Maximizing the Potential of Precision Medicine

In this Breaking Down Health Care conversation, John Hennessy, MBA, and Michael Kolodziej, MD, talk about the economic challenges of precision medicine in health care, including the differences between Medicare and commercial health plans and the logistical challenges of arranging testing. 

Read the transcript: 

John Hennessy, MBA: Welcome to Breaking Down Healthcare where we'll be discussing topics involving health care in the United States. My name is John Hennessy. I'm a consultant with the Valuate Health Consultancy. And I'm joined here by Michael Kolodziej, who is a writer with a Substack called Breaking Down Healthcare, also working with Canopy as a consultant as they grow their presence in the ePRO space. 

Mike, we've been having a conversation about precision medicine and you've got a unique background as both a practicing oncologist and having spent time as a plan medical director. Talk about the economic challenges of not just the first test, but maybe let's talk about serial testing, which is becoming a bigger part of what we do. You know, that not all precision tests are the same. There are big panels and small panels. There are different costs of these things. Talk about those challenges of getting the right panel to the right patient at the right time and how payers think about that. 

Michael Kolodziej, MD: Yeah, great. So let's split this discussion into Medicare and commercial health plans because they're completely different. So for Medicare, the coverage of panel testing is very clear. Medicare will pay for somatic mutation analysis, that's like the FoundationOne test, for all patients with advanced malignancy, period. But they'll only pay for it once at the time of diagnosis. And like every other Part B benefit is a patient co-insurance, etc, etc, etc. That coverage exists because a few years ago, FoundationOne went to the FDA and to Medicare and said, “Listen, guys, you guys do this thing where you both review at the same time called parallel review. And if the FDA says it's good enough, then Medicare says they'll cover it.” 

And it hadn't been used very much and it was a little bit of a of a gamble. But they succeeded and as a consequence, if you have an FDA-approved panel using next generation sequencing, then it will get paid for by Medicare. Now, the results of the test default to the coverage policy for the individual drugs. Which is to say, as we've discussed before, if the targeted therapy is either FDA-approved for that indication or NCCN compendium-listed or compendia-listed, Medicare will pay for it. There's no off label beyond that. So if you happen to find a weirdo mutation in a weirdo cancer, Medicare ain't gonna pay for it, period. 

That's Medicare, very clear. Commercial health plans, much less clear. Because commercial health plans for complex molecular diagnostic testing are not compelled to cover the test in the same way, for example, they are compelled to cover FDA and compendium-approved therapeutics. Each health plan does their own analysis, and that has proven to be extraordinarily frustrating for the test developers and for doctors because the commercial health plans have said, "Listen, there just isn't enough evidence to warrant testing 350 genes in a patient with most cancers because there are 350 therapies, so we're not paying for this panel." 

And they've slowly, slowly, slowly adapted so that most payers now will pay for the panel in non-small cell lung cancer. And that's because in non-small cell lung cancer, there are a fair number of mutations that make a difference. They are not going to pay for the panel and a lot of other cancers. They just not got it, because there aren't that many genes that make a difference. Now, do they do that because the test is expensive, or they do that for another reason? The answer is, it's partially because the test is expensive. You know, these tests cost about five grand, give or take. But as everyone knows, the therapies cost a heck of a lot more than that. And I would argue that the biggest reason that they have put up this roadblock is they're afraid that doctors are going to look at these mutation results and say, "Oh my goodness, there's this mutation. I can use this drug and they'll want to use it in the scenario where the drug hasn't really been adequately tested. So, it will be a lot of money for a specialty drug for a targeted therapy without any reasonable expectation of benefit. And I think that that has caused a bit of a bottleneck. 

So, commercial payer is usually only paid for once as well. Repeat testing is usually not covered. Sometimes it is, you know, we do have evidence, for example, in lung cancer, EGFR mutation where there's evolution of the EGFR mutation and no mechanism of resistance that has a therapeutic consequence. But those examples are few and far between. So, repeat testing is not often done now. Testing is often not also done in early disease, usually metastatic, although that's changing based on some data about adjuvant EGFR-directed therapy and non-small cell lung cancer where that seems to make a difference. So, we'll see continued evolution. But I think the limiting factor at this point is just more knowledge that informs coverage policy. 

Hennessy: Yeah, and even if we think about that side, the other challenge we have is the logistical challenge, which is getting the testing done. And fortunately, we have some opportunities to use liquid biopsies. But talk about the challenge of arranging for this test to happen. Someone's got to find the tumor. Someone's got to get the tumor to the lab and there's time involved. So even the, we've seen data showing that even people who have the best intent to use this, sometimes the patients don't really get access to these therapies.  

Dr Kolodziej: Yeah, so historically tissue-based testing has been the gold standard. But in order for tissue-based testing to occur, the cancer needs to be sent to the testing laboratory. In general, unless you're at Sloan Kettering or something, that testing isn't done in-house, right? So you have to submit the blocks. And it takes a couple weeks, 2-3 weeks to get a result pack. Patients hate waiting. Oh boy. I have experience from my own personal practice where I told a patient they really ought to wait, and they didn't like that and they went to another oncologist who probably gave them empiric chemotherapy. There are some studies published to show that your outcomes are superior if you wait. That is that giving the targeted therapy first yields a better outcome. Although, those are imperfect studies. 

The other issue, of course, is that the amount of tissue required may be more than is available. So, we make cancer diagnoses by bronchoscopy all the time. That is not a lot of tissue. Or cytology. That is not a lot of tissue. So, an inadequate tissue quantity can screw up the test, then you've just wasted another couple of weeks and that's why liquid biopsies have become attractive now as I think all of our listeners know, tumors to a greater or lesser extent can shed cell-free DNA into the circulation and you can actually test that for the same mutations that you test the tissue. And there are studies that look at you know the concordance rate between blood-based and tissue-based testing. It's okay. Pretty good for common malignancies like lung cancer. They don't always agree, but you can get the blood-based test back in a week and that's good because then you can act if it's positive. What to do if the tests don't agree? Nobody really knows anymore whether to believe the one that's positive because it gives you a pat? 

At any rate. The other thing that warrants at least discussion is something called a 14-day rule. And the 14-day rule says that if you are a Medicare beneficiary and your diagnosis is made as a component of an inpatient stay, all that complex pathology analysis that occurs within the next 14 days gets billed back to the DRG. Now, what that means is that your hospital where that biopsy was done or that surgery was done, they aren’t going to be so keen on sending that tissue off for complex molecular testing because they're going to get, they're going to have to pay the bill out of their DRG money. And that 14-day rule has been a real pain in the butt. Blood-based testing is not held to that 14-day standard. Huge advantage for blood -based testing because otherwise you're going to have to wait 2 weeks plus another 2-3 weeks to get it to get tissue-based results because of the 14-day rule. There have been all kinds of discussions and attempts to make that go away but it hasn't happened yet. Crazy. It just hasn't happened yet and that I think drives doctors and patients and the test developers a little bit nuts. 

Hennessy: It's interesting you talk about that and we've often talked about how multidisciplinary oncology care is. But we've often thought of it sort of on the back end. But really, the team has to be on board with next generation sequencing testing. I mean, the surgeon has to be on board. The pathologist, the interventional radiologist who is getting the biopsy. All of them have sort of their own sort of preferences or things that matter to them, which may not be sort of driven around getting this result for the patient. If I'm getting a biopsy, I want to get in and out with as little damage as possible. If I'm the surgeon, you've got to convince him to sit back and wait. And to your point, that waiting period can be very tough, particularly in community oncology, where the patients might get happy feet and want to go someplace else, despite the fact the data I think to your point shows that if you wait and you actually have a therapy that's targeted, you might have better outcomes. 

Dr Kolodziej: True. There are many potential missteps. But as I said, in the most direct path to an answer is a blood-based test which couldn't be drawn in the doctor's office and sent. And the Guardant test, for example, and the FoundationOne cell-free DNA test are both FDA approved, so they get covered for Medicare. 

Hennessy: So, talk a little bit about just the expansion of these, you know, sort of targeted therapies or targets. You know, our late colleague, Jeff Ward, used to 15 years ago say, “If I read 2 articles a day, I can only keep up with 5% of what's new in oncology.” Are these new targets, these new therapies coming faster than we have a way to sort of cogitate it? Can it work up here? Do we need some things we talked about earlier? Big data, EHRs, do we need some of those tools to sort of maximize the potential that precision medicine offers? 

Dr Kolodziej: You know, we need all those, and I think it's true that no general oncologist, no community oncologist could possibly keep up with this literature. As you know, all of the testing companies spend a lot of time developing reports that are easy to digest and implement. And so, the typical report, which I will tell you has been approved by the FDA, the FDA approves the way that is presented to patients and doctors, includes mutations that are actionable in that they have an FDA approved therapy. Mutations that might be actionable in that they have a mutation for which there is a therapy, but not in that specific clinical setting. And then clinical trials that are potentially appropriate for patients with that mutation. Now, the clinical trials piece is a little bit imperfect in the sense that they list all clinical trials everywhere in America and the world for that mutation. And in upstate New York, when I practiced sending somebody to Norway for a trial was not an appropriate and was impractical, right? But the truth of the matter is that in today's world where we're starting to think about remote clinical trials and getting patients on trials from where they live, that might be overcome, right, that there's a potential that that would get overcome. That's the only way a doctor could possibly, you know, the common stuff, you know, right, EGFR, and the rest. But the less common things are more complicated. 

Now, there are people who strongly believe that molecular tumor boards are the answer to this, whether they be in person or virtual. And there have been a smattering of studies that suggest they might be helpful. But most community oncologists, they don't have access to tap into a molecular tumor board. And I think, you know, the question really becomes, does the answer provided by the molecular tumor board really improve the outcome for the patient? As far as I'm concerned, the evidence of it makes you feel good, right? A bunch of smart guys. I do research on this gene. I know more about it than anybody else in the world. I want to know whether what comes out of their mouths ultimately impacts the patient or not. And I think the evidence for that is a little bit unclear. 

Pathways help, of course pathways help, but pathways really are a direct reflection of what that report that's generated by the testing laboratory tells you to do. It reminds you, pathways remind you to make sure that you've looked for GFR and Alk in your lung cancer patients, because it makes a difference that you treat patients with that therapy, and it makes a difference that you treat the patients with the therapy up front. But yeah, ultimately, yes, artificial intelligence, real world evidence should move the ball here to help us understand whether that's FDA approved or not FDA approved, whether that mutation means something or whether we should act on it.

Hennessy: Super. So, Mike, let's talk a little bit. We've talked a lot about NGS testing, but let's talk about, you know, not every population is the same. We have populations that have economic challenges, that have health disparities. I had a nice conversation with a young fellow at the ASCO annual meeting talking about, you know, non-white patients in their facility being less likely to opt into NGS testing. You know, we've got this great clinical promise, but talk about the challenges that subpopulations may have and sort of how that's impacting, you know, the care that they receive at the end of the day.

Dr Kolodziej: Yeah, John, there is compelling and I'll say almost embarrassing data that suggests that we don't test with the same degree of thoroughness in minority populations. And I have no idea why that is, really I don't, but it needs to be fixed. And it needs to be fixed both because it represents high quality care, right? But it also needs to be fixed because we've known for a very, very long time that minority populations do not go on to clinical trials. The differential between Caucasian middle class and minority populations is dramatic and horrific. And to the extent that we can foster a pathway towards easier enrollment of minority populations in clinical trials, particularly for targeted therapies. Boy, I'll tell you what, that's what we would define as a win-win out of the gate. And so I would say that we need to start thinking about this. 

In fact, I had a lot of discussions with my friends, Ira Klein and John Fox, both of whom work for testing companies now, and I feel that it's time for us to consider doing molecular testing on certain populations as a quality measure. Just like chemotherapy in the last 14 days of life and all that stuff, the evidence is good enough that you can say in non-small cell lung cancer, molecular testing should be done. That is a quality measure. And I think we should get behind that. 

Hennessy: I think it's really important and not just for populations for the access to the testing, but so that when we have these clinical trial results, they're reflective of the patient in front of us.  

Dr Kolodziej: Completely true.

Hennessy: It's a select subset that happened to have the means to make it happen. So 100%.  

Dr Kolodziej: Yeah, you know, looking back and saying, “Oh gosh, I wish we knew in that population that the response was different.” That's just a crappy excuse. We just need to go ahead and answer those questions from the get go. 

Hennessy: Thank you for watching this installment of Breaking Down Healthcare. We hope you've enjoyed the conversation and learned something you didn't know about health care in the United States. If you have questions or topics you'd like Mike or I to discuss, you can use the contact us feature on the website and tune in for future conversations because we've got a lot of work ahead of us.