But how mRNAs will perform in early-stage disease has not been determined, Dr. Hanash said. "We want to find neoplastic lesions before they are invasive, so there are problems with profiling patients at the time of diagnosis and predicting how candidate markers will perform for early detection," he said.
Gene profiles that are unique to cancer (not shared with chronic pancreatitis) are also making headway as biomarkers. The overexpression of S100P has been associated with pancreatic cancer and the IPMN precursor lesion, but not normal tissue. But the ability to identify this in the blood has not been realized. KRAS mutations can be detected in the blood but the mutations are also seen in pancreatitis.
The field is advancing largely due to the availability of an engineered mouse model that closely mimics human pancreatic cancer. Mouse models also offer the potential for sampling at different stages of tumor development, which can reveal unique proteomic changes. Investigators have, in fact, developed a panel of such proteins that has been validated in humans, showing the power to discriminate between persons who will develop pancreatic cancer within one year and those who will not, Dr. Hanash said. "We are now in the middle of producing assays for the most promising candidates, and we hope to come up with a set of novel markers that will have relevance for early detection."
The signaling pathway may have a name that any computer gamer would love, but the sonic hedgehog has the potential to be a great therapeutic target for ligand-dependent and ligand-independent cancers, said David Tuveson, MD, PhD, of Cambridge University and Addenbrooke's Hospital in the UK.
Activation of the hedgehog pathway by its ligand sonic has been implicated in pancreatic tumorigenesis. Sonic hedgehog (SHH) contributes to the extensive stromal reaction in pancreatic tumors, to the maintenance of cancer stem cells, and possibly to marked desmoplasia in tumors and to chemoresistance.
The value of targeting the hedgehog pathway has been demonstrated in the mouse model. "For precancerous lesions you cannot tell the mouse from the human because the model has all the same molecular, cellular, and pathologic hallmarks of human cancer," Dr. Tuveson said.
Drug resistance in pancreatic ductal adenocarcinoma seems to involve poor drug penetration as the primary problem. "Methods that modify the stroma and alter tumor perfusion are needed," he said. "Inhibitors of the hedgehog pathway are one way to accomplish this."
The experimental SHH inhibitor IPI-926 can penetrate tumors and achieve therapeutic levels, inducing a 10-fold reduction in transcription factor and clearly affecting the molecular target. Pretreatment of a tumor with IPI-926 enhances the concentration of gemcitabine(Drug information on gemcitabine) (Gemzar) in the tumor by 50%, and this translates into a measurable extension of survival in the mouse model. Mice treated with gemcitabine alone lived an average of 45 days compared with 80-plus days with the combination, Dr. Tuveson reported at the GI Cancers Symposium.
Improved survival
Robert McWilliams, MD, of the Mayo Clinic in Rochester, Minn., reported the results of a study of 605 patients with pancreatic cancer (all stages) evaluated for an association between SHH single nucleotide polymorphisms (SNPs) and survival. Carrying minor alleles in an SNP in the SHH gene rs1233556 was associated with a 16% increased survival overall (P = .033) and a 28% increase in survival among patients with metastatic disease (P = .02). The variant was also associated with the magnitude of SHH expression and the degree of desmoplasia in primary tumors (abstract 126).
"This is further evidence that the hedgehog pathway is important in pancreatic cancer and provides a paradigm for developing new sonic hedgehog pathway-based prognostic and therapeutic tools," Dr. McWilliams said. Based on the current study, it is also possible that certain genotypes may help patients who have them derive the most benefit from SHH inhibitors, he added.
