Abstract
The immunosuppressant class of drugs encompasses a wide variety of molecules with different mechanisms of action and toxicities. For example, breastfeeding has traditionally been contraindicated during the maternal use of antineoplastics. While that still holds for many drugs used for treating cancer, some of the same drugs are also used to treat inflammatory conditions such as rheumatoid arthritis, psoriasis, and inflammatory bowel disease. In some cases, only the dosage is different. This makes it impossible to make a general statement about whether immunosuppressant drugs can or cannot be used during breastfeeding.
Even among the antineoplastic drugs, distinctions have been made by some authors. 1 Use of the potent alkylating agents (e.g., cisplatin, cyclophosphamide, doxorubicin) generally contraindicates breastfeeding. But other types of small-molecule chemotherapy such as antimetabolites (e.g., mercaptopurine, methotrexate) are compatible with breastfeeding, at least with certain dosages. Newer targeted agents consist of monoclonal antibodies and various enzyme inhibitors.
In this column, the most commonly used immunosuppressants will be reviewed with respect to their use in nursing mothers. Information on the specific drugs during breastfeeding comes from LactMed, where original citations and additional information can be found.
Conventional Small-Molecule Immunosuppressants
These drugs are orally bioavailable and potentially can be absorbed by the breastfed infant if they appear in milk.
Thiopurines Azathioprine, mercaptopurine, and thioguanine are closely related thiopurines that are now used more often in inflammatory conditions such as inflammatory bowel disease and rheumatoid arthritis than in cancer chemotherapy.
Azathioprine is rapidly metabolized to the active metabolite mercaptopurine, which is further metabolized to active metabolites including 6-methylmercaptopurine, 6-thioguanine, 6-thioguanine nucleosides (6-TGNs) and others. The enzyme thiopurine methyltransferase (TPMT) is responsible for metabolism of 6-TGNs. A genetic deficiency in TPMT can lead to excessive toxicity.
Studies in women with inflammatory bowel disease, systemic lupus erythematosus or transplantation taking doses of azathioprine up to 200 mg/day have found either low or unmeasurable levels of the active metabolites in milk and infant serum. Evidence indicates a lack of adverse effects on the health and development up to 4.6 years of age in infants exposed to azathioprine during breastfeeding, but long-term follow-up for effects such as carcinogenesis has not been performed. Mothers with decreased activity of TPMT might transmit higher levels of drug to their infants in breastmilk. Poorly documented cases of mild, asymptomatic neutropenia and increased infection rates have been reported, so it might be desirable to monitor exclusively breastfed infants with a complete blood count with differential and liver function tests if azathioprine is used during lactation, although some authors feel that monitoring is unnecessary. Avoiding breastfeeding for 4 hours after a dose should markedly decrease the dose received by the infant in breast milk. Most North American and European expert guidelines, the National Transplantation Pregnancy Registry and other experts consider azathioprine to be acceptable to use during breastfeeding.
Since mercaptopurine is the primary metabolite of azathioprine, azathioprine safety data can directly be translated to mercaptopurine. Thioguanine is a downstream metabolite of these drugs, so their safety can likely also be applied to thioguanine. Very low levels of thioguanine were found in the milk of one woman taking the drug, and 37 infants were breastfed during maternal use of thioguanine for Crohn’s disease with no adverse effects.
Calcineurin inhibitors Cyclosporine and tacrolimus appear in very low amounts in milk. Hundreds of infants have been breastfed with no reports of infant harm. Most international guidelines consider these two drugs acceptable during breastfeeding.
Cladribine is used off-label in multiple sclerosis (MS). The amount of cladribine in milk is low with doses used in MS. Two patients taking cladribine for MS had relative infant dosages (RIDs) of 3% with a 10 mg/day dosage and 4.7% with 20 mg/day. Peak milk concentrations were seen at 1 and 2 hours after the dose, respectively. Data from one patient indicates that the drug is rapidly eliminated over 24 hours and undetectable in milk by 48 hours after a dose.
Although potentially toxic to the infant, cladribine is given with long periods between cycles. This might allow for intermittent breastfeeding with appropriate timing. Manufacturers recommend a 7-day (Europe) or 10-day (U.S.) abstinence period, although these times seem excessive given the published data.
Dimethyl fumarate is used in relapsing MS. It is metabolized to the active monomethyl fumarate. Two nursing mothers with relapsing–remitting MS began oral dimethyl fumarate 240 mg twice daily after discontinuing breastfeeding. Average milk levels of dimethyl and monomethyl fumarate were 2.7 mcg/L and 7.5 mcg/L, respectively. These values indicate that the infants would receive weight-adjusted RIDs of 0.007% and 0.019% of the maternal dosage. U.S. labeling does not preclude mothers from nursing while taking dimethyl fumarate, so it appears to be acceptable during breastfeeding.
Hydroxyurea is an antimetabolite used in certain noncancerous blood disorders such as sickle cell disease. Eighteen women have had milk levels analyzed after hydroxyurea intake. The infant would receive less than 4% of the mother’s weight-adjusted dosage and avoiding breastfeeding for 3 hours after the mother’s dose can decrease the infant dose by about half. In doses used for sickle cell disease, hydroxyurea appears to be acceptable in nursing mothers.
Methotrexate is a folate antagonist used as an anti-inflammatory in low doses (usually ≤25 mg/week) and an antineoplastic in high doses (100–500+ mg/m2 or about 170–850+ mg). Although high-dose methotrexate treatment contraindicates breastfeeding, the issue of low-dose therapy is more nuanced and somewhat controversial.
An old case report found no harm to a breastfed infant whose mother started receiving methotrexate 25 mg once weekly at day 151 postpartum. Some recent cases further indicate that the weekly doses used for inflammatory conditions are unlikely to be toxic to the breastfed infant. A woman with placenta accreta received intramuscular methotrexate 92 mg/day for 4 days, beginning on day 5 postpartum. The estimated daily infant dosages of methotrexate and 7-hydroxymethotrexate were estimated to be 1.2 mcg/kg and 0.2 mcg/kg, respectively. These represent infant dosages of 0.11% and 0.02% of maternal weight-adjusted dosages.
A recent abstract reported 3 postpartum women who were erroneously dispensed methotrexate 2.5 mg/day instead of methylergonovine 2.5 mg/day. They took methotrexate daily for 5, 13, and 15 days while they were breastfeeding. Although all of the women developed toxicity and required hospitalization, none of their infants had clinically observable complications.
Expert opinion and guidelines often warn against use of low-dose methotrexate as an inflammatory because of the small amount of published evidence. However, the tide may be turning, with some guidelines considering doses of ≤25 mg/week as an acceptable alternative to use in nursing mothers. Some possible maneuvers might reduce the risk of infant toxicity. Withholding breastfeeding for 24 hours after a weekly low dose of methotrexate would decrease the infant’s dose by ∼40%. If breastfeeding during long-term, low-dose methotrexate use is undertaken, monitoring of the infant’s complete blood count and differential should be considered.
Mitoxantrone is given as an intravenous infusion once every 3 months for MS. The duration of breastfeeding abstinence after a dose is not clear. In one patient, mitoxantrone was still detectable in milk 28 days after a dose of 6 mg/m2. However, she resumed breastfeeding her infant 3 weeks after the third dose of mitoxantrone at a time when mitoxantrone was still detectable in milk. The infant was apparently well at 16 months of age.
Newer Small Molecules
Janus kinase inhibitors
Janus kinase (JAK) inhibitors are orally absorbable small-molecule immune modulators, although their mechanism is more specific than older immunosuppressants. They modulate the signaling pathway at the point of JAKs.
Tofacitinib is used in psoriasis, rheumatoid conditions, and inflammatory bowel disease and is the only JAK inhibitor that has been studied in nursing mothers. Labeling recommends discontinuing breastfeeding during tofacitinib therapy and for 18 hours after the last dose of the immediate-release form or for 36 hours after the last dose of the sustained-release formulation. Two reports of mothers taking 5 and 10 mg twice daily of tofacitinib found identical worst-case RIDs of 3.4% of the maternal dosage in milk. This value was calculated using peak milk concentrations at 3.5–4 hours and is below the presumed “safe” level of 10%. Using the average milk concentration would produce an even lower RID.
In addition, three mothers have reportedly breastfed their infants while taking tofacitinib 5 or 10 mg twice daily. Two had normal growth and psychomotor development at the time of follow-up and no negative safety signals. Both received the mandatory nonlive vaccinations without complications. The third infant underwent a thorough immunological evaluation at 12 weeks of age that was normal. Two doses of the live rotavirus vaccine were given at weeks 13 and 20 of age with no adverse events reported. The infant was healthy at 12 months of age.
The evidence accumulated thus far indicates that maternal immediate-release tofacitinib does not pose a serious risk to the breastfed infant, but more evidence is needed to be certain. These finding do not necessarily apply to other JAK inhibitors (abrocitinib, baricitinib, momelotinib, ruxolitinib, upadacitinib and [overseas] filgotinib), although two of the drugs (momelotinib and ruxolitinib) are >90% plasma protein bound and unlikely to enter milk in large amounts.
Other protein kinase inhibitors
In addition to the JAK inhibitors, nearly 100 other oral protein kinase inhibitors (e.g., tyrosine kinase inhibitors) are available for a variety of conditions, but very few have been studied during nursing. Only low levels of imatinib were found in the milk of nine mothers, and a few infants have been breastfed safely during maternal imatinib therapy. One mother with chronic myeloid leukemia breastfed three infants in successive pregnancies during imatinib, nilotinib, and dasatinib therapy, respectively, without apparent infant harm. The remainder of the protein kinase inhibitors are generally considered to be contraindicated during breastfeeding. Based on their physicochemical properties, some of these drugs are likely to be found safe during breastfeeding, but clinical data are currently lacking.
Sphingosine 1-phosphate receptor modulators
This class of oral drugs includes etrasimod, fingolimod, ozanimod, ponesimod, and siponimod. They block the capacity of lymphocytes to exit lymph nodes, reducing the number of lymphocytes in peripheral blood. They are primarily used in MS, but ozanimod and etrasimod are approved for treating ulcerative colitis. Although none of them have been studied in breastfeeding, all are ≥98% plasma protein bound, meaning that ≤2% of the drug in the bloodstream is available to transfer into milk. None of these drugs is stated to be contraindicated in breastfeeding in U.S. labeling, but many guidelines, reviews, and European labeling warn against their use in nursing mothers because of the lack of data. Since they are not cytotoxic, likely almost absent from milk, not contraindicated by U.S. labeling, and their effect can be readily monitored in breastfed infants, there appears to be no strong reason to withhold breastfeeding with these drugs.
Monoclonal Antibodies
Monoclonal antibodies are directed against various inflammatory mediators, such as tumor necrosis factor (adalimumab, certolizumab, golimumab, and infliximab), integrins (natalizumab and vedolizumab), the B-cell-specific surface antigen CD-20 (rituximab), or interleukins 12 and 23 (ustekinumab). Professional guidelines are beginning to accept that all monoclonal antibodies used for inflammatory conditions are safe to use during breastfeeding, even though not all of them have been studied in nursing mothers. Monoclonal antibodies used to treat cancer have not yet followed suit. All monoclonal antibodies for cancer treatment are labeled to avoid breastfeeding during treatment and for various time periods after the termination of therapy, usually weeks to months.
There are several reasons why monoclonal antibodies are acceptable during breastfeeding. First, only minute amounts of these large molecules enter the breastmilk, and only in a small fraction of mothers. A recent simulation of passage of 79 monoclonal antibodies into milk indicates that the typical milk-to-plasma ratio range is 0.0003–0.0151. 2 Second, these proteins are ∼50% digested in the infant’s gastrointestinal (GI) tract. 3 Third, what remains of the drug is very poorly absorbed from the GI tract. Monoclonal antibodies have almost never been detected in the blood of breastfed infants unless their mothers’ received doses during the third trimester of pregnancy. And, when they are detected, the levels have decreased over time, even with continued breastfeeding. Additionally, no adverse effects have been reported in any of the hundreds of infants reportedly exposed to a monoclonal antibody in breast milk.
A minor exception to this scheme has been identified, though. That is, in the first days postpartum, the pores in the mammary epithelium are open to allow immunoglobulins to pass into milk. Likewise, pores in the neonate’s GI tract are open to allow these immunologically important factors to be absorbed. So, some authors have recently recommended that monoclonal antibody therapy that had been discontinued during pregnancy not be resumed until 2 weeks postpartum to minimize exposure of the infant to these drugs. 4 However, measurements of monoclonal antibodies in the colostrum of a few mothers have not been much higher than in mature milk.
Summary
Some immunosuppressants are generally considered acceptable to use during breastfeeding, including calcineurin inhibitors, thiopurines, hydroxyurea, and monoclonal antibodies. A few drugs have published data indicating they can be used in nursing mothers with careful infant monitoring, including cladribine, dimethyl fumarate, tofacitinib, weekly low-dose methotrexate, and possibly imatinib. Sphingosine 1-phosphate receptor modulators appear to be low-risk, but are lacking human data.
Footnotes
Author Disclosure Statement
No competing financial interests exist.
Funding Information
No funding support was provided for this work.