Precision medicine is the great healthcare promise of today and the future. Successful individual treatments regimens and research programs, at large academic medical centers and community hospitals alike, are already underway and saving lives. And with large-scale initiatives such as the National Institutes of Health's All of Us research program, which kicked off earlier this month, that bold vision will only grow in its potential to improve the health of patients and populations. Without policies in place, there are major mistakes that hospitals, researchers, clinicians and policymakers must avoid to get this right. That point stuck with me among all the optimism here at the HIMSS Precision Medicine Summit in the nation's capital late last week.
The New York company said in July that it was reshaping its business into three units: Innovative Medicines; Established Medicines, which handles older drugs that have lost protection; and Consumer Healthcare. It said Innovative Medicines will bring in most of the company's revenue and has strong growth potential due partially to an aging population that will create growing demand for new medicines.
Many public and private efforts in coming years will focus on research in precision medicine, developing biomarkers to indicate which patients are likely to benefit from a certain treatment so that others can be spared the cost--financial and physical--of being treated with unproductive therapies and therapeutic signals can be more easily uncovered. However, such research initiatives alone will not deliver new medicines to patients in the absence of strong incentives to bring new products to market. We examine the unique economics of precision medicines and associated biomarkers, with an emphasis on the factors affecting their development, pricing, and access.
In precision medicine, sometimes called personalized medicine, researchers work to identify the genetic factors that drive or contribute to a disease and build medicine that targets the downstream effects of those miscreant genes. Then, they use genomic sequencing technologies to identify just those patients who bear the distinctive genetic signatures their drug works on. More often than not, these drugs are costly, and they don't work on everyone. But when the right patients get the right medicine at the right time, treatments will be more effective and have fewer side effects.
Inferences were made using traditional biostatistics. In the early 1990s, ML emerged, whereby advanced computing programs (machines) processed huge data sets (big data) from many sources and discerned patterns among multiple unselected variables. Such patterns were undiscoverable using traditional biostatistics (1) and were used to iteratively refine (learn) layered mathematical models (algorithms). The Table lists key differences between EBM and ML.