Should Hospitals Recruit Patients for Drug Company Research Projects?

I have always thought that hospitals, as stewards of abundant genotypic and phenotypic patient information, could perhaps partner with pharmaceutical companies in identifying patients for clinical research projects. A key concern of mine, however, has always been how this could be accomplished without impinging on patient privacy and security ...

I have always thought that hospitals, as stewards of abundant genotypic and phenotypic patient information, could perhaps partner with pharmaceutical companies in identifying patients for clinical research projects. A key concern of mine, however, has always been how this could be accomplished without impinging on patient privacy and security rights. A recent article raised this issue and I provide an excerpt from it below (see: Does 23andMe Deal Mean Medical Centers Are Sitting On Data Worth Millions?):

Are academic medical centers (AMCs) sitting on troves of data that could provide researchers insight into disease, jump start pharmaceutical efforts to develop cures – and potentially generate millions of dollars of revenue in the process? That’s the question many AMCs are sure to be asking this week, following Genentech’s announcement...that the South San Francisco-based biotech will pay 23andMe $60M ($10M upfront, plus $50M if certain milestones are met) for access to genetic and phenotypic data collected by 23andMe, with particular focus on the 12,000 patients with Parkinson’s Disease, and 1,300 parents and siblings.  Herper reports 23andMe have data on 800,000 customers in all, about 600,000 of whom have agreed to donate their data for research. The 23andMe data are especially useful for researchers because of its size and because the patients are re-contactable....Consequently... the 23andMe data trove “is much better than the vast majority of academic equivalents.”....

[In] early 2014, Tarrytown, NY-based Regeneron Pharmaceuticals ...and Geisinger Health System, headquartered in Danville, PA, initiated a collaboration, where Regeneron would sequence samples from consented Geisinger patients, creating opportunity for both organizations to leverage genomic and EMR-derived data....(see: Regeneron and Geisinger Health System Announce Major Human Genetics Research Collaboration) There are a number of questions emerging here: First, are datasets comprised of rich phenotype plus genotype information actually valuable....The honest answer at this point is that nobody knows....If new data can help pinpoint targets or suitable patients, or can help a company avoid a tempting dead end, then the effort could pay for itself quickly. ....Second, how could integrated datasets be more powerful....For instance, how rich or detailed should the genetic data be?  What about EMR data, which are notoriously difficult to extract – how do you organize this information....? 

The fact that Geisinger Health has cut a deal with a pharmaceutical company caught my attention since I view it as one of the most innovative health systems in the country. Here is a brief quote from a Regeneron press release discussing some additional details about this collaboration (see: Regeneron and Geisinger Health System Announce Major Human Genetics Research Collaboration): 

During the initial five-year collaboration term, Geisinger plans to collect samples from more than 100,000 consented patient volunteers, while Regeneron, through its wholly-owned subsidiary, Regeneron Genetics Center LLC, will perform sequencing and genotyping to generate de-identified genomic data....The collaboration will benefit from Geisinger's state of the art sample collection and storage capabilities, the MyCode biorepository, and extensive electronic medical records.  Regeneron has built a team and an infrastructure to support sequencing and genotyping over the term of the collaboration.  Regeneron intends to use its translational research and functional biology capabilities, including its VelociGene technology, to validate observed human genetic associations.

The Geisinger project is very interesting in that it involves Geisinger personnel collecting samples prospectively from 100,000 consented patient volunteers. The Geisinger MyCode biorepository project is a large scale biobanking initiative integrated with its EHR. It's not clear whether Geisinger will mine retrospective clinical data from its EHR relating to its patient volunteers or such data on a prospective basis. What is abundantly clear from this news is that a major health system is embarking on a large scale genomic research project with a pharmaceutical company using patient volunteers. The goal is undoubtedly to better understand the genetic basis of disease and also generate intellectual property for Geisinger. In this era of reduced reimbursement for healthcare services, this seems like a suitable goal as long as no patient rights are infringed upon, which I anticipate will be the case.

Source: labsoftnews.typepad.com