MET TARGETING IN NSCLC

This is Mark Kris, from Memorial Sloan Kettering Cancer Center in New York City, speaking about 2 abstracts that were presented in the oral sessions on lung cancer at the American Society of Clinical Oncology (ASCO) meeting this year. These studies were of 2 drugs that target MET^[1,2]:Onartuzumab is an antibody targeting the MET protein, and crizotinib -- our old friend, our utility in-fielder of oncologic drugs -- which was originally developed to target MET. The patients in these studies were those with amplification of the MET gene.

The study looking at crizotinib^[1] in patients with amplification ofMET was quite encouraging. But the study of onartuzumab,^[2] a drug that appeared to have promise several years ago, was very discouraging. The drug demonstrated no benefit, and for the cohort of patients with epidermal growth factor receptor (EGFR)-mutant tumors, not only no benefit, but perhaps even harm.

What can we learn from these 2 different tales? The first lesson is to be very careful to specifically define the populations of patients to whom we give a targeted therapy. The EGFR story is the best one. Although virtually all patients with lung cancers have expression of EGFR protein that can be measured and quantitated, those cancer cells do not depend on EGFR for their survival. Whereas the benefits can be either nonexistent or small, when patients have those cells that are dependent, the benefits are dramatic.

That was the EGFR story. Patients with EGFR gene mutations in their tumors had dramatic responses. Those who did not have the gene mutation do not have response. By the way, that result was shown again in trials of EGFR mutation-specific drugs. For example, dacomitinib in unselected patients gives only very modest or negative results.^[3]

There is no such thing as "MET-positive lung cancer." MET can be activated, MET can be found in lung cancer cells, and those cells can have different degrees of dependence on MET. MET protein can be found in many cells, and approximately one half or more of the participants in the trial of onartuzumab had MET as determined by an immunohistochemistry test. Many fewer cells have amplification of theMET gene -- multiple copies of the MET gene, not just copies of the whole gene, but copies of MET.

Finally, there can be mutations in MET. That was not discussed at the meeting.

Using onartuzumab (a MET protein-targeted gene) to target the MET protein-overexpressing cells as identified by an immunohistochemistry test was not successful. Using crizotinib was successful for those patients whose cancers had amplification of MET, and for the highly amplified cases, the results were similar to those we see with either ALK- or ROS-positive lung cancers.

Immunohistochemistry Testing May Not Be Up to the Job

The lesson is to be very careful in definition of MET activation. Know exactly what you are looking at. Those of us who did not grow up with molecular biology as a standard course have to learn this. There is a clear difference between "MET-amplified" and "MET protein overexpression."

Finally, we have to again question the ability of immunohistochemistry tests to clearly define populations. We saw here that the use of this test did not select a group of patients that would be sensitive to this MET inhibitor therapy.

Also at ASCO, we saw that in a study of necitumumab, an EGFR-targeted agent, an immunohistochemistry test that was used to show EGFR protein over-expression was insufficient to allow better selection of that agent.^[4]

In summary, pay attention to the biology that underlies the cancer cell. By understanding that biology, we can then choose the right therapy. The mere presence of a protein does not prove dependency of cancer cells on that pathway -- a dependency that can be exploited by a drug. However, when you find those dependencies -- in this case, MET amplification -- we have a drug, crizotinib, that is available today. With the growth of next-generation testing, and on the basis of the study by Camidge and colleagues,^[1] we will be able to find those cancer cells that are MET-amplified, and those are the ones we want to target with crizotinib.