Japanese researchers have developed a new organoid biobank with which to identify molecular subtypes of rare gastroentero-pancreatic neuroendocrine neoplasms (GENs) using genotype-phenotype mapping, they reported in the November 6, 2020, online edition of Cell.

Simplified 3D versions of an organ with a realistic micro-anatomy, organoids are a valuable research resource, as they are derived in vitro from stem cells and can self-organize due to their self-renewal and differentiation abilities.

Importantly, the new study demonstrated the validity of using organoid-based modeling to investigate rare and lethal cancers such as GENs, for which there are few clinically relevant models.

Gastrointestinal (GI) tract epithelial malignancies usually differentiate into adenocarcinomas or squamous cell carcinomas from glandular or squamous epithelia, respectively.

Tumors lacking epithelial structure and neuroendocrine (NE) differentiation are classified as NE carcinomas (NECs), while NE tumors (NETs) express NE markers, but show well-differentiated morphology.

"Two lines of colonic NEC organoids were first established by Japanese researchers in 2016, but ours is the first report of established NET organoids," said study leader Toshiro Sato, a professor in the Department of Organoid Medicine at Keio University in Tokyo.

The WHO classifies GI tract NETs and NECs collectively as GEN neoplasias (GEP-NENs), which are slow growing, but are aggressive and eventually metastasize with poor survival.

"There are four known gene expression subtypes in small cell lung carcinomas, but the existence of similar subtypes in GEP-NEN was previously unknown," said Sato.

"Here, we have identified three subtypes overlapping with lung and one new subtype unique to GEP-NENs," he told BioWorld Science.

However, due to their rarity, prospective clinical trials have been difficult to conduct and GEP-NEC biology is poorly understood, although recent molecular profiling has suggested genetic involvement.

Genetic involvement

Two genetic mutations, in TP53 and RB1, occur recurrently in GEP-NECs, with genetic profiling showing they share driver gene mutations with adenocarcinomas.

The existence of mixed NE- and non-NE neoplasms further suggests that GEP-NENs and non-NENs develop from common precursors.

However, unlike GEP-NECs, GEP-NETs lack the TP53 and RB1 and other driver gene mutations found in adenocarcinomas, but acquire other characteristic mutations.

Furthermore, germ-line mutations are known to predispose to multi-focal GEP-NETs, possibly underlying independent molecular mechanisms and contributing to GEP-NEC and GEP-NET differential pathobiology.

Although mouse genetic models have shown molecular pathways contributing to GEP-NEC and GEP-NET development, these do not accurately reflect human clinical phenotypes, presumably due to species differences.

Moreover, functional assessment of GEP-NEN biology has been hampered by the lack of relevant human models and few authentic human GEP-NEN cell lines are currently available and as yet there has been no long-term propagation of GEP-NET organoids.

This is important, "because GEP-NECs can grow indefinitely in patients, so need to be propagated in the long-term in in vivo-relevant culture conditions," said Sato.

"Moreover, long-term organoid propagation is essential for their distribution to researchers and industry."

This requirement prompted the new Cell study, in which Sato and his research team generated and characterized the molecular structure of 25 patient-derived NEN organoids.

Using comprehensive molecular analysis and CRISPR-based genetic engineering, the researchers explored the molecular drivers contributing to acquisition of unique GEP-NEN phenotypes.

Notably, the GEP-NEN organoids were shown to reflect the pathohistological and functional phenotypes of the original tumors.

"Both the organoids and their xenografts exhibited similar pathology with their original tumors," Sato said.

"We also found a similar drug response of the GEP-NET organoids with their original patients to hormone therapy with a somatostatin analogue, suggesting functional recapitulation."

Moreover, whole-genome sequencing revealed frequent TP53 and RB1 genetic alterations in GEP-NECs and characteristic chromosome-wide loss of heterozygosity (LOH).

"Diploid cells contain only two alleles of each gene or two copies of a single allele, whereas a haploid gamete contains one copy of each gene, hence only one allele," explained Sato.

"Our somatic cells contain two copies of a single allele, whereas some of our GEP-NEN organoids showed near haploidy, with cells contain one copy of a single allele.

"The presence of chromosome-wide LOH indicates chromosomal instability and may promote the transformation of cancers," he said.

Transcriptome analysis was then used to identify molecular subtypes distinguished by the expression of distinct transcription factors.

Furthermore, the GEP-NEN organoids were demonstrated to gain independence from the stem cell niche, irrespective of genetic mutations, which is important.

"Stem cell niche factors confer indefinite self-renewal capacity, with most cancers depending on specific niche factors, which can be subject to molecular-targeting therapy," noted Sato.

"We showed that the GEP-NEN organoids were highly independent of stem cell niche factors irrespective of mutations, suggesting their resistance to such therapy."

Moreover, TP53 and RB1 gene knockout and overexpression of key transcription factors resulted in normal colonic epithelial phenotypes compatible with GEP-NEN biology.

"To understand a disease, it needs to be modeled and, by generating GEP-NEN organoids, we can better understand its biology and how normal cells become GEP-NENs," said Sato.

Taken together, "our findings provide a research resource with which to promote the investigation of rare cancers, such as GEN-NENs, by accessing rare live cancer materials," he said.

"Due to their rarity, it has been difficult to conduct clinical trials for GEP-NENs, but our organoid biobank, researchers can determine the therapeutic effects of potential new treatments in cultures and xenografts." (Kawasaki, K. et al. Cell 2020, Advanced publication).