What Does the Latest Research Say About Soy in Your Breastmilk?

Apr 26
9 min read

You've already pulled dairy, the most commonly recommended elimination when a baby shows signs of food reactivity. Maybe symptoms eased, but not enough. Maybe nothing changed at all. Now your provider is recommending the next step on most elimination lists, soy.

If pulling dairy felt like a full-time job, soy is another one. It hides in places dairy never went. Salad dressings, breads, "natural flavors," vitamins, even the lactation cookies you bought to support your supply. We hear you, and we want to make this part easier.

This post pulls together what the science actually says about soy reactivity in breastfed babies, including a brand new study we co-authored that adds new clarity to a question families have been asking for years.

Hands cutting tofu on a wooden plate, beside a bowl of soybeans, on a light surface. Soft focus, calming scene, pastel colors.

What Soy Actually Is and Why It Hides So Well

Soy, soybean, and soya all describe the same legume. Soybeans pack a strong protein profile, which is exactly why food manufacturers love them and why soy shows up across so much of the U.S. food supply.

Reading labels becomes your new (unwanted) hobby. You'll spot soybean oil, soy lecithin, soy protein isolate, and a long list of other soy-derived ingredients. What matters for your baby is not the ingredient name. It is whether enough intact soy protein is present to trigger reactivity.

Soybean oil and soy lecithin sit in a gray zone. The processing they go through removes most of the soy protein, which is why many soy-reactive babies tolerate them without issue. Researchers have nonetheless identified IgE-binding proteins in soy lecithin at low levels, and clinical reactions to soy lecithin in highly sensitive individuals have been reported in the peer-reviewed literature [3]. Every baby has a different threshold.

This is why we usually recommend a strict elimination first, then a careful reintroduction with professional guidance. You learn what your specific baby tolerates, rather than guessing.

A man in an apron serves food from a pot while a woman eats with chopsticks in a cozy kitchen. They appear happy. Counter has drinks and snacks.

Soy Got Easier to Spot, Sort Of

In the U.S., soy is one of the nine major allergens that must be declared on packaged food labels. Look for it in parentheses next to the ingredient (like "lecithin (soy)") or in a "Contains" statement after the ingredient list [1].

Here is where it stops being easy.

Free-from claims are not regulated the same way. The only free-from claim with an enforced FDA threshold is "gluten free," which requires the food to contain less than 20 parts per million of gluten [2]. "Soy free," "dairy free," and "wheat free" are marketing language. They mean whatever the manufacturer wants them to mean. The simple rule of thumb is this. If you see "Contains: Soy," you can trust that. If you see "soy free," you cannot.

This is the part that wears people down. For more on hidden allergens and the gaps in U.S. labeling, see our blog on hidden food allergens.

Soy also slips through under names that don't sound like soy at all. Bean curd, edamame, miso, natto, tempeh, tofu, and yuba are direct soy-based foods. Other ingredients are sometimes (not always) derived from soy, including hydrolyzed vegetable protein, textured vegetable protein, mono- and diglycerides, vegetable broth or starch, tocopherols (a form of vitamin E often sourced from soy), and natural or artificial flavors carried in soy oil. When a label is unclear, contacting the manufacturer is usually the fastest path to an answer.

Eating outside the home becomes the hardest part. Soy oil is the default frying oil at most U.S. restaurants. Sauces and dressings are built on it. The "quick dinner picked up on the way home" plan disappears for a while. It is not you. It is the food system.

Cubed tofu browning in a black frying pan, glistening with oil. The tofu is seasoned with a dark spice blend, creating a rich, savory look.

What Reactivity in a Breastfed Baby Actually Looks Like

Symptoms vary so much from baby to baby. Some show one thing. Others show a constellation. Common signs of food reactivity in breastfed babies include reflux and frequent spit up, vomiting, eczema flares, hives or rash, mucousy or bloody stools, diarrhea or constipation, poor weight gain or failure to thrive, and prolonged colic-style fussiness [4]. Severe reactions can include lethargy, paleness, or airway involvement, which always require urgent evaluation.

If you are seeing a pattern and it feels real to you, it is worth paying attention to. Reactivity in breastfed babies is well-documented, and the medical literature is finally catching up to what families have known for decades about soy in your breastmilk.

IgE-Mediated, Non-IgE-Mediated, and Why Standard Testing Often Misses the Mark

Soy can drive two distinct types of immune response, and the difference matters for how you'll get answers.

IgE-mediated reactions happen quickly, usually within minutes to two hours of exposure. These are the responses skin prick and blood IgE testing pick up. Symptoms tend to include hives, swelling, vomiting, and in severe cases anaphylaxis. Soy allergy affects approximately 0.4% of U.S. children, with onset typically in infancy [5]. In the subset of infants with IgE-mediated cow's milk allergy, around 14% will also react to soy in formal food challenge studies [6].

Non-IgE-mediated reactions are far more common in young infants, and they show up hours to days after exposure. Standard allergy testing does not detect them. Categories include food protein-induced allergic proctocolitis (FPIAP), which presents with bloody or mucousy stools in an otherwise well-appearing baby, and food protein-induced enterocolitis syndrome (FPIES), which presents with severe delayed vomiting and lethargy [7], and more.

Cow's milk and soy are the most common triggers for both, and co-reactivity is high. In FPIES, 40-60% of cases react to both cow's milk and soy. In FPIAP and the related condition of food protein-induced enteropathy, the rate is closer to 10 to 30% [7].

If you have a young baby, especially one under six months, allergy panels alone will rarely give you a complete picture. A negative skin prick or blood test result for soy does not mean soy is safe. It means IgE is not the mechanism in play. In simple terms, testing only catches part of the picture, which is why elimination and reintroduction still matter.

The most reliable diagnostic tool we have for non-IgE reactivity is still the same one families have used for decades. You eliminate the suspected trigger, you watch for change, and you confirm with a careful reintroduction. Many families see meaningful improvement within about five days of pulling the right trigger.

Can Soy Pass Into Breastmilk? What the Latest Research Reveals

This is the part of the conversation that has shifted the most since we last wrote about soy.

For years, families on a soy elimination heard the same vague advice. Pull all soy, give it weeks, hope for the best. The problem? No published study had directly measured soy protein transfer into human milk. Everyone was guessing.

We changed that.

In October 2025, our team published the first peer-reviewed study to use highly sensitive LC-MS/MS peptidomics to look for soy-derived peptides in human milk after maternal soy consumption [8]. Dr. Trill is a co-author on this study alongside researchers from the University of Idaho, Oregon State University, and Rockefeller University. Importantly, the study did not include infants with known soy reactivity, so the findings cannot rule out reactivity in that specific population. With that boundary in mind, here is what we did and what we found.

We had 24 lactating participants eliminate soy for five days, then reintroduce a soy beverage in increasing daily amounts of 175, 295, and finally 415 mL over three days. Milk samples were collected before consumption and at 2 and 4 hours after the largest dose. The study used an identical design for cow's milk in a parallel arm.

For cow's milk, we identified 121 distinct bovine-derived peptides across the samples, mainly from β-lactoglobulin and the caseins. The relative abundances were low overall, and they did not consistently rise after consumption. Some peptides actually appeared at higher levels after five days of elimination, which could suggest accidental exposure (elimination is hard) based on other previous research [8].

For soy, we initially detected eight peptides that might be soy-derived. When we ran them through the NCBI BLAST protein database, only two matched soy exclusively, and even those shared significant homology with other plant sources. The other six matched multiple non-soy foods. Most telling, when we analyzed the actual soy beverage participants drank, none of its 710 peptides matched the eight found in milk [8].

In plain language, we did not find confident evidence that soy protein from a soy beverage transfers into human milk in a detectable, soy-specific form.

This finding aligns with two earlier proteomics studies that also looked broadly at non-human peptides in human milk and reported zero confirmed soy-derived peptides [8].

Frontiers in Nutrition article header on bovine milk vs. soy peptides in human milk after maternal consumption. Includes authors and citation.

What This Soy Research Means and What It Does Not

We want to be careful here, because it would be easy to misread these findings.

What the research suggests is that soy protein from a typical soy food source may not transfer to human milk in the same robust, detectable way that cow's milk and peanut proteins do. That is genuinely meaningful, and it offers some reassurance to families who have been told to eliminate soy for weeks alongside dairy.

What the research does not say is that soy reactivity through breastfeeding never happens or that families should ignore symptoms. Several real limitations apply. Our participants did not report soy allergy or have soy-allergic infants, so we did not test the question in the most clinically relevant population. We used one type of ultra-processed soy beverage. Other soy foods, especially less-processed forms like edamame, miso, or tofu, may behave differently. Reactivity is also not always about transferred peptides. Some infants react to soy directly when they begin eating solids, even if maternal soy did not trigger reactions through milk.

A note from Dr. Trill as one of the authors on this study. If you have a baby who reacts when you consume soy, your experience is real. Our participants were not soy-allergic, and we tested one ultra-processed soy beverage, not the full range of soy foods families actually eat. What we showed is genuinely meaningful, and it is also a starting line, not a finish line. The families living this every day are the ones who need the next round of research, and that is the work I will keep pushing.

Practically, this means that if your baby's symptoms fully resolve with dairy elimination alone, soy may not need to come out. If symptoms persist on dairy elimination and you and your clinician are weighing soy as the next step, the new evidence supports a focused trial rather than weeks of restriction. Five days of strict elimination is generally enough to know whether soy is actually a contributor for your baby. If symptoms have not improved by then, soy likely is not the trigger, or it is one of several co-triggers that need to be unraveled together.

This is exactly the kind of decision-making we walk families through one step at a time.

Two children lying on a floral-patterned sheet; a baby wrapped in white sleeping, and a toddler in a white shirt smiling contentedly.

How Soy Reactivity Tends to Resolve

Most infants outgrow soy reactivity over time. A natural history study of children with IgE-mediated soy allergy found that 25% had developed tolerance by age 4, 45% by age 6, and 69% by age 10 [5]. For FPIES specifically, the majority of cow's milk and soy cases resolve by age 3 to 5 [7].

Reintroduction matters. It helps confirm whether soy is truly a trigger and prevents unnecessary long-term restriction. Once your baby has been symptom-free on a soy-free diet for a meaningful stretch, a structured ladder approach lets you trial small amounts of less-allergenic forms first, then move toward more concentrated ones. We have built reintroduction ladders for soy and the other top allergens to help guide that process.

Woman holding a white plate of green beans with red sauce, wearing a black top and colorful bracelets. Sunlit outdoor setting.

You Don't Have to Figure This Out Alone

Soy elimination is harder than it sounds in print. Reading every label, navigating restaurants, identifying hidden sources in vitamins and medications, all while running on broken sleep and watching your baby for signs of healing. We see you, and we have built this whole company around making this part of the journey less lonely.

If you want a clear plan instead of guessing, our Elimination to Reintroduction Intensive walks you through both phases with two consultations from our Director of Allergy Support, Annie, three pre-recorded courses, and a personalized elimination and meal plan.

Tracking patterns between what you eat and what your baby experiences can help you get answers faster, especially during elimination and reintroduction.

For staying current on the research as it evolves, follow us on Instagram at @free.to.feed.

You are doing real work here. Healing is rarely linear, and the science is finally catching up to your lived experience. We are proud to keep pushing both forward.