Back in 1970, I had decided to look at the structure-activity relationships of anti-estrogens. At the time, these compounds had been tested as anti-fertility agents in laboratory animals, but if anything they increased fertility. Nobody was really interested in them. I wanted to work on them as anticancer agents. When I had my Ph.D. examination in 1972, the examiner was Dr. Arthur Walpole, who worked for industry and was the patent holder of a compound that I had been reading about called ICI 46,474. When I came to America as a visiting scientist at the Worcester Foundation for Experimental Biology in Massachusetts, I was allowed to do anything I wanted. I got there in September 1972 and rang up Arthur Walpole and said, let’s turn ICI 46,474 into a breast cancer drug. He said, yes, he’d give me the opportunity and resources to do that. This drug company, now called AstraZeneca, gave me the resources and the technicians to do the first systematic study of the compound which is now called tamoxifen. That study demonstrated that tamoxifen prevented mammary cancer in rats, and became a pivotal publication. And that’s really how I got involved with tamoxifen and that’s what I’ve spent the last 30 years doing—studying anti-estrogens and breast cancer.

It was known for most of the 20^th century that one in three premenopausal women who have advanced breast cancer, metastatic breast cancer throughout their body, will respond to ovarian oblation—removing their ovaries, the source of estrogen. The whole idea is that you are withdrawing the fuel for the fire, if you like, by having no estrogen around in the body. This was known in clinical practice, but nobody really knew why it was. Then a scientist called Elwood Jensen discovered the estrogen receptor in the 1950s and in 1962, he really described the target cite specificity of estrogen around an animal body. Estrogen bound in the uterus and the vagina. These were estrogen target tissues. But it didn’t bind in heart and muscles, which weren’t estrogen target tissues. He came up with the idea that the estrogen receptor was necessary to cause estrogen-mediated events in the tissue. He also said, let’s take this one step further: if we measure the estrogen receptor in breast tumors, maybe the tumors that respond to estrogen being taken away are the estrogen-dependent tumors. So this became the start of the estrogen receptor test. You could now predict which patients would respond to an ovariectomy. This was the big advance in the 1960s. So I put all these little bits and pieces together and it suggested that an anti-estrogen that blocks the estrogen receptor will do the same thing.

Here’s an important thing and a good story: it gets involved with not only tamoxifen but also another anti-estrogen called raloxifene. In the mid-1980s, people were thinking about using tamoxifen for the prevention of breast cancer; the drug had been used since the early 1970s for breast cancer treatment. One of the concerns that I had about recommending long-term tamoxifen therapy was that tamoxifen did block estrogen action. But if estrogen is going to be good for bones and coronary heart disease, which is what people believe, maybe if we give anti-estrogens for a long time to women who are healthy, we might prevent breast cancer, but they’ll all drop dead from coronary heart disease or osteoporosis five or ten years later. You win on breast cancer, but you lose on two other diseases. So when I was at the University of Wisconsin in the mid-1980s, we did what I think is the pivotal experiment – and it’s important to mention that we did all this in animals first. Of course, it’s the clinical publication that gets all the publicity and citations because this is the way the world is made. But the paper that changed everything was published in Breast Cancer Research and Treatment in 1987 and it was called "The effects of anti-estrogens on bone in castrated and intact female rats" (Breast Cancer Research and Treatment, 10[1]: 31-5, 1987). What we found was that tamoxifen and raloxifene both maintain bone density in animals that have had their ovaries removed. You take the estrogen away and the bone density goes down. But if you take the estrogen away and treat with tamoxifen or raloxifene, you maintain bone density. We got very excited about this. We said, look, we’ve discovered selective estrogen receptor modulation, although at the time we called it target-site specificity of anti-estrogens. In one tissue, like mammary tissue, these compounds acted like anti-estrogens. But in bone, these compounds worked as estrogen. Then at Wisconsin we set up a clinical trial to see whether tamoxifen would harm bone density in women and found it maintained bone density, just like we saw in rats. That then became an important observation and the paper that’s now so highly cited.

Yes, that paper came out of the same clinical trial. We measured blood cholesterol in these women, and we found tamoxifen also does no harm to their risk of coronary heart disease. But here’s where the importance of these publications comes out. Based on the 1987 publication on animals, and seeing the way drug development was going with tamoxifen, and seeing that tamoxifen was being targeted specifically to high-risk women with breast cancer, in 1990 we came out with a paper that altered the whole of the drug development of these selective estrogen receptor modulators, known as SERMs. This was Lerner and Jordan, "Development of anti-estrogens and their use in breast cancer" (Cancer Research, 50[14]: 4177-89, 15 July 1990). In this paper, we decided to predict where we thought drug development was going for the future. We basically said, well, half the women who develop breast cancer aren’t high-risk women, they’re normal-risk women whose only risk factor is that they are getting older. These women are never going to get tamoxifen. What do we do for these women? Here’s the idea: just turn over the coin. If tamoxifen prevents breast cancer and is good for bones, why not develop a drug for osteoporosis that prevents breast cancer as a positive side effect. That’s where raloxifene comes in. It’s the first drug used for osteoporosis that is now being tested against tamoxifen for the prevention of breast cancer. These two compounds, tamoxifen and raloxifene, turned out to be different sides of the same coin. And raloxifene has the same beneficial effect on blood lipids. Both lower LDL, the bad cholesterol, and that’s supposed to be good for preventing coronary heart disease. Raloxifene is being tested at the moment in women with high cholesterol and high risk of coronary heart disease to see if we can protect them from developing coronary heart disease.

In the early 1970s, although there was enthusiasm everywhere for the conquest of cancer, there was general agreement that it was chemotherapy that was going to destroy cancer cells and cure cancer. Anti-hormones like tamoxifen were assumed to be complete nonstarters. So the challenge was overcoming the skepticism in the clinical community that these particular agents were going to be of any use as therapeutic agents. I think the second challenge has been really with the interdisciplinary aspects of SERMs. For example, that 1987 paper looking at tamoxifen and raloxifene on laboratory animals is never cited by the osteoporosis people, even though it’s in the refereed literature.

Well, for example, I’m a member of the American Association of Cancer Research, so when it comes to my scientific papers I prefer to publish in my professional journals: in particular, Clinical Cancer Research and Cancer Research. I want to publish once or twice a year in those journals. If I’m working in conjunction with clinicians and they want to submit to the New England Journal of Medicine, then we’ll do that. Although I’m normally not in that loop. Because I was educated in Britain, I also publish in some journals in the United Kingdom and Europe. I look at my research program as being global rather than parochial. So I will publish in the British Journal of Cancer or the European Journal of Cancer. It is based on that sort of logic. These are good European journals and they often ask me to write editorials or commentaries. The same would be true for the Journal of the National Cancer Institute. I have two or three articles that I’m very proud of in the JNCI.

What I have always done is tackle a problem based on what I consider to be a sound hypothesis, but I’m trained as a pharmacologist, so I want to develop concepts and medicines for treating human beings. Everything I have done in my career is research in the laboratory that has been targeted specifically to how I can help to improve medicine for cancer or osteoporosis or other diseases. My general message is that all my work has been specifically targeted to help alleviate human diseases, but using animal models and laboratory concepts to achieve that aim. What I’ve always wanted to do is look at the good, the bad, and the ugly for these drugs—like the old Clint Eastwood movie.

This work now gives us the possibility of multi-functional medicines. You may be able to develop a medicine that maintains bone density, prevents breast and endometrial cancers, and lowers the risk of heart disease. Now what we have to do is marry molecular biology and medicine and then marshal together all the mechanisms of how these drugs work. It’s what I called, in an article in Scientific American, making designer estrogens ("Designer estrogens," Scientific American, 279[4]: 60-7, October 1998). Now we have tremendous clues in this area and we can go off in lots of different directions to develop drugs to treat a whole host of different diseases associated with menopause.

Dr. V. Craig Jordan
Northwestern University School of Medicine
Robert H. Lurie Comprehensive Cancer Center
Lynn Sage Breast Cancer Research Program
Chicago, IL, USA