Transcriptional variations in the wider peritumoral tissue environment of pancreatic cancer

Peritumoral tissue (PTT): increasing need for naming convention

Various terms are used to describe non-malignant tissue located in the proximity of a tumor, belonging to the organ from which the tumor originated. Traditionally, these tissues, sometimes called "normal adjacent tissue" have been used as controls in cancer studies, and were considered representative of morphologically healthy, non-cancerous tissue. However, with the advancement of OMIC technologies, such tissues are increasingly recognized to be distinct from both tumor and healthy tissues. Furthermore, properties, characteristics, and role of these tissues in cancer formation and progression is increasingly studied. In order to make future research in this area more harmonized and more accessible, as well as to counter the widespread perception of normalcy, we are advocating the need for standardized naming convention. For this purpose, we propose the use of neutral and comprehensive term "Peritumoral Tissue" along with the acronym "PTT". While significant amount of data on these tissues are publicly available, reuse of such data remains limited due to a lack of information on sample collection procedures. In order to facilitate future reuse of the data, we suggest a list of features that should be documented during sample collection procedures. These recommendations can aid the definition of Standard Operating Procedures.

DNA-methylation signature accurately differentiates pancreatic cancer from chronic pancreatitis in tissue and plasma

OBJECTIVE

Pancreatic ductal adenocarcinoma (PDAC) is a lethal malignancy. Differentiation from chronic pancreatitis (CP) is currently inaccurate in about one-third of cases. Misdiagnoses in both directions, however, have severe consequences for patients. We set out to identify molecular markers for a clear distinction between PDAC and CP.

DESIGN

Genome-wide variations of DNA-methylation, messenger RNA and microRNA level as well as combinations thereof were analysed in 345 tissue samples for marker identification. To improve diagnostic performance, we established a random-forest machine-learning approach. Results were validated on another 48 samples and further corroborated in 16 liquid biopsy samples.

RESULTS

Machine-learning succeeded in defining markers to differentiate between patients with PDAC and CP, while low-dimensional embedding and cluster analysis failed to do so. DNA-methylation yielded the best diagnostic accuracy by far, dwarfing the importance of transcript levels. Identified changes were confirmed with data taken from public repositories and validated in independent sample sets. A signature of six DNA-methylation sites in a CpG-island of the protein kinase C beta type gene achieved a validated diagnostic accuracy of 100% in tissue and in circulating free DNA isolated from patient plasma.

CONCLUSION

The success of machine-learning to identify an effective marker signature documents the power of this approach. The high diagnostic accuracy of discriminating PDAC from CP could have tremendous consequences for treatment success, once the result from still a limited number of liquid biopsy samples would be confirmed in a larger cohort of patients with suspected pancreatic cancer.

Promoter Hypermethylation Promotes the Binding of Transcription Factor NFATc1, Triggering Oncogenic Gene Activation in Pancreatic Cancer

Studies have indicated that some genes involved in carcinogenesis are highly methylated in their promoter regions but nevertheless strongly transcribed. It has been proposed that transcription factors could bind specifically to methylated promoters and trigger transcription. We looked at this rather comprehensively for pancreatic ductal adenocarcinoma (PDAC) and studied some cases in more detail. Some 2% of regulated genes in PDAC exhibited higher transcription coupled to promoter hypermethylation in comparison to healthy tissue. Screening 661 transcription factors, several were found to bind specifically to methylated promoters, in particular molecules of the NFAT family. One of them-NFATc1-was substantially more strongly expressed in PDAC than control tissue and exhibited a strong oncogenic role. Functional studies combined with computational analyses allowed determining affected genes. A prominent one was gene ALDH1A3, which accelerates PDAC metastasis and correlates with a bad prognosis. Further studies confirmed the direct up-regulation of ALDH1A3 transcription by NFATc1 promoter binding in a methylation-dependent process, providing insights into the oncogenic role of transcription activation in PDAC that is promoted by DNA methylation.

Efficient, high-performance semantic segmentation using multi-scale feature extraction

The success of deep learning in recent years has arguably been driven by the availability of large datasets for training powerful predictive algorithms. In medical applications however, the sensitive nature of the data limits the collection and exchange of large-scale datasets. Privacy-preserving and collaborative learning systems can enable the successful application of machine learning in medicine. However, collaborative protocols such as federated learning require the frequent transfer of parameter updates over a network. To enable the deployment of such protocols to a wide range of systems with varying computational performance, efficient deep learning architectures for resource-constrained environments are required. Here we present MoNet, a small, highly optimized neural-network-based segmentation algorithm leveraging efficient multi-scale image features. MoNet is a shallow, U-Net-like architecture based on repeated, dilated convolutions with decreasing dilation rates. We apply and test our architecture on the challenging clinical tasks of pancreatic segmentation in computed tomography (CT) images as well as brain tumor segmentation in magnetic resonance imaging (MRI) data. We assess our model's segmentation performance and demonstrate that it provides performance on par with compared architectures while providing superior out-of-sample generalization performance, outperforming larger architectures on an independent validation set, while utilizing significantly fewer parameters. We furthermore confirm the suitability of our architecture for federated learning applications by demonstrating a substantial reduction in serialized model storage requirement as a surrogate for network data transfer. Finally, we evaluate MoNet's inference latency on the central processing unit (CPU) to determine its utility in environments without access to graphics processing units. Our implementation is publicly available as free and open-source software.

A Transcriptionally Distinct Subpopulation of Healthy Acinar Cells Exhibit Features of Pancreatic Progenitors and PDAC

Pancreatic ductal adenocarcinoma (PDAC) tumors can originate either from acinar or ductal cells in the adult pancreas. We re-analyze multiple pancreas and PDAC single-cell RNA-seq datasets and find a subset of nonmalignant acinar cells, which we refer to as acinar edge (AE) cells, whose transcriptomes highly diverge from a typical acinar cell in each dataset. Genes upregulated among AE cells are enriched for transcriptomic signatures of pancreatic progenitors, acinar dedifferentiation, and several oncogenic programs. AE-upregulated genes are upregulated in human PDAC tumors, and consistently, their promoters are hypomethylated. High expression of these genes is associated with poor patient survival. The fraction of AE-like cells increases with age in healthy pancreatic tissue, which is not explained by clonal mutations, thus pointing to a nongenetic source of variation. The fraction of AE-like cells is also significantly higher in human pancreatitis samples. Finally, we find edge-like states in lung, liver, prostate, and colon tissues, suggesting that subpopulations of healthy cells across tissues can exist in pre-neoplastic states. SIGNIFICANCE: These findings show "edge" epithelial cell states with oncogenic transcriptional activity in human organs without oncogenic mutations. In the pancreas, the fraction of acinar cells increases with age.

Qualitative diagnostic signature for pancreatic ductal adenocarcinoma based on the within-sample relative expression orderings

BACKGROUND

Pancreatic ductal adenocarcinoma (PDAC) accounts for about 90% of pancreatic cancer, which is one of the most aggressive malignant neoplasms with a 9.3% five-year survival rate. The pathological biopsy is the current golden standard for confirming suspicious lesions of PDAC, but it is not entirely reliable because of the insufficient sampling amount and inaccurate sampling location. Therefore, developing a robust signature to aid the accurate diagnosis of PDAC is critical.

METHODS

Based on the within-sample relative expression orderings of gene pairs, we identified a qualitative signature to discriminate both PDAC and adjacent samples from both chronic pancreatitis and normal samples in the training datasets and validated it in other independent datasets produced by different laboratories with different measuring platforms.

RESULTS

A six-gene-pair signature was identified in the training data and validated in eight independent datasets. For surgical samples, 96.63% of 356 PDAC tissues, 100% of 11 pancreatitis tissues of non-cancer patients, and 23 of 24 normal pancreatic tissues were correctly classified. Especially, 59 of 60 cancer-adjacent normal tissues of PDAC patients were correctly identified as PDAC. For biopsy samples, all of 11 PDAC biopsy tissues were correctly classified as PDAC.

CONCLUSION

The signature can distinguish both PDAC and PDAC-adjacent normal tissues from both chronic pancreatitis and normal tissues of non-cancer patients even when the sampling locations are inaccurate, which can aid the diagnosis of PDAC.

Blood biomarkers for differential diagnosis and early detection of pancreatic cancer

Pancreatic cancer is currently the most lethal tumor entity and case numbers are rising. It will soon be the second most frequent cause of cancer-related death in the Western world. Mortality is close to incidence and patient survival after diagnosis stands at about five months. Blood-based diagnostics could be one crucial factor for improving this dismal situation and is at a stage that could make this possible. Here, we are reviewing the current state of affairs with its problems and promises, looking at various molecule types. Reported results are evaluated in the overall context. Also, we are proposing steps toward clinical utility that should advance the development toward clinical application by improving biomarker quality but also by defining distinct clinical objectives and the respective diagnostic accuracies required to achieve them. Many of the discussed points and conclusions are highly relevant to other solid tumors, too.

Fibroblasts from Distinct Pancreatic Pathologies Exhibit Disease-Specific Properties

Although fibrotic stroma forms an integral component of pancreatic diseases, whether fibroblasts programmed by different types of pancreatic diseases are phenotypically distinct remains unknown. Here, we show that fibroblasts isolated from patients with pancreatic ductal adenocarcinoma (PDAC), chronic pancreatitis (CP), periampullary tumors, and adjacent normal (NA) tissue (N = 34) have distinct mRNA and miRNA profiles. Compared with NA fibroblasts, PDAC-associated fibroblasts were generally less sensitive to an antifibrotic stimulus (NPPB) and more responsive to positive regulators of activation such as TGFβ1 and WNT. Of the disease-associated fibroblasts examined, PDAC- and CP-derived fibroblasts shared greatest similarity, yet PDAC-associated fibroblasts expressed higher levels of tenascin C (TNC), a finding attributable to miR-137, a novel regulator of TNC. TNC protein and transcript levels were higher in PDAC tissue versus CP tissue and were associated with greater levels of stromal activation, and conditioned media from TNC-depleted PDAC-associated fibroblasts modestly increased both PDAC cell proliferation and PDAC cell migration, indicating that stromal TNC may have inhibitory effects on PDAC cells. Finally, circulating TNC levels were higher in patients with PDAC compared with CP. Our characterization of pancreatic fibroblast programming as disease-specific has consequences for therapeutic targeting and for the manner in which fibroblasts are used in research. SIGNIFICANCE: Primary fibroblasts derived from various types of pancreatic diseases possess and retain distinct molecular and functional characteristics in culture, providing a series of cellular models for treatment development and disease-specific research.

SST gene hypermethylation acts as a pan-cancer marker for pancreatic ductal adenocarcinoma and multiple other tumors: toward its use for blood-based diagnosis

Aberrant DNA methylation is often involved in carcinogenesis. Our initial goal was to identify DNA methylation biomarkers associated with pancreatic cancer. A genomewide methylation study was performed on DNA from pancreatic ductal adenocarcinoma (PDAC) and endocrine pancreas tumors. Validation of DNA methylation patterns and concomitant alterations in expression of gene candidates was performed on patient samples and pancreatic cancer cell lines. Furthermore, validation was done on independent data from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO). Finally, droplet digital PCR was employed to detect DNA methylation marks in cell-free (cf) DNA isolated from plasma samples of PDAC patients and cancer-free blood donors. Hypermethylation of the SST gene (encoding somatostatin) and concomitant downregulation of its expression were discovered in PDAC and endocrine tumor tissues while not being present in chronic pancreatitis (inflamed) tissues and normal pancreas. Fittingly, treatment with a somatostatin agonist (octreotide) reduced cell proliferation and migration of pancreatic cancer cells. Diagnostic performance of SST methylation in a receiver operating characteristic curve analysis was 100% and 89% for tissue and plasma samples, respectively. A large body of TCGA and GEO data confirmed SST hypermethylation and downregulation in PDAC and showed a similar effect in a broad spectrum of other tumor entities. SST promoter methylation represents a sensitive and promising molecular, pan-cancer biomarker detectable in tumor tissue, and liquid biopsy samples.