Amalgam of Biomarkers to Determine Which Patients Get Checkpoint Inhibitors

Jeffrey S. Weber, MD, PhD


April 03, 2017

This feature requires the newest version of Flash. You can download it here.

Hello. I am Dr Jeffrey Weber. I am the deputy director of the Laura and Isaac Perlmutter Cancer Center at the New York University Langone Medical Center.

Today we are going to discuss biomarkers for the use of checkpoint protein inhibitors. Much of this work has gone on in the melanoma field. I believe it will have enormous applicability for the use of these drugs, which is now becoming much more widespread and used in various histologies, including kidney cancer, head and neck cancer, non-Hodgkin's lymphoma, and non-small cell lung cancer, all of which have achieved FDA drug approvals in the past year or so.

There have been two very important sets of data that have come out over the past 6-8 months. One of them comes from UCLA, the groups of Roger Lo and Tony Ribas, who have done an extensive series of analyses of both pretreatment and on-treatment samples from patients receiving PD-1 inhibitors.[1] They find that there are so-called JAK1 and JAK2 mutations in a surprising number of the tumors that both have innate and adaptive resistance to the use of these drugs. They also find that there are beta-2 microglobulin deletions or mutations that render beta-2 microglobulin inactive, which means that a T cell could not possibly recognize a tumor. These are genomic changes, some of which occur at the onset of treatment, which would be innate resistance; some of which become or occur with the use of these drugs, which would be adaptive resistance. Again, these are things we will have to look out for as markers for resistance at the get-go whenever we use these drugs.

In addition, the group of Lo and Ribas has looked at RNA expression data, with some very interesting findings.[2] They have actually defined what is called an IPRES signature of genes, often largely related to wound healing, the mesenchymal-epithelial transition, and a variety of other genes, some of which are interferon gamma inducible. IPRES, by the way, stands for innate resistance to PD-1 antibody. Here they find that when you look at a variety of genes expressed in tumors prior to the use of these drugs, there are a number of genes that are able to predict resistance. This is very important because, for example, wound-healing genes seem to be very highly represented in this and it is possible that other important inflammatory genes that relate to the gamma interferon pathway may also be highly represented.

A set of groups from MD Anderson, headed by Pam Sharma and Jen Wargo, have also found that not only can they detect pretreatment gene signatures and sets of mutations, but they can also detect on-treatment gene mutations and alterations in genes that seem to be associated with outcome.[3] They find that in patients who express mutations in genes that are in a so-called interferon gamma pathway, where those gene pathways are compromised. Again, no surprise—those patients do not respond well to either ipilimumab or nivolumab. They have looked at both, including sets of ipilimumab-treated patients and then sets of ipilimumab-treated patients who progressed and then were treated with PD-1 antibodies. They found that there are certain common mutations and gene alterations, or alterations and gene expression, in these very important inflammatory and immune pathways.

Some of these have been seen by different groups, which would strongly suggest that these data are very important and that if we could find a way to upregulate these gamma interferon pathway genes, we might be way ahead of the game in treating patients with melanoma and other cancers with PD-1 blockade. There is no doubt at the end of the day that a number of other biomarkers, such as PD-L1 staining or the mutational or neo-antigen load, will also play an important role in predicting who will do well with these drugs and who will not. Unfortunately, there is no single marker today that is very useful in predicting outcome after the use of these important antibodies. In the future, I suspect that it will be an amalgamated biomarker of both host-related (that is, genetic mutations) and tumor-related gene expression alterations that will tell us who should be treated with these drugs and who should not.

Again, this is Dr Jeffrey Weber. Please feel free to call in to give your views and ask questions. Thank you very much for your attention.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: