Transcriptomic Profiling Identifies DCBLD2 as a Diagnostic and Prognostic Biomarker in Pancreatic Ductal Adenocarcinoma

Abstract

Background: Accumulating evidence showed that the elevated expression of DCBLD2 is associated with unfavorable prognosis of various cancers. However, the correlation of DCBLD2 expression value with the diagnosis and prognosis of pancreatic ductal adenocarcinoma (PDAC) has not yet been elucidated.Methods: Univariate Cox regression analysis was used to screen robust survival-related genes. Expression pattern of selected genes was investigated in PDAC tissues and normal tissues from multiple cohorts. Kaplan–Meier (K-M) survival curves, ROC curves and calibration curves were employed to assess prognostic performance. The relationship between DCBLD2 expression and immune cell infiltrates was conducted by CIBERSORT software. Biological processes and KEGG pathway enrichment analyses were adopted to clarify the potential function of DCBLD2 in PDAC.Results: Univariate analysis, K-M survival curves and calibration curves indicated that DCBLD2 was a robust prognostic factor for PDAC with cross-cohort compatibility. Upregulation of DCBLD2 was observed in dissected PDAC tissues as well as extracellular vesicles from both plasma and serum samples of PDAC patients. Both tissue and extracellular vesicle-encapsulated DCBLD2 expression have significant diagnostic value. Besides, DCBLD2 expression is correlated with infiltrating level of CD8 + T cells and macrophage M2 cells. Functional enrichment revealed that DCBLD2 may be involved in cell motility, angiogenesis, and cancer-associated pathways.Conclusions: Our study systematically analyzed the potential diagnostic, prognostic and therapeutic value of DCBLD2 in PDAC. All the findings indicated that DCBLD2 may play a considerably oncogenic role in PDAC with diagnostic, prognostic and therapeutic potential. These preliminary results of bioinformatics analyses need to be further validated in more prospective studies.

The Network Zoo: a multilingual package for the inference and analysis of biological networks

“…Among the top ten associations (Figure 2A), we found that the targeting by the TF melanocyte inducing transcription factor (MITF) was associated with changes in promoter methylation, in particular, we found a significant association between the MITF targeting score, computed as the TF weighted outdegree, and promoter hypermethylation of Discoidin, CUB and LCCL Domain Containing 2 ( DCBLD2 ) (Figure 2A), a gene that has been suggested to trigger oncogenic processes in melanoma through Epidermal Growth Factor Receptor ( EGFR ) signaling [41]. This finding consistent with the identification of MITF as a key driver of melanoma [42, 43].…”