Sunday, June 16, 2013

Predictive scans

A recent study funded by the National Institute of Mental Health shows that brain scans before and immediately after the initiation of a treatment hold valuable information to predict efficacy. Both the discovery and the practice of medicine, still, largely depend on trial and error. Multi-factorial complex responses of the biological system to CNS therapies have resulted in significant noise in the collected behavioral data – creating havoc both in R&D and in healthcare providers, not to mention the payers. The process of inventing and applying medicines for mental health seem fully antiquated in the presence of available technologies.

Pharmaceutical R&D and healthcare, albeit being technologically advanced in certain dimensions, have been lagging in the analysis of available information. PET scans, an old technology, show brain activity in highly analyzable matrix, providing an almost instantaneous path to measure efficacy of a drug. Such scans may also provide a method to determine optimal dose, something that researchers were forced to ignore. The 7 billion specimens of humans across the world show such genetic diversity that it seems unlikely that popping a standardized amount of the NCE within standardized time intervals, will be optimal treatment.

It is time both R&D and the practice of medicine embraced the tools of analytics – something boring engineers have been accustomed to for many decades and it has resulted in handsome returns.