Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Nov 1.
Published in final edited form as: J Pathol. 2020 Sep 19;252(3):252–262. doi: 10.1002/path.5515

Molecular characterization of organoids derived from pancreatic intraductal papillary mucinous neoplasms

Bo Huang 1,2,#, Maria A Trujillo 1,#, Kohei Fujikura 1, Miaozhen Qiu 1,3, Fei Chen 4, Matthäus Felsenstein 1,5, Cancan Zhou 1, Michael Skaro 1, Christian Gauthier 1, Anne Macgregor-Das 1, Danielle Hutchings 1, Seung-Mo Hong 1,6, Ralph H Hruban 1,7, James R Eshleman 1,7, Elizabeth D Thompson 1, Alison P Klein 1,4,7, Michael Goggins 1,7,8, Laura D Wood 1,7,*, Nicholas J Roberts 1,7,*
PMCID: PMC8162794  NIHMSID: NIHMS1703650  PMID: 32696980

Abstract

Intraductal papillary mucinous neoplasms (IPMNs) are a commonly identified non-invasive cyst-forming pancreatic neoplasms with the potential to progress into invasive pancreatic adenocarcinoma. There are few in vitro models with which to study the biology of IPMNs and their progression to invasive carcinoma. Therefore, we generated a living biobank of organoids from 7 normal pancreatic ducts and 10 IPMNs. We characterized 8 IPMN organoid samples using whole genome sequencing and characterized 5 IPMN organoids and 7 normal pancreatic duct organoids using transcriptome sequencing. We identified an average of 11,344 somatic mutations in the genomes of organoids derived from IPMNs, with one sample harboring 61,537 somatic mutations enriched for T->C transitions and T->A transversions. Recurrent coding somatic mutations were identified in 15 genes, including KRAS, GNAS, RNF43, PHF3, and RBM10. The most frequently mutated genes were KRAS, GNAS, and RNF43, with somatic mutations identified in 6 (75%), 4 (50%), and 3 (37.5%) IPMN organoid samples respectively. On average, we identified 36 structural variants in IPMN derived organoids, and none had an unstable phenotype (>200 structural variants). Transcriptome sequencing identified 28 genes differentially expressed between normal pancreatic duct organoid and IPMN organoid samples. The most significantly upregulated and downregulated genes were CLDN18 and FOXA1. Immunohistochemical analysis of FOXA1 expression in 112 IPMNs, 113 mucinous cystic neoplasms, and 145 pancreatic ductal adenocarcinomas demonstrated statistically significant loss of expression in low-grade IPMNs (p < 0.0016), mucinous cystic neoplasms (p < 0.0001), and pancreatic ductal adenocarcinoma of any histologic grade (p < 0.0001) compared to normal pancreatic ducts. These data indicate that FOXA1 loss of expression occurs early in pancreatic tumorigenesis. Our study highlights the utility of organoid culture to study the genetics and biology normal pancreatic duct and IPMNs.

Keywords: Intraductal papillary mucinous neoplasm, pancreas, pancreatic, precursor, cancer, organoids, DNA, RNA, mutations, expression

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is a lethal disease with a 5-year survival rate of only 9 percent, and PDAC is predicted to become the second leading cause of cancer-related death in the United States by 2030 [1,2]. PDAC develops from non-invasive precursor lesions in a process that takes many years, if not decades [3]. These pancreatic cancer precursor lesions can be microscopic, as is the case for pancreatic intraepithelial neoplasia (PanIN), or macroscopic, as is the case for intraductal papillary mucinous neoplasms (IPMNs). As IPMNs are macroscopic, and incidental pancreatic cystic lesions suggestive of IPMNs are identified in up to 2.6% of the population when imaged by contrast-enhanced multi-detector computed tomography, IPMNs present an ideal opportunity for early clinical intervention [4].

PDACs have been extensively characterized by high-throughput genomic, transcriptomic, and proteomic approaches. This has led to many new insights in pancreatic tumorigenesis, including the identification of genetic alterations driving tumorigenesis, molecular subgroups that may respond to specific therapies, and evolutionary features underlying pancreatic cancer progression indicating a broad time window of opportunity for early detection [5–12]. However, comparatively little is known about the genetics and biology of pancreatic precursor lesions such as IPMNs. Previous genomic studies to characterize pathogenic germline variants and somatic alterations in IPMN have used whole exome or targeted gene panels and identified key drivers previously associated with PDAC tumorigenesis (KRAS, CDKN2A, TP53, SMAD4) as well as drivers of the IPMN pathway (GNAS, RNF43) [13–15]. Furthermore, microarray-based RNA expression analyses in IPMNs highlighted differentially expressed genes associated with histologic grade and putatively, therefore, IPMN progression [16,17]. Similarly, immunohistochemistry to detect expression of candidate genes has been used to identify differentially expressed proteins in IPMN compared to normal duct and invasive carcinoma [16,18]. While these studies provided important insights into the origin and development of IPMNs, they did not comprehensively characterize somatic alterations that occur in the genome and transcriptome of these earliest stages in pancreatic tumorigenesis.

The development of model systems to study pancreatic tumorigenesis is essential to understand the physiologic effects of genetic alterations and of changes in gene expression. Unfortunately, there are few reported examples of in vitro models of normal pancreatic duct and pancreatic precursor lesions such as IPMN, and those that do exist are either experimentally immortalized cells or derived from patients with concurrent invasive carcinoma [19,20]. Furthermore, while several genetically engineered mouse models of IPMN have been described, in vivo models are expensive and time consuming to generate and do not have the genetic complexity present in human IPMN [21,22]. This lack of disease appropriate models has contributed to the paucity of research into pancreatic cancer precursor lesions.

Three-dimensional organoid culture is a recently described ex vivo model system, in which small clusters of neoplastic cells are grown in matrix gels, maintaining similar appearance and functionality to the original tissue. From the pancreas, organoids can be generated from small amounts of tissue such as endoscopic fine-needle aspiration or biopsy samples in a short period of time, thereby allowing for additional testing (such as pharmacotyping) to complement the pathological diagnosis [23]. Organoid culture can also be used to enrich ductal or neoplastic cells for DNA or RNA analyses without contamination by associated mesenchymal or immune cells, which are typically eliminated in early passages [24]. With its combination of fidelity to three-dimensional structure and ease of experimental manipulation, the organoid model represents a valuable tool for several lines of investigation, including validation of genetic alterations required for cancer progression and correlation of in vitro behavior with treatment response and outcome in vivo. However, the ability to culture and propagate organoids from human pancreatic precursor lesions has not been systematically investigated.

In this study, we generate a living biobank of organoids derived from 7 normal human pancreatic ducts and 10 human IPMN samples. We also present the characterization of somatic alterations in these IPMN organoids using whole genome sequencing. Furthermore, we use transcriptomic sequencing to identify genes differentially expressed between normal pancreatic duct and IPMN organoids.

Materials and methods

Additional details are provided in Supplementary material and methods

Study participants and ethics statement

This study was reviewed and approved by the Johns Hopkins University Institutional Review Board. All patients gave written informed consent. Seventeen patients were enrolled in this study. All patients had undergone pancreatic resection at the Johns Hopkins Hospital between January 2016 and September 2018 (supplementary material, Table S1). IPMN organoids were derived from 10 patients, designated H23 to H79. Normal pancreatic duct organoids were derived from macro-dissected, grossly normal, main pancreatic duct from 7 unmatched patients, designated N22 to N92.

Whole genome sequencing

Whole genome sequencing was performed as previously described [25]. DNA from eight IPMN organoids (in passages 1–4) and matched primary non-neoplastic duodenal tissue were whole genome sequenced by the Next Generation Sequencing Core of the Johns Hopkins Sidney Kimmel Comprehensive Cancer Center as previously described [25]. In brief, sequence libraries were generated using the TruSeq DNA Nano Library Prep Kit (Illumina, San Diego, CA, USA; catalog no. 20015964) and sequenced on a HiSeq2500 (Illumina) to generate 2×150 bp reads.

Whole exome sequencing

Whole exome sequencing was performed as previously described [25]. Briefly, paired IPMN (tumor) and duodenum (normal) DNA from H55, H68, and H79 were extracted from FFPE tissue and whole exome sequenced by the Next Generation Sequencing Core of the Johns Hopkins Sidney Kimmel Comprehensive Cancer Center. Exome capture and library preparation were performed with Agilent SureSelect Target Enrichment V5-post (Agilent, Santa Clara, CA, USA). DNA libraries were sequenced on an Illumina NovoSeq instrument to generate 2 × 150 bp PE sequence reads.

RNAseq

RNA-Seq was performed as previously described [25]. RNA from IPMN organoids and normal pancreatic duct organoids (in passages 2 or 3) were sequenced by Genewiz Inc. (Plainsfield, NJ, USA). In brief, mRNA was enriched by polyA selection before random priming, cDNA synthesis, A-tailing, adapter ligation, and PCR amplification of final sequence libraries. Libraries were sequenced on an Illumina HiSeq instrument to generate 2×150-base pair (bp) reads.

Immunohistochemistry

Tissue microarrays (TMAs) containing IPMNs, mucinous cystic neoplasms (MCNs), PDAC, and normal pancreatic ducts were used to verify RNAseq findings at the protein level by immunohistochemistry (IHC). TMAs were grouped by lesion type and IPMNs were grouped by grade. TMAs included tissue samples from 112 patients with IPMN (61 IPMN-LG and 51 IPMN-HG), 113 patients with MCN (103 MCN-LG and 10 MCN-HG/CA), 145 patients with PDAC (77 well-moderately differentiated and 68 poorly differentiated. Normal pancreatic duct on IPMN, MCN, and PDAC TMAs were scored and used for comparisons, 49 normal pancreatic ducts were identified on the IPMN TMA, 99 on the MCN TMA, and 102 on the PDAC TMA. Immunolabeling for FOXA1 was performed on formalin‐fixed, paraffin (FFPE) embedded sections on a Ventana Discovery Ultra autostainer (Roche Diagnostics, Indianapolis, IN, USA). In brief, following dewaxing and rehydration on board, epitope retrieval was performed using Ventana Ultra CC1 buffer (Roche Diagnostics; catalog no. 6414575001) at 96 °C for 64 min. The primary antibody, anti‐FOXA1 (1:300 dilution; Abcam, Cambridge, MA, USA; catalog no. ab170933) was applied at 37 °C for 60 min. Primary antibodies were detected using an anti-rabbit HQ detection system (Roche Diagnostics; catalog no. 7017936001 and 7017812001) and Chromomap DAB IHC detection kit (Roche Diagnostics; catalog no. 5266645001). Sections were counterstained with Mayer’s hematoxylin, dehydrated and mounted.

IHC results were evaluated using a H-score, a semi-quantitative, non-blinded approach. In brief, more than 100 cells were counted for each case, and the H-score was subsequently generated by adding the percentage of strongly stained nuclei (3X), the percentage of moderately stained nuclei (2X), and the percentage of weakly stained nuclei (1X), giving a possible range of 0–300 [26]. The score was independently obtained by two authors (BH and KF) and the mean of the two values used for statistical analysis. Inter-observer differences were < 5%.

Results

Whole genome sequencing

Eight IPMN derived organoids and matched normal tissue samples were analyzed by whole genome sequencing (supplementary material, Table S1). All IPMN organoids had either an oncogenic KRAS or an oncogenic GNAS mutation as determined by PCR amplification and Sanger sequencing prior to whole genome sequencing analysis to confirm the neoplastic origin of the samples (supplementary material, Table S2). The IPMN organoids sequenced included three from IPMNs with low-grade dysplasia (IPMN-LG) and five from IPMNs with high-grade dysplasia (IPMN-HG) (supplementary material, Table S1). The mean number of reads per sample was 356,581,475 (range: 304,627,462 – 434,765,537) and the mean alignment rate was 99.6% (range: 97.3% – 99.8%). The mean coverage was 33.8X (range: 27.2 – 41.8X). The mean of bases covered at ≥ 10X was 98.2% (range 97.8% – 98.9%) and covered at ≥ 20X was 93.7% (range: 88.5% – 96.9%) (supplementary material, Table S3).

A total of 90,751 somatic mutations (single base substitutions and INDELs) were identified across all eight IPMN organoids sequenced. The mean number of somatic mutations per sample was 11,344 (range: 1,419 – 61,537) (Figure 1A). One sample (H79) had a much higher number of somatic mutations than the rest of the cohort (61,537); when this sample was excluded, the mean number of somatic mutations per sample was 4,173 (range: 1,419 – 8,927). After excluding H79, there was no significant difference in the number of somatic mutations in organoids derived from IPMN-LG and IPMN-HG (mean number of somatic mutations 2,932 and 5,104 respectively; p = 0.3712). The mutation signatures and transition/transversion patterns are similar among all IPMN derived organoids except for H79 (supplementary material, Figure S1, and Figure 1B). Furthermore, while the mutation signatures of the 7 IPMNs (when H79 was excluded) were predominated by putative age-related changes such as C->T transitions, the mutation signature of H79 was unique, with enrichment of T->C transitions and T->A transversions. To test for the possibility that the increased mutations were a result of a mismatch repair defect, we tested H79 organoid sample for microsatellite instability using a PCR-based assay, which did not demonstrate microsatellite instability in this sample.

Figure 1.

Figure 1.

Open in a new tab

Somatic mutations identified in whole genome sequenced IPMN organoids. (A) The number of somatic mutations in coding and non-coding regions by sample. (B) Profile of transition/transversion in somatic mutations by sample. (C) Genes with recurrent somatic coding mutations by sample.

The mean number of somatic mutations in coding regions was 52 (range: 29 – 90) excluding H79. The mean number of non-silent somatic mutations was 36 (range: 16 – 69) (supplementary material, Figure S2). We found that 16 genes had non-silent mutations in ≥ 2 IPMN organoid samples, including H79. The most commonly mutated genes were KRAS (6 IPMN organoid samples), GNAS (4 IPMN organoid samples), and RNF43 (3 IPMN organoid samples). The remaining 13 genes (CYP4Z1, DNAH9, HLA-DQB2, KIAA1109, MUC4, MUC12, PHF3, RBM10, RXFP2, SLC7A8, SLC9A3, ZNF260, ZNF835) were somatically mutated in two IPMN organoid samples (Figure 1C). Somatic mutations were also identified in established pancreatic cancer driver genes CDKN2A p.H83Y (1 IPMN organoid sample) and TP53 p.R174X (1 IPMN organoid sample) (supplementary material, Table S4). Moreover, inactivating somatic mutations were identified genes in that may play a role in pancreatic tumorigenesis (AXIN2, INPPL1, HES1, and RREB1). Importantly, average allele frequencies for coding and splice site somatic mutations ranged from 0.11 for H40 to 0.38 for H60 (supplementary material, Table S4), suggesting ongoing heterogeneity in organoid culture, an observation supported by previous studies [27].

Of note, IPMN organoid sample (H40) had a KRAS p.G12V mutation at 0.0455 allele fraction by whole genome sequencing in addition to a KRAS p.R73R mutation at 0.0968 allele fraction. This observation may be due to subclonal mutation or the presence of contaminating non-neoplastic cells in the organoid culture. Confirming the neoplastic origin of the organoids from H40, we identified 45 coding somatic mutations in this sample with a mean allele frequency of 0.11 (range: 0.06 – 0.26), including in MUC4, MUC12, and PHF3, genes that were somatically mutated in at least one other IPMN organoid sample (Figure 1C). As expected, somatic mutations included known oncogenic hotspot mutations such as KRAS p.G12D (4 IPMN organoid samples), KRAS p.Q61R (1 IPMN organoid sample) and KRAS p.Q61H (1 IPMN organoid sample), as well as GNAS p. R201H (3 IPMN organoid samples) and GNAS p.R201C (1 IPMN organoid sample) (supplementary material, Figure S3). The only other recurrent mutation was SLC7A8 p.T79P in 2 IPMN organoid samples (supplementary material, Figure S3 and Table S4).

We identified a total of 250 somatic SVs in the 8 IPMN organoid samples. The mean number of somatic SVs per sample was 31 (range: 9 – 82). These included 40 duplications, 111 deletions, 58 inversions and 41 translocations (supplementary material, Figure S4 and Table S5). 16 SVs involving the same gene were identified in ≥ 2 samples, including 2 SVs occurring at the same positions in CMPK1–CLIC5 and DSTYK. The most frequently observed SVs were a duplication encompassing CCNYL3 and ANKRD26P1, a deletion of KIAA1671, and an inversion involving LINC02254, each in three IPMN organoids. Another recurrent SV (2 IPMNs) with the most supporting reads (PR>20 and SR>20) was an inversion in the gene DSTYK, encoding a dual serine/threonine and tyrosine protein kinase that is thought to regulate cell death (supplementary material, Figure S5 and Table S5). As the number of SVs per IPMN organoid sample was < 100 and had no obvious pattern of distribution in the genome, none would meet the classification of unstable used in whole genome sequenced PDAC [8].

As pathogenic germline variants in pancreatic cancer susceptibility genes and hereditary cancer predisposition genes have been identified in patients with surgically resected IPMN, we assessed germline variants in sequenced IPMN organoids [15]. However, no pathogenic germline variants in 94 hereditary cancer predisposition genes were identified in this sample set.

Whole exome sequencing

We conducted whole exome sequencing of FFPE IPMN and non-tumor archival tissue from H55, H68, and H79. Identity between whole exome sequenced tissue samples and whole genome sequenced organoid samples was confirmed by comparing germline variant calls. Interestingly, no somatic mutations were shared between FFPE IPMN and organoid sample for H55 and H68, indicating that a different subclone was sequenced from FFPE tissue than grew out in organoid culture. FFPE IPMN and organoid from H79 were both hypermutated, an uncommon finding in IPMN, with 369 and 342 coding or splice site somatic mutations identified in each sample respectively (supplementary material, Table S4 and Table S6) [28]. 58 mutations identified in the FFPE IPMN were present in the organoid sample, highlighting that different subclone(s) of H79 were cultured than were whole exome sequenced.

RNA-Seq

RNA from five IPMN organoids and seven normal pancreatic duct organoids were sequenced. IPMN organoid samples include three from patients with IPMN-LG and two from patients with IPMN-HG (supplementary material, Table S1). Normal duct organoid samples were derived from two patients with branch duct IPMN, four patients with pancreatic neuroendocrine tumors, and one patient with metastatic clear cell renal cell carcinoma. In each case, the harvested pancreatic duct was grossly normal main duct, and analysis of sequence reads in IGV covering exon 2 and exon 3 of KRAS, and exon 8 and exon 9 of GNAS, did not identify any oncogenic mutations, confirming the non-neoplastic origin of the normal pancreatic duct organoids used for RNA-Seq. The mean number of reads per sample was 47,248,301 (30,775,100 – 60,882,990) and the mean quality score is 37.62 (37.52 – 37.75). 89.67% (89.33% – 90.15%) of bases with quality score is above 30 (>=30). The mean alignment rate is 97.72% (97.53% – 98.02%) (supplementary material, Table S7)

Differential gene expression analysis identified 28 genes (29 transcripts) as aberrantly expressed in IPMN derived organoids compared to normal pancreatic duct organoids. These included 15 transcripts representing 14 genes upregulated in IPMN organoids, and 14 transcripts representing 14 genes downregulated in IPMN organoids (Figure 2, Figure 3, and supplementary material, Table S8). Of these genes, CLDN18 was the most significant upregulated gene (p = 6.89E-06, q =0.031) and FOXA1 was the most significant downregulated gene (p = 6.10E-08, q = 0.0009).

Figure 2.

Figure 2.

Open in a new tab

Volcano plot showing expression profile for transcripts in IPMN organoids versus normal pancreatic duct organoids. Q value of 0.05 indicted by red dashed line. Transcripts with q value < 0.05 and fold-change ± 4 highlighted in red with gene symbol.

Figure 3.

Figure 3.

Open in a new tab

Heatmap of 29 differentially expressed transcripts in IPMN organoids versus normal pancreatic duct organoids grouped by sample.

Fusion gene analysis identified 10 fusion events in the RNA-Seq data, including 4 putatively somatic fusion events present only in IPMN organoid samples and 1 putative germline fusion event present in both IPMN and normal pancreatic duct organoid samples. One IPMN organoid sample (H60) had both whole genome sequence and RNA-Seq data available for analysis. In this case, 3 fusions identified by RNA-Seq were not supported by whole genome sequence data, highlighting the discrepancy between fusion and rearrangement calling pipelines. (supplementary material, Table S9).

Immunohistochemistry

To validate FOXA1 expression observed in RNA-Seq data we conducted immunohistochemistry on FFPE IPMN samples from 4 patients. The IPMNs which showed decreased FOXA1 expression in the RNA-Seq analysis also showed decreased protein expression by immunohistochemistry, with either diffuse moderate (rather than strong) expression or focal areas of loss of expression. In order to validate the most differentially expressed genes from our RNA-Seq analysis, we created a human tissue microarray containing IPMN tissue from 112 patients. Because expression of CLDN18 has already been evaluated at the protein level in IPMNs [18], we focused on FOXA1, which was the most significantly downregulated gene in IPMN compared to normal pancreatic duct. In our TMA sections, immunohistochemistry demonstrated strong and consistent nuclear FOXA1 expression in both normal pancreatic ductal and acinar cells. When all IPMNs were considered together, FOXA1 expression (as determined by H-score) was not significantly lower in IPMNs of any grade compared to normal pancreatic ductal cells (Figure 4, p = 0.68). However, low-grade IPMNs had significantly lower expression compared to normal pancreatic duct (Figure 4, p < 0.0016). These results suggest that FOXA1 downregulation at the RNA level leads to decreased protein expression in low-grade IPMNs.

Figure 4.

Figure 4.

Open in a new tab

The expression of FOXA1 in normal pancreas and IPMNs. (A) H&E and (B) immunohistochemical image of normal pancreatic duct and surrounding acinar cells showing strong nuclear expression of FOXA1. (C) H&E and (D) immunohistochemical image of IPMN showing variable weak-moderate expression of FOXA1. (E) H&E and (F) immunohistochemical image of IPMN showing strong expression of FOXA1. (G) H-score for immunohistochemical staining of FOXA1 in IPMN of any grade compared to normal duct and (H) IPMN-HG compared to IPMN-LG and normal duct.

To ascertain the role of FOXA1 in other pancreatic neoplasms, we evaluated FOXA1 expression in MCNs and PDACs, as determined by H-score. A statistically significant decrease in FOXA1 expression was observed for MCNs (supplementary material, Figure S6, p < 0.0001) and PDACs (supplementary material, Figure S7, p < 0.0001), compared to normal pancreatic duct irrespective of histologic grade. These data indicate that decreased FOXA1 expression is present in a broad range of pancreatic neoplasms.

Discussion

We created a unique set of normal pancreatic duct and IPMN organoids. We then used whole genome sequencing to identify recurrent somatic mutations and structural variants in these organoids. As expected, prevalent somatic mutations in KRAS, GNAS, CDKN2A, TP53, and RNF43 were observed. We also identified 13 other genes with recurrent somatic mutations in IPMN (CYP4Z1, DNAH9, HLA-DQB2, KIAA1109, MUC4, MUC12, PHF3, RBM10, RXFP2, SLC7A8, SLC9A3, ZNF260, and ZNF835). RBM10, encoding an RNA binding protein, has been shown to be mutated in multiple cancer types including IPMN [29]. Of note, several other genes with inactivating somatic mutations in our study may play a role in the development of IPMN. These include AXIN2, a Wnt pathway regulator, INPPL1, a component of FGFR signaling, HES1, a target of Notch signaling that can affect Hedgehog signaling, and RREB1, a regulator of zinc homeostasis recurrently mutated in PDAC [30–34]. Additional studies to determine the prevalence of somatic alterations in these genes in larger cohorts of IPMNs and their functional role in pancreatic tumorigenesis are warranted.

Interestingly, one high-grade IPMN organoid sample (H79) had an excess of somatic mutations relative to the other IPMNs, and a mutation spectrum with a predominance of T->C transitions and T->A transversions. Previous reports have suggested a higher prevalence of microsatellite instability in IPMNs compared to PDACs [28]. However, we did not identify pathogenic germline variants or somatic alterations in mismatch repair genes in this sample, and microsatellite testing indicated that the tumor was microsatellite stable. Exogenous carcinogens have previously been associated with specific somatic mutation spectra, for example alcohol and aristolochic acid [35,36]. Further investigation to determine the prevalence and etiology of hypermutated IPMNs, and whether this phenotype is associated with high-grade histomorphology, is warranted given the potential clinical implications of somatic mutation burden on therapy [37].

Our germline analysis of 94 hereditary cancer predisposition genes did not identify any pathogenic variants in this cohort, however, the reported prevalence of such variants in patients with surgically resected IPMN is 7.3% [15]. As such, this observation was not unexpected. Multiple variants of unknown significance were found in the assessed hereditary cancer susceptibility genes; however, as none were associated with somatic mutations or copy number alterations, their contribution to tumorigenesis is unknown.

Analysis of RNA-Seq data for gene fusions identified 4 putatively somatic fusion events present only in IPMN organoid samples (supplementary material, Table S8). Previous reports of intraductal oncocytic papillary neoplasms (IOPN) note frequent fusions between ATP1B1–PRKACB, DNAJB1–PRKACA, or ATP1B1–PRKACA [38]. Interestingly, our analysis did not identify any fusions involving these genes, confirming the previous finding that these fusions are a specific feature of IOPNs.

We identified 28 genes that were differentially expressed between IPMN organoids and normal duct organoids. Importantly, CLDN18, an integral component of the cell membrane and tight junctions, was the most statistically significant upregulated gene and had previously been shown to be over expressed in IPMNs using immunohistochemistry, validating our approach [18]. The most statistically significant downregulated gene was FOXA1, a regulator of apoptosis through downstream mediators including BCL2 and CDKN1B as well as a mediator of epithelial-to-mesenchymal transition [39–41]. We assessed FOXA1 expression using immunohistochemistry and confirmed a statistically significant decrease in protein expression in IPMN, MCN, and PDAC compared to normal pancreatic duct [41]. The association of FOXA1 and other differentially expressed genes with early pancreatic tumorigenesis is intriguing, and studies in larger cohorts will be necessary to determine whether these genes differentiate IPMNs from other types of precursor lesions or are associated with progression to PDAC.

Our study also provides important data about the feasibility of organoid culture of normal and precancerous epithelial cells derived from human pancreatic resection specimens. We cultured several specimens of normal pancreatic duct and both low-grade and high-grade IPMNs in the organoid system. We were able to isolate adequate DNA and RNA at enriched neoplastic cellularity for genomic and transcriptomic analyses. This highlights the value of organoid culture as a source of enriched biological molecules (including nucleic acids and proteins) for analysis of uncommon, difficult to harvest, or paucicellular neoplastic lesions. In addition, the success of organoid culture of both normal and precancerous pancreatic epithelium highlights this system as a potential tool for experimental manipulation of such human cells in culture.

Though our study represents a unique combination of organoid culture and molecular analysis of human IPMN specimens, it has a number of limitations. First, we and others have demonstrated that organoid cultures of pancreatic precursor lesions and invasive carcinoma may contain many of the same mutations present in original lesion [23,25]. However, our organoid samples were derived from a single piece of resected IPMN and therefore subclones established in organoid culture may not capture the genetic heterogeneity of the whole lesion [42,43]. This was demonstrated by our observation of low concordance between somatic mutations identified in single pieces of FFPE IPMN and organoid samples and is consistent with previous studies demonstrating driver gene heterogeneity in IPMNs [42]. Of note, organoid culture maintains a diverse population of sub-clones. Our observation that average allele frequencies for coding and splice site somatic mutations in our organoid samples range from 0.11 to 0.38 indicate that our organoid cultures also show this diversity. Second, as pancreatic resections are always performed for some underlying pancreatic disease or neoplasm, the non-neoplastic duct samples may not be truly normal. In each case, the harvested pancreatic duct was grossly normal and derived organoids did not have a KRAS or GNAS mutation. Third, while our study provides insight into the genetics of IPMNs, it was not powered to identify somatic alterations that drive development of high-grade IPMN from low-grade lesion. In the future, larger studies of low-grade and high-grade lesions will be needed to deconvolute the genetic basis of this evolution.

In summary, our study documents a biobank of IPMN and normal pancreatic duct organoids, characterized by genomic and transcriptomic sequencing, a unique resource for the interrogation of these precancerous lesions. Using whole genome sequencing and RNA-Seq we identified potential drivers of IPMN tumor development. These included 12 genes with recurrent somatic mutations and 28 genes that were differentially expressed when compared to normal pancreatic duct. Our results demonstrate that organoid culture is an effective tool for analyzing small precursor lesions at –omics scale to identify critical molecular alterations in precious human tissue samples.

Supplementary Material

tS1-S3,S7

Table S1. Analyses performed on normal and neoplastic human pancreatic organoids.

Table S2. KRAS and GNAS hotspot mutation in IPMNs by Sanger sequencing

Table S3. Data statistics for whole genome sequencing

Table S7. Data statistics for RNA-Seq

tS4-6,S8,S9

Table S4. Somatic mutations identified in coding region and splice sites of IPMN organoids

Table S5. Structural alterations identified in IPMN organoids

Table S6. Somatic mutations identified in coding regions and splice sites from whole exome sequence data of H79

Table S8. Differentially expressed transcripts and genes identified in organoids

Table S9. Fusion genes identified in organoids

fS1-S7

Figure S1. Mutational signature for IPMN organoids by sample

Figure S2. Functional consequence of coding and splicing mutations in IPMN organoids by sample

Figure S3. Somatic mutations identified in KRAS, GNAS, RNF43 and SLC7A8 genes in IPMN organoids by amino acid change

Figure S4. Structural alterations and somatic mutations identified in IPMN organoids by chromosomal location

Figure S5. Structural alterations and somatic mutations identified IPMN organoids by chromosomal location and sample

Figure S6. H-score for immunohistochemical staining of FOXA1 in MCN compared to normal duct.

Figure S7. H-score for immunohistochemical staining of FOXA1 in PDAC compared to normal duct.

supinfo

Acknowledgements

Rolfe Pancreatic Cancer Foundation; Susan Wojcicki and Denis Troper; NIH/NCI P50 CA62924; NIH/NIDDK K08 DK107781; NIH NCI R00 CA190889; Sol Goldman Pancreatic Cancer Research Center; Burroughs-Wellcome Fund; the Maryland-Genetic, Epidemiology and Medicine Training Program (MD-GEM); DFG-German Research Foundation; BIH-Charité Junior Clinician Scientist Program; Buffone Family Gastrointestinal Cancer Research Fund; Kaya Tuncer Career Development Award in Gastrointestinal Cancer Prevention; AGA-Bernard Lee Schwartz Foundation Research Scholar Award in Pancreatic Cancer; Sidney Kimmel Foundation for Cancer Research Kimmel Scholar Award; AACR-Incyte Corporation Career Development Award for Pancreatic Cancer Research; American Cancer Society Research Scholar Grant; Emerson Collective Cancer Research Fund; Joseph C Monastra Foundation; The Gerald O Mann Charitable Foundation (Harriet and Allan Wulfstat, Trustees); Art Creates Cures Foundation.

Footnotes

Data availability statement

RNA-Seq data has been deposited in the European Genome-phenome Archive (EGA) with accession code XXupdate at proof stageXX. Whole genome sequence data is available on request from the corresponding authors.

Conflicts of interest

RHH has the right to receive royalty payments from Thrive Earlier Diagnosis for the GNAS in pancreatic cysts invention. LDW is an Associate Editor of The Journal of Pathology. No other conflicts of interest were declared.

References

  • 1.Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin 2019; 69: 7–34. [DOI] [PubMed] [Google Scholar]
  • 2.Rahib L, Smith BD, Aizenberg R, et al. Projecting cancer incidence and deaths to 2030: the unexpected burden of thyroid, liver, and pancreas cancers in the United States. Cancer Res 2014; 74: 2913–2921. [DOI] [PubMed] [Google Scholar]
  • 3.Iacobuzio-Donahue CA. Genetic evolution of pancreatic cancer: lessons learnt from the pancreatic cancer genome sequencing project. Gut 2012; 61: 1085–1094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Laffan TA, Horton KM, Klein AP, et al. Prevalence of unsuspected pancreatic cysts on MDCT. AJR Am J Roentgenol 2008; 191: 802–807. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Yachida S, Jones S, Bozic I, et al. Distant metastasis occurs late during the genetic evolution of pancreatic cancer. Nature 2010; 467: 1114–1117. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Jones S, Zhang X, Parsons DW, et al. Core signaling pathways in human pancreatic cancers revealed by global genomic analyses. Science 2008; 321: 1801–1806. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Biankin AV, Waddell N, Kassahn KS, et al. Pancreatic cancer genomes reveal aberrations in axon guidance pathway genes. Nature 2012; 491: 399–405. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Waddell N, Pajic M, Patch AM, et al. Whole genomes redefine the mutational landscape of pancreatic cancer. Nature 2015; 518: 495–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Bailey P, Chang DK, Nones K, et al. Genomic analyses identify molecular subtypes of pancreatic cancer. Nature 2016; 531: 47–52. [DOI] [PubMed] [Google Scholar]
  • 10.Hoskins JW, Jia J, Flandez M, et al. Transcriptome analysis of pancreatic cancer reveals a tumor suppressor function for HNF1A. Carcinogenesis 2014; 35: 2670–2678. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Chen R, Yi EC, Donohoe S, et al. Pancreatic cancer proteome: the proteins that underlie invasion, metastasis, and immunologic escape. Gastroenterology 2005; 129: 1187–1197. [DOI] [PubMed] [Google Scholar]
  • 12.Zhou Q, Andersson R, Hu D, et al. Quantitative proteomics identifies brain acid soluble protein 1 (BASP1) as a prognostic biomarker candidate in pancreatic cancer tissue. EBioMedicine 2019; 43: 282–294. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Wu J, Matthaei H, Maitra A, et al. Recurrent GNAS mutations define an unexpected pathway for pancreatic cyst development. Sci Transl Med 2011; 3: 92ra66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Amato E, Molin MD, Mafficini A, et al. Targeted next-generation sequencing of cancer genes dissects the molecular profiles of intraductal papillary neoplasms of the pancreas. J Pathol 2014; 233: 217–227. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Skaro M, Nanda N, Gauthier C, et al. Prevalence of germline mutations associated with cancer risk in patients with intraductal papillary mucinous neoplasms. Gastroenterology 2019; 156: 1905–1913. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Sato N, Fukushima N, Maitra A, et al. Gene expression profiling identifies genes associated with invasive intraductal papillary mucinous neoplasms of the pancreas. Am J Pathol 2004; 164: 903–914. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Jury RP, Thibodeau BJ, Fortier LE, et al. Gene expression changes associated with the progression of intraductal papillary mucinous neoplasms. Pancreas 2012; 41: 611–618. [DOI] [PubMed] [Google Scholar]
  • 18.Tanaka M, Shibahara J, Fukushima N, et al. Claudin-18 is an early-stage marker of pancreatic carcinogenesis. J Histochem Cytochem 2011; 59: 942–952. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Ouyang H, Mou L, Luk C, et al. Immortal human pancreatic duct epithelial cell lines with near normal genotype and phenotype. Am J Pathol 2000; 157: 1623–1631. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Kamiyama H, Kamiyama M, Hong SM, et al. In vivo and in vitro propagation of intraductal papillary mucinous neoplasms. Lab Invest 2010; 90: 665–673. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Ideno N, Yamaguchi H, Ghosh B, et al. GNAS(R201C) induces pancreatic cystic neoplasms in mice that express activated KRAS by inhibiting YAP1 signaling. Gastroenterology 2018; 155: 1593–1607 e1512. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Kopp JL, Dubois CL, Schaeffer DF, et al. Loss of Pten and activation of Kras synergistically induce formation of intraductal papillary mucinous neoplasia from pancreatic ductal cells in mice. Gastroenterology 2018; 154: 1509–1523 e1505. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Boj SF, Hwang CI, Baker LA, et al. Organoid models of human and mouse ductal pancreatic cancer. Cell 2015; 160: 324–338. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Broutier L, Andersson-Rolf A, Hindley CJ, et al. Culture and establishment of self-renewing human and mouse adult liver and pancreas 3D organoids and their genetic manipulation. Nat Protoc 2016; 11: 1724–1743. [DOI] [PubMed] [Google Scholar]
  • 25.Felsenstein M, Trujillo MA, Huang B, et al. Generation and characterization of a cell line from an intraductal tubulopapillary neoplasm of the pancreas. Lab Invest 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Ishibashi H, Suzuki T, Suzuki S, et al. Sex steroid hormone receptors in human thymoma. J Clin Endocrinol Metab 2003; 88: 2309–2317. [DOI] [PubMed] [Google Scholar]
  • 27.van de Wetering M, Francies HE, Francis JM, et al. Prospective derivation of a living organoid biobank of colorectal cancer patients. Cell 2015; 161: 933–945. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Lupinacci RM, Goloudina A, Buhard O, et al. Prevalence of microsatellite instability in intraductal papillary mucinous neoplasms of the pancreas. Gastroenterology 2018; 154: 1061–1065. [DOI] [PubMed] [Google Scholar]
  • 29.Furukawa T, Kuboki Y, Tanji E, et al. Whole-exome sequencing uncovers frequent GNAS mutations in intraductal papillary mucinous neoplasms of the pancreas. Sci Rep 2011; 1: 161. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Cancer Genome Atlas Research Network. Electronic address aadhe, Cancer Genome Atlas Research N. Integrated Genomic Characterization of Pancreatic Ductal Adenocarcinoma. Cancer Cell 2017; 32: 185–203 e113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Mazzoni SM, Fearon ER. AXIN1 and AXIN2 variants in gastrointestinal cancers. Cancer Lett 2014; 355: 1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Fafilek B, Balek L, Bosakova MK, et al. The inositol phosphatase SHIP2 enables sustained ERK activation downstream of FGF receptors by recruiting Src kinases. Sci Signal 2018; 11. [DOI] [PubMed] [Google Scholar]
  • 33.Schreck KC, Taylor P, Marchionni L, et al. The Notch target Hes1 directly modulates Gli1 expression and Hedgehog signaling: a potential mechanism of therapeutic resistance. Clin Cancer Res 2010; 16: 6060–6070. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Su J, Morgani SM, David CJ, et al. TGF-beta orchestrates fibrogenic and developmental EMTs via the RAS effector RREB1. Nature 2020; 577: 566–571. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Hoang ML, Chen CH, Sidorenko VS, et al. Mutational signature of aristolochic acid exposure as revealed by whole-exome sequencing. Sci Transl Med 2013; 5: 197ra102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Chang J, Tan W, Ling Z, et al. Genomic analysis of oesophageal squamous-cell carcinoma identifies alcohol drinking-related mutation signature and genomic alterations. Nat Commun 2017; 8: 15290. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Le DT, Durham JN, Smith KN, et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science 2017; 357: 409–413. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Singhi AD, Wood LD, Parks E, et al. Recurrent Rearrangements in PRKACA and PRKACB in Intraductal Oncocytic Papillary Neoplasms of the Pancreas and Bile Duct. Gastroenterology 2020; 158: 573–582. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Song L, Wei X, Zhang B, et al. Role of Foxa1 in regulation of bcl2 expression during oxidative-stress-induced apoptosis in A549 type II pneumocytes. Cell Stress Chaperones 2009; 14: 417–425. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Williamson EA, Wolf I, O’Kelly J, et al. BRCA1 and FOXA1 proteins coregulate the expression of the cell cycle-dependent kinase inhibitor p27(Kip1). Oncogene 2006; 25: 1391–1399. [DOI] [PubMed] [Google Scholar]
  • 41.Song Y, Washington MK, Crawford HC. Loss of FOXA1/2 is essential for the epithelial-to-mesenchymal transition in pancreatic cancer. Cancer Res 2010; 70: 2115–2125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Fischer CG, Beleva Guthrie V, Braxton AM, et al. Intraductal Papillary Mucinous Neoplasms Arise From Multiple Independent Clones, Each With Distinct Mutations. Gastroenterology 2019; 157: 1123–1137 e1122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Felsenstein M, Noe M, Masica DL, et al. IPMNs with co-occurring invasive cancers: neighbours but not always relatives. Gut 2018; 67: 1652–1662. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Tsai S, McOlash L, Palen K, et al. Development of primary human pancreatic cancer organoids, matched stromal and immune cells and 3D tumor microenvironment models. BMC Cancer 2018; 18: 335. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Li H Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. arXiv:13033997v2 [q-bioGN] 2013. [Google Scholar]
  • 46.Broad Institute. Picard Toolkit. GitHub Repository http://broadinstitutegithubio/picard/. Last accessed 01/12/2019.
  • 47.McKenna A, Hanna M, Banks E, et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res 2010; 20: 1297–1303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Cibulskis K, Lawrence MS, Carter SL, et al. Sensitive detection of somatic point mutations in impure and heterogeneous cancer samples. Nat Biotechnol 2013; 31: 213–219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Wang K, Li M, Hakonarson H. ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic Acids Res 2010; 38: e164. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Fu Y, Liu Z, Lou S, et al. FunSeq2: a framework for prioritizing noncoding regulatory variants in cancer. Genome Biol 2014; 15: 480. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Layer RM, Chiang C, Quinlan AR, et al. LUMPY: a probabilistic framework for structural variant discovery. Genome Biol 2014; 15: R84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Talukder AK, Ravishankar S, Sasmal K, et al. XomAnnotate: Analysis of Heterogeneous and Complex Exome- A Step towards Translational Medicine. PLoS One 2015; 10: e0123569. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Skidmore ZL, Wagner AH, Lesurf R, et al. GenVisR: Genomic Visualizations in R. Bioinformatics 2016; 32: 3012–3014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Rosales RA, Drummond RD, Valieris R, et al. signeR: an empirical Bayesian approach to mutational signature discovery. Bioinformatics 2017; 33: 8–16. [DOI] [PubMed] [Google Scholar]
  • 55.Ou J, Zhu LJ. trackViewer: a Bioconductor package for interactive and integrative visualization of multi-omics data. Nat Methods 2019; 16: 453–454. [DOI] [PubMed] [Google Scholar]
  • 56.Gu Z, Gu L, Eils R, et al. circlize Implements and enhances circular visualization in R. Bioinformatics 2014; 30: 2811–2812. [DOI] [PubMed] [Google Scholar]
  • 57.Andrews S FastQC: a quality control tool for high throughput sequence data. Available online at: http://www.bioinformaticsbabrahamacuk/projects/fastqc. Last accessed 01/12/2019.
  • 58.Martin M Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnetjournal 2011; 17.1: 10–12. [Google Scholar]
  • 59.Kim D, Langmead B, Salzberg SL. HISAT: a fast spliced aligner with low memory requirements. Nat Methods 2015; 12: 357–360. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Pertea M, Kim D, Pertea GM, et al. Transcript-level expression analysis of RNA-seq experiments with HISAT, StringTie and Ballgown. Nat Protoc 2016; 11: 1650–1667. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61.Pertea M, Pertea GM, Antonescu CM, et al. StringTie enables improved reconstruction of a transcriptome from RNA-seq reads. Nat Biotechnol 2015; 33: 290–295. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62.Fu JFA, Collado-Torres L, et al. ballgown: Flexible, isoform-level differential expression analysis. R package version 2.16.0 Available online at https://www.bioconductor.org/packages/release/bioc/html/ballgown. Last accessed 01/12/2019.
  • 63.Wickham H. ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York 2016. [Google Scholar]
  • 64.Kolde R. pheatmap: Pretty Heatmaps. Implementation of heatmaps that offers more control over dimensions and appearance. Available online at: https://cranr-projectorg/web/packages/pheatmap/indexhtml. Last accessed 01/12/2019.
  • 65.Haas BJ, Dobin A, Li B, et al. Accuracy assessment of fusion transcript detection via read-mapping and de novo fusion transcript assembly-based methods. Genome Biol 2019; 20: 213–228. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66.Dixon JR, Xu J, Dileep V, et al. Integrative detection and analysis of structural variation in cancer genomes. Nat Genet 2018; 50: 1388–1398. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 67.Lagstad S, Zhao S, Hoff AM, et al. chimeraviz: a tool for visualizing chimeric RNA. Bioinformatics 2017; 33: 2954–2956. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 68.Murphy KM, Zhang S, Geiger T, et al. Comparison of the microsatellite instability analysis system and the Bethesda panel for the determination of microsatellite instability in colorectal cancers. J Mol Diagn 2006; 8: 305–311. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

tS1-S3,S7

Table S1. Analyses performed on normal and neoplastic human pancreatic organoids.

Table S2. KRAS and GNAS hotspot mutation in IPMNs by Sanger sequencing

Table S3. Data statistics for whole genome sequencing

Table S7. Data statistics for RNA-Seq

NIHMS1703650-supplement-tS1-S3_S7.docx (21.1KB, docx)
tS4-6,S8,S9

Table S4. Somatic mutations identified in coding region and splice sites of IPMN organoids

Table S5. Structural alterations identified in IPMN organoids

Table S6. Somatic mutations identified in coding regions and splice sites from whole exome sequence data of H79

Table S8. Differentially expressed transcripts and genes identified in organoids

Table S9. Fusion genes identified in organoids

NIHMS1703650-supplement-tS4-6_S8_S9.xlsx (171KB, xlsx)
fS1-S7

Figure S1. Mutational signature for IPMN organoids by sample

Figure S2. Functional consequence of coding and splicing mutations in IPMN organoids by sample

Figure S3. Somatic mutations identified in KRAS, GNAS, RNF43 and SLC7A8 genes in IPMN organoids by amino acid change

Figure S4. Structural alterations and somatic mutations identified in IPMN organoids by chromosomal location

Figure S5. Structural alterations and somatic mutations identified IPMN organoids by chromosomal location and sample

Figure S6. H-score for immunohistochemical staining of FOXA1 in MCN compared to normal duct.

Figure S7. H-score for immunohistochemical staining of FOXA1 in PDAC compared to normal duct.

NIHMS1703650-supplement-fS1-S7.pdf (8.1MB, pdf)
supinfo
NIHMS1703650-supplement-supinfo.docx (42.6KB, docx)

RESOURCES