Abstract
Gastric cancers gain R-spondin independence by multiple mechanisms including CDH1/TP53 comutations.
Major finding: Gastric cancers gain R-spondin independence by multiple mechanisms including CDH1/TP53 comutations.
Approach: 37 patient-derived gastric cancer organoids were generated and phenotypically and genomically characterized.
Impact: Therapeutic targeting of Wnt may be effective in R-spondin independent gastric cancers.
Gastric cancer is a heterogeneous disease comprising a range of histopathological appearances and genomic aberrations. Disease models that reflect this phenotypic diversity in human gastric cancer are lacking. To generate diverse tumor models, Nanki, Toshimitsu, Takano, and colleagues established a collection of 37 patient-derived gastric cancer organoids covering histologically and molecularly diverse tumors. A subset of the organoids was genomically characterized using whole-exome sequencing, copy number analysis, and microsatellite instability analysis, facilitating classification by genetic subtype. Further, the transcriptomes were characterized using microarray and methylation microarray analyses. Wnt and R-spondin are required for the growth of normal gastric epithelium organoids, but 12 of the gastric cancer organoids were rendered WNT/R-spondin independent by several mechanisms: self-secretion of Wnt, APC mutations, or epigenetic Wnt/R-spondin regulation. Overall, 24 organoids were Wnt ligand dependent, and these had a high frequency of deletions in RNF43 and ZNRF3, which encode proteins that ubiquitinate the Wnt receptors. Of the 15 R-spondin independent organoids, 10 lacked RNF43 alterations. These 10 harbored frequent CDH1 mutations and CDH1/TP53 comutations, representing an alternative mechanism to gain R-spondin independence in the absence of RNF43 or ZNFR3 mutations. Xenotransplantation of 10 of these patient-derived gastric cancer organoids into the immunodeficient mice allowed for evaluation of the porcupine inhibitor (Porcn-i), which suppresses active Wnt ligand production. Porcn-i treatment reduced the growth of R-spondin independent tumors, but not Wnt-independent tumors or tumors with constitutive WNT activation. The generation of gastric cancer organoids provides insight into their diverse molecular and histopathological features, and suggests the potential for therapeutic targeting of Wnt in a subset of gastric cancers.
Note: Research Watch is written by Cancer Discovery editorial staff. Readers are encouraged to consult the original articles for full details. For more Research Watch, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/CDNews.