Interpreting PD-L1 Status Among Key Melanoma Topics at ASCO

Jason J. Luke, MD, FACP

Key areas in the treatment of melanoma discussed at the 2015 ASCO Annual Meeting included selecting appropriate immunotherapy regimens based on PD-L1 status and improving surgical practices. OncLive sat down with Jason J. Luke, MD, assistant professor of Medicine at University of Chicago Medicine, to provide deeper insight on both topics.

OncLive: Immunotherapy was a hot topic at this year’s ASCO Annual Meeting. Which studies presented in this space did you find the most exciting?

Dr Luke: There was an exciting advancement in combination immunotherapy presented, which was a randomized study [CheckMate-067] of ipilimumab plus nivolumab versus nivolumab or ipilimumab alone. It was determined that there was an increase in progression-free survival (PFS) with the combination, but what was really interesting was that, in a subgroup analysis, it was found that there is a group of patients that we potentially should be treating with nivolumab alone based on PD-L1 staining.

This was an important finding because there are a lot of toxicities with the combination. If we could figure out how to test this, we could give patients who were PD-L1—positive nivolumab alone and give patients who are PD-L1–negative the combination. However, we don’t have the long-term survival data at this point, and this was not incorporated as the original endpoint of the trial, so there is still a lot to learn. If it turns out that long-term survival is much longer, even in the PD-L1–positive patients, we may still give them the combination.

What does that study tell us about PD-L1 as a biomarker?

The clinical trial suggested that patients who were biopsied and shown by immunohistochemistry to have staining for PD-L1, defined as >5% PD-L1 staining in the microenvironment by the Dako antibody (used by Bristol-Myers Squibb only) had improvement with the PD-L1 antibody alone.

This really highlights the complexity of PD-L1. The molecule of PD-L1 has been very difficult to develop immunohistochemical assays for. All of the different companies with drugs have various assays that test different things, so it is not as if a patient who is positive will be positive using another assay. That is very confusing to patients and makes it very difficult. It is also confusing as to how we are going to take this forward, because there is no assay everyone is using.

It is also confusing because PD-L1 is an inducible ligand; therefore, it’s not always there or not there. It gets turned on or off. It is plausible that if you do a tumor biopsy, depending on where you put the needle in the tumor, you might not find the PD-L1, even if it’s there. Because of this, potentially 10% to 15% of patients, who are negative for PD-L1, still have responses to PD-L1 antibodies.

In addition, PD-L1 is really just a part of the description of the microenvironment in the tumor. The model that we are using is known as the T-cell inflamed tumor microenvironment. This suggests that the T cells are able to get into the microenvironment, but they are just not able to complete the job and kill the cancer. In melanoma, we believe that is in approximately 40% of patients, which correlates fairly well with PD-L1, but not exactly. We can test that by doing gene expression profiling, but that is a research test. We are still working on if we can build this in to become a clinical test. While PD-L1 is useful, it is not the whole story. We will need further research to understand this.

Moving to a different area, what did you learn about emerging surgical practices in melanoma at ASCO, and what will it mean for the surgical standard of care?

On the surgical side, there was a lot discussed at ASCO about how much tissue should be removed during surgery.

One study looked at whether a surgical margin of 1 or 3 cm was most appropriate for removing the original melanoma. It was not clear, statistically, if removing more was better than less. The current standard of care for large melanoma is to remove 2 cm and, based on the findings presented at ASCO, this seems to still be appropriate.

Another abstract presented at ASCO looked at lymph node removal and whether we should do a complete lymph node dissection if there is a positive sentinel. That data did not clearly suggest that we should be doing that in all patients. It improves local relapse, but does not improve long-term survival. In the era of modern therapeutics that can fight melanoma, perhaps local recurrence is not as important as it used to be and we need to focus more on long-term survival. We may be removing lymph nodes more often than necessary.

This is an area of interest that requires more research. I don’t think our surgical management techniques will change based on this at the moment, but we may be heading toward a future where we do less surgery and remove fewer lymph nodes.