Geisinger’s Strategy for Clinical NGS Implementation

Mana Chandhok:
Hi, everyone. Welcome to this podcast from Cambridge HealthTech Institute for the Next Generation Diagnostics Summit, which runs August 15th to the 18th in Washington, DC. I'm Mana Chandhok, an associate producer. We have with us today one of our speakers from the clinical NGS applications and interpretations track. Dr. Marc Williams of the Geisinger Health System. Dr. Williams, thank you for joining us.

Marc Williams:
My pleasure.

Mana Chandhok:
In your talk, you will discuss strategies used by Geisinger Health System to effectively use genomic information in clinical care. Can you give us a brief overview of what those strategies are?

Marc Williams:
The key for our genomic medicine program is to understand how genomic information improves outcomes and ultimately adds value to the healthcare delivery system. To recognize and realize that, what we need to do is develop an infrastructure so we know that genomic information can be consistently delivered in a reproducible way with high reliability. Then we also have to define what are the outcomes of interest from the returned results and set up a measurement system so that we can measure those outcomes over time. Then we can we ask did we achieve the outcomes we were anticipating and if so, what was the impact from the perspective of the patient, from the perspective of the provider, from the perspective of the healthcare system. We then put those together to try and realize the value of the overall process. So that's the brief overview of the strategy, and, as you can imagine, there are a lot of details that are involved to make that happen. Geisinger is well positioned to do this for several reasons. First of all, we have already established a high reliability clinical care system so a lot of the things that we need in order to measure and capture outcomes are in place and so we can leverage the existing tools, customize them a bit to what we're doing, but we don't have to create things from scratch.

Second of all, the culture of the organization supports this type of high performance, high reliability care. We do not have to make the case to leadership or to support services like information technology, the data analysts, etc. about why we need to do this in a different way from all of the other care that's being done. We have the assets in terms of our data warehouse and other informatic tools that allow us to collect the data, but probably most importantly we have a partnership with our patients that is longstanding. Many of our patients reside in the same community over the course of their life, they receive from Geisinger really from cradle to grave. That puts us in a position that very few organizations can achieve, which is to monitor outcomes over the course of a very long timeline, from years to decades. What that means is that we have to establish programs that can capture data over that time frame, meaning that the programs have to be able to be sustainable without individuals being champions. So creation of that infrastructure is critically important but something that we are very committed to.

Mana Chandhok:
What are some challenges that you have encountered along the way?

Marc Williams:
Some of the challenges are technical. At the present time electronic health records don't accept genomic information as what we call structured data. What that means is that it can be very difficult to build the clinical reminders that we would like to get out to our patients and to providers to make sure that recommendations based on the genomic information actually take place over time. So we have to use manual processes to get that data into the electronic health record. We hope at some point that this will take place and electronic health records will be able to accept these data, but right now that requires a lot of extra effort on our part. We've not had much in the way of push back, either from our patients or providers. I think they recognize that this is a very important area, and it's been gratifying to see the engagement that we've had with patients and providers. We interact with them on a regular basis to assess how the program is doing. As you might expect, though, people's lives are busy, they have a lot of things going on, and as a consequence, we probably do not have as much engagement as we would like because we need to be respectful of people's time and effort.

The last challenge is one that would be common to most organization, as a health care delivery system we have resources that have certain limitations. We don't have an infinite amount of money to spend on things. So there's some return of results that we would be very interested in doing, like pharmacogenomics where we just don't have the resources available to be able to do the confirmatory testing that's necessary to be able to return the result. That's something that we're exploring different ways that we might be able to move this forward as we think this information is very high value. We have it, at least in some cases, from the sequencing and we want to be able to use it.

Mana Chandhok:
What are some new applications of next generation sequencing that you are most excited about?

Marc Williams:
I think what we're realizing, even in the early days of the return of results is just how limited clinical identification of a high penetrant Mendelian single gene conditions really is. So if you think about things like hereditary breast and ovarian cancer related to BRCA 1 and 2, Lynch syndrome, hereditary colorectal and endometrial cancer, familial hypercholesterolemia, a lipid disorder that increases the risk for cardiovascular disease, we are really not identifying nearly as many of these folks clinically as are out there. Now some of it relates to the absence of an informative family history but some of it is also that the information is just not collected in a systematic way.

By using the genome sequence information to identify individual that carry a pathogenic variant in one of the genes that are associated with one of these conditions, we're identifying a large number of people that had not been previously identified and are intervening in their care in a way that at least based on our preliminary, primarily anecdotal information, seems to be improving outcomes. This is really exciting and I think it's reinforcing the importance of the project that we're doing and reinforcing also how critical it is that we capture the outcomes from this return of results in an efficient way. But I think if our preliminary results continue to hold true, there really is the opportunity that this approach will revolutionize how we think about identification of individuals at risk for genetic disorders.

Mana Chandhok:
Dr. Williams, thank you for your time and insights today.

Marc Williams:
You're very welcome. It was a pleasure.

Mana Chandhok:
That was Dr. Marc Williams of the Geisinger Health System. He'll be speaking at the Clinical NGS applications and interpretations track at the upcoming Next Generation Diagnostic Summit, which runs August 15th to the 18th in Washington, DC. I'm Mana Chandhok, thank you for listening.