Transcript: Industry Standards Needed for Next-Generation Sequencing in Cancer Care

Transcript:

Gordon Kuntz: Welcome to Oncology Innovations, a Journal of Clinical Pathways podcast, focusing on candid discussions with innovators aiming to advance quality and value through the cancer care ecosystem. I'm your host Gordon Kuntz. I'm a consultant with almost 20 years experience in oncology clinical pathways and the business of oncology. I've worked with oncology practices, pharma, payers, GPOs, and pathway developers, basically every aspect of the oncology ecosystem. I'm very excited about today's podcast because I'm joined by Dr. Ira Klein, VP of medical affairs for payer relations at Tempus Labs. I've had the pleasure of knowing and working with Ira for over 15 years.

Today, we'll be discussing the need for industry standards in NGS and CGP and how the lack of standards is hurting patients and providers in terms of access and administrative waste, and is hurting payers in subverting care to wasteful therapies and curving use of clinical trials. Ira, you chose an interesting topic. Your background seems particularly well suited for this topic after nine years at Aetna, including the last couple as national medical director, largely focused on oncology programs. You spent five years in a quality and value role with Janssen Oncology and are now with Tempus working with payers.

Let's kick things off by talking about where things are today. What kind of standards currently exist for NGS and CGP testing? And don't all labs have to be CLIA-certified?

Dr Klein: It's a complicated question, Gordon, because the labs that are generally used are CAP CLIA-certified and cap is the College of American Pathology. And CLIA is really a regulation. It's the Clinical Laboratory Improvement Amendment, I think, of 1988. And these are complimentary sets of standards that exist as, I believe, CMS and the Medical Specialty Societies' compromise over what we can do to create access for all of America for laboratory testing versus let's have a known and agreed upon set of standards that sets a quality threshold bar for what we're going to do when we approach patients.

So for CAP, the College of American Pathologists, they have a biorepository accreditation process program. And that takes the first, I would say the best sets of practices from all of the prior existing programs. And that includes things happening in other parts of the Western world, the OECD, what's happening at the National Cancer Institute and what CMS had prior. And so, you have this lab accreditation program in which the goal is to standardize the processes and it takes the approach that I am familiar with from way back, many years ago, when I was in health system administration and I had to get my health system, my hospital through the JCAHO accreditation.

And it's a checklist type of process and it lines up and compliments the clinical laboratory improvement amendments requirements as well. So you have what CAP requires, and that really controls for the, I would say, pre-analytic variables and specimen handling. And that includes how is the specimen procured, what is this preparation in shipping, what's the chain of custody, what's the storage, when are samples released, what do we believe is correct to forward signal in terms of what they want as tumor content, how the prep is done and what are the standards of that prep.

So if you want to be CAP-certified and also get your CLIA certificate, you're going to have to specify these things. And they're going to have to line up with what is required through these programs that comes from past standards and past experience of what good looks like. And so, I'm a general internist by training and I've been through many regulatory processes, but I'm not a pathologist. So I'd have to say that maybe I stated a Holiday Inn Express last night and I'm playing one on TV, but I could tell you as we move through this conversation that CAP, CLIA standards are not trivial.They're not the be-all end-all, but they do create the beginnings of a regulatory framework, which maybe needs in some ways to catch up with the rapidly advancing technology that we're having in comprehensive genomic profiling and cancer care. It's always a race between regulatory frameworks and innovation and technology to figure out which one we want to favor. Do we want to favor rapid progress with some less control mechanisms, or do we want to put the screws down on progress and take a very conservative approach to what we're going to do as testing available to the general public per an existing certification process?

So it's a balance, and I will say that I think that CMS has done as good a job as possibly could be done. Others may differ. We need some great lab pathologists to give commentary.

Gordon Kuntz: But it sounds like the sort of premise of the topic is that may not be enough. I mean, it sounds like that's sort of along the front end handling and sort of making sure that things are collected and transported and stored and that the specimen itself is preserved and collected properly. What other kinds of standards? Why is that not enough, I guess, is sort of to finish the thought? And then, what other kind of standards do you think are needed or are you thinking we might want to look at?

Dr Klein: Well, I mean, I talked about CAP and I talked a little bit about the CLIA certificates of which there's a bunch of different types. And they definitely are the key to regulating physician office laboratory environments. And you need that CLIA certificate if you're going to get testing performed, and that's really a lot about proficiency testing to say, can we actually take the specimen through a set of processes in an environment that will support those processes and does the result come out to be something that's reasonable?

Actually, it's still more process-oriented, not exactly results-oriented for the specific testing, but it is a set of standards that is not trivial to attain. And beyond the proficiency testing that happens at regular intervals to get this CLIA certification, you also have to submit a lot of paperwork that states that you have the right type of personnel in the lab to do this work. And also, you have to have very discrete and clear descriptions of the processes that are followed as for any kind of regulatory certification process.

Gordon Kuntz: Right.

Dr Klein: And so, you pay a fee that's based on the complexity of this and the number of tests you do per year and you set yourself up for serial proficiency testing, and then that's how it goes. It's not the be-all end-all, but it's an existing framework.

Gordon Kuntz: Yeah. Okay. There's an article for the Journal of Clinical Pathology that I've heard reference a number of times that indicated that only 37% of labs successfully identified genetic mutations during ingest testing in samples with known genetic mutations. They were created in with CRISPR. With the standards that you're referring to and sort of advancing this sort of thing, how would those address what are certainly perceived as quality issues?

Dr Klein: That's a really tough question. One other standard that I did not mention when I was talking about CAP and CLIA is New York state certification. And that is somewhat more rigorous look at some of the elements of sample procurement and processing that can be gained as an additional certification. And I don't want to go into details of that certification, but suffice to say that CMS feels it's significant enough to list it as part of the standard, or I should say the policy for a lab being able to do and get paid for comprehensive genomic profile.

If you look at the CMS website, I think it's standard 90.2. It states that to be able to process a test of large numbers of genes, greater than 50 genes, the lab has to have not only CAP and CLIA, but it could also have New York state certification. And so CMS recognizes that as another standard. And then we're talking about the treatment of advanced or metastatic cancers and the use of comprehensive genome profiling. But back to your original question, which I sort of skirted around a little bit regarding an aspect of testing that is a little bit troublesome, and it gets to that balance and trade off between supporting rapid technology advances versus regulatory oversight.

And this is the question of analytic validity. And you are citing an article that took a sort of artificially constructed sample that had both a physical product and then a computational product that was put through a series of commonly known and widely used sets of laboratories that do genomic profiling. And I believe it was a third of the vendors came out in pretty good shape, getting greater than 80 or 85% agreement. A third were hovering around 50% and a third did terribly, less than 50%.

 Now I'd have to reread the article to get the exact statistics, but the bottom line is not everyone's analytic validity is as good as it could be to give you confidence that the results would not only be, that the results would be reproducible. That's all analytic validity says. It says, if you're an archer and you're shooting an arrow at a target, you'll hit the same spot every time. It may not be in the middle, but you'll hit the same spot every time.

And the three main tenets of testing that are important are the analytic validity, like we just discussed. Then there's the analytic utility, which means, does the test actually have meaning? Is it applicable? And then the last, I'm sorry, is it logical? Meaning, does the test, test for the disease you're looking to treat? And then last is clinical utility, which is does it work? Is it applicable? So if you sum it up, you say that you've got to have tests that are replicable, meaning analytic validity, tests that are logical, meaning it's analytic utility, they actually test for the disease. And are they applicable? Do they actually help in the management of the patient, which is the clinical utility? And as you are pointing out, some of the testing on the market does not have the total analytic validity confidence that we’d like from those who've then looked at it in a synthetic testing environment. And that is not something that is fully done with all of the certification processes today.

Now, again, I would say we need to consult regulatory pathologists for the full story, but it's clear that we are allowing technology to move ahead and we are not totally set on what standards we're going to lock down in this environment. And it may be that per the original designation of CLEA, there's this balance between access to care and innovation and confidence that the testing is not harming individuals and that it's helping individuals. So there is always this balance.

Gordon Kuntz: So that sort of leads us back to our topic, which is really, what standard should the industry be considering? What should they be promoting for themselves to improve access for patients and reduce administrative waste?

Dr Klein: Well, I think as the industry matures, this is going to happen because people who use the testing know, and I think people in the business are not ever intentionally putting out a test that has, for example, a negative predictive value for a screening test that is too low, which would be a disaster. And same for specific gene identification of somatic mutations. What is the positive predictive power of the test? This is not intentional, but we're talking about complex bioinformatics that is done in addition to just the mechanical sequencing of particular genes.

I'm not going to say that's a total commodity, but it is possible to purchase sequencing machines. What is not so easy to purchase is the individuals who have incredible technical expertise that make sure that you are looking at the specimens in the proper way. That if you are going to receive a formal and fixed paraffin embedded specimen, that you may want to do some micro dissection to make sure that your tumor content is high enough to get a good yield. You want to make sure that your bioformatics team is doing the right sets of calculations.

So this is as much a bioinformatics exercise as it is a lab exercise. And I think that the winners in this game over time will be those who can gather the right bioinformatics teams that can then generate the data that the industry has confidence in. And I think that's really going to be key as we move on, because what we're finding is that there are a lot of ways to approach this area of biology. And I know that the audience here is probably going to be a little smarter about this than me, but may not be genetic pathologists. So I'll just say that you get into the minutia of what is the number of gene reads you're going to do at the DNA level. What are the number, if you're doing RNA, and maybe you're doing whole transcriptome analysis? How many times are you reading that as well?

Is it 500 depth? Is it 2,000 depth? Is it 100 depth? And what does it matter? What are you getting as you're going deeper? What are you gaining in sensitivity and specificity? When you have to submit, and this gets back to your original question about what else can we do? When you want to submit to the FDA and in a dossier, you have to tell them for everything you're measuring, what's the sensitivity? What's the specificity? What's the positive, predictive value? And that goes for everything. For example, we know that what's important in comprehensive genomic profiling is not only the single nucleotide variants and their mutations, but what about insertions? And what about deletions? What about gene rearrangements infusions? And we also need to look at copy number variants. We need to look at other factors like microsatellite instability. And then we need to look at a series of other calculatable parameters of the tissue micro environment, which include tumor mutational burden, and a few other factors that can be calculated.

So these things will go into a dossier and then that goes to the FDA and it follows the same process, although specific for this test or these types of tests that you see with 510K applications for medical devices. So you do have to go and put in analytic validity information, you have to put in clinical validity information. And of course, paramount to all this, what is the clinical utility of your test? And I won't say that getting the FDA certification means that, oh, one test is better than another test because in reality that may or may not be true. I'll just say that it takes a fair amount of time, money, and resources in order to assemble this set of documents to submit to the FDA. So if a company is doing fantastic work from a technology perspective, they may find that the trade off is too much and they're running too fast to get this done now.

And they may say, well, someday we're going to do it, but we know our test is great. The only question is what are, as you had pointed out earlier, the clear regulatory signs that say "your test is great" from analytic validity, analytic utility, clinical utility. And some are saying, particularly if you've went and done all the work to get your FDA approval, will say, well, that shows we're great. I don't necessarily sign on to that. Although I will say, if you can get through that exercise, it's probably a reassured process. But to say that you don't have it, that means you're not great, that's not true either. And even CMS understands that, which is why on their policy for comprehensive genomic profiling, they talk about CAP, CLEA, and New York state certification.

Gordon Kuntz: Right. So tie it back for me. Is it the FDA certification that would help, like a greater access to FDA certification? Would that help improve access for patients or are there other things the industry could be doing to improve access for patients and then reduce the administrative costs and waste associated with this kind of testing?

Dr Klein: I don't think that is really the key. I think that's something that a company could do, but I think that what really we're talking about in the public payer space is simply recognizing that the technology is advancing,  ,inimal standards have to be set, continuous surveillance has to be done, and policies will be set on the basis of what's observed. Now, CMS is fairly conservative. And like I said, it's CAP, CLEA, or New York State. And they're not saying in the local Medicare administrative carriers that you need the FDA certification.

They state that in their policy, that the Medicare administrative carriers of MAX have that jurisdiction to make those determinations. On the private payer space, the coverage is really all over the place. And I think that's really the problem when we are talking about, as you say, Gordon, access to care and understanding what's great. It's very frustrating because while ASCO has a definition for precision oncology and they have a definition for next gen sequencing, the private payers have interpreted this in many different ways and they put in many flavors of restrictions as a result. So it's not necessarily something I can puzzle out for you in a logical manner.

Gordon Kuntz: But I was kind of hoping you would here, actually. That would be very helpful if you could do that. Because I think, yes, there's a wide variety. And as I've talked to oncologists, they're very frustrated with not just CMS, but private payers as well, maybe especially private payers as they've gotten in the Medicare advantage space as well as the commercial space. There's such a variety of policies. And I think among large employers, there's just a lack of understanding, a lack of knowledge. They have a thousand other things to think about besides this. Whereas you and I may be talking about this a lot, but I think fundamentally again, as I've talked to some payers, one of their key issues is are you building a... The analogy that was used is, are you building a house with an inaccurate measuring tape, right? Where it measures something different every time, why would you build a house that way? I mean, that makes no sense at all.

And so, how can we establish some sort of a standard or way to reassure or communication maybe is all it is to reassure payers and self-fund employers that the testing that's being done is in fact as you described, actionable, accurate, and meaningful. And I think that's one of the fundamental issues that is driving this huge amount of variance that's out there in the market today.

Dr Klein: I think they have to look to the medical specialty societies and they have to devote some resources to understanding what would otherwise go into a dossier to say, oh, can we see what your analytic validity is? What your analytic utility is? And what your clinical utility is? And have someone at the organization who can review a couple of different companies worth of information. I can't say whether they should devote resources to that, but I think if they're interested in this, they should. I think that's a more nuanced way of doing it and would probably yield better results because if you're really thinking about this and comparing it to, for example, MRIs. So if you're getting a comprehensive genomic profile test and you look at the Medicare price, which is I think $2,920 or somewhere around $3,000, and then you think about, what's the price for a MRI of the abdomen and pelvis with contrast? It's probably not that far off, depending on where you go for your test.

So I think there needs to be sort of... And there's probably a hundred times more of those tests done than these tests. And yet I find a lot more hand waving with these tests and it would be nice to sort of… let's titrate the resources to what we're doing here, because what you're really thinking about with comprehensive genomic profiling is how good is this test in terms of the depth and breadth of it? How good is the reporting to the practicing oncologist who is busy, is seeing a lot of patients and has to make a lot of tough decisions? How good is the support? Can you call someone up from the testing lab and get a response within 24 hours?

How good is the turnaround time of the test and what does that do to inform care? And that care involves hundreds of thousands of dollars of drugs, could be between $80,000-$140,000 per year. So I actually think there's a lot more hysteria over this than is warranted. I think there should just be some logical assessments of various options that are open and not a series of difficult to comprehend policies instead. Because it doesn't make sense.

Gordon Kuntz: That's the defining feature of our healthcare system. Isn't it?

Dr Klein: Yeah.

Gordon Kuntz: Confusing.

Dr Klein: Exactly. Yeah.

Gordon Kuntz: So we also mentioned as part of our topic clinical trials, and I know that Tempus does a lot of work around aligning patients to clinical trials, but are there standards for testing companies that could be instituted to help to promote clinical trials on a much broader basis?

Dr Klein: Well, clinical trials have a lot of definitions, and a research effort can be a research effort for research sake. A clinical trial can be a clinical trial that marches a drug over stepwise to FDA approval. A clinical trial can be something that looks at a treatment strategy based on a testing strategy that then becomes a new standard of care. And I think that we've settled on a regulated free market approach to this.

And I don't think that's a bad thing, because like I said earlier on, there's this trade off between how you want to spur innovation and how you want to regulate and maintain the safety and health of the American public. Now, the FDA has gone all the way to creating its own, I believe it's called the Oncology Innovation Center. I can't remember the name, but they have their own group looking at oncology drugs, because it's known that in the past, the difficulty in marching through a standard process may take years, and when something comes out of clinical trials and is clearly in markedly better than existing therapy, patients who are dying of that disease don't want to wait those extra few years to get the drug.

So there is an expedited approval process on the drug side, and I think on the testing side, we're kind of in the same boat, where we've not stamped out everything that doesn't have 510K approval by the FDA, and we're letting those technologies grow and we're letting those processes improve. Because just like on the drug side, we don't want to impede the ability for innovative drugs to get to patients. We don't want to impede innovative testing getting to patients. And you have to hand it to our regulators for letting that happen. They don't do everything right, but it's actually a reasonable compromise if you think about the alternatives.

Gordon Kuntz: Okay.

Dr Klein: Oh, one more thing on the clinical trial side. It's that what really is novel and important in access to clinical trials is, if you're on the side of setting those trials up, standardize them. Democratize them. Let them be available not just to National Cancer Institute centers. Create a set of processes that's not an exhaustive set of hoops for administrators and practicing oncologists to jump through. And if you can do that, then all of a sudden you can have clinical trials happening in sites that are not major cities.

And the importance of that is that by democratizing the trials and hopefully moving them out to the community oncology space, you then create the appropriate demographics that you want in a clinical trial. Trials done in large big city centers tend to unintentionally cherry pick higher income white subjects. They don't look like America. So if you want to improve clinical trials, make it easier to get into a clinical trial, then you'll be able to have those trials in areas of the country that will recruit the demographics that more look like America. Then the treatment strategies and the drugs will reflect what's needed to manage all of America, not just part of America.

Gordon Kuntz: Right, that's a great point. I know from a pathway standpoint, aligning to and recruiting patients for clinical trials, or identifying patients to be recruited for clinical trials, more appropriately, is especially difficult in part because the criteria are not uniformly presented.

Every clinical trial defines the desired patient population in a little bit different way, and in several conversations I've had recently, the notion that you could use pathways or an EMR to automatically select a group of patients, whether it's from a genomic side or simply from a disease side, into a clinical trial, or again, let them know that the clinical trial was available, is almost impossible in today's world, because the definitions of what that patient population, the recruiting criteria are just all over the board. So I think that's actually not a bad place to start when one thinks about standardizations around clinical trials as well.

So testing companies, should they be involved... And again, when we talk about standards, it may be regulatory standards, it may simply be industry agreed upon standards of conduct or care, or data in fact. Is that something that there should be an outside body doing that? Is that something that the testing companies and the testing industry themselves should be doing, and maybe police themselves a little bit better? You mentioned clinical societies, the FDA, CMS. Is this a government oversight issue? Is it an industry agreement issue?

Dr Klein: I think it's more of an industry agreement issue. And you have to think about the stakeholders here, because you have so many of them, and all of them don't have, actually, the same aligned incentives. You have the testing companies, the lab testing companies, diagnostics companies. They have a set of interests in creating super high quality testing that has the right analytic and clinical features, and differentiating themselves in the marketplace over their competition. Which, by the way, it's a fiercely competitive market.

So they have their agendas. You have the providers who want the best care for individual patients and they want it quick and they want it understandable. You have the medical societies that are supporting them, who want to be able to give them some kind of protection and framework so that their work environment is not out of control and that they're moving the science forward. Then you have the health insurance companies, who are really in the business of controlling costs at a population health basis, and they want to know that these tests don't actually add additional expense to the total cost of care. Which we're finding out now they don't.

We're finding out now, actually, that comprehensive testing upfront is probably the most economical strategy for overall cancer care management. But that really hasn't penetrated through most of the payers, because the analytics and the data exercise you have to do is fairly rigorous to get through the pull through for the total cost of care. But those are big stakeholders, and then I left out the most important stakeholder, which is the patient, who is scared and just wants to get better. So you have all these stakeholders with these different agendas and different incentives, and aligning them all under one set of rules may actually unfairly hurt one of those stakeholders more than the other.

And so this is where I said there's this balance of trying to create some rules, some regulatory framework. And as I've said earlier, I'm not too happy about the cacophony of discordance in private payer policies on genomic testing. But I do think that as the evidence comes in about what is an appropriate strategy, particularly when I earlier compared this to MRIs, which is being ordered like water, this is actually a much cheaper way to manage care than if the pricing of the test is reasonable, then the overall result is going to be great for the system. And you have to try to bake that into it, but you can't actually regulate it with just one set of rules.

Gordon Kuntz: So maybe the issue is there's sort of three sets of “guidances.” We can make up a new of word here for that. Three sets of guidances for what you're talking about. One is industry regulation, which has the authority of a licensure or certification or something like that, which might be required to receive payment. There's possibly standards, which the industry could agree on, that don't have that same weight, but a bit of a Good Housekeeping seal of approval that maybe they come up with themselves. And then it seems like there are also potentially really just agreed upon guidelines or principles. And my guess is, as you move from the principles through to the regulation, as you said, it may disadvantage some, it may... there be less and less agreement as you get more and more of the stick approach to it.

John Fox, when we saw him recently, made the comment that genomic testing offers the possibility for every patient having the choice of the fullest range of treatments possible. Which I think is a principle that it's kind of hard to dispute, right? Patients want it. Oncologists want it. Testing companies want it. Payers at least better say they want it, or that's how you hit the front page of the paper. And certainly I would think that government regulators would want it as well. So maybe that's one of those principles that you start with and say, "Does everybody agree that we should know what's possible?" And then we can decide what to do and how to pay for it separately.

The standardization that's an industry piece might come around data standards or access to clinical trials or those sorts of things, and then the regulation is, if it's insufficient at the industry level, maybe the regulation comes into play that says you absolutely positively have to do these things. Obviously we already have the regulation that's there, but maybe it goes further. And maybe that's a way to think about this and try and get the world aligned.

Dr Klein: Yeah, I think that's a great summary of how this is a series of interlocking processes that doesn't ignore the incentives and the desires of each set of entities involved in this, but tries to balance out everybody's needs in a fair way that keeps the business going forward as both a science and a business. And John's quote, I think, works very well, because it syncs with what ASCO says in their definition of precision oncology, which is the use of molecular biomarkers to aid the diagnosis, prognosis, or treatment of cancer.

So what do we do to get this for everybody? And then their NGS definition, a technology that performs massively parallel DNA sequencing to detect genetic alteration. So it fits with the medical societies' ideas of how we're going to do comprehensive genomic profiling. And then as you said, there has to be some other compartment that looks at the business, and then some other compartment that looks at, "Okay, how are we going to regulate minimum standards?" And then of course, there's also the coding and billing end, which I'm not sure we have enough hours to discuss.

Gordon Kuntz: Yes, there's that. Which, having worked with some of that data, is really quite a mess. So yeah, that I think deserves attention. Whole separate conversation around that.

Dr Klein: Yeah.

Gordon Kuntz: But how widespread do you think, maybe among your testing company colleagues, obviously not just at Tempus, but in other companies among payers, certainly I know that would be welcome from a provider standpoint, I would think. Is there an appetite for having the kind of conversation we just talked about, where we can arrive on a set of principles that we can all agree on and then maybe think about, how can the industry self regulate a little bit more, maybe to forestall the pain that often comes with more government regulation for the sake of government regulation. Is there an appetite for doing something about this or do people just not want to talk about it?

Dr Klein: No, I think it's a sliding scale. I think the desire to talk about it and create new sets of processes, I'm not sure anyone wants new regulations but I think new sets of rules of engagement, the appetite for that is probably determined by where you are as a testing company. How new you are, what your economic resources are, how great you think your technology is in differentiating yourself from others in the business versus where you are in maturity. If you're a more mature company that has a set of processes that you've decided, well, we're not really going to invest too much more in that process, we're already established in the marketplace and that's fine.

So I think when you think about who wants to talk about it in the NGS industry, you probably have to go back to basic business and think about businesses and the life cycle of those businesses. Almost like the BCG four box model of who's incredibly new, who's growing rapidly, who's sort of just a standard ongoing business and who's just doing terribly, not understanding the marketplace. Different companies are in different places. Their desire and their appetite to manage or influence what's going to happen in the future from any kind of regulatory or rules perspective is going to be based on where they are in that four box model.

If you're the cash cow and you're pulling in money with the way things are today, you don't want anything to change. If you feel you have something new to offer the marketplace, but you're seeing some theoretical barriers to entry you may want the marketplace to change.

Gordon Kuntz: But wouldn't better acceptance on the part of payers through potentially again, let's just focus on the industry standards and common principles, isn't that sort of a rising tide lifts all boats kind of situation?

Dr Klein: I think so. I think for most companies in this space who got into this space because they believed in science, that is true. So I would say it's probably a 70-30 equation, where probably 70% would definitely sign onto your statement.

Gordon Kuntz: Well, I'll call that a majority then. So what are the next steps? What are you going to do about this, I guess is my question, what are the next steps to making this model that we just talked about, are there next steps? How does one move that forward?

Dr Klein: I really think the next steps follow the model that you really nicely laid out before, whereby there are certain zones of play that different stakeholders will play in to move their agendas forward. You talked about the sort of science testing environment. You talked about the payment and of population health environment. And then you talked about the regulatory and certification environment. And I think all three of those environments will move forward with different input from different stakeholders. One of the clear drivers of moving all three of those forward is going to be the weight of aggregated data that is now coming through to show that this is a clear and new way of approaching cancer care that is not a trendy niche. This will be the foundation of future cancer care.

As that becomes the reality for everyone because different stakeholders will absorb that fact at different times, because sometimes it's not convenient to absorb that fact early. If you weren't paying for something and now you have to pay for it, and maybe you're not analyzing your database well enough to know that you were actually paying more for it before, and you just didn't know it. But as every stakeholder absorbs the weight of the process data that turns out in a population health environment to show that precision medicine care is actually better planned care, better outcomes care, less medical waste care, and a greater uptick in clinical trials care. All of those factors actually weigh in on the side of cost savings.

When we look at the other end of the cost equation on the drug side, whereby the drugs may get a little cheaper, they're probably not going to get a lot cheaper. But we would like to pick the right agents, particularly when we're thinking about targeted therapy for tumor markers versus immuno-oncology, the immuno-oncology drugs are a little bit more expensive. Some of the earlier targeted therapies are becoming generics. Over time the use of comprehensive genomic profiling will better allow for the sorting out of appropriate therapy that is also economical therapy.

I think Gordon, you and I lived through the pathway revolution of 15 years ago and you were a key voice in that. I think that this is the next version of that type of turn of the wheel in what we do in health care policy.

Gordon Kuntz: Right. Great. Is there anything else that you'd like to share with us? I know it's a very broad topic and we could go on for hours, but anything else about comprehensive genomic testing and NGS testing that you'd like to share?

Dr Klein: I just wanted to say that there are levels of complexity even in the operational side that people don't often think about. Which is, the management of, for example, a solid tumor block from the time it comes out of the operating room where it's taken out from a human being, prepped and moved to a local pathology lab, and then possibly over to some either local or centralized industrial level processing facility for comprehensive general profiling. One of the things that turns out to be key is are you handling that tissue right? And I mentioned early on that you've got the formalin-fixed, paraffin-embedded that has certain specifications. You don't want your local pathologist who might be doing some cursory immunohistochemistry, well that would come off a slide prep, but taking some other sample, running a few genes and then sending it off somewhere else. And then it gets to that somewhere else place and you've discovered that you don't have enough tissue to do the testing.

And then the oncologist who's looking for guidance only has the results of the immunohistochemistry or maybe either some fish testing done or whatever and it comes back quantity not sufficient. Then you say, what do you do now? It's really beyond the scope of today's conversation. But when we think about advancements in the science of comprehensive general profiling, we as an industry are getting better at liquid biopsy. So if something comes back Q and S, wouldn't it be great if the sample was paired with a couple of tubes of blood so that the liquid biopsy could be run if the solid can't be run. And we're looking at now rates of concordance, it may not be the full span of testing, it may not be the same depth because you get many less cells out of liquid than you get out of solid, but you're getting concordance rates that are getting over 80, 85%.

So, in an environment where your choice is either to guess or to use a test that has 85% accordance with a solid tumor genomic profiling, I would say that's a pretty good option. And so, as we bake into those various stakeholders, including the testing companies, the people who pay for it and the people who regulate it on the behalf of the public, we have to start thinking about how the technology advances for the options when the handling of the specimens is not great.

Gordon Kuntz: Right, right. Yeah. Lots of steps in the chain. Well, Ira, thank you so much for your time today. That wraps up this episode of Oncology Innovations. Thank you, Ira, for joining us today and as always, thank you to the Journal of Clinical Pathways for producing this episode.

Please download, rate, and review and subscribe to the podcast. For more episodes, you can visit www.journalofclinicalpathways.com, or you can find us on Apple Podcasts, Google podcasts, and Spotify. Also, be sure to share on Oncology Innovations with a friend or colleague. Let me know your reactions to episodes, questions, or recommended topics for future episodes. You can find me on LinkedIn, or you can send an email to Gordon@GordonKuntz.com to request specific topics and guests. See you next time.