Video 37 min

A Strategic Approach to Implementing Genomic Testing in Clinical Oncology

September 28th, 2022


The below text is a transcript from the webinar. Because it is a transcript, there may be oddities that arise from the process of translating speech into text. We recommend accessing the recording, above, to gain full context. 

Overview and Introductions

Speaker 1: Dr. Sidiropoulos is the Director of Genomic Medicine in the department of pathology and laboratory medicine at the University of Vermont Medical Center.

Speaker 1: She completed her anatomic and clinical pathology residency at Dartmouth-Hitchcock Medical Center, her cytopathology fellowship at the University of Vermont Fletcher Allen Health Care, and her molecular genetic pathology fellowship at the University of California in San Francisco.

Speaker 1: She is also a board-certified clinical informaticist. The University of Vermont has really been with PierianDx since the very beginning, and we've worked very closely together to help build an NGS program from the ground up.

Speaker 1: We're really excited to have Dr. Sidiropoulos here with us today, sharing her insights on what makes an NGS testing program successful. So with that, I'll hand over to her to get into the content of her webinar today.

Dr. Sidiropoulos: Thank you, Sara, and thank you for having me address this important topic today. So I will be discussing, as Sara mentioned, a strategic approach to implementing genomic testing based on my background and experience and certainly what we have accomplished here at the University of Vermont.

Dr. Sidiropoulos: My webinar objectives are listed on the screen. The first is to provide a foundational background on clinical NGS implementation. Secondly, to describe how care pathways are central for successful genomic testing programs.

Dr. Sidiropoulos: And I'd like to also explain what I am calling a 360° Implementation Consideration and go over what those considerations are.

Dr. Sidiropoulos: I'd also like to discuss important functions of a tumor board, and explore ways to implement reflex testing, which is something we've very recently have done and are continuing to do in our health network.

Deciding to Implement Genomic Testing

Dr. Sidiropoulos: So I'd like to give a little bit of background and really a frame of reference so you can determine what content of this webinar may apply to your practice.

Dr. Sidiropoulos: We made a fundamental decision. I returned from San Francisco to Vermont in late 2013, and at the time, I can say, about three years earlier, there was very little molecular genetic pathology practice outside of what was being done in microbiology.

Dr. Sidiropoulos: So there was no focus, there was no division within our department of molecular genetic pathology. And so, with a new Chair, Debra Leonard, and myself, prior to me starting, we had to make a fundamental decision with the resources we had at hand: what we were going to do?

Dr. Sidiropoulos: And were we going to develop a test menu to get on an equal playing field with most of the labs out there in terms of classic molecular genetic pathology and then, or simultaneously, bring on next-generation sequencing as a backbone for our genomic medicine program?

Dr. Sidiropoulos: Or, with the resources at hand, were we gonna skip over classic molecular genetic pathology, as most of us know it, and move straight to genomic medicine?

Dr. Sidiropoulos: And so we made the decision to skip over classic molecular genetic pathology and move to genomic medicine. A lot of this was based on resource evaluation: what did we have at hand?

Dr. Sidiropoulos: And we were quite fortunate at the time to have a brand new chair who had negotiated, specifically, to build a genomic medicine program, and so for that, we were quite fortunate.

Dr. Sidiropoulos: But also, I have these two maps up on the screen. Our area, just so you're aware, the population of Vermont is listed here: a little over 600,000 people.

Dr. Sidiropoulos: The lower part of the state, Dartmouth-Hitchcock sits on the other side of the line there, so they take a lot of the population in the lower part of the state, but we handle the rest of the population of Vermont, in general.

Dr. Sidiropoulos: And then we also cover New York state, and specifically upper New York state. And so even though the population is quite large in New York, you can see, based on the population per square mile, we have relatively low volume.

Dr. Sidiropoulos: At the time, also, there was increasingly more ancillary molecular information that was being requested from increasingly limited amounts of tissue, and I know that almost everyone on this webinar understands that scenario.

Dr. Sidiropoulos: So, essentially, what this led us to in terms of looking at what we had at hand and what our landscape was, is we determined we were gonna use next-gen sequencing technology as our backbone.

Dr. Sidiropoulos: And we had a site visit at Genomics and Pathology Services at Washington University to learn a little bit about what they were doing with their genomic and pathology services.

Evaluating Solutions

Dr. Sidiropoulos: And that's when we were introduced to the CGW, and this was before, I think, PierianDx was actually formally a company. We were asked if we would be interested in beta testing that.

Dr. Sidiropoulos: I think my chair and I both looked at each other and we felt quite fortunate to have that opportunity, because when we had taken a tour of their institution.

Dr. Sidiropoulos: They had, essentially, a whole floor of people just to build and maintain their bioinformatics and reporting. And we knew that we couldn't recreate that wheel, nor did we have to anymore.

Dr. Sidiropoulos: So that's how we got involved with the CGW from the get-go, and that's a little bit of our background.

Dr. Sidiropoulos: What happened leading up to my arrival and then that continued within the first six months is there, we did a complete and a thorough stakeholder evaluation.

Dr. Sidiropoulos: And I think it was during that evaluation and engaging our stakeholders and leadership that we, and also the people who we consider to be our clients, we were able to secure resources with a pretty detailed five-year business plan that I put together with a lot of help from some expert business-planning folks at our institution.

Dr. Sidiropoulos: And we not only got leadership approval on the business plan, which then really secured our resources, but also we set out, at the time, and our department was doing a new strategic planning process for the department at large, and a significant section of that strategic plan involved the genomic medicine program.

Dr. Sidiropoulos: The way the genomic medicine program was laid out, the way we envisioned it, is that we would have our clinical genomic medicine operation.

Dr. Sidiropoulos: That would, essentially, generate a bunch of data that we intended to collect and mine in various different ways, including the actual genomic data, having a biobank, being able to access pertinent healthcare data that went along with the samples that we had, and really, in hopes to ...

Dr. Sidiropoulos: Research was obviously at an academic institution, one of our key tenets, as well, and we wanted to put that right into the strategic plan.

Dr. Sidiropoulos: And the genomic education is another key tenet, and that underlies, really, the entire operation and strategic plan.

Dr. Sidiropoulos: And so our vision, with all of this, was to improve health and healthcare, drive genomic value and implementation research, and to provide genomic medicine education.

Dr. Sidiropoulos: We discussed the provision of genomic medicine education for ourselves and our colleagues, for patients, and for laypeople at large.

Dr. Sidiropoulos: So we've been quite active in various spheres, and of course, that includes trainees and medical students and even, we teach in the undergraduate honors college course, as well.

Dr. Sidiropoulos: So we're trying to be pretty global in getting the message out there about what we're doing.

Genomically-informed Clinical Care

Dr. Sidiropoulos: So this is another fundamental key development and operations philosophy for us here at the University of Vermont Medical Center. And we really wanted to go beyond the concept of the lab just building an assay and putting it out there for use and move towards this concept of genomic care pathways.

Dr. Sidiropoulos: And so we fundamentally believe that genomically-informed clinical care involves strategic integration of the best genomic technology, as we judge it, with people and processes beyond the laboratory to realize the promise of precision medicine for each unique patient.

Dr. Sidiropoulos: So in our business plan, we have laid out three care pathways: one for oncology, one for pharmacogenomics, and one for inherited disease.

Dr. Sidiropoulos: We decided to start with the oncology care pathway first, and just at a very high level, we intended to develop ... we weren't at a place yet where we were thinking we would go with exome or genome to cover everything.

Dr. Sidiropoulos: And also based on reimbursement, we decided that at the time when we got started and wrote the business plan that we would develop a portfolio of next-gen sequencing-based assays that would cover solid tumors, being one, and that's called our "gene panel solid tumors", a hemepath panel, and then an inherited cancer risk panel.

Dr. Sidiropoulos: We specifically broke out inherited caner risk because we felt that, and I will get into this on an additional slide, that there were ancillary considerations in terms of up-front and how informed consent would fit into a work flow, and on the back end, making sure that we had the appropriate amount and structure of genetic counseling at our institution.

Dr. Sidiropoulos: It exists, but with the volumes that we anticipated, the way the genetic counseling stood, it may not have been as accessible for patients who needed it without increasing our STE, if you will. So in our business plan, we had included iterative additions of genetic counselors.

Implementation Considerations

Dr. Sidiropoulos: So I'd like to sort of switch now to implementation considerations, and I'll go over some at a high level and some we'll drill down.

Dr. Sidiropoulos: Institutional leadership, as a consideration, taking into consideration the end users, and by that I mean sometimes people call end users "clients" or "clinical providers", and I would even consider other sections and other sections of the lab and the professionals in those other sections.

Dr. Sidiropoulos: I'd like to talk a little bit about patients and that consideration, and reimbursement is a big one for all of us.

Dr. Sidiropoulos: In terms of institutional leadership, I can't say enough about our experience with early identification and engagement of key institutional leaders.

Dr. Sidiropoulos: I think that, for me, I cannot say enough that the leader of our department, our chair, had engaged with the college of medicine leadership and the medical center leadership to gauge that support, ensure that support, and I think that the discussions that were had sort of identified what local barriers may exist, and I think that's a really important thing.

Dr. Sidiropoulos: I think at this point, had I been in a situation knowing what I know now, being charged with starting a genomic medicine program and if that was my hire, my task, I think that I would turn around to the person hiring me and say, "Okay, now in order for me to complete my task, I'd like your assistance in engaging and identifying the correct leaders".

Dr. Sidiropoulos: So for anyone out there that's maybe at a stage of the game where, 'cause I know I listened to a lot of webinars as a trainee, certainly I would take that into consideration, and laying that groundwork up front is very important.

Dr. Sidiropoulos: I also think therein lies the opportunity for a strategic selection of pilot projects. Each institution has their own PHO-sci of interest; some places it's cardiology, some places it's oncology. And I would say, identify what those projects are and set yourself up for success in terms of how you wanna launch your program.

Dr. Sidiropoulos: And lastly, I'd like to say that it always, I feel, and this is in my opinion, behooves someone in a position who's trying to drive forth a genomic program or next-gen clinical next generation sequencing.

Dr. Sidiropoulos: There's a lot of people who don't quite buy it yet, and I certainly think it behooves you to create a transdisciplinary team and approach. That can also then be used to interface with leadership, as opposed to taking a unilateral approach from the lab.

Dr. Sidiropoulos: Having that clinical laboratory connection and engaging with the leadership in that way can be quite powerful, in my experience.

Dr. Sidiropoulos: Now switching gears to end users and end user considerations. End users, again, I said earlier, I would really define that as "clinicians who are using the report at the back end or clinical providers, and also, I think, a lot of folks in the lab".

Dr. Sidiropoulos: So for us, especially with oncology testing, anatomic pathologists and being board-certified in anatomic and cytopathology, when you generate those diagnoses on those cases, you feel an ownership for those cases.

Dr. Sidiropoulos: And I think that people get a little bit nervous and shy away from genomics as the unknown, but I think it would be behoove the molecular pathology team or the genomic medicine team in the lab to try and engage and keep a pathway of open conversation with the anatomic pathologists and engage them and have them be involved in seeing what those reports are at some level, or just let them know the reports are being done.

Dr. Sidiropoulos: Because I think they're the ones who go to the front line a lot of the time at tumor boards, and so it's really important to make them feel like they're part of this process and include that in that genomic education.

Dr. Sidiropoulos: One of the things that we did is, before we design any assay, and I'm sure a lot of you do this as well, but it's really important to define report content, that's important to the end user.

Assay Design

Dr. Sidiropoulos: We specifically, with the development of our gene panel solid tumor assay, for us, it wasn't right to pull an assay off the shelf; for some people, it is. But we certainly worked through and medically vetted all of the genes and target areas of our assay with, not only internally our pathologist team who looks at solid tumor samples, but also, once that was refined at that level, we took it to a group of medical oncologists in the cancer center to vet it there as well.

Dr. Sidiropoulos: That way, there were no surprises. 'Cause obviously, we're all very aware that new genes with significance are coming out at all times, but we have to lock something down in development and then develop the assay and bring it live.

Dr. Sidiropoulos: That way, you avoid that sort of criticism on the back end, like, "Hey! You forgot this gene". And it's like, "Well yeah, it just came out last month. So no, it's not on the panel that we agreed to develop a year ago", and that can be considered for version two.

Dr. Sidiropoulos: So I would just say it's important, I think, to engage end users. One other thing that, in a lot of the discussion with the end users, and again, this is that care pathway, we wanted to avoid impeding clinical work flow, and that's not only the clinicians who are out there with patients, but that was also the anatomic pathologist.

Dr. Sidiropoulos: I would be surprised if most of the people in the audience have a LIMS or a system that you use in molecular to support your molecular work flow, that's the identical system that also supports the anatomic pathology work flow.

Evaluating the Workflow

Dr. Sidiropoulos: And so we took stock, and we'll drill down a little deeper into this, but we took stock in terms of "what electronic system are most people living in during the day, and how do we let them live in that one system without having to click to others to do their work?" And so we really explored connectivity deeply as we implemented our first genomic assay.

Dr. Sidiropoulos: The other thing that I think is important is the tumor sample work flow. Pathologists are using, typically, an anatomic pathology-based module, and those don't typically connect all that well to the genomic or molecular pathology modules that are out there.

Dr. Sidiropoulos: So there's this disconnect between their electronic world and our electronic world.

Dr. Sidiropoulos: Their electronic world and our electronic world. So it's important to evaluate connectivity there. I think it's also important to take stock of, what are your anatomic pathologists comfortable doing? I think there is a lot of unknowns around, everybody like to ask the question, "How much do I need on this sample to pass it through and make sure it's successful?"

Dr. Sidiropoulos: We actually took that. We have three molecular pathologists here. We actually decided we would be the ones to do that tumor context sample evaluation. It's been very well-received by our anatomic pathologist, because I think it caused them a lot of consternation. And we decided to take that on as a value-added service for our genomic medicine group.

Dr. Sidiropoulos: Whether that's right for other groups out there, I'm not so certain. But that is what we have done. Also, I'd like to say that it's important to build processes that support clinical workflow. This isn't only in bio-informatics build, but we involve the laboratory information team and the institutional informatics leadership, and the electronic health record team in evaluating and building some of our connectivity requests.

Dr. Sidiropoulos: What we ended up doing and a really interesting feature of what we did, was our LIMS feeds back to our electronic health record. Because our electronic health record was not quite ready to have a dashboard feature that could alert our clinicians when the final report would be ready or posted.

Dr. Sidiropoulos: I know that Vanderbilt had presented something really interesting in terms of a dashboard feature. The reason why I wanted to try and recreate something like that was that we have clinical teams that are trying to schedule oncology patients for follow-up visits.

Dr. Sidiropoulos: And what is more frustrating? You're not only sick as a patient. You're scared. You're anxious. But to make it all the way into your doctor's office, and you think you're going to come out with a care plan. Finally, you can act on your cancer. And that's a relief for a lot of people. It's scary, but it's also a relief.

Dr. Sidiropoulos: You work yourself up to that moment only for the doctor to find out the molecular pathology report isn't back yet. So we're going to have to reschedule your visit. We desperately wanted to avoid that scenario for our patients. And so, our lens was feeding back to electronic health record to give a date of report so then the patient could be scheduled accordingly to avoid that scenario.

Dr. Sidiropoulos: So, we also, I think my fundamental statement on all of this is that intelligent and informed design of the electronic information systems and reporting can then mitigate loss productivity for clinical and laboratory staff and improve patient care experience.

Dr. Sidiropoulos: We also wanted to avoid the scenario of the clinicians constantly calling in to the lab and people having to deglove and run around, and figure out when the report was going to be complete. So we built this into our electronic systems. And we've had quite a nice reception of all of this.

Dr. Sidiropoulos: So, drilling a little deeper, this is a little bit about how it felt for us. We sat down pretty early on, myself and our technical specialist. We had a really small team and first, and we said, "Okay, let's sit down and draw out what this electronic workflow will look like. What should connect to what, and when, and where?"

Dr. Sidiropoulos: We brought together me, as the molecular pathologist, key technical staff, and a technical director in molecular pathology, managers, and supervisors from anatomic pathology and clinical pathology, the laboratory information team, and representatives from the electronic health record team.

Dr. Sidiropoulos: We walked into the room and we said, "Hello. Here is this wonderful pie in the sky vision of what we want." We quickly realized that everybody was speaking a different language in the room. It was quite humbling for me and it was I think humbling for everybody. We ended up building a great team, and people were very patient with one another in learning the right sort of buzz phrases.

Dr. Sidiropoulos: What happened was, we took that nice little package if you will, that nice, tight bottled yarn, and we unraveled it. I mean, it was completely unraveled. It was shocking to see for me what couldn't connect to what. The things that I had never considered. And this is despite being a well-trained pathologist in various sections of the lab.

Dr. Sidiropoulos: I just never knew that it wasn't possible to connect this or that on these systems. Why couldn't AP talk to CP? I just couldn't even believe some of this stuff. So the evolution, and we went through sort of the ordering and [inaudible 00:26:30] prophecies, the internal genomic medicine/anatomic pathology laboratory workflows. Then we also took a real deep dive into how the reporting was going to look.

Dr. Sidiropoulos: In the end, after bringing all these people together and completely making a mess of our initial pie in the sky vision, what we ended up doing was we could walk out of that room, walk out of those sessions. We were able to present an ideal, or present the workflow. This was a workflow that had the ideal connectivity, because there was in no way, shape, or form any way I was going to let go of the idea of cutting out any sort of manual entry into this whole process.

Dr. Sidiropoulos: So I wanted to keep that vision of ideal connectivity there. But it also, in this map, if you will, had identified all the basic challenges of connectivity. And so, we could identify piece by piece, iteration by iteration which ones we were going to knock off the list. And we're still making our way towards that ideal vision.

Workflow Considerations for Patients

Dr. Sidiropoulos: With patients, I would like to really talk about some workflow considerations that I took into account for patients. This was the concept, especially when we're talking about genetics and genomics, patient autonomy and quality of care.

Dr. Sidiropoulos: The patient autonomy issue, I'm pretty certain probably all of you in the audience are aware that some patients, and you yourself, some of you may want to know my whole genome. Tell me everything, even the things I can't act upon, I want to know. And other people don't want to know.

Dr. Sidiropoulos: You can imagine the scenario of a woman who may be estranged from her family or has bad relations and doesn't want to speak with her sisters or other women, or her family. And you unearth this heavy load of having a BRCA mutation. She wasn't expecting anyone opening Pandora's Box for her and now she's burdened with the concept of, not only how does she deal with this herself, but then having this burden of should she or should she not tell the rest of her family. Maybe she didn't want to know that in the first place.

Dr. Sidiropoulos: I thought it was really important to make sure, if we were going to include genes that had a real strong potential of inherited cancer risk in our assay, that we had adequate and appropriate informed consent right up front and that that was taken into consideration before an assay was ever ordered. Then we wanted component make sure we had appropriate ancillary genetic counseling support on the backend.

Dr. Sidiropoulos: In order for us to go live as quickly as possible, we didn't have the opportunity to do that informed consent process and genetic counselor right up front. We determined we would break out our assay building as having a panel that was clinically validated and designed specifically to try and avoid uncovering inherited cancer risk. We validated it in that way. Then we were going to build a separate panel that was for inherited cancer risk.

Dr. Sidiropoulos: That's how we decided to do that. I didn't feel it was right to include a lot of the panels I saw off the shelf had genes that I thought could unearth some of those questions for patients and put a burden on patients. We certainly didn't have informed consent upfront. I didn't feel right about doing that until I built that the right way. So that was our process. It doesn't mean it's the right process or the wrong process. It's just what we decided to do, and that's the reason why.


Dr. Sidiropoulos: We also, with our strategic plan of education, constantly talking to the clinicians about what we were proposing and how we were proposing to take this ... deal with inherited cancer risk with our assay design. We had complete buy-in from the clinicians on that. We certainly have been active in our local marketing, if you will, and educational theories that we have for patients about what we're doing here and why.

Dr. Sidiropoulos: We certainly wanted to make sure we had the right amount of genetic counseling support in place before we started ... before we turned on the faucet and started pushing patients down that route. Because the last thing we wanted patients to have to deal with was having a long wait to get in to see their genetic counselor. So that's part of why we structured it the way we did. And again, I'd like to repeat that we think that intelligent and informed design of the electronic information systems and reporting can mitigate loss productivity for clinical and laboratory staff and improve the patient care experience.

Dr. Sidiropoulos: Again, I don't need to repeat what we talked about earlier in terms of how we have insured that patients are being scheduled for their follow-up oncology care once they have a complete ancillary molecular profile in place. And we did that with the help of electronic design.

Dr. Sidiropoulos: So the next topic is a big one. Reimbursement. I decided before I even took my molecular pathology boards, when I went into my own practice, that I was going to be proactive instead of reactive around this issue.

Dr. Sidiropoulos: I'd like to put up in the screen now what I call the ... everyone calls it the promise of genomic medicine. And I very recently changed that phrase for myself to the hypothesis of genomic medicine.

Dr. Sidiropoulos: So the scenario is something like this. You have a patient that's diagnosed with cancer, metastatic cancer. The classic pathway is, it has usually a defined cost around it. Let's say molecular pathology, genomics isn't involved, and so you get a standard chemotherapy regimen. We all know that that has many effects in the body besides just targeting the cancer, which is why this person is highlighted in red here. It can be quite toxic.

Dr. Sidiropoulos: It is not uncommon that patients who are treated in this way end up in the hospital with complications from their chemotherapy, and that those hospital stays are quite expensive.

Dr. Sidiropoulos: With provision testing, molecular testing, genomic testing, and precision therapies, obviously there's an upfront cost in doing that. But the issue is is that you inform therapy either by avoiding targeted therapies when they are expected to be futile based on a negative profile, or you give the targeted therapies to patients who would otherwise be going this top route. You mitigate a lot of those unnecessary side effects. The hypothesis is that typically, you can avoid some of the untoward effects and hospitalizations that happen from classic chemotherapy.

Dr. Sidiropoulos: Again, I would say, in general, a lot of the payers at least consider this a hypothesis currently. This is where we get into, and I warn you, the dreaded circular argument that a lot of us in the clinical lab face and can't stand. This is ongoing, and it's been ongoing for years. I used to sit in the audience at a lot of the conferences we have nationally and listen to a lot of the early adopters of genomics.

NGS Test Reimbursement

Dr. Sidiropoulos: It was hard enough bringing this testing into the clinical lab and validating this testing, and being one of the first people to do it, and having to create the wheel of your bio-informatics and reporting. That was a challenge in and of itself. Never mind trying to triangulate all of the reimbursement and all of the other economic stuff that has to go along with the laboratory testing so you could actually prove your hypothesis in the previous screen, that this precision pathway would end up costing equal or less money to the healthcare system versus the top pathway.

Dr. Sidiropoulos: So the payers will say ... we can say, "Here's our promise of genomic medicine." And they'll say, "Show me the evidence that proves that." And I think a lot of us, there's a couple examples out there. There have been a couple examples out there. But in terms of the real robust data that we need to support reimbursement at a larger, more global scale, that just really doesn;t exist yet in my opinion.

Dr. Sidiropoulos: I decided before I even started practice, like I said, that I wanted to try and escape this circular argument. I dind;t know how at the time, but I really wanted to escape the circular argument. And I'm sure most of you out there would also like to escape the circular argument. Because I think a lot of us, as molecular pathology and people in this field, we are medical professionals. I think that we do see medical and clinical value in a lot of this testing. I think we have to do our job to prove it, though.

Dr. Sidiropoulos: So there are three key things that shaped my escape, or my attempt to escape the circular argument to date when I was reflecting on this. One is certainly, when we visited genomic and pathology services at [inaudible 00:37:17], we took note that they had a whole prior authorization team. And that their group put the resources in to doing prior authorization themselves. And I thought, "Brilliant, I'm going to do that." Because that seems to be working relatively well for them. Instead of saying, "Who at this institution can do prior authorization for me?" We actually hired someone to do prior authorization. She's been marvelous and she has educated me every step of the way. It's been quite valuable me, as well.

Dr. Sidiropoulos: There's Intermountain Healthcare, and I remember the first time I heard, I believe his last name is pronounced Nadode, Dr. Nadode. He's an oncologist at Intermountain Healthcare. I heard him speak about some of the health economic work that they were doing at Intermountain Healthcare in terms of informing, genomic ally informing cancer care there. They recently published, and I put up the PubMed ID here because I was moved by what they were doing and I wanted to try and recreate some of that work here in Vermont.

Dr. Sidiropoulos: The bottom line was that they did a retrospective analysis of precision medicine outcomes in patients with advanced cancer, and that it revealed improved progression-free survival without increased healthcare costs. So I want to add to this literature.

Dr. Sidiropoulos: The other thing that made a big mark for me recently, the National Academies of Sciences, Engineering, and Medicine put together a committee on policy issues in the clinical development and use of biomarkers for molecularly targeted therapies. I put a picture of this, the cover of this publication on the screen. And really, it includes ten goals to advance the development and appropriate clinical use of biomarker tests for molecularly targeted therapies.

Dr. Sidiropoulos: I heard about this, there was an early bird session at the Association for Molecular Pathology meeting in North Carolina this year. That's the first time I heard about this publication. So there's two recommendations I want to drill into with this that may help you at your institution and your discussions with your payers. And how maybe you're going to creatively partner in the future.

Dr. Sidiropoulos: One thing that this publication did was it acknowledged that generating evidence of overall clinical benefit of any biomarker test should be viewed as a continuous process. The recommendation 5A that was put out is that when there's existing evidence of clinical utility, when that's sufficient for initial use of a biomarker test. A lot of you may be saying, "Well, what is that?" That's a little bit up for interpretation. And you may have to argue about what that evidence is. That may get you into a conversation to have to argue your point.

Dr. Sidiropoulos: When there is that existing evidence of clinical utility, if it's sufficient for initial use of biomarker tests for molecular targeted therapy, payers should develop reimbursement models that support the ongoing collection of data within a rapid learning system.

Dr. Sidiropoulos: So what this boils down to, there was a phrase of how this boiled down for me. And it's coverage with evidence development. And that's exactly what we did at the University of Vermont Medical Center with Blue Cross/Blue Shield.

Dr. Sidiropoulos: What I will put out there for you, though, because I realize we might be a little bit different, 80% of our private payer landscape is Blue Cross/Blue Shield Vermont.

Dr. Sidiropoulos: And so, before we brought our first test live, I called the medical director at Blue Cross/BLue Shield Vermont and had a discussion with him about what our assay is, what genes are on it, how we medically vetted all of these, and came together as a team, a transdisciplinary team here and created this assay. And how it would be impactful for various types of cancer. And that we were going to take on the activity of prior authorization ourselves. And that we would also be doing health economic analysis ongoing with doing this testing.

Dr. Sidiropoulos: I basically asked him to engage with us for coverage with evidence development. And that's what we've been doing here. And so, they agreed with prior authorization, that they would pay for our genomic assay.

Dr. Sidiropoulos: So currently, they also, this committee, also made the statement, correctly so, that currently data on biomarker tests and patient outcomes are not collected in a way that can inform clinical practice. Part of the problem is that we have clinical data sitting in the EHR and maybe that's put off into a data warehouse somewhere that the lab doesn't touch. And then you have your lab data and that may sit in various places for you, and your molecular genomic data may sit in a very different place than your, say, anatomic pathology or lab medicine data. And the charge data and all of the reimbursement type data may sit in a very different place.

Dr. Sidiropoulos: So how do you ... It's hard enough to accurately and do timely mining and analysis of the data in one of those spheres, never mind triangulating all of that and then having someone sophisticated ... I wasn't trained as a health economist or an economist for that matter, despite all the training that I've done. So building the right team to analyze that data, to set up those systems and analyze that data. And so recommendation 6a from this committee was that they called out that electronic health record LIS vendors, Laboratory Information System vendors, and relevant software developers, and I'll say, such as clinical genomic data generating vendor systems such as PierianDx, right? They could be considered that. Should facilitate and enable the capture and linkage of biomarker test, molecularly targeted therapies, and longitudinal clinical patient data in the Electronic Health Record.

Dr. Sidiropoulos: And so they go on further to say that the information should be structured in the EHR, that it should include biomarker test specimen requirements; test result and interpretation; treatment prescribed and test ordered; and longitudinal clinical patient data. And I would say, are these health systems in the market in those three groups that they called out ready to meet this recommendation? And I would say no, in most instances probably not.

Dr. Sidiropoulos: So now, if I go back a slide, we're sitting in a place where I told BlueCross BlueShield, please cover us, and I promise you I will do the evidence development and so I'm sitting in a place where I'm saying, "Geez, I don't think everybody's ready to do this." But what I did find is a creative partnership with PierianDx. I will say that we have found PierianDx willing to partner creatively with us. We partnered with Pierian to enable sophisticated searchability to support and catapult our clinical operations, and to drive our health services research in regards to the economic impact of implementing genomics in oncology, and the tool we ultimately are using is Genospace.

Dr. Sidiropoulos: And so what we're able to do is take our genomic data that sits in PierianDx, we're able to take a lot of our other information we capture in our LIMS, which is the sun quest molecular module, and then take the clinical data and some of the charge data, and we're able to ... We decided to use Genospace as the tool to put this all in. And Pierian was really instrumental in helping me get this whole system in place. And really what that's doing is it is going to be, as we're designing this, our clinical searchability tool, and permit us to do the health services research that I wanted to do to escape the circular argument and that now I am obligated to do based on our relationship with BlueCross BlueShield of Vermont.

Dr. Sidiropoulos: And so I will say that it's been a huge step in the right direction, having a report format that is locked down in the way that Pierian permits us to do, and that the fidelity of this report format, as we generate it to be concise, accurate, and easily digestible, which facilitates through the interface, then, usability in a very, very busy clinic that facilitates action. But it's not enough, and I would suggest finding these creative partnerships for yourself to think about how you will be able to analyze and mine your data to catapult your service further.

The Mission of the Tumor Board

Dr. Sidiropoulos: So what has been absolutely ... Many, many people have tumor boards, and everybody sets them up in very different ways. Ours certainly ... The key mission, I think, of our tumor board is sort of a 360 degree genomic education. What we do is we dive deep down into our cases and we are able ... It's been challenging but very exciting to have all walks of knowledge as it pertains to genomics in the same room. And really, I start out with housekeeping on every single genomic tumor board. The only shame in the game is not asking the question. And we go there with a spirit of education, and certainly we have learned a lot about how people are using our reports, how they're interpreting our reports, and again, we've been able to partner with Pierian, give them that feedback, and work with them to continuously improve our report format and that has resulted in a much greater usability on the back end. And that's been very exciting to watch that in real time. So I would encourage tumor board development.

Dr. Sidiropoulos: The other thing that we've done most recently, and I'm sure most of you follow the professional guidelines out there. We decided, based on some of the professional recommendations that are out, especially right now with non-small cell lung cancer, we decided to target several transdisciplinary teams. The cancer care at our institution, and probably at a lot of yours out there are done via tumor boards. Or very specific transdisciplinary groups that focus on a certain or several tumor types. And so we targeted lung cancer, non-small cell lung cancer specifically, advanced age melanoma, and metastatic colorectal cancer, and either have already or are in the process of creating reflex testing. And we bring the guidelines to these transdisciplinary teams and discuss the clinical utility and quality of care that may come from reflex protocols, and certainly we consider the reimbursement landscape for this code that we use for five to 50 genes.

Dr. Sidiropoulos: And once we have ... The way this has typically worked at a very high level is with a transdisciplinary discussion around the guidelines and clinical care, we vote as a team to either implement the reflex protocol or not. And then it's really handed off to me and our molecular pathology group to say, "Let us handle all of the workflow issues." And so we interface with compliance, right? Is it legal to do so? How do we write the right policies into our documents to do this? We obviously work with our informatics teams to make sure that all of the informatics are in place for reflex testing to occur. And certainly there's a big education component, so before we bring a reflex test live, obviously the people who are supposed to be doing the reflex and the rest of the pathology department, specifically anatomic pathology, are educated, the residents are educated, about what to do and what to expect.

Dr. Sidiropoulos: So with that, as we just get closer and closer to the end of the hour, I would like to say that if you are going to do this, doing those three reflex protocols really has us projecting a doubling of our volume, and so I would say laboratory preparedness for increased volume is key. And you have to be ready for the scalability of sign out. And here is, again, another place where we have partnered creatively with Pierian to work on developing interpretation services. Everybody has different ways that they do their sign out, and I would just say that I've been pretty impressed by how Pierian has been trying to develop a portfolio of solutions that are right for various different groups.

Identifying NGS Stakeholders Key to Success

Dr. Sidiropoulos: So in the end, I would like to ... I think that was a relatively ... I went over a comprehensive strategic approach, at least some of what we try to do here, and so I would just like to reiterate, sort of in conclusion, that I recommend assessing your local landscape. Even if you're already in the game, I would say it's worth it if you haven't done it already to assess the landscape as you move forward. Identify those stakeholders and gain that leadership support. Identify and, if possible, align efforts with the institutional vision. Its important to identify barriers to implementation, and to define a strategic plan if at all possible. There are times when it's very nice, in the course of a conversation that seems to be going off course, to say, "Let's anchor ourselves back to our strategic plan. What is our mission here?" And that was really helpful to do. And it gets everybody back on the same page, and redirects the focus of what we're trying to do.

Dr. Sidiropoulos: And it's very obviously important, I probably don't have to tell this group, a realistic assessment of what your resources are. If you don't already have this transdisciplinary team concept, I would consider forming it, even if you have an informal transdisciplinary team, and leverage that transdisciplinary team as you go out to especially engage with leaders and stakeholders. I would strongly consider the concept of "Care Pathway" development and reflex ordering, and certainly if you're gonna do that, you wanna be proactive on reimbursement because the last thing you wanna find yourself doing is increasing all the testing you're doing without getting paid. And then I would say develop a dedication to demonstrating your value, and to education. We have found that to go very far for us and it's been very well received.

Dr. Sidiropoulos: Do not underestimate clinical informatics. I was grandfathered in to take this board exam because at least 20 percent of my clinical practice in the past three years has been strictly clinical informatics. I never anticipated that to be the case, but it is the case. I'm glad I did it, and certainly I would say grab a seat at that table, because it's all in the information and what you can do with it. And really look for those creative partnerships at this point, because that committee put out that publication, and it's absolutely fantastic in calling out some of the future directions. I just don't think everybody's quite there yet. And so at this point it may take creative partnerships. And we have found that creative partner in several key spheres with PierianDx.

Dr. Sidiropoulos: And so I would say, if you lock all this down, just be prepared to deliver. And with that, we have a couple minutes, but I will stop there. But thank you for listening.

Speaker 1: Thank you Dr. Sidiropoulos, I think we're up against the hour mark, but we're spend a few minutes on a couple of questions. So the first one is, "What recommendations do you have for effectively and efficiently working with IT? How did you engage with your IT department to successfully integrate systems?"

Dr. Sidiropoulos: Okay. So we engaged right away. I think the lowest hanging fruit, for us, was finding sort of a stakeholder in the LIS group. So in the Laboratory Information Support team. They are a part of the larger IT at our institution, so when we came together with someone there ... And so there was a key leader there who helped me understand the landscape in conversations with him. And he also was able to inform me how the organizational structure at the institution, how did that exist for IT? And he was really a key person to inform me, "If we wanna get this done, Nicki, we have to talk to this person." And then you have that next conversation, and that's someone outside your department. So it's really an organic process, but understanding the organizational hierarchy of IT at your institution is really, really important.

Dr. Sidiropoulos: So I would say, if you don't know, and a lot of us don't, start with someone who manages IS in the lab, or informatics in your lab. And certainly they'll have a grip on what the world looks like in terms of IT outside of your lab. And that's how we did it. And it took a little bit of time, but it certainly I think facilitated the right conversations as the right time, and didn't leave any key people out, which can then leave a bad taste in people's mouths. So that's how I managed it.

Speaker 1: Thank you. Next question is about the future of genetic medicine. So in your opinion, does the future of genetic medicine include fewer but larger [inaudible 00:56:49] or numerous, smaller, more focused [inaudible 00:56:52]?

Dr. Sidiropoulos: I mean, that's a great question, and I think we debate that all the time in our own lab. I think it probably depends a lot on ... At this point, you could say, well, the smaller, more focused [inaudible 00:57:06] you may be able to get those paid for, to some degree, based on how much they're costing your lab to do. It might be a more cost efficient way of doing things currently. But then again, based on your volume and your informatics approach, there's some people who are saying, "Go for the larger [inaudible 00:57:26] and use your bio informatics to drill down and report only on some." So I mean, that's a really timely question, and I think there are other considerations that will come down to what is your local landscape to help you answer that question at this specific moment in time.

Speaker 1: Thank you, thank you. The next question is about your interaction with BlueCross BlueShield. So when you spoke with the medical director at BlueCross BlueShield of Vermont? What concerns of objections did he raise, and how did you address those to gain agreement for coverage?

Dr. Sidiropoulos: Yeah, so I think medically ... What we've basically ... And approach we've taken is if I'm gonna be called to explain our [inaudible 00:58:18] and why it's clinically relevant, you know, we medically vet every target we put on our panel, meaning that it's gonna go on the panel if I have publications and evidence to prove something has diagnostic, prognostic, or predictive value in the human population. So I had, for every single gene on that panel, I could give him a reference that was relevant to say why it was clinically relevant, and why it should be on this test. I also had ... So that was really important. So medical vetting of your targets is very important and being able to produce that evidence for the person on the other end of that call is very important.

Dr. Sidiropoulos: I also think I was able to come together and I said, "All of the medical professionals who are involved in treating cancer care at this institution have taken part in designing this [inaudible 00:59:16]. So again, it was that sort of transdisciplinary ... This isn't just me as a lab director, and you could say, "Oh, well you're interested in driving your own mission." Well no, this was actually a whole transdisciplinary team and I was just a representative on that call. So I would say do your homework before you get on a call like that.

Speaker 1: Mm-hmm (affirmative).

Dr. Sidiropoulos: And then be ready to think about generating that evidence.

Speaker 1: Okay. I think we have time for one last brief question, which is, "Can you provide an estimate of the number of patients that have gone through your oncology panel?"

Dr. Sidiropoulos: Sure. So again, I laid out the landscape for you in terms of what our population is here. So this should all be viewed within that. Our first year for that one test, we had put through 200 cases, and so that's anticipated this year to go to 400 with some of the reflex pathways that we put in. So that's for that one test in this region of the country, the type of volume we were seeing.

Dr. Sidiropoulos: Hello?

Speaker 1: Okay. I think we're out of time, so the questions we will respond to directly to the folks asking the questions. Thank you so much, Dr. Sidiropoulos, for this insightful presentation and for all of the webinar attendees for joining. The recording will be available on the PierianDx website on Monday, and will be sent out to the attendees as well. So thank you all, and goodbye.

Dr. Sidiropoulos: Thank you.

Subscribe Now!

Stay Up To Date with Pierian!