Recent sequencing analyses have shed light on heterogeneous patterns of genomic aberrations in human gastric cancers (GCs). To explore how individual genetic events translate into cancer phenotypes, we established a biological library consisting of genetically engineered gastric organoids carrying various GC mutations and 37 patient-derived organoid lines, including rare genomically stable GCs. Phenotype analyses of GC organoids revealed divergent genetic and epigenetic routes to gain Wnt and R-spondin niche independency. An unbiased phenotype-based genetic screening identified a significant association between CDH1/TP53 compound mutations and the R-spondin independency that was functionally validated by CRISPR-based knockout. Xenografting of GC organoids further established the feasibility of Wnt-targeting therapy for Wnt-dependent GCs. Our results collectively demonstrate that multifaceted genetic abnormalities render human GCs independent of the stem cell niche and highlight the validity of the genotype-phenotype screening strategy in gaining deeper understanding of human cancers.

No abstract

BACKGROUND AND AIMS: Diffuse-type gastric cancer (GC) is currently subdivided into signet-ring cell carcinoma (SRCC) and non-SRCC, referred to as poorly cohesive carcinoma not otherwise specified (PCC-NOS). Although these subtypes are considered to be independent, they often coexist in the same tumors, raising a question of whether they clonally differ or not. To tackle this question, we established an experimental platform for human diffuse GC that enables accurate modeling of histologic subtypes. METHODS: Seven patientderived diffuse GC organoid lines were established, characterized by histopathologic analysis, in situ hybridization, and gene expression analysis. For genetic modeling of diffuse GC, we knocked out CDH1 and/or TP53 in human normal gastric organoids. Green fluorescent protein-labeled GC organoids were xenotransplanted into immune-deficient mice for in vivo assessment. RESULTS: PCC-NOS organoids transformed into SRCC-like structures on removal of Wnt and Rspondin from the culture medium. This morphologic change paralleled downregulation of Wnt-target and gastric stem cell genes, including LGR5, and elevation of differentiation markers, such as KRT20 and MUCs. The association between Wnt target gene expression and histologic subtypes was confirmed in 3 patient-derived GC tissues. In vivo, single clone-derived organoids formed tumors that comprised 2 distinct histologic compartments, each corresponding to SRCC and PCC-NOS. The transition from PCC-NOS to SRCC histology reflected the abundance of surrounding Rspondin-expressing fibroblasts. CONCLUSIONS: SRCC and PCC-NOS were clonally identical, and their morphology was regulated by extracellular Wnt and R-spondin expression.Our results decoded how genetic mutations and the tumor environment shape pathohistologic and biologic phenotypes in human diffuse GCs.

Immune checkpoint inhibitors (ICIs) have been approved for various types of cancer; however, they cause a broad spectrum of immune-related adverse events (irAEs). The association between the development of irAEs and the clinical benefit remains uncertain. We aimed to evaluate the association of irAEs and the treatment efficacy in real-world practice. Patients and Methods: We conducted a retrospective study on patients with recurrent or metastatic non-small-cell lung cancer, malignant melanoma, renal cell carcinoma, or gastric cancer who received anti-PD-1/PD-L1 antibodies (nivolumab, pembrolizumab, or atezolizumab) at the Keio University Hospital between September 2014 and January 2019. We recorded treatment-related AEs from medical records and graded them using the Common Terminology Criteria for Adverse Events version 4. We performed an overall survival (OS) analysis using a Cox proportional hazards model and the shared frailty model. Results: Of 212 patients eligible for this study, 108 experienced irAEs and 42 developed multiple irAEs. The median OS was significantly longer in the irAEs than in the no-irAE group (28.1 months vs 12.7 months; hazard ratio [HR], 0.49; 95% confidence interval [CI], 0.33-0.73; P = 0.0004). Moreover, the OS of patients with multiple irAEs was significantly longer than that of patients with a single irAE (42.3 months vs 18.8 months; HR, 0.473; 95% CI, 0.346-0.647; P < 0.0001). Conclusion:Our single-center retrospective study revealed a significant tendency associating the development of multiple irAEs with favorable prognoses.

No abstract