A Charles River panel discusses innovations that might help make tumor models more robust and predictive
The low percentage of candidate drugs that make it to market is well-documented. One study published by Nature Drug Discovery found that between 2013-2015, 24% of candidates in Phase II and Phase III trials failed to meet safety endpoints, while 54% of the candidates got shelved because they didn’t work. A separate analysis reported in 2016 by PAREXEL, a life science consulting firm, identified 38 trials enrolling over 145,000 patients that failed to show efficacy.
How can we improve upon these results? A panel at the recent AACR cancer meeting in Chicago earlier this month looked at the need for robust and predictive clinical models in oncology as one way of minimizing failures in the clinic.
“The best clinical model would reflect human tumor biology as closely as possible,” says Sabine Gorynia, PhD, a Managing Director of Discovery Services at Charles River and the moderator of the panel. The discussion looked at the preclinical models that have the best chance of identifying effective drug candidates and combinations and how much information can be discerned from studying the molecular data of tumor models—in other words is the tremendous amount of data we can collect about tumor behavior leading us to drugs that are more able to outfox the tumors. The discussion also looked at a novel tool from Philip’s Research that identifies activated signaling pathways in cancer cells.
The other panelists included Thomas Metz, PhD, Research Director, Charles River; Anja van de Stolpe, PhD, Medical Scientific Advisor, Philips Research; and Neil Williams, PhD, Senior Director of Immunology, Charles River.