Patients with advanced pulmonary adenocarcinoma exhibiting overexpression or mutation of epidermal growth factor receptor tend to respond better to targeted therapy with tyrosine kinase inhibitors such as gefitinib and erlotinib. There is no consensus regarding how these neoplasms should be routinely tested for epidermal growth factor receptor (EGFR) and whether the results of immunohistochemistry (IHC), mutation analysis and fluorescent in situ hybridization correlate with each other or are independent predictive variables. We tested 100 pulmonary adenocarcinomas from patients with stage III or IV disease for EGFR abnormalities using IHC, PCR and fluorescent in situ hybridization (FISH) and compared the results using κ and other statistical methods. The sensitivity of each test to detect an EGFR abnormality and its negative predictive value to estimate the presence of an abnormal test result by the other two methods were calculated. Abnormal EGFR test results were found in 62, 40 and 24% by IHC, FISH and PCR, respectively. κ statistics yielded poor concordance between the results of the EGFR tests (κ=0.3, and 0.2 for IHC and PCR and for PCR and FISH, respectively). Strong membranous immunoreactivity in more than 90% of the tumor cells was found to correlate with amplification or polysomy. PCR when used as a single test is likely to underestimate the presence of EGFR abnormalities that may significantly predict response to tyrosine kinase inhibitors. The need to standardize the approach to EGFR testing in patients with advanced pulmonary adenocarcinoma is discussed.
