The Tell-Tale Gene

Despite all we understand about colorectal cancer, there are still a number of ongoing mysteries. One of these – namely, why up to half of patients receiving cetuximab treatment fail to respond – may be inching closer to a solution. Geoffrey Liu, the Alan B. Brown Chair in Molecular Genomics and an Associate Professor of Medicine, Medical Biophysics, and Epidemiology at the University of Toronto’s Dalla Lana School of Public Health, recently published a study revealing that a polymorphism in the FCGR2A gene can serve as a biomarker to predict which patients may benefit from cetuximab.

An initial trial conducted a decade ago established that the drug was most effective in patients whose tumors exhibited a RAS mutation (1) – but that alone didn’t explain the response pattern. The new research builds on that work by indicating that, in patients with wild-type KRAS, cetuximab extends survival by an average of 5.5 months in those with two FCGR2A H alleles. Patients with one H and one R allele gain only a 2.8-month benefit, and in patients with two R alleles, survival is extended by only 1.6 months (2).

How does FCGR2A function as a biomarker for cetuximab response?

The fragment C gamma receptor (FCGR) is the binding site for immunoglobulins in the antibody-dependent cellular cytotoxicity pathway. Therefore, any monoclonal antibody therapy – including cetuximab – can work through several mechanisms. Although one mechanism is simply that cetuximab “mops” up EGFR ligands, there has been more evidence recently that one of its main modes of action is through antibody-dependent cellular cytotoxicity. Binding assays and other functional assays performed by other groups have suggested that there is a differential binding capability dependent on the FCGR2A polymorphic variant, with the R allele showing significantly less binding capabilities.

Will there soon be a clinical test for the biomarker?

The translation into a clinical test will not be difficult. As a single polymorphism, it can fall into any number of next generation sequencing or tumor panel tests, the same way that we test for somatic mutations in KRAS, EGFR, and other genes.  

How might this change the day-to-day work of pathologists involved in colorectal cancer care?

We anticipate that, if further validated prospectively, the FCGR2A polymorphism will be performed concurrently with RAS mutation testing, the same way that ALK translocations and EGFR mutations are often reflexively tested in metastatic lung cancer patients. Of course, as both RAS and FCGR2A are genomic changes, it will be easier to multiplex the testing in metastatic colorectal cancer patients than it is to use genetic testing for EGFR mutations and IHC/FISH for ALK in lung cancer.

What are your testing recommendations for patients with colorectal cancer?

We don’t recommend testing for FCGR2A yet clinically. RAS testing, however, should be ongoing. We are now in the process of figuring out whether there is an “ideal” patient profile for FCGR2A testing, or whether all colorectal cancer patients should be tested routinely.

What are the next steps for your work?

Now that we have found this new association between FCGR2A and cetuximab response, we will need to test further in additional patient groups to ensure that our marker is useful across ethnicities and other populations. We will also be developing CLIA-certified methods of testing FCGR2A using multiplex technologies that can run RAS mutation testing (KRAS, NRAS, and so on) simultaneously.

Were there any surprises during your research?

We were surprised to find that there were so many challenges in genotyping FCGR3A (which was not significantly associated with clinical outcomes). The presence of a pseudogene required us to be careful in our selection of primers and platforms for testing, and we wonder if some of the prior publications on this other polymorphism might have been confounded by inaccuracies generated by this pseudogene.