Identification of prognostic lipid droplet-associated genes in pancreatic cancer patients via bioinformatics analysis

Abstract

Abstract Background Pancreatic cancer is the fourth leading cause of cancer deaths in the United States both in females and in males, and is projected to become the second deadliest cancer by 2030. The overall 5-year survival rate remains at around 10%. Cancer metabolism and specifically lipid metabolism plays an important role in pancreatic cancer progression and metastasis. Lipid droplets can not only store and transfer lipids, but also act as molecular messengers, and signaling factors. As lipid droplets are implicated in reprogramming tumor cell metabolism and in invasion and migration of pancreatic cancer cells, we aimed to identify lipid droplet-associated genes as prognostic markers in pancreatic cancer. Methods We performed a literature search on review articles related to lipid droplet-associated proteins. To select relevant lipid droplet-associated factors, bioinformatics analysis on the GEPIA platform (data are publicly available) was carried out for selected genes to identify differential expression in pancreatic cancer versus healthy pancreatic tissues. Differentially expressed genes were further analyzed regarding overall survival of pancreatic cancer patients. Results 65 factors were identified as lipid droplet-associated factors. Bioinformatics analysis of 179 pancreatic cancer samples and 171 normal pancreatic tissue samples on the GEPIA platform identified 39 deferentially expressed genes in pancreatic cancer with 36 up-regulated genes (ACSL3, ACSL4, AGPAT2, BSCL2, CAV1, CAV2, CAVIN1, CES1, CIDEC, DGAT1, DGAT2, FAF2, G0S2, HILPDA, HSD17B11, ICE2, LDAH, LIPE, LPCAT1, LPCAT2, LPIN1, MGLL, NAPA, NCEH1, PCYT1A, PLIN2, PLIN3, RAB5A, RAB7A, RAB8A, RAB18, SNAP23, SQLE, VAPA, VCP, VMP1) and 3 down-regulated genes (FITM1, PLIN4, PLIN5). Among 39 differentially expressed factors, seven up-regulated genes (CAV2, CIDEC, HILPDA, HSD17B11, NCEH1, RAB5A, and SQLE) and two down-regulation genes (BSCL2 and FITM1) were significantly associated with overall survival of pancreatic cancer patients. Multivariate Cox regression analysis identified CAV2 as the only independent prognostic factor. Conclusions Through bioinformatics analysis, we identified nine prognostic relevant differentially expressed genes highlighting the role of lipid droplet-associated factors in pancreatic cancer.

Keywords

• Bioinformatics
• GEPIA
• Lipid droplet-associated genes
• Pancreatic cancer
• Lipid metabolism