Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS
The Truth About Cancer
Take One Step: A PBS Health Campaign
The Truth About Cancer + Take One Step: A Conversation About Cancer with Linda Ellerbee  

Interview: Dr. George Demetri
< Interviews with Experts

George Demetri

Dr. George Demetri is director of the Center for Sarcoma and Bone Oncology at Dana-Farber Cancer Institute, director of the Ludwig Center at Dana-Farber/Harvard Cancer Center, and executive director for Clinical and Translational Research at the Ludwig Institute for Cancer Research. He is also Associate Professor of Medicine at Harvard Medical School.

Interview Content

Defining Cancer & Metastasis

GARMON: What is cancer?

DEMETRI: Cancer represents many different diseases, but at its heart cancer is an uncontrolled growth of cells, which then attain the quality of being able to spread through the body and get in the way of the normal body working mechanisms.

GARMON: Can you define what a metastasis is?

DEMETRI: A metastasis is a group of cancer cells that have broken away from wherever they started in the body and are now living somewhere else inside the body. The problem with metastases are that they get in the way of the normal functioning of the organ in which they find themselves. So breast cancer cells can metastasize or spread to the lung and get in the way of the lung function. It could spread to the bone and cause pain. It could spread to the brain and cause major problems.

GARMON: Now can you define what a distant metastasis is? There is a lot of talk about how much lower your chances are of survival in the common cancers once you get a distant metastasis versus a regional one. So in that context, can you explain a distant metastasis please?

DEMETRI: Over the last 50 years a lot has been researched, mainly by surgeons, who have studied how cancers start in one place but then move elsewhere through the body either close to where they started, often in a lymph node, or far from where they started. Let's say if it starts in the breast, does it wind up in the liver or the brain, something very far from where it started? When it goes very far from where it started, that's a distant metastasis. The close ones, like a node metastasis or a regional metastasis may have important information about whether a cancer cell has the potential to spread distantly through the body and run the risk of killing the patient.

GARMON: When you have a distant metastasis for the leading causes of cancer deaths, for the common cancers, what then becomes the survival rate or the ratio—you could say 1 out of 10. How many patients can actually live more than 5 years once they have a distant metastasis in common cancers?

DEMETRI: If a cancer has already spread distantly in the body it really represents a grim prognosis for that patient. Generally with the common cancers survival rates beyond 10 years let's say from the time of a distant metastasis would occur only in less then 1 out of 10 patients, maybe 1 out of 20. So it's a very rare thing. We don't know what makes those patients lucky enough to live that long, but the vast majority of patients, 9 out of 10 maybe even 19 out of 20, would die from their cancer once it has spread distantly.

Back to Top

Understanding Chemotherapy Side Effects

GARMON: Why do conventional chemotherapy treatments cause side effects? You were using examples of what are the cells in the body that are rapidly—

DEMETRI: Growing, dividing.

GARMON: I was going to say dividing, but—

DEMETRI: Right. The standard cancer chemotherapy drugs are generally poisons that can kill dividing cells, cells that are making copies of themselves. Cancer cells make copies of themselves, but so do some normal cells. Normal cells like the cells responsible for the hair or responsible for the lining of the mouth or for the bone marrow that create new blood cells. This process of cell division is critical so that we can make blood cells and it's why cancer chemotherapy commonly knocks down the production of those blood cells because the normal dividing cells are inhibited by the chemotherapy. This is also why older cancer chemotherapy actually works because it can kill dividing cells non-specifically. So it has all those side effects on normal cells as well.

GARMON: Well, let's just use why you lose your hair as one example of one of the side effects. Okay, so why do conventional chemos cause that?

DEMETRI: Traditional cancer chemotherapy drugs cause side effects because they non-specifically hurt dividing cells. Cancer cells are dividing, but so are a lot of normal cells in the body specifically cells like the ones that give rise to the hair. They are constantly dividing. It's part of how we produce hair. Other cells in the body also divide, such as the cells lining the gastrointestinal tract, so this accounts for two of the more common side effects with chemotherapy, the standard kind of chemotherapy. Hair loss, because the dividing normal cells aren't able to keep up with hair production or mouth sores or other intestinal ailments where the lining cells of the gastrointestinal tract can't keep up with the needs of the body.

Back to Top

The Leading Causes of Cancer Deaths

GARMON: Can you tell me what are the leading causes of cancer deaths?

DEMETRI: In the United States right now the leading causes of cancer deaths are due to the common cancers. Things like lung cancer, colorectal cancer, breast cancer, pancreatic cancer. We certainly see in this list also blood cancers like leukemias and lymphoma as well as prostate cancer.

Back to Top

Results Since Nixon Declared the War on Cancer

GARMON: Since Nixon declared the war on cancer in the United States how much real progress have we made?

DEMETRI: Since the United States under Richard Nixon's presidency in 1972 declared the war on cancer full force, our country has spent a lot of cancer research and there has been a tremendous growth in what we understand about what makes cancer cells tick. It's really though only in the last 5 or 7 years that we have been able to apply that knowledge to develop new therapies and already we have seen tremendous examples of how that has come forward with effective new drugs, but admittedly they have only had a rather marginal impact on outcomes on survival for patients with the common cancers. Those of us in the field remain extremely excited that the next decade will be quite different. That we are going to be able to understand cancer better; put patients into categories and pick drugs for them on a very personalized basis which is very different than what we have done in the last 10 or 20 years. We have in many ways treated all breast cancers the same. We are getting better at personalizing our treatments based on the science that has come out in the last 30 or 35 years of investment in science that this country has made.

Back to Top

How Cancer Begins

GARMON: So how—in the simplest way possible, how does cancer begin?

DEMETRI: There is still a lot of research going on about what makes cancer start? What is the first thing that happens and one interesting thing is that it is one cell that makes a mistake. One cell makes a mistake in its DNA, that gives rise to a protein that is not working properly and that leads to the cell growing and dividing abnormally and eventually not sticking where it's supposed to stick and go spread through the body. At its heart, it's all about a cell making a mistake in the code of life, in the DNA. There are different places in the DNA it can make that mistake that can give rise to different proteins that are mutant that can give rise to different signals that drive that cancer cell to act aberrantly. To act as a rogue cell inside the body and because there are so many different proteins that the cell can make a mistake on it means that the common cancers are very complicated and they are probably making mistakes in many different kinds of proteins which is one of the things that accounts for the fact that one person's cancer is often very different than another person's cancer even if it's the same name, cancer. Even if they both have breast cancer or even if they both have lung cancers. The cells may have made many different mistakes.

GARMON: But is it no less true and I am really asking this, I'm not asking you to say it. Is it true that for no matter how many circuits are involved or you know aberrant proteins, doesn't just the process begin with one mutation in one cell?

DEMETRI: The question—cancer really starts with one cell. One interesting question is whether that cell makes one mutation, but is that one mutation enough to cause cancer or do you need a certain threshold number of mutations? Do you need a certain number of mistakes to lead to cancer? There is good evidence that cells are making mistakes all the time but the body fights them off. The body can recognize that that cell is a mutant and destroys it or if the cell makes the wrong mistake it actually commits suicide and dies. We have so many protective systems in our body to prevent cancer it's remarkable.

GARMON: So I'm not saying that leads to cancer, I'm just saying isn't it accurate to say that the process begins—it has to begin with one mutation, right? Then that mutation couldn't lead to another mistake which could lead to another which could lead to multiple mutations which could lead to no cell death, which could lead to no—so doesn't it always begin though with one thing—one error in your DNA or we don't know?

Back to Top

Targeted Therapies & Smart Drugs

GARMON: So a lot of what I read and a lot of what people say is, yes, okay we haven't had the greatest number of battles that we would like on the war on cancer since Nixon declared the war on cancer, but trust me we are on the brink now. Now everything—all that $48 million a year that your taxpayer dollars went into that investment, trust me, it's going to pay off now. But what I think in my very naive and uneducated way is well, I feel like I heard that when Interleukin was—you know there was a buzz there and Interferon and that was going to you know wash over many different cancers and it ended up not, so I feel like I have lived through the hype. I am old enough to have lived through the hype of different cancer eras and my question is what if the same thing happens with targeted therapies and smart drugs? What if it doesn't help but a couple of cancers?

DEMETRI: You know this is a very important issue of expectations of the public. I think many of us grew up in the era of the war on cancer and when there is a war maybe one day you will wake up and there is a peace treaty and the war is over. That's not the right expectation here. If we are saying there are thousands of different cancers, what is more likely to happen in this era of smart drugs and targeted therapies is that we will see the war won one battle at a time, one type of cancer, one group of patients at a time. I doubt we are going to wake up someday and say breast cancer is gone. We will however wake up and say we have an important benefit for women with Her-2 positive breast cancer or we have an important new drug for a type of breast cancer that really makes those women live longer. That is where the future is going and it may not be as exciting as the public waking up someday and saying cancer is gone, let's move on to diabetes and arthritis, but it's probably more accurate and that's our job as scientists and doctors to explain that to the public. We are on the brink of multiple, wonderful discoveries and wonderful new drugs. I am confident about that.

GARMON: But what if we are not. I mean because I haven't—do you know what I mean? I mean I know this is different than Interferon and Interleukin but as a lay person, I can only tell you I have heard this excitement before and it didn't pan out. What if smart drugs don't pan out? What if they are great for four little you know subclasses of cancers? It doesn't really manage a lot of common cancers?

DEMETRI: Well, no, let's get to again a matter of expectations. I don't think that these drugs will routinely be good for all common cancers. I think it's going to be developing an arsenal of drugs that we then match specifically to individual types of patients. I think that is absolutely inevitable. Everything we know about cancer says it will happen that way. That is exactly what is going to happen. It may not happen tomorrow. It may not happen in two years, but that is the way we are going to fix the cancer problem. It is a very exciting prospect.

GARMON: You are absolutely 100% confident with a scientist's certainty that the concept of smart drugs, solving problems in this way, isn't going to turn out to be the Endostatin of this decade or the limited role that Interferon proved to be able to play in cancer when it was thought it would play a bigger role.

DEMETRI: That's right. I am absolutely confident that the new generation of drugs once we learn to use them for the right patients—that's all key to understand what disease we are treating. If you treat an infection with insulin, you are not going to get anywhere. It's the wrong drug for the wrong disease and in many ways that's where we have been with some of our drugs. We are giving the wrong drugs to the wrong patients and as we test new drugs if we give a PI3 Kinase inhibitor to a patient who doesn't have the PI3 Kinase pathway turned on, it's the wrong drug for the wrong patient. But as we get smarter, as we make better diagnoses and we say this is turned on, this drug shuts it off it has that mathematical certainty that if we match them something great will happen.

Back to Top

Detection vs. Treatment

GARMON: But if you put more research money in early detection you might have the technology for it. I mean what's the—are you mixing up the cart and the horse.

DEMETRI: I don't know that that's true.

GARMON: What if we just put all of our money, next year all $48 million—billion dollars or whatever into early detection? You don't think you would pop out of that year with a better technology for early detection?

DEMETRI: I'm not sure. I am honestly not sure. That is like saying what if you wanted to build a nuclear reactor in 1492. You could put the world's GNP of 1492 into that problem, they wouldn't have done it. The technology wasn't there to do it. So I think this is one about where are we as a society putting our resources for maximal impact and can we learn in the advanced disease setting key lessons that will inform our thinking about prevention. That will inform our thinking about early detection. I will give you an example, one of the most exciting projects we are working on now in patients with metastatic disease is can we develop a more accurate blood test to really predict—this whole field of proteomics is about that. So I think we are mixing and matching because all of us, all of us in our work are trying to get as early as possible because obviously that is going to be the biggest impact. I don't know if that is a fair way of answering the question, but I do feel pretty strongly about that.

GARMON: It's nonetheless—it's a poor relation in the cancer field. I mean there is no question that it is less valued right now.

DEMETRI: I wouldn't say valued. I would say that realistic scientists are successful because they know what they can do. They know—they push the edge of technology, but they know that you know we are not going to try to do the kind of transponders that you see on Star Trek because we don't have the technology. It's good in a TV show, but it's not real right now. You know you go to MIT labs thinking about where we are going to be 50 years and, yes, they are getting there but I think it is also important in science to try to think about the breakthroughs but also work incrementally and we are seeing a lot of that incremental work lead to breakthroughs once in a while.

GARMON: So what is real in early detection? In all cancers, where is there incontrovertible evidence that if you screen or you know early detect a cancer you can decrease mortality? Which ones is it incontrovertible, the evidence?

DEMETRI: I think it's still controversial in many of the big cancers, even in prostate cancer which is one of the most commonly screened ones with the PSA, the prostate specific antigen test. There is still controversy about how to use that rather simple test.

GARMON: But do you have a short list of ones where the NCI or you would agree that—

DEMETRI: I think most of us look at breast cancer and colon cancer as tumors where early detection probably makes a difference probably makes a big difference and where early detection programs work. In other countries, in Asia for example, screening early detection for stomach cancer because stomach cancer is much more common over there, have worked to make people live longer. So I do believe early detection is important and I think in many ways it will be customized as to what you are trying to detect early. It would make no sense to try to detect early very rare disease unless you have a very, very simple test to do it that is painless and cheap.

Back to Top