Asian Pac J Cancer Prev, 15 (17), [7459][7460][7461][7462][7463][7464][7465] Introduction Nasopharyngeal carcinoma (NPC) is a common head and neck malignant neoplasm, especially in southern China where the major pathological type of NPC is nonkeratinizing undifferentiated carcinoma (World Health Organization Type III) (Jemal et al., 2011;Huang et al., 2012). Several factors have been proved to enhance the risk of NPC, including Epstein-Barr virus (EBV), smoking, alcohol consumption, salt-preserved foods and family history of cancer, etc (Chien et al., 2001;Jia et al., 2010;Ji et al., 2011;Hashim et al., 2012).Extensive investigations have showed that the ABO blood group might be associated with risk to certain malignancies (Iodice et al., 2010;Gates et al., 2011;Joh et al., 2012;Wang et al., 2012;Zhang et al., 2014). A previous study suggested that blood type A or AB was associated with an increased risk of NPC and male patients with blood group A had higher rates of distant

Background: Differentiating morphologic features based on hematoxylin-eosin (HE) staining is the most common method to classify pathological subtypes of non-small-cell lung cancer (NSCLC). However, its accuracy and inter-observer reproducibility in pathological diagnosis of poorly differentiated NSCLC remained to be improved. Materials and Methods: We attempted to explore the role of immunohistochemistry (IHC) staining in diagnosing pulmonary squamous cell carcinoma (SQCC) with poorly differentiated features by HE staining or with elevated serum adenocarcinoma-specific tumor markers (AD-TMs). We also compared the difference of epidermal growth factor receptor (EGFR) mutation rate between patients with confirmed SQCC and those with revised pathological subtype. Logistic regression analyses were used to test the association between different factors and diagnostic accuracy. Results: A total of 132 patients who met the eligible criteria and had adequate specimens for IHC confirmation were included. Pathological revised cases in poor differentiated subgroup, biopsy samples and high-level AD-TMs cases were more than those with high/moderate differentiation, surgical specimens and normal-level AD-TMs. Moreover, biopsy sample was a significant factor decreasing diagnostic accuracy of pathological subtype (OR, 4.037; 95% CI 1.446-11.267, p=0.008). Additionally, EGFR mutation rate was higher in patients with pathological diagnostic changes than those with confirmed SQCC (16.7% vs 4.4%, p=0.157). Conclusions: Diagnosis based on HE staining only might cause pathological misinterpretation in NSCLC patients with poor differentiation or high-level AD-TMs, especially those with biopsy samples. HE staining and IHC should be combined as pathological diagnostic standard. The occurrence of EGFR mutations in pulmonary SQCC might be overestimated.