Monoclonal Antibodies

The conventional model of drug passage into breastmilk states that large molecules such as maternal immunoglobulins can pass into colostrum because spaces between mammary epithelial cells are open. Then, these pores begin to close at the onset of lactogenesis II and are fully closed by 10 to 14 days postpartum when tight junctions are formed between the cells, prohibiting the direct passage of “large molecules.” Very small molecules such as alcohol and urea can still pass through the aqueous channels between cells and equilibrate rapidly between maternal plasma and breastmilk. As the theory goes, moderate sized molecules can cross the semipermeable membrane formed by the epithelial cells and tight junctions, but have to pass across the cells. Their passage is governed by factors such as pKa (favoring passage of weak bases), protein binding (favoring passage of poorly bound drugs), and molecular weight (favoring smaller molecules). Highly charged molecules such as heparins and very large molecules such as monoclonal antibodies are said to be prohibited from passing into breastmilk. The cutoffs between small molecules, moderate sized molecules, and large molecules are not well defined, but the cutoff between molecules that can pass into milk and those that cannot is said to be around 800 Da.

Given this model, conventional wisdom indicates that biosynthetic monoclonal antibody drugs should not pass into breastmilk because they have molecular weights of around 140,000 Da. Furthermore, monoclonal antibodies are proteins and would be digested in the infant's gastrointestinal tract. And, after the initial few days of life, the infant's gastrointestinal tract becomes impermeable to large molecules, so monoclonal antibodies should not be absorbed, creating a third line of defense against systemic exposure of the infant to monoclonal antibody drugs.

Reality always seems to be messier than our models might suggest. As early as 2004, traces of etanercept and infliximab were found in the breastmilk of mothers who were receiving the monoclonal antibody for treatment of rheumatoid arthritis flare.^1,2 Since that time, several other cases have been reported, in which a monoclonal antibody given to the mother was detected in breastmilk. How could this be? One possibility is that mastitis, even subclinical mastitis or other inflammatory conditions, can open the tight junctions between mammary epithelial cells. We have no direct evidence of this phenomenon with monoclonal antibodies, but studies in the late 1980s on the treatment of human mastitis with beta-lactam antibiotics indicate that this can happen. Penicillin V concentrations in the milk from mastitic breasts were higher than the drug concentrations in unaffected breasts in one study. ³ A study from Japan compared passage of cefuroxime into the breastmilk of women with and without acute mastitis. ⁴ Not only was the cefuroxime concentration higher in mastitic breasts but milk from the contralateral breast also contained higher concentrations of the drug than did the breastmilk of women without mastitis who were given the drug.

Other articles have found traces of antitumor necrosis factor (TNF) monoclonal antibodies in the serum of breastfed infants. These findings are tricky to interpret because the half-lives of monoclonal antibodies are generally quite long, and if a mother received the drug too close to delivery, the baby could be born with the drug in the bloodstream. Several cases have been reported in which this seems to have been the reason for detectable monoclonal antibodies in the infant's serum. Could a monoclonal antibody be excreted into milk and absorbed by a breastfed infant? At the present time, only one case has been reported in a partially breastfed infant whose mother was receiving infliximab. ⁵ The infant was 4 months old and had a subtherapeutic plasma infliximab level of 1.7 mcg/L, 2.2% of the simultaneous maternal plasma concentration, which was quite elevated. Perhaps the high maternal level and gastrointestinal inflammation from partial breastfeeding contributed.

Another interesting property of monoclonal antibody excretion into breastmilk is its time course. Most drugs appear in breastmilk within a few minutes and rise to a peak concentration in an hour or two. Measurements of sequential levels of several monoclonal antibodies in breastmilk indicate that milk levels can increase for several days after a dose. For example, intravenous natalizumab used for multiple sclerosis was undetectable in breastmilk after a 300 mg dose until day 14 when a concentration of 333 mcg/L was measured. A peak level of 1.01 mg/L was detected on day 20 and a second dose was given on day 29. Milk levels increased to a maximum of 2.83 mg/L on day 50. The latter represented a surprisingly high relative infant dose of 5.3%. ⁶ In another patient, ipilimumab levels in breastmilkpeaked 10 days after the first intravenous infusion and 4 days after the second infusion. ⁷ These cases illustrate the difficulty of timing of breastfeeding with respect to maternal dosage with monoclonal antibodies. Nevertheless, the concentrations of most monoclonal antibodies found in breastmilk are very low, and the second two lines of defense, protein hydrolysis and poor oral absorption, are presumably still intact.

When monoclonal antibody records were initially written for LactMed, standard language was used in most drug records because most of the monoclonal antibodies had not been (and still have not been) studied during breastfeeding. The language used in the records of drugs with no information reflects the standard model already described such as this statement on ustekinumab: “Because ustekinumab is a large protein molecule with a molecular weight of about 149,000, the amount in milk is likely to be very low and absorption is unlikely because it is probably destroyed in the infant's gastrointestinal tract.” For drugs with information on the excretion into milk, the first part of the statement is modified accordingly, but the portion concerning absorption by the infant is left the same. Finally, a statement is given regarding the advisability of using the drug during breastfeeding. If experts in the field have provided opinions, these are stated. Otherwise, the general statement is provided: “Until more data become available, ustekinumab should be used with caution during breastfeeding, especially while nursing a newborn or preterm infant.”

This was where LactMed stood until the data about prolonged natalizumab excretion and infliximab absorption were published and concern was raised about the small amounts that appear in milk. Since then, LactMed records have taken into account the package insert warnings against use during breastfeeding. If the package insert states that the drug should be “used with caution” during breastfeeding, the standard wording mentioned was used and no additions were made to the “Summary of Use during Lactation” section. If the package insert warns that the drug should not be used during breastfeeding, the following statement was added to the end of the Summary: “The manufacturer recommends that breastfeeding be discontinued during therapy.” This statement reflects both our lack of knowledge about the clinical relevance of the small amounts of some monoclonal antibody drugs in breastmilk and the medicolegal consideration of failing to have a statement representative of the package insert's warning. In general, most of the warnings against use of monoclonal antibodies in nursing mothers are found in package inserts of drugs used to treat cancers, whereas most package inserts for monoclonal antibodies used for other conditions such as rheumatoid arthritis or inflammatory bowel disease merely recommend caution.

It is important to note that no adverse reactions in breastfed infants have been reported with the use of any monoclonal antibody in nursing mothers, although use has been limited outside the areas of rheumatoid arthritis and inflammatory bowel disease. Several expert guidelines published in late 2015 and early 2016 uniformly state that monoclonal antibodies used to treat inflammatory bowel disease and rheumatoid arthritis (mostly anti-TNF agents) are acceptable to use in mothers who are breastfeeding.^8–11 A similar consensus has not been reached on monoclonal antibodies used to treat cancer or multiple sclerosis.