by ecancer reporter Clare Sansom

The prognosis for patients diagnosed with colorectal cancer has improved significantly in the past thirty years, mainly due to improvements in the treatment options available.

One of the most important advances has been the development of monoclonal antibodies directed against the EGFR receptor as therapy for advanced disease.

However, these drugs are only effective in the subset of about 60-65% of colorectal cancer patients whose tumours do not contain mutations in the gene KRAS.

Even in these patients, resistance to anti-EGFR antibodies invariably emerges months after the therapy begins.

The molecular mechanisms through which resistance to EGFR-directed antibody therapy develops in these patients have until very recently been poorly understood.

Now, however, two papers published in the same issue of Nature have implicated the emergence of new mutations in the KRAS gene in the acquisition of this resistance.

In the first study, led by Alberto Bardelli at the Institute for Cancer Research and Treatment, Turin, Italy, colorectal cancer cell lines that had wild-type KRAS and were known to be sensitive to anti-EGFR antibodies were treated continuously with cetuximab [1].

When, as expected, cetuximab-resistant variants emerged, both parental and resistant cell lines were tested for mutations and copy number changes.  Observed increases in KRAS activation in the resistant cell lines were accompanied by both amplification of and point mutations in this gene.

The researchers then examined the parental cells for changes to KRAS and found that a very small proportion of these cells contained some of these changes, although, interestingly, one mutation – G12R – was observed in resistant cells. These and other results indicated that resistance could be acquired both through the selection of KRAS mutant clones and through the development of new mutations.

Deep sequencing of tumour cells from colorectal cancer patients who had developed resistance to an anti-EGFR antibody using the 454 and BEAMing technologies also showed that many had developed mutations in KRAS; these were not present in samples taken from the same patients pre-treatment or in tumours from patients treated with cytotoxic chemotherapy only.

Furthermore, these mutations could be found in blood samples taken from patients many months before disease progression was observed using the standard method of radiography.

The second study, led by Luis Diaz from Johns Hopkins Kimmel Cancer Center, Baltimore, Maryland, USA, involved the retrospective genetic analysis of serum taken from 24 patients with advanced colorectal cancer without initial KRAS mutations who had been treated with the anti-EGFR antibody panitumumab [2].

Serum samples from four patients with KRAS mutant tumours were used as controls; as expected, none of these achieved disease control with panitumumab. Samples were obtained from the study patients before treatment initiation and every four weeks until disease progression.

Nine (38%) of the 24 study patients with initial wild-type KRAS were found to develop detectable mutations in this gene during therapy; there were no significant differences in progression-free or overall survival between patients who developed mutations and those who did not. Generally, mutations were first observed 5-6 months after therapy initiation, which was before disease progression could be detected using radiography. Two patients developed two different KRAS mutants during the course of treatment, and one developed four.

Diaz and co-workers then used this longitudinal data to develop a mathematical model of the evolutionary processes in the tumours. They estimated growth rates for the tumours and used a branching process model to test the null hypothesis that there were no KRAS-mutant cells in the tumours before treatment initiation.

Each tumour that developed mutations was modeled separately, and the results showed that all tumours would have almost certainly included at least one cell containing a mutation before treatment began. This further suggested that tumours observed to have no KRAS mutations are likely to contain small sub-clones of cells that do, and that resistance to anti-EGFR antibodies can arise from the selective expansion of these sub-clones.

Taken together, the results from both these studies indicate that mutations in KRAS arising in tumours assessed as wild-type for this gene arise during therapy with anti-EGFR antibodies, leading to resistance. These mutations are detectable months before tumour progression is observed and targeting the KRAS pathway with, for example, an inhibitor of the kinase MEK may lead to a delay in or even a reversal of this resistance.

References

[1] Misale, S., Yaeger, R., Hobor, S. and 23 others (2012). Emergence of KRAS mutations and acquired resistance to anti-EGFR therapy in colorectal cancer. Nature, published online ahead of print 14 June 2012. doi:10.1038/nature11156

[2] Diaz, L.A. Jr., Williams, R.T., Wu, J. and 10 others (2012). The molecular evolution of acquired resistance to targeted EGFR blockade in colorectal cancers. Nature, published online ahead of print 14 June 2012. doi:10.1038/nature11219