Biomarkers for Prediction of Sensitivity to EGFR Inhibitors
Epidermal growth factor receptor (EGFR) Inhibitors have shown promising results in patients with advanced NSCLC who previously have failed on chemotherapy. A major problem is how to select the patients, who will benefit from treatment, and who will not. The predictive role of EGFR protein expression assessed by IHC is still debated. Specific EGFR gene mutations have been identified associated with response to gefitinib (Iressa®), but seem not to be associated with stable disease. Other biomarker studies are described, which are associated with sensitivity to EGFR inhibitors. Increased EGFR gene copy number based on FISH analysis is demonstrated to be a good predictive marker for response, stable disease, time to progression, and survival. However, EGFR mutation is a better predictor of clinical outcome in gefitinib-treated patients than the EGFR gene copy number. EGFR/ FISH seems to be the best predictive iomarker for clinical benefit from EGFR inhibitors in NSCLC. Prospective large scale clinical studies must identify the most optimal paradigm for selection of patients.
Previously, tumor biopsies have been used for EGFR genotyping in NSCLC as it has been difficult to detect the low levels of specific mutations shed from the tumor into the blood against the high background of normal DNA. Testing DNA isolated from blood, rather than tumor tissue, would be better for predicting responses to gefitinib, erlotinib (Tarceva) and other cancer therapies. If EGFR mutations can be observed in serum DNA, this could serve as a noninvasive source of information on the genotype of the original tumor cells as compared to direct sampling of the tumor and could influence treatment and the ability to predict patient response to gefitinib. Patients with EGFR mutations seem to have better outcomes with gefitinib (Iressa) treatment, in terms of progression-free survival, overall survival, and response, than those patients without EGFR mutations.
A phase I, dose-escalation, combination drug study of subjects with NSCLC receiving celecoxib and erlotinib (Tarceva) has identified proteins that may eventually act as biomarkers guiding advanced lung cancer treatment (Krysan et al. 2008). All of these advanced lung cancer patients had previously received conventional treatments without success. About half of the patients had positive outcomes on the therapy defined as tumors that did not grow or that decreased in size by more than 30%. Use of ELISA to analyze patients’ serum protein levels at various time points over 2 months revealed an intriguing link between treatment response and the levels of four proteins: soluble E-cadherin, a matrix metalloprotein (MMP), a tissue inhibitor of MMP, and CCL15. These proteins act downstream of COX-2, an enzyme causing inflammation that can hinder cancer treatment by vascularizing tumors and making them more resilient. COX-2, which is overexpressed in 80-85% of lung cancers also seems to hinder the effect of drugs such as erlotinib, which target EGFR on tumor cell surfaces. Combination therapy with the COX-2 inhibitor cele- coxib is intended to restore tumor cell sensitivity and enhance erlotinib’s efficacy, since only 15% of patients respond to erlotinib alone. In this study, the patients who showed positive outcomes after eight weeks also had lower sEC, TIMP-1, and CCL15 serum levels, while patients with low MMP-9 levels before treatment showed the best treatment results. The latter result indicates the usefulness of MMP-9 as a treatment biomarker for determining the appropriateness of the new combination drug treatment. A larger, phase II, multi-center study will verify their preliminary association between tumor size and protein levels during combination drug treatment.
In 2008, the NIH started a 4-year phase III study, Marker Validation for Erlotinib in Lung Cancer (MARVEL), to validate if choosing patients with NSCLC for Erlotinib based on the presence or absence of EGFR activation has a meaningful benefit over the standard chemotherapy the patients received. The trial (NCT00738881) was terminated.
