Occurrence and Co-Occurrence of Regulated and Emerging Mycotoxins in Foods Marketed to U.S. Toddlers

1. Introduction

Mycotoxins are low-molecular-weight secondary metabolites produced by filamentous fungi under favorable environmental and storage conditions [1,2]. Since the 1960s, hundreds of mycotoxins have been identified, with dietary ingestion representing the primary route of human exposure [1]. Inhalation and dermal contact may also occur in mold-contaminated indoor environments, such as water-damaged buildings, and in occupational settings, including bakeries and breweries, making complete avoidance difficult [2,3,4].

Historically, outbreaks of mycotoxin-related foodborne illness were often regionally confined before the expansion of global trade [2,5]. However, mycotoxin contamination is now widespread, with an estimated 60–80% of cereal grains and nut crops affected worldwide, posing risks to both human and animal health [1]. As a result, humans are continuously exposed to low- and, at times, high levels of dietary mycotoxins, with infants and young children among the most disproportionately affected due to lower body weight and higher relative toxin exposure [6,7,8].

The primary mycotoxin-producing genera affecting the human food supply include Aspergillus, Penicillium, Fusarium, and Alternaria species [2]. These fungi produce a chemically diverse range of toxic compounds, including aflatoxins, ochratoxins, trichothecenes (e.g., deoxynivalenol, nivalenol, HT-2 toxin, and T-2 toxin), zearalenone, fumonisins, enniatins, beauvericin, alternariol mono-ether, and alternariol [2]. Many of these metabolites have been shown to adversely affect multiple organ systems, including the gastrointestinal tract, liver, kidneys, nervous system, reproductive organs, and epithelial and immune tissues [3,4,9,10].

Within the context of organ-specific toxicities, many mycotoxins have acute and chronic health effects. For example, at low but chronic doses, the commonly occurring trichothecene mycotoxin deoxynivalenol (DON) may stimulate immune activation, inhibit protein synthesis, compromise the intestinal mucus barrier, increase intestinal permeability, and suppress appetite, thereby potentially contributing to growth stunting in children [11,12,13,14,15,16,17,18]. In higher, more acute exposures, DON has been linked to immune suppression, nausea, vomiting, abdominal pain, diarrhea, intestinal hemorrhage, and fever in both animal models and human outbreaks [5,9,11,12,13,16,19,20,21]. These effects have led some researchers to hypothesize that chronic trichothecene ingestion may contribute to inflammatory gastrointestinal (GI) disorders, including inflammatory bowel disease (IBD) [3,13,19,20,21].

Following systemic absorption, some mycotoxins have been found to cross the blood-brain barrier. DON and ochratoxin A (OTA) can cross the blood–brain barrier and have been shown to damage astrocytes and glial cells in animal and in vitro models [15,22]. Detoxification differences between developmentally typical children and their siblings with autism spectrum disorder (ASD) have raised concerns about their potential role in ASD [14,15,22,23,24,25].

Other mycotoxins also exhibit distinct organ-specific effects. Cyclopiazonic acid, an unregulated mycotoxin, has been linked to GI and neurological toxicity, including human outbreaks of Kodua poisoning [26,27,28,29,30]. Aflatoxins are well-established hepatotoxins and hepatocarcinogens, while enniatins and Alternaria toxins, including alternariol (AOH) and alternariol monomethyl ether (AME), may also contribute to liver pathology, although their roles remain less well defined [31,32,33,34,35,36,37,38]. Zearalenone (ZEA), AOH, and AME exhibit hormone-disruptive activity, and ZEA has been associated with reproductive abnormalities, including links to polycystic ovarian syndrome in animal models [2,37,39,40,41,42,43,44,45]. Fumonisins exposure has been implicated in the development of neural tube defects in infants and possibly esophageal cancer in humans [2,46,47]. Aflatoxin B1, fumonisins, and DON have recently been under investigation for their potential role in growth stunting in children [48,49].

In addition to direct toxicity, many mycotoxins disrupt intestinal microbial communities in a manner analogous to antibiotics. DON alters Bacteroides and Firmicutes populations [50,51,52]. Aflatoxins reduce lactic acid–producing bacteria. OTA targets lactobacilli, including Lactobacillus reuteri [53]. Beauvericin exhibits broad antimicrobial activity against Gram-positive and Gram-negative organisms, potentially contributing to dysbiosis [54,55,56,57].

Collectively, these findings highlight the potential for foodborne mycotoxins to negatively impact the human microbiome and overall health. Despite decades of research and the identification of many foodborne mycotoxins with known toxicity, numerous fungal metabolites are still underrecognized in clinical and public health fields [9,30,58,59].

Exposure can start in utero, as mycotoxins have been found in amniotic fluid and linked to impaired fetal growth [60]. In a prospective cohort study, Tan et al. (2023) reported that mothers in the highest tertile of DON intake were more likely to deliver infants with lower birth weight and reduced birth length relative to gestational age (all P-trends < 0.05) [61]. Postnatally, exposure may continue through contaminated breast milk, infant formula, and cow’s milk, with additional increases as solid foods are introduced [6,8,61,62,63,64].

Following weaning, toddlers consume a broader range of solid foods, many of which are grain-based products such as cereals, crackers, breads, and snack foods. Because grains are common sources of mycotoxins and are consumed more frequently relative to body size, this transition may further increase exposure compared with infancy [7,65]. These risks may be amplified by developmental vulnerabilities, including immature, less resilient microbiomes, underdeveloped metabolic and immune systems, reduced detoxification capacity, and lower body weight [8,66].

Compounding this concern, mycotoxins are imperceptible to humans by taste, odor, or visual inspection, rendering avoidance difficult without advanced analytical methods such as liquid chromatography–tandem mass spectrometry [67]. Historically, most surveillance efforts have focused on a limited subset of regulated mycotoxins, which may be why only a small number are currently subject to regulatory limits in the United States (U.S.). Currently, only limited data are available on the occurrence and co-occurrence of regulated and emerging mycotoxins in the U.S. food supply.

To address these gaps, the present study evaluates the occurrence and co-occurrence of 34 regulated and emerging mycotoxins in processed foods marketed for U.S. toddlers, providing a snapshot of potential early-life dietary exposures (see Table 1 for more information on the analytes tested). Specifically, this investigation seeks to address the following questions: (1) which of the 34 mycotoxins tested are U.S. toddlers likely to be exposed to through commonly available processed foods, and (2) whether observed contamination levels exceed established European tolerable daily intake (TDI) values when adjusted for toddler body weight, thereby potentially increasing risks to child health. European TDI values were used where available, as no established pediatric TDI values are available in the U.S.

2. Materials and Methods

To assess for multi-mycotoxin contamination, 118 food products were screened for 34 mycotoxins using liquid chromatography–tandem mass spectrometry (LC–MS/MS) . A validated multi-analyte method was applied for sample preparation and quantification as described by Kresse et al. [67] (See Table 1 for analytes tested). Quality control measures were implemented throughout the analytical process.

2.1. Sample Selection and Procurement

Foods specifically marketed for infant and child consumption, including cereals, snacks, pastas, first foods, juices, and basic cooking ingredients, were selected for analysis. For baby food products, only vegetable-based, non-dairy items were included, as defined a priori by the study team. Cereal grain-based foods and some non-cereal grain ingredients were selected for testing. Non-cereal-based ingredients included chocolate powder, soy-based formula, almond flour, coconut flour, arrowroot starch, and tapioca starch, and miscellaneous other products (e.g., first food beans and apple juice).

Products were purchased at retail locations and by mail order in the Midwestern United States (Missouri) and the Northeastern United States (Massachusetts, New Hampshire, and Maine) between 1 June 2024 and 1 October 2025. In the absence of a formal randomization protocol, a standardized systematic sampling approach was implemented to reduce selection bias when purchased directly from individual stores. Specifically, the second item in each product row was selected, provided that the packaging was intact. Additionally, only one product was selected from each product category (e.g., cornflakes) on the shelf without regard to brand. However, because two separate people procured the food items, duplicate products from different geographic regions and different lots were sometimes selected.

To supplement retail sampling and expand geographic coverage, a mail-order procurement strategy was also implemented to obtain commercially available food products for mycotoxin analysis. Products marketed for infant and toddler consumption, along with some basic cooking ingredients, were identified through major online retailers and manufacturers' websites via Google search and purchased using a standardized selection approach. Within each product category, one item from every other available brand was selected to reduce selection bias.

Prior to laboratory submission, all products were inspected to confirm package integrity, absence of visible damage (e.g., rips or tears), and validity of expiration dates to ensure sample quality. Details for each product were also noted, including manufacturer lot numbers and storage facility numbers, when available. However, such details were not available for all products procured.

2.2. Sample Shipment and Laboratory Analysis

After procurement, samples were sent via U.S. Postal Service Priority Mail to Trilogy Analytical Laboratory (Washington, MO) for analysis of 34 foodborne mycotoxins using LC-MS/MS procedures. Once received by the laboratory, samples were stored in their original, unopened packaging under ambient, dry conditions consistent with manufacturer recommendations. This was done to minimize potential contamination or degradation prior to analysis. All samples were logged with the receipt date, product type, brand, lot number (if available), and country of origin if disclosed on the label. Samples were stored for no more than two weeks until testing. Samples were tested in groups based on matrix (e.g., packaged cereals).

2.3. Reagents and Standards

To test for 34 mycotoxins, reference standards for all targeted mycotoxins (≥98% purity) were manufactured by Trilogy Analytical Laboratory (United States) or purchased from Romer Labs (United States). Stock solutions were prepared in 1 mg/ml concentrations in LC–MS grade acetonitrile, methanol, and water. LC–MS grade acetonitrile, methanol, and water were obtained from Honeywell. Formic acid (≥99% purity, LC–MS grade) was obtained from Millipore Sigma.

2.4. Sample Preparation

Diverse food subsets required varying methods of preparation. For products prone to heterogeneity (e.g., cereals, mixed snacks), incremental portions were taken from multiple locations within the homogenized matrix to ensure representativeness. Liquid products were mixed thoroughly prior to aliquoting.

Samples were then prepared for testing using the following methods: Samples were ground using a Retsch GM200 mill or homogenized using a high-speed blender (Osterizer) until a uniform particle size was achieved. 25-gram samples were then weighed in an extraction flask, and 100 mL of acetonitrile–water (80:20, v/v) was added.

Extraction was performed on an orbital shaker (6000, Eberbach) for 90 min at room temperature (20–22 °C). The samples were then filtered with a Whatman No. 1. The filtered extract was then purified with a QP1100 SPE (Trilogy Analytical Laboratory, United States). The purified extracts were then diluted with water-acetic acid (2% v/v) and injected into the LC-MS/MS system.

2.5. LC–MS/MS Instrumentation and Parameters

For LC-MS/MS analysis, the Shimadzu SIL-40 series system was configured with a near-zero refrigerated autosampler (temperature maintained at 8 °C), a gradient pump, and a column oven. This was combined with the SCIEX 7500 mass spectrometry system by an Electrospray Ionization Interface (ESI). The preinstalled 6-port valve on the MS-MS instrument was configured to direct the analyte flow to waste. The SCIEX OS acquisition software controlled all systems. For data evaluation, the SCIEX OS software was applied [108].

2.5.1. Chromatographic Separation

Separation was performed using a Phenomenex Kinetex column (3 mm x 100 mm; 2.6 μm; 100 Å). The column was stored in the column oven. The temperature was set to 40 °C, and the injection volume was 5 μL.

For the separation, a gradient program was used. The gradient profile started at 5% B, held for 1.0 minute, and then decreased linearly to 50% B within 6.0 minutes. This was maintained for 12.0 minutes before being returned to 5% B in 4.0 minutes. The starting conditions were held for 4.0 min before the next injection.

The mobile phases for the analysis were A: water-0.1% Formic Acid-4 mM Ammonium Formate and B: Methanol-0.1% Formic Acid-4 mM Ammonium Formate. Flow rate was 0.5ml/min.

2.5.2. Mass Spectrometry Conditions

The parameters for MS/MS were as follows: ESI positive with interface temperature 350 °C, ion spray voltage 1700 V, curtain gas 40 psi, collision gas 9 psi, ion source gas 1 set at 40 psi, and gas 2 set at 60 psi. The mass spectrometry worked in the scheduled multiple reaction monitoring (sMRM), and the two most intensive transitions were measured when the relevant substance was eluting to the MS/MS. The dwell times were optimized automatically by Sciex OS Software.

Analyses were performed in scheduled multiple reaction monitoring (sMRM) mode. For each analyte, the two most intense precursor–product ion transitions were monitored within a 1.2 min retention window. Dwell times were automatically optimized by Sciex OS software ([3.0, Sciex]) [108] to ensure adequate data points across each peak. The window of one MSM signal was 1.2 minutes, and the total cycle time was 1.1 s [108].

Analytes required both positive and negative ionization polarities. Complete MS/MS operating settings are provided in Table 2 [67].

2.5.3. Method Validation

Method performance was evaluated in accordance with FDA guidelines. Linearity was assessed using matrix-matched calibration curves prepared in triplicate across a range. Coefficients of determination (R²) exceeded 0.99 for all analytes (See supplemental verification data).

Limits of detection (LOD) and quantification (LOQ) were determined based on signal-to-noise ratios of 3:1 and 10:1, respectively, using spiked blank matrices. Recoveries were assessed by fortifying blank matrices. Mean recoveries ranged from 70% to 120%, with relative standard deviations (RSD) ≤15%. Matrix effects were evaluated by comparing slopes of calibration curves prepared in solvent versus matrix extracts. If mycotoxins were detected, but below the typical LOQ, the data were considered not quantifiable and discarded from the total dataset.

2.6. Statistical Analysis and Comparison to Established Tolerable Intake Levels

For each sample, data were entered into a Microsoft Excel (Version 2601 Build 16.0.19628.20132) spreadsheet and evaluated using descriptive statistics [109]. Data were then converted to micrograms per kilogram (µg/kg) for ease of clinical interpretation and comparison with TDI and current regulatory guidance levels. The mean number of samples in which at least one mycotoxin was detected was also calculated for each tested category. R statistical software (version 4.5.2) running under RStudio [110] and add-on packages were then used to further evaluate the data [111,112,113,114]. A simulated Pearson’s chi-squared test was conducted to determine the distributions of mycotoxins across all groups.

Median and maximum detection levels were computed for mycotoxin contamination levels; mean rates were calculated for the number of positive samples. Maximum and median detection levels were then evaluated against body weight (bw) to assess pediatric risk using pediatric portion sizes (e.g., 57 grams (g) or approximately two ounces), with current TDI recommendations using an algorithm [115]. This equation was developed by the Italian National Institute of Health, the Department of Innovation in Biology at the University of Tuscia, Italy, and an Italian food company to produce processed wheat-based products with child-safe DON levels [115]. This equation provides a method for comparing the median and maximum mycotoxin levels found in food samples with established TDIs.

Exposure ng/kg bw/ day = Mycotoxin Contamination level (µg/kg) × Consumption data (g)/bw (kg)

For consistency, mean toddler weights-for-age for a 24-month-old were obtained from the Centers for Disease Control and Prevention (CDC) growth charts. The 50th percentile for girls and boys combined (12.5 kg/bw) was input into the equation [116].

Because the product was reported in nanograms, the units were converted to micrograms by dividing the product in nanograms by 1000 to provide intake levels in µg/kg bw/day for comparison with the TDI limit for each toxin. TDI limits have not been established for some of the mycotoxins tested for. In such cases, an accurate risk assessment was not possible.

Microsoft Excel was used to evaluate if toddlers would exceed TDI exposure limits using the following equation [109]:

Toddler Exposure =TDI (µg/kg bw/day) /Actual intake (µg/kg bw/day)

This method allowed easy comparison of mycotoxin exposure levels in foods with established TDI limits. All pediatric exposure levels above the TDI limit guidelines were considered a risk to child health.

3. Results

One hundred and eighteen (118) processed foods and ingredients were analyzed. A broad range of foods was sourced for testing, including snack foods, first foods and formulas, packaged cereals, and pastas (n = 72 samples), as well as raw ingredients used in food preparation (n = 46 samples). Contamination rates differed across commodities. The categories of food samples tested are listed in Table 3.

Of 118 samples of food and basic ingredients tested, 88.1 percent (95% CI = (81.1, 92.8)) were contaminated with at least one mycotoxin; the median was 4.0, the mean was 3.8, and the maximum was 13 mold contaminants. In addition, of the 34 mycotoxins tested, 26 were detected at varying levels within the sample set. Roquefortine C, Griseofulvin, Aflatoxin G2, Diacetoxyscirpenol, Fusarenon-X, Neosolaniol, Citrinin, a-Zearalenol, and b-Zearalenol were not detected in any product. Out of 4,012 tests (118 samples × 34 toxins), 445 (11.1 %) were positive for some mycotoxin. A simulated Pearson’s chi-squared test indicates that the proportions of products with at least one mycotoxin detected were approximately equal across food types (P=0.61). However, the types of mycotoxins found differed by commodity. Figure 1 shows the distribution of regulated, unregulated, and emerging mycotoxins detected in foods. Table 4 presents the number of samples contaminated within each mycotoxin category, the level of quantification (LOQ) in parts per million (ppm), the highest and median amounts found in each category, and the type of food in which the highest amount was detected.

Table 5 demonstrates the distribution of mycotoxins by chemical structure across food types. Cyclic hexadepsipeptides (e.g., ENNs and BEA) were most abundant across all food categories, followed by Di-furanocoumarins (e.g., aflatoxins) and then trichothecene mycotoxins (e.g., DON, NIV, HT-2, etc.).

Because the levels and types of mycotoxins detected varied by commodity, the results are broken down below into food category subgroups for ease of interpretation. It is important to note that among all the mycotoxins detected, only aflatoxins, DON, and fumonisins are regulated in the U.S.

3.1. Snack Foods (n=27):

Eighty-five percent of snack foods were contaminated with at least one mycotoxin. The mean number of mycotoxins per sample was 4, with a maximum of 6. Five out of 27 snack samples were contaminated with either type A or type B trichothecene mycotoxins. The trichothecene mycotoxins detected in these samples included 15-Ace, DON, NIV, T-2 toxins, and HT-2 toxin. The highest NIV level recorded in the study was observed in a snack food at 19,070 μg/kg. Six samples were contaminated with AFB1 (22%), 64% were contaminated with at least one ENN mycotoxin, and 67% with BEA. Thirty-three percent of snack samples were also contaminated with AME. Lastly, FB1, FB2, and CPA were detected together in a single corn-based vegetable stick snack.

3.2. Toddler First Foods (n=7):

In the toddler first food category (n = 7), 6 samples contained at least 2 mycotoxins (86%), with the highest number being 8. Apple juice marketed for infants and toddlers contained the highest levels of the emerging mycotoxins EnnA1, EnnA, and EnnB1 recorded in the study (100.7 ppb, 95.6 ppb, and 82.5 ppb, respectively). In addition, 57% contained the emerging mycotoxin AME, and a fruit puree sample contained AOH and AME.

3.3. Processed Cereal Products (n=25)

Eighty-eight percent of processed, packaged cereals contained at least one mycotoxin. Moreover, packaged cereals were often contaminated with multiple mycotoxins from various mold species, with a mean of 4 and a maximum of 7. In this category, 44% were infected with at least one ENN and 68% with BEA. Aflatoxins were also detected in 20% of cereals, but at very low levels well below the FDA limit of 20 μg/kg. CPA was also present in 16% of samples, all of which were made from corn. Twenty-eight percent of the samples were contaminated with trichothecene mycotoxins. HT-2 toxin, a type A trichothecene, was detected at 45.44 μg/kg in one packaged cereal. Additionally, the highest level of the type B trichothecene DON was found in packaged cereal at 2,499 μg/kg, which exceeded FDA pediatric limits. Only two packaged cereals contained either AOH or AME, with AOH at 6.0 ug/kg and AME at 3.6 μg/kg. AOH, AME, ENNs, BEA, and CPA are currently not monitored in the US or Europe.

3.4. Pasta Samples (n = 13):

The pastas tested were all wheat-based, with 92% of the samples containing at least one mycotoxin. OTA was detected in three of the samples (23%). Seventy-seven percent of samples contained at least one of the ENNs, and 38% of the samples contained BEA. HT-2 toxin, DON, and AFB1 were detected in 3 different samples. Lastly, the Alternaria species mold toxin AME was detected in 62% of samples.

3.5. Raw Ingredients (n=46):

Raw ingredients are often considered minimally processed because they typically undergo fewer food processing steps than ultra-processed foods. A total of 46 ingredients were tested. Among the tested ingredients were wheat flour, corn flour, almond flour, coconut flour, arrowroot and tapioca starch, and chocolate powder.

3.5.1. Grain-Based Ingredients

Wheat flour (n = 10) was contaminated with two or more mycotoxins in 90% of samples, with a maximum of 13 mycotoxins detected in one sample. Furthermore, 30% of samples were contaminated with one or more trichothecene mycotoxins, including DON, NIV, or HT-2. Fifty percent of samples contained with low levels of aflatoxins, and 70% contained at least one ENN and one or more Alternaria species mold toxins. BEA was detected in one sample. Additionally, fumonisins were detected at low levels in 40% of samples, and OTA and ZEA were detected in 1 sample each.

Corn flours were found to be more contaminated than wheat flour. Corn flours (n = 21) were found to be 100% contaminated with three or more mycotoxins. In addition, 100% of the samples were contaminated with AFB1, 71% with AFB2, and 19% with AFG1, with 43% exceeding FDA guidance levels, raising concerns about hepatotoxicity. The highest detected concentration was 273.8 μg/kg for combined AFB1, AFB2, and AFG1 in corn flour, while the median concentrations among samples were 1.3, 1.8, and 2.0 μg/kg, respectively.

The precursor to aflatoxin, STC, was also detected in one corn flour sample. CPA, a mycotoxin that frequently co-occurs with aflatoxin, was detected in 24% of samples, with the highest concentration being 3,804 μg/kg. Additionally, 38% of samples contained at least one type A or type B trichothecene mycotoxin, or both. Fumonisins were detected in 57% of samples, EnnA1 in 71%, and AOH and AME were detected in 19% of samples.

Overall, corn-based flours were the most severely contaminated commodities. Corn flours contained the highest levels of AB1, AFB2, T-2, HT2, ZEA, fumonisins, CPA, DOM, and DON-3-Glucoside found during this survey, and at times, AFB1 and AFB2 levels exceeded FDA and European regulatory limits.

3.5.2. Non-Grain Ingredients:

Almond flour samples (n=4) all contained trace amounts of aflatoxins, but at significantly lower levels than those found in corn flour and within FDA regulatory limits. One sample contained all four ENNs, BEA, and AFB1. Coconut flour (n = 3): One sample was free of mycotoxins, and two other samples were contaminated with all four ENNs and BEA.

Arrowroot and tapioca starches (n = 4) were contaminated in 50% of the samples. One tapioca sample was contaminated with all four ENNs, BEA, and AFB1, and arrowroot was contaminated with low levels of 15-ace (i.e., 100 µg/kg). Two samples were found to be free of mold toxins.

Chocolate powder (n = 4) was also tested due to its popularity as an ingredient. Four samples were tested, and all contained trace amounts of various contaminants. All samples contained at least one mycotoxin, typically trace amounts of AFB1 or AFG1, but were within FDA limits. One sample contained six mycotoxins, including all four ENNs, BEA, and AFB1.

3.6. Comparative Mycotoxin Detection Across Heterogeneous Sample Matrices

In general, ENNs were present in only 26% of samples, and BEA in 33% of non-grain samples. Trace amounts of aflatoxins were detected in 60% of non-grain ingredient samples, but all levels were well below the FDA and European safety guidance levels. In addition, non-grain flours and ingredients were not significantly contaminated with trichothecene mycotoxins compared with grain-based foods. Only one sample contained 100 µg/kg of 15-ace, which is below European pediatric guidance levels, and no other trichothecene mycotoxins, fumonisins, or ZEA were detected in non-grain ingredients. In addition, Alternaria spp. molds AOH and AME were not found in any of the non-grain ingredients. They were only detected in processed grain-based ingredients and ultra-processed foods.

In total, corn-based products had the highest contamination rates of trichothecenes, ZEA, fumonisins, aflatoxins, and cyclopiazonic acid. OTA was most frequently detected in wheat-based foods, with only one chocolate sample containing trace amounts. The ENNs and BEA were detected across all categories, but were most prevalent in processed cereals, snack foods, and apple juice. AOH and AME were most frequently found in wheat flour, pasta, and toddler foods (i.e., with prevalences of 70%, 61%, and 57%, respectively), and are currently not regulated in the US or Europe. The extent to which regulated mycotoxins exceeded European TDI limits at the highest and median detected levels is shown in Table 6.

4. Discussion

Based on this survey's findings, pediatric exposure to mycotoxins through processed foods appears to be common rather than incidental. Through the routine consumption of processed foods, toddlers are likely exposed to a broad spectrum of toxins. Importantly, in this survey, both regulated and emerging mycotoxins often occurred together in the same food products, raising concerns that simultaneous exposures could heighten overall toxicity and health risks, especially in young children [13,122,123]. Furthermore, many share biological effects, such as impairing mitochondrial function, inhibiting protein synthesis, dysregulating immune responses, increasing reactive oxygen species (ROS) and lipid peroxidation, depleting glutathione reserves, and activating cellular death pathways like apoptosis and ferroptosis, potentially further compounding risks associated with co-occurrence [36,42,85,93,124]. While little data is available on the combined effects of co-occurring mycotoxins, multi-mycotoxin effects will be discussed where possible. Otherwise, the major mycotoxins detected, along with their health-associated risks, are discussed separately within the context of TDI limits.

4.1. Aflatoxins and Sterigmatocystin (STC)

Aflatoxins and STC are structurally related polyketide-derived mycotoxins produced by Aspergillus species molds. Aflatoxins can contaminate most major crop plants and are well-established Group 1 hepatocellular human carcinogens [7,31,125]. However, interindividual susceptibility to aflatoxin-induced carcinogenesis is believed to be influenced by host factors, including genetic polymorphisms in DNA repair pathways and co-existing hepatitis infections, which are difficult to ascertain in risk assessments [32,33,98]. Aflatoxins may also contribute to growth stunting in children through inhibition of protein synthesis, among other mechanisms [49].

Increasing evidence also indicates that aflatoxins, particularly AFB₁ and aflatoxin M₁ (AFM₁), adversely affect intestinal and immune health. Animal and human cell line studies have demonstrated that aflatoxins disrupt intestinal epithelial integrity by promoting inflammation, compromising tight junctions, and increasing intestinal permeability [100,101,102,126]. When aflatoxins co-occur with other mycotoxins, such as the trichothecenes, deleterious effects on gut barrier function may be amplified, potentially contributing to immune suppression, abdominal pain, vomiting, and GI distress [100,123,127].

Mechanistically, aflatoxins have also been shown to induce ferroptosis, a regulated form of cell death distinct from apoptosis, necrosis, and autophagy [124]. Ferroptosis is driven by iron-dependent lipid peroxidation and requires intracellular ferrous iron (Fe²⁺), which catalyzes the formation of ROS via Fenton chemistry [124]. This process leads to peroxidation of polyunsaturated fatty acids in cell membranes, cellular and mitochondrial injury, and depletion of intracellular glutathione reserves [124]. Dietary exposure to AFB₁ may therefore promote early cellular death, hepatic injury, and disruption of iron homeostasis at sufficiently high exposure levels. Notably, DON, the ENNs, and OTA, among other mycotoxins, have similarly been shown to induce ferroptosis in mammalian cells, suggesting a shared mechanism of toxicity potentially affecting iron homeostasis among these co-occurring contaminants [92,124]. All were well-represented in these samples.

Regulatory limits for aflatoxins have been established in both the U.S. and Europe. In the U.S., the FDA permits up to 20 μg/kg total aflatoxins in foods, and Europe permits 2-12 μg/kg for AFB1 and 4-15 μg/kg total [128]. The Joint FAO/WHO Expert Committee on Food Additives (JECFA) has evaluated aflatoxins on multiple occasions (e.g., 1997, 2001, 2017) and concluded that a TDI or provisional maximum tolerable daily intake (PMTDI) cannot be established for aflatoxins due to their genotoxic and carcinogenic properties [98]. Nonetheless, a proposed intake range of 0.017–0.082 μg/kg bw/day for total aflatoxins has been suggested based on two earlier animal studies in 1987 and 1989 that evaluated immune system effects in mice and observed significant reductions in white blood cell counts at these levels [32,98]. For risk characterization purposes, the lower bound of 0.017 μg/kg bw/day was considered the more conservative and protective estimate for young children and was applied in the present TDI analysis.

Using the maximum observed contamination level of 273.8 μg/kg combined aflatoxins, a toddler consuming just a 57-g sample of the affected product could ingest 1.25 μg/kg bw/day—nearly 73-fold higher than the proposed limit value of 0.017 μg/kg bw/day. Even at median combined contamination levels of AFB1, AFB2, and AFG1 (e.g., 5.1 μg/kg), which are well below FDA regulatory levels, exposure exceeded the lower reference value by approximately 1.4 times for a small child, and such exposures can accumulate throughout the day. These findings underscore the need for improved surveillance, more robust exposure data, and stronger regulatory oversight to better protect young children from potentially harmful aflatoxin exposure.

STC and aflatoxins share a common xanthone–difuran core structure. STC is a biosynthetic precursor of aflatoxin production. STC has been classified by the International Agency for Research on Cancer as a possible human carcinogen (Group 2B) and is recognized for its immunotoxic properties [104]. Experimental studies have shown that STC induces hepatic and renal toxicity and causes chromosomal damage in both in vitro and in vivo models [104,121]. However, due to limited occurrence and exposure data in humans and animals, the European Food Safety Authority (EFSA) CONTAM Panel has been unable to fully characterize the associated human health risk [121]. In this survey, STC was detected in a single corn flour sample at 5.7 μg/kg, underscoring the need for larger, more representative datasets to support meaningful exposure assessment and risk characterization.

4.2. Trichothecene Mycotoxins

Foodborne trichothecenes are primarily produced by Fusarium molds. Trichothecene mycotoxins possess a tricyclic 12,13-epoxytrichothec-9-ene (EPT) nucleus, which is the principal determinant of toxicity [31]. Multiple trichothecene mycotoxins were detected across the samples analyzed. Globally, trichothecenes most commonly infect wheat, corn, barley, rye, oats, millet, triticale, rice, sorghum, grain-derived alcoholic beverages, soy, tea, and dried spices [76,81]

Trichothecene mycotoxins are well known for their adverse effects in animals and humans, including gastroenteritis, nausea, vomiting, anorexia, impaired immune function, fever, headache, and growth retardation [9,11,12,13,14,16,19,20,21,56,57,129]. Chronic exposure to DON, one of the most common mycotoxins in foods, has been shown to compromise epithelial barrier integrity, disrupt intestinal mucosal immunity, alter gut microbiota homeostasis, impair mitochondrial function, and promote inflammation by increasing ROS and oxidative stress [6,124,130]. Due to their harmful effects on intestinal structures and immune regulation, trichothecenes are believed to play a role in IBD and related GI disorders [6,16,31,123,131,132,133]. DON and other trichothecene mycotoxins, such as T-2 and HT-2 toxins, have also been found to induce ferroptosis in intestinal and other organ tissues in animals and in vitro, potentially altering redox reactions and iron metabolism, which could be particularly detrimental to a developing child [124,134].

Although trichothecenes such as DON and its acetylated derivatives were detected less frequently in this study than in some previous surveys, their presence was nonetheless notable and often occurred alongside other mycotoxins [6]. The highest concentration of DON, a type B trichothecene, was 2,499 μg/kg, which is 5 times the FDA limits and 12 times the European limit for toddlers. Several DON-related metabolites and modified forms—including 15-Ace, 3-Ace, DON-3-Glu, and DOM—were detected in varying amounts. Most have been found to have effects similar to those of the parent compound DON, except for DOM, which may be less toxic [135,136].

To better conceptualize risk for toxicity, a toddler consuming a 57-g serving of a cereal product contaminated at 2,499 μg/kg DON would exceed the TDI by 11.4-fold [75,118]. Even at a median DON level of 300 ug/kg, a single meal will exceed a small child's TDI by approximately 1.4-fold, potentially causing adverse effects.

The type B trichothecene NIV was also detected in three samples, with the highest concentration of 19,070 μg/kg in a toddler snack product and a median level of 7,700 μg/kg found in corn flour. NIV’s mode of action is similar to that of DON; however, the FDA has not set limits for this toxin. While NIV is not routinely monitored in the U.S., the European Union has established a TDI of 1.2 μg/kg bw/day [137]. At the highest detected concentration, a toddler consuming a 57-g serving would ingest approximately 72 times the TDI, while at the median concentration, intake would exceed the TDI by approximately 29-fold. These exposure levels are especially concerning due to their effects on the human GI tract and their co-occurrence with other mycotoxins in the same samples.

Type A trichothecenes T-2 and HT-2 toxins were also identified. These compounds are considered more potent than many type B trichothecenes and have been found to target the GI, epidermal, immune, and hematopoietic systems [119]. These toxins are readily absorbed through the GI tract, skin, and lungs [58]. At high enough exposure levels, T-2 and HT-2 toxins can be lethal, as demonstrated during outbreaks of alimentary toxic aleukia (ATA) in the Orenburg region of the former Union of Soviet Socialist Republics during World War II, where mortality rates approached 50% among children under 10 years of age consuming contaminated grains [5].

T-2 toxin is rapidly metabolized to HT-2 toxin following ingestion, and the two compounds are therefore considered together in risk assessments [119]. Considering the toxicity of these compounds, the EFSA has established a TDI of 0.02 μg/kg bw/day for the combined intake of T-2 and HT-2 toxins [119]. HT-2 toxin was detected in eight samples, and T-2 toxin in two samples. The highest T-2 toxin level detected was 49.6 μg/kg in corn flour, and the highest HT-2 toxin level was 71.2 μg/kg, also found in corn flour. The median concentrations of T-2 and HT-2 were 8.9 μg/kg and 29.9 μg/kg, respectively. Based on a 57-g serving size, the estimated exposure for a toddler would exceed the TDI by approximately 11.3-fold and 16.2-fold at the highest detection levels for T-2 and HT-2 toxins, respectively, and by 2-fold and 6.8-fold at the median detected levels. T-2 and HT-2 toxins co-occurred in one corn flour sample, exceeding the TDI by 27.5-fold, and were detected alongside other mycotoxins in wheat flour, snack, and cereal samples, further raising concerns for child safety.

Despite their documented toxicity, NIV, T-2, and HT-2 toxins, as well as acetylated derivatives and masked forms of DON, are not currently regulated or routinely monitored in foods in the U. S., in contrast to European regulatory frameworks, raising serious concerns for child safety [75,118].

4.3. Fumonisins (FB1, FB2, and FB3)

The fumonisins are a family of water-soluble mycotoxins primarily produced by Fusarium fungi and characterized by a long-chain amino polyol structure (i.e., polyketide). Fumonisins are typically detected alongside other Fusarium-produced mycotoxins in cereal grains, as they were in this study [46].

The International Agency for Research on Cancer has identified FB1 as possibly carcinogenic [9]. Studies have shown that FB1, one of the three fumonisins believed to impact human health, can significantly increase the risk of esophageal cancer in humans [46]. It has also been found to interfere with folic acid and sphingolipid metabolism, potentially increasing the risk of neural tube defects in infants [30].

Eating foods with high concentrations has also been demonstrated to affect intestinal health, leading to abdominal pain, borborygmi, and diarrhea in humans and to increase the number of lymphocytes and monocytes in the ileum and cecum of the intestines [46]. Fumonisins have also been found to increase intestinal permeability (i.e., leaky gut) and the growth of harmful bacteria, such as pathogenic E. coli, in various parts of the intestines (e.g., ileum, cecum, and colon) [47]. Furthermore, fumonisins, along with aflatoxins, AME, and CIT, have also been implicated in disrupting the infant microbiome, potentially increasing the abundance of Klebsiella and Clostridium difficile taxa [138].

Twenty-one samples were positive for FB1, with the highest level of 2000 μg/kg in and a median level of 500 μg/kg in corn flour for FB1, FB2, and FB3 combined. FB1 was the most prevalent of the three. The EFSA established a tolerable TDI for fumonisins in 2018 of 1.0 µg/kg bw/day [117]. Using a 57-g serving to calculate risk showed that a toddler would consume approximately 9-fold the TDI at the maximum and 2.3-fold the median amount, demonstrating a potential risk to child health.

Fumonisins were frequently identified alongside other Fusarium-produced mycotoxins, such as DON, T-2 and HT-2 toxins, and Aspergillus mold toxins, such as aflatoxins and CPA, in this study, raising concerns about synergistic and additive effects of these toxins on multiple organ systems. The Sanitation Hygiene Infant Nutrition Efficacy (SHINE) trial examined the effects of fumonisins, DON, and aflatoxins on enteric dysfunction and growth stunting in infants and children [48]. Thus far, aflatoxins have been evaluated. However, no comprehensive analysis of multi-mycotoxin exposure across all outcomes has been reported.

4.4. Ochratoxin A (OTA)

OTA is a chlorinated, low–molecular–weight mycotoxin with a distinctive isocoumarin–amino acid hybrid structure produced by Aspergillus and Penicillium fungal species [30]. Chronic exposure to OTA has been shown to damage renal tubular cells and impair kidney function, and was historically implicated in Balkan endemic nephropathy, a fatal chronic kidney disease affecting populations in Bulgaria and surrounding regions [9,30]. Subsequent studies have linked OTA exposure to testicular cancer, urinary tract tumors, and potential neurotoxicity, including a proposed contributory role in ASD [8,9,22,23,25,96].

OTA was detected in six samples, with a maximum concentration of 2.5 μg/kg in corn flour and a median concentration of 1.2 μg/kg in wheat-based pasta. The EFSA has established a tolerable weekly intake (TWI) of 0.120 μg/kg bw/week, corresponding to an average daily intake of approximately 0.017 μg/kg bw/day when averaged over a week. Based on this benchmark, the estimated exposure derived from the maximum OTA from a single 57-g serving is approximately 0.0114 μg/kg/bw day, or 67% of the daily-equivalent limit, and 32% of the median level, suggesting that intake from single food items may fall below current safety thresholds [120].

However, these risk estimates do not account for cumulative exposure from multiple foods consumed throughout the day or week. Nor do they consider OTAs’ limited metabolic clearance in humans. OTA is frequently detected in human biomonitoring studies, a finding likely attributable to its prolonged biological half-life and resistance to enzymatic degradation [139]. Compared with the structurally related ochratoxin B, OTA persists longer in the body, in part due to the presence of a chlorine atom at the 5-position of the isocoumarin ring, which increases chemical stability and reduces metabolic breakdown, possibly increasing the risk to child health [139].

Despite its persistence, nephrotoxicity, carcinogenic, and possibly neurotoxic potential, OTA is not regulated in U.S. foods, highlighting a critical gap in food safety oversight for a widely encountered mycotoxin.

4.5. Zearalenone (ZEA)

ZEA is a fungal resorcyclic acid lactone produced by Fusarium fungi whose chemical structure closely mimics endogenous estrogens. ZEA is a potent mycoestrogen with greater estrogenic activity than many other foodborne endocrine disruptors, including genistein, bisphenol A, and phthalates [45]. ZEA has been extensively studied for its adverse reproductive effects in humans and wildlife, and has historically been used in pharmaceutical applications due to its estrogenic properties [2]. Kincade et al. [45] conducted a socioeconomic and dietary predictors US pregnancy cohort study and found that women who ate more processed diets had higher placental levels of hormone-disrupting ZEA.

In addition to its endocrine-disrupting effects, ZEA has been shown to affect GI health. At low doses, it increases intestinal expression of inflammatory cytokines and may suppress tumor suppressor gene activity [43]. In line with this, Abassi et al. [86] found that ZEA appears to promote the growth of human colon cancer cells in vitro. Furthermore, its metabolites, α- and β-zearalenol, have been reported to increase intestinal permeability, potentially enhancing susceptibility to infection, although these metabolites were not detected in the present sample set [44].

Despite its potential adverse effects in humans, ZEA is not routinely monitored or regulated in U.S. foods, highlighting a significant oversight gap for this biologically active mycotoxin that is relevant to pediatric health. However, ZEA is regulated in Europe, and a TDI of 0.25 μg/kg bw/day has been established [45].

ZEA was detected in 9 samples, with the highest concentration observed in corn flour at 126.6 μg/kg. It was also detected in processed cereal products and wheat flour, with a median level of 32.2 μg/kg. Using the TDI as a benchmark, a toddler consuming a 57-g serving of corn flour contaminated at the highest observed level would exceed the TDI by approximately 2.3-fold, while the same serving at the median level in processed cereal would account for approximately 59% of the TDI in a single serving, which could quickly add up due to cumulative exposures throughout the day, again raising concerns about child safety.

4.6. Emerging Mycotoxins

Emerging mycotoxins were among the most frequently detected toxins in this survey. These fungal secondary metabolites are increasingly detected in food and feed but are not yet routinely regulated or comprehensively risk-assessed, despite evidence of biological activity and potential health concerns. As a result, they lack established maximum limits or guidance values in most countries. Furthermore, data on chronic toxicity, particularly in vulnerable populations like children, on metabolites generated, and on the synergistic effects of multiple mycotoxins consumed together, remain sparse. Consequently, TDIs for AOH, AME, BEA, the ENNs, and CPA mycotoxins detected here have not been established.

4.6.1. Alternariol (AOH) and Alternariol Mono-Ethyl Ether (AME)

AOH and AME are structurally related dibenzopyranone mycotoxins—AME being the O-methylated, more lipophilic form of AOH—produced mainly by Alternaria species (spp.) [140]. Among Alternaria spp. mycotoxins, AOH, AME, and tenuazonic acid are the most extensively studied due to their cytotoxic and genotoxic properties, with AOH and AME considered among the most toxic [42,140].

Mechanistic studies show that both toxins exhibit endocrine-disrupting effects, are genotoxic and cytotoxic in vitro, and can induce liver, kidney, intestinal, and splenic toxicity, as well as immune dysfunction, at repeated low µg/kg bw/day exposures [37,41,42,141]. These effects are initiated in part by DNA strand breaks, oxidative stress from increased ROS generation, and inhibition of topoisomerase I and II, enzymes critical for DNA replication and repair [42].

In addition, co-exposure to AOH and AME, frequently observed in this study, has been shown to potentiate cytotoxic effects in human intestinal and hepatic cell lines compared with either compound alone [37]. Both have also been found to increase pro-inflammatory signaling by influencing the NF-κB signaling pathway, and to accumulate in the gut, possibly disrupting gut pH balance and microbiota diversity [42]. Evidence also suggests that these toxins exhibit greater cytotoxic, genotoxic, and estrogenic effects when they occur together or with other mycotoxins, such as DON and ZEA [42].

Occurrence data for AOH and AME in the U.S. are limited. Most analytical surveys have occurred in Europe, Asia, and Africa, but only limited occurrence data are available from North America, primarily from Canada [42,140]. However, based on this survey’s findings, chronic low-level dietary exposure to AOH and AME may be a concern in the U.S., given their potential for toxic effects. Worldwide, AOH and AME have been detected in fruits, vegetables, cereals, oilseeds, and processed products such as juices and tomato-based foods, with the highest prevalence reported in cereal grains and tomato products [42]. In this study, AOH and AME were among the most prevalent contaminants detected in grain-based snacks, pasta, and processed wheat flour, but they were not detected in non-grain flours or chocolate powder.

The EFSA has identified significant data gaps regarding the long-term health effects of AOH and AME and has noted that, despite their frequent co-occurrence with other mycotoxins, these compounds remain unregulated worldwide [140]. Of particular concern is potential exposure among infants and young children, for whom EFSA has proposed a conservative Threshold of Toxicological Concern (TTC) of 0.0025 μg/kg bw/day [140]. Given a 57-g serving at the maximum and median AOH concentrations of 28.4 μg/kg and 6.0 μg/kg, respectively, a toddler would exceed the TTC by 51.8- and 10.9-fold, respectively. Likewise, given the maximum and median AME concentrations of 36.4 μg/kg and 3.4 μg/kg, a toddler would exceed the proposed TTC by 66.4-fold and 6.2-fold, respectively. Collectively, these findings highlight the need for expanded toxicological research, improved exposure surveillance, and enhanced regulatory oversight by U.S. food safety authorities to better characterize risk and protect vulnerable pediatric populations.

4.6.2. Cyclopiazonic Acid (CPA)

CPA is an indole-tetramic acid mycotoxin produced by several Aspergillus and Penicillium species, including Aspergillus flavus, a common foodborne fungus capable of simultaneously producing aflatoxins and CPA [26,27]. Cereal grains, nuts, oil seeds, dried figs, milk, cheese, and processed meats are the foods found to be most contaminated with CPA [26].

Although CPA has been extensively studied in animal models—including rats, pigs, guinea pigs, poultry, and dogs—human toxicological data remain limited. Experimental animal studies indicate that even low-dose exposure can induce inflammation of the GI tract, liver, and kidneys, as well as neurological disturbances [27,28]. Reported toxic effects include severe GI necrosis similar to necrotizing enterocolitis, tremors, and other neurologic manifestations [27]. In addition, CPA has been shown to impair immune function in human cell lines and has been implicated as a possible causative agent in Kodo (Kodua) poisoning outbreaks associated with mold-contaminated millet in northern India [26].

Mechanistically, CPA is of particular concern because it acts as a potent and specific inhibitor of endoplasmic reticulum Ca²⁺-ATPase and exhibits strong metal-chelating activity, thereby disrupting calcium homeostasis and potentially impairing mineral absorption and sufficiency in children [27,142].

Burdock and Flamm [29] and DeWaal [28] discussed establishing a safe CPA TDI limit for humans at the turn of this century. DeWaal proposed that a safe TDI intake limit would be 0.1 μg/kg/bw/day [27,28,29]. However, because of sparse in vitro and animal data on chronic toxicity, a TDI limit was never established. Regardless, using the proposed number may provide general information about the level of toxicity a toddler may experience from its occurrence.

In the present survey, CPA was detected in contaminated corn flour and corn-based products at high and median concentrations of 3804 µg/kg and 134.8 µg/kg, respectively. Using DeWaal’s proposed conservative intake level, a toddler consuming just a 57-g serving would exceed the suggested safety threshold by more than 173-fold at high levels and by 6.2-fold at the median level. Given the documented toxic effects of CPA in animal models, including GI injury and, in some cases, necrotizing enterocolitis-like pathology, this level of exposure is concerning [27,28,29]. These findings underscore the urgent need for additional research to better characterize human health risks associated with CPA exposure, particularly in infants and toddlers, and to inform evidence-based regulatory guidance in the U.S. for this emerging mycotoxin.

4.6.3. Enniatins A, A1, B, B1, (ENNs) and Beauvercin (BEA)

The ENNs and BEA are emerging cyclic hexadepsipeptide mycotoxins produced primarily by Fusarium spp. and Beauveria bassiana [90]. These toxins are more commonly synthesized in milder climates by species such as Fusarium avenaceum and can proliferate during post-harvest transport and under suboptimal storage conditions [143]. ENNs and BEA are lipophilic ionophores that readily integrate into cellular membranes, forming cation-selective pores that disrupt ionic homeostasis [34,36,90,144]. By facilitating the transmembrane movement of mono- and divalent cations, these toxins induce mitochondrial membrane depolarization and dysfunction—effects demonstrated in vitro in human cell lines and in vivo in animal models [34,36,91].

Toxicological studies indicate that ENNs and BEA exhibit a broad range of adverse biological effects, including cytotoxicity, mitochondrial damage, antimicrobial activity, intestinal toxicity, and hepatotoxicity [34,35,65,90,92,93,94,130]. Proteomic analyses of acute EnnB and BEA exposure in rat liver have demonstrated hepatotoxic potential, with altered expression of proteins involved in metabolic regulation, oxidative stress reactions, mitochondrial function, and acetylation pathways, leading to metabolic disturbance [29]. Complementing these findings, Wang et al. [144] reported that exposure to ENNs and BEA in 3D HepaRG human hepatic cell models disrupted glycerophospholipid and sphingolipid metabolism, promoting hepatic lipid accumulation, oxidative stress, and chronic inflammatory signaling. Similar metabolic derangements have been described in the development of liver diseases such as non-alcoholic fatty liver disease (NAFLD), steatohepatitis, and possibly hepatocellular carcinoma [144,145].

Experimental evidence further suggests that ENNs and BEA may induce ferroptosis as a mechanism of liver injury [36,124,144]. Söderdam et al. [92] demonstrated that EnnB and BEA were cytotoxic to salmon hepatocytes at low concentrations and disrupted glutathione metabolism while enriching ferroptosis-associated pathways. Although human exposure data remain limited, these findings warrant further investigation into the potential links between ENNs and BEA exposure, iron dysregulation, liver disease, and anemia, particularly in pediatric populations. In addition, both toxin groups have been associated with microbiome disruption and immunomodulatory effects, and because ENNs and BEA frequently co-occur, combined exposures may produce synergistic toxic effects, potentially increasing health risks in young children [55,90,94,130].

In this study, ENNs and BEA were frequently detected, occurring in 43% and 42% of samples, respectively. Worldwide, ENNs have been reported in grain surveys, with occurrences ranging from 12% to 100% [90]. BEA has been similarly reported in cereal grains, including wheat, rye, oats, barley, and rice, at prevalence rates ranging from 40% to 90% globally [55,90,91].

At present, insufficient toxicological data are available to establish substance-specific TDIs for the ENNs or BEA. Consequently, the EFSA has applied a TTC of 1.5 μg/kg bw/day for ENNs, while cautioning that tolerable exposure levels for BEA may be substantially lower due to its genotoxic potential, and has suggested a TTC of 0.0025 μg/kg bw/day [146]. Notably, the highest levels of ENNs and BEA in this survey were 308.1 μg/kg of combined ENNs in baby apple juice and 190.3 μg/kg of BEA in processed cereal. A 57-g serving of EENs at this level would correspond to approximately 94% of the TTC and a 347-fold exceedance for BEA. Median concentrations across all samples were 41.6 μg/kg for ENNs and 8.3 μg/kg for BEA, corresponding to approximately 13% of the TTC for ENNs and a 15-fold exceedance for BEA. Overall, the concentrations and detection frequencies observed in this study underscore the need for expanded surveillance, refined exposure assessments, and targeted toxicological research in the U.S., particularly to evaluate risks to infants and toddlers [147].

4.7. Public Health and Policy Implications

Mycotoxins have contaminated human food supplies for millennia, and mycotoxin-associated illnesses have been documented across diverse geographic regions [5,9,30,58,59]. In modern globalized food systems characterized by large-scale sourcing, prolonged storage, extensive processing, and long-distance transport, this longstanding problem is increasingly transboundary—particularly in settings with inadequate Hazard Analysis and Critical Control Point (HACCP) controls [42]. Climate change further amplifies these risks, as temperature shifts, drought, and excessive rainfall create conditions that favor fungal infection and mycotoxin production in crops, which may then be distributed across regional and international markets [148,149,150]. Even crops that appear unaffected at harvest may become contaminated during transport, storage, or processing if hygiene and climate controls are insufficient, increasing the likelihood of mycotoxin entry into the food supply [88,148,151].

Many foodborne contaminants—including several emerging mycotoxins such as AOH, AME, CPA, ENNs, and BEA—are not routinely monitored or regulated in many regions (including the U. S.), largely due to limited human toxicological and epidemiological data. As a result, these compounds remain understudied despite growing evidence of their widespread occurrence in commonly consumed foods and their frequent detection in this survey.

Furthermore, contamination may arise from multiple ingredients in a single product, including primary components such as cereal flours and minor additives such as spices or flavorings. This complexity can allow multiple fungal species to contaminate the same food matrix, resulting in co-occurring mycotoxins with potentially additive or synergistic effects. Growing recognition of these cumulative exposures has prompted the Food and Agriculture Organization of the United Nations and the World Health Organization to re-examine global agricultural and food-processing practices related to mycotoxin contamination [11].

Several interrelated agricultural factors are believed to contribute to increasing fungal infections in crops, including climate change, suboptimal crop rotation practices, the heavy use of agrochemicals that upset soil ecosystems, and insect damage, among other agronomic variables [151,152,153,154,155,156,157,158,159]. Additionally, although good agricultural practices remain crucial for prevention, evidence indicates they may become progressively inadequate as climatic conditions evolve. In a 15-year longitudinal study in Croatia and Serbia, Kos et al. [150] demonstrated that shifts in temperature, precipitation, and humidity were associated with increased DON contamination across multiple cereal commodities.

Amidst climate-driven and agronomic pressures that promote fungal growth and mycotoxin formation, chemical control methods have demonstrated limited success. Pesticides and fungicides have not reliably suppressed mycotoxin production and may contribute to human toxicity, promote the emergence of azole-resistant fungal species, and increase CO₂ emissions by disrupting soil ecosystems, thereby further exacerbating climate-related pressures on fungal growth [141,152,160,161,162,163]. Moreover, metal-chelating herbicides such as glyphosate may paradoxically promote the growth of fungal pathogens (e.g., Fusarium spp.) in human food crops by killing off competing microbes and weakening plant immune defenses by making various chelated metals, such as manganese, unavailable for mounting an immune response [155,164,165]. Kremer and Means [157] demonstrated that infection rates for three mycotoxin-producing Fusarium spp. were increased by two- to five-fold in soybeans treated with glyphosate at recommended application rates. Potentially exacerbating this problem, certain fungicides may paradoxically enhance mycotoxin biosynthesis in pathogenic fungi. For instance, strobilurin fungicides have been shown to increase DON concentrations in harvested grain by approximately six to eighteen percent following application [155,157,158,159,164,166,167]. In contrast, biological approaches, including mycotoxin-degrading microorganisms and restoration of soil microbial ecosystems, represent promising alternatives for long-term mitigation [53,127,168,169,170,171]. Adsorbent materials such as clay-based products, activated charcoal, and seaweed-derived biopolymers have also shown some efficacy for protecting animals fed contaminated feed [172].

Beyond crop management, post-harvest handling further influences the risk of mycotoxin contamination. Global trade facilitates the transboundary movement of fungal spores, with the country of origin strongly influencing both fungal species distribution and contamination levels [141,161]. Fungal spores can also persist on agricultural and food-processing equipment and within storage facilities despite routine cleaning, and their germination is strongly influenced by temperature and humidity [88,152]. In large-scale storage systems with limited climate control, maintaining conditions unfavorable to mold growth is often impractical, particularly for fungi that require humidity levels below 20% for suppression [173]. Extended storage periods prior to processing, driven by regional food and food-processing needs, further increase the likelihood that contaminated commodities enter the global food supply [152].

Once present in foods, many mycotoxins are difficult, if not impossible, to eliminate due to their heat-stability and resistance to pH extremes, making prevention the most effective mitigation strategy [76,174,175]. Furthermore, industrial use of preservatives to prevent or mitigate toxigenic fungi and mycotoxins may not be as effective as once thought. While many fungi can be killed by chemical preservatives, their use may induce stress responses that trigger mycotoxin release. Zhelifonova et al. [176] found that a 0.015% sodium nitrate solution did not impair CPA production, and a 0.015% sodium benzoate and 0.02% potassium sorbate solution caused a 1.5-fold increase in CIT, CPA, and mycophenolic acids, and a 1.7-fold and 2.6-fold increase in CPA and mycophenolic acids, respectively. Lee et al. [177] found that the use of preservatives as food additives may pose additional health risks, as they may disrupt the human microbiome and have been linked to the development of IBD in pediatric populations. Accordingly, robust agricultural practices must be complemented by effective post-harvest controls, including storage and transport management and HACCP-based food safety systems that protect the public [42].

While some argue that completely preventing mycotoxin exposure from processed foods might be impossible, especially for vulnerable groups like children, significant reductions in exposure can still be achieved [152,173]. Indeed, promoting the increased consumption of organically farmed, fresh, whole foods harvested locally over processed foods may reduce many chances for children to eat foods contaminated during transit, production, and storage, and could help keep such infections contained within a region [8].

Protecting vulnerable pediatric populations will also likely require more protective regulatory thresholds, inclusion of emerging mycotoxins in monitoring frameworks, and strengthened oversight of food manufacturing and processing practices. Enhanced agricultural surveillance, including oversight of crop rotation, pest management strategies, harvest timing, and on-farm storage, may enable earlier detection and prevent highly contaminated commodities from entering food-processing streams.

Finally, research on foodborne mycotoxins has historically remained largely siloed from clinical and public health practice, limiting translation of scientific findings into prevention and risk-reduction strategies. Consequently, healthcare professionals often receive little formal education regarding dietary mycotoxin exposure or its potential relevance to pediatric health. This gap is increasingly concerning in light of rising rates of chronic pediatric conditions—including IBD, IBS, ASD, and certain liver diseases, among others—whose etiologies remain incompletely understood and may overlap with biological effects observed in chronic mycotoxin exposure. Addressing these issues will therefore require coordinated, cross-sector collaboration among healthcare providers, researchers, agricultural scientists, food producers, and policymakers to develop and implement comprehensive, long-term strategies that reduce exposure and protect child health.

4.8. Limitations and Future Research Implications

This study identified multiple regulated, unregulated, and emerging mycotoxins in foods commonly consumed by U.S. children, suggesting the potential for chronic, low- to high-dose dietary exposure during a sensitive developmental period. The frequent co-occurrence of mycotoxins suggests that toddlers may experience daily combined exposures, with cumulative intake increasing throughout the day. However, interpretation of these findings must account for several limitations within established EFSA risk assessment frameworks [146,178].

First, this cross-sectional survey of commercially available processed and ultra-processed foods is not representative of the broader U.S. food supply. Regional, seasonal, and supply-chain variability—recognized by EFSA as critical determinants of exposure—were not captured, underscoring the need for larger, randomized, nationally representative studies to improve generalizability and the accuracy of risk assessment [178].

Second, although 34 mycotoxins, including masked or modified forms, were quantified by LC-MS/MS, the analytical panel does not cover the full range of fungal secondary metabolites, and additional compounds may contribute to additive exposure.

Third, the toxicological significance of several detected emerging mycotoxins, including AOH, AME, ENNs, BEA, and CPA, remains uncertain because tolerable daily intake values have not been established due to data gaps, which should be addressed in future research [1,140,178,179]. Furthermore, the additive and interactive effects of these and other, more comprehensively studied mycotoxins detected here may be relevant in the context of chronic, possibly synergistic exposures, but evaluating mixture toxicity was beyond the scope of this study, given current limitations.

Fourth, estimated intakes were compared with available TDIs using a mean toddler body weight derived from U.S. growth charts to improve population relevance. However, this approach does not capture inter-individual variability in body weight, dietary patterns, or metabolism, and therefore represents a conservative screening-level assessment rather than a comprehensive probabilistic exposure assessment as defined by EFSA [178].

Future research in the U.S. should focus on examining contamination rates in foods frequently eaten by children, using more nationally representative surveillance, broader analytical coverage, and multistage, stratified random sampling methods to ensure that the results can be generalized to U.S. children at large. Packaged cereals, pasta, pizza, dairy products, and even drinking water should be assessed to gauge the extent of mycotoxin prevalence in these commodities and to provide data to develop a mitigation plan to protect vulnerable youth. Additionally, future research should compare infection rates between sustainably and organically grown foods and industrially farmed foods to better inform mitigation efforts. Emerging interventions, such as the use of food- and human-safe microorganisms capable of degrading mycotoxins in agricultural and processing environments, also warrant further investigation [171].

Finally, because dietary exposure typically involves mixtures, in vitro studies using human cell lines and relevant mycotoxin combinations may help clarify synergistic, additive, or antagonistic effects [36]. Direct exposure studies in children are ethically impermissible; however, mechanistic and dietary intervention studies aimed at reducing mycotoxin intake are feasible. Such studies would monitor clinical, developmental, validated urinary and blood mycotoxin markers, and microbiome-related outcomes. They can also help evaluate individual differences that more accurately reflect real-world conditions and exposures, and clarify their connection to the development of chronic diseases. Collectively, these efforts are essential to inform science-based risk assessment and regulatory decision-making to protect infants and young children from dietary mycotoxin exposure.

5. Conclusions

Foodborne mycotoxins remain underrecognized and understudied in the U. S., despite growing evidence of their widespread presence in the food supply. This survey detected a broad spectrum of regulated and emerging mycotoxins in frequently consumed grains, snack foods, pasta, first foods, and processed ingredients, to which many toddlers may be routinely exposed during early childhood—a critical developmental window—raising concerns about potential adverse health effects. Some children may routinely exceed safe intake levels for multiple mycotoxins simultaneously, yet data on the combined, potentially synergistic effects of these exposures remain limited.

These findings highlight critical gaps in surveillance, toxicological characterization, and risk assessment for pediatric populations. Improved mitigation strategies across agricultural production, storage, and food processing are needed to reduce contamination at its source and throughout the supply chain. Expanded monitoring of foods marketed to infants and young children, refinement of regulatory frameworks, and greater integration of emerging mycotoxins into risk assessment processes are warranted. Addressing these challenges will require coordinated efforts among healthcare providers, scientists, public health professionals, farmers, food manufacturers, and policymakers to develop evidence-based strategies that more effectively protect infants and children from dietary mycotoxin exposure.