Host — Dan Keller
Hello, and welcome to Episode Seventy-seven of Multiple Sclerosis Discovery, the podcast of the MS Discovery Forum. I’m Dan Keller.
Pregnancy and the postpartum period present special concerns to women with MS. Dr. Annette Langer-Gould, a neurologist and epidemiologist at Kaiser Permanente in Los Angeles, investigates ways to lessen the risk of relapses in these women. We discussed the effects of breastfeeding, among other topics, when we met at the ECTRIMS meeting last fall in Barcelona.
Interviewer – Dan Keller
In terms of pregnancy and breastfeeding in MS, what are you looking at?
Interviewee – Annette Langer-Gould
We're studying modifiable risk factors for postpartum relapses in women with multiple sclerosis. And specifically, we are looking at starting therapy shortly after delivery, whether that can reduce the risk of postpartum relapses, whether breastfeeding, particularly breastfeeding exclusively, could reduce the risk of postpartum relapses, and whether vitamin D levels play any role in increasing or decreasing the risk of postpartum relapses.
And are these women who are on disease-modifying therapy throughout pregnancy or not?
No. In our population, a little over 60% were treated prior to pregnancy. But we do have a decent number of women who had decided to never go on disease-modifying therapies before, and almost all of them stopped disease-modifying therapies either shortly before or when they find out that they're pregnant.
In terms of each of those outcomes, what are you finding?
We haven't analyzed the data for the vitamin D yet, but in the German pregnancy registry, we just published the data in exclusive breastfeeding, and once again showed that exclusive breastfeeding does protect against postpartum relapses. In that population, actually 96% of the women had been on some sort of DMT prior to pregnancy, and none of them were treated throughout pregnancy. We also found that resuming DMTs does not seem to have a big effect on reducing the risk of relapses, particularly in the first six months postpartum.
Is that in women who are exclusively breastfeeding or not?
Ah, so that's a good question. So there is no good safety data on taking the medications during breastfeeding. And therefore, many clinicians and most patients are concerned about potential theoretical risks. So behaviors are actually mutually exclusive. Women typically will either breastfeed or resume medications early in the postpartum course. The other thing we find in the Kaiser population is that there are still a fair number of women who neither breastfeed exclusively or resume their medications, which presents sort of an interesting opportunity. If we could show that one or the other behaviors is protective, perhaps we could encourage either exclusive breastfeeding or resuming DMT.
If women are not breastfeeding, do you have an idea of the time course of resumption of risk for relapse?
Yes, so the concern about postpartum relapses really is about having a relapse in the first three to four months postpartum. If we look over at the whole pregnancy year, and that's about 30% to 40% of women. So this is actually still the best defined risk period for having a relapse and actually the only clear trigger—with perhaps the exception of upper respiratory tract infections—of relapses. So we know that having just had a baby or having an upper respiratory tract infection is a pretty strong predictor of having a relapse. So it presents sort of a unique opportunity to also look at other biological factors, like vitamin D, which is why we're interested in it, to see if any of these things have a strong role in relapses as well.
If women are breastfeeding postpartum, what is the hormonal profile like? Is this almost like an extension of pregnancy?
For women who breastfeed exclusively, meaning that they breastfeed to the point of suppressing their ovaries and not resuming menstruation—so that essentially there's no regular meal that's being replaced by formula or by table food in the baby—they have very high prolactin levels. So it's actually a little bit different than being postmenopausal, in the sense that they have very high prolactin levels. And they have incredibly low nonpulsatile FSH and LH levels. In the postmenopausal period, there occurs a very high FSH and LH levels. The similarity, though, is that they both have bottomed-out estradiol and progesterone levels, in both women who are breastfeeding to the point of suppressing menses and also postmenopausal women. And of course the other similarity is that there's no ovulation occurring, either during pregnancy, during exclusive breastfeeding, or after menopause.
So it sounds like breastfeeding is really a hypothalamic pituitary suppressant as opposed to in menopause, where you still have those cranking away, but just no response from the ovaries.
Can this be used in any clinical sense? Do you see an application?
The most obvious direct way to translate these findings is that that, if you have a woman with MS in front of you and she is pregnant and she tells you she'd like to breastfeed, we certainly have no good reason to discourage her. And that if anything, I would suggest that the data we've already published would point to the fact that we may want to encourage exclusive breastfeeding, provide them with lactation counseling, and also sort out exactly what the optimal duration of exclusive breastfeeding may be for these women. Is it really only eight weeks, which we had defined arbitrarily? Or does longer duration of exclusive breastfeeding have additional suppressive properties? And that would, of course, have implications in the United States for things like maternity leave and work accommodations to allow that to continue, if it has a strong therapeutic effect for the mother.
What's the relapse rate among postmenopausal women compared to postpartum women?
So relapse rate declines with age. And so it typically in postmenopausal women, although there's not great data, we would expect them to have relapse rates of less than 0.3 per year, Annualized relapse rates of less than 0.3 per year. And in postpartum women, that first three to four months, the annualized relapse rate exceeds one.
But men also have a decline in relapse rate as they age, too. So you can't attribute it to lower estradiol.
Exactly. Yeah, I think it's far more complicated than just a simple sex hormone effect. You know, that was sort of our first instinct from pregnancy or the reason pregnancy must be protective. It has to have something to do with estradiol or the very high progesterone levels. And that's what prompted the postpartum study and also the estradiol randomized control trial. And both of those, of course, disappointingly have been negative. In isolation, the sex hormones associated with the protective effect of pregnancy don't really have a protective effect on inflammation. It's probably more of a combination of factors that play into modulating the immune response.
Where do you go from here?
I think that if we are able to reproduce the findings, looking at this population-based source, that early resumption of DMTs is not particularly helpful, but perhaps it may be later in the postpartum year, and that exclusive breastfeeding is, again, protective, then I think the next step really is to establish the safety of some of these medications during lactation. For several of them, there's really no biologically plausible reason to think that they would have an effect on the baby, as they're not likely to be absorbed through the gut or enter into the baby's bloodstream.
Examples of that would be the large molecules like Copaxone, the interferons, and also the infusion medications, Tysabri (natalizumab), and rituximab as well. Although you may be able to detect them in breast milk, they are such large molecules that they would not diffuse across the baby's stomach and into the bloodstream. Think about it. If the mom has to take it as a pill, it is very likely to be transmitted to the baby. If the mom has to take it as an infusion or injection, very unlikely that oral route through the baby would have any effect.
How sensitive is this effect to, as you said, exclusive breastfeeding? Can you start introducing formula, or it's all or none?
That's a really good question. So we did look at that also in the German pregnancy registry. So first of all, women tend to have very defined behavior. They tend to decide to supplemental feed with formula very, very early, before they've even established their full milk supply. So to back up even further, a healthy woman gives birth to her child. Usually menstruation will resume two months after delivery, not one month. So it does take the HPA gonadal axis a little chance to recover from those high-circulating hormones of pregnancy. And in women who introduce supplemental feedings, particularly early, we also see the very same thing; that they will resume their period at two months postpartum.
Actually, most of the work done in this field has been done by nutritionists who are in developing countries who are interested in knowing what you should do if you see a starving mother and a starving baby. Who should you feed?
It turns out that if you feed the baby, the mother's ovarian function will resume. So any regular supplemental feedings and very quickly their prolactin levels will drop. The pulsatility of the FSH and LH secretion will return. Ovulation returns, and so does menses. It's essentially sending the mother's body a signal that the baby no longer needs nutrition from the mom to survive, so she's ready to have another child. So the right thing to do in that situation would be feed the mom, and let her nurse the child. Biologically, it's very interesting. Even though some breastfeeding is better than none for the baby, in terms of the effect on the mother's HP [hypothalamic-pituitary] ovarian axis, some supplemental feeding is just like all supplemental feeding.
Have we missed anything or anything interesting to add?
So I guess I would say just in general, women's, and now even men's, desire to have naturally-born children has taken on a new significance with a lot of the small molecule agents, because we need to consider family planning and discuss it much earlier, as small molecules are likely to have an effect even if they get pregnant accidentally on the developing fetus. This is a challenge we haven't had before, because large molecules won't cross the placenta in the first trimester. And the first trimester is the critical period for organ development.
So it's sort of new era for MS neurologists, where we really, really have to think carefully about which medication we put them on if they're planning on having children soon. So I’d strongly encourage that you have that conversation very early and have it with every followup visit. I typically will ask them, are you planning on having children within the next two years? And if they say, no, I ask what kind of birth control they're on, or in some cases they're in same-sex couples. That's obviously an exception. And if they are not on a reliable form of birth control, I think you need to think twice about giving the small-molecule agents—so the pills, basically.
Should MS neurologists work with high-risk OB/GYNs?
I think for the most part it's not necessary, because women with MS, they don't have abnormal complications at pregnancy. I think there are certainly situations that we're running into now. If they get pregnant accidentally on fingolimod, teriflunomide, or Tysabri, we do need to work with them, mostly for the baby. So you may want to do more intense early screening if the mother is culturally open to the idea of having an abortion. You may want to do more fetal ultrasounds, perhaps even a fetal MRI, if there's suspicion of major malformations early on in pregnancy.
And also for the Tysabri, really, it's not so much about organogenesis, but if they've had later exposure to Tysabri during pregnancy, which unfortunately on occasion has been necessary to control rebound disease activity during pregnancy, that, you know, we have seen hematological abnormalities in some of these children, so far none with clinical complications. Only one child had a subclinical intraventricular hemorrhage that resolved. It's still concerning. Our experience is very small, and we would certainly highly recommend that those women give birth in a hospital that has a neonatal intensive care unit available and a pediatrician on call to examine the child and also make sure that the child doesn't have a severe thrombocytopenia or anemia at birth.
Do the different drugs have different risks for fetal malformations or other dysfunctions?
Yes. So teriflunomide, or Aubagio, is the most concerning medication because if a woman gets pregnant on that accidentally, it is, you know, a category X drug because it can interfere with neural tube development. And although you can chelate to get the medication out very quickly, the safety data from other indications, you know, the rheumatoid arthritis and lupus literature, is not particularly reassuring in terms of fetal outcomes. So I think that's sort of the number one to stay away from if a woman is planning on getting pregnant. And it's also one where, you know, there is some concern, although not strong evidence, that it may also affect the offspring of men with MS who are on the medication.
In terms of the other ones, of course, again, small molecules in fingolimod has about a 15% to 16% major fetal malformation risk with early pregnancy exposure. It has a very long half-life. So even if they stop the medicine the minute they find out they're pregnant, it takes over two months for it to be cleared, which means that the baby has seen it now through the entire first trimester. That can have significant effects, both on cardiac and brain development. And then with dimethyl fumarate, we haven't seen—now of course, this is a very new drug, so we don't have nearly as much experience—we have not seen any major malformations, but there was concern in the animal models that it could interfere with cognitive development. In particular, the rats had maze-finding difficulty.
Is alemtuzumab indicated at all? It seems to have a long tail.
I'm not sure what the half-life of alemtuzumab, but it's probably similar to other monoclonal antibodies, which is usually around 15 to 20 days. So monoclonal antibodies don't cross the placenta in the first trimester, because it's a very large molecule. So large molecules only get across if there's an active transport system. For antibodies, there is an active transport system, because it's very important that the child be born with a high dose of antibodies received from the mother to help protect them during the early part of their infancy while their own immune system is still developing.
So we see maternal antibodies being transported, and of course, monoclonal antibody medications would be dragged along with that during second trimester. And it goes up in elliptical fashion, with very, very high amounts being pumped across the placenta in third trimester. And they also, of course, have a very delayed clearance mechanism, both the fetus really has no clearance mechanism, and then even the neonate has a very slow clearance mechanism. So in TNF alpha studies, if the drug is given during third trimester, it's typically not cleared until about six to nine months postpartum.
So you also have to be concerned that a baby exposed would have some of that medication hanging around during the early neonatal period and give some thought to whether or not their immunization scheme would need to be adjusted, as the cautionary tale there would be TNF alpha exposure during pregnancy.
There was a case reported of a woman who had very severe rheumatological disease, had discussed with her rheumatologist the potential risks and benefits of taking it throughout pregnancy, opted to take it throughout pregnancy. And then living in an endemic area for tuberculosis, the baby got the BCG vaccine and got disseminated mycobacterium and died. And that, you know, was probably directly related to impaired immunity from the TNF alpha antagonist. And sure enough, the baby was born with fairly high cord levels and also had very high levels still remaining in the blood in the neonatal period.
So it's not just once the baby's born, it's like the drug is out. So drugs like alemtuzumab and rituximab, the way in which they work, even though the drug could be long gone, but the effect of the medication works very long time. So those are actually good choices for women with highly active disease who are planning on getting pregnant. And you have concerns about rebound. I mean, we typically use rituximab because it's obviously much safer than alemtuzumab and seems to do a fairly good job. But you know, these aren't medications we should be giving while they're pregnant, but probably not a big effect in crossing the placenta and on the baby if they're used prior to pregnancy.
If they can plan that well and get a pulse of that early, and then get pregnant a few months later.
Yes. Yeah, that's always the trick, right? And they do get pregnant accidentally on just about everything we put them on. So the infrequent infusion medications is the easiest because you can ask about last menstrual period. And you can ask about birth control use, and you can do a pregnancy test the day of, a quick urine dipstick and find out so that you don't accidentally infuse a pregnant woman. Of course with Tysabri, when you're giving them an infusion every month, it gets a little tricky. Usually people just kind of get tired of it. The nurses forget. The doctor forgets to order it, although it's not necessarily bad practice if you know you have a patient who is not on a reliable form of birth control.
Very good. I appreciate it. Thank you.
Thank you for listening to Episode Seventy-seven of Multiple Sclerosis Discovery. This podcast was produced by the MS Discovery Forum, MSDF, the premier source of independent news and information on MS research. MSDF’s executive editor is Carol Cruzan Morton. Msdiscovery.org is part of the nonprofit Accelerated Cure Project for Multiple Sclerosis. Robert McBurney is our President and CEO, and Hollie Schmidt is Vice President of Scientific Operations.
Msdiscovery.org aims to focus attention on what is known and not yet known about the causes of MS and related conditions, their pathological mechanisms, and potential ways to intervene. By communicating this information in a way that builds bridges among different disciplines, we hope to open new routes toward significant clinical advances.
We’re interested in your opinions. Please join the discussion on one of our online forums or send comments, criticisms, and suggestions to email@example.com.
For Multiple Sclerosis Discovery, I'm Dan Keller.