Petition Requesting Enhanced Warnings on Dangers of Lubiprostone (Amitiza)

November 26, 2012, FDA denial of petition on Amitiza

May 6, 2009

Joshua Sharfstein, M.D., Acting Commissioner
U.S. Food and Drug Administration
5600 Fishers Lane
Rockville, MD  20857 

Dear Dr. Sharfstein:

Public Citizen, representing more than 65,000 consumers nationwide, hereby petitions the Food and Drug Administration (FDA), pursuant to the Federal Food, Drug, and Cosmetic Act 21 U.S.C. Section 355(e)(3), and 21 C.F.R. 10.30, to immediately take the following actions with respect to Amitiza (lubiprostone; Sucampo Pharmaceuticals Inc. and Takeda Pharmaceuticals America, Inc.):

  • add a black box warning regarding the risk of abortion to the product label;
  • change the pregnancy category from C to X;
  • contraindicate nursing while taking the drug;
  • require the distribution of an FDA-approved Medication Guide for all patients;
  • mandate a “Dear Doctor” letter.


Current Indications

Lubiprostone was approved for the treatment of chronic idiopathic constipation in adults on January 31, 2006 (48 µg/day), and later approved for irritable bowel syndrome with constipation in women 18 years of age and older on April 29, 2008 (16 µg/day).

However, in women in whom lubiprostone was used to treat irritable bowel syndrome, there was a difference in response between the treated and placebo group of only 6%. In fact, the response to the drug was so low that FDA has requested that the sponsors do another study at a higher dose.[1] A somewhat larger benefit between treated and placebo was seen in women in whom lubiprostone was used to treat chronic idiopathic constipation (20% to 30%).[2] However, other less hazardous drugs for each of these indications are available for pregnant women.

Mechanism of Action

Lubiprostone is a prostaglandin E1 analog in the same family as misoprostol. The lubiprostone mechanism of action emphasized by the manufacturers involves the activation of CIC-2 chloride channels in plasma membranes leading to an extrusion of chloride ions and concomitant fluid secretion.[3] Although the sponsors focus on CIC-2 activity in the intestinal tract, these channels have been found in all rat tissues examined: skeletal muscle, heart, brain, lung, kidney, pancreas, stomach, intestine, and liver.[4] Thus, it is not surprising to find adverse effects in tissues other than the intestinal tract. There is, in fact, an interest in using lubiprostone’s CIC-2 agonist activity in the treatment of cystic fibrosis[5] where the activation of these channels at the apex of ciliated airway cells leads to an increase in fluid secretion diluting the heavy mucus that obstructs small airways in this disease.[6] The fluid secretion into the lungs may be related to the development of dyspnea (difficulty in breathing) that can be an adverse effect of this drug in otherwise healthy patients.

Misoprostol is a drug approved by the FDA for reducing the risk of NSAID-induced gastric ulcers and is widely used off-label in medical abortions to induce uterine contractions after an initial treatment with mifepristone, a progesterone antagonist that terminates pregnancy. The sponsors’ awareness of the similarities between lubiprostone and misoprostol was presumably the reason for doing abortion studies and comparing misoprostol and lubiprostone binding. These studies are not typically performed as part of a new drug application.

In Vitro Data: Studies submitted in the original lubiprostone new drug application showed that lubiprostone and misoprostol had almost identical potencies in causing contractions in isolated guinea pig ileum (longitudinal) muscle.[7] Unfortunately, the sponsors did not compare lubiprostone and misoprostol in isolated guinea pig uterine muscle which would have explored their comparable abilities to cause abortions. An example of such a study is one comparing misoprostol-induced uterine in vitro contractions with those induced by prostaglandin receptor agonists other than lubiprostone.[8]

The sponsor (Sucampo) did fund a study on a commercial source of human smooth muscle uterine cells cultured on glass cover slips, cells that the authors admit "have not been specifically tested for EP [prostaglandin] receptor subtypes."

They measured calcium, cyclic AMP, and plasma membrane potential. This does not appear to be a validated measure of uterine response, and, again, misoprostol was not tested along with lubiprostone.[9]

In Vivo Data: Guinea pigs and primates (humans and monkeys) are the appropriate models for testing prostaglandin analogs in pregnancy (rats are refractory to prostaglandins when progesterone levels are high as they are in pregnancy).[10] Studies in pregnant guinea pigs show clearly that lubiprostone causes dose-dependent abortions, beginning at very low drug levels (two times the human exposure at the maximum recommended human dose), a level of exposure likely within individual variations in drug metabolism. A study in pregnant monkeys was also suggestive of lubiprostone’s abortifacient effects. However, the sponsors never compared lubiprostone and misoprostol in either animal model.

1) Guinea Pig Abortion Study[11]

The abortifacient activity of lubiprostone is evident in the data from the pregnant guinea pig study. A preliminary range-finding study showed that abortions occurred in 4% of pregnant females at a dose of 5 micrograms per kilogram (µg/kg). That study was followed by a second, larger study showing dose-dependent abortions beginning at 1 µg/kg. Abortions occurred in 0%, 3.8%, 8.0%, and 19% (0, 1, 10, 25 µg/kg) (see Figure 1). There were 22 to 26 pregnant guinea pigs for each dose. The 1 µg/kg dose in guinea pigs is approximately the same exposure as the 48 µg dose in women.

Figure 1: Impact of lubiprostone dose upon guinea pig abortion rates

Impact of Lubiprostone Dose Upon Guinea Pig Abortion Rates

2) Monkey Abortion Study[12]

Even though, according to the Pharmacology Reviewer, the doses were not chosen appropriately,[13] there was one abortion at 10 µg/kg and two early deliveries (one at 10 µg/kg and one at 30 µg/kg) with neither outcome in the control group (see Table 1). Since lubiprostone caused no adverse effects in the pregnant females themselves, the early deliveries and abortions cannot be blamed on maternal toxicity and must represent direct toxicity to the developing fetus.

Table 1. Impact of lubiprostone dose upon monkey abortion rates



0 µg/kg

10 µg/kg

30 µg/kg

Number of pregnant monkeys








Early delivery





3) Human Data

There were six pregnancies in the placebo-controlled clinical trials.[14] The label states that, of the six women who became pregnant in clinical trials, four delivered healthy babies when, in fact, one of the “healthy babies” had bilateral club feet. Two outcomes were unknown as one patient was lost to follow-up and one pregnancy was electively terminated. Following the completion of the clinical trials, eight more pregnancies have been reported to the FDA (through September 2008), one with an unspecified congenital anomaly.[15]

In the clinical trials for chronic constipation, almost 2% of women (four out of 239) treated with lubiprostone became pregnant even though women of childbearing potential had to agree to use at least two methods of   contraception, including one barrier method, to enroll in the trial. The requirements for strict precautions against unintended pregnancy were presumably based on the analogy with misoprostol. When pregnancies occur even under the stringent requirements of a clinical trial, it is clear that the warnings and other systems to prevent pregnancy are inadequate. The numbers of women exposed to lubiprostone was not provided for the IBS-C trials.

Consult Reviews on Abortifacient Property of Lubiprostone[16]

The results of the in vivo and in vitro studies caused the reviewing division (Division of Gastrointestinal Products) to ask for a formal consult with the FDA’s Division of Reproductive and Urologic Products.   The two consultant reviewers (Medical and Pharmacology) emphasized the importance of the guinea pig data as the appropriate animal model to study abortifacient drugs because of the ability of guinea pigs to respond to prostaglandins during pregnancy. The Pharmacology Reviewer concluded that, in addition to qualitative similarities, “There also seems to be a reasonable correspondence of abortifacient potency in guinea pigs and that reported in humans [for drugs in general].” The Medical Officer concurred and added that, “The main safety concern is probably the inadvertent exposure to lubiprostone in early pregnancy when the drug might be used by a woman who did not know that she was pregnant.”

The consulting Medical Officer went on to suggest that, “Labeling for lubiprostone should include many of the recommendations presently found in the boxed warnings for misoprostol”. These are:[17]

  • “A negative pregnancy test within 2 weeks prior to beginning therapy;”
  • “The use of effective contraceptive measures;”
  • “Ensuring that the women receive adequate information about the potential risks of lubiprostone for a pregnancy;”
  • “Women should wait until their next menstrual period before beginning use of this product [lubiprostone]”.

Current lubiprostone label: physician information

Warnings and Precautions Section: The only advice currently offered is that, “Women who could become pregnant should have a negative pregnancy test prior to beginning therapy . . . and be capable of complying with effective contraceptive measures”. There is no additional information as to why this is recommended or on the potential serious consequences in failing to warn patients. Misoprostol users, on the other hand, are referred to the black box warnings concerning abortion and premature birth.

Pregnancy Section: Buried in the first paragraph is a mention of “fetal loss” in guinea pigs with no indication to the reader that this means “abortions”, as is the case in the Pregnancy Section of the misoprostol label. The lubiprostone label should state the conclusion of the Pharmacology Reviewer that, “Administration of RU-0211 [lubiprostone] to the pregnant guinea pigs [a valid human model] was associated with dose-dependent abortions.” Also, completely missing in the label is any mention of early deliveries and abortions in monkeys. The Pregnancy section concludes that drug use is a risk/benefit calculation without having provided the reader with information on either side of this equation.

Current Lubiprostone Label: patient counseling information

Misoprostol: Patient information in the misoprostol label begins with the warning not to take the drug if pregnant (“see boxed WARNINGS”). It also has a warning to avoid pregnancy while on the drug, cautioning that it is possible to become pregnant even if on birth control, and what to do if this happens (stop taking the drug and contact your physician immediately).

Lubiprostone: For women taking lubiprostone, on the other hand, the patient label focuses on dosing instructions. There is no information at all on the risks to a pregnant woman. The only other information for patients is a list of the commonest adverse events: severe nausea, diarrhea, and dyspnea.

It is clear that current warnings are not adequate, as evidenced by the occurrence of pregnancies even in the strictly controlled clinical trial setting. As the Medical Officer stated in his review, “Despite the fact that pregnant women were excluded from all clinical trials of RU-0211 [lubiprostone] . . . , four pregnancies were reported during the development of RU-0211.”[18] Public Citizen feels that it is imperative that FDA take the following steps to remedy this situation.

Public Citizen Specific Recommendations

Public Citizen concurs with the conclusions of both review divisions that lubiprostone is a potential abortifacient and recommends that their suggestions for safe use in women of child-bearing potential be included in the label.  We therefore request that FDA immediately implement the following label changes:

1) Black Box Warning: FDA should immediately require a black box warning for lubiprostone similar to that for misoprostol, i.e., require that women have had a negative pregnancy test, be capable of complying with contraceptive measures, and have received both oral and written warnings about the risks in pregnancy.

2) Pregnancy Category: The pregnancy category should be changed from C to X in order to alert women to the potential hazards to the fetus. The FDA defines pregnancy category C to mean that an adverse effect was observed in animal tests but the drug may be used during pregnancy if the potential benefit justifies the potential risk to the fetus. The risk to the fetus (abortion) with lubiprostone does not justify the small benefit provided to the mother, and for which other, safer drugs could be used temporarily. This mandates that the drug, like misoprostol, be placed in category X.

3) Nursing Mothers: The statement that “because of the potential for serious adverse reactions in nursing infants from lubiprostone, a decision should be made whether to discontinue nursing or to discontinue the drug . . .” is not helpful. Although the sponsors did not perform the usual animal study to test for drug secretion in milk, many drugs are secreted in milk, and the potential for serious adverse reactions in infants would appear to outweigh the benefits.  Indeed, diarrhea was seen in nursing infants on a similar drug and diarrhea and nausea were common adverse events in patients taking lubiprostone. Unless the sponsors can demonstrate that lubiprostone is not secreted in milk, nursing should be contraindicated, as it is for misoprostol.

4) Medication Guide: In order to insure that patients receive the needed information, FDA should write a Medication Guide for consumers. As is the case for misoprostol, the following should be included: SPECIAL NOTE TO WOMEN: Amitiza may cause abortion or premature labor if given to pregnant women. Women should be told that they must not be pregnant, must employ an effective contraception method, and should not give the drug to anyone else.

5) Dear Doctor Letter: The manufacturers should immediately send a letter to all Health Care Professionals with all this information. Health Care Professionals should be encouraged to report to the FDA all cases of adverse events related to pregnancy As is the case for misoprostol, the following should be included: SPECIAL NOTE TO WOMEN: Amitiza may cause abortion or premature labor if given to pregnant women. Women should be told that they must not be pregnant, must employ an effective contraception method, and should not give the drug to anyone else.

Environmental Impact Statement

Nothing requested in this petition will have an impact on the environment.


We certify that, to the best of our knowledge and belief, this petition includes all information and views on which this petition relies, and that it includes representative data and information known to the petitioners which are unfavorable to the petition. 


Elizabeth Barbehenn, Ph.D.
Research Associate 

Sidney M. Wolfe, M.D.
Public Citizen’s Health Research Group


Comparison of drug labels




Black box present



Pregnancy Class

Category C

Category X

Warnings and Precautions


“In guinea pigs, lubiprostone has been shown to have the potential to cause fetal loss. Amitiza should be used during pregnancy only if the potential benefit justifies the potential risk to the fetus. Women who could become pregnant should have a negative pregnancy test prior to beginning therapy with Amitiza and should be capable of complying with effective contraceptive measures.”

[There is no way for a reader to know that “fetal loss” in guinea pigs means “abortion”, that it was dose-related beginning at exposures comparable to that in  humans, and that guinea pigs are predictive of humans exposed to abortifacients.]

[Death of a fetus must surely be a cause for contraindication, i.e., a category X rating in pregnant women.]

[The label provides no solid information as to how to calculate a potential benefit/risk ratio for either patient or health professional.]

Black Box Warnings:



“May be prescribed if the patient

  • has had a negative pregnancy test within 2 weeks prior to beginning therapy,

  • is capable of complying with effective contraceptive measures.

  • has received both oral and written warnings of the hazards of misoprostol, the risk of contraceptive failure and the danger to other women of childbearing potential should the drug be taken by mistake

  • will begin Cytotec only on the second or third day of the next normal menstrual period.”

Use in specific populations

Teratogenic effects

“Lubiprostone was not teratogenic in rats or rabbits. In guinea pigs, lubiprostone caused fetal loss at . . . 2 and 6 times the highest recommended human dose . . .”

[The reader is not told that rats and rabbits are not valid models for prostaglandin in reproductive toxicity tests: guinea pigs and monkeys are valid models but were tested for sensitivity to abortion and not for teratogenicity.]

“. . . during clinical testing of Amitiza, six women became pregnant. . . .” Four women delivered healthy babies [they omit information about bilateral club feet in one], one was lost to follow-up and one was electively terminated.

“Amitiza should be used during pregnancy only if the potential benefit justifies the potential risk to the fetus. If a woman is or becomes pregnant while taking the drug, the patient should be apprised of the potential hazard to the fetus.”

[There is no information available for a woman to calculate a risk/benefit. Neither is there information as to the urgency of contacting a health professional.]


“Several reports in the literature associate the use of misoprostol during the first trimester of pregnancy with skull defects, cranial nerve palsies, facial malformations, and limb defects.”


"See Pregnancy"

See boxed WARNINGS


The only information available is that under “Pregnancy”.

SPECIAL NOTE TO WOMEN: Cytotec may cause abortion . . . premature labor, or birth defects if given to pregnant women.”

Nonteratogenic effects

No information

See boxed WARNINGS” “may endanger pregnancy (may cause abortion)”

Nursing mothers

“Because of the potential for serious adverse reactions in nursing infants from lubiprostone, a decision . . . whether to discontinue nursing or to discontinue the drug”

[There is no information available on which a woman could base a decision.]

“Should not be administered to nursing mothers because the potential excretion of misoprostol acid could cause significant diarrhea in nursing infants.”


[2] Medical Officer Review; web pages 63-64; Accessed February 24, 2009.

[3] Pharmacology review; web page 8; Accessed February 23, 2009.

[4] Thiemann A, Grunder S, Pusch M, and Jentsch TJ. A chloride channel widely expressed in epithelial and non-epithelial cells. Nature 1992;356:57-60.

[5] MacDonald KD, McKenzie KR, Henderson MJ, et al. Lubiprostone activates non-CFTR-dependent respiratory epithelial chloride secretion in cystic fibrosis mice. Am J Physiol Lung Cell Mol Physiol 2008;295:L933-L940.

[7] Linda Reid, Ph.D. Memo of consult. November 5, 2005. web page 236. Accessed February 5, 2009.

[8] Terry KK, Lebel WS, Riccardi KA, et al. Effects of gestational age on prostaglandin EP receptor expression and functional involvement during in vitro contraction of the guinea pig uterus. Prostaglandins, Leukotrienes and Essential Fatty Acids. 2008;78:3-10.

[9] Cuppoletti J, Malinowska DH, Chakrabarti J, et al. Effects of lubiprostone on human uterine smooth muscle cells. Prostaglandins and other lipid mediators. 2008;86:56-60.

[10] Memo of 11/5/05, Linda Reid, Ph.D., Supervisory Pharmacologist, p.236. Accessed February 17, 2009.

[12] Linda Reid, Ph.D. Memo of consult. November 5, 2005. web page 236.  Accessed February 5, 2009.

[13] The doses were based on those used in a rat study rather than on an actual test of the drug in pregnant monkeys.

[14] Lubiprostone drug label. 2008. Accessed January 22, 2009.

[15] FDA Adverse Event Reports analyzed by Public Citizen.

[16] Medical Officer Review; web page 16; Accessed February 24, 2009.

[17] Medical Officer Review; web page 20; Accessed February 24, 2009

[18] Medical Officer Review; web page 12; Accessed February 24, 2009

[19] Amitiza drug label. Amitiza drug label. Accessed January 22, 2009. Now available at:

[20] Cytotec drug label. Accessed January 22, 2009.