Pregnancy Impact on IBD

Several studies, including large population-based studies, have shown that females with CD and UC may have greater risk for adverse pregnancy outcomes, including preterm birth and low birth weight. Most of these data have been retrospective; however, a recent prospective 1:1 case control study evaluating 332 pregnant women with IBD found that women with IBD had similar pregnancy outcomes to women without IBD.

Disease activity at conception is an important determinant of disease activity throughout pregnancy, because studies have found that only one-third of patients with quiescent IBD at conception have a flare, whereas most of those with active disease at conception continue to have active disease through pregnancy. In a large population-based cohort study from Northern California, most patients with CD and UC had inactive or mild disease at conception, and had good outcomes throughout pregnancy. However, more recent data suggest that patients with UC have an increased risk of disease flare during pregnancy as compared with nonpregnant patients with IBD. A prospective multicenter case-control study followed 209 pregnant patients with IBD and found that pregnant patients with UC were significantly more likely to have a disease flare than nonpregnant patients with UC. Only 65% of patients with UC in remission at conception remained in remission.

Disease Activity Impact on Pregnancy

The effect of disease activity on pregnancy outcomes is a key issue to consider. Studies have suggested that increased disease activity can lead to worse pregnancy outcomes, including fetal loss, preterm birth, and low birth weight. A more recent large cohort study of women with CD found that disease activity during pregnancy was associated with preterm birth. Another population-based case-control study evaluating 1209 women with UC and 787 women with CD found that the risk of venous thromboembolism was increased in patients with UC, particularly in those with flaring disease. Furthermore, women with IBD had increased risk for an emergency cesarean section, and in patients with CD, the risk increased with disease activity. In a prospective case-control study in pregnant women with and without IBD, patients with UC who required hospitalizations had a higher risk of preterm birth, small-for-gestational-age birth, and cesarean section compared with those who only required ambulatory care.

In summary, significant evidence exists to suggest that the most important factor for an expectant mother with IBD and her unborn child is to have adequate disease control before conception. This may be especially important in patients with UC. With this in mind, physicians and patients should focus on disease response and remission, and in patients with moderate to severe IBD, biologic therapy may be required.

Anti-TNF Medications in Pregnancy

Anti–TNF-α medications are monoclonal antibodies or antibody fragments to TNF, an important cytokine in pathogenesis of inflammation in IBD. Infliximab is a chimeric mouse/human IgG1 monoclonal antibody against TNF, and in large clinical trials was found to be effective for induction and maintenance of remission in UC and CD. Adalimumab is a fully human IgG1 anti-TNF monoclonal antibody, and in large clinical trials was found to be effective for induction and maintenance of remission in UC and CD. Certolizumab pegol is a polyethylene glycolylated Fab’ fragment of humanized anti–TNF-α monoclonal antibody, and in large clinical trials has been shown to have clinical efficacy in CD. Golimumab is a fully human monoclonal antibody to TNF, and in large clinical trials found to be effective for induction and maintenance of remission in UC.

Anti–TNF-α Antibodies and Placental Transfer

An important consideration in the evaluation of the safety of anti-TNF antibodies administered during pregnancy is when the fetus will be exposed to the medication. IgG antibodies are the main subclass of antibodies that are transferred across the placenta, with selective binding of the Fc gamma portion of the IgG at the surface of the syncytiotrophioblast layer. Additionally, an Fc receptor neonatal molecule is bound to the IgG and is responsible for the transfer of the IgG from mother and neonate. Studies have shown that IgG1 subclass antibodies likely increase linearly through pregnancy, with fetal blood concentrations steadily increasing from second trimester until delivery, with most antibodies being transferred in the third trimester.

Infliximab, adalimumab, and golimumab are complete anti-TNF antibodies with an intact Fc portion of IgG1 subclass, and should therefore cross the placenta through the previously described mechanisms. These anti-TNF medications have been shown to cross the placenta, because studies have shown detectable umbilical cord blood concentrations of anti-TNF medication at delivery. A recent study evaluating the outcomes of pregnant patients with IBD on infliximab or adalimumab found detectable cord blood levels in most patients, even those who stopped anti-TNF medication greater than 10 weeks from delivery. Recently, a prospective multicenter observational study evaluated pregnant patients with IBD treated with infliximab or adalimumab. Infliximab levels from mother at delivery and cord blood in 11 newborns and 8 mothers (last dose given at mean gestational week 28; range, 24–31 weeks) were higher in the cord blood than the mother, with a mean ratio of 3.8 (range, 0.9–10.7). Adalimumab levels were drawn in five mothers and newborn cord blood at delivery, and similar levels were found in the mothers and newborn. Mahadevan and colleagues evaluated 31 pregnant women on infliximab, adalimumab, and certolizumab. The median level of infliximab in the cord blood was 160% that of the mother, and the adalimumab median drug level was 153% that of the mother. Drug levels could be detected up to 6 months after delivery in the infants. Because certolizumab is a polyethylene glycolylated Fab’ fragment, it does not have an Fc portion, and therefore should not be transported actively across the placenta. The only transfer of the antibody fragment that occurs will likely be passive, and in a much smaller amount than actively transported antibodies. In the previously mentioned study by Mahadevan and colleagues, infants of 10 pregnant women on certolizumab were found to have very low drug levels, with a median level of 3.9% of the mother.

Anti–TNF-α Medications in Pregnancy

Although newborns will likely be exposed to certain anti-TNF medications in utero, all anti-TNF medications are Food and Drug Administration (FDA) pregnancy class B ( Table 1 ). In evaluation of available animal data, toxicity studies using doses several hundred times higher than the recommended human dosing of infliximab and adalimumab showed no evidence of teratogenic effects. Data available from each individual medication are presented next, and further studies combining data from all anti-TNF medication use in pregnancy are then summarized.

Effect of Anti-TNF Medication on Newborns

As noted, anti-TNF levels have been detected in umbilical cord blood in multiple studies, often at levels higher than maternal anti-TNF levels. Although most data have not shown deleterious effects of this, immunosuppression of the mother can lead to adverse outcomes. A recent case report documented a fatal case of disseminated bacille Calmette-Guérin (BCG) infection after BCG vaccination in an infant born to a mother treated with infliximab throughout her pregnancy. However, nonlive vaccines are likely safe, and it is currently recommended that infants exposed to anti-TNF therapy in utero can undergo vaccinations according to regular schedule, because no adverse effects have been noted and vaccination responses seem appropriate. A few recent case reports have documented appropriate vaccine responses in infants exposed to anti-TNF medications in utero.

Bortlik and colleagues recently prospectively evaluated the long-term impact of maternal anti-TNF medication on 25 exposed children (infliximab = 22, adalimumab = 3) until median gestational age 26 weeks (range, 17–37 weeks). Twelve mothers (48%) were on concomitant thiopurines. All children had normal growth and all but one had normal psychomotor development. Twenty-three (92%) had normal vaccination according to protocol, and 15 children had tuberculosis vaccination (BCG vaccine within 1 week of birth) without serious complication. Immunologic evaluation was performed in 17 children, cellular immunity was normal in all infants, and all had detectable serologic response to vaccination. Additionally, the largest amount of data available, from the large prospective PIANO registry, shows no increased risk of infections from biologic therapy exposure in infants to 1 year. There was a small but significant increase in infections in children detected at 12 months in the combination therapy–treated group (excluding certolizumab pegol–treated patients given the minimal placental transfer; RR, 1.35; 95% CI, 1.01–1.80). These infants will be followed prospectively for 4 years, and further data are forthcoming.

Therefore, with limited data that are available, children exposed to anti-TNF therapy likely will have normal development and minimally increased risk of immunodeficiency syndromes. Children exposed to combination therapy, however, may be at slight increased risk of infections in the first and second year of life, although documented infections were mild. Given the recent data regarding improved outcomes in patients on combination therapy (both an immunomodulator and anti-TNF medication) compared with either medication alone in patients with moderate to severe IBD, more patients will likely become pregnant while on combination therapy with thiopurine and anti-TNF medications. More data regarding combination therapy throughout pregnancy are needed.

Based on available data, infants exposed to anti-TNF medications in utero will likely mount appropriate immune response to inactivated vaccinations and these can be given according to regular schedule. Live vaccines, however, should be avoided for at least the first 6 months in infants.

How Long to Continue Anti-TNF Therapy During Pregnancy

Opinions in the literature regarding this question differ. Because most placental transfer of immunoglobulins, and thus anti-TNF medications, occurs in the third trimester, it had been suggested in the literature to stop anti-TNF medication before gestational week 30. However, older studies have suggested cessation of therapy could worsen patient outcomes in pregnancy, but this was before the era of biologics. Also, it has been well documented that stopping anti-TNF medications is associated with a high risk of recurrence of luminal and perianal disease. To answer this question, a previously mentioned study was conducted to evaluate if patients with IBD in remission could safely stop their anti-TNF medication in the second trimester. Five women had active disease in their second trimester and continued infliximab to weeks 30 to 34. Twelve women were in remission and stopped infliximab in the second trimester (between weeks 18 and 27; average, week 23), and 11 patients on adalimumab were in remission and stopped in their second trimester (stopped between weeks 21 and 27; average, week 22). Only two women, both in the adalimumab-treated group, had a flare in their third trimester (at weeks 30 and 36). Those who stopped the anti-TNF inhibitor greater than 10 weeks before delivery still had detectable cord blood levels. However, the cord blood levels were significantly lower in the patients who had their infliximab stopped greater than 10 weeks before delivery compared with those who continued through delivery.

However, stopping anti-TNF medication in the second or third trimester leads to continued exposure of medication for the fetus, because the previously mentioned studies have shown cord blood levels persist if mothers last received the medication as early as gestational weeks 16 to 20. Certainly, if a patient has active disease, she should continue her biologic therapy throughout pregnancy, because much of the available literature detailed now supports the medications’ safety in pregnancy. Also, newer literature showing that patients with UC have an increased risk of flare, even if they are in remission at conception, compared with nonpregnant patients with UC, would cause more hesitation for medication cessation in this group. However, if a patient is in clinical remission, it may be reasonable to consider cessation of anti-TNF medications in the second trimester, and physicians and patients should consider options carefully.

Anti-integrin Antibody Therapy in Pregnancy

Natalizumab is an IgG4 antibody that inhibits the α ₄ integrin adhesion molecule. IgG4 antibodies cross the placenta in the second and third trimester, although not as preferentially as IgG1. There is limited knowledge of the use of natalizumab during pregnancy and lactation. In a small study of pregnant guinea pigs given seven times the human dose based on body weight, there was a small reduction in pup survival. A primate study showed hematologic effects on the fetus (mild anemia, decreased platelet count, reduced weight of the liver and thymus, and increased splenic weight) at a dose of natalizumab that was 2.3-fold higher than typical clinical dosing in humans. No teratogenic effects were noted in either animal study.

In one study of 143 pregnancies reported in patients with CD or multiple sclerosis who received natalizumab in their first trimester, no increased numbers of birth defects were noted. A preliminary data review of the prospective Tysabri (natalizumab) Pregnancy Exposure Registry in 2011 following patients on natalizumab showed 277 known pregnancy outcomes. The rate of spontaneous abortions (11%) was consistent with the expected rate in the US population. Malformations occurred in 23 pregnancies in 21 of the 277 women, but they were isolated, some were confounded by other risk factors, and they did not suggest a drug-related pattern. The ongoing multicenter PIANO study includes a cohort of six women with natalizumab exposure in pregnancy with no increased rates of adverse events of congenital anomalies or abnormal infant growth or development. Recently, a study evaluated two pregnant women with relapsing multiple sclerosis treated with natalizumab until pregnancy week 34. Both newborns had uneventful deliveries with normal birth weights and were followed to week 12 of age. One infant had no adverse events. One infant had a possible minor intracerebral hemorrhage detected on a screening ultrasound, which normalized by week 12. This infant also had mild anemia and thrombocytopenia, which normalized by week 12. Blood was evaluated in both infants at weeks 2 and 12 of age and found impaired in vitro CXCL-12–induced T-lymphocyte chemotaxis rate, which normalized by week 12. The authors noted that similar effects on in vivo chemotaxis of T lymphocytes have been demonstrated in adults treated with natalizumab, but the clinical significance of this is yet unclear.

Currently, natalizumab has an FDA pregnancy category class C. A recent World Congress of Gastroenterology position statement highlights that the safety of natalizumab in pregnancy is unknown, with available human data showing no increased risk of birth defects. The authors state that physicians and patients should judge the risk versus the individual patient needs with use of natalizumab during pregnancy.