Copy number loss of KDM5D may be a predictive biomarker for ATR inhibitor treatment in male patients with pulmonary squamous cell carcinoma.

A limited number of patients with lung squamous cell carcinoma (SCC) benefit clinically from molecular targeted drugs because of a lack of targetable driver alterations. We aimed to understand the prevalence and clinical significance of lysine-specific demethylase 5D (KDM5D) copy number loss in SCC and explore its potential as a predictive biomarker for ataxia-telangiectasia and Rad3-related (ATR) inhibitor treatment. We evaluated KDM5D copy number loss in 173 surgically resected SCCs from male patients using fluorescence in situ hybridization. KDM5D copy number loss was detected in 75 of the 173 patients (43%). Genome-wide expression profiles of the transcription start sites (TSSs) were obtained from 17 SCCs, for which the cap analysis of gene expression assay was performed, revealing that upregulated genes in tumors with the KDM5D copy number loss are associated with 'cell cycle', whereas downregulated genes in tumors with KDM5D copy number loss were associated with 'immune response'. Clinicopathologically, SCCs with KDM5D copy number loss were associated with late pathological stage (p = 0.0085) and high stromal content (p = 0.0254). Multiplexed fluorescent immunohistochemistry showed that the number of tumor-infiltrating CD8+ /T-bet+ T cells was lower in SCCs with KDM5D copy number loss than in wild-type tumors. In conclusion, approximately 40% of the male patients with SCC exhibited KDM5D copy number loss. Tumors in patients who show this distinct phenotype can be 'cold tumors', which are characterized by the paucity of tumor T-cell infiltration and usually do not respond to immunotherapy. Thus, they may be candidates for trials with ATR inhibitors.

Comparison of ASCL1, NEUROD1, and POU2F3 expression in surgically resected specimens, paired tissue microarrays, and lymph node metastases in small cell lung carcinoma.

Subtypes of small cell lung carcinoma (SCLC) are defined by the expression of ASCL1, NEUROD1, and POU2F3 markers. The aim of our study was to explore the extent to which the intratumoral heterogeneity of ASCL1, NEUROD1, and POU2F3 may lead to discrepancies in expression of these markers in surgical samples and their matched tissue microarray (TMA) and lymph node (LN) metastatic sites. METHODS AND RESULTS: The cohort included 77 patients with SCLC. Immunohistochemical examinations were performed on whole slides of the primary tumour, paired TMAs, and metastatic LN sites. Samples with H-scores >50 were considered positive. Based on the ASCL1, NEUROD1, and POU2F3 staining pattern, we grouped the tumours as follows: ASCL1-dominant (SCLC-A), NEUROD1-dominant (SCLC-N), ASCL1/NEUROD1 double-negative with POU2F3 expression (SCLC-P), and negative for all three markers (SCLC-I). In whole slides, 40 SCLC-A (52%), 20 SCLC-N (26%), 15 SCLC-P (20%), and two SCLC-I (3%) tumours were identified. Comparisons of TMAs or LN metastatic sites and corresponding surgical specimens showed that positivity for ASCL1, NEUROD1, and POU2F3 in TMAs (all P < 0.0001) or LN metastatic sites (ASCL1, P = 0.0047; NEUROD1, P = 0.0069; POU2F3, P < 0.0001) correlated significantly with that of corresponding surgical specimens. CONCLUSION: The positivity for these markers in TMAs and LN metastatic sites was significantly correlated with that of corresponding surgical specimens, indicating that biopsy specimens could be used to identify molecular subtypes of SCLC in patients.

Comprehensive clinicopathological characteristics and mucin core protein expression profiles of bronchiolar adenoma.

AIMS: Bronchiolar adenoma (BA) is a novel entity in the 2021 WHO classification of lung tumours. The expression profile of mucin core proteins in BAs is not clear. The aim of this study was to clarify the expression profiles of mucins and to validate the clinicopathologic and molecular features of BAs. METHODS AND RESULTS: We examined the clinicopathological, immunohistochemical, and molecular features of 20 BAs. Our cohort comprised 10 proximal and 10 distal BAs. Only seven of 18 patients (39%) were accurately diagnosed with BA at the time of intraoperative consultation. The frequent genetic alterations were BRAF V600E (35%) and KRAS (30%) mutations, which were mutually exclusive. The expression of MUC1, MUC4, and MUC5B was observed in all cases and that of MUC5AC and MUC6 was observed in nine (45%) and five (25%) cases, respectively. MUC4 was diffusely expressed in 18 cases. In contrast, MUC1, MUC5AC, MUC5B, and MUC6 displayed a patchy expression pattern. These expression patterns were similar to that of bronchiolar epithelium in normal lung tissue. In addition, overexpression of MUC1 and MUC4 on the entire cell surface was not observed in any of the BAs, suggesting their benign nature. CONCLUSION: BA commonly exhibits diffuse MUC4 and patchy MUC1 and MUC5B expression. Its unique expression pattern is probably attributed to mucin expression specific to the bronchiolar epithelium. These results confirm the clinicopathologic and molecular characteristics of BA, including the difficulty in intraoperative frozen section diagnosis and the broad morphologic spectrum of BA derived from the bronchiolar epithelium.