Abstract
Small intestinal neuroendocrine tumors, siNETs, are a group of rare cancers that arise from neuroendocrine cells in the lining of the jejunum and ileum, which are either classified as tumors, siNETs, or small intestinal neuroendocrine carcinomas, siNECs. Current treatment strategies for low-grade tumors include surgical resection, peptide radionucleotide receptor therapy, and somatostatin analogues, while high-grade and recurrent tumors may receive cytotoxic chemotherapy. These limited treatment options are linked to the lack of representative models that can both reflect the biology of the tumor and are amenable to mid-to-high throughput experimentation. Cell line generation is challenging considering the indolent nature of primary lesions, although some attempts have been successful using a variety of methods and include the primary P-STS line and those derived from metastatic lesions, including GOT1, CNDT2.5, and HC45. Patient-derived modeling, including organoids and xenografting, have allowed for multicellular and 3D representations of the original tumor. These specific models allow for multicellular populations derived from the tumor, providing better tumor representation for use in drug screening and in vitro assays. Currently, there are limited, although increasing, published models of siNETs implanted as xenografts in mice and zebrafish. As these cellular and animal models provide insights into siNET biology, theragnostic modeling has provided key information on the clinical progression and treatment of this disease. Significant strides toward more representative models have been made throughout the last decade. In this review, details of these attempts as well as future directions and strategies for more robust models will be addressed.
Introduction
Small intestinal neuroendocrine tumors (siNETs) are a rare malignancy of the midgut, located within the jejunum and ileum between the Ligament of Treitz and the ileocecal valve (Modlin et al. 2004). Although uncommon overall, these slow-growing tumors are the most frequent malignancy in the small intestine and are increasing in incidence and prevalence in the United States, with an annual incidence of about 1.05 per 100,000 (Dasari et al. 2017). There has been a rise in the incidence of this pathology, which can be attributed to better diagnostic strategies, imaging, and clinical awareness (Dasari et al. 2017). Recently, the incidence of siNETs has been reported at 1.46 per 100,000 (White et al. 2022). In some cases, these tumors can contribute to carcinoid syndrome (Kerstrom et al. 2005). Carcinoid syndrome is a paraneoplastic disease caused by the release of hormones, primarily serotonin, which results in flushing, diarrhea, and, rarely, carcinoid heart disease. In the past, these tumors were generally were classified as midgut carcinoids; however, the nomenclature of these malignancies has been reclassified to reflect molecular and cellular differences within this group as siNETs and small intestinal neuroendocrine carcinomas (siNECs). Modeling is vital for the improvement of imaging techniques, diagnosis, targeted therapeutics, and patient outcomes for siNETs of all grades and metastatic statuses. There have been some analyses of broad gastro-entero-pancreatic neuroendocrine tumor models published previously (Kawasaki et al. 2018); however, this review aims to provide an in-depth portrayal of small intestinal NET models on a molecular, cellular, and organismal level. Notably, functional small cell and ampullary duodenal NETs will not be included in this review, as they are genetically and clinically separate from other siNETs.
SiNETs are inherently indolent in comparison to many other cancers, including cancers in the neuroendocrine category which originate in other tissues. SiNETs have a low mutational burden, with CDKN1B locus alteration, the most commonly observed mutation, present in fewer than 10% of cases (Francis et al. 2013). However, hemizygous loss of chromosome 18q, which contains the SMAD4 gene, has been reported to be present in 74% of patients in one study (Hofving et al. 2021a ). This relatively low mutation burden complicates the creation of models, which often depend on multiple genetic abnormalities to improve establishment; alternatively, increased knowledge of the somatic copy number alterations that underlie this cancer may provide insights for improved model creation. Models that represent the biology of SiNETs are essential for preclinical and translational studies that improve patient outcomes. In addition, given the relative scarcity of fresh tumor samples, models are necessary to recapitulate this neoplasm with sufficient breadth and depth to improve the diagnosis, treatment, and prognosis of patients with this disease. Overall, the best models to study siNETs are those that reflect the genetic alterations, histological markers, and behavior of most tumors in this category.
Current treatments for siNETs are similar to NETs of other origins, including the pancreas, colon, and rectum (Pavel et al. 2020). SiNETs are graded in the same fashion as other malignancies, via the WHO system; grade 1 (G1) corresponds to a Ki-67 of less than 3%, grade 2 (G2) to a Ki-67 of 3–20%, and grade 3 (G3) above 20%. In G1 and G2, primary lesions are surgically resected, with a lower risk of recurrence. Well-differentiated tumors are also candidates for PRRT and somatostatin analogue therapy, which targets SSTR2-expressing tumor cells. In higher grades, Ki-67 index >20%, cytotoxic chemotherapy is the primary symptom-attenuation treatment option. Therefore, there is a strong need for more effective and varied treatment strategies to improve outcomes and the quality of life for siNET patients, especially those with G3 siNETs and siNECs. Powerful modeling is a requirement for such advancement in therapy. From least to most complex, these models include cell lines, spheroids, organoids, and animal models; however, as complexity and tumor representation increase, researchers are limited by throughput (Fig. 1). The overall aim of this review is to provide a detailed discussion of the utilization and limitations of siNET models and address areas of future research in this field.
Schematic describing the proposed development and progression of small intestinal neuroendocrine tumors, as well as current models at an increasing level of complexity and disease representation. Image created with Biorender.com.
Citation: Endocrine Oncology 4, 1; 10.1530/EO-24-0038
Cell lines
Monolayer culture of immortalized cells derived from tumor samples is one of the oldest and most utilized techniques to model cancers. They can be used in a high-throughput format that is amenable to large-scale genetic interrogation, drug screening, and various molecular assays. However, due to the slow-growing nature of low-grade siNETs, deriving cell lines from primary tissue has been challenging. Most lesions will be positive for Chromogranin A, Synaptophysin, and Neuron-specific Enolase (NSE), among others, but there are examples of siNETs that do not express one or a combination of these biomarkers (Table 1). To date, there has only been one published cell line derived from a non-functional human primary lesion and three cell lines derived from liver metastases (Table 2). These lines vary in doubling time, immunohistochemical profile, and original patient pathology, which reflect the heterogeneity of this pathology broadly.
Immunohistological markers of siNETs and siNET model development.
| Marker | Function | Relevance to siNETs and siNET models | Cellular location | References |
|---|---|---|---|---|
| Synaptophysin (Syp) | Integral membrane protein in neural synaptic vesicles | Expressed by almost all siNETs with high senstitivity | Tumor cell intracytoplasmic vesicles | Miettinen (1987), Andersson et al. (2016) |
| Chromogranin A (CgA) | Acidic secretory pro-hormone protein from neurons and most neuroendocrine cells | Main biomarker for the clinical diagnosis of siNETs, with high specificity and medium-to-high sensitivity | Cytoplasmic secretory granules | D’Amico et al. (2014), Andersson et al. (2016), Di Giacinto et al. (2018) |
| Somatostatin receptor 2 (SSTR2) | Membrane receptor for the cyclic peptide, growth-hormone-inhibiting hormone somatostatin | Found in a majority of siNETs, functional target for siNET therapy | Within the cytoplasm and within the plasma membrane | Papotti et al. (2002), Watanabe et al. (2022) |
| Neuron-specific enolase (NSE) | Isoenzyme of the glycolytic enzyme enolase, secreted in late-stage neural maturation | Upregulated in tumors and analyzed in coordination with Syp and CgA | Dispersed throughout the cytoplasm, occasionally in the nucleus and on the cell surface | Isgro et al. (2015), Mjones et al. (2017) |
| Vesicular monoamine transporter 1 (VMAT1) | Mediate amine transport into vesicles | Secreted by amine-producing neuroendocrine tumors of ileal origin | Found in large dense core and small synaptic vesicles | Erickson et al. (1996), Jakobsen et al. (2001) |
| Serotonin (5-HT) | Vasoactive hormone secreted by many endocrine organs, including enterochromaffin cells | Marker used in the diagnosis of siNET, present in >85% of tumors | Diffuse throughout the cytoplasm | Yang et al. (1983), Krishnamurthy & Dayal (1997), Blazevic et al. (2018) |
| Substance P | Neuropeptide that is widely present in neural cells and promotes tumor cell migration | Positive in some tumors and cell lines | Cytoplasmic vesicles | Hofving et al. (2018), Covenas & Munoz (2022) |
Known siNET cell lines, their origin, immunohistochemical profile, growth profile, and original publication.
| Cell line (original publication) | Origin | Immunohistochemical profile | Growth profile | References |
|---|---|---|---|---|
| P-STS | G3 tumor of the terminal ileum | Pan-CK, serotonin, CD56, Syp, HISL19, NSE | Doubling time: 4 days–1 week, grows in mildly adherent culture with some non-adherent aggregates | Pfragner et al. (2009), Hofving et al. (2018) |
| GOT-1 | G1 carcinoid tumor liver metastasis | CgA, 5-HT, VMAT1, serotonin, NCAM | Doubling time: 18 days | Kölby et al. (2001), Hofving et al. (2018) |
| CNDT2.5 | G1 liver metastasis from an ileal carcinoid | Syp, NSE, produced serotonin, SSTR1-5 | Limited published data | Van Buren et al. (2007), Barazeghi et al. (2021) |
| HC45 | G1 liver metastasis from an ileal carcinoid | CgA, CgB, SSTR2, SSTR5, Syp, VMAT 1, vitamin D3 receptor | Limited published data | Stilling et al. (2007) |
| SS-C | Jejunal somatostatinoma | SSTR2, NSE, CgA, Syp | Doubling time: 3–6 days | Galli et al. (2006) |
| STC-1 (mouse) | Tumor from the duodenum of a RIP1Tag2/RIP2PyST1 mouse | Secretin, gastrin, somatostanin, CgA | Doubling time: 54 h | Rindi et al. (1990) |
Non-functional primary siNET cell line
P-STS
In 2009, Pfragner et al. (2009) reported the development of a stable cell line derived from a siNET, specifically a primary lesion from the distal ileum of a 42-year-old woman that originated from enterochromaffin cells. P-STS cells form loose monolayers with an average doubling time of approximately 4 days. Upon initial culture, P-STS expressed pancytokeratin, cytokeratins 7, 8, 18, and 19, serotonin, NSE, CD56, PGP9.5, calcitonin, synaptophysin, and growth hormone-releasing factor. This further establishes this line as a representative model of the original lesion and its neuroendocrine phenotype (Pfragner et al. 2009). Although most cells were adherent in culture, floating cells displayed chromosomal alterations at the 18q site, which is consistent with some studies showing losses and gains of 18q play a role in tumorigenesis and tumor progression in siNETs (Kytola et al. 2001). Of interest, genetic sequencing of the P-STS line showed no alterations to the menin coding region, suggesting sporadic tumor generation without the genetic abnormalities associated with MEN1 syndrome. Hofving et al. reported that the P-STS line displayed a neuroendocrine phenotype, and were responsive to HDAC inhibitors (Hofving et al. 2018). Recently, the P-STS cell line was utilized to investigate stromal and neoplastic cell crosstalk. The authors found that integrin signaling pathways play a role in tumor and stromal cell crosstalk that contributes to tissue fibrosis (Laskaratos et al. 2021). The original patient presented with local lymph node and hepatic metastases, which resulted in the development of L-STS and H-STS (Pfragner et al. 2009), respectively; however, these lines have been shown to be lymphoblastoid and not representative of siNETs.
Metastatic siNET cell lines
GOT-1
In 2001, Kölby et al. reported the establishment of GOT-1 cells from a hepatic metastasis of a distal ileal NET of a 55-year-old woman, who also presented with lymph node metastases. The original tumor cells were positive for 5-HT, substance P, VMAT1/2, and markers for neuroendocrine phenotype, namely CgA, SYP, SV2, and NCAM, via IHC; the original tumor and the derived cell line also expressed somatostatin receptors, including SSTR2, which are the main treatment targets for these malignancies. Cell culture showed similar positive IHC results; therefore, the identity of the original GOT-1 cell line was successfully verified. This cell line is still widely used for a variety of in vitro studies; however, the doubling time of more than 2 weeks (>18 days) limits its usage in proliferation and growth studies (Kölby et al. 2001, Grozinsky-Glasberg et al. 2012). This line also displays a loss of a 1.8 Mb segment of chromosome 18; this is representative of the original neoplastic patient sample and is aligned with chromosome loss that often accompanies siNETs (Hofving et al. 2018).
CNDT 2.5
Van Buren et al. (2007) reported the generation of the CNDT 2.5 cell line in 2007, derived from the primary CNDT2 monolayer culture, which originated from a liver metastasis of a primary ileal carcinoid. CNDT2 cells were grown and then injected into nude mice. Cells from the xenograft tumor were then isolated and cultured independently. The resultant CNDT 2.5 line was validated as siNET-derived via molecular, protein, and cellular assays. Grown in culture, these cells have an approximate doubling time of 20 h, allowing for efficacious growth and proliferation studies, in vitro drug screening, and other functional assays. CNDT 2.5 cells contain neurosecretory granules and express mRNA for NSE, SYP, and VEGF and its receptors. Additionally, these cells express many targetable tyrosine kinase receptors and all five SSTRs, which were proven to be functional; these cells also contain serotonin and serotonin receptors (Van Buren et al. 2007). Notably, the original culture of these cells did not express chromogranin A nor cytokeratins. This deviation led to some debate on the origin of the cell line (Ellis et al. 2010). Chromosomal aberrations were present in this model, but on different chromosomes and loci than in other siNET models, which commonly occur at chromosome 18. The authors reported deletions in 2p and 6q and various translocations. This data may suggest an alternate origin; however, there is no definitive evidence that the cell line is misclassified beyond the loss of some cellular markers. CNDT 2.5 is still used to some degree as a model for in vitro experiments. In 2019, one study investigated transmembrane receptor type tyrosine phosphatases in these cells. The researchers found that a specific protein tyrosine phosphatase, PTPRM, acted as a tumor suppressor in siNET cells (Barazeghi et al. 2019).
HC45
Stilling et al. (2007) published the initial culture of HC45, a transformant cell line from a liver metastasis of an ileal carcinoid, in 2007. An SV40 early T-antigen insert is a notable feature of this specific line, which alters the line’s genome to immortalize the cell line for future culture and propagation. The neuroendocrine cellular phenotype was verified via IHC; this line was positive for synaptophysin, chromogranin A/B, 7B2, SSTR2, and SSTR5, and rarely VMAT1. The cells were negative for serotonin. Uniquely, the HC45 expresses the vitamin D3 receptor. Western blotting of the original culture displayed high expression of TGFβR1 and TGFβR2, as well as EGFR, which proved capable of therapeutic targeting. This profile reflects the neuroendocrine phenotype of these cells and the hepatic metastasis from which this line was cultured. Often TGFβ expression is associated with cell proliferation and tumor metastasis in the later stages of disease (Xie et al. 2018). Additionally, EGFR plays a role in cell division and growth in many cancer types (Sasaki et al. 2013). This cell line has been used as an ileal NET comparison model in a study regarding viral oncogenesis in lung NETs (Haley et al. 2010) and has been cited as an appropriate model of liver metastasis of siNETs (Carpizo & Harris 2021).
Functional and non-human siNET cell lines
SS-C
Galli et al. (2006) initially reported the generation of the SS-C cell line from a jejunal somatostatinoma that was surgically resected from a 28-year-old female. Somatostatinomas are incredibly rare among siNETs, making up less than 5% of cases. Notably, the authors mentioned that the original patient did not have any outlying hereditary or sporadic endocrine pathologies, including MEN1. Cultures of the primary tumor were positive for many neuroendocrine markers over several passages, including CgA, NSE, somatostatin, SSTRs, and gastrin, among others. These cells grow in a firm monolayer and have a doubling time of approximately 3–5 days, with an increase in days as passage number increases. This model represents a rare tumor that displays phenotype and functional differences from the large majority of siNETs.
STC-1 (mouse)
Rindi et al. (1990) reported the development of a murine cell line model of a small intestinal neuroendocrine adenoma, specifically from the duodenum, in 1990. This cell line was cultured from a RIP1Tag2/Rip2pyST1 transgenic mouse. The doubling time of this line is approximately 54 h. Upon immunohistochemical analysis (IHC), tumor cells stained positively for secretin, glicentin, gastrin, and somatostatin, albeit the latter was not ubiquitously expressed among the cells. Notably, only a select number of cells expressed both CgA and large T-antigen immunoreactivity; however, there was no mention of SYP. Although this is a murine-derived line, this calls the relative representation of the small intestinal neuroendocrine phenotype of STC-1 into question. Nevertheless, this line has been widely cited in both primary and review literature regarding both cancer and gastrointestinal metabolic disease. Most recently, STC-1 has been used experimentally as a model of cholecystokinin release in response to fermentation products (Egberts et al. 2020) and to test a cytotoxic herpes simplex virus on tumor growth in neuroendocrine cancer (Matsushima et al. 2019). Overall, this line may be useful in developing stable cell-derived xenograft tumor models in mice given inferred immuno-compatibility; however, the genetic and molecular differences between this model and human NETs must be considered.
Organoid and spheroid modeling
3D modeling presents a higher-order approach to studying siNETs and carcinomas in vitro (Fig. 2). This method provides a more representative view of how tumors behave, as opposed to a monolayer culture; yet they are still reproducible enough to perform high-throughput drug screening and cell-based assays. This section will primarily focus on patient-derived tumor organoids (PDTOs), which are 3D patient-derived cell structures present in a scaffold, often composed of extracellular matrix (ECM) components such as collagen or Matrigel. They can either self-organize into organ-like structures or form large aggregates of cells similar to tumor growth. These PDTOs can be cultured for multiple passages and stored for future experiments.
siNET 3D culture methodology. Image created with Biorender.com.
Citation: Endocrine Oncology 4, 1; 10.1530/EO-24-0038
In comparison to siNET cell lines, some of which do not accurately reflect the genetic profile, patient-derived tumor cell organoids that are cultured ex vivo can provide more powerful and translational insights in target validation and therapy evaluation. Additionally, siNET 3D culture can take on molecular, cellular, and structural characteristics that mimic the original midgut tumor. This provides an avenue for target elucidation, drug screening, and mechanistic studies of small intestinal neoplastic cells that truly reflect the tumor biology and may lead to clinical applications to improve prognosis at a faster rate.
To date, there are only a few accounts of 3D cultures of siNET tumor cell populations developed from patient tissue derived from freshly resected siNET tumors. Ear et al. (2019) developed a strategy for developing these 3D structures that support long-term culture for more than 9 months continuously, allowing for drug screening and functional studies downstream. Although the authors of this publication refer to their cultures as spheroids, the use of a dense ECM scaffold to aggregate and grow patient-derived tumor cells is more aligned with the term organoid in other publications. Nevertheless, these cultures maintain expression levels of ChgA, Syp, and SSTR2, which validate the neuronal and endocrine phenotype of these organoids over long-term culture. More recently, D’Agosto et al. (2023) cultured short-term PDTOs derived from a G2 lymph node metastatic siNET that was maintained in culture for several months. Upon characterization, the organoid cultures expressed similar siNET markers as the parent tumor sample via IHC, namely strong expression of CK8-18 and CDX2, and heterogeneous expression of CgA and SYP, and formed complex lumen-containing structures. Additionally, the proliferative index did not change significantly between organoid and tissue, maintaining the classification of G2. Compared to tumor organoids, there is a relative lack of research on the usage of 3D culture for siNET cell lines, including spheroids. Spheroids are cultured using non-adherent conditions and rely on microgravity to aggregate cells. Barazeghi et al. (2021) described in a recent publication that they successfully cultured non-adherent GOT1 spheroids to study cell proliferation when challenged with CPI-1205, a small molecule inhibitor of enhancer zeste homolog 2 (EZH2), and metformin, reinforcing recent data which suggest these tumors may be a result of epigenetic dysregulation. Most recently, Dayton et al. (2023) developed a set of lower-grade small intestinal NET organoids from both primary and metastatic lesions that were positive for chromogranin A and had similar transcriptional profiles to their parent tissues. This maintenance of gene expression patterns between patient and model underlines the value of patient-derived organoid modeling for investigations into tumoral protein signaling and drug screening. One of the tested siNET organoid lines in this investigation showed no single nucleotide variants but demonstrated structural variants; this recapitulates what is known about the genetic basis of this disease in general (Elias et al. 2021). Overall, these 3D cultures provide a translational bridge between cell lines and in vivo models, and they often maintain an epithelial phenotype and provide a more conducive growth environment, given ECM support. In contrast with animal models and larger-scale ex vivo bioreactor modeling, organoids and spheroids maintain the high-throughput, reproducible nature required for screening assays. Combined with an apparent need for cell lines that are truly representative of siNETs, and the wide genetic heterogeneity among them, organoids provide a reproducible structure to analyze each small intestinal neuroendocrine tumor as a separate entity with a variety of in vitro assays. One potential area for future research and development in small intestinal 3D modeling includes culturing primary cells in ECM on a microfluidic chip to recapitulate the tumor immune microenvironment and cell-cell interactions more accurately within these tumors (Aref et al. 2018).
Mouse models
In vivo modeling of tumor progression is one of the main hallmarks of preclinical investigation, which illuminates potential avenues toward innovation in therapeutics. There are three common routes for modeling tumors: patient-derived xenograft (PDX), cell line-derived xenograft (CDX), and genetically engineered mouse models (GEMMs), which produce sporadic or hereditary tumors over time. In recent literature, there have been successful attempts to implant both cell line- and patient-derived material into mice, which produce a tumor eligible for further experimentation (Table 3). However, due to the indolent and slow-growing nature of these tumors and their derivative cell lines, xenografting has only been sparsely successful in a select number of studies. As of this review, there are few successful accounts of a genetically engineered mouse model that produces sporadic or hereditary small intestinal lesions, although there have been successful GEMM developments in NETs of other origins, whether through selective silencing of genes or via oncogenic transformation (Forsythe et al. 2023).
Current murine models of small intestinal neuroendocrine tumors.
| Model | Type | Location | Notes | References |
|---|---|---|---|---|
| Transgenic, carrying insulin-promoted oncogenes | GEMM | Small intestine | Secreted secretin, proglucagon-related peptides, glucose-dependent insulinotropic polypeptide, among others. Large T-antigen immunoreactivity | Rindi et al. (1990) |
| GOT1 CDX/BALB/cABom-nu | Cell xenograft | Subcutaneous, back of the neck | IHC positive for major NET markers (CgA, Syp, NCAM) | Kölby et al. (2001) |
| Low-grade LM PDX/NOG | Tumor xenograft | Subcutaneous, hepatic orthotopic | Hepatic orthotopic was not successful, some cells from SQ were positive for NET markers | Hofving et al. (2021b) |
| CNDT2.5 PDX/NMRI-nude female | Cell xenograft | Subcutaneous, hind flank | Cells embedded in Matrigel, cells positive for Syp | Barazeghi et al. (2021) |
| RT2 transgenic mice, B6AF1 background | GEMM | Ileum | Tumors noted to have high expression of IGF2, which is a potential driver of tumorigenesis | Contractor et al. (2020) |
In 1990, Rindi et al. (1990) reported the development of neuroendocrine tumors in the gastrointestinal tract of transgenic mice, some of which produced hormones including secretin and proglucagon-related peptides. This represented one of the earliest reports of a murine model for this cancer type. In 2001, Kölby et al. (2001) published one of the first successful transplantations of primary cells cultured from a liver metastasis of a small intestinal neuroendocrine tumor, which were subsequently established as the GOT1 line. The researchers transplanted tumor cells into male BALB/cABom-nu mice; over time, these tumors were resected and passaged over five generations. Small intestinal NET immunohistochemical markers were positive across all generations of tumor propagation. While studying the tumor microenvironment of siNETs, Hofving et al. (2012b) reported one successful attempt at subcutaneous grafting of cells from a low-grade liver metastasis of an siNET in a NOG mouse. The authors reported, however, that most attempts at both subcutaneous and hepatic orthotopic xenografting from dissociated cells were ultimately unsuccessful; only 1 of 36 subcutaneous grafting attempts was successful. Successive undigested tumor sample xenografting from the original patient was marginally successful, with a modest amount of live siNET cells after propagation, although tissue necrosis was noted. Via IHC, the metastatic siNET cell identity was verified with positive staining for synaptophysin. Most recently, in addition to GOT1 spheroid development, Barazeghi et al. (2021) documented successful xenotransplantation of CNDT2.5 cells to investigate EZH2 inhibitor therapy for siNETs. CNDT2.5 cells were embedded in Matrigel and subcutaneously injected into the hind flank of nude female mice. In all conditions, the xenografted tumor stained positively for synaptophysin and staining for CgA was not noted. Recently, Contractor et al. (2020) have reported the development of ileal neuroendocrine tumors, specifically in the B6AF1 genetic background, from the RT2 transgenic mouse model, which usually produces pancreatic neuroendocrine tumors.
Other models of siNETs
Zebrafish are commonly used as a model of tumor angiogenesis and as a relatively reproducible animal model of neuroendocrine tumors, including those that arise from the small bowel. Gaudenzi et al. reported a successful xenotransplantation of primary culture from a liver metastasis of a low-grade (Ki67 < 2%) ileal NET in a teleost zebrafish with EGFP-labeled fli1a (Gaudenzi et al. 2017, Gaudenzi & Vitale 2019). The EGFP-conjugated fli1a promoter provides a fluorescent marker for vascular development and tumor vasculature. Anti-fibroblast processing preceded the initial transplantation of suspended NET cells into the subperidermal space of the zebrafish embryos 24 h post fertilization. The authors measured a 33% tumor-induced angiogenesis ratio in fish with the PDX culture, and 67% of fish were positive for tumor cell migration. Notably, the authors reported an unsuccessful PDX attempt of a primary G2 ileal lesion. This specific class of model provides a bridge between organoid and mouse models to address tumor microenvironmental factors that may affect treatment efficacy on an organismal level, while being relatively higher-throughput compared to other animals.
Limitations of current siNET models
Limitations of siNET models stem from the nature of the cancer from which they are derived; a slow-growing, indolent cancer generates indolent models. Additionally, the lack of single nucleotide genetic variants impacts animal model development. For example, the doubling time of GOT1 is 18 days, which presents a barrier for cell proliferation and growth assays and cell line xenografting in animal models. Another recurring issue in siNET model development is the inconsistent validation for available cell lines. Pfragner et al. reported the establishment of a continuous line, KRJ-I, from a small intestinal carcinoid tumor in 1996 (Pfragner et al. 1996). This cell line expressed both neuroendocrine markers and estrogen receptors, and these cells were morphologically representative of the original tumor; however, staining for many common NET antigens, including somatostatin receptors and VIP, was negative in the original culture. Recently, the validity of this line has been the subject of debate. In culture, KRJ-I cells grow in suspension, which is not consistent with other cell lines in this group. Additionally, these cells stained positive for lymphoblastoid markers, including CD45 and CD20. These issues call into question the utility of these cells for in vitro studies of siNETs (Hofving et al. 2018). However, this cell line is well cited in literature and mimics these tumors on a transcriptomic level (Alvarez et al. 2018, 2019). Hofving et al. classified L-STS and H-STS cell lines, which were originally identified as lymphatic and hepatic metastases from the primary lesion used to develop the P-STS line. The metastatic lines did not stain positively for many neuroendocrine markers or cytokeratins. These lines were also positive for lymphoblastoid markers and Epstein-Barr Virus DNA. Therefore, KRJ-I, L-STS, and H-STS are not representative models of siNETs (Hofving et al. 2018). Overall, examination and characterization of cell lines are important to determine which lines are suitable for experimentation, and researchers should demonstrate caution when using these cell lines.
Although CDX and PDX animal models have been recently developed using immunodeficient mice, this limitation attenuates the utility of these models for investigations into the tumor immune microenvironment in vivo. Only one genetically engineered mouse model resulted in the sporadic generation of small intestinal neuroendocrine tumors within immune-competent mice. In the study, Rindi et al. reported accounts of spontaneous generation of intestinal neuroendocrine tumors in RIP1Tag2/RIP2PyST1 mice in 1990 (Rindi et al. 1990); however, the precise location of these tumors was not noted, and this model has not been replicated since initial documentation. Although insulinomas, gastrinomas, glucagonomas, and non-functional pancreatic NETs have been grown in MEN1-mutant, RIP-Tag, and Glu-Tag murine models, there are no accounts of siNET generation (Pritchard 2022). Clinically, MEN1 patients often present with foregut and pancreatic tumors (Kamilaris & Stratakis 2019, Al-Salameh et al. 2021), but do not usually have midgut tumors in the jejunum and ileum. This may explain the relative lack of small intestinal lesions in this model specifically. Additionally, given the low mutational burden of these tumors, transgenic mouse models may not be feasible. However, unlike cell lines and organoid models, animal models would provide the most realistic representation of carcinoid syndrome, which is an important aspect of this disease (Vitale et al. 2023). This warrants further development of this class of model.
Computational modeling
Although in vitro and in vivo modeling provide necessary insights into tumor biology, computational modeling of small intestinal NET growth, grading, and therapy can help refine technology and improve patient outcomes. Radiomics is a powerful computational tool to model tumor growth for the prediction of prognosis (Fig. 3). Through a nomogram, high-depth data can be generated from radiologic scans to model tumor progression through computational algorithm training (Wu et al. 2017). Artificial intelligence and machine learning can be utilized as high-throughput tools to improve the algorithm over successive samples and provide deeper insights on prognostic factors (Lu et al. 2018).
Model schematic of radiomic modeling of small intestinal neuroendocrine tumors using segmentation of computed tomography imaging.
Citation: Endocrine Oncology 4, 1; 10.1530/EO-24-0038
Recently, this methodology has been applied to small intestinal neuroendocrine tumors. Blazevic et al. (2021) reported using radiomic analysis to predict small intestinal NET metastases in the mesentery. In this study, CT scans were digitally segmented for a region of interest and assessed by a trained clinician. Mesenteric mass and surrounding mesenteric regions were then quantified for various prognostic factors. A trained algorithm with various models interpreted the data to predict the development of adverse intestinal symptoms in patients, which the authors noted was likely due to the characteristic surrounding mesentery of the metastasis.
In addition, pharmacokinetics, the study of therapy uptake and metabolism, presents a clinically applicable mathematical model of siNET therapy and can be applied to increase the depth of in vivo studies. Given the current recommendations for peptide-receptor radionuclide therapy for siNET cases, modeling how this specific therapeutic works is a vital asset for investigating clinically relevant questions regarding absorption, toxicity, drug metabolism, half-life, and excretion. There are many different forms of PRRT, with different radiolabeling molecules, that have been subject to investigation as a therapeutic intervention for NETs.
Using 212Pb-SSTA-PBPK (physiology-based pharmacokinetic) mice as a model of alpha-PRRT, Zaid et al. (2021) demonstrated the value of combining a whole organism PBPK model for evaluating dosimetry of alpha-particle emitting therapy and toxicity in various SSTR-producing tissues with computational PK methods. Additionally, the authors analyzed renal toxicity of this treatment, which is vital in the transition of this therapy to the clinic, and found that including a kidney model could provide key insights into long-term renal toxicity as a result of PRRT. Jimenez-Franco et al. (2021) performed in silico modeling of 177Lu-DOTATATE therapy, which is currently FDA-approved for well-differentiated siNETs where SSTR expression is high. The researchers used a virtual patient PK model to investigate tumor control probability with respect to tumor perfusion and SSTR2 density; the results indicated a minimum tumor perfusion of 0.062 mL/g/min and receptor density of 55 nmol/L to establish a 99% tumor control probability. The authors noted that further computational analysis of dosimetry of PRRT is vital for optimization of this therapy for NETs.
Future directions
As previously mentioned, small intestinal NETs are known to be genetically heterogeneous and show some clustering for limited aberrations at specific gene loci, including CDKN1B (Elias et al. 2021, Wedin et al. 2022). Current therapies for siNET, beyond surgical excision, are only moderately successful, depending on differential somatostatin receptor density, tumor cell differentiation and presence or absence of mutations among patients. Current models seek to improve our understanding of these factors and provide new avenues for treatment. For example, high-throughput organoid development protocols, ideally through patient biopsy, could be cultured, then challenged with a panel of therapeutics to determine efficacy on a per-patient basis. This precision oncology approach may be useful in the future when more drugs become FDA-approved for these malignancies. In the short term, organoid biobanking of siNETs expands patient samples for larger scale research projects.
Bioreactor tumor modeling has been used recently as a 3D ex vivo system for pancreatic NET tumor modeling and could theoretically be applied to NETs that arise from the small intestine (Herring et al. 2021). This technology uses a scaffold that holds cultured tumor cells in a matrix and continuously perfuses media through the scaffold. Imaging of these scaffolds is non-invasive and can be evaluated for growth via IR-783, a specific near-infrared fluorescent dye. Aref et al. (2018) published another microfluidic model, based on using patient-derived tumor spheroids on a chip to investigate immune checkpoint biology in siNETs. A mixture of tumor, immune, and stromal cells were embedded in collagen I and perfused with media. Although these initial studies are accounts of the developments in 3D modeling in the last 5 years, future innovation in this area could further close the gap that exists in siNET modeling, extend the utility of samples, and better recapitulate the tumor environment in vivo.
Conclusion
As siNETs increase in incidence globally, there is a growing need in neuroendocrine tumor research, and specifically siNET research, for better models which provide representative insights, are cost-effective, and are higher throughput. There have been many attempts to increase the breadth and depth of siNET models (Table 4); many were unsuccessful or have failed tumor validation. Patient tissues from which these models are derived are rare, which limits development. However, organoid modeling is one of the best candidates to reflect the genetics and multicellular makeup of siNETs in larger-scale experiments when these tissues can be procured. Cell lines provide a lower-cost and broadly accessible alternative, albeit they are less representative of the original tumor. With advancements in tissue engineering, more intricate and sophisticated ex vivo modeling is also worthy of future research. Off the bench, in silico and computational models can provide translational insights with clinical applications. Given the intricacies of these specific tumors, models that are malleable yet best reflect the patient ultimately result in the most powerful data and preclinical innovation.
Overview of currently available siNET models and their advantages and limitations in basic, translational, and clinical research.
| Model | Examples | Intended use | Advantages | Limitations |
|---|---|---|---|---|
| Cell lines | GOT1 (met) (Kölby et al. 2001) P-STS (primary) (Pfragner et al. 2009) HC45 (met) (Stilling et al. 2007) CNDT2.5 (met) (Van Buren et al. 2007) SS-C (functional) (Galli et al. 2006) STC-1 (mouse) (Rindi et al. 1990) |
In vitro experimentation where fast growth is required, cell-based assays | Highest-throughput, highly reproducible and easy to generate more, easily manipulated for assays, cost-effective | Least representative, some lines harbor uncharacteristic mutations or lack biomarkers, immortalization procedures necessary to extend growth timespan |
| Spheroid | GOT1-derived spheroids (Barazeghi et al. 2021) | In vitro experimentation, drug screening, functional assays | Maintain multicellular aggregates, providing more representative drug responses than cell lines | Lack of tissue architecture, finite growth timeline, size limitations |
| Organoid | Short-term siNET PDOs (Ear et al. 2019) G2 LN Met PDOs (D’Agosto et al. 2023) Low-grade siNET Primary and Met PDOs (Dayton et al. 2023) |
Ex vivo experimentation, drug screening, long-term culture | Maintain varied tumor cell populations ex vivo, representative drug response, culture is representative of an individual patient, high throughput | Limited growth timespan, require rare tumor tissue, more time- and cost-intensive |
| Zebrafish | PDX EGFP-fli1a telost (Gaudenzi et al. 2017) | In vivo experimentation related to tumor angiogenesis, migration, and metastasis | Specialized model for studies of vasculature and migration, high xenograft establishment rate compared to other animal models | Specialized equipment requirements, limited scope of use, requires rare patient tissue |
| Mouse | Transgenic, insulin-promoted oncogenes (Rindi et al. 1990) GOT1 CDX/BALB/cABom-nu (Kölby et al. 2001) Low Grade LM PDX/NOG (Hofving et al. 2021b ) CNDT2.5 PDX/NMRI-nude female (Barazeghi et al. 2021) RT2 transgenic (Contractor et al. 2020) |
In vivo experimentation, preclinical modeling, testing therapy candidates | Most representative preclinical model, standard for translational drug efficacy studies | Most cost-intensive, lowest throughput, ethical and institutional regulations with animal use in research, lowest rate of model establishment |
| Clinical | siNET CT radiomics (Blazevic et al. 2021) siNET therapy pharmacokinetics (Zaid et al. 2021, Jimenez-Franco et al. 2021) |
Computational modeling for clinical decision-making, drug uptake studies, and macroscopic tumor analysis | Can accommodate large, already available datasets and improve patient outcomes | Limited use for basic or translational research, requires powerful computational resources |
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this review.
Funding
Funding is provided by the National Institutes of Health ZIA BC 011899.
Consent for publication
All authors have read and approved the final manuscript.
Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Author contribution statement
ASG – conceptualization, manuscript writing, figure preparation, editing. FSD – manuscript preparation, expert opinion, editing. MJP – manuscript preparation, expert opinion, editing. SSM – conceptualization, oversight, expert opinion, manuscript editing.
References
Al-Salameh A, Cadiot G, Calender A, Goudet P & & Chanson P 2021 Clinical aspects of multiple endocrine neoplasia type 1. Nature Reviews. Endocrinology 17 207–224. (https://doi.org/10.1038/s41574-021-00468-3)
Alvarez MJ, Subramaniam PS, Tang LH, Grunn A, Aburi M, Rieckhof G, Komissarova EV, Hagan EA, Bodei L, Clemons PA, et al.2018 A precision oncology approach to the pharmacological targeting of mechanistic dependencies in neuroendocrine tumors. Nature Genetics 50 979–989. (https://doi.org/10.1038/s41588-018-0138-4)
Alvarez MJ, Yan P, Alpaugh ML, Bowden M, Sicinska E, Zhou CW, Karan C, Realubit RB, Mundi PS, Grunn A, et al.2019 Reply to ‘H-STS, L-STS and KRJ-I are not authentic GEPNET cell lines’. Nature Genetics 51 1427–1428. (https://doi.org/10.1038/s41588-019-0509-5)
Andersson E, Arvidsson Y, Sward C, Hofving T, Wangberg B, Kristiansson E & & Nilsson O 2016 Expression profiling of small intestinal neuroendocrine tumors identifies subgroups with clinical relevance, prognostic markers and therapeutic targets. Modern Pathology 29 616–629. (https://doi.org/10.1038/modpathol.2016.48)
Aref AR, Campisi M, Ivanova E, Portell A, Larios D, Piel BP, Mathur N, Zhou C, Coakley RV, Bartels A, et al.2018 3D microfluidic ex vivo culture of organotypic tumor spheroids to model immune checkpoint blockade. Lab on a Chip 18 3129–3143. (https://doi.org/10.1039/c8lc00322j)
Barazeghi E, Hellman P, Westin G & & Stålberg P 2019 PTPRM, a candidate tumor suppressor gene in small intestinal neuroendocrine tumors. Endocrine Connections 8 1126–1135. (https://doi.org/10.1530/EC-19-0279)
Barazeghi E, Hellman P, Norlen O, Westin G & & Stalberg P 2021 EZH2 presents a therapeutic target for neuroendocrine tumors of the small intestine. Scientific Reports 11 22733. (https://doi.org/10.1038/s41598-021-02181-7)
Blazevic A, Hofland J, Hofland LJ, Feelders RA & & de Herder WW 2018 Small intestinal neuroendocrine tumours and fibrosis: an entangled conundrum. Endocrine-Related Cancer 25 R115–R130. (https://doi.org/10.1530/ERC-17-0380)
Blazevic A, Starmans MPA, Brabander T, Dwarkasing RS, van Gils RAH, Hofland J, Franssen GJH, Feelders RA, Niessen WJ, Klein S, et al.2021 Predicting symptomatic mesenteric mass in small intestinal neuroendocrine tumors using radiomics. Endocrine-Related Cancer 28 529–539. (https://doi.org/10.1530/ERC-21-0064)
Carpizo DR & & Harris CR 2021 Genetic drivers of ileal neuroendocrine tumors. Cancers 13 5070. (https://doi.org/10.3390/cancers13205070)
Contractor T, Clausen R, Harris GR, Rosenfeld JA, Carpizo DR, Tang L & & Harris CR 2020 IGF2 drives formation of ileal neuroendocrine tumors in patients and mice. Endocrine-Related Cancer 27 175–186. (https://doi.org/10.1530/ERC-19-0505)
Covenas R & & Munoz M 2022 Involvement of the substance P/Neurokinin-1 receptor system in cancer. Cancers 14 3539. (https://doi.org/10.3390/cancers14143539)
D’Agosto S, Fiorini E, Pezzini F, Delfino P, Simbolo M, Vicentini C, Andreani S, Capelli P, Rusev B, Lawlor RT, et al.2023 Long-term organoid culture of a small intestinal neuroendocrine tumor. Frontiers in Endocrinology 14 999792. (https://doi.org/10.3389/fendo.2023.999792)
D’Amico MA, Ghinassi B, Izzicupo P, Manzoli L & & Di Baldassarre A 2014 Biological function and clinical relevance of chromogranin A and derived peptides. Endocrine Connections 3 R45–R54. (https://doi.org/10.1530/EC-14-0027)
Dasari A, Shen C, Halperin D, Zhao B, Zhou S, Xu Y, Shih T & & Yao JC 2017 Trends in the incidence, prevalence, and survival outcomes in patients with neuroendocrine tumors in the United States. JAMA Oncology 3 1335–1342. (https://doi.org/10.1001/jamaoncol.2017.0589)
Dayton TL, Alcala N, Moonen L, den Hartigh L, Geurts V, Mangiante L, Lap L, Dost AFM, Beumer J, Levy S, et al.2023 Druggable growth dependencies and tumor evolution analysis in patient-derived organoids of neuroendocrine neoplasms from multiple body sites. Cancer Cell 41 2083–2099.e9. (https://doi.org/10.1016/j.ccell.2023.11.007)
Di Giacinto P, Rota F, Rizza L, Campana D, Isidori A, Lania A, Lenzi A, Zuppi P & & Baldelli R 2018 Chromogranin A: from laboratory to clinical aspects of patients with neuroendocrine tumors. International Journal of Endocrinology 2018 8126087. (https://doi.org/10.1155/2018/8126087)
Ear PH, Li G, Wu M, Abusada E, Bellizzi AM & & Howe JR 2019 Establishment and characterization of small bowel neuroendocrine tumor spheroids. Journal of Visualized Experiments 152 e60303. (https://doi.org/10.3791/60303)
Egberts JH, Raza GS, Wilgus C, Teyssen S, Kiehne K & & Herzig KH 2020 Release of cholecystokinin from rat intestinal mucosal cells and the enteroendocrine cell line STC-1 in response to maleic and succinic acid, fermentation products of alcoholic beverages. International Journal of Molecular Sciences 21 589. (https://doi.org/10.3390/ijms21020589)
Elias E, Ardalan A, Lindberg M, Reinsbach SE, Muth A, Nilsson O, Arvidsson Y & & Larsson E 2021 Independent somatic evolution underlies clustered neuroendocrine tumors in the human small intestine. Nature Communications 12 6367. (https://doi.org/10.1038/s41467-021-26581-5)
Ellis LM, Samuel S & & Sceusi E 2010 Varying opinions on the authenticity of a human midgut carcinoid cell line--letter. Clinical Cancer Research 16 5365–5366. (https://doi.org/10.1158/1078-0432.CCR-10-2550)
Erickson JD, Schafer MK, Bonner TI, Eiden LE & & Weihe E 1996 Distinct pharmacological properties and distribution in neurons and endocrine cells of two isoforms of the human vesicular monoamine transporter. PNAS 93 5166–5171. (https://doi.org/10.1073/pnas.93.10.5166)
Forsythe SD, Pu T, Andrews SG, Madigan JP & & Sadowski SM 2023 Models in pancreatic neuroendocrine neoplasms: current perspectives and future directions. Cancers 15 3756. (https://doi.org/10.3390/cancers15153756)
Francis JM, Kiezun A, Ramos AH, Serra S, Pedamallu CS, Qian ZR, Banck MS, Kanwar R, Kulkarni AA, Karpathakis A, et al.2013 Somatic mutation of CDKN1B in small intestine neuroendocrine tumors. Nature Genetics 45 1483–1486. (https://doi.org/10.1038/ng.2821)
Galli G, Zonefrati R, Gozzini A, Mavilia C, Martineti V, Tognarini I, Nesi G, Marcucci T, Tonelli F, Tommasi M, et al.2006 Characterization of the functional and growth properties of long-term cell cultures established from a human somatostatinoma. Endocrine-Related Cancer 13 79–93. (https://doi.org/10.1677/erc.1.00988)
Gaudenzi G & & Vitale G 2019 Transplantable zebrafish models of neuroendocrine tumors. Annales d’Endocrinologie 80 149–152. (https://doi.org/10.1016/j.ando.2019.04.013)
Gaudenzi G, Albertelli M, Dicitore A, Wurth R, Gatto F, Barbieri F, Cotelli F, Florio T, Ferone D, Persani L, et al.2017 Patient-derived xenograft in zebrafish embryos: a new platform for translational research in neuroendocrine tumors. Endocrine 57 214–219. (https://doi.org/10.1007/s12020-016-1048-9)
Grozinsky-Glasberg S, Shimon I & & Rubinfeld H 2012 The role of cell lines in the study of neuroendocrine tumors. Neuroendocrinology 96 173–187. (https://doi.org/10.1159/000338793)
Haley SA, O’Hara BA, Banerjee R & & Atwood WJ 2010 Unique susceptibility of a human lung carcinoid tumor cell line to infection with BK virus. Virus Research 149 128–132. (https://doi.org/10.1016/j.virusres.2010.01.013)
Herring B, Jang S, Whitt J, Goliwas K, Aburjania Z, Dudeja V, Ren B, Berry J, Bibb J, Frost A, et al.2021 Ex vivo modeling of human neuroendocrine tumors in tissue surrogates. Frontiers in Endocrinology 12 710009. (https://doi.org/10.3389/fendo.2021.710009)
Hofving T, Arvidsson Y, Almobarak B, Inge L, Pfragner R, Persson M, Stenman G, Kristiansson E, Johanson V & & Nilsson O 2018 The neuroendocrine phenotype, genomic profile and therapeutic sensitivity of GEPNET cell lines. Endocrine-Related Cancer 25 X1–X2. (https://doi.org/10.1530/ERC-17-0445e)
Hofving T, Elias E, Rehammar A, Inge L, Altiparmak G, Persson M, Kristiansson E, Johansson ME, Nilsson O & & Arvidsson Y 2021a SMAD4 haploinsufficiency in small intestinal neuroendocrine tumors. BMC Cancer 21 101. (https://doi.org/10.1186/s12885-021-07786-9)
Hofving T, Liang F, Karlsson J, Yrlid U, Nilsson JA, Nilsson O & & Nilsson LM 2021b The microenvironment of small intestinal neuroendocrine tumours contains lymphocytes capable of recognition and activation after expansion. Cancers 13 4 3 05. (https://doi.org/10.3390/cancers13174305)
Isgro MA, Bottoni P & & Scatena R 2015 Neuron-specific enolase as a biomarker: biochemical and clinical aspects. Advances in Experimental Medicine and Biology 867 125–143. (https://doi.org/10.1007/978-94-017-7215-0_9)
Jakobsen AM, Andersson P, Saglik G, Andersson E, Kolby L, Erickson JD, Forssell-Aronsson E, Wangberg B, Ahlman H & & Nilsson O 2001 Differential expression of vesicular monoamine transporter (VMAT) 1 and 2 in gastrointestinal endocrine tumours. Journal of Pathology 195 463–472. (https://doi.org/10.1002/path.973)
Jimenez-Franco LD, Glatting G, Prasad V, Weber WA, Beer AJ & & Kletting P 2021 Effect of tumor perfusion and receptor density on tumor control probability in (177)Lu-DOTATATE therapy: an in silico analysis for standard and optimized treatment. Journal of Nuclear Medicine 62 92–98. (https://doi.org/10.2967/jnumed.120.245068)
Kamilaris CDC & & Stratakis CA 2019 Multiple endocrine neoplasia Type 1 (MEN1): an update and the significance of early genetic and clinical diagnosis. Frontiers in Endocrinology 10 339. (https://doi.org/10.3389/fendo.2019.00339)
Kawasaki K, Fujii M & & Sato T 2018 Gastroenteropancreatic neuroendocrine neoplasms: genes, therapies and models. Disease Models and Mechanisms 11 dmm029595. (https://doi.org/10.1242/dmm.029595)
Kerstrom G, Hellman P & & Hessman O 2005 Midgut carcinoid tumours: surgical treatment and prognosis. Best Practice and Research. Clinical Gastroenterology 19 717–728. (https://doi.org/10.1016/j.bpg.2005.05.005)
Kölby L, Bernhardt P, Ahlman H, Wängberg B, Johanson V, Wigander A, Forssell-Aronsson E, Karlsson S, Ahrén B, Stenman G, et al.2001 A transplantable human carcinoid as model for somatostatin receptor-mediated and amine transporter-mediated radionuclide uptake. American Journal of Pathology 158 745–755. (https://doi.org/10.1016/S0002-9440(10)64017-5)
Krishnamurthy S & & Dayal Y 1997 Immunohistochemical expression of transforming growth factor alpha and epidermal growth factor receptor in gastrointestinal carcinoids. American Journal of Surgical Pathology 21 327–333. (https://doi.org/10.1097/00000478-199703000-00009)
Kytola S, Hoog A, Nord B, Cedermark B, Frisk T, Larsson C & & Kjellman M 2001 Comparative genomic hybridization identifies loss of 18q22-qter as an early and specific event in tumorigenesis of midgut carcinoids. American Journal of Pathology 158 1803–1808. (https://doi.org/10.1016/S0002-9440(10)64136-3)
Laskaratos FM, Levi A, Schwach G, Pfragner R, Hall A, Xia D, von Stempel C, Bretherton J, Thanapirom K, Alexander S, et al.2021 Transcriptomic profiling of in vitro tumor-stromal cell paracrine crosstalk identifies involvement of the integrin signaling pathway in the pathogenesis of mesenteric fibrosis in human small intestinal neuroendocrine neoplasms. Frontiers in Oncology 11 629665. (https://doi.org/10.3389/fonc.2021.629665)
Lu CF, Hsu FT, Hsieh KLC, Kao YJ, Cheng SJ, Hsu JBK, Tsai PH, Chen RJ, Huang CC, Yen Y, et al.2018 Machine learning-based radiomics for molecular subtyping of gliomas. Clinical Cancer Research 24 4429–4436. (https://doi.org/10.1158/1078-0432.CCR-17-3445)
Matsushima H, Kaibori M, Hatta M, Ishizaki M, Nakatake R, Okumura T, Yoshii K & & Todo T 2019 Efficacy of a third-generation oncolytic herpes simplex virus in neuroendocrine tumor xenograft models. Oncotarget 10 7132–7141. (https://doi.org/10.18632/oncotarget.27391)
Miettinen M 1987 Synaptophysin and neurofilament proteins as markers for neuroendocrine tumors. Archives of Pathology and Laboratory Medicine 111 813–818.
Mjones P, Sagatun L, Nordrum IS & & Waldum HL 2017 Neuron-specific enolase as an immunohistochemical marker is better than its reputation. Journal of Histochemistry and Cytochemistry 65 687–703. (https://doi.org/10.1369/0022155417733676)
Modlin IM, Shapiro MD & & Kidd M 2004 Siegfried Oberndorfer: origins and perspectives of carcinoid tumors. Human Pathology 35 1440–1451. (https://doi.org/10.1016/j.humpath.2004.09.018)
Papotti M, Bongiovanni M, Volante M, Allia E, Landolfi S, Helboe L, Schindler M, Cole SL & & Bussolati G 2002 Expression of somatostatin receptor types 1–5 in 81 cases of gastrointestinal and pancreatic endocrine tumors. A correlative immunohistochemical and reverse-transcriptase polymerase chain reaction analysis. Virchows Archiv 440 461–475. (https://doi.org/10.1007/s00428-002-0609-x)
Pavel M, Oberg K, Falconi M, Krenning EP, Sundin A, Perren A, Berruti A & ESMO Guidelines Committee. Electronic address: clinicalguidelines@esmo.org 2020 Gastroenteropancreatic neuroendocrine neoplasms: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Annals of Oncology 31 844–860. (https://doi.org/10.1016/j.annonc.2020.03.304)
Pfragner R, Wirnsberger G, Niederle B, Behmel A, Rinner I, Mandl A, Wawrina F, Luo J, Adamiker D, Hoger H, et al.1996 Establishment of a continuous cell line from a human carcinoid of the small intestine (KRJ-I). International Journal of Oncology 8 513–520. (https://doi.org/10.3892/ijo.8.3.513)
Pfragner R, Behmel A, Hoger H, Beham A, Ingolic E, Stelzer I, Svejda B, Moser VA, Obenauf AC, Siegl V, et al.2009 Establishment and characterization of three novel cell lines - P-STS, L-STS, H-STS - derived from a human metastatic midgut carcinoid. Anticancer Research 29 1951–1961.
Pritchard DM 2022 New mouse model suggests that some neuroendocrine tumors may originate from neural crest-derived cells. Cellular and Molecular Gastroenterology and Hepatology 14 1170–1171. (https://doi.org/10.1016/j.jcmgh.2022.08.005)
Rindi G, Grant SG, Yiangou Y, Ghatei MA, Bloom SR, Bautch VL, Solcia E & & Polak JM 1990 Development of neuroendocrine tumors in the gastrointestinal tract of transgenic mice. Heterogeneity of hormone expression. American Journal of Pathology 136 1349–1363.
Sasaki T, Hiroki K & & Yamashita Y 2013 The role of epidermal growth factor receptor in cancer metastasis and microenvironment. BioMed Research International 2013 546318. (https://doi.org/10.1155/2013/546318)
Stilling GA, Zhang H, Ruebel KH, Leontovich AA, Jin L, Tanizaki Y, Zhang S, Erickson LA, Hobday T & & Lloyd RV 2007 Characterization of the functional and growth properties of cell lines established from ileal and rectal carcinoid tumors. Endocrine Pathology 18 223–232. (https://doi.org/10.1007/s12022-007-9001-3)
Van Buren G 2nd, Rashid A, Yang AD, Abdalla EK, Gray MJ, Liu W, Somcio R, Fan F, Camp ER, Yao JC, et al.2007 The development and characterization of a human midgut carcinoid cell line. Clinical Cancer Research 13 4704–4712. (https://doi.org/10.1158/1078-0432.CCR-06-2723)
Vitale G, Carra S, Alessi Y, Campolo F, Pandozzi C, Zanata I, Colao A, Faggiano A & & On Behalf Of The Nike Group 2023 Carcinoid syndrome: preclinical models and future therapeutic strategies. International Journal of Molecular Sciences 24 3610. (https://doi.org/10.3390/ijms24043610)
Watanabe H, Fujishima F, Komoto I, Imamura M, Hijioka S, Hara K, Yatabe Y, Kudo A, Masui T, Tsuchikawa T, et al.2022 Somatostatin receptor 2 expression profiles and their correlation with the efficacy of somatostatin analogues in gastrointestinal neuroendocrine tumors. Cancers 14 775. (https://doi.org/10.3390/cancers14030775)
Wedin M, Tsoli M, Wallin G, Janson ET, Koumarianou A, Kaltsas G & & Daskalakis K 2022 Heterogeneity of small intestinal neuroendocrine tumors metastasis: biologic patterns of a series with Virchow’s node involvement. Cancers 14 913. (https://doi.org/10.3390/cancers14040913)
White BE, Rous B, Chandrakumaran K, Wong K, Bouvier C, Van Hemelrijck M, George G, Russell B, Srirajaskanthan R & & Ramage JK 2022 Incidence and survival of neuroendocrine neoplasia in England 1995–2018: a retrospective, population-based study. Lancet Regional Health. Europe 23 100510. (https://doi.org/10.1016/j.lanepe.2022.100510)
Wu S, Zheng J, Li Y, Yu H, Shi S, Xie W, Liu H, Su Y, Huang J & & Lin T 2017 A radiomics nomogram for the preoperative prediction of lymph node metastasis in bladder cancer. Clinical Cancer Research 23 6904–6911. (https://doi.org/10.1158/1078-0432.CCR-17-1510)
Xie F, Ling L, van Dam H, Zhou F & & Zhang L 2018 TGF-beta signaling in cancer metastasis. Acta Biochimica et Biophysica Sinica 50 121–132. (https://doi.org/10.1093/abbs/gmx123)
Yang K, Ulich T, Cheng L & & Lewin KJ 1983 The neuroendocrine products of intestinal carcinoids. An immunoperoxidase study of 35 carcinoid tumors stained for serotonin and eight polypeptide hormones. Cancer 51 1918–1926. (https://doi.org/10.1002/1097-0142(19830515)51:10<1918::aid-cncr2820511027>3.0.co;2-y)
Zaid NRR, Kletting P, Winter G, Prasad V, Beer AJ & & Glatting G 2021 A physiologically based pharmacokinetic model for in vivo alpha particle generators targeting neuroendocrine tumors in mice. Pharmaceutics 13 2132. (https://doi.org/10.3390/pharmaceutics13122132)
