People with type 2 diabetes (T2DM) have a significantly elevated risk of developing pancreatic cancer (PC). This well-known epidemiological connection suggests the two diseases share underlying biological mechanisms - but exactly which genes and pathways are involved has remained unclear.
Understanding this shared biology is important for two reasons. First, it could reveal new drug targets that address both diseases simultaneously. Second, it might point toward molecular changes that occur early in patients with diabetes who are developing pancreatic cancer, offering a window for earlier detection.
This study used a computational approach called Weighted Gene Co-expression Network Analysis (WGCNA) to systematically identify genes involved in both T2DM and pancreatic cancer. By analyzing multiple public gene expression datasets, the team found 44 shared genes and pinpointed S100A6 as the most central - or hub - gene connecting these two diseases.
Weighted Gene Co-expression Network Analysis (WGCNA) is a method that groups genes into modules based on how similarly they behave across many samples. Genes in the same module tend to be functionally related - they may be part of the same pathway, regulated by the same transcription factor, or involved in the same disease process.
The team analyzed six independent gene expression datasets from the GEO database - three for T2DM and three for pancreatic cancer. For each disease, WGCNA identified modules of co-expressed genes. The researchers then asked: which modules are associated with disease status, and do any of those modules overlap between T2DM and pancreatic cancer?
This cross-disease comparison yielded 44 shared genes - genes whose expression patterns were coordinately altered in both diseases. These 44 genes were then analyzed further to identify which among them occupied the most central position in the network - meaning it was most connected to other important genes. That central gene, the hub gene, is most likely to be a key driver of shared disease biology.
S100A6 (also known as Calcyclin) is a small calcium-binding protein belonging to the S100 protein family. It plays roles in cell proliferation, apoptosis (programmed cell death), and cytoskeletal organization. The fact that it emerged as the hub gene in this analysis places it at the intersection of the molecular networks disrupted in both T2DM and pancreatic cancer.
In the pancreatic cancer datasets, S100A6 was significantly overexpressed in tumor tissue compared to normal pancreatic tissue. This upregulation was consistent across datasets, lending confidence that this is a genuine feature of pancreatic tumors rather than a technical artifact. High S100A6 expression was also associated with poor overall survival in pancreatic cancer patients (p = 0.005), making it not just a biomarker of tumor identity but also a predictor of outcome.
Gene Set Enrichment Analysis (GSEA) performed on the S100A6-connected gene network revealed enrichment in cancer-related pathways as well as glycometabolism pathways - the molecular machinery that processes sugars. This glycometabolic connection is particularly interesting given that T2DM is fundamentally a disease of impaired glucose regulation, suggesting a mechanistic bridge between the metabolic dysfunction of diabetes and the growth-promoting machinery of pancreatic cancer.
To move beyond correlation and test whether S100A6 actually drives pancreatic cancer behavior, the researchers performed loss-of-function experiments in the Mia PaCa-2 cell line - a well-established pancreatic cancer cell model. They used RNA interference (siRNA) to silence S100A6, then measured the effects on cell behavior.
When S100A6 was knocked down, cell proliferation decreased significantly. Fewer cells divided over the same time period compared to control cells with normal S100A6 levels. This demonstrates that pancreatic cancer cells are at least partially dependent on S100A6 to sustain their rapid growth - a characteristic feature of aggressive cancer.
Even more striking, S100A6 knockdown also reduced cell invasion - the ability of cancer cells to break through tissue barriers, a key step in metastasis. Reduced invasion in cell culture experiments suggests that targeting S100A6 might slow not just primary tumor growth but also the spread of pancreatic cancer to other organs. These functional results transform S100A6 from a statistical correlation into a validated, mechanistically relevant target.
One of the most conceptually interesting findings from the pathway analysis was the enrichment of endodermal cell fate specification pathways among the shared genes. Both pancreatic beta cells (which produce insulin and are destroyed in diabetes) and pancreatic ductal cells (which give rise to PDAC) originate from the same embryonic tissue layer - the endoderm.
Disruptions in the developmental programs that establish and maintain endodermal cell identity may therefore create conditions that predispose cells to both diabetic dysfunction and malignant transformation. S100A6, as a hub gene in these networks, could be a molecular actor in this shared developmental vulnerability - its overexpression potentially reflecting a reversion toward or disruption of normal endodermal programming in cancer cells.
This developmental perspective offers a new lens through which to understand the diabetes-cancer connection. Rather than viewing T2DM as merely a risk factor that creates a pro-tumor metabolic environment, this analysis suggests the two conditions may share deeper molecular roots in cell identity and fate regulation. Targeting those shared roots could potentially address both diseases or at least identify patients with diabetes who are at highest risk for pancreatic cancer development.
This study demonstrates through integrated computational and experimental approaches that S100A6 is a shared molecular driver connecting T2DM and pancreatic cancer. Its overexpression in tumors, association with poor survival, and functional role in promoting cell growth and invasion make it a compelling biomarker and potential therapeutic target.
From a clinical standpoint, S100A6 expression levels in pancreatic tissue or blood could potentially be used to identify diabetic patients who are at elevated risk of having or developing pancreatic cancer. Early identification of such patients could enable more intensive surveillance, potentially catching cancer at a stage where curative surgery is still possible.
The 44 shared genes identified in this study also provide a broader resource for future research. Each one represents a possible node in the biological network connecting metabolic disease and cancer, and any one of them could emerge as important through further study. The methodological framework itself - using WGCNA to find cross-disease hub genes - is applicable to many other disease-cancer associations and could yield important discoveries beyond pancreatic cancer.