More than 200,000 women in the United States develop breast cancer each year. Approximately 70 percent of them have estrogen receptor–positive cancer. Many of these women are prescribed tamoxifen following surgical treatment. Tamoxifen is an anti–estrogen drug that almost halves the return of cancer and reduces the mortality rate by one–third in women with early breast cancer. It is so effective that it has achieved status as somewhat of a miracle drug. However, there is a large group of women who have been taking tamoxifen who might as well have been taking a sugar pill. Three Mayo research teams are collaborating to find the best drugs for these patients.

Mayo archives critical to discovery

A gene called CYP2D6 (pronounced sip–2–d–6) encodes for an enzyme involved in the metabolism of up to 25 percent of all drugs. The enzyme is present in different forms in different people — some lack it entirely. Tamoxifen is a "pro–drug," meaning that it is relatively inactive until the liver recruits the CYP2D6 enzyme and converts it into active molecules. Hot flashes are a common and uncomfortable side effect of tamoxifen and, during the past 10 years, physicians have prescribed anti–depressants to relieve them. Scientists now know that anti–depressants inhibit the activity of CYP2D6.

Putting all of these facts together, physician scientist Matthew Goetz, M.D., concluded that women with CYP2D6 variations, as well as those taking anti–depressants with tamoxifen, would have problems metabolizing tamoxifen. He suspected these groups, called poor metabolizers, would have metabolite levels too low to prevent recurrence of the cancer. To prove the hypothesis, he needed large numbers of DNA samples with corresponding clinical data from women treated with tamoxifen alone and from those taking tamoxifen who also took drugs that inhibited CYP2D6.

However, designing a new clinical trial would take at least a decade to produce results. If only some inspired researcher had collected samples and accurate data from the more than 30 years of clinical trials already conducted on tamoxifen. It was a true eureka moment for Dr. Goetz to find out that, not only did such a visionary exist, but that he was right here at Mayo. Breast oncologist and researcher, James Ingle, M.D., had conducted multiple tamoxifen studies to develop best practice guidelines in the 1980s and 1990s, and had archived exactly what was needed — DNA from tumor samples of 250 breast cancer survivors who had each taken tamoxifen for five years following surgery. Dr. Ingle shrugs off his contribution: “I just thought we’d be a lot smarter about gene technology in 10 years, so I archived the tissues. It was fortunate that we designed our studies with the strict eligibility criteria that made them useful for these studies.”

Dr. Goetz was amazed by Dr. Ingle’s foresight.

“People just didn’t save tissues in those days yet he had samples from a population of patients who had taken tamoxifen and now we have more than 15 years of clinical follow–up,” says Dr. Goetz. “Other groups may have had the same idea but it was Mayo who had the required resources.”

Drs. Goetz and Ingle collaborated with others in the Pharmacogenetics Research Network (PGRN), to complete the study, which was first published in The Journal of Clinical Oncology in December 2005 and in Breast Cancer Research and Treatment in 2007. It showed that women treated with tamoxifen who inherited a common genetic change in the CYP2D6 gene or who were co–administered medications that inhibited the CYP2D6 enzyme — the poor metabolizers — had a nearly three fold higher risk of relapse when treated with five years of tamoxifen.

The discovery’s significance is illustrated by the fact it led to an FDA recommendation to change the label of tamoxifen to incorporate the importance of genetic and drug–induced variation in CYP2D6 — the first time the FDA has recommended relabeling for the effectiveness of a drug, not just toxicity warnings.

Proving endoxifen dominance

Dr. Goetz’s interest in the field began as an oncology research fellow under the mentorship of Matthew Ames, Ph.D.

“Dr. Ames’ laboratory provided me with the training essential to understand the pharmacogenetics of drug metabolism,” says Dr. Goetz. “It was at that time that we came up with the idea of studying the metabolism of tamoxifen.”

Initially, the research team suspected a metabolite called 4–hydroxytamoxifen (4–OH tam) as the major player responsible for tamoxifen’s action. A 2003 study, however, revealed very low levels of it in the bloodstream and favored another metabolite, endoxifen, which was present in much higher amounts. The researchers concluded that tamoxifen must be converted by CYP2D6 to endoxifen to have its full effect. The early research demonstrated that like 4–OH tam, endoxifen was 100 times more potent than tamoxifen itself. However, its potential importance was related to simple pharmacology: endoxifen was present at concentrations up to 50 times higher than 4–OH tam.

Nevertheless, many scientists who had studied tamoxifen for many years were reluctant to accept the dominance of endoxifen. Their theory was that tamoxifen was saturating the estrogen receptor making the concentrations of endoxifen irrelevant.

However, when Dr. Goetz published that the rate of relapse or death in the first two years of taking tamoxifen was 32 percent for poor metabolizers compared with just two percent for extensive metabolizers, people began to take notice. There had to be a biological reason for such huge differences between patients with different metabolism.

For an explanation, Drs. Goetz and Ingle turned to molecular biologist Thomas Spelsberg, Ph.D., whose expertise includes hormone resistance in breast cancer and selective estrogen receptor modulators (SERM). His lab is funded to research tamoxifen action, including endoxifen, by the Breast Cancer Research and the Atwater Foundations. Tamoxifen is a SERM. That means it diffuses into the cell, binds to a receptor, and collocates into the nucleus before being activated.

After conducting extensive in vitro tests, Dr. Spelsberg’s lab team, which includes John Hawse, Ph.D., Malayannan Subramaniam, Ph.D., and Xianglin Wu, M.D., Ph.D., discovered that endoxifen has a completely different mechanism of action.

“We showed that endoxifen is extremely potent and is the main compound in tamoxifen metabolism, but we also found that it behaves very differently to 4–OH tam in cell culture,” says Dr. Spelsberg. “When we treated cancer cells with endoxifen, rather than binding to a receptor and co–locating to the nucleus, endoxifen actually degrades the receptor. The genes
regulated by endoxifen are largely different from those regulated by 4–OH tam.”

Dr. Goetz, wanting to understand the clinical relevance, questioned whether the same action would occur even in the presence of high concentrations of tamoxifen, as would be the case with his patients. The question inspired Dr. Spelberg’s lab to design an in vitro model system that has never been done before. They pretreated cancer cells with metabolites in amounts that mirrored compositions found in humans, then added endoxifen. At low concentrations of endoxifen, such as is observed in tamoxifen treated women who are CYP2D6 poor metabolizers, there was no effect on the estrogen receptor and no inhibition of cancer cell growth. But at high concentrations, similar to what extensive metabolizers produce, the receptor was degraded, and cancer cell growth was most potently inhibited.

“That was the evidence we needed,” says Dr. Spelsberg. “It showed that tamoxifen is activated via the CYP2D6 enzyme into a completely different molecule that has a completely different mechanism of action from other SERMs such as tamoxifen and 4–OH tam.”

These new findings suggest that many of the models of tamoxifen resistance developed over the last 30 years may be partly or totally wrong because the importance of the metabolite endoxifen was unrecognized. For example, some of the most highly cited preclinical studies comparing tamoxifen to aromatase inhibitors were performed in mice, which unlike humans do not have the CYP2D6 enzyme.

Developing endoxifen as a drug

The obvious next question is, why not just give patients endoxifen? And that is exactly what Dr. Ames’s lab is working on. They have already conducted animal studies that suggest intravenous administration of endoxifen is safe and effective and are currently testing oral administration.

“It is a compelling argument to sidestep CYP2D6 and give endoxifen directly,” says Dr. Ames. “It is more potent than tamoxifen, it works by a different mechanism, and it would be suitable for all women regardless of their genetic makeup.”

Dr. Ames has conducted pharmacology studies for cancer drugs developed by the National Cancer Institute. Recently, his laboratory played a key role in another study for which Dr. Goetz was awarded NCI funding.

“Following the release of our data in 2006, we began to test patients for the genetic change in the CYP2D6 gene,” says Dr. Goetz. “However, there remained some question as to how to use this information in the new era of the potent third generation aromatase inhibitors, drugs which were tested and demonstrated to be superior to tamoxifen.”

To answer this question, the Goetz team collaborated with investigators at the Austrian Breast and Colorectal Study Group in Vienna, who had conducted a large clinical trial in which breast cancer patients were randomized to either tamoxifen for 5 years, or tamoxifen for 2 years followed by a switch to an aromatase inhibitor. After many months of refining the techniques to genotype for CYP2D6 from paraffin embedded tumor samples, the Ames’ laboratory successfully genotyped over 600 samples for seven CYP2D6 allelic variants. On December 12, 2008, Dr. Goetz presented his findings from this study at the San Antonio Breast Cancer Symposium demonstrating that CYP2D6 poor metabolizers had a nearly 4–fold higher risk of early breast cancer recurrence compared to CYP2D6 extensive metabolizers.

One of the most intriguing findings related to the patients who switched to anastrozole, was no increased risk of breast cancer recurrence for CYP2D6 poor metabolizers in years three through five.

“This completely and totally confirmed our hypothesis because CYP2D6 is not involved in the metabolism of anastrozole,” says Dr. Goetz. “By definition CYP2D6 poor metabolizers should do better on anastrozole than tamoxifen. Poor metabolizers who were fortunate to not develop breast cancer recurrence in the first two years of tamoxifen appear to be rescued by anastrozole.”

The goal of pharmacogenomics is to figure out the best drug for every patient. Next on the agenda is to better understand the molecular mechanisms that rid the body of endoxifen. Are there environmental factors that affect it? How does a person’s genetic makeup affect the rate of removal? The Mayo research teams have been instrumental in being able to identify and prescribe drugs that are compatible with the genetic makeup of the more than 140,000 women in the United States who develop estrogen receptor–positive breast cancer, and so many more worldwide.

Mayo Clinic has a potential financial interest in technology related to this research.

— Yvonne Hubmayr, December 2008