MS. DIANE REHMThanks for joining us. I'm Diane Rehm. Blood tests already offer a number of valuable insights into a person's health, and the list is expanding. Researchers are now using blood tests to detect cancer, other kinds of illnesses, and prenatal genetic abnormalities. Joining me in the studio to talk about new frontiers for blood tests, Dr. Luis Diaz of Johns Hopkins University.

MS. DIANE REHMJoining us by phone from Palo Alto, Dr. Ash Alizadeh, assistant professor of medicine at Stanford School of Medicine. And from WXXI, in Rochester, N.Y., Mark Mapstone, senior neuropsychologist at the University of Rochester Medical Center. Do join us, questions, comments, 800-433-8850. Send us an email to drshow@wamu.org, follow us on Facebook or Twitter. And welcome to all of you.

DR. LUIS DIAZThanks, Diane. I'm so happy to be here.

MR. MARK MAPSTONEGood morning.

REHMThank you all. Dr. Diaz, this seems like an absolutely fascinating development. Explain to us how blood samples can now be used to detect cancer in the body.

DIAZWell, there's really be evolution in what type of blood samples we use for detecting cancer. We've all heard of PSA for prostate cancer.

REHMOf course.

DIAZThat's a protein released by prostate cells. So when the levels go up, we suspect there's prostate cancer. Since the early '80s, '70s, there's been an evolution in the understanding of cancer that it is a genetic disease. It is a disease caused by mutations. And all cells, when they divide or when they die, release small fragments of DNA into the circulation. If those cells happen to be cancer cells, we now have the technology to look in the blood for free-floating fragments of DNA that are mutated that could herald a cancer in the body.

REHMSo you're saying, talking about prostate cancer, for example, that that test, that PSA test, is different from what you're talking about here, in terms of a prick in the finger for blood or taking the blood from the arm.

DIAZAbsolutely. The PSA is a protein. This new test is DNA, the actual origin or the foundation of that tumor. What makes that tumor tick is released into the circulation. So, very specifically, you can precisely find the mutations that are in that tumor in the circulation. And you can use that for a variety of different things. But the most exciting and what might have the most impact is early detection, catching cancer at its earliest stages.

REHMAll right. And to you, Dr. Alizadeh, you've been doing work on blood tests for lung cancer among patients, I gather, with no symptoms at all.

DR. ASH ALIZADEHThat's true to -- but most of our work is focused on patients with more advanced stages of disease. And, full disclosure, that's, as Luis pointed out, the thing we would really like to do though is to push the technology to those with absolutely no symptoms. The majority of patients that have their disease resectable show up with absolutely no symptoms. So that's the population that we really want to address. And I can say that we've been able to detect disease in such patients, though it's early work. So I don't want to overstate, but that's where most of our effort's focused now.

REHMWell, even speculating, though, Dr. Alizadeh, how different would a blood sample that carried the DNA of lung cancer look from the DNA of breast cancer, say?

ALIZADEHSure. Well, you know, I think you're highlighting a pretty important aspect of what we're trying to do here. We all have genetic material that distinguishes us from each other, as individuals with thousands of differences, tens of thousands, hundreds of thousands of differences between individuals, depending on how related they are. But cancers also are distinguished from each other in the set of mutations that they have.

ALIZADEHSome cancers, believe it or not, are more similar to each other, even if they are farther apart in the body in where they arise. Gynecological cancers are an example. The key is in identifying a core set of mutations that can confidently detect a cancer and distinguish it. To be able to do that, to classify given cancer type, though, I think is something that needs to be developed further.

REHMStill a long way off then?

ALIZADEHWell, I think it depends on the type of cancer. There are certain genetic alterations that highlight a particular cancer, certain alterations that are what we call pathognomonic or sort of the litmus test, the acid test for a given cancer. And we use them for clinical diagnosis of certain cancers, certain genetic alterations define certain cancers.

ALIZADEHThose are few and far in between. And the key then becomes, can we leverage what we are learning from the exercise of looking at many, many genomes from tumors from patients with cancer to find those sets of mutations that are shared between groups of patients? And that's the exercise we've done so far for lung cancer.

REHMAll right. Now, turning back to you, Dr. Diaz, when we think about cancer, I have heard it said that each one of us has cancer in our body, some which may not ever mutate, some which are simply going to be there, perhaps even at autopsy. How would you distinguish -- if a blood test indicates that the presence of cancer, how would you distinguish from those you need to treat from those you don't need to treat?

DIAZYou bring up some controversial points. One of them is if we develop this test that can detect cancer at its earliest stages. Is detecting at that stage going to change the eventual outcome?

REHMRight.

DIAZWill that tumor we detect grow up to be a lethal cancer, or will it be a cancer that remains indolent and never kills a patient? So with each new technology, with each new question answered, with each new problem solved, you have a new problem, right.

REHMNew question.

DIAZA new question. So that is the question that is the follow on to early detection of cancer. There's a lot of criticism and technologies that might detect early detection because they say they may not influence survival, that's not the problem. The problem is you detect the cancer early. Now let's figure out the next question: Which cancers are lethal and which cancers can you treat less aggressively?

REHMAll right. And turning to you, Mark Mapstone, I know you've been doing research on blood tests to predict the onset of Alzheimer's. Tell us about that work.

MAPSTONESure. One of the biggest problems we face with Alzheimer's today is really that the treatments that we have available to us are not given to the patients in a timely way. That is we're really not getting the therapies to patients until they present with clinical symptoms. That is as soon as they start having memory loss.

MAPSTONEAnd so there's a real thrust in the field now to start looking at something called pre-clinical disease, which is trying to detect the disease process before the symptoms actually manifest. And so what we're trying to do is detect when the disease is sort of burning along before the symptoms start to begin. So our work…

REHMWhat is it -- what would it be in the blood then that might indicate that the onset of Alzheimer's is, say, three to five years away?

MAPSTONEWell, from our work, we look specifically at lipids or metabolites in the blood. So this is a little bit different from Luis and Ash's work where they're looking at DNA or the genetic components. We're looking at metabolites, so those downstream processes and small molecules that are the result of physiological processes.

MAPSTONEAnd what we found was that there are lipids or fats that represent cell membranes and constituents of cell membranes that are altered in the early disease stages so that when the disease is sort of simmering along, before symptoms are manifest, we can find evidence of altered metabolism of these lipids in the blood.

REHMBut, now, the same kind of question arises, doesn't it? That is, perhaps many people would have those kinds of lipids present, but does that necessarily point to the development in later life of Alzheimer's?

MAPSTONEAbsolutely. And that's a critical question, and it's a critical next step that needs to be taken. Our work, and that we've recently published, really presents a biomarker of disease where there really is no test yet, so there's -- you can't go to your doctor and order these particular tests. However, this work does need to be validated, and we hope that it can be moved out to general testing situations. But right now it's not there yet. So you're absolutely right. We need to make sure that this is specific to Alzheimer's disease. And we don't have that information yet.

REHMAnd final question before we take a short break. There aren't any really good treatments for Alzheimer's available right now. So what about the ethical issue of informing someone that these lipids are present?

MAPSTONEWell, I think you hit the nail on the head. And, I mean, that's really one of the big questions we have to deal with now. Without any available treatments that are really disease modifying in nature, that is, cures or disease modifying therapies, the use of this information is somewhat limited. And so that's why we're not advocating, yet, that this is some sort of a general test. However, we do hope to get to that next step. And we hope that this will get us to develop new treatments.

REHMAll right. Mark Mapstone, he's at the University of Rochester Medical Center. Short break. We'll be right back.

REHMAnd as we talk about the use of blood tests to determine the presence or the indication of certain diseases that could be existing in the bloodstream by virtue of identifying DNA or certain lipids, we turn now to Dr. Nichole Pardo. She's joining us from her office here in Washington, D.C. I know, Dr. Pardo, that pregnant women can now get a blood test to test for genetic abnormalities in the fetus. Explain how this works.

DR. NICHOLE PARDOYeah. So there is a new test available to mothers now that screens for the three most common chromosomal abnormalities that can exist in a fetus. Something most people are familiar with is Down syndrome, and then there are other chromosomal abnormalities where there's an extra copy of a particular chromosome called Trisomy 13 and Trisomy 18.

REHMAnd I gather you can also test for Fragile X.

PARDOWell, the Fragile X test is a bit different than the noninvasive prenatal testing we're doing for aneuploidy. It's a little bit of a different type of test. When you speak of Fragile X, typically you can screen the parents for precursors or genetic abnormalities in the parents and see if they are at risk of passing that along to their offspring.

REHMI thought there was also a blood test to indicate whether Fragile X was present in the mother carrying the child.

PARDOThere is. That test is to indicate if the mother is carrying the Fragile X that can go to the child. Exactly.

REHMAll right. OK.

PARDOBut that is a bit different than the noninvasive tests we're doing now for the aneuploidy.

REHMAll right. And what are the benefits that you would say these tests provide as compared to say amniocentesis?

PARDOThey are considered -- they're noninvasive. They are done by taking a sample of the mother's blood as early as 10 weeks in her pregnancy. There is no risk of the -- there's not a needle to the uterus or introduced to the fetus itself. So it's not invasive, therefore it doesn't pose risks of miscarriage to the fetus.

REHMAnd what types of patients are getting these tests?

PARDOIt's recommended -- we're testing or offering testing to patients who are considered higher risk for these conditions. So they're women who will be 35 or older at the time of their delivery. There are women who have a history of chromosomal abnormalities in prior pregnancies and people who are younger but they have an abnormal screening test, which is more of a standard of care that we've used up to this point.

REHMAnd does insurance cover this test?

PARDOWell, that's an interesting question, and it is variable across the board. Some insurers are covering a portion or all of it for their patients who are 35 and older, and some insurers are covering just a portion. So it is a question that patients ask, and I typically have them check with their insurance companies just to be sure to see how much of this test would be covered.

REHMI see. And, Dr. Pardo, do you see any downsides to the test?

PARDOWell, this is a screening test. It is not an absolute test. It doesn't -- it is very accurate, better than what we've had in the past. But it still has a risk of missing a few patients whose fetuses might be affected by these conditions. It's not comprehensive for all possible genetic disorders that can be in a newborn when the child is born. So I think that it requires counseling of the patient to -- for the patient to understand exactly what we're screening for. But it certainly is an improvement over doing a lot of unnecessary invasive testing, which we have done up to this point.

REHMDr. Nichole Pardo, she's a practicing physicians with Foxhall OB/GYN Associates. Thanks so much for joining us.

PARDOYou're welcome. Thank you for having me.

REHMIndeed. Dr. Diaz, do you see any downsides to that kind of testing?

DIAZSo I would think more information is better. And this technology's being tested in so many different fields in medicine. We're looking at DNA in cancer. We're looking at DNA in field medicine. We're looking at DNA in other conditions like myocardial infarction, stroke, even concussions down the road. Small fragments of DNA or RNA that may herald injury earlier than we know is going on or herald cancer earlier than we know is going on.

DIAZThe question is what we do with that information. And that's the challenge to the medical research community. It's sometimes tempting to bring this technology into the clinic very quickly. But, as you mentioned, it brings up questions that may pose ethical concerns. So it's a balance, how fast you bring this new novel technology to the patient, and you balance that with what answers can be obtained for the patient with that technology.

REHMBut, Mark Mapstone, can you talk about signs of concussion in the blood?

MAPSTONESure, sure. So there's been some recent work looking at blood biomarkers of concussion. And obviously this is an issue that's been brought to the forefront with sports related head injuries and also with military concussions as well, soldiers returning from the theater. So this is an important issue that people are starting to look heavily into. And blood, for its various reasons, makes a good fluid to interrogate for this.

MAPSTONEAnd there's a couple of recent studies out looking at particular biomarkers of concussion. And interestingly there's some overlap between the things that are being found in blood shortly after a head injury with some of the things that are investigated or traditionally thought of being involved in Alzheimer's disease, and in particular things like a tau protein, phosphorylated tau, and in addition to other markers in concussion as well. So there's a lot of work going on using the blood as a means to interrogate what's going on within the central nervous system or the CNS or the brain.

REHMAnd so if you found, for example, evidence of a concussion in the bloodstream, would that mean that the brain is still recovering or further treatment is called for? What would it indicate?

MAPSTONEWell, that's a good question, and it's one of the ones that we're left with when we start investigating in the blood for things that are going on in the CNS, is exactly how those changes are related to each other. And so I don't think that the dots have been completely connected between what's going on in the CNS and in the blood, in both concussion model and then also Alzheimer's and other sorts of neurological diseases.

REHMWould that indicate, Dr. Alizadeh, that the ability to search out these particular problems is ahead of the development of medications to treat them?

ALIZADEHI think that's a great question. I think, as Luis pointed out, collecting information and having more information is almost always better. I think the accuracy of that information is a very important part of trying to figure out how to use the information. And we've been hampered by types of tests that we've used so far for reaching decisions that we make in modern medicine based on tests that have pretty limited precision in terms of their accuracy.

ALIZADEHAnd the types of technologies we're talking about here are really potentially transformative because they raise the bar to a significantly higher level by telling us more precisely whether what we're detecting is meaningful or not. Now what to do about it, I think, is a new question. I think you're right in that sense, but having more precise information, I think, is almost always a cornerstone of what we end up acting on.

REHMDr. Diaz, so if you were performing this type of blood test and it indicated that the patient could be vulnerable to Alzheimer's, what would you say to them? What would you -- how would you transmit that information, or would you, knowing that there is no good treatment out there yet?

DIAZWell, that's where the question of when do you bring these tests into the clinical arena comes into play. When do you leave them as research and to answer research questions, and when do you bring that novel technology into the clinic so that the patient and the physician can have some conversations? That's a complex question.

DIAZAnd each disease state has to be looked at individually. In the case of Alzheimer's, I'm an oncologist, so I have very little information there. But there's similar situations in cancer. And more information's better, but there's more information that's limited still where we don't know the impact of that biomarker could be dangerous and harmful.

REHMAll right. So, Mark Mapstone, same question to you. What do you do if you have tested an individual and it turns out that those lipids are indeed present indicating either the presence or the soon-to-develop Alzheimer's? What do you do?

MAPSTONEWell, it's important to be clear that this is not a test that's available right now. And there needs to be a lot more work that needs to be done in order to get this out into the scenario that you're describing. Given that it's out there, if we do get it out as a particular test for use, I think that there is some important things that can be done with the information. And one is future planning. I think most patients do want to know what's going to happen within a reasonable certainty. But, until we get a higher degree of specificity of this test, I mean, that's going to be limited right now. Three are things...

REHMAm I incorrect in saying that Georgetown University is using this test?

MAPSTONEYou -- no, they are not, not a clinical setting. So we're going to be using it hopefully soon into developing new drugs. And so what we can do with this test is we can enroll patients or subjects into clinical trials...

REHMI see. I see.

MAPSTONE...where we know that these people are going with fairly high certainty to get the disease. And then we can test the drugs in those patients to see if the drugs have disease modifying effects.

REHMAll right.

MAPSTONESo we're not using it in a clinical setting yet.

REHMAnd you're listening to "The Diane Rehm Show." Going to open the phones now, 800-433-8850, first to James in Valdosta, Ga. Hi. You're on the air.

JAMESOh, thank you so much. You know, this is a fascinating subject. With a personal family history, I have a paternal grandmother who died with Alzheimer's and a paternal grandfather who died with Parkinson's disease. I'm only 28 years old. And I guess my question is, what should I do? Should I sit back and wait? Should I, you know, seek out testing tomorrow? Or, you know, what are my options at this point?

REHMAnd there you go, Mark Mapstone.

MAPSTONERight. So this is the sort of scenario that we're not quite prepared for yet. And so this sort of biomarker in a 28-year-old individual is not validated at this point. We don't know what giving such a test and what those results would mean in a person of this age. The Alzheimer's disease and Parkinson's are age-related neurodegenerative dementias, meaning the older you get, the more likely you are at risk of these -- the higher your risk.

MAPSTONESo in a 28-year-old, I'm not sure exactly what this test would tell you. I can tell you that there are things that our physicians have been telling us for decades, probably centuries now, about maintaining good health that have really borne out with respect to prevention or staving off dementia such as Alzheimer's disease. And these are things like maintaining your weight, not smoking, alcohol in moderation, exercise, eating well, et cetera. Those sorts of things are the things that I would do now. And we have decent evidence that those things work.

REHMAnd...

MAPSTONEAs far as the blood test...

REHMGo ahead. Sorry.

MAPSTONE...I'm not sure -- as far as a blood test, I'm not sure that that would be applicable in this situation.

REHMDr. Diaz.

DIAZYou've always heard that you should keep your mind active. What about Sudoku?

REHMI mean, you know, and people are talking about doing crossword puzzles, doing puzzles of any and every sort. And, James, you might give that a try. Thanks for calling. Good luck to you. And to Ros in Raleigh, N.C. Hi there.

ROSThank you, Diane. I just want to tell you I love your show.

REHMThank you.

ROSListen to it all the time.

REHMI'm so glad.

ROSI just wanted to -- I have a question. I had successful lung cancer surgery at Duke University two years ago for stage 1A lung cancer. I didn't have to have chemotherapy or radiation. My question is -- there's all this talk about this blood test that possibly could identify the tumor reoccurring. I go for periodic CAT scans for my checkup at Duke. And I was wondering whether this blood test could at some time in the future replace having to go through CAT scans and radiation?

REHMDr. Alizadeh.

ALIZADEHOh, that's a great question. This is the scenario where we have the most experience using our test right now. But, as Dr. Mapstone pointed out, these tests right now are still biomarkers for the most part with clinical versions, real tests being works in progress. So, in terms of actually bringing this to a patient like yourself, I think the best way to go about it is to participate in perspective studies. And at Stanford, we are doing that as part of interventional studies and as well as observational studies where patients are getting the routine standard of care.

ALIZADEHUnfortunately, at the moment at your institution, we don't have a mechanism for carrying this out. But I'm hoping that we could have this test be a component of the monitoring of patients with lung cancer. And my colleague, Max Dean, and myself have been trying to incorporate this with our clinical colleagues in lung cancer monitoring for patients with known lung cancer like yourself.

REHMDr. Diaz.

DIAZYeah, first I want to congratulate you on your early diagnosis. As you know, very different from most patients with lung cancer, you're diagnosed very late. These tests can answer a question that you have. Are you cured or not? And that's exactly what we're going to direct these at in the future.

REHMDr. Luis Diaz, he's an oncologist and researcher at Johns Hopkins Kimmel Cancer Center. Short break, right back.

REHMAnd welcome back as we talk about the use of blood testing to determine certain types of illnesses, certain type of abnormalities in a fetus or a young child. Here's an email from Chris. He says, "I don't understand how a blood test can determine where the cancer is. How do you know if it's in the lung instead of another organ?" Dr. Diaz?

DIAZThat's a very sophisticated question because that hits the next problem right on the head. If we found a mutation in your circulation that heralded cancer, we actually don't know where that's coming from. And the caller's exactly right. We would have to probably perform a PET scan, an age-appropriate screening, so colonoscopy, maybe MRI or mammogram, to try to hone in where that cancer is. But the writer or the caller is exactly right. If we find a mutation in a healthy person's blood, we don't know where that's coming from.

REHMBut how about you, Dr. Alizadeh, as far as lung cancer is concerned, is there a specific marker for lung cancer?

ALIZADEHYes. I think that's a difficult question to answer immediately, but I think that ultimately this is going to be achievable. And I agree that it's difficult to know with full certainty where the cancer is coming from unless you have a specific handle on the genetic alterations that are unique to that cancer. So, for instance, prostate cancer is -- about half or 60 percent of prostate cancers carry specific alteration where a chromosome aberration occurs. And that's essentially a hallmark of that disease and not other diseases.

ALIZADEHSo if I were to detect that in the blood of a patient without symptoms, I might consider prostate cancer much more probable than something else. And at the same time, you know, we're talking about risk here. We're unlikely to expect to find a set of mutations over prostate cancer, say, in a walnut and vice versa for gynecological malignancies in a male.

ALIZADEHFor lung cancer specifically, this is a harder question. There are certain types of genetic alterations that are thus far thought to be unique to lung cancers but a relatively minor population of lung cancers. These are genetic alterations where chromosomes get stuck to each other and drive genes that drive the cancer and can be targeted. And we have a handful of those within lung cancer, especially in younger individuals who have never smoked.

ALIZADEHBut the overall picture here is that can we identify signatures of cancers? And there's some remarkable recent studies looking at thousands of cancers that have been sequenced that signatures of alterations exist that can help us classify cancers. And I think this is the next generation of these tests might incorporate this kind of knowledge and this kind of information. And my laboratory, Max Diehn's laboratory are working on this specific question at the moment.

REHMAll right. And here's another email from Melanie who says, "My daughter battled a 22-month battle with brain cancer, and she passed in September." I'm so sorry, Melanie. She says, "Are these tests able to detect childhood cancers such as brain tumors," Dr. Diaz?

DIAZWe just completed a study in collaboration with our neurosurgery department and neuro-oncologists at Hopkins where we actually collected specimens from patients with brain cancer. Unlike other cancers in the body, the brain tumor is shielded behind what's called the blood brain barrier. And we were not able to detect these fragments of DNA in the blood circulation.

DIAZNow there are newer tests looking at small (word?) a different type of what we call nucleic acids from the cell that may be encapsulated in certain vesicles. And those may be able to be detected in the circulation. But that's still very early and investigational at this time.

REHMAll right. And for you, Mark Mapstone, the question of Alzheimer's is so huge. And as the population ages, so many of us will experience this disease. What is going on in the treatment area to try to develop something, some useful medication to help people who then realize they are going to face this?

MAPSTONEYeah, that's a great question. And I think the field is at a turning point in a way. Much of the work in Alzheimer's disease has been dominated by a certain hypothesis called the beta-amyloid hypothesis, which has been pretty dominant for the past couple of decades in which most of the therapies have been developed. And unfortunately we don't, at this point, have great treatments, and we don't have disease-modifying treatments.

REHMRight. Right.

MAPSTONEAnd so people are starting to question this as the route that we should be -- or the path that we should be on. And so there are people that are taking new stances and looking in new directions for different causes and different treatment avenues. Although most -- my impression is that a large number of pharma companies have sort of retreated to the sidelines as far as Alzheimer's goes just because of the failures in the past.

MAPSTONEThat I think that we're at a point now where, you know, some really transformative work is happening. And we might be able to entice money to come back into development of treatments based on other hypotheses. And so there's a lot of work trying to develop other ideas and novel ideas to look at how to treat and understand this disease.

REHMAnd based on the number of people who may be affected in the generation ahead.

MAPSTONEYes. So we have a large wave of our baby boomers who have just started...

REHMExactly.

MAPSTONE...they just started to turn 65 in 2010, right? So these next couple of decades are going to be major changes in our population base with the aging of our society. And as I've stated before, Alzheimer's is an age-related disease. And so your risk of getting this disease increases with age. And if we have more older adults, we're going to see more Alzheimer's. And we are woefully unprepared for this, which is...

REHMAnd the Congress has thus far not allocated sufficient monies for the kind of research that needs to be done.

MAPSTONEWell, I would agree, I think most people would agree with that. The funding is well below what it needs to be. There's certainly great advances being made in, for example, cancer. Luis and Ash doing great work, things like HIV. And these are well-funded diseases. Alzheimer's has very little funding in comparison. And it's really a problem that we need to have more resources allocated to.

REHMAbsolutely. Let's go to Kristina in Concord, N.H. Hi there.

KRISTINAHi. Thank you for taking my call.

REHMSure.

KRISTINAMy question is twofold, I'm pretty sure. Considering that Obamacare prevents health insurance companies from discriminating against preexisting conditions now, could that increase the likelihood that patients will get comfortable finding this kind of information out and treating it sooner? And also, are there dramatic potential cost implications for health insurance across the board now?

REHMDr. Diaz.

DIAZYeah, with -- this new technology is currently quite expensive. And we've gone from a technology where we can detect one mutation to now detecting several hundreds to thousands of mutations in the blood. And those tests are hundreds of dollars per patient. The next generation of tests are going to be sequencing whole genomes. And currently that's $10- to $50,000. That's beyond the health care system from doing it. But, as the technology improves, as the price comes down, this, I hope, will be inserted as standard care for many different conditions.

REHMWhat about so-called personalized medicine, Dr. Diaz? Do you see that coming sooner and sooner?

DIAZYeah, this is a branch of personalized medicine. The human genome project and then the cancer genome studies that have really allowed us to do these studies are the understanding and classifications of different tumors by their genetics. And one philosophy is that we don't look at a cancer as a breast cancer or as a colon cancer or as a lung cancer. Rather we look at it as a collection of mutations. And that collection of mutations will define how we treat that patient, will define the prognosis of that patient.

REHMI see.

DIAZAnd those mutations will define what biomarkers we use to track in the blood.

DIAZSo what about the use of other body fluids, for example, urine or saliva?

DIAZWell, one of the most exciting studies I was involved in was about a year ago with my colleagues at Johns Hopkins where we had an unusual idea, which was, can we screen for ovarian cancer or endometrial cancer and a PAP smear? And as you know, currently there exists no screening tests for either ovarian cancer or endometrial cancer. And the tests that we do have are suboptimal.

DIAZSo we took a simple PAP smear, a routine PAP smear, and, much like we looked in the blood, we interrogated for mutations. And we found two very surprising results. One is we could detect 100 percent of endometrial cancers and 40 percent of ovarian cancers. And if you think of the biology and anatomy behind this, those cells from the ovary, from the cancerous ovary, survived transit through the fallopian tube, through the uterus, and lodged in the cervix so that a simple brush could collect them.

DIAZAnd, again, so pointing this very powerful technology from the blood into bodily fluids -- and I think that we will see people using saliva, for instance, for head and neck cancers or urine for bladder cancers. And I think that that really goes in the future. And recently there was a test that was approved by a committee at the FDA for stool testing of colon cancer, using the same approach, interrogating stool for mutations that would herald the presence of early colon cancer.

REHMAnd eliminating colonoscopy.

DIAZCorrect. Well, maybe eliminating, but if it's positive, just like the previous caller alluded to, you still have to do the colonoscopy.

REHMExactly. All right. And let's go to Felice in Sudbury, Mass. You're on the air, Felice.

FELICEHello. Good morning.

REHMHi.

FELICEI just -- this is a follow-up from Kristine's (sic) question but more focused on privacy issues. When people either volunteer to have their bodily fluids or biopsy specimens tested for investigational studies, such as yours, what ensures their privacy since this is often recorded in both the medical record as well as in clinical research files? And at what point, if a test is positive and could affect the patient's decision-making, is that research data made available to them?

REHMDr. Alizadeh, do you want to comment, or Mark Mapstone?

ALIZADEHSure. I can comment with regards to cancer, and Mark perhaps with regards to...

REHMAlzheimer's.

ALIZADEH...Alzheimer's.

REHMSure.

MAPSTONESure.

ALIZADEHThus far, the subjects that are being studied in our various laboratory studies are anatomized, and we keep track of them with identifiers. We don't know precisely how to use the information that we have now in a way to adjust our clinical decision-making. And so it is anonymous process. But when built into clinical trials, these biomarkers could ultimately be used to inform decisions. Those processes need to be much more standardized for this to be useful.

ALIZADEHBut I can tell you, in terms of the privacy of the record, the genetic information here is not being used for anything aside from measuring the disease burden. And right now that is an observation made. There are certain studies going on to use that information to change therapy or make a decision to adjust therapy. And that information is shared with the patient.

REHMAll right. And before you respond Mark, let me just remind you, you're listening to "The Diane Rehm Show." Please go on, Mark.

MAPSTONESure, sure. In the field of Alzheimer's, it's very similar to that in cancer, as Ash just mentioned. The data are all anatomized. And as a matter of fact, the clinical data where the patient is actually providing, for example, responses to questionnaires and such is actually done in a separate location from where the blood samples are analyzed. And all of the information is sent via a coded number -- which the code for that is locked up in my office. So basically these are data that cannot be -- you can't figure out who the person is.

REHMThat's good to hear. That's good to hear. All right. And here's an email from George. He says, "Having had two prostate biopsies in the past year, I am eager to learn how long before I can go to my doctor and request a blood test for the prostate," Dr. Diaz.

DIAZProstate's one of the more complicated diseases currently. We have a blood test. It's called the PSA, and the genetic test in the blood may augment that and make that more specific because sometimes the PSA is wrong. But the major issue right now that we're dealing with in prostate cancer is, even if you detect the cancer, even if you have a biopsy and there's a cancer there, what do you do?

REHMExactly.

DIAZAnd what's the correct treatment? Do you go with radiation or surgery, or do you watch? And that is a conundrum that's facing...

REHMAnd you watch those numbers go up and down.

DIAZYeah, and I'm sure there are many people listening that watching the PSA is incredibly anxiety provoking. It can consume a patient. So we have to be careful that, even though we're providing more information, that we're not creating a very high anxiety situation for that patient.

REHMOK.

DIAZIt's a mental health issue.

REHMSo, finally, you are all working in very exciting areas. How soon do you think that these tests will appropriately be beneficial to patients out there who come to you and say, Dr. Diaz, I want you to take a blood sample, tell me what's coming?

DIAZWell, I think for patients with established cancer, with cancer that is stage 4 and that we're tracking the response to therapy, either chemotherapy or precision-medicine therapy, like a targeted therapy, that's going to be available probably this year. There are companies right now working on what they call CLIA-approved tests where you can send the patient's samples for this type of analysis.

DIAZThe early detection question or the question whether or not I'm cured after surgery or not is going to take longer because you have to do appropriate clinical trials in order to validate this as a good marker that actually impacts survival and outcome.

REHMAll right. And there we are. Dr. Luis Diaz, oncologist at Johns Hopkins University, Dr. Ash Alizadeh of Stanford School of Medicine, Mark Mapstone -- and he is in the field of cognitive neuroscience at the University of Rochester Medical Center -- thank you all so much. And thanks for listening. I'm Diane Rehm.

ANNOUNCER"The Diane Rehm Show" is produced by Sandra Pinkard, Denise Couture, Susan Casey Nabors, Rebecca Kaufman, Lisa Dunn, Danielle Knight and Allison Brody. The engineer is Toby Schreiner. Natalie Yuravlivker answers the phones. Visit drshow.org for audio archives, transcripts and podcasts. Call 202-885-1200 for more information. Our email address is drshow@wamu.org and we're on Facebook and Twitter. This program comes to you from American University in Washington, D.C. This is NPR.