伴有SMARCA4缺失的非小细胞肺癌9例临床病理分析

BACKGROUND AND OBJECTIVE: SMARCA4-deficient non-small cell lung cancer (SMARCA4-dNSCLC) is a rare primary lung malignancy. These diseases are not listed separately in the 2021 World Health Organization (WHO) classification of lung neoplasms, but they have special morphological, immunophenotypic and molecular genetic characteristics. This study aims to improve understanding of SMARCA4-dNSCLC by discussing the clinicopathological features, diagonosis and differential diagnosis of the disease. METHODS: The clinical and imaging data of 9 cases of SMARCA4-dNSCLC diagnosed in Shanghai Changhai Hospital from January 2020 to March 2022 were collected. The clinicopathological features were analyzed by histological and immunohistochemical staining, and the literature was reviewed. RESULTS: The median age of 9 patients was 65 years old. Six men were smokers. The average maximum diameter of tumor was 3.3 cm. Six cases had been metastasized. The imaging showed that it was an infiltrating mass with unclear boundary and 3 cases invaded the pleura. Nine cases were diagnosed as SMARCA4-dNSCLC, which mainly showed three pathological forms including classic lung adenocarcinoma, mucinous adenocarcinoma and poorly differentiated carcinoma. Poorly differentiated tumor cells are epithelioid, syncytial or rhabdomyoid, the cytoplasm was rich, the cytoplasm could be completely transparent to eosinophilic, eosinophilic globules or small abscesses could be seen, showing solid flakes, with more inflammatory cells and flake necrosis in the stroma. Immunohistochemistry showed that SMARCA4 was negative in all cases and eight cases demonstrated cytokeratin 5.2 (CAM5.2) and cytokeratin 7 (CK7) was diffusely strongly positive, p40 was negative, thyroid transcription factor-1 (TTF-1) was negative in 6 cases, partially positive in 2 cases and diffusely positive in 1 case. CONCLUSION: SMARCA4-dNSCLC is a rare subtype of lung cancer with complex and diverse pathological morphology. The characteristic of immunohistochemical phenotype can assist in the diagnosis.