HER2+ NSCLC

Hello. This is Mark Kris from Memorial Sloan Kettering. I would like to speak today about a new topic in the treatment of lung cancers, and that is lung cancers that have dependency on HER2. One thing that we have learned, and we have still a lot to learn in this area, is that there is no such thing as HER2-positive lung cancers. Each lung cancer has a different mechanism of HER2 dependency, and ultimately research will show that how we treat these cancers depends tremendously on the specific type of dependency, and in the case of mutations, the specific mutation that we find.

Probably the most that has been published about HER2 mutations at this point comes from the Lung Cancer Mutation Consortium (LCMC).^[1] We looked for many, although not all, HER2 mutations. We found an incidence of HER2 mutation of approximately 3% in the LCMC. We also found that median survival was 1.6 years from the date of diagnosis of metastatic disease. That is a slightly lower survival than we see with KRAS-mutant lung cancers and much lower than with ALK-positive or EGFR-mutant lung cancers, which have median survivals in the 4-year range.

The other observation that some of the other, older-generation oncologists have had experience with—and I will count myself in that group—is the use of testing by immunohistochemistry for the presence of HER2 protein. We found two things. Number one, the presence of high-level protein expression (2+ or 3+) is quite rare. It is not always associated with amplification, as we have seen in breast cancers. Also, the use of trastuzumab (Herceptin®), either alone or with chemotherapy, was generally not helpful in this group, even in those with a very high level of amplification. There are a few cases in whom amplification may be an important criterion, and the jury is still out on that.

We were part of a project with the University of Colorado in which we looked for amplification in cases in whom we also looked at mutation.^[2] We found the same rate of mutation in HER2 amplification—approximately 3%. In these cases, we found that it did not overlap. The other interesting thing about amplification is that our amplification was very low-level, usually between two and three, whereas in breast cancers it is much higher.

What have we learned about treating HER2-mutant lung cancers? First, we have only detected a small number of them. Using the LCMC as a model and our standard snapshot or Sequenom panels, we did not include all of the HER2 mutations nor did we include the extracellular domain mutations, both the exon-20 insertions and the extracellular domain mutations. There are more mutations out there. The nice thing about multiplex platforms is that they allow us to find these mutations, and they will also allow us to precisely identify them.

There have been reports of benefit in patients with mutations, with, for example, trastuzumab and afatinib (Gilotrif®). If you find a patient with a HER2 mutation today, in the absence of a clinical trial, the go-to drug would be afatinib. I would probably urge the same kind of schedule used in EGFR-mutant cancers. I would start with a dose of 30 mg daily. The role of trastuzumab is uncertain. At our institution we are about to begin a clinical trial of ado-trastuzumab emtansine (Kadcyla®). It makes sense to use that drug in these situations, but I am unaware of any published data about it.

When we looked at our patients with HER2-mutant lung cancers and we looked at what treatments they had received, like other patients with non-adenocarcinomas in whom no allogeneic drivers were identified, chemotherapy was the mainstay of their care. They can have long and durable remissions with chemotherapy. In cases where you find HER2 dependency, chemotherapy remains the treatment of choice. In those with HER2 mutations, it makes a lot of sense to try afatinib as well. We have to look to research to better identify these patients through our multiplex testing panels, and to test new agents in these areas.