DR LOVE: Welcome to Oncology Today — The Role of Liquid Biopsy for Cancer Screening, this is medical oncologist Dr Neil Love. For this program, I met with Dr Eric Klein from the Cleveland Clinic Lerner College of Medicine in Cleveland, Ohio, to talk about the use of liquid biopsy in cancer detection. Dr Klein also did a presentation and interview on a similar topic as part of another program focused on the management of colorectal and gastroesophageal cancers which can be accessed at ResearchToPractice.com/ASCOCRCGastric20/Presentations.

To begin this podcast, I asked him to provide an overview of the topic.

DR KLEIN: Tumors shed all sorts of things into the blood stream: DNA, RNA, exosomes. They secrete substances that modify platelets and so forth. And the technology exists now to isolate and measure those at very minute levels. So you can detect the presence of a cancer even when it’s not very big because when the cancer cells divide and die they shed these substances into the blood.

DR LOVE: So can you talk a little bit about the use of liquid biopsy and how it’s been used and what the new thoughts are about how it might be utilized in terms of early detection?

DR KLEIN: Sure. Liquid biopsy has been primarily developed previously for use in patients with known cancer, and there were a couple different approaches. One was to look at the number of actual cancer cells circulating in the blood stream of people with cancer as a measure of tumor burden, and then as a measure of tumor response. So if you have someone who has a metastatic cancer and they have lots of cancer cells shed into the blood stream, that can be measured at a certain level per cc, and you treat the patient. And the number of cancer cells goes down. That’s been shown in multiple different tumor systems to be prognostic for a good outcome.

And then you can also monitor patients who are being treated for cancer for that same signal, whether it’s circulating tumor cells or whether it’s cell-free DNA, which comes from the tumor cells. As they turn over you can measure the levels in the system and determine whether or not they’re responding to cancer, and if they have responded, and their numbers went down, or the amount of cell-free DNA went down. And if it came up again, that suggests a tumor recurrence.

But even more than that, and maybe even more exciting, is that you can actually sequence the DNA that’s in the system, in the blood, and figure out if someone with a cancer has a mutation that can be targeted with a specific agent. And I’ll tell you a story about a patient who was treated here at the Cleveland Clinic in that regard in a minute, and he’s gone public with this story, so I can tell this — so a targeted agent. And then we know that when we use targeted agents that the dominant clone, the one that has the mutation that’s being targeted, goes away, but then there are resistant clones that grow up. And so then you could use that as a trigger for switching agents.

So here’s a great story, it probably goes back a decade now. A young medical student, who was I think 22 or 23 at the time, who developed a cough one summer, and being a youngster and feeling invulnerable, ignored it for a while. Eventually, in the fall, went to see his primary care physician, had a chest x-ray, at age 23, nonsmoker, that showed widely metastatic adenocarcinoma of the lung. And he had huge disease burden, and he had I think really significant disease in his lungs and bilateral pleural effusions and so forth, that landed him in our ICU relatively quickly.

And this was just at the dawn of the targeted agent era and next-generation sequencing of tumors, again, a decade ago. And so his tumor was biopsied and sequenced, and it turned out he had a mutation in ALK, the ALK mutation, that’s typical of some leukemias and — and lung cancer. And the FDA had just approved an agent, and I don’t recall which one it was, that targets that mutation in leukemia. And so we got permission to try it out on a compassionate-use basis on this young man. And they started it, and within, I don’t know, 10 days, 2 weeks, something like that, he was off the ventilator, and a lot of his tumor regressed.

And a couple months later he walked out of the hospital, and he quit medical school, and he became a cancer researcher and joined our Lerner Research Institute. And it’s a remarkable story about how sequencing can really impact cancer.

And so that was tumor sequencing.

We’ve now gotten to the point where we can measure the same signal without having to biopsy the tumor, and we can do it in blood or urine or sputum or that sort of thing.

So one thing I didn’t emphasize was the fact that you can use a liquid biopsy on a serial basis. And so it’s hard to rebiopsy tumors. It’s not as much fun for the patients. It might require anesthesia, getting a CT scan or radiation exposure, that sort of thing. Some tumors aren’t going to be accessible, but everything is accessible from the blood because we believe that most or all tumors shed these substances in there. So one of the other advantages of liquid biopsy in cancer management, not so much in screening, is the ability to easily monitor things on a serial basis.

DR LOVE: So this man is still alive and well?

DR KLEIN: As far as I know, yes. He told this story on a website, and the website went quiet about 2 or 3 years ago, so I’m not sure what’s happened to him since. But he had a prolonged survival with a very productive many years after all of this.

DR LOVE: So you think about it, fast forward, I’ve heard dozens of cases like this, where you’re going to get a liquid biopsy, of course you’re going to send tissue, but they get the answer, I think, like 3 days, a week. And then when the patient recurs they do it again, they get the answer. It’s amazing, it’s literally become part of practice in terms of oncology practice.

DR KLEIN: It has. Yes. Right now it’s the minority of patients who benefit from that approach, but I’m an optimist about it. I think the door continues to open wider and wider and wider.

DR LOVE: So let’s talk about the focus of this work, which is cancer screening and early diagnosis. Maybe you can kind of provide an overview of where we are in general with screening and also as part of that, just preceding what you’re going to discuss in terms of the data with these liquid assays, what kind of false-positive rates we’re getting with our current screening?

DR KLEIN: Yeah, so there are 5 standardized, US Preventive Services Task Force recommended at various levels, screening paradigms. So Pap smears for cervical cancer, mammography for breast cancer, CT scans for patients at high risk of lung cancer, colonoscopy for colon cancer, and PSA for prostate cancer. False positives are prevalent for some of them, not all of them. Probably the poster child there is PSA, which is prostate specific but not prostate cancer specific. And there are a number of new markers in that space that are reducing the false positivity there. That’s not really the main one to talk about, I think.

And those screening paradigms work well. I mean they’ve all be shown to reduce mortality, even PSA. They’re not perfect though. Again, we do too many prostate biopsies for PSAs that are falsely positive. Mammography misses some high-grade cancers. Colonoscopy is not user friendly. Mammography is probably not the most pleasant thing, and so forth. And there’s opportunity to improve on all of those.

One of the interesting data was that even for those 4 cancers, excluding prostate cancer, which screening we know works, most of those cancers are still diagnosed at advanced stages, even when people are screened.

DR LOVE: What fraction of cancers overall, fraction of cancer deaths, it seems like these are the common cancers, but in reality what part of the need are we actually even addressing right now?

DR KLEIN: Yeah, so most cancers actually, for which we don’t have established screening paradigms that account for lots of the cancer deaths. So for example, if you look at the CCGA study and the 12 cancers where it performs very well, that’s ovarian, pancreas, liver, plasma cell, bladder, and a couple others, they — those cancers, again for which there’s no screening paradigm, account for two thirds of cancer mortality in the United States.

And so there’s opportunity there, even if we can only detect a subset of those cancers at an earlier stage, we should start to be able to make a dent on mortality. And part of the exciting thing to me is that these are first-generation assays. And as we use them more and wider and wider, we’ll get better understanding what they mean. The technology will continue to get better, and ultimately I believe that we will detect even more early-stage cancers than we do now, or potentially than we do now I guess.

DR LOVE: So when you think about trying to design screening intervention, what are some of the basic qualities you’re looking for? What do you want to achieve, and how do you want to get there?

DR KLEIN: You need a test that has high sensitivity to detect potentially lethal cancers. You want to avoid the PSA conundrum, which over-detects low-grade cancers that don’t need to be treated. And there is some emerging data that suggests that these sorts of tests do detect primarily, although not exclusively, biologically significant cancers. So that’s one thing.

The second thing that’s really important in the screening population is avoiding false positives. So we cannot have a simple blood test that might be used widely that tells a lot of people who don’t have cancer that in fact they have cancer. And all of the trials have very, very high specificity, meaning the false-positive rate can be set with these algorithms to less than 1%, which is probably an acceptable level for the general population, and so we can avoid that.

You want it to be easy. I mean what could be easier than getting a blood test, right? We all do that for various things, far easier than our other standard screening paradigms; colonoscopy, mammography, CT scan, and that sort of thing. It has to have pretty rapid turnaround, doesn’t have to be overnight, but within a meaningful period of time. So you need a high throughput technology. And it needs to be done in a way that is meaningful to patients and doesn’t cause them psychologic or psychic harm.

DR LOVE: I guess the other thing too that is maybe a factor in this whole thing is how useful is it to make an early diagnosis? Does it follow a typical solid tumor paradigm or more myeloma, hem-type paradigm? Any thoughts about that? I mean that kind of goes back to some pretty basic principles, Halstedian principles. Do you think that really applies? And which cancers do you think it applies the most to? It seems like colon, I mean you cure.

DR KLEIN: Yeah.

DR LOVE: Not necessarily with the others.

DR KLEIN: Well I think the cancer data are pretty clear. Survival rates are best for earlier-stage tumors. I mean that’s pretty true across all malignancies. Having said that, I think we’re going to learn a lot here. So one question is, although the answer might be obvious, is a Stage I pancreatic cancer that’s detected by a blood test and worked up with a CT scan, does that behave the same way that a Stage I pancreatic cancer that was picked up for unrelated reasons behave? and how do we treat that and so forth?

I think there’s going to be a learning opportunity here to figure that out. We don’t see many Stage I pancreatic cancers, so I’m not sure we know the best way to treat that. Maybe it’s not surgery. Maybe it’s something else. Maybe it’s something less morbid.

So this is going to open up a whole new field in all the malignancies, to answer your question. But it just seems obvious, from having managed cancer for many decades now, that earlier-stage tumors are easier to cure. It’s not only the cure rate’s higher, but the burden of cure is less, right? You need fewer modalities to cure them.

DR LOVE: So this really is an exciting thought and idea. Can you talk about what we know and what some of the major initiatives have been up to this point looking at it?

DR KLEIN: Yeah, so the published studies have adopted a very similar strategy, which was discovery strategy followed by a validation strategy. So the discovery strategy was let’s take lots of patients who have known cancers of various stages and see if the technology using a blood test can detect those cancers. And the answer was that’s doable.

And then the next question was can it detect early-stage cancers. And the answer was yes, it can do that. And then the last idea was can the signal predict in a blinded fashion what organ system is involved? And the answer to that is yes also. And so then all of the studies have then taken validation cohorts, people with known cancers or without cancer, as negative controls, and replicated the findings there.

And the algorithms rely a great deal on artificial learning, or artificial intelligence and machine learning. These are huge data sets that are put in there. We’re talking about tens of thousands of patients in some studies. And iterated over and over and over again to refine the algorithms to get great predictive value. So it’s a nice marriage, I think, of next-generation sequencing, which is exciting, microfluidics, and droplet-based RT PCR that allows you to isolate a single DNA molecule and get a signal if you’re looking for a mutation or a methylation pattern and that sort of thing. And then artificial intelligence approaches to allow you to interpret the data meaningfully.

DR LOVE: So let’s talk about the specific initiatives that kind of moved the field forward, beginning with the CancerSEEK project.

DR KLEIN: Yeah. So CancerSEEK is a test that is based on looking at methylation of cell-free DNA and protein expression, protein amino assays, that are typical for a limited set of mostly GI cancers. And that was a population of about a thousand individuals with Stage I to III cancer, many of whom, or most of whom, had symptoms. And it was a nice discovery study that showed that looking at these methylation and protein immunoassay expression levels that you can detect cancer in patients who are known to have cancer.

And you can see a signal that is typical for each individual cancer, so if you have a positive signal you can predict what organ system it’s from. And it’s pretty good sensitivity and very high specificity, again, very low false-positive rate. And all the studies have followed a similar pattern.

DR LOVE: So I know they looked at about a thousand patients with cancer, Stage I to III. What did they see in terms of stage of the tumors that they picked up? Or that they confirmed?

DR KLEIN: Yeah, so the sensitivity is best for the more advanced cancer, so it was highest for Stage III cancer. But if I recall correctly, around 40% or 45% of Stage I cancers were detected, and about 75% of Stage II cancers were detected. And again, these are cancers that often have no screening paradigm, where patients are generally not diagnosed until they’re symptomatic. So that’s usually Stage II, III, or IV.

DR LOVE: So is this initiative moving forward? What are they doing now?

DR KLEIN: So the company that has published the cancer seq data is proceeding with commercial development of a test that is intended to be used as a screening test in the population.

Now we have these 3 studies that have shown that you have a blood-based assay that can detect early-stage cancer, and if you have the signal that can predict what organ system it’s in, those were discovery and validation studies primarily, though not exclusively, in patients who had cancer.

So let’s think forward of how that test might be used. You have to go to the general population who may have an average risk of having cancer or who may have an elevated risk of having cancer by virtue of the fact that they have an inherited genetic syndrome, like BRCA or Lynch syndrome, or maybe they were a smoker, or perhaps they had an invasive cancer that was treated and cured many years ago. So those are 2 separate populations, but you can think about using this technology that can detect cancer when it’s present and predict the origin of testing these in what we would call the intended use population.

So that’s the next step for all of these tests, which is now that there’s validated algorithm and assay, let’s go to the population that we intend to use this on, the general population, who may or may not have risk factors, and see how it works. And the prevalence of cancer in the general healthy population’s pretty low. It’s only around I think 1.3%, 1.5%, something like that. So you have to do relatively large studies to determine how these tests perform in that population. And that is what’s happening now with all of these tests. That’s the next step.

DR LOVE: It seems like most or all of these are targeting methylation. I don’t know much about that. I think about methylation I think about epigenetic therapy. We’ve heard a little bit about it.

DR KLEIN: Yes.

DR LOVE: What is the process of methylation, and why has it been chosen to focus on?

DR KLEIN: So methylation is a fundamental biologic process that regulates gene expression. So genes that are methylated generally are turned off. And which specific genes are methylated and the pattern amongst different pathways is different for each cancer. So you can measure whole-genome methylation, for example, at a high level in prostate cancer versus lung cancer versus breast cancer, and they will all be different. And it is possible, using sequencing technology, to distinguish between those 3.

So methylation is when there’s a methyl group applied to certain DNA bases that turns things off. Again, it’s very specific, very characteristic, I should say, for cancer. And the pattern of methylation, which genes are methylated, is very characteristic for each kind of cancer. So that is one way to measure a cancer signal in the blood. There are other ways, and it depends on what you’re looking for. So for example, lung cancer’s a good example. If you’re looking for the presence of cancer cells in the blood or cancer-related DNA — cell-free DNA in the blood in someone with lung cancer, one common mutation is in the epidermal growth factor receptor (EGFR). And so you can measure that using a targeted sequencing assay. You can see if there’s mutated EGFR in the blood for someone with a known lung cancer.

If you look at the screening paradigm, no one had ever looked in the blood before in patients as a screen, and so one of the studies the CCGA study, actually started as a head-to-head comparison of looking at 3 different targets to see which performed best. And the 3 targets were methylation, whole-genome sequencing, and targeted mutation assays. And for the screening population it turned out that the methylation assay was the best choice, and it was multifactorial. Not only did it have slightly better sensitivity and specificity than the others, it’s, I think, a little easier to run analytically, and it’s cheaper, for example. It’s cheaper to look at methylation than it is to do whole-genome sequencing in every sample that you get.

So there are a lot of considerations that go into what’s best. But methylation does seem like a very robust signal to be able to be used in the screening population.

DR LOVE: So we talked about the CancerSEEK project that looked at people with cancer, you also talked about the PanSEER project.

What did they do?

DR KLEIN: PanSEER was a study from China which took about 600 individuals who were part of a longitudinal cohort study, who had donated blood that was in the bank. I mean it was in a freezer. And this technology came along, and they had longitudinal follow up on these 600 people. So they knew who got cancer and who didn’t get cancer over the next 4 years. And so they used a methylation assay also on the stored blood, and they were able to show that methylation provides a signal for the patients who are ultimately diagnosed with cancer. And what was really remarkable about it was that it had a 4-year lead time.

The test was able to detect the presence of cancer 4 years before individuals were diagnosed, which is pretty remarkable in terms of the sensitivity. There’s a downside here though. We’re going to be creating a new diagnosis, healthy patient, has a blood test that detects cancer, may have the appropriate diagnostic workup, CT scan, that sort of thing, and no cancer is found.

And we have no idea what to do with that yet. We don’t know if those are false positives, and there will be some of those. We don’t know if the tests are so sensitive that they pick up microscopic cancers before they become clinically evident, and how do we deal with the psychic injury and worry that that causes patients, and how do we follow those patients and so forth? And that’s something that’s a conundrum that this field has created, and we’ll have to figure that out going forward.

DR LOVE: So the last study you talked about in your talk was the CCGA study. Can you talk about that initiative?

DR KLEIN: Yeah. So that stands for circulating cell genome atlas, and this was a discovery and validation study. So for example, the CancerSEEK study as published was really a discovery study, although using that platform there are validation studies published also. So CCGA took 15,000 individuals, 70% of them had known cancer, so they were the discovery set. 30% were not known to have cancer, so they were the negative control set is the way to think about it, and compared the 3 assays that I mentioned, targeted mutation assays, whole-genome sequencing, and methylation, and found that a methylation-based signal could predict the presence of more than 50 cancers, that’s really remarkable, and had about 43% sensitivity for all cancers across all stages, and about 67%, I think it was 67% sensitivity for 12 cancers that together account for two thirds of cancer-related deaths.

All of the studies I talked about, including CCGA, had about a 40% sensitivity for Stage I cancers and 60% to 75% sensitivity for Stage II cancers. Those are the ones you want to target for screening. And that was iterated across other cohorts of the patient subsets of these 15,000 patients, or participants, in order to validate the algorithm. And like the other studies, CCGA had a very high ability to predict not only the presence of cancer, but when the signal was present, what organ system was involved. That’s pretty interesting, actually.

So if you think about 1 potential use for this sort of test, as a primary care physician you might see a patient who’s got kind of some vague symptoms, isn’t feeling well, may not have really significant physical findings, may or may not have abnormalities on a blood test and so forth, and you scratch your head and say, “Gee, I don’t know. Is this a chronic viral infection? Is this an autoimmune disease? Could this be cancer?” Imagine, just imagine if we had a blood test that could rule in or out the presence of cancer within a week or so of drawing the blood. You could really focus your workup after that.

So if you were to see a patient like that and have a blood test that suggested the presence of cancer, and not only that it suggested it’s kidney cancer, then you go right to an ultrasound or a CT scan, and you make that diagnosis pretty quickly.

On the other hand, and this is still to be worked out whether this is possible, you might be able to rule out the presence of cancer pretty quickly in a patient who’s not feeling well and reassure the patient that it’s not cancer, or at least at present they don’t have cancer, and do a different workup. And so one of the things that will be looked at in forthcoming studies will be how these tests serve diagnostic efficiency in patients who have kind of vague symptoms and so forth. I mean there are a lot of exciting possibilities here.

DR LOVE: So it’s interesting, when I look at the list of cancers that they’ve been able to detect, I see lymphoma, plasma cell neoplasm, but I don’t see any leukemias and other hematologic cancers. And we haven’t really traditionally thought about early screening or detection of those, what’s the reason you can’t pick it up early?

DR KLEIN: Yeah, I’m not sure. You might be able to. Those were not the target of the study. The focus of the study was mostly on solid tumors.

DR LOVE: Yeah, that makes sense of course.

DR KLEIN: Yes.

DR LOVE: So anything else you want to say about the data? And also where things are heading, what the next steps are.

DR KLEIN: Yeah, what’s interesting to me is in the 3 studies is that the findings are all remarkably consistent on how well a liquid biopsy, based on next-generation sequencing primarily, although not exclusively, can detect early-stage cancers. And so I think as a field we need to move this forward in the “healthy population” to see how it performs, to be sure that we can keep the false-positive rate at a very low, acceptable level, and to understand what the meaning of these tests are.

So I want to be clear about one thing. Some of these tests are likely to be available on the market sometime next year.

DR LOVE: Wow.

DR KLEIN: I think, and those will be, at least initially those will be offered as a lab-developed test, so out-of-pocket expense not covered by insurance. I think most of the companies are moving toward FDA approval, as well, so who knows how long that’ll take. But these tests are going to be out on the market.

It’s critical for people to understand that these tests are not meant as a replacement for the established screening tests that we have, the 5 that I talked about earlier. At present, they are an adjunct.

I think one interesting question that I’m particularly intrigued by is it possible that 5 or 10 years from now this kind of blood test could replace colonoscopy or mammography or PSA. Having said that, none of these tests really perform very well in prostate cancer, I will tell you that.

But in any event, the question is whether a blood test like this can replace or establish screening paradigms. I mean wouldn’t that be wonderful, having had a recent colonoscopy, that you don’t have to prep in order to be screened for colon cancer? You’d have a blood test instead, that sort of thing. I don’t know if that’s going to be possible or not. It certainly is not possible today. So again, to re-emphasize, in the short-term these are intended to be an adjunct not a replacement for standard screening tests.

And here’s another question that I think is very interesting that we don’t have an answer to yet, which is if you’re healthy and no risk factors, or even if you have risk factors, and you have a negative test, how long is that good for? If I have a negative test now, a negative blood test, does that mean I’m currently cancer free and will be cancer free for the next year, 2 years, 3 years, 5 years, that sort of thing? We don’t know that.

So one of the studies, the CCGA study that enrolled these 15,000 participants, is going to follow all of them for 5 years. So in the subset of patients, the 30% who did not have cancer, we will know, at least with a 5-year predictive ability, if you had a negative blood test initially, and you’re cancer free in 5 years, we’ll be able to say 80% of the people who had a negative blood test didn’t develop cancer over 5 years. Or we will say if you have no risk factors maybe it’s a good predictor that you’ll be cancer free 100% in 5 years, but if you have risk factors like BRCA syndrome, maybe it says your risk goes up at year 3, better get screened with mammography or an abdominal ultrasound or CT at year 3. I mean there’s a lot to be learned still.

DR LOVE: Yeah, I guess it’s going to be interesting — start thinking about the issues of cost effectiveness —

DR KLEIN: Yes.

DR LOVE: — in terms of potentially — it could be a huge cost savings, theoretically.

DR KLEIN: So it could be, at various levels. So one thing to think about is again, this is pie in the sky, this isn’t real, but maybe we eliminate the need for screening mammography. Maybe we can shut down screening mammography, and you only use mammography to get anatomic detail on known tumors. And so we can stop investing in, potentially, in the machinery, in the personnel who do that and repurpose them for doing other things. So that’s 1 potential cost saving. It’s probably cheaper to do a blood test than it is a mammogram, although probably not initially. The price of these eventually, with so many entities in the market the price is going to go down. But maybe it’ll be cheaper than established screening tests. And even if it’s not cheaper it’s better.

Second thing is it is cheaper to cure earlier-stage tumors than later-stage tumors because you don’t need as much therapy, right? You might be able to cure an early-stage tumor with radiation alone or surgery alone and later-stage tumors often take both or chemotherapy on top of it, that sort of thing.

And then the third thing is diagnostic efficiency, which I talked about a little bit earlier. So if you have someone who you think might have cancer, and you get to the diagnosis in a week doing 1 CT scan because the blood test predicted they had a kidney cancer, instead of doing a PET scan or something else, or lots of other blood tests, you might save money that way.

And then the last thing is return to work and return to normal life and so forth. If you can have a test that allows you to rule in the presence of cancer quickly, figure out what cancer it is quickly, treat it when it’s early stage, patients will recover faster and get back to work quicker. And so if you look at it from that perspective there’s a potential cost saving there too.

Now all of those details haven’t been worked out because we don’t really know what the performance characteristics of these tests are in the intended use population, the general population, or those who have elevated risk of cancer. We need to work all that out.

DR LOVE: Fascinating. Anything else you want to add to what we’ve talked about today? I was going to ask you, you mentioned prostate cancer. Is the issue that it can’t pick it up diagnostically or maybe PSA is so good that it’s not going to have an impact? Or what’s the reason?

DR KLEIN: I can only speak from my experience in participating in CCGA. We entered more than a thousand participants from the Cleveland Clinic in that study, and we did look at prostate cancer, and it just didn’t perform well for early-stage prostate cancer. So it just does not seem like it’s going to be a good screening test for early-stage prostate cancer, at least based on that initial experience. And there are new markers on the market, just speaking of prostate cancer screening.

So PSA has a high false-positive rate because it’s prostate specific not prostate cancer specific. But there are a number of new markers that are based on measuring different isoforms of PSA. There are microRNAs. There are methylation sequences in urine, that sort of thing, that have improved sensitivity and specificity over PSA. So the screening paradigm for prostate cancer, I think is moving from recommending a biopsy for every man who has a PSA above 3 or 4 to doing a reflex test, a blood test or urine test or, in some countries, an MRI of the prostate before deciding on a biopsy. And then MRI has allowed us to do more targeted biopsies to allow us to come to the correct diagnosis quicker, reduces the sampling error in the prostate.

DR LOVE: I don’t know if it’s technologically possible, but can you fantasize or envision a day where you would do self-testing? We do so many analyses right now, let people just go out and do it.

DR KLEIN: So that already exists. I think you can have Cologuard^® sent to your house.

DR LOVE: Right.

DR KLEIN: And I know that lots of entities are working on self COVID tests, for example, either saliva based or nasal swab based, so absolutely you could do that. I mean if you get someone to draw your blood. Or if it’s a urine or sputum sample, you certainly can collect those at home and send them into a lab.