MS. DIANE REHMThanks for joining us. I'm Diane Rehm. A long term study cast new doubts on the value of annual mammograms. According to the research, annual screening leads to increased diagnosis, but does not reduce the number of cancer deaths. Joining me by phone from Helsinki, Finland, Dr. Anthony Miller, professor emeritus at the University of Toronto. From a studio in Newton, Massachusetts, Daniel Kopans. He's professor of radiology at Harvard Medical School. And by phone from Boston, Shannon Brownlee of the Lown Institute.

MS. DIANE REHMJoining us by phone from Washington, D.C., Dr. Ranit Mishori, associate professor at the Georgetown University School of Medicine. I know this is a subject of interest to many people. I hope you'll join us, 800-433-8850. Send us your email to drshow@wamu.org. Follow us on Facebook or send us a tweet. And welcome to all of you. Let me go first to Dr. Anthony Miller. He joins us from Helsinki, he's professor emeritus at the University of Toronto. Dr. Miller, you're the lead author on this study. Tell us about the scope and the findings.

DR. ANTHONY MILLERWell, this was a study in which we recruited just over 89,000 women in 1980 to 1984 and randomized them with their consent to mammography and skilled breast examination screening on the one hand and the control group on the other hand. Because of the evidence that we had at the time, the control groups were different by age. Women age 40 to 49 in the two trials were allocated to a single breast examination in the screening center, and then they were followed annually by mail. And they were expected to go and see their doctor if they had any concerns.

DR. ANTHONY MILLERWomen age 50 to 59 in the control group were asked to come back each year and have the annual breast examinations, but not mammography. So the older age group, we specifically focused on the role of mammography to what extent does it reduce deaths when added to a breast examination compared to the breast examination alone. The breast examinations were given by nurses who were very carefully trained, and the mammograms were given in carefully selected centers throughout Canada. The machines were carefully assessed, and the radiologists participated in a great deal of interaction.

REHMSo tell me about the findings. Are they an indication that we're doing too many yearly breast exams from your perspective?

MILLERWell, certainly if the breast exams include mammography, from my perspective, we are doing far too many, and one of the reasons for that, in addition to failing to find, in our study, any benefit in terms of reduction of deaths from breast cancer, we found something else, which we've been suspecting for some time. And that is a number of the breast cancers that were found by mammography. Not many of these -- these cancers could not be felt -- would never have presented in that woman's lifetime.

MILLERThey would be growing so slowly that they would never have caused problems. This fits in with what we know about the biology of breast cancer because it has a very long natural history. But this means that about 22 percent of women who were -- whose breast cancers were detected as a result of the screening should never have been treated, and yet they were. And so women have these -- whatever complications come from treating unnecessarily.

REHMAnd does treatment, in your mind, include biopsies to determine whether, you know, a suspected lump is, in fact, cancerous?

MILLERWell, as it happens, I'm referring to what happened after the biopsies confirmed that the pathologist -- on examining the material under the microscope, the pathologist confirmed this was cancer. In addition to these unnecessary cancer diagnosis, there were nearly five times as many women who, in fact, had a biopsy because of a suspicion of breast cancer, and they were not found to have breast cancer. So this is unnecessary biopsies in addition. So there is a potentially a substantial degree of harm from using mammography.

REHMDr. Miller, you talked about women age 50 to 59. Did you also study women 60 and older?

MILLERNo, we didn't at the time. We have felt that the information we would get from women age 50 to 59 would apply to older women. There's a major shift in the way breast cancer occurs and behaves in women under the age of 50 and over the age of 50, which, in fact, coincides approximately with the age of menopause. But there isn't that change in the behavior of breast cancer between women at the time of their 60s, and we felt that whatever we learned from women in their 50s would apply to women in their 60s.

REHMAnd that is the voice of Dr. Anthony Miller. He's professor emeritus at the University of Toronto and lead author on the study we've been talking about regarding breast cancer. Shannon Brownlee, turning to you, you're with the Lown Institute that's a nonprofit in Boston working to improve healthcare and health. The research is clearly going to add to the debate as to when and how frequently women should have mammograms. What's your thinking?

MS. SHANNON BROWNLEEYou know, this is a study that's obviously been a long time coming, and yet there have been hints for quite a long time that mammography is not as effective as the public has been led to believe. And it really has to do with sort of how we think about cancer and how we imagine it works. And we have thought for so long that all little cancers start to grow and become bigger cancers and, if you don't catch it early, then that cancer is inevitably going to eventually kill you.

MS. SHANNON BROWNLEEAnd what we have been learning over the last couple of decades is the biology of cancer is much more complicated than that. And so what we have done with mammography is we have gotten really, really good at catching a lot of cancers that, in fact, are not necessarily going to lead to harm ever in the woman's lifetime, or they wouldn't lead to harm even if they were caught later in the woman's life.

MS. SHANNON BROWNLEEAnd so I think that one of the most important things about this study is that it shows there is actual harm from mammography, that we are catching a lot of cancer and treating a lot of women who don't need to be treated, and those treatments are not entirely benign.

REHMHow do you see -- or do you see a comparison here with prostate cancer?

BROWNLEESo prostate cancer, as we've spoken before, Diane, is kind of the poster child for the problem of overdiagnosis and overtreatment. The PSA test was a blood test that has been very widely used, especially in the United States, to catch prostate cancer early. But we know that what it does is it catches cancers that would not have bothered the man, and the treatment itself, removal of the prostate, often leads to a lot of harm for men.

BROWNLEEAnd it's really a parallel situation with breast cancer. The difference between the two is, in some ways, we're sort of catching the PSA wave before it completely overwhelms the way we think about things whereas mammography had become such a part of the medical landscape. I mean, we have advocacy groups out there advocating for mammography. We have clinics trying to find free mammography for poor women.

BROWNLEEI mean, it's just become an axiom, catch it early, and use a mammogram and mammograms are good. So I think what this study does is it really adds a very important piece of information to this ongoing debate about mammography. And I hope that, at some point, we start to be a little more rational about it.

REHMRational. When you say rational, what do you mean?

BROWNLEEI mean that belief drives this so much of the time. So I'll give you an example. When the U.S. preventive services task force recommended against regular mammograms for women between 40 and 50 -- and this has happened a couple of times over the last 20 years that I've been watching the mammography debates.

BROWNLEEThe U.S. preventive services task force said that there isn't evidence to say that there's benefit from mammography for women between 40 and 50. There's lots of evidence for harm, and it should not be done regularly. Well, the breast cancer advocacy groups just went into high gear and said, this was outrageous. This was anti-women. This is a plot to save money. I mean, there were all sorts of arguments that were made.

REHMAll right. And we'll take a short break here. When we come back, we'll hear from Dr. Daniel Kopans, professor of radiology at Harvard Medical School.

REHMAnd just to reiterate, we're talking about a recently-published study indicating annual mammogram screenings for women age 40 through 59 lead to more diagnoses of breast cancer. But the screenings apparently do not reduce breast cancer deaths. The research tracked nearly 90,000 women for 25 years, and, of course, the study is adding to the debate already underway over benefits of yearly mammograms.

REHMJoining us now is Dr. Daniel Kopans. He's professor of radiology at the Harvard Medical School. Dr. Kopans, you've been a strong voice urging women to get annual mammograms. Explain why.

DR. DANIEL KOPANSWell, Diane, first of all, thank you for letting me participate in the discussion.

REHMOf course.

KOPANSI would point out, I've been sitting here for 20 minutes listening to a lot of, unfortunately, misinformation, and I'll try and address that as quickly as I can.

REHMAll right.

KOPANSI think the bottom line is mammography screening began in the United States in the mid-1980s. Prior to that, the death rate for breast cancer had been unchanged for 50 years. Now each year more than 30 percent fewer women die of breast cancer primarily because of early detection. What's interesting is that people who oppose mammography screening don't actually take care of women with breast cancer. If you speak to medical oncologists, they will tell you that the therapy has improved, but therapy saves lives when breast cancers are found early.

KOPANSIn terms of Dr. Miller's study -- and I would say hello to Tony. I haven't seen him in many years. I was actually invited by Tony and his associate Dr. Kanelia Banes (sp?) to review the quality of the mammograms in this trial. I would point out to your listeners that this trial was done in the 1980s, and it was theoretically designed to test high-quality mammography screening to see what benefit it had.

KOPANSUnfortunately, for whatever reasons, the quality of mammography -- as I found when I was asked to review with two other radiologists and as was published, the quality of mammography was poor. And it was poor for a number of reasons. First of all, instead of using state-of-the-art equipment as you would expect in a state-of-the-art study, they used old mammography machines. And at least one of them was a second-hand machine. And many of them were outdated -- they used outdated technology, problem number one.

KOPANSThe next problem was that they did not use grids. If you take a mammogram without a grid -- a grid is a way of reducing what's called scatter radiation. A mammogram without a grid is like looking through a window that's covered with fog. You can barely see what's through the window. If you don't use a grid, you can barely see what's in the breast. That's problem number two.

KOPANSProblem number three, they had poor positioning. They resisted, for whatever reason, to use the -- what was in the modern positioning, which is now what we use all the time. And they used positioning that did not include the entire breast. So cancers at the back of the breast were missed. And the combination of all these is evidence in their own report. They point out that only 32 percent of the cancers that were detected among the screened women were detected by mammography.

KOPANSNow, even at that time, but certainly now, we find 60 percent or more of the cancers are detected by mammography. Another piece of information that shows that the quality of the mammography in this trial was really poor is that the cancers detected by screening were 1.9 centimeters in average diameter. The cancers among the women who weren't getting mammograms was 2.1.

KOPANSThat was a difference of 2 millimeters, which is about, you know, a 16th of an inch. It's very, very little difference. So the mammography was terrible, and they did not pick up small cancers. That's problem -- a series of problems that make up one of the reasons why this trial has never really been given a lot of validity among those of us who understand the trials.

REHMAll right. Dr….

KOPANSThe next problem -- they had 20 minutes, so let me try and get...

REHMOK.

KOPANSThe next problem is, in a randomized control trial, the rules -- and these aren't my rules. These are rules you can go in textbooks and go online and look up papers. When you do a randomized control trial, the goal is to divide a large group -- in our case -- in this case of women -- a large group of women into two groups that will have exactly the same outcomes if you were to leave them alone.

KOPANSSo, in other words, you divide the large group in half. And if you've done it randomly, then there'll be the same number of deaths in both groups from breast cancer. There'll be the same number of women diagnosed with breast cancer and so on. The way to do that is critical, and that is you can't know anything about the women before you decide or before you assign them to be in one group or the other.

KOPANSThis again, not my rule. This is fundamental, and you know this, Diane, as well as I do. In the Canadian study, they first did a clinical breast examination on all the women. So they knew who had lumps because there were women who had cancer already, and they knew which women actually had large lymph nodes in their armpits, which indicate that the breast cancer has spread out of the breast.

KOPANSThe examinations, as Dr. Miller stated, were done by trained nurses and some physicians, but the information was given to the coordinators. And this was documented in several papers. The coordinators who were assigning the women to be in the screen group or the control group knew who had lumps and who had positive lymph nodes in their underarms. So that's a huge violation of the randomized control trial.

KOPANSNow, had they selected random numbers or some other way to assign women to one group or the other, it might not have been so bad. But they assigned them on open lists. That meant that the coordinators who were assigning them had -- line one was a mammogram, line two was a control group, line three was a mammogram, line four was a control group and so on.

KOPANSSo it certainly was possible and probably happened that, for probably good reasons -- I don't think these coordinators really felt that they were trying to damage the trial -- but they could skip a line and make sure that a woman with an advanced cancer got a mammogram. Now, that was actually proven to be the case by a fellow at the National Cancer Institute named Tyrone back in the 1990s who showed...

REHMAll right. Dr. Kopans, I have a couple of questions for you.

KOPANSWell, let me just finish, Diane. They had 20 minutes, Diane.

REHMNo, I'm sorry. I'm sorry, Dr. Kopans. I really want to carry on this conversation.

KOPANSAll right. Go ahead. The trial is a corrupted trial.

REHMPlease, please -- OK. And that's your conclusion. I appreciate that.

KOPANSWell, that's the scientific evidence.

REHMOK. When an abnormality is found in the breast through mammogram, what is it that can actually be known about the patient prognosis? How do you tell from a mammogram if what you see and determine is an abnormality that's showing up? How do you know whether that's going to lead to a life-threatening tumor?

KOPANSWell, first of all, the statement that was made that there's a lot of overdiagnosis of invasive breast cancers is simply not true. And I'd be happy to tell you the data on that, but that's probably another talk. But there is no...

REHMThat's a long -- but your conclusion I accept.

KOPANSThere is no overdiagnosis of invasive cancer. But, that said, to answer your question, mammography -- all screening tests are to filter out -- or filter in individuals who are more likely to have cancer based on what you see on the screening tests. This is the same with Pap smears, for example. And so what has to happen is, when we see something on a mammogram, a patient is recalled for additional evaluation.

REHMOK.

KOPANSAnd this is the same rate, again, as women recalled for Pap smears. It's no different. And we do additional x-ray imaging, take a few extra pictures. Most of the time, we can say there's nothing there. It was just a superimposition of normal tissue in the patient. Everything's fine, and we'll see you next year. In about one to 2 percent of women who have a screening mammogram, we see something that concerns us enough to recommend a biopsy.

KOPANSAnd so those are now done under imaging guidance with local anesthesia and a needle. And about 30 to 40 percent of those turn out to be breast cancer. So the way you prove that something is breast cancer requires a biopsy. I would point out that, when biopsies were done just because of palpable abnormalities, which many opponents to screening are saying, oh, wait until you have a lump, there are many, many lumps that aren't cancer.

KOPANSAnd only 15 percent of lumps that are biopsied, because you can feel them, turn out to be cancer. And when they are biopsied, they're larger and at later stage (unintelligible)...

REHMAll right. I do want to bring in another voice here, but you've been hearing Dr. Daniel Kopans. He's professor of radiology at the Harvard Medical School. Turning to you now, Dr. Ranit Mishori, associate professor of family medicine at the Georgetown University School of Medicine, what are you now telling your own patients about the value of annual screening with mammogram?

DR. RANIT MISHORIGood morning, Diane. As you could tell from Dr. Kopans' response, this is a very, very complicated issue.

REHMIndeed.

MISHORIAnd it's complicated on the population level. Most of these studies are population-based studies. But the question becomes, what happens with the individual patient that walks into your door? How do you get through all of this conflicting information?

REHMOf course.

MISHORIGranted, all studies have some weaknesses. But I think this is one study in a series of studies that show that the bottom line is mammograms are imperfect. And it's very, very hard to make the right call. I usually have a conversation with patients. We -- I tell them that screening is a choice, but I want them to make an informed choice.

MISHORINow, we usually don't have time to go into all the biostatistical and probabilities that are involved in many of these studies. But there are some tools to talk to patients about their own risk and help them make a shared decision with me, with other clinicians, as to whether they should go ahead and have that mammogram or not.

REHMDr. Ranit Mishori is associate professor of family medicine at the George University School of Medicine. And you're listening to "The Diane Rehm Show." Now, turning back to you, Dr. Anthony Miller, as the lead author of this study, you've heard what Dr. Kopans has said about what he and other professionals see as the flaws in your study. How do you respond?

MILLERWell, Dan Kopans -- and hello, Dan -- has made these comments over the years, and we have repeatedly responded to him. As he well knows, we had a referenced radiologist and a referenced physicist. They did their very best with the technology available in the 1980s to ensure that the mammograms were the best that could be obtained.

MILLERAnd what is more, one of the radiologists who shared with him the review he talked about -- and I don't think I should mention his name, but he was a very senior radiologist who had participated in the U.S. breast cancer detection demonstration project -- had come up to advisors before the study started. And the positioning we used was what he advised us to use for standard North American positioning at the time.

MILLERWhen one of our referenced radiologists later asked us to change, we did change almost at once. He seems to be pegging a lot of -- well, a lot on the fact that, in his view, only 30 percent were detected by mammography. Well, that's true. But, in fact, it's because in what he has been talking about, 60 percent, you're only dealing with far -- the poorer breast examinations that were used in our study.

MILLERWe used highly skilled examiners, very carefully trained. And at one time I thought that what we had found was that when you do that, you get very good results, as good as mammography. Now I'm not sure that you will do so.

REHMAll right. Let me turn back to Dr. Kopans for a moment because you, I gather, Dr. Kopans, invented 3-D mammography. What do you see as the advantages of that approach to mammography? Does that make all the difference?

KOPANSWell, I appreciate your asking that, Diane. Just in response to what Prof. Miller just said, in fact, the referenced physicist has written in the Journal of the National Cancer Institute that the quality of the mammography in that trial was below what was even being done in the community. So just to reemphasize, it was very poor quality mammography. And you can get in touch with -- Martin Yaffe would be happy to discuss that with you in Toronto.

KOPANSAnyhow, in terms of digital breast tomosynthesis, this -- we've been constantly trying to improve our ability to detect small breast cancers and reduce the death rate even further. And we did develop a technique at Mass. General that allows us to remove some of the normal breast tissue that gets in the way on our mammograms. When you look at a standard two-dimensional mammogram, even a good one, not the kind that we're talking about, and you -- it's like a book with clear pages. You can hold it up to the light and see the words on the pages, but they're all superimposed one on top of the other.

KOPANSDigital breast tomosynthesis will let us look at each page individually. So cancers that are hidden behind normal tissue are much easier to see. And in addition, things that confuse us that we briefly talked about earlier, things that confuse us will be less confusing, and we won't recall as many women for additional evaluations.

REHMAll right. I want to turn now -- pardon me -- back to Shannon Brownlee. We know that breast cancer diagnosis is up. What are the overall trends in breast cancer deaths, Shannon?

BROWNLEESo if you look across the population of women in the United States, breast cancer deaths are declining. And the real question is, what is causing that decline? Is it because we are detecting more and more and smaller and smaller breast cancers? Or is it because our treatments are getting better or some combination? There have been some studies that have tried to parse out what's the real -- what deserves the credit. And there's growing thought that in fact it is mostly that we have better and better treatments.

BROWNLEEWe have better drugs. We have fewer women who are on hormone replacement therapy which clearly had an effect on the rate of breast cancer in this country. So is it screening? I think that's still up for debate. But I don't think that we can say it's because of screening, and we also know it's causing harm.

REHMAll right. All right. Shannon Brownlee of the Lown Institute. That's a nonprofit in Boston working to improve healthcare and health. Shannon, thanks for joining us. We'll be right back.

REHMAnd welcome back as we talk about a new study from Canada regarding mammography and the number of mammograms and the spacing thereof, how often they can identify potentially deadly cancers, and how often they lead to unnecessary treatment. With me from Helsinki is Dr. Anthony Miller. He's the lead author of the study, professor emeritus in Toronto. Dr. Ranit Mishori, she's at the Georgetown University's School of Medicine. And Dr. Kopans, professor of radiology at the Harvard Medical School.

REHMHere is our first email from Barbara in Kenilworth, Ill., who says, "What about the inconsistencies of who reads the screenings and the shortage of radiologists across the country? Aren't there too many screenings to read by too few radiologists?" Dr. Kopans?

KOPANSDiane, that's an important question. And so far, radiologists have risen to the task, and we seem to be keeping up with the number of women being screened. Now, a lot of women don't participate in screening, and if they do -- which I hope they will -- we will figure out how to deal with that. But that's an important question. And just to go back to the study we're talking about, the radiologists in that study were not specifically trained for reading mammograms, and technologists, as I said, were not trained to actually position the breast.

KOPANSI'd like to make one other comment. Your previous guest said that the U.S. Preventative Services task force said there was no benefit for screening women in their 40s. In fact, that's not true. The U.S. Preventative Services task force -- if you read them carefully -- say that the most lives are saved by annual mammography beginning at the age of 40.

KOPANSThey just decided to advise women that they didn't think that the women would want to suffer the anxiety of being called back for additional evaluation, and that's why they shouldn't participate in their 40s. The death rate has declined for women in their 40s. The randomized control trials clearly show a benefit for screening women beginning at the age of 40. They always have. And Prof. Miller's study really is an outlier in all those things.

REHMDr. Miller, is your study an outlier?

MILLEROur study is different because we tested different things, but our study is particularly different because it was done in the era of modern treatment for breast cancer. And the Swedish studies, with which Kopans is using to bolster his arguments, were not. They didn't have good adjuvant treatment at that time. And in fact he's wrong about the decline in breast cancer mortality in the United States and Canada. It's because of good treatment. It's not because of screening.

REHMAll right. Dr. Mishori, I want to come back to you. What do you see as the risks of getting annually screened?

MISHORIWell, there are many -- this is a very emotional and deeply personal issue for many women. And I know Dr. Kopans -- he talked about the fact that some women are called back. There are. But studies have showed that about 30 percent of women -- up to 30 percent of women do get called back for suspicious results.

MISHORIAnd this is a very, very stressful time for a woman when she gets a call from the mammography center saying, listen, there's something wrong. And it should not be discounted. This is one of the harms. There are other harms. There are harms of the follow-up test and biopsies that need to occur when somebody has a suspicious finding.

MISHORIBut this is not to be discounted because women, at that point, feel that they may have a fatal disease. They may have a condition that will affect their lives and their children's lives and their families. So this is news that many women receive, and it affects them sometimes for days, sometimes for weeks until they have that biopsy or a more certain result of a follow-up biopsy or a follow-up mammogram.

REHMAnd, Dr. Mishori, what is it that makes a woman's prognosis, who may have been told she has breast cancer -- what makes it so hard to predict how serious that breast cancer is going to be or whether even without treatment she might live a long life?

MISHORIDiane, I'm not an oncologist, and this is not my specialty. So when a woman is diagnosed with breast cancer and when I help to usher in that diagnosis, I send her to an oncologist. But what we do know is there are different types of breast cancer. Breast cancers can have different manifestations in how aggressive they are. So you don't know what kind of breast cancer a woman has until she has the definitive diagnosis, which comes through a biopsy.

MISHORIAt that point, you know, depending on the type of breast cancer that she has, there are different treatments. I also beg to differ with Dr. Kopans about the mortality rate, just as Dr. Miller was saying. I think the treatments have improved dramatically over the last few decades. And in my opinion, and from the studies that I have read, this is my understanding, is that the mortality differences are primarily related to different types of treatment that are offered these days.

REHMAll right. I'm going to open the phones now. I'm going first to David in New York City. Hello, you're on the air.

DAVIDHello, there. Thank you for taking my call.

REHMSure.

DAVIDI'm a fan. I wanted to say that when I hear Dr. Kopans speak about the randomized trail from Canada that did not show a benefit, what I'm hearing from him is that there were methodological flaws, that there were problems in the study. And I think we can accept that there were probably problems in that study and many others, but what patients are looking for when they walk into an exam room and a physician or scientist, ostensibly, recommends a mammogram, I think they're expecting that there has been lots of randomized trials that prove that mammograms save lives.

DAVIDAnd what Dr. Kopans is suggesting is that we haven't been able to prove that they don't save lives. Instead of saying he has proof that they do save lives, he's saying there are 600,000 women in randomized trials, including the Canadian trial, and that we have been unable to show a benefit, but he believes and hopes and sort of wishfully believes, I think, that they might still save lives. And we just haven't seen it. So I think it's an important distinction.

REHMAll right. Dr. Kopans, do you want to respond?

KOPANSYes. I'd love to respond to that. First of all, that's not true. The scientific evidence clearly proves a benefit. The U.S. Preventative Services task force has said this. All other studies of the science show that the women in the screened groups have fewer deaths than women in the control groups from breast cancer. And I would also take issue with the comments that have been made that studies suggest that it's therapy that's saving lives. In fact, there are no studies that have proven that therapy is saving lives.

KOPANSI personally believe -- there's where belief comes in -- that therapy has improved and is saving lives, but the studies certainly in Europe have shown that when you introduce screening, the death rate goes down. Even if women have all had access to modern therapy, it's not the therapy that's doing it. What's interesting is no one on this call actually takes care of patients with breast cancer, except I do up at the beginning of their problem. None of the major oncology groups in this country have come out in support of the U.S. Preventative Services task force's guidelines. And the reason…

REHMAll right. Dr. Miller, do you want to respond to that?

MILLERWell, first, I think one of the problems here is that Dr. Kopans takes care of women with breast cancer and doesn't know epidemiology. But the other is that there has been a big assessment by a number of people, and there's absolutely no question that treatment has improved the outcome of breast cancer all over the world.

MILLERAnd as to my study having methodological flaws, it was evaluated by a number of international experts and found to be probably the best study performed, meticulous. It may be that we made an error in bringing people like Dr. Kopans and others to come and look at our mammograms, but nobody else did. And at that period in the use of mammography in North America, I suspect exactly the same conclusion would be reached.

REHMDr. Miller, I want to clarify something. How old is your study?

MILLERIt began -- we started planning in 1972. We did pilot studies. We had a lot of experts to advise us. We actually started the study in 1980. We finished recruiting women in 1984. We finished screening in 1988.

REHMAll right. And you're listening to "The Diane Rehm Show." I want to take a call from Addie in Greensboro, N.C. You're on the air.

ADDIEHi, Diane. Thank you for having me on.

REHMSure.

ADDIEI'm calling because I want to talk a little bit about the issue of false negative tests by mammograms. My breast cancer -- three years ago, I was diagnosed with invasive breast cancer. And it was missed by at least three years of mammograms. And, subsequently, I was instrumental in getting the law changed here in North Carolina as it pertains to women being informed about their individual level of breast density and how that relates to their mammogram. There's only right now, I believe, 14 states that have had these laws passed. Do all believe that the FDA should make a regulatory change?

REHMAll right. Thanks for your call. Dr. Kopans?

KOPANSYeah. No. I think the caller makes a very important and unfortunate point. And that is that, as important as mammography screening is in reducing deaths from breast cancer, it's not the ultimate answer to breast cancer. Those of us who have been in the field for years recognize this, and that's why we're trying to constantly improve on it.

KOPANSDigital breast tomosynthesis may, we think, will find a lot of the cancers, such as the woman who just called in, find them earlier. But there are still going to be cancers that we can't find early enough. We're looking at ultrasound screening, magnetic resonance imaging screening. My concern is that women are going lose access to mammography screening because of a lot of the misinformation that's out there. And, unfortunately, with all due respect to my friend Tony, as a result of a methodologically-flawed trial that comes up with additional misinformation, I think that would just be unfortunate.

REHMAll right. Dr. Miller, I want to ask you about the exposure to radiation through annual mammography. Do you think that's potentially harmful?

MILLERI think it's unlikely to be. I had done other research which showed that the effects of radiation breast cancer was largely restricted to women under the age of 35. We took a great deal of care to insure that the radiation dose given to women was as low as possible. There has been some research, particularly on cells, to suggest that there could be an effect, even though the low dose of radiation give in modern mammography, and in our trial. But I doubt if that has had a major effect in our population.

REHMHere's an email from Brandon who says, "The problem is our ability or inability to differentiate between problem versus non-problem cancers. This differentiation has seen great advance with genomics now in regular use. This study is based on exams performed almost 30 years ago." Dr. Miller?

MILLERWell, it is certainly true we need to find out which are the most important breast cancers that we worry women about and which are the unimportant ones, the overdiagnosis ones, that grow so slowly that they wouldn't harm women. And I believe there is a great deal of research going on to try and do this, but we're not there yet. And for the moment, part of the problem here is that we find these shadows which turn out under the microscope to look like cancer but don't behave like cancer.

REHMAll right. And finally, Dr. Mishori, does this study affect what you will say to your patients about annual mammograms?

MISHORIDiane, this study, as I said before, is one of several studies that have shown how imperfect mammography is. And this is also echoed by the caller before, who was in her 40s, and her cancer was not detected for three mammograms. So the answer to your question is, no. I'm going to continue to counsel my patients to think about this choice of screening very, very carefully, based on personal risk, based on personal value, based on the science that is available.

MISHORIThe other thing I wanted to mention is a lot of the proponents of early detection -- we need to think about their financial incentives. And I believe Dr. Kopans has several patents pending on imaging studies. And I think that new imaging studies should come out, but I just wonder if this is not something that should enter into the equation when we listen to different experts debate this issue.

REHMDr. Kopans?

KOPANSI think that's a perfectly valid point. I would also point out that Prof. Miller was paid by money in his research to do the National Breast Screening Study of Canada. I've never raised that as an issue, saying that he couldn't be objective based on that, but that's true. I do have a patent on digital breast tomosynthesis, but it's held by the Massachusetts General Hospital.

REHMAll right. And finally, Dr. Miller, based on this study what would you tell women about annual mammography?

MILLERPlease, Diane, can I just defend myself? I wasn't paid by the National Cancer Institute of Canada to do this study. I was paid by the University of Toronto to be a professor in the University of Toronto. And it was part of my research to do this study.

REHMThank you.

MILLERWhat would I tell women? I'd tell them to place less reliance on mammography. I'd tell them to be very well aware of their breasts. I personally believe there's much more data in favor of breast self-examination then many people now accept and that it is extremely important that, if women are concerned about their breasts, if they think there's an abnormality, they not only have mammograms, but they are examined by somebody who knows the signs of early breast cancer.

REHMAll right. And we'll have to leave it right there. Dr. Anthony Miller of the University of Toronto, Dr. Daniel Kopans of Harvard University Medical School, and Dr. Ranit Mishori of the Georgetown University School of Medicine -- a very controversial subject -- thank you all so much for joining me. And thanks for listening. I'm Diane Rehm.