How adaptive I-SPY 2 breast cancer study silenced critics
The initial results from I-SPY 2 are a fine rebuttal to industry skeptics who doubted the academic groups in charge had the ability to quickly run a robust clinical trial across multiple tumor types. In less than two years, the adaptive Phase II study in breast cancer has enabled decisions to advance two compounds to Phase III in specific patient subgroups - and to stop investigation of those compounds in a combined 10 other subgroups.
The goal of I-SPY 2 is to test unapproved compounds in up to eight different biomarker-defined subgroups of patients with neoadjuvant breast cancer and use Bayesian statistics to predict specific groups in which a compound has at least an 85% chance of success in a Phase III trial.
The study is managed by the Biomarkers Consortium, a public-private partnership of the non-profit Foundation for NIH (FNIH) and industry.
When I-SPY 2 started in 2010, only three companies were willing to include their compounds, and neoadjuvant breast cancer was not the lead indication for any of them. Three other companies provided financial support, but no compounds.
"I think when we started, there were a lot of people that didn't think we could really pull it off, in terms of testing multiple drugs across multiple subgroups in a short period of time," said lead investigator Laura Esserman, a professor in the department of surgery and radiology and director of the Carol Franc Buck Breast Care Center at the UCSF Helen Diller Family Comprehensive Cancer Center.
Since then, four new compounds from three companies have been added, and more companies have joined the waiting list.
Researchers behind I-SPY 2 credit both the encouraging results and industry's increased interest in participating to an operational formula that included the development of a master protocol, the upfront registration of trial sites, and the rapid integration of emerging data to quickly advance or drop new drug candidates.
While most academic-led trials can take up to two years just to get started and even longer before they report results, Esserman said the standing master protocol allowed patients to be enrolled quickly and ensured the data were collected in a consistent manner.
Additionally, investigators were able to collect data across multiple subgroups in near real-time. That speed allowed them to determine which programs or cohorts should be dropped in as little as six months, and which should be advanced to