Colorectal cancer has a relatively poor prognosis unless it is identified before it has spread, although the prospects for patients with this disease have recently improved following the development of antibodies targeted to the epidermal growth factor receptor (EGFR).

Two of these, cetuximab and panitumumab, are available as standard therapies.

In almost all cases, however, their use is limited by the development of resistance.

A group of researchers led by Clara Montagut and Joan Albanell from Hospital del Mar, Barcelona, Spain, has now investigated the mechanisms through which colorectal cancer cells develop resistance to cetuximab.  

They began by treating metastatic colorectal cancer cells continuously with the antibody.

This cell line had amplifications in EGFR but was wild type for KRAS (which is wild type in all colorectal tumours that are sensitive to anti-EGFR antibodies) and several other kinases. 

After five months, they were able to isolate three separate clones that were resistant to cetuximab but morphologically similar to the parent tumour cells.  

These clones had no mutations in KRAS or the other kinases that were wild type in the original cell line, and they retained sensitivity to panitumumab.

The researchers developed a hypothesis that an EHFR mutation that affects the epitope that binds cetuximab but not the one that binds panitumumab might be responsible for the introduction of this resistance. 

They therefore sequenced the coding region of the EGFR gene from both cetuximab-resistant and cetuximab-sensitive cells. 

This revealed one base substitution, C1476A, which caused an amino acid change from serine to arginine (S492R).

This mutation from a small polar amino acid to a much larger positively charged one is located in the EGFR ectodomain and so might well be able to interfere with cetuximab binding.

Testing 156 colorectal tumours isolated from patients who had not been exposed to antibody therapy revealed no instances of this mutation. 

Ten further samples were isolated before and after treatment from patients who had initially responded to cetuximab therapy but then relapsed. Sequencing EGFR in these tumour samples revealed that the S492R mutation had arisen in two patients during treatment.

One of these was the C1476A mutation already observed; the other, affecting the same codon, was A1474C. In both cases this appeared to be a somatic mutation, found in tumour tissue only. 

The EGFR protein was found to be over-expressed in both these tumour samples before and after treatment, and neither tumour sample contained mutations in any of the kinases KRAS, BRAF and PIK3CA. 

One of these two patients found to harbour this mutation was deceased by the time it was identified. The other patient was considered appropriate for a trial of monotherapy with panitumumab.

This patient showed a reduction of over 50% in the volume of liver metastases two months after starting panitumumab treatment, and disease progression after five months’ treatment.

This single result indicates that the point mutation identified in EGFR that confers resistance to cetuximab is of clinical relevance, and that identification of this mutation in the tumour of a patient with cetuximab resistance would suggest that that patient might respond to panitumumab. 

This is a fully human antibody, whereas cetuximab is a chimeric mouse-human antibody, and the drugs have different side effect profiles despite being generally considered of equivalent efficacy. 

Montagut, Albanell and their co-workers recommend that patients with colorectal cancer who fail cetuximab therapy should be tested for the S492R mutation, and that it would also be useful to determine whether it also arises in other tumour types that develop cetuximab resistance.

Reference

Montagut, C., Dalmases, A., Bellosillo, B. and 16 others (2012). Identification of a mutation in the extracellular domain of the Epidermal Growth Factor Receptor conferring cetuximab resistance in colorectal cancer. Nature Medicine 18(2), 221-223 doi: 10.1038/nm.2609