Clinical features of infantile‐onset soy‐induced food protein‐induced enterocolitis syndrome

Food protein-induced enterocolitis syndrome (FPIES) is a non-IgE-mediated food allergy that primarily affects infants and young children, characterized by repetitive vomiting 1–4 h after ingestion of the triggering food, often accompanied by pallor and lethargy, followed by diarrhea within 5–10 h.1 The range of causative foods varies by geographic region, and soy is recognized as a major trigger of FPIES in many parts of the world. In a large U.S. cohort of 160 acute and chronic FPIES cases, soy accounted for 41%.2 Soy has long been implicated in chronic FPIES during early infancy, especially in relation to soy formula feeding (“liquid soy FPIES”).3 Chronic soy FPIES is often associated with cow's milk FPIES but typically resolves earlier, often by 6 months of age.4 In recent years, acute soy FPIES during the weaning period (“solid soy FPIES”) has become increasingly recognized.5 In an Australian cohort, soy was the second most common trigger after rice.5 Soy is widely consumed as a general food ingredient, in infant formulas, and as a seasoning such as soy sauce (shoyu) and miso in Asia. Nevertheless, data on the clinical presentation and prognosis of soy FPIES remain scarce. The present study aimed to describe the clinical characteristics, natural course, and prognostic factors of acute soy FPIES in infants. We retrospectively reviewed children ≤18 years of age who underwent soy oral food challenge (OFC) at the Tokyo Metropolitan Children's Medical Center between April 2018 and March 2025. Eligible cases were identified from the institutional OFC database of the Allergy Department. Patients were included if they met the full 2017 international consensus diagnostic criteria for acute soy FPIES (major criterion plus ≥3 minor criteria).1, 6 Demographic and clinical characteristics, including the type of soy products triggering reactions, were extracted from medical records. Serum total IgE, soy-sIgE, and Gly m 4-sIgE were collected (ImmunoCAP). All soy products (e.g., tofu, soy milk, and natto) were eliminated. However, as part of routine clinical practice, soy sauce and miso were not restricted if they had been regularly tolerated before diagnosis. Follow-up OFCs were scheduled according to international guidelines, usually 12 months after the last reaction,1 with earlier assessment in selected cases; second follow-up OFCs, when required, were performed after an interval of 12–18 months. All OFCs were performed in hospital using tofu as the standardized soy challenge food. The OFC consisted of a single dose administered on 1 day, with the initial dose determined by the attending physician according to the severity of prior reactions, ranging from 50 to 250 mg of protein (1 g to 5 g of tofu). Patients were observed for at least 4–6 h after ingestion. In some patients, a repeat OFC with an increased dose was subsequently performed before proceeding to home reintroduction. After a negative OFC, soy intake was gradually reintroduced at home using a standardized stepwise protocol. Intake was increased from 1 g to 5 g and 10 g, with each dose administered on three separate occasions, followed by stepwise escalation to the final target dose (3/8 × 3, 1/2 × 3, 3/4 × 3, and full dose ×3), provided no symptoms occurred. The final target dose was 50 g for children aged 1–2 years and 100 g for children aged ≥2 years. Tolerance was defined as a negative OFC followed by regular home ingestion without symptom recurrence. Confirmation of tolerance was performed at scheduled outpatient follow-up visits, supplemented by structured telephone interviews when necessary, typically 4–8 weeks after the OFC, in line with routine clinical care. After confirmation of tolerance to tofu, caregivers were advised to introduce other soy products (e.g., soy sauce and miso) at home if these products had not been previously consumed; in selected cases (e.g., prior symptoms with miso or caregiver request), an additional OFC using miso was performed before the tofu-based OFC. Statistical analyses used Kaplan–Meier curves and log-rank tests (R 4.5.0). Written informed consent for soy OFC was obtained from all parents and/or guardians. Additionally, an opt-out process was implemented by posting information about the study on the hospital's website, thereby providing an opportunity for individuals to decline participation. This study was approved by the ethics committee of Tokyo Metropolitan Children's Medical Center 2024b-161. Out of 7097 OFCs during the study period, 114 were soy OFCs. Amont these, 19 patients underwent 31 OFCs for the diagnosis or tolerance assessment of acute soy FPIES. Two did not fully meet diagnostic criteria, leaving 17 patients (14 males) for analysis. All cases were acute FPIES during the weaning period; no chronic soy FPIES cases were identified (Figure S1). Table 1 summarizes the patients' characteristics. Median age at onset was 6 months. All patients experienced vomiting, with lethargy in 15/17 (88%), pallor in 12/17 (71%), and diarrhea in 8/17 (47%). No hypotension or hypothermia was observed. Most presented to the emergency room, though only 3/17 (18%) required IV fluids. Eleven (65%) had severe and six (35%) had moderate symptoms. Three (18%) patients had concomitant FPIES to other foods (fish, egg yolk, or egg yolk + cow's milk), and eight (47%) had eczema. Neither soy-specific IgE nor Gly m 4–specific IgE was detected. As shown in Table 2, all 17 patients reacted to tofu. Before tolerance to tofu was confirmed, natto was introduced at home in three patients, whereas miso was ingested either by OFC in three patients or at home in eight patients; reactions occurred in all patients who ingested natto (3/3) and in three of 11 patients (3/11) after miso ingestion. Ten patients tolerated both soy sauce and miso, and none reacted to soy sauce. None had been exposed to soy formula or soy milk before diagnosis. Thirteen patients achieved tolerance at their first follow-up OFC (mean age 17.3 months), and four at their second OFC (mean age 27.5 months) (Figure S2). Detailed patient-level information, including the timing of symptom onset, the timing and administered dose of the tofu-based OFC, and the timing of symptom induction associated with natto and miso ingestion, is summarized in Table S1. No post-OFC relapses occurred. Tolerance rates were 53% at 18 months, 77% at 24 months, and 100% at 30 months (Figure S3). The median duration from onset to tolerance confirmation was 12 months. The presence of diarrhea at initial presentation was significantly associated with delayed tolerance acquisition (median age at tolerance: 24.5 vs. 15.0 months, p = .02) (Table S2 and Figure 1). In an additional analysis, tolerance or regular consumption of fermented soy products (miso and/or soy sauce) was associated with an earlier acquisition of tolerance to soy (median age at tolerance: 14.0 vs. 21.0 months, p = .044) (Table S2). This study provides the first detailed description of infantile-onset acute soy FPIES in Japan. All patients developed symptoms during the weaning period, and all reacted to tofu. The clinical phenotype, vomiting with frequent lethargy and pallor but absence of hypotension, resembles that of hen's egg and wheat FPIES in Japanese infants7 but differs from peanut and fish FPIES cohorts reported in other countries where systemic symptoms were less common.8, 9 Approximately half of the patients presented with diarrhea, a frequency comparable to that reported for wheat-induced FPIES in Japan and within the range reported internationally.7-9 This pattern suggests that while systemic symptoms may be more dependent on host factors, gastrointestinal symptoms may partly reflect antigen-specific properties. Further cross-national studies are needed to explore these mechanisms. An important dietary finding was that all patients tolerated soy sauce and most tolerated miso. Fermentation may reduce allergenicity by degrading proteins, as seen in IgE-mediated soy allergy.10 Given soy sauce's cultural importance, unnecessary avoidance may impair quality of life. Our results support its continued use in most patients, even in those with severe FPIES, provided there is no prior history of reaction. However, in IgE-mediated soy allergy, reactions to soy sauce have been reported to vary by product type and manufacturing process,11 indicating that soy sauce cannot be uniformly regarded as hypoallergenic to soy FPIES. Similarly, miso may require more caution, particularly in patients who experienced severe reactions to tofu, as partial fermentation may leave some allergenic proteins intact. The prognosis of infantile-onset acute soy FPIES in this cohort was favorable, with complete resolution by 30 months of age. This is earlier than typical resolution ages for egg, wheat, peanut, or fish FPIES,12 suggesting a shorter natural course; however, the reasons for this difference remain unclear. The presence of diarrhea at initial presentation was associated with delayed tolerance, consistent with a recent report in cow's milk FPIES.13 This may reflect more extensive intestinal inflammation, leading to prolonged disease activity. Nevertheless, the small sample size limits the ability to draw definitive conclusions, and further multicenter studies with larger cohorts are needed to better clarify these associations. A key strength of this study is the strict diagnostic criteria and uniform OFC-based assessment of tolerance. This ensures high diagnostic specificity. Limitations include the small, single-center design. Additionally, our findings reflect Japanese dietary practices, where soy formula is used only in therapeutic contexts, which may also account for the lack of chronic soy FPIES in this cohort. In conclusion, infants with acute soy FPIES demonstrated that the median age at tolerance acquisition was 20 months. The presence of diarrhea at the initial presentation was associated with delayed tolerance, indicating that gastrointestinal symptom patterns may influence disease trajectory. Notably, soy sauce was tolerated by all patients and miso was tolerated by most patients, indicating that fermentation may reduce the allergenicity of soy products. These findings provide new insights into the clinical course and dietary management of acute soy FPIES and may support individualized care strategies, including the timing and selection of foods for OFC. Naoki Kajita: Conceptualization; methodology; investigation; visualization; writing – original draft; formal analysis. Yuki Murata: Investigation; writing – review and editing. Ryosuke Kagami: Investigation; writing – review and editing. Kumiko Morita: Investigation; writing – review and editing. Hideaki Morita: Writing – review and editing; supervision. Koichi Yoshida: Supervision; writing – review and editing; investigation; project administration. We are indebted to Mr. James R. Valera for his critical reading of the manuscript. No funding. The authors have no conflicts of interest to declare. The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1111/pai.70342. Figure S1. Figure S2. Figure S3. Tables S1–S2. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.