Corey Hoffman, a predoctoral student in the Medical Center, was the winner of the second annual “America’s Got Regulatory Science Talent” student competition held on Feb. 10 at the Clinical and Translational Science Institute. Hoffman will soon travel to Maryland to speak with the Food and Drug Administration about his project. This competition is organized by the University of Rochester Regulatory Science program, led by Joan Adamo, Ph.D. and Scott Steele, Ph.D. and held in conjunction with the Center for Excellence in Regulatory Science and Innovation at the University of Maryland.
Tell us a little bit about your proposal.
Right now, there are certain areas in regulatory science where the FDA wants to advance science, and within each area there are a few more specific focuses. The focus that stuck out to me is a clinical one that centers on personalized medicine, and looking for new types of biomarkers.
So what we want to know is: Is there something within the human body, a protein or DNA sequence that might be different in a healthy individual than someone who might be at a predisposition for a disease or who might have a disease. And how can we identify methods for finding these and then validating them?
Can you give an example of that?
There was one study on people with leukemia, and there was a small subset of patients that had a more aggressive disease and didn’t respond to chemotherapy. So the study was asking if there was something genetic that was different in those individuals, and it showed that there was indeed a commonality. So that was the example I used in the presentation.
What I’m proposing is to take advantage of all the clinical trials being done that are using sequencing like that, and put them all into a database. This could allow others to prevent redundancies in future trials and could be a reference point for future clinical studies or basic science studies.
Very cool. In a perfect world, what do you see as the potential outcomes from this type of database?
There are two potential outcomes that I could see. The first is that you could screen a patient’s genome, so that when patients do respond well in a trial, maybe there’s a genetic element that they share. Or, if there’s a commonality in people that don’t respond, then you could screen future patients ahead of time before you even offer the treatment. That way you wouldn’t have to put them through something if you knew ahead of time that they weren’t going to respond.
Secondly, certain individuals don’t respond well to certain drugs because of how their body metabolizes the drugs. This is determined by things called CYP genes, and certain people have mutations in these genes which affect how they respond to the drug. So you could screen people for mutations in their CYP gene to give indications for drug response, and who is and isn’t a good candidate for a certain drug.
Are there any challenges you see to developing a database like this?
I think the biggest concern before a database is ever constructed is protecting the patient health information of these genetic sequences, and I think that’s something that remains unexplored. There needs to be efforts that go into de-identifying the genomes. I’m not a geneticist, but it seems possible that a certain sequence of DNA could be unique to an individual that it could tell you who that person is. So, how can we take advantage of genetic information without putting any risk to the patients? And if these genome sequences are in the database, who has access? Those are two things that need to be considered.