Pharmacovigilance and Pregnancy

Nov 21, 2012
Bart Cobert

Pharmacovigilance, Drug Safety and Regulatory Affairs Author & Expert

Obviously, the safety of drugs in pregnant and breast-feeding women, the embryo/fetus and, less directly, the male partner are of paramount importance. Yet this is an area where little is known and little can be known. Unless drugs are specifically designed and indicated for use in pregnancy, it is not ethical or feasible to do clinical trials in pregnant women or in women actively breast feeding. Thus the only real data in this population comes from accidental or intentional exposure in pregnant women. This means limited anecdotal data. Where there is exposure, it is usually not feasible to work up the mother as most lab and invasive tests contraindicated unless there is a very strong indication for them. So no x-rays, scans, biopsies etc. in almost all cases. If there is exposure and no adverse event (AE) then usually no tests can or should be done. How the male partner who is exposed (rather than the mother) is worked up is also not clear or standardized.

Unless some sort of reliable techniques can be developed to study drugs without exposing either the mother or baby to harm, we will not have more data in the future.

Of course, animal and other non-human, non-clinical data are obtained but this is of limited use in extrapolations to humans.

So we are left with a grab bag of individual cases in pregnant and lactating women and babies/children:
• Inadvertent exposure during the clinical trials and after marketing
• Overdoses with and without AEs during trials and in the post-marketing setting
• Intentional exposure usually against medical advice
• Abortions and miscarriages
• Measurement of drug levels in milk after the mother stops breast feeding (if feasible)

Companies must maintain a “pregnancy registry” of all exposed women (or the exposed male partner and the mother) and the patient must be followed to term with the outcome duly recorded. The health of the baby should also be followed as long as possible (as some AEs with long latency periods appear only much later – the classic case being DES which produced vaginal cancer in the daughters of the exposed women some dozen years later when the daughters hit menarche).

These registries are included in Annual Reports (DSURs/IND Annual Reports) and Periodic Safety Update Reports (PSURs/PADERs). FDA issued a guidance for industry in 2002 on pregnancy registries and this is useful to review. The registries may be simple data listings or complex studies with protocols and hypothesis testing. See:

Most companies enter the cases in the safety database whether an AE occurred or not as this is the most feasible operational way to track pregnancy cases and then to produce the line listings, reports and case summaries.

Labeling regarding use in pregnancy is precise in most jurisdictions but since there is little data the labeling doesn’t help very much:

In the US, pregnancy categories are as follows. Similar ones are used elsewhere:

Category A: Adequate, well-controlled studies in pregnant women have not shown an increased risk of fetal abnormalities to the fetus in any trimester of pregnancy.

Category B: Animal studies have revealed no evidence of harm to the fetus; however, there are no adequate and well-controlled studies in pregnant women.
Animal studies have shown an adverse effect, but adequate and well-controlled studies in pregnant women have failed to demonstrate a risk to the fetus in any trimester.

Category C: Animal studies have shown an adverse effect, and there are no adequate and well-controlled studies in pregnant women.
No animal studies have been conducted, and there are no adequate and well-controlled studies in pregnant women.

Category D: Adequate, well-controlled, or observational studies in pregnant women have demonstrated a risk to the fetus. However, the benefits of therapy may outweigh the potential risk. For example, the drug may be acceptable if needed in a life-threatening situation or for a serious disease for which safer drugs cannot be used or are ineffective.

Category X: Adequate, well-controlled, or observational studies in animals or pregnant women have demonstrated positive evidence of fetal abnormalities or risks. The use of the product is contraindicated in women who are or may become pregnant.

Most drugs would fall into a B, C or X category. Many companies would not even want to consider a softer rating for legal liability reasons.

FDA proposed changes in 2008 for pregnancy and lactation labeling but given the lack of data, these changes again are not very helpful.

So what to do? Basically when a signal comes along or a problem arises one must scrounge. Here are a couple of examples:

Example 1

Some years ago we marketed a new combination of two known drugs for hepatitis. One drug was clearly teratogenic and the other drug was clearly abortifacient – that is, the first produced birth defects and the second abortions/miscarriages in the animals studied. Both were clearly either a category C or X (as above) depending upon the country’s labeling. The goal was clearly that pregnant women should NOT use the drug. Although this was before REMS or RiskMAPs were in place, the equivalent of ETASUs (elements to assure safe use) were put in place including special packaging, a black box warning, patient education, negative pregnancy testing and two methods of contraception were required/advised.

Nonetheless, over several years we had hundred cases of women who became pregnant while using the product. All were duly noted in our registry and reported to FDA and other health agencies. Changes were made in the warning, packaging etc. The surprise was that there were no fetal abnormalities though there were some miscarriages and some voluntary abortions. Some folks in the company gingerly broached the idea that maybe we could soften the labeling and allow some use in pregnant women! Clearly that idea was a non-starter as a sample of a few hundred exposures was hardly sufficient to consider letting pregnant women use the product. As the rule of 3 notes, if 300 hundred pregnant women deliver babies with no fetal defects, all one can say is the incidence of such defects would be no greater than 1/100 with a 95% confidence interval – not reassuring at all.

So the points here are that great care must be taken if there is a known developmental toxicity in animals and that anything can happen: good or bad. It is also impossible to prove a drug is NOT toxic to the fetus.

Example 2

The second interesting pregnancy issue revolved around a blockbuster drug for allergies. This drug sold several billion dollars for several years and had a rather benign safety profile with a B pregnancy category. It was used by some women who became pregnant. Over the course of the years, a few cases of hypospadias (a birth defect in which the opening of the urethra is on the underside, rather than at the end, of the penis) were reported in newborn babies. All were mild and easily repaired surgically. What to do? Obviously not a terrible birth defect but still one to be avoided. A long and expensive work up was done.

The first thing was to examine the cases as thoroughly as possible. This was done but nothing of note was found.

The next thing needed was a larger sample size and birth registries around the world were sought. Although there are many such registries, few have sufficiently detailed records to be useful. The best one was the Swedish Birth Registry. As Sweden has national health coverage for the essentially the whole population almost all births were captured and a good history (medical, drug use, smoking etc.) obtained. The Registry personnel were kind enough to supply data and help us work out this epidemiologic puzzle. Seven cases were found in the registry with a surprise. Two of the cases were in identical (male obviously) twins! Was this one case or two cases?! Although one could argue this either way it certainly changed the incidence and risk calculations! After much work it was more or less concluded that the risk and incidence of hypospadias was the same in our drug as in most of the other drugs in the class.

And finally, an animal study was done. It turned out the best animal to study in terms of embryology and determining the time of penis formation when the risk of a malformation was greatest with drug exposure was the fetal mouse! So mother mice were exposed to the drug and studied. The offspring were examined either at birth or were sacrificed before full gestation for study. As you can imagine it is hard to study mice, even harder to study fetal mice and yet even harder to study the penis of a mouse fetus!! Nevertheless, the study was successfully completed and there appeared to be minimal effect on the mouse fetus. The conclusion after all the evidence came in was that the drug was not a likely cause of hypospadias. This set of investigations took over a year to complete.

So this is an example of a fascinating epidemiologic, embryologic and PV workup of a pregnancy issue.

Finally, in the face of all this uncertainty, one must note that there is a website that actually does give detailed information on what drugs can and should not be used in pregnancy: Motherisk of the Hospital for Sick Children of the University of Toronto, Canada – This is site well worth examining. See, in particular, the section on Drugs in Pregnancy ( The unit has collected data since 1985, done studies themselves and reviewed global literature and come to some conclusions about which drugs can be used. The information represents their own views and not necessarily those of the manufacturers, Health Canada, myself or anyone beyond Motherisk. Some recommendations indeed may be considered controversial. It also appears that the full information is not available on computers outside of Canada for liability reasons. Nonetheless, an interesting site. There are many other sites (in the US) that cover drugs and pregnancy but they tend to be simply a compilation of the approved drug labeling and do not make recommendations beyond that.

So pregnancy and drug use is a complex and difficult area that looks like it will not get any easier in the future.