The chronic autoimmune inflammatory diseases affect mainly women at their reproductive age. In the past, women with this affection would have abstained from becoming pregnant due to their low quality of life. Nowadays, with the emergence of new therapies for the treatment of immune-mediated inflammatory diseases, pregnancy can be achieved. However, the safety of using therapies during pregnancy and lactation is still a worldwide scientific concern. Biological therapies have revolutionized the era of immune diseases, improving the patient’s outcomes, and improving control over the disease, relapse and complications. In the last 20 years, several treatment regimens have been developed and new immunosuppressive molecules have been introduced into the therapy of autoimmune diseases. But, just as for any kind of treatment addressed to a woman who wants to obtain a pregnancy, the possible adverse effects on fertility, zygote quality, pregnancy complications such as abortion, fetal malformations, intrauterine growth restriction, premature birth, and long-term effects on the newborn should be taken into account.
Biological drugs are IgG-like immunoglobulin molecules that differ in biochemical structure, half-life and transplacental passage(1). Because antibodies are large proteins, they cannot cross the placenta by simple diffusion, instead they need an active transport across the placenta via Fc receptors on trophoblast(1,2,3). These receptors are expressed in the placenta starting with 13-14 weeks of gestation, which is why these molecules do not cross the placental barrier before this term(4). Starting from the second trimester and in the third trimester, these IgG-like molecules easily cross the haemochorial barrier, sometimes reaching higher levels in fetal blood compared to maternal levels(2,3,5). Another important feature of these molecules is their half-life ranging from 9 to 23 days, with the dissimilarity that in the newborn they have a prolonged half-life and they typically vanish from the serum in a range from a few weeks to about 6 months of life(6). Kane and Acquah reported that the levels of IgG in the cord blood are much higher in the full term neonates compared with the maternal circulating levels(7). Unfortunately, there are few studies on the effects of biological therapy on pregnancy and fetus due to ethical causes. The biological therapies approved to use in inflammatory autoimmune mediated diseases include: TNFa inhibitors like infliximab (Remicade®), adalilumab (Humira®), etanercept (Enbrel®), certolizumab pegol (Cimzia®), golimumab (Simponi®), IL-6 inhibitor tocilizumab (Actemra®), anti-CD20 antibody rituximab (Mabthera®), IL1 receptor antagonist anakinra (Kineret®) and a T cell co-stimulation modulator named abatacept (Orencia®). From this list, the most used in pregnancy are the TNF inhibitors, which have been in use for the last 15 years. Searching the data from literature and from registries, we can conclude that the biological therapy is relatively safe when the medication is chosen according to the half-life, depending on the placental passage and the possible adverse effects on the fetus and the newborn which are known. The actual guidelines recommend obtaining adequate disease control to ensure a good materno-fetal outcome. Formerly, the recommendation was that all biological therapies be discontinued prior to conception, leaving behind the dilemma: how to manage an unplanned pregnancy due to the lack of safety data in pregnancy.
Because TNF plays an important role in immunity against bacterial and viral infections, anti-TNF molecules will be associated with the risk of infection(8). The pregnancy itself being a state of immunosuppression, in association with biological therapy may increase the risk of opportunistic infections or intracellular infections such as Listeria monocytogenes(9-11). The TNFa inhibitors are considered of category B due to the absence of risk in animal studies. In humans, we only can use observational studies to follow anomalies after exposure to TNFa inhibitors. Searching the persistence of TNFa inhibitors in the serum of the infants, the studies have shown the persistence of 3-6 months for infliximab and for at least 11 weeks postpartum for adalimumab(4,12-14). Cheent and Nolan have described a case of a healthy baby born from a mother treated with infliximab during pregnancy for Crohn’s disease who was vaccinated at 3 months of age with BCG vaccine and who subsequently died at 4.5 months of age due to BCG dissemination(15). After this case report, The World Congress of Gastroenterology in 2011 proposed a position statement: “Vaccination of infants exposed to biological therapy in utero should be given at standard schedules, except for live-virus vaccines, which are best not given if circulating biologic agents are detectable in the infant”(16).
Carter and his collaborators published in 2006 a case report about an infant delivered from a woman treated with etanercept during pregnancy. This infant was born with some structural defects named VATER association - e.g., vertebral anomalies, anal atresia, cardiac defects, tracheoesophageal fistula and/or esophageal atresia, renal anomalies and limb defects(17). Also, Carter published in 2009 a report using FDA database up until 2005 and found 40 infants born from mothers treated with anti-TNFa (etanercept or infliximab), in which VATER abnormality occurred more often than other structural anomalies(18). Hyrich et al., using the British Society of Rheumatology Biologics Register, reported 32 pregnancies, out of which: 9 patients stopped anti-TNFa before conception, 23 received biological therapy at the time of conception, two continued in the first trimester and one patient continued the treatment throughout pregnancy. From the 23 pregnancy, there were 14 live births without any structural defects, 6 first trimester abortions and 3 elective pregnancy first trimester terminations(19).
In 2004, the manufacturer of infliximab created a cohort database of 131 women exposed to infliximab and showed no significant differences in rates of miscarriages, therapeutic abortions or live births compared to general population(20). Lichtenstein published in 2012 the conclusions from TREAT registry of patients with Crohn’s disease: there were no significant differences in rates of structural anomalies compared with general population(21). Nielsen also drew the same conclusions when studying the relationship between IBD treated with anti-TNF and general population in rates of congenital anomalies, intrauterine growth restriction, prematurity, and miscarriage(22). Although initially there was a suspicion that the using of anti-TNFa may be associated with VATER anomaly, later the meta-analyses contradicted this theory. However, there is still little data on the effects that this therapy may have on the pregnancy and the long term effects on infants.
Due to the different composition of anti-TNFa, it is important to individualize them.
The transplacental passage of these molecule is important in the 2nd and 3rd trimesters, and at birth the cord blood levels are often higher than the maternal serum values. It is also important to know that it persists in fetal blood up to 6 months postpartum(4). Infliximab was not detected in breast milk and breastfeeding is not forbidden(23,24). The data collected so far about the administration of infliximab in pregnancy have shown that there is no association with increasing rates of abortion, prematurity or structural anomalies in newborns, compared with the general population. Exposed newborns have normal response to non-live vaccines, but live vaccines are banned until 6 months of life, or until infliximab in no longer detected in serum(13,25). It can be used in rheumatoid arthritis, ankylosing spondylitis, psoriatic arthritis, Crohn’s disease, ulcerative colitis, and plaque psoriasis.
It is a recombined human IgG1 with high affinity and specificity that binds to soluble TNFa, interfering with cell surface receptors. The OTIS (Organisations of Teratology Information Specialists) database showed no evidence of an association with congenital defects or malformations for infants exposed in utero to adalimumab(26). It is considered to be low risk and compatible to use during pregnancy at least in first and second trimester(16). It can be used in rheumatoid arthritis, psoriatic arthritis, ankylosing spondylitis, Crohn’s disease, ulcerative colitis, and psoriasis.
Etanercept binds to soluble and cell surface TNF, blocking its receptors. Based on the reports of OTIS, etanercept seems not to increase the risk for structural anomalies mare than in general population. Etanercept can be used in rheumatoid arthritis, juvenile idiopathic arthritis, psoriatic arthritis, ankylosing spondylitis, and plaque psoriasis. The benefits of maternal treatment appear to be greater than the risk of fetal harm(27).
Certolizumab pegol (Cimzia®)
This molecule differs from other biological agent through its chemical structure, being a pegylated antigen binding fragment (Fab) antibody without a Fc region. Because his lack on Fc region, it cannot be actively transported by the Fc placental receptor, resulting a minimal transplacental diffusion(28). As a result, the fetal levels of IgG in umbilical venous blood are lower compared to infliximab and adalimumab. This suggests that it may be used as a treatment during late gestation without exposure to the neonate(29). Up to date, three studies on placental transfer of anti-TNFa have been published, two investigator-initiated and one industry sponsored pharmacokinetic study. Mahadevan et al.(4) have published a study in women with Crohn’s disease studying by comparison maternal and umbilical cord blood levels of adalimumab, infliximab and certolizumab pegol. The data reported finding a low level of placental transfer of certolizumab pegol compared with the other two drugs investigated(4). Forger et al. reported similar data for certolizumab pegol placental transfer in women treated for rheumatic arthritis and ankylosing spondylitis(30). This year, two industry-sponsored pharmacokinetic studies about certolizumab pegol were published. CRIB was the first industry-sponsored pharmacokinetic study evaluating placental transfer of certolizumab, and CRADLE was a industry-sponsored clinical lactation study evaluating the transfer of certolizumab into mother milk. Both studies were prospective, postmarketing and multicentric designed to study the safety of using certolizumab as a biological agent and fullfill an adequate disease control of the chronic inflammatory diseases such as rheumatoid arthritis, axial spondyloarthrtis, Crohn’s disease, psoriatic arthritis etc. The CRIB study concluded that there is no to minimal placental transfer of certolizumab from the mother to the newborn due to a lack of in utero fetal exposure during the third trimester of pregnancy(31). CRADLE findings suggest that the level of certolizumab ingested by the newborn by breastfeeding is minimal and indicate that the treatment can be safety continued being compatible with breastfeeding(32).
This human IgG1 monoclonal antibody has high affinity and form stable complexes with human transmembrane forms of TNF. It can be used in rheumatoid arthritis, ankylosing spondylitis and psoriatic arthritis. Only animal data exist among using this drug in pregnancy and show no development defects(33).
Rituximab is a monoclonal antibody against CD20 with B cells depletion effect. Is a drug reserved to the severe cases failed to control by anti TNFa(34). It is considered to be a category C drug, with limited data available(35). The literature data raise the concerns about hematological side effects and infections on newborn(36).
This monoclonal antibody directed against interleukin 6 receptors is approved since 2010 for rheumatoid arthritis, polyarticular juvenile idiopathic arthritis and systemic juvenile idiopathic arthritis. In animal model, no teratogenicity was demonstrated, but still it is considered a category C drug by the FDA. In literature can be found 33 reported cases of pregnancy exposed to tocilizumab. From these 33 cases, 13 had an elective abortion, 7 had spontaneous miscarriage, 10 were healthy term neonate and one child died due to acute respiratory distress syndrome, while two cases were lost from surveillance(37).
It is a recombinant interleukin 1 receptor antagonist approved by FDA in 2001 for rheumatoid arthritis. In animal studies, anakinra showed no teratogenity, but in humans there is a lack of information. FDA considers anakinra a category B drug. Berger et al. reported in 2009 a single case of a woman treated throughout the pregnancy with anakinra and who experienced placental retentions but without fetal abnormalities(38). Another two cases of exposure to anakinra in first trimester exist, with no complications reported(40).
It is an inhibitor of T-cell activation. Abatacept is a class C drug. It can be used to treat adult and juvenile rheumatoid arthritis and juvenile idiopathic arthritis. There are inadequate data about using abatacept during pregnancy. There have been 7 cases reported of exposure to abatacept during gestation. All these cases were also exposed to methotrexate. In 6 cases the pregnancy was discontinued spontaneously or therapeutically, and in one pregnancy the drugs were stopped after pregnancy confirmation and a healthy fetus was born(41).
The latest recommendations in the scientific field
The British Society for Rheumatology(42) (BSR) recommends cessation of using infliximab after 16 weeks of gestation, and etanercept and adalimumab may be continued until the end of the second trimester. If these three drugs are continued later in pregnancy, the physician should take into consideration that live vaccines should be avoided in the newborn up to 7 months of life. Certolizumab pegol can be used in all three trimesters due to its low placental transfer. Golimumab seems safe to use in the first trimester. Breastfeeding is permitted, but with caution. BSR recommends stopping rituximab 6 months before conception. There is limited evidence on the use of rituximab in the first trimester; the second and third trimester exposure is associated with neonatal B cell depletion and regarding safety in breastfeeding there are no data. Regarding the use of tocilizumab, the data is limited and it is recommended to be stopped at least 3 months prior to conception; there are no data about using it during lactation. There is also limited evidence about anakinra - it seems that unintentional exposure in the first trimester of pregnancy is not dangerous for the fetus. Regarding abatacept, the BSR recommendations are that is safer, as for anakinra. The EULAR(43,44) consider that infliximab and adalilumab can be continued up to 20 weeks of gestation, and if it is necessary, adequate disease control can be used throughout pregnancy. Etanercept can be continued up to 32 weeks of pregnancy, and if indicated throughout the pregnancy. Rituximab can produce B cell depletion in neonats if it is used later than the first trimester. Anakinra can be used before and during pregnancy, when there is no other therapeutic option. Regardind breastfeeding, EULAR statement is based on evidence: infliximab, adalilumab, golimumab, etanercept, certolizumab are compatible with lactation, but there is no data about the safety of breastfeeding during the treatment with rituximab, anakinra, abatacept, tocilizumab, so cautions should be taken.
There are several registries which track the effects that biological therapy can have during a pregnancy, such as PIANO registry, and OTIS autoimmune diseases in pregnancy studies. The addition of as many cases as possible will lead to a conclusion regarding the safety of this therapy before conception, during gestation and during lactation. The existing data is encouraging. The infectious risk due to immune depletion should be considered both for mother and child, and the balance of the benefit of maternal therapy with possible fetal complications should always be analyzed. Pregnant women should be informed of the risk of opportunistic infections such as Listeria.
Anti-TNF drugs are considered of low risk and compatible with conception and pregnancy during the first and second trimesters. Breastfeeding is permitted when using biological therapy.
But no biological agents can be assumed to be safe since a lack of association is hard to prove(38). Live vaccines should be avoided for at least 6 months of life in child’s exposed in utero to biological agents. Although it may seem tempting to draw conclusions generally positive, the long-term effects of the exposure to biological drugs remain unknown.