Small-cell lung cancer (SCLC) is the most aggressive and lethal subtype of lung cancer, for which, better understandings of its biology are urgently needed. Single-cell sequencing technologies provide an opportunity to profile individual cells within the tumor microenvironment (TME) and investigate their roles in tumorigenic processes. Here, we performed high-precision single-cell transcriptomic analysis of ~5000 individual cells from primary tumors (PTs) and matched normal adjacent tissues (NATs) from 11 SCLC patients, including one patient with both PT and relapsed tumor (RT). The comparison revealed an immunosuppressive landscape of human SCLC. Malignant cells in SCLC tumors exhibited diverse states mainly related to the cell cycle, immune, and hypoxic properties. Our data also revealed the intratumor heterogeneity (ITH) of key transcription factors (TFs) in SCLC and related gene expression patterns and functions. The non-neuroendocrine (non-NE) tumors were correlated with increased inflammatory gene signatures and immune cell infiltrates in SCLC, which contributed to better responses to immune checkpoint inhibitors. These findings indicate a significant heterogeneity of human SCLC, and intensive crosstalk between cancer cells and the TME at single-cell resolution, and thus, set the stage for a better understanding of the biology of SCLC as well as for developing new therapeutics for SCLC.

Neoadjuvant immunochemotherapy (nICT) has dramatically changed the treatment landscape of operable esophageal squamous cell carcinoma (ESCC), but factors influencing tumor response to nICT are not well understood. Here, using single-cell RNA sequencing paired with T cell receptor sequencing, we profile tissues from ESCC patients accepting nICT treatment and characterize the tumor microenvironment context. CXCL13⁺CD8⁺ Tex cells, a subset of exhausted CD8⁺ T cells, are revealed to highly infiltrate in pre-treatment tumors and show prominent progenitor exhaustion phenotype in post-treatment samples from responders. We validate CXCL13⁺CD8⁺ Tex cells as a predictor of improved response to nICT and reveal CXCL13 to potentiate anti-PD-1 efficacy in vivo. Post-treatment tumors from non-responders are enriched for CXCL13⁺CD8⁺ Tex cells with notably remarkable exhaustion phenotype and TNFRSF4⁺CD4⁺ Tregs with activated immunosuppressive function and a significant clone expansion. Several critical markers for therapeutic resistance are also identified, including LRRC15⁺ fibroblasts and SPP1⁺ macrophages, which may recruit Tregs to form an immunosuppressive landscape. Overall, our findings unravel immune features of distinct therapeutic response to nICT treatment, providing a rationale for optimizing individualized neoadjuvant strategy in ESCC.

Pulmonary sarcomatoid carcinoma (PSC) is a rare subtype of lung cancer with poor prognosis. Here, we perform multi-omics analysis of 56 PSC samples, 14 of which are microdissected to analyze intratumoral heterogeneity. We report the mutational landscape of PSC. The epithelial and sarcomatoid components share numerous genomic alterations, indicating a common progenitor. We find that epithelial-mesenchymal transition (EMT) plays important roles in the carcinogenesis of PSC. The pan-cancer analysis reveals high tumor mutation burden and leukocyte fraction of PSC. Integrated molecular classification shows three subgroups with distinct biology, prognosis and potential therapeutic strategies. Actionable mutations are enriched in C1 and C2, patients in C3 have a significantly longer overall survival, and C1 and C2 exhibit T-cell inflamed microenvironments. The three subgroups show molecular similarities to specific subtypes of conventional lung cancer. In conclusion, our study reveals the molecular characteristics and provides entry points for the treatment of PSC.

Primary small cell carcinoma of the esophagus (PSCCE) is a lethal neuroendocrine carcinoma. Previous studies proposed a genetic similarity between PSCCE and esophageal squamous cell carcinoma (ESCC) but provided little evidence for differences in clinical course and neuroendocrine differentiation. We perform whole-exome sequencing, RNA sequencing and immunohistochemistry profiling on 46 PSCCE cases. Integrated analyses enable the discovery of multiple mechanisms ofRB1disruption in 98% (45/46) of cases. The transcriptomic landscape of PSCCE closely resembles small cell lung cancer (SCLC) but differs from ESCC or esophageal adenocarcinoma (EAC). Distinct gene expression patterns regulated byASCL1andNEUROD1define two molecular subtypes, PSCCE-A and PSCCE-N, which are highly similar to SCLC subtypes. A T cell excluded phenotype is widely observed in PSCCE. In conclusion, PSCCE has genomic alterations, transcriptome features and molecular subtyping highly similar to SCLC but distinct from ESCC or EAC. These observations are relevant to oncogenesis mechanisms and therapeutic vulnerability.

Backgroud: Recent studies have indicated that programmed cell death‐ligand 1 (PD‐L1) and cluster of differentiation 47 (CD47) play an essential role in tumor immune evasion and may serve as potential targets for combined immunotherapy. The aim of our study was to evaluate the PD‐L1/CD47 expression status in lung squamous cell carcinoma (LUSC) and lung adenocarcinoma (LUAD), and explore its survival impact and relevance with the immune microenvironment.Methods: The specimens from 190 LUSC and 240 LUAD patients who underwent intent‐to‐treat surgeries were retrospectively collected for immunohistochemistry assays of PD‐L1, CD47, cluster of differentiation 8 (CD8), and cluster of differentiation 68 (CD68).Results: A total of 96 (22.3%) and 296 (68.8%) cases were positive for PD‐L1 and CD47 expression, respectively, and 80 (18.6%) of them demonstrated the co‐expression of PD‐L1/CD47. The rate of PD‐L1/CD47 co‐expression was 23.7% in LUSC, significantly higher than the 14.6% in LUAD (p = 0.018). The median overall survival (OS) for all patients was 55.9 months (range 2.0–146.0 months). The univariate analysis showed that patients with positive CD47 expression (LUSC p = 0.003, LUAD p = 0.036) and PD‐L1/CD47 co‐expression (LUSC p = 0.023, LUAD p = 0.004) exhibited significantly worse prognosis. The multivariate analysis demonstrated that PD‐L1/CD47 co‐expression was an independent prognostic factor for OS (LUSC hazard ratio [HR] 1.922, 95% CI 1.245–2.969, p = 0.003; LUAD HR 1.549, 95% CI 1.015–2.364, p = 0.043). PD‐L1/CD47 co‐expression was associated with high CD8‐positive T‐lymphocyte density in LUSC (p = 0.004) and LUAD (p = 0.043), and with high CD68‐positive macrophage density in LUSC (p = 0.026).Conclusions: PD‐L1/CD47 co‐expression was an independent prognostic factor for LUSC and LUAD patients and may serve as a potential predictive biomarker for combined dual‐targeting immunotherapy.