As published in Hai Lights:
[Editor's note: On 29 October 1998, the US Food and Drug Administration approved the drug tamoxifen for use in reducing the short-term incidence of breast cancer in women deemed at high risk for the disease. This action broke new ground in prescribing potent medicines for the prevention of a possible, future illness. Below author Sharon Batt takes readers behind the scenes to the FDA committee that considered the evidence and explains how this drug with serious, known risks for women's health gained approval.]
On September 2, I witnessed a pivotal moment in the history of women's health. An expert panel advising the American drug regulatory agency, the Food and Drug Administration (FDA), reviewed a request to approve the drug tamoxifen as a safe, effective way to prevent breast cancer. The application was filed by Zeneca Pharmaceuticals, manufacturer of Nolvadex® and the sole company with the right to sell tamoxifen in the US.
After hearing some four hours of evidence, pro and con, the committee's eleven members wrestled with the fact that a clinical trial testing tamoxifen on healthy women had not proven the drug to be either a safe or an effective means of preventing breast cancer. Nonetheless, the members voted, in full view of the packed hearing room, to recommend FDA approval of the company's application (nine yeas, two abstentions). The main concession to reality was the recommendation that Zeneca only be allowed to promote the drug for "short-term risk reduction" of breast cancer, not for "prevention"--a subtle distinction which fell by the wayside even as accounts of the committee's decision surfaced in the next morning's news.
The hearing was the most recent plot twist in breast cancer's "chemoprevention" drama, that is, the effort to stop breast cancer with a pill. For a six-year period which ended last April, most of the activity has been offstage. If the sudden flurry of front-and-centre action caught you off guard, you're not alone. But keep your eye on this story. Tamoxifen is a women's health parable for the '90s.
Chemoprevention on trial
Seven years ago, the National Cancer Institute announced its plan to fund a North America-wide clinical trial involving 16,000 healthy women, defined as at higher-than-average risk for breast cancer. Half would be given tamoxifen, the other half a placebo. The budget was impressive: US$68 million, not counting the cost of the drug, which would be supplied by Zeneca. (Modifications to the trial later reduced the price-tag to US$60 million and recruitment eventually stopped at 13,388). Researchers in the UK, Italy and Australia-New Zealand mounted similar trials about the same time.
The National Women's Health Network (NWHN) intervened to try and stop the North American study. The Network was formed 25 years ago, in response to the reproductive issue crisis. Advocates of a science-based approach to women's health, its directors saw the plan to give a potent drug to healthy women as both scientifically questionable and a dangerous precedent in medicalising disease prevention.
Tamoxifen as treatment
Tamoxifen has been used in breast cancer treatment for twenty years, with some success. Women who have already had cancer in one breast are at high risk of developing cancer in their other breast. In post-menopausal women with estrogen-sensitive tumours, taking tamoxifen for up to five years reduces the risk of breast cancer in the opposite breast by 47%.
The drug's downside, which includes an increased risk of endometrial cancer, blood clots, cataracts, and a host of "minor" effects such as hot flashes, depression, weight gain and reduced libido, is par for the course in a cancer treatment. In fact, tamoxifen is one of the more user-friendly and effective breast cancer treatments. Doctors who treat breast cancer regard the drug highly.
Disease substitution vs. disease prevention
Physician Adriane Fugh-Berman of the NWHN's board agreed: "The risk/benefit ratio for women with a life-threatening cancer definitely favors tamoxifen." Where she parted company with the prevention trial's advocates was in giving the same drug to healthy women. "Acceptable risks for the sick are not acceptable risks for the well," she wrote in a 1991 critique of the trial. Cancer specialists tend to have a skewed perspective of safety, she noted, because they are so accustomed to slash-and-burn treatments. In preventive medicine the norms are entirely different: risks should be vanishingly small, as in vaccination or vitamin-D enriched milk. Trading a lowered risk of breast cancer against an increased risk of endometrial cancer, she argued, amounted to disease substitution not disease prevention.
The trial's proponents countered that the women being recruited were not actually healthy--they were at high risk of breast cancer. The verbal maneuver reminds one of President Clinton's attempts to redefine sex. Eligibility criteria for the North American trial defined "high risk" as that of an average sixty-year-old woman, specifically, a 1.66 percent chance of developing breast cancer during the next five years of her life. Younger women, from age 35 upwards, could enter if additional factors, such as a family history of breast cancer or a personal history of breast biopsies, elevated their statistical risk to the age-60 level.
The trial results
When the NWHN failed to rally enough support to stop the North American trial, public debate over tamoxifen as a preventive died down. For most of the last six years, trialists here and abroad have gone about the business of accruing subjects in relative quiet. (One notable exception was the trial's suspension, in 1994, to revise the consent forms because they failed to mention that tamoxifen carries a risk of potentially fatal endometrial cancers.)
Last April, in a bolt from on high, the North American researchers announced that they had halted their study more than a year before the scheduled date because the benefits of tamoxifen were so striking they could no longer justify denying the drug to the women in the placebo group. The announcement made banner headlines worldwide. After an average of about four years on tamoxifen, 85 women in the treatment group had developed breast cancer, compared to 154 women in the placebo group. Without a doubt, tamoxifen had reduced the number of breast cancer cases.
The results also confirmed the prediction of "disease substitution," however. Women who took tamoxifen had significantly more endometrial cancers, blood clots and cataracts. The tally in deaths from either breast cancer or a complication linked to tamoxifen was five to six: five women in the control group died of breast cancer, compared to three deaths from breast cancer and three from blood clots in the lung in the tamoxifen group. Among the less serious side effects, women on tamoxifen were more likely to suffer from uncomfortable hot flashes and vaginal discharge.
Whether the drug had actually prevented breast cancer was not clear. An equally plausible interpretation was that the onset of small, undetected cancers had been delayed. If this were the case, the women may have gained no greater advantage than if they were treated after a diagnosis. Or, if the tamoxifen does prevent cancers over the long haul, the survival benefit may be less striking than the reduction in incidence, because the tumours that respond to tamoxifen tend to be more readily treatable. Only long-term follow-up could answer these questions.
Get ready for STAR
This trial won't give us that long-term data, however. At the same time the researchers declared victory, they announced that they had offered women in the control group the choice of taking tamoxifen. Their rationale was that withholding the drug would be unethical, given the effect demonstrated. The trialists went further: they encouraged women in the control group to volunteer for another chemoprevention trial set for launch this fall.
Dubbed the "Study of Tamoxifen Against Raloxifene"or STAR, the new trial will compare tamoxifen with a pharmacological cousin, developed by Eli Lilly and currently approved for prevention of osteoporosis. If (as seems likely) many of the placebo-group women opt either to take tamoxifen or to enter the new trial, long-term data for the original trial will be useless.
Europe: a different story
To confuse the picture even more, researchers in Great Britain and Italy rushed early results of their own tamoxifen prevention trials to press this summer and neither study showed evidence of fewer cancers from tamoxifen. When carefully examined, the contradiction is not as baffling as appears at first blush. The North American study showed clearly that tamoxifen does not reduce the incidence of estrogen-negative tumours. Younger women are more apt to have estrogen-negative tumours and both European trials had more younger women than the North American trials.
Furthermore, all the women in the British study had a family history of breast cancer. Although family history was not rigorously defined, the British results raise a red flag about women with a genetic predisposition to breast cancer. With the advent of genetic testing, women with a mutated BRCA gene are a well-defined group that would be highly motivated to take a drug that could prevent breast cancer (many resort to prophylactic mastectomies); yet they are more apt to have estrogen-negative tumours and tamoxifen may not lower their risk of breast cancer.
In contrast to the other two studies, women in the Italian trial were not selected to be a high risk. About 26 % dropped out in the first year, weakening the power of the study and perhaps contributing to the negative result. On the positive side, the high drop-out rate suggests that women at low risk have a sensible aversion to tolerating unpleasant drug side effects in the hopes of preventing breast cancer.
A final difference which has been noted between the North American and European studies is the differential use of hormone-replacement [HRT] drugs, which could interact with tamoxifen. North American women were instructed not to take HRT during the trial. In the UK study, 41% of the women took HRT; in Italy 14% did so.
The message from Europe is that we still don't know enough to give women helpful advice about tamoxifen for chemoprevention. The British intend to continue their trial for many more years and, along with the Australians, may turn up some answers about long-term effects (the Italians stopped their trial because of the high drop-out rate). Researchers in the UK, still recruiting volunteers, made no secret of their chagrin when their North American colleagues declared the whole question history. The senior researcher from the British team, Trevor Powles, testified at the FDA hearing and was visibly frustrated that the Americans had so cheerfully closed the door on the long-term results of their own trial and thrown away the key.
The North American researchers, rather than worrying about tamoxifen's inadequacies, simply admit the drug is wanting. They defend the tamoxifen prevention trial as a necessary first step, not an end in itself, and describe tamoxifen as the precursor of better, smarter hormone pills.
The SERMs vision
Tamoxifen is gaining fame as the first of an emerging class of drugs known in the trade as Selective Estrogen-Receptor Modulators (SERMs). SERMs-enthusiasts envision a medical equivalent of hormonal smart-bombs, designer pills which will manipulate women's hormonal systems to put more estrogen here and less there, rolling together all the benefits of birth control, breast cancer prevention, and hormone replacement--with none of the nasty drawbacks. SERMs arose from the accidental discovery that tamoxifen blocks estrogen in certain parts of the body but stimulates the same hormone in other places.
Researchers realised that estrogen-receptors, the switches controlling whether or not estrogen enters a cell, are different at each of the many places in the body that use estrogen. Conceivably, estrogenic drugs could be designed to block estrogen's access to some sites, but enter cells in places that benefit from estrogen. The ultimate hormonal designer drug would inhibit cancers in the reproductive organs, strengthen aging bones and lend suppleness to arteries.
Raloxifene, or Evista, Eli Lilly's entry in the breast cancer chemoprevention stakes, is a post-tamoxifen SERM. Currently being tested in clinical trials as a hormone-replacement therapy, the only results available are very preliminary. Lilly has nonetheless managed to create enough buzz about raloxifene to suggest the drug could trump both Premarin as a hormone replacement drug and tamoxifen as a chemopreventive for breast cancer. Raloxifene's promise is based largely on the two-year results of a study in which women on raloxifene had a 50 % reduced incidence of breast cancer compared to women taking the placebo. The company has highlighted this claim, based on a mere 35 cases of cancer in a trial of some 6,000 women, while downplaying the finding that raloxifene produced ovarian cancer in mice and rats.
In other words, raloxifene's claim to greatness as a breast cancer-preventive is as shaky as tamoxifen's. The so-called STAR trial (the acronym is more redolent of advertising hype than of scientific inquiry) will pit the ambitious rookie against the flawed veteran. Can we really believe women's health will be the winner?
The FDA hearing
At the FDA hearing, representatives from Zeneca and senior members of the research team presented their data, putting the results in the best possible light. Also arguing for approval of Zeneca's application was a woman from the Cancer Research Foundation of America, who acknowledged (as required the rules) that her organisation had received funding from Zeneca. Dr. Trevor Powles, from the British research project, answered questions about his own trial and made clear his distress about the North American trial. Fifteen years of follow-up would be needed to know if the effect was truly prevention, he said, pointedly. If the North American trial hadn't been unblinded, with all the trials together "we could have had useful data by 2005".
I was one of eight intervenants who urged that tamoxifen not be approved as a preventive for breast cancer. We echoed each others' concerns that the trial had shown short-term reduction in the risk of breast cancer, but hadn't demonstrated prevention. We reminded the committee of the risks (Ann Fonfa from New York opened her presentation with a taped voice from the grave saying, "Thank you for saving me from breast cancer, too bad I died of a pulmonary embolism.") We expressed our fears of exaggerated advertising claims targetting busy GPs eager to believe in a magic pill and frightened women eager to take one.
It's about advertising, not access
American advocates had briefed me on what approval of the company's application would mean. The issue was not, they explained, one of access to the drug for women at high risk of breast cancer. That right exists now. Because tamoxifen is already approved as a breast cancer treatment, any physician can prescribe the drug "off label" to a patient he or she believes might benefit from it. Given the results of the trial, such a prescription would not be an unreasonable decision for either patient or physician, provided that they both were aware of the drug's limitations and risks, and understood the need for monitoring.
Approving the drug for preventive use would mean that the drug company could advertise the product as proven effective for prevention, with a risk/benefit ratio favouring safety. Given the huge prospective market of healthy women (29 million Americans, if you restrict the numbers to those who meet the eligibility criteria of the trial), one could expect marketing to be intense. American law allows direct-to-consumer advertising, which is currently illegal in Canada, but high on the drug industry's wish list.
A subtle shift
As the eleven committee members began their deliberations, they seemed sobered by their responsibility. On the first question, whether the trial showed the drug was effective in preventing breast cancer, they voiced their concerns about the word "prevention", and finally settled on a rewording: the drug "reduced the short-term incidence of breast cancer." The next question, whether the trial showed a favourable risk-benefit ratio for the prevention of breast cancer, also caused the committee difficulty. Having jettisoned the word "prevention", the benefits were nebulous, and the risks, for some women, were obviously considerable. The vote was negative. By now the company reps were slumping glumly in their seats and the advocates were exchanging wide-eyed glances.
A subset to the risk/benefit question was whether the committee could identify a subpopulation with a favourable risk/benefit ratio. Again they had problems. The data don't point to any identifiable subset of women who would clearly benefit from taking the drug. As its members pondered this, the committee's mood changed, as if turning on an invisible pivot. One member said it would be a shame if the application was rejected and the researchers looked bad, because their trial was excellent and had yielded a wealth of data. Another member mentioned the 13,000 "courageous volunteers". The risk/benefit ratio was found to be favourable after all, and the committee voted to approve the application. The rest of the discussion slipped to damage control, via package inserts and working with the company to ensure "an extra educational effort".
The FDA has promised to rule on the tamoxifen-for-prevention application by November, and the agency usually accepts an advisory committee's recommendations. From where I sat, the committee members completely abdicated their responsibility to protect the public. No one I spoke to, however, expected anything different. This was strange, given that the procedure had all the appearance of a model exercise in democracy. Those of us arguing for a public safety decision were able to speak our minds, we were heard (committee members frequently quoted our arguments) but we lost all but a tiny corner of turf: the veto on the word "prevention". What happened?
In retrospect, the wide-open American hearing was deceptive. Behind the scenes, the drug company, the researchers and the regulators work in close contact (Zeneca's senior medical director opened his presentation by acknowledging his "colleagues" on the research team and at the National Cancer Institute). If the FDA blocked approval of a drug after the National Cancer Institute had invested US$60 million in a high-profile trial to test it, some career fur would certainly fly.
A milestone...on what road?
The media are complicit as well. Americans who sat down the next day with their morning coffee and their daily papers read an account of health policy moving smartly ahead. The Chicago Tribune stated that the panel had "urged the Food and Drug Administration to widen its war against breast cancer by approving a drug to prevent the disease...". The front page New York Times story said that if the FDA accepted the committee's recommendation, "...approval would be a milestone in efforts to prevent cancer."
A milestone, perhaps, but not for advocates of cancer prevention. As we head into a new millennium and the impending bombardment of hype about SERMs, I'll keep a bottle of old fashioned skepticism next to my bed.
Sharon Batt is author of Patient No More: The Politics of Breast Cancer, published by Gynergy Books and A New Look at Breast Cancer--Beyond Early Detection, published by DES Action Canada. She is director of policy and research for Breast Cancer Action Montreal.
This article originally appeared in the DES Action Canada newsletter. It is reprinted with permission.
Ann is interviewed at the hearing
|Remember we are NOT Doctors and have NO medical training.|
This site is like an Encyclopedia - there are many pages, many links on many topics.
Support our work with any size DONATION - see left side of any page - for how to donate. You can help raise awareness of CAM.