[Influence of Immunohistochemistry Scoring Criteria in Detecting EGFR Mutations].

BACKGROUND:Appropriate immunohistochemistry (IHC) scoring criteria can guarantee the reliability of mutation detection results. Most current studies suggest that the "four-grade criterion" may be the best among all scoring systems. The aim of this study is to discuss the influence of different four-grade immunohistochemistry scoring criteria on the test results. METHODS:Three different four-grade immunohistochemistry scoring criteria were respectively used to evaluate the EGFR status of 83 cases of non-small cell of lung cancer (NSCLC) samples. The sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), the agreement of each criterion compared with the gold standard, and whether statistical significance exists between each criterion were calculated. RESULTS:No statistical difference was found among the three criteria in detecting EGFR mutations. All three criteria exhibit considerably better specificity than sensitivity. For samples with scores of "3+", PPV could reach up to 100%. CONCLUSIONS:No definite best criterion is among different four-grade scoring criteria. Regardless of the kind of criterion used, the specificity of IHC method in detecting EGFR mutations is markedly better than the sensitivity. For samples with scores of "3+", mutation status can be confirmed, and samples can receive EGFR-TKI therapy directly.