Citizen Petition to the Food and Drug Administration.

The following Citizen's Petition was filed this year with the U.S. Food and Drug Administration by Americans United for Life. This recent historical document contains a great deal of valuable information about the health effects of RU 486, a drug which is being proposed to induce abortions but which has not yet received FDA approval. The Petition was filed because of concerns that the FDA may put RU 486 on a "fast track" of approval, overlooking normal safeguards.

For the convenience of the reader, footnotes have been placed as endnotes.--Ed.

Table of Contents

A. ACTION REQUESTED

B. STATEMENT OF GROUNDS

1. FDA's Statutory Mandate

a) FDA May Only Approve An NDA That Demonstrates Drug Safety

b) FDA May Only Approve An NDA Supported By Substantial Evidence Of Effectiveness

2. Foreign Data Must Comport With U.S. Standards To Support NDA Approval

3. FDA Should Deny Approval Unless Safety / Effectiveness Concerns Are Adequately Resolved

a) RU 486 Alone Is Ineffective And Poses Significant Adverse Effects, Complications, And Potential Risks

b) RU 486 In Combination With Misoprostol Presents Separate Risks And Safety Concerns

c) RU 486/PG: Increases Severity Of Adverse RU 486 Effects And Presents Separate Complications

d) Special Patient Populations Present Unique Risks

e) Incidence Of Complications May Be Underestimated

f) Poor Patient Compliance With Procedure And Follow-up

4. FDA Must Require Clinical Data On Unanswered Safety Concerns

a) Aborting Mothers

b) Children Born After Exposure To RU 486 And/Or Misoprostol

5. If Approved, Safety Mandates Labeling Limits

a) Indications And Usage

b) Contraindications

c) Warnings

d) Precautions

e) Adverse Reactions 

6. Dispensing Controls

a) Administration Only In Accredited Ambulatory Facilities/Hospitals

b) Administration By Physicians Only

c) Dispensing/Distribution Controls

C. ENVIRONMENTAL IMPACT

D. ECONOMIC IMPACT

E. CERTIFICATION

February 28, 1995

By Hand Delivery

Dockets Management Branch (HFA-305)

Center for Drug Evaluation and Research 

Food and Drug Administration

Department of Health and Human Services 

Room 1-2312420 Parklawn Drive

Rockville, MD 20852

CITIZEN PETITION

Petitioners, Americans United For Life, and Hon. Thomas J. Bliley, Jr.(Chairman, Committee on Commerce); Hon. J. Dennis Hastert (Committee on Commerce); Hon. Cliff Steams (Committee on Commerce, Committee on Veterans Affairs); Hon. Jack Fields (Committee on Commerce); Hon. Paul E. Gillmore (Committee on Commerce); Hon. Henry Hyde (Chairman, Committee on the Judiciary); Hon. Ed Bryant (Committee on the Judiciary, Committee on Agriculture); Hon. Bill Barrett (Committee on Economic and Educational Opportunities, Committee on Agriculture); Hon. Jim Talent (Committee on Economic and Educational Opportunities); Hon. Steve Largent (Committee on Science); Hon. Duncan Hunter (Committee on National Security); Hon. Mike Parker (Committee on Budget); Hon. Jim Bunning (Committee on the Budget); Hon. John Murtha (Committee on Appropriations); Hon. Barbara Vucanovich (Committee on Appropriations); Hon. Jim Lightfoot (Committee on Appropriations); Hon. Enid G. Waldholtz (Committee on Rules); Hon. Nick J. Rahall, II (Committ ee on Resources, Committee on Transportation and Infrastructure); Hon. Christopher H. Smith (Committee on International Relations, Committee on Veterans' Affairs); Hon. Andrea Seastrand (Committee on Science, Committee on Transportation and Infrastructure); Hon. Todd Tiahrt (Committee on National Security); Hon. Linda A. Smith (Committee on Resources, Committee on Small Business); Hon. Dan Coats (Committee on Labor and Human Resources); and the following individuals: Laurence M. Demers, M.D., Camilla Hersh, M.D., Donna J. Harrison, M.D., Earle W. Lingle, Ph.D., Eugene F. Diamond, M.D., J. Walter Sowell, Ph.D., and Joel Bind, Ph.D., hereby submit this citizen petition ("petition") under section 701 of the Federal Food, Drug, and Cosmetic Act ("the FDCA or Act"), 21 U.S.C. 371 (1988 & Supp. 1993), and its implementing regulations, 21 C.F.R. 10.25 and 10.30 (1994). Petitioners specifically request that the Commissioner of the Food and Drug Administration ("the Commissioner") refuse to approve any new drug applic ations ("NDA") submitted pursuant to section 505(b) of the Act, 21 U.S.C. 355(b) (1988), for RU 486 (mifepristone) for use as a pharmaceutical abortifacient. (1)

A. ACTION REQUESTED

Petitioners request that the Commissioner refuse to approve any NDA for RU 486 for use as a pharmaceutical abortifacient that does not contain adequate evidence that the drug has undergone nonclinical and clinical safety and effectiveness trials. The basis for petitioners' request is the statutory mandate of the Food and Drug Administration ("FDA") to withhold approval of any NDA that lacks sufficient data to establish that a drug is safe and effective for its intended use. (2) Approval of any NDA that is devoid of the appropriate safety and effectiveness data would not only be an express violation of the FDCA, but also an arbitrary and capricious agency action.

Petitioners also are concerned that RU 486 could be approved in the United States ("U.S.") based largely on foreign data, with only limited safety data generated from studies conducted in the U.S. Because approval based on possibly invalid foreign data and limited safety data would expose patients to significant and unreasonable adverse health risks, petitioners respectfully request that FDA consider the following factors in reviewing any NDA for RU 486.

* Foreign data that has not undergone a validation review by FDA may be unreliable.

* Because of the gravity of the potential risks of RU 486, it is imperative that all safety concerns in the potential populations affected by RU 486 be adequately addressed by NDA-generated data.

* Adequate directions for use of RU 486 cannot be provided in approved product labeling, unless direct safety/effectiveness issues are sufficiently resolved and defined by reliable scientific research.

* Unlike treatment for Acquired Immunodeficiency Syndrome ("AIDS"), advanced metastatic refractory cancers or other severely debilitating life-threatening diseases, alternatives to the RU 486/prostaglandin ("RU 486/PG") abortion method are currently available in the U.S. Because alternative abortion methods exist, no novel or urgent medical situation is present that requires FDA to expedite review or approval of an NDA for RU 486 as an abortifacient.

* The availability of alternative abortion methods should be considered by FDA in performing a risk/benefit evaluation of RU 486 as an abortifacient.

In light of these considerations, petitioners specifically request that:

(1) FDA audit all foreign data submitted in support of any NDA for RU 486 as an abortifacient;

(2) No NDA for RU 486 as an abortifacient be approved by FDA unless the safety and efficacy concerns presented in this petition are adequately resolved;

(3) FDA not approve RU 486 in the absence of clinical and nonclinical data that fully evaluate the potential adverse effects on the health of women who take RU 486 and/or misoprostol and any children born after exposure to these drugs;

(4) FDA refrain from adjusting and/or expediting the NDA approval process for any NDA for RU 486 for such use;

(5) In the event that FDA determines that an NDA for RU 486 meets the stringent statutory and regulatory application requirements, the approved conditions for use should be strictly limited. In particular, the drug product's labeling should bear adequate directions for use, and complete contraindication, complication and adverse reaction information. In addition, the drug product should be accompanied by an approved patient package insert. The labeling further should provide that administration of the drug must be limited to patients that are under the direct supervision and care of licensed physicians practicing in ambulatory care facilities or hospitals that meet the standards of the Joint Commission on Accreditation of Healthcare Organizations.

B. STATEMENT OF GROUNDS

This petition concerns FDA's statutory obligation to approve only those NDAs that contain adequate evidence that the proposed drug is safe and effective for its intended use. Because there appear to be a number of unresolved safety and/or effectiveness questions associated with RU 486, FDA is statutorily obligated to withhold approval of any NDA for RU 486, a potentially harmful and toxic drug product.

1. FDA'S Statutory Mandate

Specifically, section 505(b)(l)(A) of the FDCA requires that an applicant submit as part of an NDA, full reports of investigations that establish a drug is safe and effective for its intended use. (3)

a) FDA May Only Approve An NDA That Demonstrates Drug Safety

Section 505(d) indirectly defines the necessary safety evidence to support approval of an NDA. This section provides, in relevant part, that FDA must refuse to approve any NDA that does not include: (1) "adequate tests by all methods reasonably applicable to show whether the drug is safe for use" under its labeled conditions, (4) or (2) sufficient information to determine whether the drug is safe for use under the labeled conditions. (5) Petitioners assert that a reasonable interpretation of these statutory requirements is that safety data on all segments of the population affected by the administration of the drug must be provided before FDA may lawfully approve RU 486.

The segments of the population affected by RU 486 include the aborting women and their subsequent born children. (6) Petitioners are unaware of any published data on the effects of RU 486 on subsequent born children, and have identified only minimal data on certain potential health risks to users, such as the risk of breast cancer after induced abortion. Absent full reports of such critical data, FDA has a statutory obligation to refuse to approve an NDA for RU 486 as an abortifacient.

Because the agency has limited statutory authority to require controlled post-marketing studies, FDA relies on premarket research to evaluate the risk/benefit ratio of a drug and its potential post-approval risks. Prescribing physicians and the general public therefore depend on premarket trials as the source of reliable information on a new drug. FDA should require such data on RU 486 with regard to all population segments. Sufficient safety data is required not only for FDA to approve the drug, but also to (1) develop adequate directions for use of RU 486 as an abortifacient; (2) provide physicians with safety and efficacy information so that they may prescribe an optimal abortion regimen; and (3) provide sufficient information to assist patients in making well-informed medical decisions.

b) FDA May Only Approve An NDA Supported By Substantial Evidence Of Effectiveness

Section 505(d)(5) provides, in relevant part, that FDA must refuse to approve an NDA when "there is a lack of substantial evidence that the drug will have the effect it purports or is represented to have under the conditions of use prescribed, recommended, or suggested in the proposed labeling." (7) Section 505(d) defines "substantial evidence" to mean evidence consisting of adequate and well-controlled investigations, including clinical investigations, by experts qualified by scientific training and experience to evaluate the effectiveness of the drug involved. (8) In defining "substantial evidence," FDA has stated that a showing of clinically significant evidence of effectiveness is required. (9) Further, the Supreme Court has recognized that substantial evidence of effectiveness necessarily entails a showing of some benefit to the patient. (10) Petitioners therefore assert that an NDA applicant for RU 486 must establish that the drug product is clinically effective as an abortifacient. Without such evidenc e, FDA may not lawfully approve the NDA.

2. Foreign Data Must Comport With U.S. Standards To Support NDA Approval

FDA regulations require foreign clinical studies submitted by an NDA applicant to be well-designed, well-conducted, and performed by qualified investigators. The trials also must be conducted in accordance with ethical principles acceptable to the world community, or the foreign country's standards. (11) Additionally, foreign clinical data must be applicable to the U.S. population and U.S. medical practice, and validated through on-site inspections and/or submissions of case records or additional background data and information. (12) Foreign clinical data that falls to meet the above criteria cannot be accepted by the agency in support of drug approval.

Aware that most of the available data on RU 486 has been generated to secure foreign approvals, petitioners are concerned that an NDA applicant may attempt to rely on this data to support U.S. approval. Because foreign clinical trials may not have been conducted under adequate and well-controlled conditions, and/or under conditions that are representative of the U.S. population of potential RU 486 users, the agency should carefully examine the origin, design and patient population of each foreign trial proffered by an NDA applicant to support approval of RU 486. The agency also should conduct in-depth validity audits of each foreign clinical trial relied upon by an RU 486 NDA applicant. Only those clinical trials that comport with all statutory and regulatory requirements may lawfully be considered by the agency to support approval of RU 486. Petitioners believe that some or all of the known foreign clinical data may not comport with U.S. standards (13) and, therefore, any NDA which relies in whole or in par t upon foreign studies most likely will fail to meet the substantial evidence standard.

3. FDA Should Deny Approval Unless Safety/Effectiveness Concerns Are Adequately Resolved

a) RU 486 Alone Is Ineffective And Poses Significant Adverse Effects, Complications, And Potential Risks

As a single-entity abortifacient, RU 486 is relatively ineffective. (14) There is a high incidence of incomplete abortions (15) and ongoing pregnancies (16) when RU 486 is used during the first seven weeks of gestation. An even higher failure rate is observed in patients with greater body mass. (17) Incomplete abortion requires surgical intervention. (18) Incomplete abortions can also cause complications such as heavy bleeding and intrauterine infection, and may lead to pelvic inflammatory disease and infertility. (19) Further, when a pregnancy was continued after unsuccessful RU 486 administration, one child had severe deformities (20) including sirenomelia, a cleft palate and lip, micrognathia and hygroma. (21) Sirenomelia is a rare congenital malformation in which the lower extremities are fused.

As stated, one of the serious side effects of RU 486 is: the occurrence of excessive bleeding, requiring emergency curettage and sometimes blood transfusion. From published data it would appear that the risk of this complication is [approx.] 1-2% when RU 486 is used alone. (22,23)

This has led researchers to conclude that the risk of heavy bleeding is a serious complication that necessitates easy access to a hospital. (24)

b) RU 486 In Combination With Misoprostol Presents Separate Risks And Safety Concerns

When used in a two-step procedure with misoprostol, a prostaglandin, the abortifacient rate of RU 486 has been shown to increase. (25) U.S. test protocols generally use a 600 mg dose of RU 486 followed by 400 ug of misoprostol administered orally. However, three major issues have arisen with respect to the safety and effectiveness of this particular combination. First, researchers found that despite the increase in effectiveness demonstrated over RU 486 as a single entity, the effectiveness of the two-drug combination declines significantly after the 7 week gestation period. In women less than or equal to 49 days amenorrhea, the complete abortion rate is 97.5%. (26) This rate drops to 89.1% in women at 50-63 days, and 84.4% in women at 57-63 days. (27) As a result, researchers have concluded that the combination of RU 486 (200 or 600 mg) and oral misoprostol (600 ug) is effective for inducing abortion only in women of less than 50 days amenorrhea. At gestations greater than 56 days, "this combination may res ult in too many incomplete abortions to be clinically acceptable." (28)

Second, approximately thirty percent of patients do not abort the embryo prior to leaving the clinic or hospital (four hours after misoprostol administration). (29) This occurs because the pharmacological lag time for misoprostol may exceed the normal four-hour monitoring period of a typical medical abortion protocol. This is a serious public health concern since "the majority of women experience some pain during the passage of the fetus and heavy bleeding requiring resuscitation is most likely to occur at this time." (30)

Third, researchers in a British study found that the RU 486/oral misoprostol combination results in a higher incidence of ongoing pregnancies (4% of 121 subjects) than RU 486 followed by other prostaglandins, such as geme-prost or sulprostone (0.2%). (31) The results of that study are supported by the outcome of a large, clinical investigation in France. The results of the French study, which analyzed the effectiveness of RU 486/oral misoprostol in 488 women (study 1), confirmed that the ongoing pregnancy rate with 400 ug oral misoprostol is four times higher (four women, 0.8%) than that found in Britain using gemeprost (1 woman, 0.2%). (32) In conclusion, although misoprostol increases the abortion rate for RU 486, its introduction into an abortion regimen still presents considerable risks and safety concerns for the pregnant woman.

As discussed in more detail below, the use of prostaglandin analogs presents additional safety risks to the woman as compared to RU 486 used as a single-agent abortifacient.

c) RU 486/PG: Increases Severity Of Adverse RU 486 Effects And Presents Separate Complications

Even if effectiveness was not an issue in the use of RU 486 and misoprostol, prostaglandins are known to intensify the uterine cramping and pelvic pain associated with abortion to a point requiring administration of narcotics. (33) Prostaglandins can also cause severe gastrointestinal complications. (34) In addition, life-threatening cardiovascular complications can result from prostaglandin use, including death. (35) Two common effects of prostaglandins are a decrease in pulse rate (36) and low blood pressure which may lead to cardiac arrest. One study reported a slight decrease in the mean systolic and diastolic blood pressure four hours after misoprostol administration. (37) In addition, six women (1.2%) had "a substantial but transient decrease in blood pressure" (the systolic blood pressure fell by more than 30 mm Hg and the diastolic blood pressure by more than 15 mm Hg).38 Although transient, this is clinically significant hypotension, an adverse event that should be carefully examined.

d) Special Patient Populations Present Unique Risks

Because of common pre-existing conditions and disease states, RU 486 may never be a safe and effective abortifacient for certain patients. Specifically, RU 486/PG abortion poses substantial adverse risks to women with: asthma, epilepsy, diabetes, glaucoma, adrenal insufficiency, kidney disease, liver disorder, pulmonary disorder, cardiovascular disease, gastrointestinal disorders or intestinal disease, addisonian crisis susceptibility, prior use of steroid medication, prior use of some non-steroidal, anti-inflammatory medications like aspirin, recent use of hormonal contraception or presence of an IUD, recent Caesarean section, anemia, sickle cell anemia, hematologic or coagulation disorders, evidence of threatened abortion or ongoing spontaneous miscarriage, ectopic pregnancy, uterine fibroids or uterine anomalies. (39)

Other women may be subject to a greater risk of adverse reactions with medical abortion because of a "higher incidence" of complicating conditions. Four population segments may be at increased risk from this abortion method. The first population segment is African-American women. African-American women have a higher incidence of uterine fibroids than Caucasian women. (40) Thus, they may be at greater risk of retained products of conception. In cases where submucous uterine fibroids are present, these women may be at greater risk of excessive bleeding and related complications.

The second population segment is composed of Native-American (American Indians and Alaskan Natives) and Mexican-American women. These women have a much higher incidence of diabetes than that of the general U.S. population. (41) Consequently, Native-American and Mexican-American women with diabetes may be subject to a greater risk of complications from the gastrointestinal side effects of RU 486/PG abortion, such as loss of appetite, nausea, vomiting, and diarrhea. These side effects, although not severe m the average healthy American woman, are much more debilitating and may progress to more severe complications in a diabetic. Also, because Native-American women present more commonly with non-insulin dependent diabetes, their diabetic condition may be latent and remain undetected at the time of RU 486 administration.

Obese women present the third population segment. Grimes, et al., 1990 found that the risk of failure of RU 486 as a single agent abortifacient was 2.9 times greater for obese women than for women in the lowest body mass group studied. As a result, the researchers concluded that more pharmacokinetic research should be conducted to determine optimal dosing, which may not be the same for all patients. (42) The observation that obesity adversely affects RU 486 efficacy also was reported in a WHO study of RU 486 with sulprostone. The WHO study reported that subjects for whom the method failed were significantly heavier than those for whom the method resulted in complete abortion. (43) Heavier women also appear to have an earlier onset of bleeding. (44)

The fourth population segment is Asian-American women. These females may be at an increased risk of heavy bleeding and associated complications following RU 486/PG abortion. (45) Researchers to date have not conclusively identified the cause of the greater blood loss in this group of women.

The only potential population segment for whom RU 486/PG abortion might reach an acceptable level of safety is healthy women between the ages of 18 and 35, (46) who are not overweight and do not smoke. However, even this population will not be free of substantial adverse experiences, such as:

* heavy bleeding requiring medical intervention (up to 4% of users); (47)

* pelvic or abdominal pain (up to 94.1% of users); (48)

* infection (5% of users); (49) and

* cardiovascular conditions and accidents (up to 1.2% of users). (50)

Because of the known contraindications, complications and adverse effects of RU 486 with or without a prostaglandin, petitioners believe that, on balance, a proper risk/benefit analysis requires FDA to refuse to approve an NDA for RU 486, absent compelling safety and effectiveness data.

e) Incidence Of Complications May Be Underestimated

There is some statistical information on the number of women who suffer from immediate side effects of RU 486/PG abortion, such as fainting, pain, vomiting and diarrhea. (51) However, published studies to date provide little information on the incidence of secondary complications discovered shortly after drug administration (e.g., infection) or even later (e.g., PID, infertility). This is partly due to study protocol deficiencies that require follow-up only for a very limited time period after the abortion procedure. (52) Data also are unavailable or incomplete because of the number of patients who fail to return for follow-up examinations. (53) The lack of, or partially-completed, follow-up visits by patients impedes accurate reporting of latent complications and adverse experiences. Given the above factors, petitioners urge FDA to consider the published incidence of complications for RU 486/PG abortion to be underestimated.

f) Poor Patient Compliance With Procedure And Follow-up

Even with physician monitoring of the RU 486 abortion process, poor patient compliance with the procedure and follow-up program is a serious safety and effectiveness concern. An acceptable level of safety for RU 486 or RU 486/PG abortion is contingent on strict patient compliance, and adherence to an established follow-up program. (54) Lack of an effective means to ensure an adequate level of compliance in the treatment population is a serious drawback of medical abortion. Even under the carefully controlled conditions of a clinical trial, patient non-compliance has been a problem. The UK Multicentre Trial, 1990 reported that 9 women were lost to follow-up before investigators could confirm that the abortion was complete; 9.35% failed to return for follow-up two days after administration of the abortifacient; and 21.77% did not return nine days after receiving the drugs. (55,56)

Failure to return and/or receive the prostaglandin analog significantly Increases the risks associated with taking an ineffective dose of RU 486 alone. (57) In particular, taking RU 486 without a prostaglandin can be expected to increase the risk that surgical intervention will be necessary or that other complications will arise. (58) Furthermore, rhesus negative women who do not return for the prostaglandin and an anti-D immunoglobulin injection are at risk for rhesus isoimmunization and its associated complications in subsequent pregnancies. (59) Untimely PG administration (i.e., a time lapse between administration of RU 486 and the prostaglandin less than 36 hours or greater than 48 hours) also reduces the method's effectiveness. (60) As demonstrated by the UK Multicentre Trial and the Ulmann study, a significant number of women--even in well-organized health care programs--are at risk of receiving improper care and may be exposed to additional health hazards because of poor compliance. (61)

Petitioners assert that there is no reason for FDA to expect that the prospects for patient compliance will be any better here in the U.S. than observed overseas. Dr. Suzanne Poppema, owner of a Seattle abortion clinic, is currently participating in the clinical trials of RU 486/misoprostol. With regard to patient follow-up, Dr. Poppema commented that even though U.S. clinics routinely include follow-up visits in the price of an abortion "we're lucky if 30% to 40% of these patients ever return." (62) Without assurance of patient compliance, safe and effective medical abortion cannot be provided.

4. FDA Must Require Clinical Data On Unanswered Safety Concerns

To meet its statutory obligation, FDA must require an RU 486 applicant to submit clinical data that addresses all outstanding safety concerns in all population segments affected by RU 486. The population segments affected by RU 486/PG abortion include the aborting mothers, children born after a failed RU 486/PG abortion as well as any children conceived and born after their mother was exposed to these drugs.

a) Aborting Mothers

A review of published data indicates that the potential adverse effects on the subsequent health of an RU 486 user have not been fully investigated. In particular, for the aborting mother there is a paucity of data in the following areas: (63)

(1) Abortion/Breast Cancer Link

Significant data evidencing a link between induced abortion and breast cancer have now been gathered from many countries, despite major differences in study populations (e.g., Asian, Caucasian, and African-American women) and study designs. Although the link is not yet universally acknowledged, studies which refute the link are seriously flawed. For example, the studies of Vessey, et al., 1982 in England and Gandra, et al., 1993 in Portugal are confounded by the inclusion of spontaneous abortion data, (64) and a study on induced abortion in Sweden was only able to generate an odds ratio significantly less than 1 by gross omissions. (65) As a result, petitioners suggest that FDA carefully review the design of any studies submitted by an RU 486 NDA applicant that evaluate the abortion/breast cancer link. Petitioners also request that FDA refuse to approve any RU 486 NDA submitted without methodologically sound studies with statistically significant data on the association between abortion and breast cancer. Pe titioners believe that this data is critical for a comprehensive risk/benefit evaluation of RU 486.

To date, fifteen other epidemiological studies (66) report data on induced abortion and breast cancer risk. All of them are consistent with increased risk. Seven of these (67) are weakened by the lack of age matching of cases and controls, (68) one each in the former USSR (Dvoirin & Medvedev, 1978), Italy (La Vecchia, et al., 1993) and Denmark (Ewertz & Duffy 1988), and four in the US (Brinton, et al., 1983; Rosenberg, et al., 1988; Moseson, et al., 1993; Daling, et al., 1994). Nevertheless, only two of these do not report an elevated relative risk ("RR") estimate (La Vecchia, et al., 1993 [RR=0.9]; Moseson, et al., 1993 [RR=1.0]); while two of the seven report elevated risks that do not achieve statistical significance (Brinton, et al., 1983; Rosenberg, et al., 1988). (69) One reports a relative risk of 1.71 without showing the presence or absence of statistical significance (Dvoirin & Medvedev, 1978); and two report significantly elevated risks (Ewertz & Duffy, 1988; Daling, et al., 1994).

The other eight studies (70) do compare breast cancer patients with age-matched controls, one in Sweden and Norway (Adami, et al., 1990), two in Japan (Nishiyama, 1982; Hirohata, et al., 1985), two in France (Le, et al., 1984; Andrieu, et al., 1994) and three in the US (Pike, et al., 1981; Howe, et al., 1989; Laing, et al., 1993). Five of them report significantly elevated overall risks of breast cancer with induced abortion. Only Hirohata, et al., 1985, Adami, et al., 1990, and Andrieu, et al., 1994 do not report a significant overall risk (RR=1.5, 0.9, and 1.1 respectively). However, Andrieu, et al., 1994 do report a significant relative risk of 7.1 among women with two or more abortions and a positive family history of breast cancer. This interaction of two risk factors is underscored by the disturbing findings of Daling, et al., 1994, who also report greater risks for aborted women with a positive family history. (71) In particular, these authors report an incalculably high risk elevation for such women who had an induced abortion before age 18.

A weakness in study design shared by all but one of the case-control studies is the reliance on patient recall, rather than on prospective data. A bias towards more truthful reporting of induced abortion history by patients versus controls has been suggested. (72) However, the one case-control study based on prospective, computerized data (by the N.Y. State Dept. of Health; Howe, et al., 1989) reports an overall relative risk of 1.9 for induced abortion, similar to most other studies. These authors also report looking for and finding no evidence of response bias. Daling, et al., 1994 also critically evaluate and discount the response bias hypothesis in their report on Washington State women in the Journal of the National Cancer Institute.

One limitation of most studies on induced abortion and breast cancer risk is the lack of post-menopausal patients who were exposed to induced abortion. However, the recent study of Laing, et al., 1993 reports on a patient population mostly over age 50. It shows an increasing relative risk with increasing age, with the relative risk equal to 4.7 in patients over age 50.

In addition to the substantial body of epidemiological evidence for the link between induced abortion and breast cancer, the biological basis for an increased risk of breast cancer following abortion is well documented. (73) A woman's first full pregnancy causes hormonal changes which permanently alter the structure of her breast. Before a woman's first pregnancy, her breasts consist mostly of connective tissue surrounding a branching network of milk ducts, but with relatively few milk-producing cells. With pregnancy, estrogen and other hormones flood the mother's system causing breast cells to proliferate. The network of milk ducts begins to bud and branch, developing "terminal end buds."

Terminal end buds have been shown to be more susceptible to neoplastic transformation. (74) This effect is attributed to the fact that terminal end buds are composed of actively proliferating epithelium. In animal experiments using the carcinogen DMBA, the highest DMBA-DNA interaction is associated with the structure with the highest proliferative rate. In vitro experiments using human breast tissue have corroborated this observation. (75)

Skepticism concerning the biologic mechanism underlying the risk-enhancing effect of induced abortion has come recently from Dr. Lynn Rosenberg. Dr. Rosenberg cites the "inconsistent" nature of the association between spontaneous abortion and breast cancer as reason to doubt the underlying mechanism of early pregnancy interruption. (76) If susceptibility is increased by prevention of the tissue maturation that accompanies a full term pregnancy, induced or spontaneous abortions should have the same, risk-enhancing effect. It is true that many studies including the most recent American ones (Laing, et al., 1993; Daling, et al., 1994) have shown no association between spontaneous abortions and breast cancer, as has Rosenberg's own research. (77) It is entirely plausible that the earlier studies which did show an association were confounded by the misreporting of induced abortion. (78) Lehrer, et al., 1993 have also found a positive association between breast cancer and spontaneous abortion in women heterozygous for a variant of the estrogen receptor gene. (79)

One hypothesis advanced by Daling, et al., 1994 suggests that spontaneous abortions may occur earlier in gestation than most induced abortions, thereby not conferring increased risk. However, their own data still show significantly elevated risks for women who aborted their pregnancy prior to 8 weeks gestation.

There is substantial evidence to support an alternative hypothesis for the clearly emerging dichotomy between the effects of spontaneous versus induced abortion on subsequent breast cancer risk. That is, the same immune mechanism employed by the body to defend against cancer may also be responsible for many spontaneous abortions (and even for the induction of normal labor at term). (80) Specifically, Dr. M.R. Lentz, et al., in California have shown that soluble tumor necrosis factor (TNF) receptor protein is elevated both in cancer patients (81) and pregnant women. (82) Removal of this protein fraction through selective plasmapheresis has been used successfully in the treatment of human cancers (83) and in the reliable induction of labor and abortion in animals. (84)

Evidence from human clinical experience of heightened immune responsiveness toward cancer and the fetus in spontaneous abortion is provided by the detailed Canadian study of Clark and Chua, 1989. In their series of 154 patients diagnosed with breast cancer while pregnant, only 20% of patients who delivered at term survived 20 years, while 40% of patients who aborted spontaneously survived 20 years. In stark contrast, all patients who underwent "therapeutic abortion" died of the breast cancer within 11 years. (83)

Thus, while more research is needed to detail the long term effects of interrupted pregnancies vis-a-vis breast cancer risk, it is already clear that spontaneous and induced abortion are different events; and that the latter is associated with an increased risk of subsequent breast cancer. Therefore, petitioners believe that at a minimum, FDA has a statutory obligation to require testing that conclusively answers whether RU 486/PG abortion amplifies susceptibility to breast cancer in laboratory animals, in order to evaluate the risk to women. Moreover, since recent studies, especially the NCI-funded study of Daling, et al., 1994, have suggested a particularly strong link in women with a positive breast cancer family history, there is a critical need for retrospective studies on the interaction of induced abortion and family history as risk factors. Petitioners believe that the applicant should be required to provide adequate data on this point in order for the FDA to complete a comprehensive risk/benefit eva luation for RU 486.

(2) Effects on Compromised Patients

Further data is needed to define the effects of RU 486 use in patients who are adrenal compromised, and in patients with liver or kidney disease. RU 486 interferes with cortisol binding to hypothalamic-pituitary tissue, inhibiting the negative feedback mechanisms, resulting in a compensatory increase in serum levels of cortisol and corticotropin. Simultaneously, RU 486 binds to peripheral cortisol receptors, blocking the effect of circulating cortisol (86) Thus, there is a potential for an inappropriate response to stress in users. (87) However, petitioners are not aware of any studies investigating this potential problem. Also, advocates of RU 486 suggest that the infrequent use of RU 486 is unlikely to result in clinical hypocortisolism, but petitioners are aware of only one clinical study (conducted in healthy human males) that may support this contention. (88)

(3) Optimal Dosage Undetermined

The minimum effective dose of RU 486 in combination with a prostaglandin has still not been determined. (89) There is disagreement in the scientific literature about optimal dosage for RU 486. This issue should be settled prior to approval of any NDA; thus, FDA should require optimal dosage studies.

A study by Grimes, et al., 1990 indicates that further pharmacokinetic research is needed on the use of RU 486 as an abortifacient in obese women. The Grimes study demonstrated an increased risk of failure of RU 486 as a single-agent abortifacient in women with higher body mass. (90) This indicates that the dose of RU 486 required to induce a complete abortion may change with increasing body mass; or possibly that safety and effectiveness of RU 486 cannot be achieved in the obese population.

(4) Systemic Build-Up From Repeat Usage

Petitioners are unaware of any published studies on the effects of repeat usage of RU 486 alone or in combination with a prostaglandin. RU 486 is fat soluble and has been found in adipose tissue. (91) This raises the issue of whether RU 486 can accumulate and be stored in adipose tissue or other body tissues; and if so, what triggers its release. Researchers have detected unmetabolized RU 486 in plasma up to 10 days after single oral administration of 200 mg. (92) This suggests that RU 486 accumulates in the tissues. (93) If RU 486 can accumulate in the tissues after a single oral dose, the impact of serial use of the drug must be examined. This is justified since the most recent figures indicate that over 42% of women having abortions in the United States have had 1 or more previous induced abortions. (94)

Furthermore, an account of the RU 486/PG abortion experience by a woman who had two medical abortions within a four-month time period indicates that this issue warrants further investigation. After her second RU 486/PG abortion, this woman suffered from extreme fatigue for almost a month, along with an "extreme amount of bleeding" between administration of RU 486 and the prostaglandin. (95) For this individual, the RU 486/PG abortion experience was much more debilitating the second time. Thus, further data, particularly drug disposition and half-life studies, are needed on the systemic build-up of RU 486 when used more than once in a short period of time. If systemic build-up does occur, RU 486, because of its progesterone-like activity, may have an effect on tissue outside the uterus including the fallopian tubes, vagina, ovaries, breasts, and parts of the central nervous system, such as the hypothalamic-pituitary gland, respiratory center, and perhaps cortical function.

(5) Blood Loss In Asian-American Women

Although researchers have been unable to identify the cause, Asian women appear to be affected differently by RU 486 than Caucasian women. In particular, Asian women may be at increased risk of heavy bleeding and its associated complications with medical abortion. A World Health Organization ("WHO") study found a significantly longer duration of bleeding in Chinese women than in non-Chinese women. (96) Another study, performed by Chan, et al., compared blood loss in Chinese women after vacuum aspiration with blood loss after administration of RU 486/gemeprost. The researchers found that Chinese women who aborted using RU 486/gemeprost had a significantly greater degree of blood loss than those who aborted by vacuum aspiration. (97) In contrast, a study involving Caucasian women did not demonstrate a significant difference in median blood loss between subjects undergoing surgical and medical abortions. (98) Based on these results, researchers speculated that "the discrepancy might be due to a racial differenc e" since "changes in the coagulation system in the Chinese women as a result of hormone treatment are different from those in Caucasian women." (99) Chan, et al., also speculated that differences in blood loss might be explained by differences in abortion procedure. Specifically, the use or non-use of syntocinon during vacuum aspiration may have an effect. (100)

Petitioners assert that the above hypotheses must be tested by valid scientific study before FDA considers approval of any RU 486 NDA. Without such data, adequate directions for the use of RU 486 in women of Asian descent cannot be provided in approved labeling. Also, and perhaps most importantly, data that conclusively demonstrate an adverse effect or an increased risk factor with use of RU 486 that is specific to a racial group weighs strongly against approval of RU 486.

(6) Increased Vulnerability To Infection/Disease

The risk of immunosuppression must gain greater attention as RU 486 abortion studies have reported infection following RU 486 administration. If women receiving RU 486 become immunosuppressed (due to RU 486 alone, RU 486 and RU 486-elevated cortisol, or RU 486-elevated cortisol and misoprostol), they will be more susceptible to infections, particularly genitourinary infections.

After administration of a single dose of RU 486, cortisol levels rise and remain significantly elevated for at least 72 hours. (101) Unpublished data also indicates that misoprostol produces an increase in serum cortisol in women. (102) Cortisol, as a type of glucocorticoid, is known to adversely affect the body's defenses against disease and injury. (103) In addition, RU 486 itself may have a negative effect on the immune system. The few studies that have examined this possibility are limited and inconclusive but suggest that further research is necessary. (104)

For instance, Van Voorhis, et al., reported that women taking RU 486 in abortifacient doses achieve serum concentrations of RU 486 that act in an immunosuppressant manner, and further augment the suppression of immune function caused by cortisol. (105) The Van Voorhis study indicates that further research, in particular, in vivo studies in women and in vitro studies on a broad assortment of white blood cell types, should be conducted. This research is needed to clarify the immunosuppressive activity of RU 486 itself and the synergistic effects of cortisol/RU 486. Petitioners urge FDA to consider the potential adverse impact on our growing AIDS population of approving RU 486 without definitive answers in these areas.

Two studies have reported enhanced immunosuppression with misoprostol or a PGE1 analog used in conjunction with immunosuppressant drugs. (106) The Moran, et al., study mentions unpublished data that misoprostol acts to enhance the immunosuppressant activity of steroids. (107) Because RU 486 elevates blood cortisol (a steroid), further investigation into the possibility that cortisol and its immunosuppression may be enhanced in the presence of misoprostol is important. Such research should be designed to demonstrate the extent of immunosuppression, the molecular mechanisms involved and the subpopulation of white cells (e.g., monocytes, lymphocytes, eosinophils, etc.) affected.

(7) Interference With Transport Of Other Drugs/Hormones

Since the human alpha 1-acid glycoprotein (AAG or orosomucoid) is the main plasma transport protein for basic drugs (e.g., imipramine), acidic drugs (e.g., warfarin), and other ligands (e.g., progesterone), it is imperative to study the interactions of RU 486 and its prime plasma protein transporter--AAG. RU 486 binds strongly to AAG variants. (108) Petitioners are unaware of any studies that explore competitive binding of RU 486 and other drugs or hormones. This research is required to rule out the possibility that RU 486 may compete with other drugs or hormones for AAG transport, thereby slowing or blocking the transport of these medications or hormones.

(8) Impact On Future Fertility

Although there are anecdotal accounts of a return of fertility in women who have taken RU 486 with, or without a prostaglandin, (109) petitioners are unaware of any scientific data on the long-term effects of these drugs on the future child-bearing potential of women.

In light of the foregoing discussion, petitioners urge FDA to reflect on the known risks and the potential emergence of post-approval risks to RU 486 users. Pre-marketing studies and studies supporting foreign approval of RU 486 may not be adequately designed to detect or measure serious adverse effects that are infrequent or latent (like those associated with DES). Petitioners assert that there are sufficient quantifiable safety concerns and unquantifiable potential health risks that warrant requiring further study of the effects of RU 486 on users. In view of the availability of alternative abortion methods, there is no compelling reason for FDA to approve an NDA for RU 486 as an abortifacient, particularly when the known adverse effects and potential health risks to women are taken into consideration.

b) Children Born After Exposure To RU 486 And/Or Misoprostol

Although RU 486 has been used to abort human pregnancies since 1982, there are limited data on the effects of RU 486 or misoprostol on an embryo that is carried to term despite administration of these drugs to the birth mother. For subsequent born children, there are insufficient clinical data in the following areas:

(1) Risk Of Congenital Malformation In A Continued Pregnancy

The risk of congenital malformations in children born after a failed RU 486 abortion must be defined further. Since a significant percentage of pregnancies continue despite the administration of RU 486 or RU 486 with a prostaglandin, (110) the issue of deformities is not theoretical.

Risk Associated with RU 486--Research has shown that both mifepristone, and probably its major metabolite, cross the placenta. (111) There is limited and contradictory data on whether the direct action of RU 486 on "the trophoblast/placenta might result in retardation of the embryos and contribute to the birth of children with abnormalities." (112) In animals, two studies have demonstrated teratogenic activity of RU 486. A study by van der Schoot & Baumgarten, involving the neonatal administration of RU 486, resulted in behavioral and anatomical defects in male and female rats. (113) In female rats, development of the reproductive tract was permanently affected. Abnormalities occurred in oviduct and ovarian capsule structure, as well as the development of anovulatory polyfollicular ovaries during adulthood. There was interference, after some delay, with normal ovarian cyclicity and reduced fertility (smaller litters). A temporary suppression of adrenal gland growth was noted in both male and female rats. In males, permanently retarded testicular size and growth, delay of puberty and reduced growth around puberty occurred. Male rats also exhibited a reduced capacity to ejaculate. This deficiency in male sexual behavior resulted in relative infertility. Finally, early exposure to RU 486 resulted in the expression of female sexual behavior in adult males.

A study of rabbits reported various anomalies in fetuses that were not aborted following mifepristone administration. These included skull deformities, non-fused eyelids, absence of closure of the vertebral canal, and extremely small size. (114) On the other hand, no evidence of teratogenicity was found in a limited trial of monkey embryos exposed to RU 486. (115) Nevertheless, researchers cautioned that "this study and our conclusions are limited by the small numbers of observations, as well as the particular conditions tested here." (116) "[W]e cannot be assured by the data presented here that exposure to RU 486 is never teratogenic.

In humans, there have also been mixed reports. Dr. Ulmann, head of Roussel Uclaf's division of endocrinology, reported that six children born after failed mifepristone abortions are normal. (118) However, one fetus exposed to RU 486 in early pregnancy had multiple abnormalities, including sirenomelia (fused lower extremities), a cleft palate and lip, micrognathia and hygroma. (119) Although the observations were reported to Roussel Uclaf at the time of occurrence, data necessary for an objective evaluation of these cases, such as follow-up studies on lost-to-view women who have taken mifepristone and other pharmacovigilance data, are still unavailable today. (120)

Risk Associated with Prostaglandins--Prostaglandins have been reported to cause deformities in both human beings and animals. (121) In particular, misoprostol has been associated with two specific types of anomalies in children. Five cases of a frontal and/or temporal defect in the skull without other anomalies were found in babies born to women who had taken 400-600 ug of misoprostol orally and/or vaginally in the first trimester. (122) Gonzalez, et al., reported seven children with limb abnormalities, four of whom also had a diagnosis of mobius sequence. (123) The mothers all had taken 200-1800 ug of misoprostol orally and/or vaginally between 4 and 12 weeks amenorrhea to attempt abortion. (124) It is still unclear as to whether these deformities were directly caused by misoprostol. (125) Fonseca, et al., concluded that "[al deleterious effect of misoprostol plus mifepristone on the development of the fetus cannot be ruled out. Firm evidence on freedom from congenital malformations is needed before this dru g [misoprostol] can be promoted for use in pregnancy termination." (126) Therefore, petitioners assert that additional data must be collected to quantify the risk of congenital malformations in children born after failed RU 486 abortion with or without a prostaglandin.

(2) Risk of Congenital Malformation in Future Generations

RU 486 and two of its major metabolites cross the blood-follicle barrier of human pre-ovulatory follicles. Cekan, et al., found high concentrations of RU 486 and its metabolites in the follicular fluid. (127) This raises serious questions on the effect of RU 486 on the thousands of immature eggs in the ovaries of a woman that has been exposed to the drug.

(3) Risk Of Premature Delivery In Continued Pregnancies

Furthermore, data must be collected to assess the risk of premature delivery in a continued pregnancy due to the softening and dilation of the cervix caused by RU 486.128 Petitioners are unaware of any such published data.

(4) Related Risks

A study conducted by Wiedemann, et al., raises questions about the potential effects of RU 486 (and its metabolites) on fetal brain development, in utero fetal sleep patterns and fetal cortical development. (129) Petitioners are unaware of any studies conducted on the foregoing effects, or on the effect of fetal exposure to RU 486 on an infant's normal postpartum sleep patterns.

In addition, petitioners are unaware of any published studies on the potential carcinogenic, teratogenic, reproductive or behavioral post-birth effects of fetal exposure to RU 486 and its metabolites. Given the tragic experience with diethylstilbestrol (DES), these potential risks to subsequent born children exposed to RU 486 in utero should be quantified by clinical data. Also, the possible adverse effect on fetal development of a rise in fetal cortisol levels induced by RU 486 should be studied. (130)

Petitioners believe that there exists adequate concern for the safety of subsequent born children to postpone the approval of RU 486 pending: (1) the development and implementation of a voluntary RU 486 pregnancy registry, and (2) the generation of data on the potential mutagenic, carcinogenic, and reproductive effects of RU 486 on children born to women that received RU 486 during their pregnancy.

5. If Approved, Safety Mandates Labeling Limits

Notwithstanding the above discussion, should the Commissioner find that an NDA for RU 486 as an abortifacient meets all statutory and regulatory requirements, petitioners request that the Commissioner place stringent limitations on the terms and conditions for use of the drug. To accomplish this objective, the agency should require that: (1) the labeling of the drug bear the limitations and conditions set forth below; and (2) the drug be accompanied by patient brochures to assist patients in making well-informed medical decisions.

a) Indications And Usage

In order to minimize adverse health risks, the following information must be included in the labeling for RU 486:

(1) Approved only for use by physicians in ambulatory care facilities or hospitals that meet the standards of the Joint Commission on Accreditation of Healthcare Organizations.

(2) Approved only for use as an abortifacient in conjunction with prostaglandin analogs.

(3) Approved only for termination of pregnancies documented by ultrasound to be intrauterine and within 49 days amenorrhea.

Based on a review and medical expert evaluation of published scientific literature, petitioners believe that ambulatory care facilities or hospitals that meet the standards of the Joint Commission on Accreditation of Healthcare Organizations are required to control the risks of RU 486/PG abortion. This procedure requires the available surgical staff, resuscitation equipment, and adequate blood stores that only an accredited ambulatory care facility or hospital can provide. (131)

Further, although petitioners believe that the RU 486 abortion procedure cannot reach a controllable level of safety regardless of the approved conditions for its use, petitioners request that in the event that RU 486 is approved as an abortifacient, it be approved only for use in conjunction with prostaglandin analogs. RU 486 used in combination with a prostaglandin analog achieves a far more acceptable level of safety and effectiveness, than as a single agent.

Furthermore, the RU 486 abortion regimen requires precise gestational dating to minimize the risk of incomplete abortion and excessive bleeding. Effectiveness of RU 486/oral misoprostol decreases from approximately 96% within 49 days amenorrhea to only 89% within 50-63 days amenorrhea. (132) Also, the amount of blood loss increases significantly with advancing gestation. (133) A sonogram provides the most accurate measure of gestational age presently available. (134) Requiring vaginal ultrasonography or a pelvic ultrasound scan as part of the procedure is essential in order to avoid the risks associated with RU 486/oral misoprostol after 49 days amenorrhea.

A sonogram also provides an effective means of confirming that the pregnancy is intrauterine prior to administering RU 486. Women with ectopic pregnancy should not receive RU 486. The drug is believed to act primarily at the endometrium and myometrium, making the drug ineffective for extrauterine pregnancy. (135) An untreated ectopic pregnancy puts a woman at continued risk of serious complications, such as severe extrauterine hemorrhage. (136) Requiring a sonogram reduces the risk of the drug being used inappropriately in a patient with ectopic pregnancy.

b) Contraindications

The subpopulation of potential RU 486 users for whom the safety of RU 486/PG abortion cannot reach a controllable level is extensive. If RU 486 is approved, petitioners believe that the following contraindications must be noted in the labeling for the drug.

(1) Contraindicated for termination of pregnancy greater than 49 days amenorrhea.

(2) Contraindicated for termination of pregnancy in women with asthma, adrenal insufficiency, cardiovascular disease, coagulation or clotting disorders or women receiving anticoagulants.

(3) Contraindicated for women under age 18 years. (137) Not recommended for women over age 35 years.

(4) Contraindicated in pregnancies in women with recent unhealed

Caesarean section.

(5) Ectopic Pregnancy--RU 486 should not be used if ectopic pregnancy is suspected. Vaginal ultrasonography or pelvic ultrasound scan is required to confirm inrauterine pregnancy prior to RU 486 administration. Use in ectopic pregnancy may result in severe extrauterine hemorrhage.

c) Warnings

(1) Clotting Disorders/Anticoagulants--RU 486 may induce considerable uterine blood loss in women with a clotting disorder or receiving anticoagulant drugs such as sodium warfarin (Coumadin). Women with clotting disorders should not receive RU 486.

(2) Cardiovascular Reactions--Serious cardiovascular events have occurred with the RU 486/sulprostone abortion method (3 myocardial infarctions (1 fatal), 3 cases of severe hypotension). One patient death has been reported with misoprostol, in addition to clinically significant hypotension. Myocardial infarction has not been reported as a complication of RU 486/vaginal gemeprost. RU 486 plus any prostaglandin should not be used in women who smoke heavily, women over 35, or women who have any other increased risk of cardiovascular events.

(3) Any infection within the female reproductive tract should be treated prior to abortion. Failure to do so could result in a life-threatening bloodstream infection from the induction of uterine contractions and bleeding during medical abortion.

(4) Uterine tumors and endometriosis, if severe, could produce increased bleeding, retained pregnancy products, and possibly intraperitoneal bleeding.

(5) Acetaminophen may enhance pain during medical abortion and should be avoided. Drugs that affect prostaglandin synthesis, such as aspirin and non-steroidal anti-inflammatory drugs, may affect the efficacy of RU 486/PG abortion and should be avoided until follow-up.

(6) Induction of abortion is associated with an increased risk of developing breast cancer later in life. This risk is even higher in women with first or second degree family history (sister, mother, grandmother, or aunt) of breast cancer.

d) Precautions

Based on a review of the medical literature and medical evaluation, petitioners believe that in order to ensure a controlled level of safety and effectiveness for RU 486 administration, the following information for prescribing physicians should be included in the product's labeling. In addition, petitioners believe that the following patient information must be provided in patient brochures to enable pregnant women considering abortion to make informed medical decisions.

(1) Information For Physicians

* RU 486 should be used with caution in women with epilepsy, intestinal disease, hematological disorders, women who are anemic, immune compromised, or have liver or kidney disease.

* Women with submucous uterine fibroids may be at greater risk of excessive bleeding. Women with uterine tumors or uterine anomalies may be at increased risk of retained products of conception and extra care should be taken to make sure the endometrial cavity is clear and the abortion is complete.

* Women of Asian descent may be at increased risk of heavy bleeding. Studies have shown greater blood loss from medical abortion in these women, as compared to Caucasian women.

* African-American women have a much higher incidence of uterine fibroids than Caucasian women and may be at greater risk of retained products of conception or excessive bleeding.

* Native-American (American Indians and Alaskan Natives) and Mexican-American women have a higher incidence of diabetes than the general U.S. population and may be subject to greater risk of complications from the gastrointestinal side effects of medical abortion. Women with diabetes require careful management of fluid, electrolyte and blood glucose levels.

* Physician monitoring--RU 486 may only be administered by a licensed physician. Patients must be monitored by a physician for at least 4-6 hours following administration of a prostaglandin analog. Because of the risks associated with this abortion method, administer the drugs only where resuscitation equipment is immediately available.

* Contraceptives--An IUD must be removed before administering RU 486.

* Anticoagulants--Patients taking aspirin, NSAIDs or anti coagulation drug products have an increased risk of serious bleeding with RU 486.

* Obesity--An increased risk of RU 486 abortion failure has been noted in women with higher body mass.

* Rh (-)--Rhesus-negative women should receive anti-D immunoglobulin at the time of prostaglandin administration.

* Continued pregnancy-Teratogenic effects of RU 486 have been reported in rats and rabbits. In humans, sirenomelia (fused lower extremities) has been reported in an instance of continued pregnancy following RU 486. Seventeen instances of malformation have been reported with the use of the prostaglandin analog, misoprostol.

(2) Information For Patients

* Compliance--Full compliance with your physician's orders is required for a safe and effective abortion procedure with RU 486. RU 486 is prescribed along with a prostaglandin analog in a two-step process. Appointments for taking the drugs and follow-up visits must be set with your prescribing physician. It is important that you return for every scheduled visit. Failure to return or to follow your physician's orders may result in an incomplete abortion or continued pregnancy, a need for surgery, severe bleeding, severe pelvic pain or other dangerous complications.

* Risks of medical abortion--Studies indicate that induced abortion is associated with an increase in the risk of developing breast cancer. Physicians prescribing this product have a professional responsibility to provide you with individualized counseling before performing an abortion. This counseling should take into consideration your individualized breast cancer risk profile. Based on the most current research, your having a family history of breast cancer (affected sister, mother, grandmother or aunt) may put you at even greater risk of developing breast cancer if you abort this pregnancy. Your doctor should explain how your choices affect your breast cancer risk to help you decide whether to complete this pregnancy or abort it.

e) Adverse Reactions

The following adverse reactions must be noted in the approved labeling and patient information brochures for the drug. For RU 486 as a single agent abortifacient, or used in combination with a prostaglandin analog:

(1) Gastrointestinal--nausea

vomiting

diarrhea

abdominal pain

(2) Genitourinary system--uterine cramping

pelvic pain

vaginal bleeding (excessive, prolonged)

vaginal discharge 

endometritis 

salpingitis 

frequent urination 

(3) Central nervous system--headache

dizziness

sleep disruption/insomnia

(4) Skin-- skin rash

(5) Miscellaneous-- fatigue syndrome 

loss of appetite

thirst

For RU 486 administration with prostaglandin analogs:

* Cardiovascular-- myocardial infarction

ventricular fibrillation 

coronary spasms

severe hypotension

anaphylactic bronchospasm 

6. Dispensing Controls

a) Administration Only In Accredited Ambulatory Facilities/ Hospitals

As noted previously, petitioners believe that if RU 486 is approved, use of the drug must be limited to administration by physicians only in ambulatory care facilities or hospitals that meet the standards of the Joint Commission on Accreditation of Healthcare Organizations. There is a trend in the health care industry for midwives or physician's assistants to deliver infants at home. There is also a mounting campaign to permit non-physicians to perform surgical abortions, or to teach self-induced abortions. Petitioners are concerned that, if RU 486 is approved, a similar trend may develop for medical abortion. The complications and side effects of RU 486, alone or with prostaglandin administration, make it necessary for RU 486 to be administered in an accredited ambulatory facility or hospital. Researchers have emphasized that RU 486/PG should only be used in clinics where emergency facilities are available. (138) A home-abortion trend would most likely result in an increase in maternal mortality and morbidit y.

b) Administration By Physicians Only

Because of the serious complications and side effects of the RU 486 abortion process, petitioners believe that the drug labeling must require "Administration By Physician Only," rather than "Dispensing By Physician Only." Close physician supervision is required to ensure proper administration and monitoring of the RU 486/PG procedure. Approval of the drug only for use by physicians in an accredited medical facility will reduce the occurrence of physicians delegating administration of the drug to other medical staff.

c) Dispensing/Distribution Controls

RU 486 is a unique drug. Conditions which improve its effectiveness, i.e., administration in conjunction with a prostaglandin analog, are known to increase the risk of serious complications. Thus, medical abortion with RU 486 and a prostaglandin requires several visits to an ambulatory care facility or hospital, a precise, possibly individualized dosing scheme, and close physician monitoring. Women should not be led to believe that RU 486 abortion is a simple procedure or that it is conducive to self administration or administration by anyone other than a licensed physician. In light of this, petitioners believe it is necessary for FDA to require strict distribution and use controls, similar to those used for narcotic administration, to prevent the abuse and/or misuse of RU 486. (139)

C. ENVIRONMENTAL IMPACT

Petitioners believe that the actions requested herein qualify for a categorical exclusion from the requirement of issuance of an environmental assessment under 21 C.F.R. 25.24(a)(11)(1994). in any case, petitioners do not believe that there will be any substantial environmental impact from the relief requested in this petition.

D. ECONOMIC IMPACT

Petitioners will provide data concerning the economic impact of this proposal if requested to do so by the Commissioner pursuant to 21 C.F.R. 10.30(b).

E. CERTIFICATION

The undersigned certify that, to the best of his/her knowledge and belief, this petition includes all information and views on which the petition relies, and that it includes representative data and information known to the petitioners which are unfavorable to the petition.

Respectfully submitted,

Paige Comstock Cunningham

President

Americans United For Life

343 S. Dearborn Street

Suite 1804

Chicago, IL 60604

[phone]

Clarke D. Forsythe

Vice President and General Counsel

Americans United For Life

343 S. Dearborn Street

Suite 1804

Chicago, IL 60604

[phone]

Gary L. Yingling

Counsel For Petitioners

McKenna & Cuneo

1575 I Street N.W.

Washington, D.C. 20005

[phone]

Notes

(1.) The comments submitted herein are limited to RU 486 as an abortifacient drug product. This petition does not, therefore, oppose or address in any manner the use of RU 486 in the treatment of diseases, such as breast cancer or meningiomas (brain tumors).

(2.) See FDCA 505(d), 21 U.S.C. 355(d).

(3.) 21 U.S.C. 355(b)(1)(A).

(4.) See FDCA 505(d)(1) (emphasis added), 21 U.S.C. 355(d)(1).

(5.) See FDCA 505(d)(4) (emphasis added), 21 U.S.C. 355(d)(4).

(6.) The phrase "subsequent born children" is defined for the purpose of this document as children born as a result of a pregnancy continued after a failed RU 486 abortion as well as any additional children conceived and born after exposure to RU 486.

(7.) 21 U.S.C. 355(d)(5).

(8.) 21 U.S.C. 355(d).

(9.) See Warner-Lambert Co. v. Heckler, 787 F.2d 147, 155 (1986).

(10.) See United States v. Rutherford, 442 U.S. 544, 553 n.9 (1979); Warner-Lambert Co. v. Heckler, 787 F.2d at 155 (discussing United States v. Rutherford).

(11.) See 21 C.F.R. 312.120, 314.106 (1994).

(12.) See 21 C.F.R. 314.106 (1994).

(13.) Dr. Meredeth Turshen has raised concerns about the accuracy of the complication and failure rates being reported by foreign studies funded by Roussel Uclaf (the French manufacturer of RU 486), in light of the results that have been obtained by some independent researchers, which have not been published. Dr. Turshen was a fellow at INSERM (the French equivalent of NIH) during 1989-90. See Comments by Dr. Turshen at "Contraceptive Technology: Promises and Polities" workshop at the annual meeting of the Am. Public Health Assoc. (Oct. 2, 1990); "Researcher suggests aide effects of RU. 486 may be underreported" Am. Medical News, Oct. 26, 1990, at 8; Boston Herald, July 31, 1992, at 25.

(14.) An average value for the frequency of complete abortion with RU 486 alone is approx. 63%. WHO Task Force On Post-Ovulatory Methods For Fertility Regulation, Termination Of Early Human Pregnancy With RU 486 (Mifepristone) And The Prostaglandin Analogue Sulprostone: A Multi-Centre, Randomized Cemparison Between Two Treatment Regimens, 4 Hum. Reprod. 718, 719 (1989) (hereafter "WHO Task Force, 1989"). See Table 1 for comprehensive statistics on complete abortion rates.

(15.) Grimes, Mifepristone (RU 486) For Induced Abortion, 3 Women's Health Issues 171, 172 (1993) (hereafter "Grimes, 1993"); Grimes, et al., Early Abortion With A Single Dose Of The Antiprogestin RU-486, 158 Am. J. Obst. Gyn. 1307, 1308 (1988) (hereafter "Grimes, et al., 1988") (10% failure with 600 mg dose in women less than or equal to 49 days from last menstrual period ("LMP"); and Shoupe, et al., Pregnancy Termination With A High And A Medium Dosage Regimen Of RU 486, 33 Contraception 455 (1986) (hereafter "Shoupe, et al., 1986") (90% failure in high dose group--400 mg/day for days (N=5) or 200 mg/day for 4 days (N=5)--all women within 49 days of first day of LMP). For statistics on incomplete abortion rates, see Table 2.

(16.) With RU 486 alone, from 8.3% to 46.3% of pregnancies continue. See Table 9 for reported statistics.

(17.) Grimes, et al., found that the risk of failure from RU 486 alone for women in the largest body mass group studied was 2.9 times greater than that of women in the lowest body mass group studied. Grimes, et al., Predictors Of Failed Attempted Abortion With The Antiprogestin Mifepristone (RU 486), 162 Am. J. Obst. Gyn. 910, 913-14 (1990) (hereafter "Grimes, et al, 1990") (risk with a Quetelet's index [weight(kg)/height(m2)] greater than 23.81 was 2.9 times the risk with index less than 20.25); see also Grimes, 1993, at 172 (1993). See discussion infra Section 3.d.

(18.) See Chan, et al., Blood Loss in Termination Of Early Pregnancy By Vacuum Aspiration And By Combination Of Mifepristone And Gemeprost, 47 Contraception 85 (1993) (hereafter "Chan, et al., 1993") (medical termination of pregnancy carries risk of incomplete abortion that requires surgical intervention).

(19.) See Raymond, et al., RU 486: Misconceptions, Myths And Morals 38 (1991) (hereafter "Raymond, et al., 1991") ("Incomplete abortions ... necessitate that the products of conception are removed by conventional abortion methods. Incomplete evacuation can be accompanied by severe bleeding.... This adverse effect of RU 486/PG abortion may lead to ... pelvic inflammatory disease (PID) from infection, to infertility, and possibly uterine cancer.") For statistics on the incidence of infection with pharmaceutical abortion, see Table 3.

(20.) See Pons, et al., Development After Exposure To Mifepristone In Early Pregnancy, 338 Lancet 763 (1991) (hereafter "Pons, et al., 1991") (although the researchers couldn't determine whether the abnormalities were related to RU 486, they concluded that "a deleterious effect of mifepristone cannot be ruled out,"). See also infra Section 4.b.(1) and accompanying footnotes.

(21.) Micrognathia is an abnormality characterized by smallness of the jaw, especially the underjaw. Stedman's Medical Dictionary 875 (5th ed. 1982). Hygroma is a cystic swelling, usually of the neck area, containing serous fluid. See Stedman's Medical Dictionary 668 (5th ed.); Dorland's Illustrated Medical Dictionary 789 (28th ed. 1994).

(22.) Swahn, et al., Effect Of Oral Prostaglandin E2 On Uterine Contractility And Outcome Of Treatment in Women Receiving RU 486 (Mifepristone) For Termination Of Early Pregnancy, 4 Hum. Reprod. 21, 27 (1989) (hereafter "Swahn, et al., 1989"). See also, Zheng Shu-rong, RU 486 (Mifepristone): Clinical Trials In China, 149 Acts Obst. Gyn. Scand. Suppl. 19 (1989) (hereafter "Zheng Shurong, 1989") (4 patients (1.34%) suffered heavy bleeding, necessitating emergency curettage after receiving RU 486); The RU 486 Collaboration Group, Termination Of Early Pregnancy By RU 486 Alone Or In Combination With Prostaglandin, 25 Chinese J. Obat. & Gyn. 31 (1990) (clinically significant [177.4 ml] mean blood loss after complete abortion reported).

(23.) Researchers have noted that the total amount of blood lost by women treated with RU 486 is not reduced with a RU 486/PG regimen. Swahn, et al., 1989, at 27 (1989); Rodger & Baird, Blood Loss Following Induction Of Early Abortion Using Mifepristone (RU 486) And A Prostaglandin Analogue (Gemeprost), 40 Contraception 439 (1989) (hereafter "Rodger & Baird, 1989"); Cameron, et al., Therapeutic Abortion In Early Pregnancy With Antiprogestogen RU 486 Alone Or In Combination With Prostaglandin Analogue (Gemeprost), 34 Contraception 459 (1986). But see, Zheng Shu-rong, 1989 (reporting a lower volume of blood loss with RU 486/PG regimen [52 ml (RU 486/PC) v. 117 ml (RU 486)] in women who aborted within 49 days amenorrhea).

(24.) Thong & Baird, Induction Of Abortion With Mifepristone And Misoprostol in Early Pregnancy, 99 Br. J. Obst. Gyn. 1004, 1006 (1992) (hereafter "Thong & Baird, 1992"). See also Couzinet, et al., Termination Of Early Pregnancy By The Progesterone Antagonist RU 486 (Mifepristone), 315 N. Eng. J. Med. 1565, 1569 (1986); SitrukWare, et al., The Use Of The Antiprogestin RU 486 (Mifepristone) As An Abortifacient In Early Pregnancy--Clinical And Pathological Findings; Predictive Factors For Efficacy, 41 Contraception 221, 239-40 (1990); El-Refaey & Templeton, Early Induction Of Abortion By A Combination Of Oral Mifepristone And Misoprostol Administered By The Vaginal Route, 49 Contraception 111, 113-14 (1994) (hereafter "El-Refaey & Templeton, 1994"); Rodger & Baird, 1989, at 444; UK Multicentre Trial, The Efficacy And Tolerance Of Mifepristone And Prostaglandin in First Trimester Termination Of Pregnancy, 97 Br. J. Obst. Gyn. 480, 485 (1990) (hereafter "UK Multicentre Trial, 1990"); Chan, et al., 1993, at 85; Wu , et al., Clinical Trial On Termination Of Early Pregnancy With Ru 486 in Combination With Prostaglandin, 46 Contraception 203, 209 (1992) (hereafter "Wu, et al., 1992").

(25.) For statistics on complete abortion rates using RU 486 alone, see Table 1 (average complete abortion rate [approx.] 63%); compare to complete abortion rates of 84.4% to 99% using RU 486/oral or vaginal misoprostol up to 63 days amenorrhea (Table 4). Complete abortion rates of 75.3% to 99% using RU 486 in combination with other prostaglandins up to 63 days amenorrhea are reported in Table 5.

(26.) McKinley, et al., The Effect Of Dose Of Mifepristone And Gestation On The Efficacy Of Medical Abortion With Mifepristone And Misoprostol, 8 Hum. Reprod. 1502 (1993) (hereafter "McKinley, et al., 1993") (regimen of either 200 or 600 mg RU 486 followed by 600 ug misoprostol 48 hrs. later).

(27.) Id.

(28.) Id. at 1502. See also Ulmann, Warning On Low Dose Mifepristone Use, 6 PharmacoEconomica 90 (1994) ("Currently available information suggests that the efficacy of misoprostol following mifepristone is significantly lower in pregnancies above 49 days of amenorrhea. The efficacy rate drops to only 90% for pregnancies between 49 and 63 days of amenorrhea, a value that is medically unacceptable.").

(29.) Auheny & Baulieu, Contragestion With RU 486 And An Orally Active Prostaglandin, 312 C.R. Acad. Sci. Paris (III) 539 (1991) (31%); Thong & Baird, 1992 (21%); Peyron, et al., Early Termination Of Pregnancy With Mifepristone (RU 486) And The Orally Active Prostaglandin Misoprostol, 328 N. Eng. J. Med. 1509 (1993) (hereafter "Peyron, et al., 1993") (33.2% in study land 25.4% in study 2); McKinley, et al., 1993 (36.4% and 26.4% in the two groups given 200 or 600 mg RU 486); Thong, et al., What Do Women Want During Medical Abortion?, 46 Contraception 435 (1992) (hereafter "Thong, et al, 1992") (29%).

(30.) Thong, et al., 1992, at 440.

(31.) See Thong & Baird, 1992, at 1006, reporting: The 3% incidence of ongoing pregnancies with this regimen is slightly higher than that reported using a combination of mifepristone with gemeprost or sulprostone (Sylvestre et al. 1990; UK Multicentre Trial, 1990). We have previously reported ongoing pregnancies in two out of 21 women (up to 56 days amenorrhea) who were given 200 mg mifepristone followed by 200 ug and 400 ug of misoprostol (Norman et al. 1991). Therefore, in a consecutive series of 121 women treated with mifepristone and misoprostol in our centre, the ongoing pregnancy rate was 4% (95% CL 0.6-8.6%). In a consecutive series of 470 pregnancies (<56 days) terminated with mifepristone (50-600 mg) and gemeprost (0.5 mg-1 mg) in our institution, there has been only one ongoing pregnancy (0.2%). If the higher incidence of ongoing pregnancies is confirmed by a larger study..., the clinical usefulness of this combination of mifepristone and oral misoprostol for routine clinical use would be in doubt.

Two other British researchers reported a3% (5 patients) ongoing pregnancy rate in a series of 150 patients receiving oral misoprostol 800 ug after 200 mg RU 486. EI-Refaey & Templeton, 1994, at 112.

(32.) Peyron, et al., 1993.

(33.) For data on RU 486/orel misoprostol abortion, see Table 6 (pain reported in from 79.1 to 85% patients; analgesia needed in from 12.5 to 57.1% patients). For data reported for RU 486 with other prostaglandins, see Table 7. See also WHO, Pregnancy Termination With Mifeprislene And Gemeprost: A Multicenter Comparison Between Repeated Doses And A Single Dine Of Mifepristone, 56 Fertility & Sterility 32,39 (1991) (hereafter "WHO, 1991") ("As anticipated, mifepristone-induced uterine contractions and vaginal bleeding were associated with lower abdominal pain, which became almost universal after gemeprost administration."); UK Multicentre Trial, 1990, at 484 (where 16% reported mild, moderete or severe pain the first 24 hours after mifepristone compared to 84% who reported pain 2 hours after gemeproat); Thong & Baird, 1992, at 1005 (11% reported abdominal pain before misoprostol administration and 85% 2 hours afterwards); McKinley, et al., 1993, at 1504 (53.6% reported abdominal pain before misoprestol adminis tration and 79.1% 2 hours afterwards).

(34.) Zheng Shu-rong, 1989, at 22 (reporting a higher incidence of abdominal pain resulting from uterine cramping and diarrhea in women given RU 486 plus a PG); Thong & Baird, 1992, at 1005-06 (reporting an expected increase in PG-related side effects 2 hrs. following administration of misoprostol, including vomiting, pain, faintness and diarrhea); McKinley, et al., 1993, at 1504 (reporting increases in vomiting, pain, diarrhea and fainting 2 hrs. after taking misoprostol); UK Multicentre Trial, 1990, at 484 (where 3% and 0.5% reported vomiting and diarrhea during the first 4 hours after mifepristone compared to 26% and 13% during the first 4 hours after gemeprest). See also Sachs, A., Abortion Pills en Trial, Time, Dec. 5, 1994, at 45-46.

(35.) See Ulmann, et al., Medical Termination Of Early Pregnancy With Mifepristone (RU 486) Followed By A Prostaglandin Analogue, 71 Acta Obst. Gyn. Scand. 278 (1992) (hereafter "Ulmann, et al., 1992") (reporting 3 myocardial infarctions (1 fatal) and 3 cases of severe hypotension after sulprostone injection); Anonymous, A Death Associated With Mifepristone/Sulprostone, 337 Lancet 969 (1991) (discussing the name death reported in Ulmann, et al., 1992). See also institute of Medicine, Clinical Applications Of Mifepristone (RU 486) And Other Antiprogetins 27 (1993) (hereafter "Institute of Medicine, 1993") (reporting one patient death during the first trial of RU 486 with oral misoprostol).

(36.) WHO Task Force On Post-ovulatory Methods Of Fertility Regulation, Termination Of Pregnancy With Reduced Doses Of Mifepristone, 307 BMJ 532 (1993) (hereafter "WHO Task Force, 1993") (reported significant decrease in pulse rate during the first few hours after gemeprost (3-4 beats/min.); WHO, 1991 (reported significant (P<0.001) decrease in pulse rate during 4-hour period after gemeprost).

(37.) Peyron, et al., 1993, at 1511.

(38.) Id, (researchers attributed the hypotension to a vagal reaction secondary to painful uterine cramps).

(39.) This list is based on expert review of the medical literature cited in Attachment 2.

(40.) Wilcox et al., Hysterectomy in The United States, 1988-1990, 83 Obst. & Gyn. 549 (1994) (fibroid tumor was reported as the primary diagnosis for 61% of African-American women and 29% of Caucasian women having hysterectomy); Kjerulff, et al., Hysterectomy And Race, 82 Obst. & Gyn. 757 (1993) (in study of more than 53,000 hysterectomies, African-American women were more than twice as likely to have a diagnosis of uterine fibroids as Caucasian women).

(41.) Gohdes, et al., Diabetes In American Indiana, 16 Suppl. 1 Diabetes Care 239 (1993); Gohdes, Diabetes In American Indians: A Growing Problem, 9 Diabetes Care 609 (1986); Freeman, et al., Diabetes In American Indians Of Washington, Oregon, And Idaho, 12 Diabetes Care 282 (1989); Knowler, et al, Diabetes Mellitus In The Pima Indians: Incidence, Risk Factors And Pathogenesis, 6 Diabetes/Metabolism Reviews 1 (1990); Knowler, et al., Diabetes Incidence And Prevalence In Pima Indians: A 19-Fold Greater Incidence Than In Rochester, Minnesota, 108 Am. J. Epidem. 497 (1978); Gardner, et al., Prevalence Of Diabetes In Mexican Americans--Relationship To Percent Of Gene Pool Derived From Native American Sources, 33 Diabetes 86 (1984); Carter, et al., Diabetes Mortality Among New Mexico's American Indian, Hispanic, And Non-Hispanic White Populations, 1958-1987, 16 Suppl. 1 Diabetes Care 306 (1993).

(42.) Grimes, et al., 1990, at 913-14 (risk with a Quetelet's index [weight(kg)/height(m2)] greater than 23.81 was 2.9 times the risk with index less than 20.25); see also Grimes, 1993, at 172.

(43.) See WHO Task Force, 1989, at 722, 724 (subjects where RU 486/sulprostone failed weighed 65.7 + or - 10.2 kg; those with complete abortion, 54.7 + or - 8.3 kg; and those with incomplete abortion, 54.5 + or - 10.6 kg). But see Thong & Baird, 1992 (complete abortion rate not influenced by body mass index; however, the study focused on the clinical efficacy of RU 486/misoprostol abortion in only 100 subjects, not on predictors of failed attempted abortion).

(44.) WHO, 1991, at 35, 38 (women with vaginal bleeding prior to gemeprost administration had significantly greater weight (56.1 + or - 7.5 kg) and ponderal index (2.14 + or - 0.26) than those who started to bleed after gemeprost (53.7 + or - 7.4 kg; 2.06 + or -0.23).

(45.) See discussion infra Section 4.a.(5) and accompanying footnotes.

(46.) One study found a higher rate of failure in patients over the age of 34. Thonneau, et al., Analysis Of 369 Abortions Conducted By Mifepristone (RU 486) Associated With Sulprostone In A French Family Planning Center, 61 Fertility & Sterility 627, 629-30 (1994) (hereafter "Thonneau, et al., 1994") ("The patient characteristics that correlated with failure were age over 34 years (6% <35 years versus 14% >34 years; P=0.05)....") ("We are not aware of other studies indicating that age (>35 years) is a risk factor for failure. However, age over 35 has been considered to be a risk factor for morbidity....").

(47.) For data on RU 486/oral misoprostol abortion, see Thong & Baird, 1992 (4% of users: 1 woman needed emergency curettage for heavy bleeding due to incomplete abortion; 3 women were given intramuscular ergometrine for heavy bleeding at the time of expulsion of products); Peyron, et al., 1993 (0.4% of patients (2 women) in study 1 had hemorrhage requiring hemostatic curettage, with one patient needing a blood transfusion 9 days after misoprostol when hemoglobin fell from 13.0 g/dl to 6.1 g/dl); McKinley, et al., 1993 (5.5% had a drop in hemoglobin >2 g/dl, but no blood transfusions). For additional data on this complication with RU 486 and other prostaglandins, see Table 8.

(48.) See Table 6 for pain data on RU 486/oral misoprostol abortion. For pain data on RU 486 with other prostaglandins, see Table 7.

(49.) For infection data on RU 486 with prostaglandins other than misoprostol, see Table 3.

(50.) For data on RU 486/oral misoprostol abortion, see Peyron, et al., 1993, at 1511 (6 women (1.2%) had a "substantial but transient decrease in blood pressure (more than 30 mm Hg for the systolic pressure and 15 mm Hg for the diastolic pressure) attributable to a vagal reaction secondary to painful uterine cramps."). For data on RU 486 with other prostaglandins, see Ulmann, et al., 1992 (reporting 3 myocardial infarctions (1 fatal) and 3 cases of severe hypotension after sulprostone injection).

(51.) FDA should consider that complications with pharmaceutical abortion may result regardless of whether the abortion is complete. See Birth Control Trust, Mifepristone In Practice: Running An Early Medical Abortion Service, 38 (1994) (hereafter "Birth Control Trust, 1994") (an early complication rate of 14% among women who had a medical abortion (RU 486/gemeprost) even though 94% had a complete abortion); A World Health Organization ("WHO") study reported that 2.6% of the women with complete abortions (92.7%) required antibiotics to prevent or cure suspected genitourinary infection during a six week follow-up period. WHO, 1991, at 37. See also, WHO Task Force, 1989, at 722 (1.3% of subjects w/complete abortion (88.8% of subjects) given antibiotic therapy because of clinically suspected endometritis).

(52.) The three largest studies only followed patients for approximately one week alter PG administration. Ulmann, et al., 1992, at 279 ("approximately one week later... the final outcome of treatment was evaluated."); Aubeny, RU 486 Combined With PG Analogs In Voluntary Termination Of Pregnancy, 7 Adv. Contraception 339, 341 (1991) ("fourth visit took place on day 8-12 after expulsion."); Wu, et al., 1992, at 204 ("On the 8th day after medication, the woman returned to the clinic for evaluation of the result .... If complete abortion could not be confirmed, further follow-up on the 14th day was required."). The remaining studies typically follow patients until the onset of their next menstrual period (one, two and four or six weeks after the RU 486 abortion procedure). See, e.g., UK Multicentre Trial, 1990 and the WHO studies.

(53.) See infra section 3.f. and accompanying footnotes for statistics on lack of patient compliance.

(54.) Rodger, et al., Induction Of Early Abortion With Mifepristone (RU 486) And Two Different Doses Of Prostaglandin Pessary (Gemeprost), 39 Contraception 497, 501 (1989) (hereafter, "Rodger, et al., 1989") ("Careful follow-up is essential following treatment to exclude the presence of a continuing pregnancy."); Peyron, et al., warned that "[e]ctopic pregnancy is difficult to detect very early, and its possible occurrence makes a follow-up visit 8 to 15 days after the treatment mandatory ...." Peyron, et al., 1993, at 1512; Ulmann, et al., 1992, at 283 (RU 486/PG abortion is an "acceptable alternative to surgical procedures, provided that ... the protocol recommended by the manufacturer is strictly followed."); Rodger & Baird, Induction Of Therapeutic Abortion in Early Pregnancy With Mifepristone In Combination With Prostaglandin Pessary, Lancet ii: 1415, 1417-1418 (1987) ("The occurrence of incomplete abortion after medical termination of pregnancy ... makes careful follow-up a necessity.").

(55.) UK Multicentre Trial, 1990. A study by Ulmann, et al., indicates a lower, but still significant incidence of noncompliance: (1) 0.3% of women given RU 486 were lost to follow-up prior to PG administration; and (2) 0.8% never received a PG even though they had not aborted. Thus, 1.1% of women who received RU 486 as part of an RU 486/PG protocol did not return and/or refused to take the prostaglandin analog. In addition, 2.6% of the women in the study were lost to follow-up after RU 486/PG administration. Ulmann, et al., 1992, at 280 (There were other protocol violations reported in this study: 11.6% took PG either before or after the protocol time (36-48 hr. after RU 486 intake); and 13.6% had pregnancies beyond the 7 week protocol cut-off (i.e., more than 49 days amenorrhea calculated from the first day of the last menstrual period). Id.

(56.) See also Grimes, et al., 1990 (2.5% of patients were lost to follow-up after RU 486 administration; study did not include PG); Peyron, et al., 1993 (2.8% of the women in study I did not return for follow-up after RU 486/misoprostol administration; and 27.6% of the women in study 2 who had not aborted within 4 hours after 400 ug of misoprostol declined to take an additional 200 ug dose); Henshaw, et al., Comparison Of Medical Abortion With Surgical Vacuum Aspiration: Women's Preferences And Acceptability Of Treatment, 307 BMJ 714 (1993) (4% of patients in a study comparing medical to surgical abortion did not return for follow-up visit 16 days later); Hill, at a)., The Efficacy Of Oral Mifepristone (RU 38,486) With A Prostaglandin El Analog Vaginal Peasary For The Termination ion Of Early Pregnancy: Complications And Patient Acceptability, 162 Am. J. Obst. Gyn. 414 (1990) (7%, 15%, and 13% of patients did not attend for follow-up at 7, 14, and 28 days after PG administration, respectively); Thonneau, et al., 1994, at 628-29 ("Sixteen patients (4.3%) did not return for follow-up on day 14, and no data about either efficacy or complications are available for these women."); WHO Task Force, 1993, at 534 ("Two women refused the gemeprost pessary .... Six other women received both mifepristone and gemeprost and attended the follow up visit one week later, but they defaulted from attending further follow up visits and attempts to contact them failed."); Indian Council of Medical Research Task Force On Hormonal Contraception, A Multicentre Clinical Trial With RU 486 Followed By 9-Methylene-PGE2 Vaginal Gel Far Termination Of Early Pregnancy: A Dose-Finding Study, 49 Contraception 87, 91, 97 (1994) (4 patients (0.88%) did not come back for the PG gel: 3 for "personal reasons" and 1 was lost to follow-up); Maria & Stampf, Termination Of Early Pregnancy Using Mifepristone one In Combination With Prostaglandin Analogs, 149 Acts Obst. Gyn. Scand. Suppl. 31 (1989) (4.9% (13 patients) were lost from study and did not rece ive the follow-up exam on day 10); Swahn & Bygdeman, Termination Of Early Pregnancy With RU 486 (Mifepristone) In Combination With A Prostaglandin Analogue (Sulprostone), 68 Acta Obat. Gyn. Scand. 293, 294 (1989) (hereafter "Swahn & Bygdeman 1989") (0.85% (one woman) withdrew from study 24 hours after RU 486 administration--prior to PG administration--due to nausea); WHO Task Force, 1989, at 720 (0.4% (one woman) stopped RU 486 and did not receive the PG "for personal reasons unrelated to the treatment ...." And 0.4% (one) was lost to follow-up.); Broome, Using Mifepristone In A Family Planning Clinic, 20 Br. J. Family Planning 11 (1994) (11.2% failed to keep hollow up appointment. "This rate has become much worse in the last few months...").

(57.) Ulmann, et al., 1992, at 280-81 (reporting that the success rate was significantly lower (88.6% instead of 95.3%) in the absence of PG administration and the incidence of ongoing pregnancy was higher (4.6% as opposed to 1.2% overall).

(58.) Id. at 281. Ulmann reported 4.2% of patients who did not receive the PG required vacuum aspiration or D&C because of incomplete abortion (compared to 2.8% of total patients). Also, 2.6% of the group not receiving PG required a hemostatic surgical procedure (compared to 0.7% of total patients).

(59.) The protocol followed in most published studies calls for administration of an anti-D immunoglobulin injection for Rh negative women at the time of PG administration. See Thong, et al., Changes In The Concentration Of Alpha-Fetoprotein And Placental Hormones Following Two Methods Of Medical Abortion In Early Pregnancy, 100 Br. J. Obst. & Gyn. 1111 (1993); Urquhart & Templeton, Reduced Risk Of Isoimmunisation In Medical Abortion, 335 Lancet 914 (1990).

(60.) Id. at 280-81. Ulmann reported that the time lapse between RU 486 and PG intake had a significant effect on the complete abortion rate. This rate was highest (95.8%) at 36-48 hours, but dropped to 92.8% at less than 36 hours, and 93.9% at more than 48 hours.

(61.) UK Multicentre Trial, 1990; Ulmann, et al., 1992, at 283. See also Wu, et al., 1992, at 209 ("It seems that the treatment regimen when followed carefully plays an important role on the effectiveness of the method.").

(62.) "With RU-486, Will More Physicians Provide Abortions?," Amer. Med. News, Apr. 12, 1993, at 3.

(63.) There is some published medical literature on RU 486 which medical experts assume is scientifically valid. However, this assumption may not be warranted and therefore FDA should not rely on the published literature without verification of the raw data supporting the publication.

(64.) Vessey, et al., Oral Contraceptive Use And Abortion Before First Term Pregnancy In Relation To Breast Cancer Risk, 45 Br. J. Cancer 327 (1982) (this study is composed almost entirely of women who had a spontaneous abortion, with "only a handful" of patients who had undergone induced abortion.); Gandra, et al., Risk Factors For Breast Cancer: A Case-Control Study, 6 Acta Medica Portuguesa 129 (1993).

(65.) Lindefors-Harris, et al., Risk Of Cancer Of The Breast After Legal Abortion During First Trimester: A Swedish Register Study, 299 Br. Med. J. 1430 (1989) (authors achieved a borderline significant negative association only by inexplicably excluding women aborted after age 30, and by comparing aborted women to the general population rather than to a bona fide control group. Since the general population had a 20% higher nulliparity rate than the study population of aborted women (49% versus 41%), the known protective effect of parity could account for the apparent protective effect of abortion.).

(66.) Several studies are published in more than one report.

(67.) See Dvoirin & Medvedev, Role of women's reproductive status in the development of breast cancer, Methods And Progress In Breast Cancer Epidemiology Research 53 Tallinn, Estonia (1978) (hereafter "Dvoirin & Medvedev, 1978"); La Veechia, et al., Long-Term Impact Of Reproductive Factors On Breast Cancer, 53 Int. J. Cancer 215 (1993) (hereafter "La Vecchia, et al., 1993") (This study has appeared in at least four separate reports.); Ewertz & Duffy, Risk Of Breast Cancer In Relation To Reproductive Factors In Denmark, 58 Br. J. Cancer 99 (1988) (hereafter "Ewertz & Duffy, 1988"); Brinton, et al., Reproductive Factors In The Aetiology Of Breast Cancer, 47 Br. J. Cancer 757 (1963) (hereafter "Brinton, et al., 1983"); Rosenberg, et al., Breast Cancer In Relation To The Occurrence And Time Of Induced And Spontaneous Abortion, 127 Am. J. Epidem. 981 (1988) (hereafter "Rosenberg, et al., 1988"); Moseson, et al., The Influence Of Medical Conditions Associated With Hormones On The Risk Of Breast Cancer, 22 Int. J. E pidem. 1000 (1993) (hereafter "Moseson, et al., 1993"); Dating, et al., Risk Of Breast Cancer Among Young Women: Relationship To Induced Abortion, 86 J. Nat'l Cancer Inst. 1584 (1994) (hereafter "Daling, et al., 1994").

(68.) Controls are generally younger than the cancer patients so relative risk estimates tend to be underestimated and confidence intervals are wider.

(69.) This study reports relative risks of 1.2-1.3, despite the fact that the median patient age was 12 years greater than the median control age.

(70.) See Adami, et al., Absence Of Association Between Reproductive Variables And The Risk Of Breast Cancer in Young Women In Sweden And Norway, 62 Br. J. Cancer 122 (1990) (hereafter "Adami, et al., 1990"); Nishiyama, The Epidemiology Of Breast Cancer In Tokushima Prefecture, 38 Shikoku Med. J. 333 (1982) (hereafter "Nishiyama, 1993"); Hirohata, et al., Occurrence Of Breast Cancer In Relation To Diet And Reproductive History: A Case-Control Study In Fukuoka, Japan 69 Natl. Cancer Inst. Monogr. 187 (1985) (hereafter "Hirohata, et al., 1985"); Le, et al., "Oral Controceptive Use And Breast Or Cervical Cancer:) Preliminary Results Of A French Case-Control Study, Hormones And Sexual Factors In Human Cancer Aetiology 139 (Elsevier, Amsterdam 1984) (hereafter "Le, et al., 1984"); Andrieu, et al., Familial Risk Of Breast Cancer And Abortion, 18 Cancer Detection & Prevention 51 (1994) (hereafter "Andrieu, et al., 1994"); Pike, et al., Oral Contraceptive Use And Early Abortion As Risk Factors For Breast Cancer In Yo ung Women, 43 Br. J.) Cancer 72 (1981) (hereafter "Pike, et al., 1981"); Howe, et al., Early Abortion And Breast Cancer Risk Among Women Under Age 40, 18 Int. J. Epidem. 300 (1989) thereafter "Howe, et al., 1989"); Laing, et al., Breast Cancer Risk Factors In African-American Women: The Howard University Tumor Registry Experience, 85 J. Nat'l Med. A. 931 (1993) (hereafter "Laing, et al., 1993").

(71.) But see Parazzini, et al., Menstrual And Reproductive Factors And Breast Cancer In Women With Family History Of The Disease, 51 Int. J. Cancer 677 (1992) (a large case-control study in Milan, Italy). In Italy, over three-quarters of legal abortions occur among women who have already had one or more children. See FigaTalamanca, et al., Epidemiology Of Legal Abortion In Italy, 15 Intl. J. Epidem. 343 (1986). This means that these women are at a lower risk of breast cancer anyway, because of the previous live birth.

(72.) Lindefors-Harris, et al., Response Bias In A Case-Control Study: Analysis Utilizing Comparative Data Concerning Legal Abortions From Two Independent Swedish Studies. 134 Am. J. Epidem. 1003 (1991); Lindefors-Harris, et al., Risk Of Cancer Of The Breast After Legal Abortion During First Trimester: A Swedish Register Study, 299 Br. Med. J. 1430 (1989).

(73.) For a general discussion of the biological changes in breast tissue during pregnancy, see Russo, et al., Differentiation Of The Mammary Gland And Susceptibility To Carcinogenesis, 2 Breast Cancer Res. Treat. 5 (1982) (hereafter "Russo, et al, 1982").

(74.) Brooks S.C. & Pauley R.J., Breast Cancer Biology, Encyclopedia Of Human Biology (R. Dulbecco, ed. 1991). A full-term pregnancy protects against breast cancer by bringing breast cells into their specialized forms. These mature cells have almost no vulnerability to cancer. Abortion interrupts this process, leaving terminal end buds (immature cells) suspended in high risk transitional states.

(75.) Russo, et al., 1982.

(76.) Rosenberg, Induced Abortion And Breast Cancer: More Scientific Data Are Needed, 86 J. Nat'l Cancer Inst. 1569 (1994).

(77.) Rosenberg. et al., 1988.

(78.) Soini, Risk Factors Of Breast Cancer In Finland, 6 Intl. J. Epidem. 365 (1977); Hadjiauchael, et al., Abortion Before First Livebirth And Risk Of Breast Cancer, 53 Br. J. Cancer 281 (1986).

(79.) Lehrer, et al., An Estrogen Receptor Polymorphism And A History Of Spontaneous Abortion--Correlation In women With Estrogen Receptor Positive Breast Cancer But Not in Women with Estrogen Receptor Negative Breast Cancer Or In Women Without Cancer, 26 Breast Cancer Res. Treat. 175 (1993).

(80.) Lentz, The Phylogency Of Oncology, 2 Mol. Biother. 137 (1990).

(81.) Gatanaga, et al., Identification Of TNF-LT Blocking Factor(s) In The Serum And Ultrafiltrates Of Human Cancer Patients, 9 Lymphokine Res. 225 (1990a); Gatanaga, et al., Purification And Characterization Of An Inhibitor (Soluble Tumor Necrosis Factor Receptor) For Tumor Necrosis Factor And Lymphotoxin Obtained From The Serum Ultrafiltrates Of Human Cancer Patients, 87 Proc. Nat'l Acad. Sci. 8781 (1990b).

(82.) Lentz & Saltonstahl, Apheresis Of Law Molecular Weight Protein Fraction And The Onset Of Labor, 5 J. Clin. Apheresis 62 (1990) (hereafter "Lentz & Saltonstahl, 1990").

(83.) Lentz, Continuous Whole Blood Ultrapheresis Procedure In Patients With Metastatic Cancer, 8 J. Biol. Response Modif. 511 (1989).

(84.) Lentz & Saltonstahl, 1990.

(85.) Clark & Chua, Breast Cancer And Pregesncy: The Ultimate Challenge, 1 Clin. Oncol, 11 (1989).

(86.) Spitz & Bardin, Clinical Pharmacology Of RU 486--An Antiprogestin And Antiglucocorhicoid, 48 Contraception 403 (1993) (hereafter "Spitz & Bardin, 1993"); Weiss, RU 486: The Progesterone Antagonist, 2 Arch. Fam. Med. 63(1993).

(87.) See, e.g., Healy, Clinical Status Of Antiprogesterone Steroids, 3 Clin. Reprod. & Fertility 277, 284(1985) ("RU 486-induced blockade of the cortisol receptor may prevent the usual glucocorticoid stress response to anaesthesia and surgery in a patient who needs curettage after RU 486 treatment. This might make anaesthesia complex in such patients.").

(88.) See Laue, et al., Effect Of Chronic Treatment Will: The Glucocorticoid Antagonist RU 486 In Man: Toxicity, Immunological, And Hormonal Aspects, 71 J. Clin. Endocrin. & Metab. 1474(1990) (hereafter "Laue, et al., 1990") (in study designed to examine immune function, blockade of cortisol receptors with RU 486 in 11 healthy males was associated with marked compensatory elevations of plasma ACTH and cortisol; RU 486 (10 mg/kg/day) was administered twice a day for 7-14 days; however, one subject developed signs and symptoms consistent with adrenal insufficiency).

(89.) Brogden, et al., Mifepristone: A Review Of Its Pharmacodynamic And Pharmacokinetic Properties, And Therapeutic Potential, 45 Drugs 384,405 (1993) (hereafter "Brogden, et al. 1993"); Heikinheimo, Anitiprogesterone Steroid RU 486: Pharmacokinetics And Recptor Binding In Humans. 69 Acts Obstet. Gynecol. Scand. 357(1990).

(90.) See Grimes, et al., 1990, at 913-14 (reporting that risk of failure of RU 486 as single-agent abortifacient was 2.9 times greater for obese women than for women in the lowest body mass group studied). See also infra Section 3.d. and accompanying footnotes.

(91.) Spitz & Bardin, 1993, at 409; Heikinheimo, Antiprogesterone Steroid RU 486: Pharmacokinetics And Receptor Binding In Humans, 69 Acts Obstet. Gyn. Scand. 357(1990).

(92.) Lahteenmaki, et al., Pharmacokinetics And Metabolism Of RU 486, 27 J. Steroid Biochem. 859 (1987); see also, Heikinheimo, et al., Pharmacokinetics Of The Antiprogesterone RU 486: No Correlation To Clinical Performance Of RU 486, 123 Acta Endocrinologica 298 (1990).

(93.) Avrech, Mifepristone (RU 486) Alone Or In Combination With A Prostaglandin Analogue For Termination Of Early Pregnancy: A Review, 56 Fertility & Sterility 385,386 (1991).

(94.) CDC, "Abortion Surveillance--United States, 1990," 42 (SS-6) Morbidity & Mortality Weekly Report 29 (Dec. 17, 1993).

(95.) See Birth Control Trust, 1994, at 43.

(96.) WHO, 1991, at 35 (when both study groups [repeated doses and single dose RU 486] were combined, the difference between Chinese and non-Chinese subjects was significant [Chinese women: median--12 days bleeding; range--3 - 45 days, 95th percentile: 37.6 days, n=115; non-Chinese women: median: 10 days bleeding; range 2-54 days, 95th percentile: 23.8 days, n=224; p<0.01]). But see, WHO Task Force, 1993, at 534 (reporting no significant difference between Chinese women and non-Chinese subjects).

(97.) Chan, et al., 1993 (2 women (2.08%) required emergency suction evacuation for heavy bleeding; because of heavy bleeding, researchers concluded that "strict supervision is mandatory" for RU 486/PG abortion).

(98.) Id. at 93 (commenting on a study conducted by Rodger & Baird, see Rodger & Baird, 1989).

(99.) Id. at 93 (citing Wong, et al., The Effect Of Oral Contraceptives On Coagulation And Fibrinolytic Parameters In The Chinese--A Prospective Study, 48 Thromb. Haemostas. (Stuttgart) 263 (1982).

(100.) Id.

(101.) WHO, 1991, at 37; Spitz & Bardin, 1993, at 411. See also WHO Task Force, 1989, at 721; Swahn & Bygdeman, 1989, at 298; Shoupe, et al., 1986, at 457; Swahn, et al., 1989, at 24.

(102.) Herting & Nissen, Overview Of Misoprostol Clinical Experience, 31 Dig. Diseases & Sci. 47S, 51S (Feb. 1986 Suppl.) (hereafter "Herting & Nissen, 1986").

(103.) Namely, glucocorticoids reduce fluid movement from the blood to the tissues; decrease the permeability of blood vessels. This limits leukocyte migration to the site of tissue injury and reduces the body's ability to fight invading bacteria. Also research suggests that glucocorticoids affect transport of glucose, amino acids, and RNA in lymphocytes; and down regulate gene transcription within the lymphocyte. They also affect the intracellular ability of neutrophils to destroy ingested microorganisms and affect protein expression in lymphoid tissues and lymphoid function. Schulster, et al., Molecular Endocrinology Of The Steroid Hormones 282 (1976).

(104.) Laue, et al., 1990; Van Voorhis, et al., The Effects Of RU 486 On Immune Function And Steroid-Induced Immunosuppression In Vitro, 69 J. Clin. Endocrin. & Metab. 1195 (1989) (hereafter "Van Voorhis, et al., 1989"); Bertagna, et al., Peripheral Antiglucocorticoid Action Of RU 486 In Man, 28 Clin. Endocrin. 537 (1988); Emilie, et al., Inhibition Of In Vitro Immunosuppressive Effects Of Glucocorticosteroids By A Competitive Antagonist RU-486, 8 Immunology Letters 183 (1984).

(105.) Van Voorhis, et al., 1989.

(106.) Redgrave, et al., An In Vitro Comparison Of The Immunosuppressive Potential Of Synthetic Prostaglandin Analogues, 23 Transplant. Proceed. 346 (1991) (an in vitro study which reported that a PGE1analog enhanced the immunosuppressant effects of cyclosporine); Moran, et al., Prevention Of Acute Graft Rejection By The Prostaglandin E1 Analogue Misoprostol In Renal-Transplant Recipients Treated With Cyclosporine And Prednisone, 322 N. Eng. J. Med. 1183, 1187 (1990) (hereafter "Moran, et al., 1990") (reporting that misoprostol acted in a synergistic manner with two known immunosuppressant drugs--cyclosporine and prednisone (a glucocorticoid) in an in vivo study of kidney transplant recipients). By itself, misoprostol has not demonstrated any adverse effects on the immune system. See Waymack, et al., Effect Of Prostaglandin E On Immune Function In Normal Healthy Volunteers, 175 Surg. Gyn. & Obstet. 329 (1992); Herting & Nissen, 1986, at 51S; Moran, et al., 1990, at 1187.

(107.) Moran, et al., 1990, at 1187

(108.) Herve, et al., Evidence For Differences In The Binding Of Drugs To The Two Main Genetic Variants Of Human Alpha 1-Acid Glycoprotein, 36 Br. J. clin. Pharmac. 241 (1993); Bree, et al., Comparison Of Drug Binding Capacities Of Three AAG Glycan Variants Of Human Origin, 300 Prog. Clin. Biol. Res. 405 (1989).

(109.) See Grimes, et al., 1988, at 1311.

(110.) With RU 486 alone, from 8.3% to 46.3% of pregnancies have continued. See Table 9 for complete statistics. After RU 486/oral misoprostol administration, from 0.45% to 9.5% of pregnancies have continued (see Table 10 for statistics); and after medical abortion with RU 486 and other prostaglandins, from 0.4% to 6.2% of pregnancies have continued (see Table 11 for statistics).

(111.) Hill, et al., Transplacental Passage Of Mifepristone And Its Influence On Maternal And Fetal Steroid Concentrations In The Second Trimester Of Pregnancy, 6 Hum. Reprod. 458 (1991) (hereafter "Hill, et al., 1991"); Hill, et al., The Placental Transfer Of Mifepristone (RU 486) During The Second Trimester And Its Influence Upon Maternal And Fetal Steroid Concentrations, 97 Br. J. Obst. Gyn. 406 (1990b) (hereafter "Hill, et al., 1990b"); Frydman, et al., Transplacental Passage Of Mifepristone, ii Lancet 1252 (1985).

(112.) Raymond, et al., 1991, at 76-79.

(113.) van der Schoot & Baumgarten, Effects Of Treatment Of Male And Female Rats In Infancy With Mifepristone On Reproductive Function In Adulthood, 90 J. Reprod. Fert. 255 (1990).

(114.) Jost, Animal Reproduction--New Data On The Hormonal Requirement Of The Pregnant Rabbit: Partial Pregnancies And Fetal Abnormalities Resulting From A Treatment With A Hormonal Antagonist Given At Sub-Abortive Dosage, 303 C.R. Acad. Sci. III 281 (1986); Silvestre, et al., Voluntary Interruption Of Pregnancy With Mifepristone (RU 486) And A Prostaglandin Analogue, 322 N. Eng. J. Med. 645 (1990) (hereafter "Silvestre, et al., 1990") (deformities were reportedly "attributed to uterine contractions secondary to decreased progesterone activity.").

(115.) Wolf, et al., Tolerance Of Perinidatory Primate Embryos To RU 486 Exposure In Vitro And In Viva, 41 Contraception 85 (1990) (hereafter "Wolf, et al., 1990"); see also, Raymond. et al., 1991, at 78.

(116.) Wolf, et al., 1990, at 90.

(117.) Id. Furthermore, one study in mice found that RU 486 retarded embryonic development in vivo and acted directly on the embryo, interfering with its development in vitro. Yang & Wu, RU 486 Interferes With Egg Transport And Retards The In Vivo And In Vitro Development Of Masse Embryos, 41 Contraception 551 (1990).

(118.) Institute of Medicine, 1993, at 28.

(119.) See Pons, et al., 1991 (although the researchers couldn't determine whether the abnormalities were related to RU 486, they concluded that "a deleterious effect of mifepristone cannot be ruled out."). For explanation of micrognathia and hygroma see infra footnote 21.

(120.) See Pons & Papiernik, Mifepristone Teratogenicity, 338 Lancet 1332(1991). In addition, the published studies on human fetal exposure to RU 486 during the second trimester are inconclusive. Hill, et al., 1991 found no statistically significant changes in fetal concentrations of progesterone, oestradiol or cortisol, but a significant increase in fetal aldosterone occurred 4 and 24 hours after drug intake. However, researchers concluded that "[t]he importance of the increased fetal aldosterone levels is uncertain. In view of the small number of patients in each group this may have occurred by chance and similarly because of the study size, any effect of mifepristone on the other parameters studied cannot be totally excluded." Id. at 461. See also Hill, et al., 1990b.

(121.) Silverstre, et al., 1990; Raymond, et al., 1991 at 89-90; Schonhofer, Brazil: Misuse Of Misoprostol As An Abortifacient May Induce Malformations, 337 Lancet 1534(1991); Collins & Mahoney, Hydrocephalus And Abnormal Digits After Failed First-Trimester Prostaglandin Abortion Attempt, 102 J. Pediatrics 620 (1983).

(122.) Fonseca, et al., Misoprostol And Congenital Malformations, 338 Lancet 56 (1991); fonseca, et al., Misoprostol Plus Mifepristone, 338 Lancet 1594 (1991) (hereafter "Fonseca, et al., 1991").

(123.) Mobius syndrome is characterized by congenital facial diplegia and a developmental bilateral facial paralysis associated with oculomotor or other neurological disorders. See Stedman's Medical Dictionary 1391 (5th ed. 1982).

(124.) Gonzalez, et al., Limb Deficiency With Or Without Mobius Sequence In Seven Brazilian Children Associated With Misoprostol Use In The First Trimester Of Pregnancy, 47 Am. J. Med. Genetics 59 (1993) (hereafter "Gonzalez, et al., 1993").

(125.) Many children exposed to misoprestol in utero appear normal. Schuler, et al., obtained information on 17 babies born to women who took misoprostol as an abortifacient during the first trimester and did not abort. No major malformations were found, although one child had a preauricular tag. Schuler, et al., Teratogenicity Of Misoprostol, 339 Lancet 437 (1992). "However,... we cannot evaluate the exposure risk since we do not know the drug effect in embryos (or fetuses) that are aborted." Gonzalez, et al., 1993, at 64.

(126.) Fonseca, et al., 1991.

(127.) Cekan, et al., Levels Of The Antiprogestlin RU 486 And Its Metabolites In Human Blood And Follicular Fluid Following Oral Administration Of A Single Dose, 4 Hum. Reprod. 131 (1989) (researchers note that "the morphological appearance and cleavage rate of the oocytes fertilized in vitro were not affected by the treatment with RU 486"). However, the fertilized eggs were only developed to the four-to eight-cell stage. And "the developmental capacity of the oocytes after fertilization in vitro could not be fully determined, since the cleaving embryos were not replaced in a recipient uterus." See Raymond, et al., 1991, at 75 (citing Messinis & Templeton, The Effect Of The Antiprogestin Mifepristone (RU 486) On Maturation And In-Vitro Fertilization Of Human Oocytes, 95 Br. J. Obst. & Gyn. 592 (1988) (hereafter "Messinis & Templeton, 1988"). Interestingly, the women who received RU 486 had fewer eggs which fertilized in vitro as compared to the controls, although the difference was not statistically significa nt. Messinis & Templeton, 1988, at 593.

(128.) Spitz & Bardin, 1993, at 417; Grimes, 1993, at 173.

(129.) Wiedemann, et al., in a limited study, examined the role of glucocorticoids in sleep and demonstrated that RU 486 as a glucocorticoid receptor blocker disrupts sleep patterns. See Wiedemann, et al., Antiglucocorticoid Treatment Disrupts Endocrine Cycle And Nocturnal Sleep Pattern, 241 Eur. Arch. Psych. Chin. Neurosci. 372 (1992).

(130.) Hill, et al., 1991 reported "a trend to higher... fetal cortisol concentrations 24 and 48 h after treatment, [however,] this increase failed to reach statistical significance..." Id. at 460-61.

(131.) See additional discussion infra Section 6.

(132.) See Table 4.

(133.) Rodger & Baird. 1989, at 445 ("It is likely that as pregnancy advances and the fetus and placenta increases in size, there is a larger vascular area from which bleeding can occur.").

(134.) According to D. Danforth & J. Scott, there are several methods used to determine gestational age. Because of the unreliability of alternative methods, sonography is recommended as a standard abortion practice for assessing gestational age. Alternative methods include: (1) patient's menstrual history (least reliable method--predictive only to within a margin of 3 weeks with 90% confidence, even if date of last menstrual period is known with certainty); (2) pelvic examination (inaccurate to plus/minus 2 weeks; with retroverted uterus (30% of women) inaccuracy reaches plus/minus 4 weeks); (3) maternal perception of fetal movement (only useful as "rough estimate"); and (4) measurement of fundal height (useful only to corroborate other clinical estimations of gestational age). Danforth & Scott, Obstetrics & Gynecology 263,365-66 (5th ed. 1986); Hern, Abortion Practice 69-70, 109, 207 (1984) (sonography "appears to be considerably more accurate than are menstrual dates and even a careful examination by an ex perienced physician." Any doubt concerning length of gestation should be checked by real-time ultrasound examination.).

(135.) Rodger, et al., 1989, at 501 ("It is important that an intrauterine pregnancy be established prior to treatment as mifepristone seems ineffective in the disruption of ectopic pregnancy."); Weiss, RU 486: The Progesterone Antagonist, 2 Arch. Fam. Med. 63, 66(1993); Levin, et al., Mifepristone (RU 486) Failure In An Ovarian Heterotopic Pregnancy. 163 Am. J. Obst. Gyn. 543(1990); Baulieu, Contragestion And Other Clinical Applications Of RU 486, An Antiprogesterone At The Receptor, 245 Science 1351 (1989).

(136.) WHO Task Force, 1993 (1 patient in the study had undiagnosed tubal pregnancy which ruptured 2 weeks after an RU 486/PG abortion).

(137.) Petitioners are unaware of any studies demonstrating safety and/or effectiveness of RU 486 in the pediatric population.

(138.) UK Multicentre Trial, 1990, at 485 ("the procedure needs to be clinic based, and preferably hospital based, in view of the small but definitive risk of severs hemorrhage."); Roger, et al., 1989, at 501 ("Hospital admission ... for four hours following prostaglandin administration is advisable."); Wu, et al., 1992, at 209 ("It should be emphasized that RU 486 in combination with PG be used only in clinics where emergency facilities are available."); Brogden, et al., 1993, at 404 ("Mifepristone should be administered in an environment where suitably experienced medical personnel and resuscitation equipment are immediately available,"); Thonneau, et al., 1994, at 627 ("the risk of maternal morbidity associated with sulprestone and also the risk of fetal malformations in cases of continued pregnancy indicate that this method should only be used in specialist centers."). See also footnote 24 infra.

(139.) See Testimony Before the Subcomm. on Regulation, Business Opportunities, and Technology of the House Comm. on Small Business, 103d Cong., 2d Sess. (May 16, 1994). Petitioners also urge FDA to compare the misuse of misoprostol as an abortifacient in Brazil where in distribution is not carefully regulated with the carefully controlled distribution of RU 486 in France. See Costa & Vessey, Misoprostol And Illegal Abortion In Rio do Janeiro, Brazil, 341 Lancet 1258 (1993); Coelho, et al., Misoprostol And Illegal Abortion In Fortaleza, Brazil, 341 Lancet 1261 (1993).

Leanne McCoy served as Special Counsel for the Citizens Petition. She is an attorney with Americans United for Life, a public-interest law firm and educational organization. She concentrates on drafting model legislation in the area of wrongful life/wrongful birth and abortion-clinic regulations. She represented four Illinois state's attorneys who objected to the 1989 settlement of Ragsdale v. Turnock before the U.S. District Court for Northern Illinois. She also co-authored a brief in Planned Parenthood v. Casey before the U.S. Supreme Court.