Clinical Applications of Mifepristone (RU486) and Other Antiprogestins: Assessing the Science and Recommending a Research Agenda (1993)

DAVID GRIMES, M.D., Professor and Vice Chair,

Department of Obstetrics, Gynecology, and Reproductive Sciences, University of California, San Francisco

I would like to review with you what we have heard this morning and what I think are some of the key points. I will begin with a discussion of the use of mifepristone (RU 486) for first-trimester abortion.

The first scientific issue relates to the implied danger of mifepristone. A recurrent theme in review articles to date is that mifepristone for abortion requires ''strict medical supervision" and that hospital facilities are required. There is, to my knowledge, no scientific foundation for either assertion. Antiabortion elements have used the call for strict medical supervision to imply that the drug is dangerous. Extensive documentation confirms that abortion with mifepristone alone or with misoprostol is very safe and poses no unique medical challenges to a gynecologist or family physician. Routine medical care is warranted. Instead of strict medical supervision, what is required is strict administrative supervision to ensure that each abortion is completed. This follow-up is a logistical, not a medical, challenge.

Second, appropriate medical backup for the small percentage of women who have an incomplete abortion with heavy bleeding is suction curettage on an outpatient basis. Operating room facilities and general anesthesia are not necessary. Because patients may develop heavy bleeding during nights or weekends, facilities for suction curettage

should be available around the clock. While hospital emergency departments are open at all hours, they are often very expensive—and very hostile—locations to perform suction curettage. Your committee might want to consider after-hour availability of suction curettage in clinics or physician's offices that might be dispensing antiprogestins.

Spontaneous abortion is a prototype for expulsion of uterine contents without medical supervision, and the morbidity and mortality are negligible. Over the past five years, I have been in charge of the emergency room in the largest maternity hospital in North America, where I have personally supervised the care of about 2,000 patients with spontaneous abortions each year. The morbidity is really negligible. So, with medical supervision in a doctor's office, the safety should be even greater.

Given that we are going to be using prostaglandins along with RU 486, how can we make this regimen less clumsy, less cumbersome? A very interesting study from France published in the fall of last year showed that among women who had received this combination regimen of mifepristone plus sulprostone, three times as many women were unsatisfied with this regimen as with the curettage offered under either local or general anesthesia. This shows that we have a long way to go in making this more palatable for women in Europe and in the United States.

The second issue that generated debate this morning is the practice of four visits for mifepristone plus prostaglandin abortions. The scientific basis for two of these visits is debatable. This practice evolved in countries with very different health care systems for abortion services than the United States. In France, the United Kingdom, and Sweden, abortions take place in publicly supported facilities. In contrast, most abortions in the United States take place in free-standing abortion clinics. The response of public hospitals in the United States to the Roe v. Wade decision in 1973 was tantamount to default. Moreover, since 1977, the federal government has been prohibited from paying for abortions.

The European model of abortion provision is unlikely to be compatible with the United States health care delivery system because of the cumbersome requirement for multiple visits. In the United States, abortion services are currently limited largely to metropolitan areas, with more than 80 percent of U.S. counties having no provider. For

example, there is only one provider in South Dakota, and more than half of all Wyoming women who have abortions cross state lines for care. Women disadvantaged by lack of transportation or difficulties related to child care costs will find it difficult to make four separate visits for a first-trimester abortion. For example, this might require over 1500 miles of round-trip travel for many women in rural America. Use of this regimen would effectively deprive these women of this option, if one assumes that they could afford such care.

Medical science supports the need for two contacts with a health care provider for mifepristone abortion. The first would be for counseling, evaluation, and initiation of treatment. A second contact is important to confirm that the abortion has been completed and, if not, to perform a curettage. A second visit could be done at the initial facility or, if the travel distances were prohibitive, at a provider closer to home. A quantitative pregnancy test or a vaginal ultrasound examination can be used for this purpose. Women with a failed attempted abortion might be referred back to the initial provider.

Medical decisions about antiprogestins should be based on scientific evidence. Importation of abortion practices from dissimilar health care systems, without firm evidence of benefit, may be counterproductive. At the present time, a single visit to an abortion provider poses a hardship for many women in the United States. If mifepristone abortion is offered only in the European model, the service will be so cumbersome and expensive that many U.S. women will not have access to it. Rather than improving access to safe, legal abortion, this model would aggravate the existing disparity in access to service for disadvantaged women. To bring antiprogestins to our country and then to place them out of reach of women would be a cruel hoax.

I believe that randomized controlled trials could be done comparing the safety and acceptability of mifepristone abortions with varying numbers of visits. This would place clinical recommendations on a more firm scientific footing.

Third, we have heard discussed the appropriate dose of antiprogestins for the various indications. A common theme is the fact that we don't really know what the optimal dose is, and that is because we still lack sophisticated pharmacokinetic studies. As we have heard from several speakers, the dose-finding studies today have tended to use arbitrary doses with round multiples, such as 100, 200, 600 mg. They featured small sample sizes and, as we have heard from Dr. Nieman (see Appendix B3), have looked at relatively few outcome measures.

We need elegant pharmacokinetic studies correlating serum levels,

areas under the curve, and outcomes. What we now appear to know is that the 600-mg dose of mifepristone that we have been using appears far in excess of what is actually required, as you heard from Dr. Bygdeman this morning (see Appendix B5).

Does the dose make a difference? The pharmacology is very, very difficult. We don't see a linear relationship between dose and serum levels. Likewise, we don't see a linear relationship between dose and response. Nonetheless, it has been shown that the regimen does make a difference, and I will share with you our experience in Los Angeles.

Our experience using 13 different regimens (nearly all with mifepristone alone) can be divided into three broad categories: a single 600-mg dose, divided doses for a week, and all other regimens. Using regression analysis, we found three factors predictive of success for abortion: (1) body mass index, (2) initial beta human chorionic gonadotropin level, and (3) the regimen. The seven-day regimen was about twice as likely to result in failure as the other regimens and about six times as likely to fail as the single 600-mg dose. Each of these differences was highly statistically significant.

There is no free lunch in medicine, and we pay a price for everything we do. What should be the role of the prostaglandins with RU 486? After publication of the pivotal paper of Drs. Bygdeman and Swahn in 1985, the routine use of prostaglandins with RU 486 became common. We know that most of the noxious side effects and, indeed, the serious morbidity and mortality related to this regimen, result from the prostaglandins and not the RU 486—an example of the tail wagging the dog. The same week that a woman died in France with the RU 486 and prostaglandin regimen, we had a maternal death in Los Angeles of a 37-year-old woman who received a single, vaginal E2 suppository for abortion.

Since the prostaglandins cause most of the noxious side effects, the question we must ask ourselves is whether increasing the efficacy from the 80 to 90 percent range to the 90 to 100 percent range is worth the inconvenience, expense, and morbidity involved.

An important point that we heard raised this morning is that RU 486 may be most useful for augmenting second-trimester abortions. This has not received much attention in the United States. Premedication 36 to 48 hours before beginning an induction abortion cuts the time in half, and the important advantage here is that this now makes induction abortion

potentially an outpatient procedure instead of an inpatient procedure. At the present time in the United States, dilatation and evacuation (D&E) dominates second-trimester abortions, but requires substantial skill, in contrast to induction abortions. However, induction abortions are very expensive because they require hospitalization. Now we have the option with RU 486 for outpatient second-trimester labor induction.

The relative value of the antiprogestins for contraception in comparison to the current low-dose oral contraceptives cannot be judged until we know about the long-term effects of, let's say, continuous administration of an antiprogestin. The health benefits of low-dose oral contraceptives are really quite compelling. So, on public health grounds, one could make a much stronger case for using contemporary oral contraceptives as opposed to antiprogestins.

It has been shown by our colleagues in Edinburgh and Manchester that mifepristone has great promise as a post-coital contraceptive. In these two randomized controlled trials sponsored by the World Health Organization, about 600 women have received RU 486 without a single failure. Importantly, there was much less nausea and vomiting with this regimen as compared with the birth control pill regimen widely used in North America. The downside, however, was a substantial delay in onset of next menses.

We have also seen that mifepristone will soften and dilate the cervix, and its side effects are fewer than with prostaglandins. However, as you heard this morning, the need for prolonged pretreatment—36 to 48 hours—limits the usefulness of this method and makes this less convenient than osmotic dilators.

So, osmotic dilators, such as the traditional laminaria, Lamicel, which is a polyvinyl alcohol sponge impregnated with magnesium sulfate, or the newer Dilapan, which is hygroscopic plastic, will continue to dominate practice here.

Finally, I think we need to hear more about work with RU 486 analogues, other antiprogestins such as onapristone, lilopristone, and literally scores of other pristones, that await study.

In summary, I would leave you with six points. First, with regard to the safety issue, I think that issue has largely been resolved. Second, I think we all concur that better dose-finding studies are needed. Third, we need a way to come up with a less cumbersome regimen for administering RU 486, particularly in conjunction with a prostaglandin analogue. Fourth, I think a great potential use for RU 486 and similar drugs is to augment second-trimester abortions. Fifth, another very exciting potential is use as a post-coital contraceptive. Sixth, and finally, we need much more work, such as we have heard this morning, with these other scores of compounds.