Until 1986, Premarin wasn’t such a big deal. Isolated from pregnant mares, the complex mixture now known to contain more than 50 estrogens was introduced in 1942 by Wyeth Ayerst (now a division of American Home Products, based in Philadelphia) as a treatment for hot flashes and other symptoms of menopause. It was a successful product for the manufacturer, but its best days were yet to come.
In 1986, the FDA announced that Premarin and other short-term-acting estrogens were effective in combating the bone loss associated with osteoporosis. Suddenly, a drug that women were taking over the short term was the treatment of choice for a long-term, chronic problem.
“There was not a large market for anyone to do anything about until Premarin was granted the osteoporosis indication,” says Kenneth Phelps, vice president of clinical operations at Duramed Pharmaceuticals, Inc. (Cincinnati, OH). Duramed had a generic version of Premarin, known simply as conjugated estrogens USP, which it hoped to exploit in this newfound niche. “The market burst open, and Premarin seized the day. [There was talk that] literally every postmenopausal woman should take Premarin for the rest of her life,” Phelps says.
The new indication promised a staggering market, but it also touched off a scientific debate. Suddenly, the FDA was faced with a witches’ brew product that had been demonstrated safe and effective for short-term use but was otherwise poorly understood. Initial testing showed that Premarin tablets followed a slow-release bioavailability pattern, while the existing generics were immediate release. “They said, ‘whoops, we can’t have that because there could be a lot [of unknown long-term ramifications of different modes of release],’” says Phelps.
“We had certified that the generics were interchangeable with Premarin, but when we looked at it from the point of view of the science, that was not the case,” says Janet Woodcock, director of the FDA’s Center for Drug Evaluation and Research (CDER).
The agency immediately set about determining the bioavailability of Premarin and its pharmaceutical properties. The results led to a long struggle between members of the generic industry, Wyeth Ayerst, and the FDA. Women’s advocacy groups got in on the act too, because the battle’s outcome would determine the future alternatives to a drug whose sales to American women exceeded $1.5 billion in 1999.
A different standard
When Premarin was introduced in 1942, federal law required Wyeth Ayerst to prove only its safety, not its efficacy. But in response to the 1962 amendments to the Food, Drug, and Cosmetic Act, the FDA examined existing products. In 1972, the agency published a Federal Register notice announcing that Premarin and other estrogen products had been effective in the treatment of menopausal symptoms. The agency also decided that two estrogens—estrone sulfate and equilin sulfate—were primarily responsible for Premarin’s activity, and it laid the groundwork for Abbreviated New Drug Application (ANDA) submissions of generic versions.
But the scrutiny of the 1980s painted a different picture. The immediate finding was a disparity in bioavailability between Premarin and its generic versions, but the complexity of the mixture—Premarin may contain 50 or more estrogens with some level of biological activity—soon cast doubt on whether the generic versions of the drug were true pharmaceutical equivalents.
The FDA conducted detailed analyses of the blood and urine concentrations of estrogens in women taking Premarin and eventually concluded that the synthetic generics then on the market could no longer be considered generics.
In 1991, the FDA dropped the bomb and withdrew approval of all ANDAs for generic estrogens, leaving Premarin once more alone on the market. That left companies such as Barr Laboratories and Duramed, which had been producing generics of the drug, out in the cold. In 1992, the U.S. Pharmacopeia released a monograph stipulating that three additional estrogens must be included in any generic version.
The lure of a share of Premarin’s huge market drove Duramed and Barr to try again, and both companies submitted new ANDAs in 1994 and 1995, respectively. “[We knew] it was going to be a difficult accomplishment to get a product that would be bioequivalent to Premarin. It was costly and high-risk, and Barr and Duramed were the only two combatants that were interested to give it a shot,” says Duramed’s Phelps.
The good news was that the ANDA required only the demonstration of bioequivalence and adequate manufacturing controls, and no additional evidence of efficacy.
The bad news was that more roadblocks lay ahead. By 1995, Barr Laboratories had developed a synthetic that included estrone sulfate and equilin sulfate and submitted an ANDA in July of that year. But in November 1994, Wyeth Ayerst—which had been conducting its own research into the components of Premarin—filed a citizens’ petition with the FDA, claiming that it had found another critical component of Premarin: delta-8,9-dehydroestrone sulfate (DHES). The company had been working on the project even before the FDA asked it to characterize Premarin’s ingredients in the early 1990s, says Mike Dey, president of ESI Lederle (St. Davids, PA), a division of Wyeth Ayerst, whose parent company is American Home Products.
“We were concerned that the agency would apply a different standard for conjugated estrogens than they required for other generics, because of the complexity of the product . . . that they would look at resolving this by lowering the standards as opposed to working through the difficult scientific questions to make sure the standards were consistent across all products,” says Dey. Osteoporosis is a particular problem because it is a so-called silent disease that doesn’t reveal itself until the onset of bone weakness. “Because it is silent, women are unaware if the therapy isn’t working until their bones fracture. If the agency were to relax its standards, women could no longer be sure they were getting the efficacy needed to protect them from osteoporosis,” says Ley.
The company had collected preclinical evidence and conducted two clinical trials that showed DHES had a unique clinical profile. The FDA took the new information under consideration, and in July 1995, the agency’s Fertility and Maternal Health Drugs Advisory Committee declared that no one had demonstrated whether estrone sulfate and equilin sulfate were sufficient to account for Premarin’s activity. The agency then took an unusual stance by deciding to put the burden on generics manufacturers to prove that their products were pharmaceutically equivalent to Premarin.
By 1997, “FDA again changed its position and said only a natural product will do,” says Bruce L. Downey, chairman and CEO of Barr Laboratories.
That position was outlined in a May 1997 memo. Still, the FDA didn’t come to that decision lightly. A philosophical debate raged within the organization, according to Roger L. Williams, who was the director of the Office of Generic Drugs in the early 1990s and later a deputy director of the CDER from 1994 until this year, when he was named CEO and executive vice president of U.S. Pharmacopeia.
“[The agency] concluded that . . . the burden falls on the generic [manufacturer] to demonstrate pharmaceutical equivalence,” says Williams. “I would argue that the pioneer always has the burden of proof, that whatever is in there is what’s creating safety and efficacy. Part of the problem was that Premarin was approved at a time when (the government) only required proof of safety, not efficacy. When FDA upgraded Premarin’s efficacy status in 1971, the agency concluded that two primary ingredients created the efficacy.”
Williams couldn’t persuade his colleagues. “So in some respects the burden is on the generic manufacturer if they wish to leave things out (of a generic), but we won’t reduce that to absurdity and require trace contaminants,” says Woodcock. “But . . . there is an issue of sheer potency, because you might have small amounts of 10 or 20 estrogens and you have to add them up to have the same estrogenic potency as your innovator product.”
A tall order for generic producers? Woodcock admits that it is. “We said in the memo that we thought it was more plausible that a naturally derived estrogen product could be found to be pharmaceutically equivalent to Premarin—it’s hard to imagine a synthetic one,” she says.
Forging ahead
That pronouncement effectively ended the generic scramble, at least for the moment. In March of this year, the FDA released a draft guidance document for chemical analysis of Premarin, which could spark another round of generics (www.fda.gov/cder/guidance/3668dft.pdf). Still, Williams thinks the prospects for a Premarin generic are bleak. “There are a lot of products available to treat vasomotor symptoms of menopause. . . . I think people are trying to develop natural-product generics [isolated from a different source than Premarin], but it will be difficult.”
Duramed decided to push on, filing an ANDA for its product—dubbed Cenestin—that was eventually approved in March 1999 for the treatment of hot flashes. It doesn’t have the status of a generic, however, which dampens its prospects. Still, Phelps is optimistic. Cenestin is based on a controlled-release delivery system that he believes is superior to the bioavailability properties of Premarin and could reduce side effects such as breast soreness. “Perhaps with Cenestin’s more favorable blood level profile, women might find better tolerance of our drug,” he says, noting that women often stop taking Premarin because of side effects such as breast tenderness, bleeding, or concern about developing breast cancer.
Not surprisingly, Downey is irritated at the outcome. “I can think of no other case where the standards were changed in this way, and this happened twice. I think that American Home Products has paid an extraordinary amount of attention in the political arena and have armed themselves with an extremely gifted legal and lobbying talent. It was a triumph of politics over science,” he says. Nevertheless, Barr continues research on synthetics in hopes that the FDA will change its position, as well as on alternative natural sources that may have a better shot at being considered as a generic.
Duramed’s Phelps is more sanguine. “As a stockholder you feel that FDA pulled the rug out from under us. Is that fair? There’s a sense that you invested based on the best advice of the government, and then you lost because of the process. But as a scientist, you learn something every day . . . to fix standards is impossible. The best understanding holds until you learn something new. From a scientific point of view, Wyeth Ayerst took advantage of that [by continuing to search for new active elements and challenging approvals],” he says.
Woodcock is sympathetic but firm. “I know there’s some bitterness because the community went along for a long time [developing generic candidates] from an existing monograph [the 1972 document]. Meanwhile, the science had changed quite a bit, and our knowledge of estrogens and receptors increased. . . . That was not fair to the people who had developed a new product based on the monograph,” she admits.
Lessons learned
As with all facets of drug discovery, the science marches on. And although the controversy over Premarin has been painful for both the industry and the FDA, it may yet bear fruit.
Estrogens have applications far beyond relieving the symptoms of menopause and osteoporosis—studies suggest protective effects against cardiovascular disease, Alzheimer’s disease, and other conditions. And, a more intricate understanding of estrogens could assist in the fight against estrogen-responsive forms of breast cancer.
In 1995, researchers discovered a second estrogen receptor in rats, ER-b. Its physiological role hasn’t been pinpointed, according to Dey, but it generally resides in different tissues than ER-a.
“Some researchers believe they can modulate the effects of one over the other, while others believe that they are distinct receptors that function independently . . . if we can establish selectivity for the beta receptor, we can compare what genes are turned on compared to the alpha receptor. Many of the components of Premarin have different binding affinities to the two receptors,” says Dey.
That suggests that all of the hand-wringing over a Premarin generic may not have been entirely in vain. The efforts to tease out compounds and understand the biological activities of myriad components of the witches’ brew could pay big dividends as the significance of the two receptors in disease becomes increasingly clear. “If we can find a molecule or molecules that are selective for alpha or beta and can enhance the efficacy and minimize the side effects, then that would be a very exciting outcome of all of this,” says Dey.
Other osteoporosis drug therapies
Althought the time-honored treatment to help prevent and alleviate postmenopausal osteoporosis is estrogen replacement therapy, questions about the hormoneÕs effects on other body systems have led to the development of other drugs. Among these are selective estrogen receptor modulators (SERMs), one of which, raloxifene, has recently been approved by the FDA for the prevention of osteoporosis. In 1995, alendronate, a bisphosphonate compound, became the first nonhormonal prescription medicine used to treat osteoporosis in postmenopausal women. By helping to inhibit bone breakdown, the drug leads to a cumulative increase in bone density. In recent clinical trials, risedronate, another bisphosphonate compound, reduced by up to 58% the risk of hip fracture in women with low bone mineral density. As their principal side effect, bisphosphonate compounds can cause digestive disorders.
