In pancreatic

cancer, IGF1 may function as a growth factor (63). IGF1 is upregulated in human pancreatic cancer tissue, with serum levels elevated in pancreatic cancer patients (64), (65). We recently noted that genetic variations in the IGF axis pathway are prognostic in advanced pancreatic cancer (66). After genotyping 41 SNPs from 10 IGF-axis genes Inhibitors,research,lifescience,medical in over 700 advanced pancreatic cancer patients, we noted that SNP of the IGF1R, IGF2R, and IRS1 gene were significantly associated with ALK inhibitor survival. In a current study that includes an IGF1R-directed antibody, MK-0646 we have noted a correlation between IGF1/IGFBP3 ratio and response. These findings will be confirmed in a wider

cohort of patients and a prospective, biomarker-driven study is planned (67). Biomarker validation Biomarker-driven therapeutic clinical trials can include the co-development of the biomarker and the study agent, particularly when the Inhibitors,research,lifescience,medical biomarker is relatively novel. The goal is to have appropriate validation Inhibitors,research,lifescience,medical before the marker can reach clinical applications; but validation is a cumbersome process for which standards are not clearly established. Critical issues that need to be addressed for the validation studies include the specificity and reproducibility of the marker. In the case of pancreatic cancer, this is further complicated by inter-patient heterogeneity and difficulty in obtaining representative sampling from the primary tumor site (pancreas). Regulatory guidance in this regard will be imperative in the development of biomarker driven targeted therapies for pancreatic cancer. Clinical trial Inhibitors,research,lifescience,medical design for targeted agents The use of a panel of biomarkers as potential predictive tools for the enrollment of patients on clinical trials with targeted agents requires Inhibitors,research,lifescience,medical innovative

clinical trial design beyond the traditional simple randomization. These traditional trial designs are based on the ‘frequentist’ principles. Frequentist trial designs are based on the probability of observing results as being disparate from the expected or the ‘null hypothesis’. In these frequentist designs, a p value is defined as the probability that the observed results are sufficiently disparate from Phosphatidylinositol diacylglycerol-lyase the controls and a p value of <0.05 is generally considered as significant. The advantage of the traditional randomized trials is that these are relatively easy to implement and they are scientifically robust and focused. However, the latter is also a potential disadvantage as these trial designs are inflexible, limiting innovation or modification as the trial proceeds. Furthermore, traditional randomized trials tend to be large and expensive wherein some patients are needlessly exposed to inferior therapies.