MoIDX Open Public Meeting - May 30, 2019 - JF Part A

Dr. Gary Oakes:
Welcome to our Open Meeting, first one that we've had for MolDX since the 21st Century Cures Act.

We've elected to go ahead and have them in the different jurisdictions until we can get CMS to say that we can also have these meetings. Because as you know, there's four contractors involved. Palmetto, WPS, CGS Admin, and Noridian. So it would just make sense that we could have one meeting and then Paul could be out on the golf course instead of having to be here today. [inaudible].

So, with this, Paul Gerrard was kind enough to be able to join me today to help explain some of the nuances that I probably have no clue about and answer some of the high-tech questions that you might have. So, Paul, thanks for joining us. Christine is the one who facilitated all this and give her a big round of applause.

OK. So, on the record, we are being recorded. The recording will be transcribed.

I can't speak any louder. Do I need to stand over here closer to the phone? Yeah, we can do that.

So, the recording will be transcribed. The recording is an MP3, I think. It's going to be in that format, and the transcription will become part of the LCD record, which is publicly viewable.

So, with that, choose your words very nicely. You can't berate the Medical Directors. No, you can if you want to, but that's more for the transparency.

Also, to be able to speak today, you have to have completed a conflict of interest form. That's the new requirement because we need to know where you're coming from. All right. And with that, any other thoughts that you have?

Dr. Paul Gerrard:
Try and choose my words wisely.

Dr. Gary Oakes:
Yeah, you always do. So, let's get underway and do you have the changer or is [inaudible]. Well, Paul, this is more in year realm because you do this every day, so I'll defer to you.

Dr. Paul Gerrard:
OK. So just before we get into talking about the LCD's, one of the things we want to talk about was how MolDX goes about making coverage decisions. We got a statement of work for Medicare in terms of how we should evaluate tests for coverage decisions.

If you look at the way if you look at a number of NCDs on diagnostic tests, generally speaking, including the NGS NCD, Medicare references this hierarchical framework for diagnostic tests by Fryback and Thornbury. And if you did see but there's a reference here as a paper in 1991 that contains in detail, and basically it starts down at level one go with technical quality of the tests. Now, now, they were talking about imaging tests rather than in vitro tests, but it starts at level one with regard to technical quality and then it goes up higher and higher in levels until we finally get to the impact on patient outcomes. The idea being that for a test to be reasonable and necessary it has to reach level 5, but these lower levels are the necessary building blocks upon which you get to level 5. Each of these lower steps are not sufficient but are (or) necessary to reach that level.

Next slide please. Next slide please. I think it's trying.

OK, so the diagnostic tests really must be expected to improve our patients' outcomes will be considered reasonable and necessary. Now when it comes to in vitro diagnostic tests, we implement this framework using the ACCE model, which doesn't describe that hierarchy, but nevertheless captures elements of technical quality and going into how the test impacts the patient's analytical validity, clinical validity, and then clinical utility.

And really, this comes from the idea that a diagnostic test must be expected to improve patient outcome from the regulations and from a court case, Kort versus Burwell, where that particular interpretation of the regulation was upheld.

Next slide, please.

[inaudible].

Yell closer and maybe yell in the direction of the phone.

Dr. Paul Gerrard:
OK. So, with that, the first policy is MolDX Next-Generation Sequencing for Solid Tumors. So, this is a policy that does not describe a specific test. Rather, it describes specifically those first levels of the Fryback and Thornbury hierarchy, and how we will go about considering coverage of next generation sequencing tests and solid tumors for cancer within those first two rungs of the hierarchy.

The LCD is meant to fit within the component of the NGS NCD that describes contractor discretion. Now the NGS NCD is under reconsideration, so that may have an impact on the final LCD, but we wrote it with the present NCD as written in mind. And basically, it requires submission of validation documents to MolDX to ensure that an adequate validation of an NGS test has been performed.

That if someone does not submit those coverage documents, then we will not know that an adequate validation has been performed and the test will not be eligible for coverage. That does not establish clinical utility of the test. That may require additional work, but once again, clinical utility has to build upon the lower levels of technical quality and validity. And so that's really what this LCD gets at.

All right. Next slide please. And I don't know, should we take questions now or wait until the end or?

Dr. Gary Oakes:
Your call.

Dr. Paul Gerrard:
If there are questions, I guess feel free to raise your hand as they come along or at the end if you have questions, feel free to raise your hand then.

So, the next LCD is the Next-Generation Sequencing LCD for Myeloid Malignancies and Suspected Myeloid Malignancies. So once again this particular LCD describes how we go about evaluating technical quality of NGS tests in myeloid malignancies.

The clinical utility may not necessarily be established just based on the technical quality alone, although for biomarkers that are either required as part of an FDA approved, or required as part of the FDA label of a drug, or biomarkers that have an NCM two way or higher indication, we will accept that clinical utility has already been established.

As with the solid tumor LCD, this requires submission of documents to MolDX, by which will verify that the test has been adequately validated and does what is necessary to be considered reasonable and necessary.

Next slide please.

OK, this LCD Drug Interaction Testing, so this LCD does not describe a single test. This LCD is a non-coverage for a class of tests. We are currently only aware of one test like this available on the market, but this LCD would be applicable to any similar tests that either are on the market that we are unaware of, or that could come to the market. This LCD non covers drug assays that are particularly designed for identifying medication interactions, a drug interaction test in this LCD is a test that does essentially 2 things. One is that it identifies drugs in a person's system. The second is that it tests for interactions among those drugs.

If a lab is running a test just to identify unknown substances, that is not a drug interaction test. So, this LCD does not speak to those kinds of tests. It speaks only to drug interaction tests having those components.

It is a non-coverage policy because at this point in time, the evidence that has been developed that we are aware of to support the use of such a test supports the use of such a test for indications that are not part of a defined benefit category.

As a diagnostic test, which is a defined benefit category that we are able to cover as a MAC, we are unaware of evidence supporting that it is reasonable and necessary.

And next slide please.

OK. LCDs regarding proprietary tests.

Next slide please.

So Guardant360 Plasma-Based Comprehensive Genomic Profiling in Solid Tumors. This LCD expands the coverage of Guardant360 giving it pan cancer, solid tumor coverage for non-CNS malignancies.

The coverage decision was made based on evidence showing that the test performed similarly across mutation types and types of cancers. It also defines in the NGS NCD there is a requirement that a test can only be used one time for the same primary cancer.

And the definition of primary cancer is not given in the NCD, so we provide clarification on what we, as a contractor, interpret primary cancer to be based on cell line and treatment response in a particular malignancy.

Next slide please.

Prospera. So Prospera is a cell free DNA blood test to assess for kidney allograft rejection and transplantation. This policy provides limited coverage of Prospera test to be used as a non-invasive test when a physician believes a biopsy would otherwise be warranted to assess rejection.

Next slide please.

TruGraf Blood Gene Expression Test. So, this is another test for kidney transplantation. This test provides limited coverage to the TruGraf gene expression test to assess for kidney allograft rejection following the transplant. It provides coverage for the TruGraf blood test as an alternative to surveillance biopsies.

Next slide please.

Decipher Prostate. So, there are, we have two LCDs that provide limited coverage to the decipher prostate test.

There are pre-existing LCDs regarding the decipher. What we have done traditionally in MolDX, and we may be changing this in the future, but for right now we're stuck with the convention, is for a given test with separately defined indications in a different disease state. We have issued separate LCDs, and so we have two different LCDs for the Decipher test. One LCD provides coverage for the test in low-risk intermediate prostate cancer and the other in, I should say unfavorable intermediate risk prostate cancer and favorable intermediate risk prostate cancer.

The and as I said, we also have some separate policies that covers that are not part of this cycle that cover decipher for other indications.

Next slide please.

The Pigmented Lesion Assay this is a limited coverage policy for an adhesive patch biopsy of a pigmented skin lesion that is intended to help rule out melanoma and the intent is to avoid unnecessary biopsies. It is limited to be used on one pigmented skin lesion.

So, if a patient comes in covered in moles, the provider cannot use this on multiple skin lesions. It may only be ordered by a dermatologist, and the dermatologist must have a plan in the record to follow up with the patient regarding the lesion that has been, so has a biopsy using the attached biopsy.

Next slide please.

OK. So, any comments or questions on the policies?

OK. Then I will turn it back over to Doctor Oakes.

Dr. Gary Oakes:
Walking over this way because the mic, apparently, it's sweet spot right over here Paul, so.

You don't have any questions? You're all afraid of being on record. You know that's something that we'll all have to get used to but it's what they do with the med CACs, and it's been very useful with transparency, so gratefully have embraced it. We've been working to refine it, but those going to speak today about a conflict of interest, right? We want to know where you're coming from.

Wouldn't be used against you, but it would help us better interpret the comments.

We have two presentations, so if we can go to the next slide.

And who's the lucky fellow on this one?

Yeah, if you would, sir. Probably do need to be a little louder.

Dr. Elai Davicioni:
Can you hear me now?

[inaudible]

Dr. Elai Davicioni:
Hello, Hello hello, I'll speak louder. Well, thanks for having me today. Next slide please.

Gary Oakes:
And the slowness is because this is a teleconference that anyone could have signed up for, so it's coming from the mother station.

Dr. Elai Davicioni:
Thanks. So as doctor, my name is Elai Davicioni, I'm one of the founders and the Chief Scientific Officer for Decipher Biosciences, the recent called the Genome DX. I'm speaking about the two proposed LCDs, both in prostate cancer at initial diagnosis, favorable intermediate and unfavorable intermediate risk disease.

Next slide please.

So, I think it's important first to remember that their prostate cancer is pretty heterogeneous. And so NCCN guidelines and others already use biomarkers to stratify the patients from very low risk all the way to very high risk of disease. And I'll talk about those biomarkers in a minute.

Next slide please. It's animation, so can can you go to the next slide? Note to self, don't do animation on these meetings.

Why is patient risk stratification important? Because we don't, we do risk adaptive therapy. You know, when we practice or when we treat people with cancer. Prostate cancer is no different. So, the important thing to note is most people diagnosed have lower disease, fewer people have high-risk disease.

Can you advance please?

And with low risk, we try to offer less therapy and with high risk, we try to offer more therapy because we really want to prevent people developing metastatic disease. That's really what kills people diagnosed with prostate cancer. And that's also what costs the healthcare system the most money. About 90% of the cost of treating prostate cancer are when people are already in a metastatic state, so screening and patient risk stratification are important tools to try to prevent that from happening.

Please advance. Yes, advance again, please. I will not do this.

So intermediate risk prostate cancer, you can keep advancing this.

Intermediate prostate risk prostate cancer is actually the catch all, you know, pathologist doesn't know what to call something, they tend to call it intermediate risk. And that's just kind of a trait I guess of the human brain.

But for intermediate risk, guidelines tell us we can offer pretty much any therapy. It's extremely confusing to patients, but also to providers. So, you can go from intermediate risk prostate cancer, all the way from active surveillance, which is really just following the cancer to actually very expensive multiple lines of therapy, radiation with brachytherapy and hormone therapy. And so again, if we wanted to risk adaptive therapy properly, you need to have good risk stratification tools.

Next slide please.

So, risk stratification is not new. I think you recognize with already some decades ago and what's important to note in prostate cancer, the most famous one is the neutral risk classification. This is really the foundation of what NCCN practice guidelines and others, the AUA practice guidelines, how they stratify patients and these models to model, they were they were originally developed to predict biochemical recurrence.

And you can see here they're actually employing three biomarkers, the grade of the tumor, which is a histologic biomarker, the PSA levels, the prostate specific antigen protein biomarker, but also the clinical stage. And not in some respects is also a biomarker, it's really the size and location.

So originally came up with the three strata. They've since subdivided intermediate risk into favorable and unfavorable, and that's why, the why the proposal LCD looks is separated into two.

Can you please advance?

So, these, this biomarker-based risk stratification has always have been adopted by NCCN and slightly modified over time.

Next slide please.

So, an important thing about any biomarker is evaluating its performance. So, one of the tools, one of the metrics we use, is the index the area end of the curve from rock analysis, and what we know about existing biomarkers. So first a coin flip this on the same page with the MAC, the Coin Flip Index is 0.5. And what we know about prognostic biomarkers and prostate cancer is they have around, from a coin flip to around 0.71 is perfect accuracy.

So, I think we all know, we realize these are suboptimal tools for risk stratification because they're actually not that good at predicting events. And actually, they were first proposed as predictors of biochemical recurrence.

Biochemical recurrence is when the PSA comes back after initial prostate cancer therapy.

Please advance.

But this is not a surrogate for survival, and this is actually one of use have biomarkers that actually predict survival endpoints. Actually, metastasis with survival is one of those improvements to be a surrogate for overall survival.

We believe this because what we're trying to do is improve survival from prostate cancer, not a surrogate of survival. And we're not, really treating that PSA value, we're trying to treat a person.

Please advance.

So, if you look at biomarkers in NCCN and then the NCCN risk groups and treat NCCN risk group as a prognostic biomarker-based mode, l across multiple data sets that are being published in the literature. You can see the index is really sub optimal, 0.65 is better than a coin toss but not much better.

Next slide please.

And I think NCCN has recognized this, and they continue to follow the publications around Decipher. And so now you can open up NCCN and there's multiple indications for populations of men with prostate cancer where they recommend using genomic biomarkers, such as Decipher.

Next slide please.

So, what is Decipher? Decipher is, as I mentioned, the prognostic test. It uses a whole transcription assay, and it reports, it uses actually just 22 genes of that assay, and it reports a score from zero to one which is translated into probabilities of survival events with [inaudible] from prostate cancer, but mainly clinicians use the risk groupings, which have been demonstrated to have utility in different practices. So low, intermediate or average risk and high risk.

Next slide.

So last year in the Journal of Clinical Oncology, Spratt and colleagues published this study, I think, which really shows how physicians should use information like Decipher in conjunction with established prognostic models such as the [inaudible].

Next slide, please.

What you can see here, so these are cumulative incidence of metastasis on the left, prostate cancer specific mortality on the right, and they show stratification for survival events based on the NCCN risk groupings.

And what you can see is the, between NCCN low and favorable intermediate, there's actually no differences in survival between unfavorable and high risk. There aren't differences, so this basically tells you that you don't have good risk stratification because if you have different risk groups, they should have different outcomes. You can see these patients have been followed for quite a long time.

Next slide please.

And just to further reiterate the points, when you look at the predictions for metastasis, again based on the C-index, you could see that the prognostic models based on clinical variables, pathological variables, CAPRA is another one of these, along with NCCN. They really aren't that much better than the coin flip.

Next slide please.

So, the authors put together a framework where you could see you have. It's easy enough that a urologist can do it. Senior authors are radiation oncologists. It's an easy summation method. You basically take your NCCN risk group, which you know from PSA, their grade, and their clinical stage. You get the Decipher risk scores low, intermediate, and high.

Each one has different points. You add up the points and you get this combined NCCN Decipher risk model. And what you can see, when you add the genomic information.

Next slide please.

As you can see, you have now risk groups that clearly have very differentiated survival characteristics for both of the endpoints that were looked at, and now we can see how applying risk adaptive therapy low clinical genomic risk group patients. These are these are men that we can withhold or use more conservative approaches whereas the people with the highest risk, we really need to intensify therapy and these are the people that actually want to send to some of the novel clinical trials for novel (agent). And these are the people that we need to we need to know their risk ahead of ahead of time so that we can do something about it.

So that's really the principle of risk adaptive therapy and better biomarker, does it helps you do it better.

So, the next slide please.

And this is another independent validation cohort. People treated people at initial diagnosis, so you could see from time from biopsy those with low clinical genomics really don't have any advances even out to 15 years after initial diagnosis.

Next slide please.

I'm going to skip this slide, just in the interest of time, Doctor Oakes.

An important part of any new model is you have to see how it changes the existing framework. So, if you have a new prognostic biomarker and it does not reclassify what you know already from the old the old system or old framework, then there's really no use in having it, and it's probably used for the US public to pay for it, Medicare to pay for it. And what you can see again from the Spratt paper is about 2/3 of patients were reclassified either to lower,

Please advance.

Either to lower or higher risk groups. Importantly for intermediate risk patients.

Can you advance?

So that's the box is highlighted in red.

Please advance.

You can see that 44% of the favorable intermediates are classified into low risk.

Reclassified as low risk and these are people that most clinicians, most providers would recommend surveillance. So really following the tumor. But on the other hand, 44% of the unfavorable intermediate risk are actually classified to higher.

Typically, for high-risk prostate cancer, we want to extend the duration of therapy for longer durations of hormone therapy, more intense multimodal therapy to treat this disease.

Next slide please.

Recently, a group from the state of Michigan, they have a statewide registry that's actually funded by Blue Cross Blue Shield of Michigan. They published a study that looked at Decipher and Prolaris as well and looked at what would be what's the impact comparing people, men that got the test versus men that didn't get the test over the last few years in the state of Michigan. They found actually that in what's highlighted in red, even among men with unfavorable disease, When, when no genomics was used, the majority of these patients were treated with radical therapy. Very few of them were offered active surveillance and treated with active surveillance. But you could see in the in the middle bar over here, that's probably hard for you to see this, that when the test came back with the low-risk results, so below the threshold, 28% of men were offered active surveillance.

We can see already that genomics does have an impact in what physicians do and what they end up recommending and offering their patients.

Next slide please. Yeah, please advance to the next slide. Thanks.

Specifically for unfavorable intermediate risk, current guidelines recommend four to six months. Men do not like to have chemical cancellation under any circumstances; it causes a lot of concern. It actually has significant both near and long term side effects. So, this is actually a study by a Canadian clinical trial group, and what they sought of to do was use more modern, more intense radiotherapy with cold hormone therapy from patients. So, these are all patients that by guidelines are recommended to get radiation with or without brachytherapy, radioactive seeds that are implanting in the prostate and four to six months of hormone therapy. Again, none of these patients got hormone therapy. And they tested to see how well Decipher could stratified this patient population.

Next slide please. So that, please advance.

Really what they found in the study is like those with a low Decipher risk did not have any of them.

The interpretation is that when you have a low Decipher, even withholding therapy in this case, hormone therapy and brachytherapy can actually get a very good outcome. And that's not the case for the high Decipher patients. So clearly undertreated those patients in the context of this.

Next slide. And again, please, yeah.

They looked at and you see, and you can see incorporating the NCCN information. Established biomarker plus the new genomic biomarker Decipher. You can get predictions actually close to about 90% accuracy, which is closer to one and more where we want to be for accurate risk stratification.

Next slide. I'll skip this study. Just I'm looking at the time here. Please advance thanks. The next slide.

So really, what does this mean in summary?

Using Decipher specifically for men with intermediate risk disease, we can really identify a pretty large group where we can probably safely offer them less aggressive therapy, such as active surveillance, and also the emission of costly and potentially costly that is having lots of side effects. Other therapies, such as brachytherapy and hormone therapy. And also, important specifically for the Medicare population, is to identify those most at risk of developing metastasis, intervening with those patients with more intensive regimen upfront and that is the goal of preventing metastatic disease from happening. So, I think this will also greatly benefit not only the Medicare beneficiaries, but also the healthcare system overall.

And I think that's, that's the end.

Next slide. So, the last slide. Yeah, you may want to just advance.

So here in summary, just gives you an idea of how people are actually using this information clinically when the clinical genomic risk is low, the event rate is going to be low, so providers are more comfortable recommending options like surveillance, which patients seem to like a lot. But when the risk is high, that's when they start to use multimodal therapy and then newer agents, and actually also even recommending patients more for clinical trials, which is a big thing.

I think the next slide is the last slide. Thank you, and I won't do those animations next time.

Dr. Gary Oakes:
Well, we're all learning from this.

Dr. Elai Davicioni:
Am I supposed to take questions or how is this done?

Dr. Gary Oakes:
We can take a moment and take questions if there's any in the audience. Are there any questions from anyone on the phone?

Hearing none, let's move along. Thank you.

Dr. Burkhard Jansen:
Good morning and good afternoon.

Thank you very much to the MolDX team and to the Noridian team for a very thorough review of our test.

What I would like to do over the next 10 minutes is one to briefly summarize the key highlights of the test and key elements of the LCD. Then I would like to briefly spend a few minutes on our current standard of care of pigmented lesion management and address why improvement is necessary. And then last but not least, I'd like to go through a few highlights on data packages and some technical elements of the test.

Next slide please.

DermTech's pigmented lesion assay is a non-invasive gene expression test for pigmented lesions with clinical suspicion of melanoma that helps clinicians rule out melanoma and the need for surgical biopsy. And from Doctor Gerrard earlier we heard that patient benefit is an absolute key element, and our pigmented lesion assay certainly meets that criteria. The test detects melanoma more accurately and early; it guides biopsy decisions and avoids unnecessary surgical procedures. It's a non-invasive test.

You see at the bottom of the slides, that we hope to contribute to a paradigm change in these three easy steps, non-invasive sample collections as simple robust assays and the molecular pathology report in 48 to 72 hours. While lots of people talk about the non-invasive nature of the test, the key differentiation of the test really comes from its superior performance.

Next slide please.

We very much appreciate that we heard from the MolDX team, from the Noridian team and from other MAC contractors that these contractors will provide limited coverage for the pigmented lesion assay. Again, a test that is indicated, and I'm taking this language directly from the LCD, that the test is indicated for use on melanocytic skin lesions with one or more clinical or historical characteristics suggestive of melanoma when a clinician trained in the clinical diagnosis of skin cancer is considering the need for biopsy to rule out melanoma.

Next slide please.

Again, to us, the highlight of the case is really its clinical performance, and you can see that here, very clearly, when you compare the performance of the test. In turn healing pathways to the historical diagnosis pathway, which relies on visual inspections plus histopathology. You can see that the negative predictive value of the historical diagnosis pathway is around 83%, whereas the negative predictive value of DermTech's pigmented lesion assay is above 99%.

The test effects biopsies, the number needed to biopsy comes down by about a factor of 10. The number needed to excise is significantly reduced as well, because we all know that biopsies in pigmented lesion assay management very often trigger additional surgical procedures, re-excisions, and other steps.

And last but not least, the test does not only improve pigmented lesion management, it also saves money. At a cost per lesion tested of around $650, PLA saves systems about $300 based on a cost per lesion. A cost to fully adjudicate the pigmented lesion of $947 as summarized and highlighted in a health economic study published recently in JAMA Dermatology.

Next slide please.

And now, moving very briefly to the second part of my talk, I just want to briefly remind everybody what current treatment is, lesion management looks like, it is obviously suggestive at both levels. It is suggestive at the visual inspection level, and then it's again subjective at the histopathology level. At the visual level, if you look at these two lesions, they pretty much look the same to probably most folks in the audience. They look the same to a group of high school kids. They look the same to pathology. They look the same to clinic clinicians and clinical dermatologists with decades of experience. It's very difficult to differentiate those lesions based on morphological criteria. And therefore, in up to 90% of cases, biopsies are done on benign regions that are not melanoma, and this is where the real difference comes in.

Next slide please.

Issues with our gold standard of histopathology. It's very nicely summarized by a number of recent publications, and the one that's probably most telling is a very large study by Elmore ET al, including 187 pathologists and genetic pathologists. And the study demonstrates that histopathology for early-stage melanoma is exceedingly poor. Histopathology for freckle is really quite good. Histopathology for really nasty advanced melanoma is also pretty good, but if you look at lesions as melanoma in situ lesions and early-stage melanoma, where appropriate, management can make the best, the most difference, you can see here that the sensitivity, for melanoma in situ and stage 1A is only around 65%. Fairly close to the coin flip that Elai Davicioni only mentioned in another context a few minutes ago.

And again, if you look at the right side of the slide, you have an example here, I don't know whether this is really easy for folks to read, but what you have here is the melanoma in situ cases based on a consensus diagnosis. And 36 pathologists and genetic pathologists looked at that lesion and came up with 18 different diagnoses covering a spectrum from pretty much a freckle, at the top of the slide, to advanced melanoma at the bottom.

So again, I hope we can all as a group, appreciate that improvements are needed.

Next slide please.

I want to briefly touch upon another key element that differentiates the pigmented lesion aspect, and that key element is its non-invasive sample collection platform depicted here on the slide where we create an ability to assess 100% of a pigmented lesions by covering the entire surface of the lesion where cells migrate up through various levels. And that is very, very different from the 1 to 2% that routine histopathology looks at.

In histopathology, very often you are dealing with a surgical biopsy that may not be complete. And then the pathologist looks at only 1 to 2% of that lesion, so he looks at a fraction of a fraction. And that's how our gold standard.

Next slide please.

Just very briefly on targets, we put a lot of effort into optimizing the targets. We narrowed things down from over 300 targets to 17 to 2. That really drives the performance of the test.

And again, in the interest of time, I'd just like to focus briefly on frame one of the two targets our test uses. And I just want to mention that crane is not something that that only DermTech focuses on. PRAME is a very well recognized target. That Myriad test also includes and is also a key element of cancels uveal Melanoma test. So well validated gene expression target that drives the performance of this test.

Next slide please. I think we've had that slide before. Can we just go? Oh, yes. Thank you.

So again, the PLA is a very highly validated solution, and we are very proud that the foundation of this test is now summarized in 14 peer reviewed publications from laboratory analytical validation, clinical validation to utility to ongoing registry studies, all published in leading dermatology journals with very, very reasonable numbers, as you can see on the right.

Next slide.

Once that, and also, you know, basically culminating in one study that I want to just spend another moment to briefly describe our long term follow-up utility and registry study now over 2,300 patients and what we found in those large patient tests was that the PLA really offers simplified actionable information, confirming the test high negative predictive value that we talked about when we started this presentation.

What you see here is that clinicians follow the guidance of the test. We had 0% missed melanomas in the follow up period on PLA negative tests. On a large registry study where the patient, where the where the clinician population consisted of dermatologists, of primary care physicians or physician assistants, and of nurse practitioners. We could see that all these clinicians follow the guidance of the test. Again, a lot of effort has gone into various levels of complexity, but the test readout is very, very simple. If the test is positive, it's either linked or trained or both are detected.

Folks, biopsy and to check the biopsy specimen to give the pathology to establish the diagnosis. If the PLA test is negative, if LINC and PRAME are not detected, clinicians follow again the guidance of the test and do not biopsy, reducing the overall biopsy rate and again, coming back to the patient benefit that we heard about earlier, reducing the biopsy rate by 90%.

Next slide please.

And with that, I think with that we've reached the end of the presentation and we are happy to take questions.

Dr. Gary Oakes:
Wasn't being rude, really, we were chatting with the mothership to make sure everything was going smooth. I would like to ask both of our speakers to quickly speak to your conflicts of interest.

Dr. Burkhard Jansen:
Absolutely. My name is Burkhard Jansen. I'm DermTech's chief medical officer and a dermatologist by background. I'm employed by DermTech, and I'm also a shareholder in DermTech.

Dr. Elai Davicioni:
My name is Elai Davicioni, I'm the chief scientific officer of Decipher Biosciences, employed and a shareholder. Those are my conflicts.

Dr. Gary Oakes:
Thank you. And both the gentleman did complete the formal written conflict of interest. Thank you for that didn't mean to put you on the spot. But I should have done that on the front end and that was my fault. Any other questions or concerns?

OK. Can we go to the next slide? Yeah, let's keep going.

Krista Babbitt:
That's the last one.

Dr. Gary Oakes:
That's the last one. So, when does our comment period end?

Krista Babbitt:
July 15th.

Dr. Gary Oakes:
OK, so there you have it. Is there, let me stop for a moment. Is there anyone on the phone who would like to speak at this time? Anyone on the phone with questions please?

OK, so the comment period begins today, 45 days out, no less, no more. We have to get special written dispensation vow if it's longer. So, in that time frame, we will work with MolDX Doctor Gerard Bein-Willner to make any changes that we deem necessary.

Then at the end of the 45 days, the final drafts or final policies will be promulgated on our web pages. And then 45 days from that, they'll be in order so.

All right. Next open public meetings should be determined. Why? Because we don't know when. And actually, we do know there is one coming up June 18th. That we'll deal with another, but that's technique that's not, never mind.

With that open meetings now we'll be held more ad hoc, you know, in the past we've done it on a pretty cyclical basis.

That's no longer going to be the case when we get to a point where our policy is ready to be presented, we will hold an open meeting.

With that, much like today, they will present evidence that we considered. All of that will be contained in our bibliography and huge chunk of it will be in the policy itself.

We'll start the policy that says coverage is, and then we'll go down and help define why we reach that consideration and that's where the open meetings and still CAC meetings when we hold them will be useful to have those make sure that we're using the right evidence.

The ultimate goal is the best care for the beneficiary. We heard that from both the gentleman today and that's my premise, if I come to work every day, I look out for the beneficiaries. A lot of them can't take care of themselves with the entire role. And then I'm trying to make sure that providers get paid correctly so they have the providers take care of the beneficiaries, right?

That's my two goals in the day and then the third goal is trying to make everybody else happy.

[inaudible]

Krista Babbitt:
Dr. Oakes, I can validate July 15^th.

Dr. Gary Oakes:
July 15, yes questions?

Unknown participant:
What is the time interval between the draft, end of the draft comment period, and then til the final is issued.

Dr. Gary Oakes:
Well, that depends on how many comments we get and to whether or not we have to revise the policy. Sometimes that may take a couple of weeks to go through, sometimes that may take a month to go through. But when the final policy is issued there will be degree, we might need to make changes to the policy.

My recollection of the program integrity manual there is not a specific date [inaudible].

Ahh I couldn't hear the question. So, the question was when the comments end, what is the time frame till the final policy is issued. That depends on how many comments we get and to what degree we need to make changes to the policy. So, but once it hits the final, our notice will be up on the web page and that begins the 45-day notice period.

Paul, anything to add to what we've talked about?

Thank you so much for your input and help.

Greatly appreciate it.

OK, love that. I got closer to the phone.

That will adjourn the meeting. We'll adjourn the recording. You can feel free to speak now.