Despite improvements in surgical techniques and adjuvant chemotherapy, the overall mortality rates in pancreatic cancer have generally remained relatively unchanged and the 5-year survival rate is actually below 2%. This paper will address the importance of achieving an early diagnosis and identifying markers for prognosis and response to therapy such as genes, proteins, microRNAs or epigenetic modifications. However, there are still major hurdles when translating investigational biomarkers into routine clinical practice. Furthermore, novel ways of secondary screening in high-risk individuals, such as artificial neural networks and modern imaging, will be discussed. Drug resistance is ubiquitous in pancreatic cancer. Several mechanisms of drug resistance have already been revealed, including human equilibrative nucleoside transporter-1 status, multidrug resistance proteins, aberrant signaling pathways, microRNAs, stromal influence, epithelial-mesenchymal transition-type cells and recently the presence of cancer stem cells/cancer-initiating cells. These factors must be considered when developing more customized types of intervention ("personalized medicine"). In the future, multifunctional nanoparticles that combine a specific targeting agent, an imaging probe, a cell-penetrating agent, a biocompatible polymer and an anti-cancer drug may become valuable for the management of patients with pancreatic cancer.
Keywords: Artificial neural networks; Biomarkers; Imaging; Nanomedicine; Pancreatic cancer; Personalized medicine.