlimab. It was combined with nivolumab, a PD-1 drug, in this study. This is a study of patients with advanced melanoma who progressed on prior PD-1 or PD-L1 therapy. They may have had an initial response to PD-1/PD-L1, but all of them must have had defined progression on PD-1. We saw a response rate of 12% in all-comer patients. It was a higher response rate in patients who expressed LAG-3 in their tumor microenvironment.
We may start seeing the emergence of a LAG-3 biomarker in the tumor microenvironment. We are trying to figure out what it means to be LAG-3–positive. What does that say about a tumor microenvironment? However, it is interesting that there might be a biomarker for a new checkpoint inhibitor in development. We are waiting on randomized studies to see if LAG-3 improves outcomes compared with PD-1 alone. Obviously, we are interested in patients who have not progressed on PD-1 and determining how effective LAG-3 can be in combination to PD-1. At least it is something good for our PD-1–refractory patients.
Can you discuss ongoing combinations with talimogene laherparepvec (T-VEC; Imlygic)?
Oncolytic viral therapies, meaning injecting a tumor site with a virus and adding it to checkpoint blockade, is something that we are also excited about if we have patients with lymph nodes or tumors in the skin that we can inject. When T-VEC was combined with ipilimumab in a randomized study, it showed that it improved outcomes compared with ipilimumab alone. Many say that if you are going to be giving ipilimumab and you have an injectable tumor site, such as a lymph node or subcutaneous metastasis, you may as well give T-VEC with ipilimumab if that can be done. With PD-1, we do not know whether adding T-VEC to PD-1 improves outcomes over PD-1 alone.
The nonrandomized data look promising with T-VEC plus PD-1. However, the groups of patients in that study are a different group than those who are treated in most clinical trials because these are all patients with injectable disease. It is hard to compare nonrandomized T-VEC data with PD-1 compared with other studies. We are very eager and are currently recruiting patients to phase III studies of T-VEC with PD-1 agents. Hopefully, that will show benefit in a randomized study.
Are we getting to a point where we investigate triplets or quadruplet combinations?
We are certainly moving towards more and more drugs in these combinations. We must remember that need to demonstrate benefits in randomized studies before we get too excited about building triplets and quadruplets. It will be interesting to see but, from a cost of care and toxicity perspective, it will be important to be convinced that more is better.
Also, we need think about de-escalating patients from certain combinations. Perhaps we need to give a few doses of a 2- or 3-drug combination to patients and, if they start to respond, pare down one or 2 of those drugs. That is an open question that we are starting to explore with ipilimumab and nivolumab in combination, but it is not something that has been flushed out for a lot of these new targets.
Is pseudoprogression something that we see commonly in melanoma with some of these checkpoint inhibitors?
Pseudoprogression can occur in patients with melanoma. We think about it mostly in patients who are treated with immunotherapy drugs. Pseudoprogression comes in a lot of different varieties. It is important to keep in mind what kind of tumor growth we are talking about when we discuss pseudoprogression.
I would emphasize that most patients for whom you believe are really having progressive disease on immunotherapy, will unfortunately continue to progress. I tell oncologists and patients that if the tumors are all really growing, it is unlikely that we will get a later response if it initially isn't showing some favorable effects. Pseudoprogression, in many patients, is something we see if we have a mixed response. Some tumors are disappearing, but a few others are growing, so we may have impressive responses in all of the existing disease at baseline. One thing I would try to offer those patients is new treatments sooner rather than too late.
What biomarker research is being done to determine who is best fit to receive these therapies?
We wish we knew exactly who we had to treat with each individual drug. The only biomarker we have in advanced melanoma still is the BRAF
mutation, which says a patient could be a candidate for BRAF/MEK combination therapy.
PD-L1 is another biomarker that has been extensively explored in patients with melanoma. There is a lot of debate on if we should use it for certain decision-making issues and there is controversy about that.
At the moment, the rest of biomarkers are all research. It is important that we do research in this area.