KRAS mutations predict response to EGFR inhibitors

Five antiepidermal growth factor receptor therapies have been approved for the treatment of solid tumors. However, response rates are relatively low. Several biomarkers that enrich for patients with tumors most likely to respond to these therapeutic agents have been identified. Mutations in the intermediate signal transduction pathway member KRAS also selects patients with tumors depending on signaling through this pathway. However, because KRAS acts downstream of the EGF receptor, somatic changes in this gene can be used as a marker to exclude patients unlikely to benefit from anti-EGFR therapy. Recent clinical data have provided substantial evidence that KRAS mutational status should be utilized as a diagnostic marker for predicting that response to anti-EGFR therapies in colorectal and non-small cell lung cancer.

Introduction

Tumor cells become dependent on factors that encourage cell growth while inhibiting negative regulatory mechanisms [1]. This dependence may arise as a result of the selective advantages of somatic changes in tumors occurring through a variety of mechanisms at any step of the complex regulatory pathways controlling cell proliferation and survival. Indeed, increased signaling through any pathway can be caused by increasing the amount of ligand binding to the receptor (autocrine or paracrine stimulation), an increase in the numbers of receptors (gene amplification), or by constitutive activation of the receptor or an intermediate pathway member (through somatic gene mutation or translocation). In addition, other mechanisms exist for activating this pathway such as overexpression or mutations in the guanine nucleotide exchange factor family. The epidermal growth factor receptor (EGFR) is a transmembrane tyrosine kinase that signals through at least two parallel intracellular pathways to regulate cellular proliferation and survival [2]. One of these pathways, the MAP kinase pathway (MAPK), regulates the G1 checkpoint and controls cellular proliferation. The MAPK pathway transmits signals from the transmembrane receptor, EGFR, to the nucleus via a series of intermediate genes including RAS, RAF, and MEK (Figure 1) [2, 3]. The EGFR–MAPK signal transduction pathway thus provides an ideal opportunity to dissect the role of individual somatic changes in the tumor to predict the outcome of therapy targeted at the EGFR receptor.

Five drugs have been approved that target EGFR thereby inhibiting the EGFR–MAPK signal transduction pathway. These include two monoclonal antibodies (cetuximab and panitumumab) that act by blocking binding of ligands to the extracellular domain of EGFR, two small molecules that bind to the intracellular kinase domains of EGFR (erlotinib and gefitinib) and one that binds to both EGFR and HER2 (lapatinib). Gefitinib was withdrawn from the U.S. market in 2005 because of phase 3 clinical trial results that showed no survival benefit in the overall non-small cell lung cancer (NSCLC) patient population [4]. However, survival benefit was established in the subgroup of Asian female nonsmokers. Genomic analyses of tumor biopsies from patients enrolled in clinical trials of these anti-EGFR agents and in silico analyses of genomic databases have revealed frequent mutations in the EGFR gene or its downstream signal transduction pathway members KRAS and BRAF (reviewed in [3, 5]), amplification of the EGFR gene [6•, 7] and increased amounts of the EGFR ligands epiregulin and amphiregulin [8••, 9•].

The somatic changes in the EGFR–MAPK pathway are used to predict response to anti-EGFR therapies and have been developed as companion diagnostics to include or exclude patients from treatment with these agents. Mutation of the EGFR kinase domain, amplification of the EGFR gene, or increased expression of the EGFR ligands amphiregulin and epiregulin identify tumors that have become dependent on signaling through this pathway and are likely to benefit from therapy. By contrast, mutations in the KRAS gene, while also identifying dependency on EGFR signaling pathway, suggest that tumors will not benefit from treatment with anti-EGFR agents because they act upstream of the activating mutation in the signal transduction cascade. KRAS is the first example where molecular characterization of a downstream signal transduction pathway member, as opposed to the drug target itself, may be developed as a companion diagnostic to guide the use of all drugs in the anti-EGFR class. Furthermore, this example highlights the importance of developing diagnostic markers that predict both sensitivity and primary resistance to antitumorigenic agents. Analysis of the combined data from the studies included in this review shows that the response rate to anti-EGFR therapy is less than 3% in patients with KRAS mutant tumors as opposed to 35% in colorectal cancer (CRC) and 20% in NSCLC with wild-type KRAS.