Two decades of monitoring acrylamide in breakfast cereals: Impact of market reformulation and compliance with EU regulation

Abstract

This study assesses acrylamide levels in 60 ready-to-eat breakfast cereals marketed in Spain in 2025 and compares them with data from 2006 and 2018 to evaluate trends related to reformulation, processing type, and grain composition. Acrylamide concentrations ranged from <15 to 569 μg/kg, with the highest levels in puffed cereals and products made with wheat and spelt. Median concentrations have decreased notably since 2006, with 95 % of samples now complying with EU benchmark levels. Paired comparisons of cereals from the same brands (2018 vs . 2025) suggest a general reduction in acrylamide content, especially in store-brand products. Estimated dietary exposure was 0.007 μg/kg bw/day, posing no neurotoxic or carcinogenic risk for average consumers, although high consumers may exceed the safety threshold. The results highlight the impact of formulation and processing on acrylamide formation and underscore the importance of continued monitoring and targeted mitigation strategies in cereal manufacturing.

Highlights

• •Acrylamide analyzed in 60 cereals marketed in Spain in 2025.
• •Median acrylamide levels have decreased significantly since 2006.
• •Puffed cereals and wheat-based products showed the highest acrylamide levels.
• •95 % of samples comply with current EU benchmark levels.
• •Acrylamide content in Spanish marketed breakfast cereals dropped 61 % since 2006.

1. Introduction

Breakfast cereals are widely consumed across all age groups due to their convenience, long shelf-life, and appealing sensory and nutritional attributes. They are typically produced from cereal grains such as maize, wheat, rice, or oats, and often contain added sugars, dried fruits, cocoa, and other ingredients that enhance flavour and consumer acceptance (Shrestha et al., 2024). While especially popular among children and adolescents, often representing a regular part of their daily breakfast routine, they are also commonly consumed by adults. Households with children are the most frequent purchasers, reflecting the product's strong popularity among school-aged children in Europe (Bellisle et al., 2018; Cuadrado-Soto et al., 2020; Inchley et al., 2020). Adult consumption, while somewhat lower, remains significant, particularly among those seeking quick and convenient breakfast options (Bellisle et al., 2018). In recent years, the adult consumer segment has driven market demand for cereals with improved nutritional profiles, such as those formulated with alternative grains or pseudocereals, enriched with fibre, protein, or whole grains, and characterized by reduced sugar content or clean-label ingredients. Consequently, the compositional and technological characteristics of breakfast cereals have evolved considerably (Croisier et al., 2021).

Breakfast cereals can be categorised based on the processing technology used, which influences their physical structure, sensory characteristics, and nutritional value. The main types include flaked, extruded, puffed, granola, muesli and coated cereals (Lewis et al., 2021). These manufacturing processes allow for a wide variety of shapes, textures, and densities, but also influence the formation of process contaminants such as acrylamide, due to the different thermal treatments applied during production (Stadler & Gökmen, 2024).

Acrylamide is primarily formed via the Maillard reaction between reducing sugars and the free amino acid asparagine, especially at temperatures above 120 °C under low-moisture conditions (EFSA (European Food Safety Agency), 2015). Its presence in heat-treated foods has raised public health concerns, as acrylamide has been classified as “probably carcinogenic to humans” (Group 2 A) by the International Agency for Research on Cancer (IARC (International Agency for Research on Cancer), 1994).

In response to growing concern, the European Commission adopted Regulation (EU) 2017/2158, which established benchmark levels for acrylamide in various food categories, including breakfast cereals, and requires food business operators to adopt mitigation strategies (EC (European Commission), 2017). In parallel, FoodDrinkEurope (FDE) developed the Acrylamide Toolbox, offering practical guidance to food producers on reducing acrylamide levels through ingredient optimisation and processing adjustments (FDE (FoodDrinkEurope), 2019). These regulatory efforts have coincided with reformulation trends in the cereal industry and public health initiatives aimed at improving the nutritional profile of processed foods. As a result, manufacturers have increasingly adopted strategies to reduce sugar, salt, and acrylamide content, particularly in breakfast cereals targeting children and health-conscious adults (Croisier et al., 2021; MAPA (Ministerio de Agricultura, Pesca y Alimentación), 2025). This aligns with broader public health policies encouraging food reformulation to reduce intake of nutrients of concern, such as sugars, sodium, and saturated fats, particularly in products consumed by children (Santos et al., 2021).

In this context, continuous monitoring of acrylamide levels in breakfast cereals is crucial not only to verify compliance with regulatory thresholds but also to assess the long-term effectiveness of mitigation strategies implemented by the industry (EFSA (European Food Safety Agency), 2015). Previous studies conducted by our research group reported a 60 % reduction in acrylamide concentrations in Spanish breakfast cereals between 2006 and 2018, leading to a measurable decrease in estimated dietary exposure. However, despite this progress, 15 % of the products analyzed in 2018 still exceeded the benchmark levels established by Regulation (EU) 2017/2158 (Mesías et al., 2019), highlighting the need for ongoing assessment of current market conditions.

This study evaluates the acrylamide content in a representative selection of commercially available ready-to-eat breakfast cereals currently marketed in Spain. The results are compared with historical data to assess changes in acrylamide levels over the past two decades (2006–2025), and to identify trends related to evolving product formulations, such as grain type, fibre content, and sugar levels. Compliance with current EU benchmark levels is also assessed. In addition, the study provides updated estimates of acrylamide dietary exposure from breakfast cereals in the Spanish population and applies the Margin of Exposure (MOE) approach to characterize the associated health risk, in line with EFSA recommendations.

2. Material and methods

2.1. Reagents and chemicals

Potassium hexacyanoferrate (II) trihydrate (98 %, Carrez I) and zinc acetate dihydrate (>99 %, Carrez II) were purchased from Sigma (St. Louis, MO, USA). [¹³C₃]-labelled acrylamide (99 % isotopic purity) was obtained from Cambridge Isotope Laboratories (Andover, MA, USA). Formic acid (98 %) was sourced from Panreac (Barcelona, Spain), and deionized water was produced using a Milli-Q Integral 5 purification system (Millipore, Billerica, MA, USA). Reversed-phase Oasis HLB cartridges (30 mg, 1 mL) were acquired from Waters (Milford, MA, USA), and 0.45 μm cellulose syringe filters were obtained from Análisis Vínicos (Tomelloso, Ciudad Real, Spain). All other chemicals, solvents, and reagents used were of analytical grade or HPLC grade.

2.2. Samples

A total of 60 commercially packaged breakfast cereal samples, marketed across Spain and produced by more than 20 manufacturers, were purchased from various supermarkets in February 2025. The selection included products from both prominent commercial brands and widely distributed store-brand lines, reflecting those with substantial market presence in the Spanish breakfast cereal sector. To minimize potential confounding effects, cereals containing chocolate coatings, dried fruits, or added nut were deliberately excluded, and only uncoated, ready-to-eat cereal formats were selected for analysis. Whenever possible, brands assessed in previous longitudinal studies were prioritised to ensure comparability across sampling years. Although all products corresponded to distinct commercial items, brand identities were anonymised to prevent attribution of specific acrylamide levels to manufacturers, thereby ensuring neutrality in the interpretation of results.

Samples were coded and characterized based on the following attributes: processing type (rolled, extruded, granola, flakes, biscuits, sticks and puffed), major cereal (oat, maize, rice, rye, spelt, wheat or mixed), type of grain (refined or wholegrain), gluten-free status (yes/no), without added sugar (yes/no), presence of honey (yes/no), presence of sweeteners (yes/no), and intended consumer population (children / general). Nutritional data, as declared on packaging labels, were recorded and included energy (kcal/100 g), protein (g/100 g), carbohydrates (g/100 g), sugars (g/100 g), fats (g/100 g), saturated fats (g/100 g), fibre (g/100 g), and salt (g/100 g). Furthermore, samples were classified using European Commission criteria for nutritional claims (EC (European Commission), 2006): source of protein (≥ 12 % energy from protein), high protein (≥ 20 % energy from protein), low sugar (< 5 g/100 g), sugar-free (< 0.5 g/100 g), source of fibre (> 3 g/100 g), and high fibre (≥ 6 g/100 g). All samples were homogenized by fine grinding, vacuum-sealed in polyethylene containers, and stored at 4 °C until analysis.

2.3. LC-ESI-MS-MS determination of acrylamide

Sample extraction and clean-up were performed according to Mesías and Morales (2015), with minor modifications. Ground sample (0.500 g) was mixed with 9.4 mL of water in polypropylene centrifuge tubes. The mixture was spiked with 100 μL of a 5 μg/mL [¹³C₃]-acrylamide methanolic solution (internal standard) and homogenized (Ultra Turrax, IKA T10 basic, Germany) for 10 min. Subsequently, 250 μL each of Carrez I (15 g potassium ferrocyanide/100 mL water) and Carrez II (30 g zinc acetate/100 mL water) solutions were added. Samples were centrifuged at 9000 ×g for 10 min at 4 °C. The aqueous phase was purified using Oasis-HLB cartridges (preconditioned with 1 mL methanol and 1 mL water). A 1 mL aliquot of the supernatant was loaded at 2 mL/min; the initial drops were discarded and the remainder collected. The eluate was filtered through a 0. 22 μm membrane into amber LC–MS vials.

Analyses were performed on an Agilent 1200 LC system coupled to an Agilent triple quadrupole mass spectrometer (Agilent Technologies, Palo Alto, CA, USA). Chromatographic separation was achieved using an Inertsil ODS-3 column (250 × 4.6 mm, 5 μm; GL Sciences Inc., Tokyo, Japan) maintained at 30 °C. Isocratic elution was carried out with 0.2 % formic acid in water at a flow rate of 0.4 mL/min, with an injection volume of 5 μL. Detection was conducted in positive electrospray ionization mode, with a spray needle voltage of 1.0 kV, a nebulizing nitrogen flow of 12.0 L/min, and a source temperature of 350 °C.

Quantification was performed in Multiple Reaction Monitoring (MRM) mode, monitoring transitions of m/z 72.1 → 55.1 for acrylamide and m/z 75.1 → 58.1 for the [¹³C₃]-labelled internal standard. Fragmentor voltages were set at 50 V for acrylamide and 76 V for the internal standard, with corresponding collision energies of 11 V and 8 V, respectively. Method accuracy was verified through participation in the Food Analysis Performance Assessment Scheme (FAPAS). The most recent proficiency tests included coffee (test ID 30117), crispbread (30118), and potato crisps (30133), with z-scores of −0.1, 0.1, and − 0.2, respectively, confirming the high reliability of the analytical method. The limit of quantification (LOQ) was 15 μg/kg. All samples were analyzed in duplicate, and results were expressed as μg acrylamide per kg of sample.

2.4. Evolution of acrylamide content in Spanish breakfast cereals (2006–2025)

The acrylamide content in breakfast cereals marketed in Spain and analyzed in the present study was compared with data previously reported by our research group in 2018 (n = 60) (Mesías et al., 2019) and 2006 (n = 60) (Rufián-Henares et al., 2006), in order to assess the temporal evolution of acrylamide levels in this food category. Overall, the mean and 95th percentile values obtained for the set of samples analyzed in different years were compared. More specifically, to assess the evolution of acrylamide levels in cereals from the same commercial brand between 2018 and 2025, a Wilcoxon signed-rank test was conducted. This analysis included 34 paired samples, selected from a total of over 60 collected observations.

2.5. Evaluation of acrylamide exposure from breakfast cereals

Estimated dietary exposure to acrylamide from breakfast cereals was calculated by combining the average acrylamide concentrations obtained in the current study with per capita consumption data (1.51 kg/person/year) reported by the Spanish Ministry of Agriculture, Food and Environment (MAPA (Ministerio de Agricultura, Pesca y Alimentación), 2025). An average body weight (bw) of 70 kg was assumed for intake estimation, with results expressed as μg/kg body weight/day (EFSA (European Food Safety Authority), 2012).

Health risk associated with acrylamide intake was assessed using the Margin of Exposure (MOE) approach, which is the preferred method for substances that are both genotoxic and carcinogenic (EFSA, 2015). The MOE is calculated by dividing the benchmark dose lower confidence limit (BMDL₁₀) by the estimated dietary exposure. In this study, the BMDL₁₀ values recommended by EFSA were applied: 170 μg/kg bw/day for neoplastic effects (carcinogenicity) and 430 μg/kg bw/day for non-neoplastic effects (specifically neurotoxicity).

2.6. Statistical analysis

Given the non-normal distribution of acrylamide concentration data and heterogeneity in group sizes and variances, all statistical analyses were performed using non-parametric methods. Group-wise comparisons of acrylamide levels by attributes with two variables were conducted using the Kruskal–Wallis H test, followed by post hoc pairwise analyses using Dunn's test with Bonferroni-adjusted significance thresholds (α = 0.05) to control for Type I error inflation. Group-wise comparisons of acrylamide levels by attributes with more than two variables were conducted using the Mann–Whitney U test. Median and interquartile range values were reported as measures of central tendency, and effect sizes were calculated using rank epsilon squared (ε²), classified according to standard thresholds for small, moderate, and large effects. To evaluate compliance with benchmark levels within regulatory classification groups, one-sample Wilcoxon signed-rank tests were applied to assess whether acrylamide concentrations fell below regulatory benchmark levels and the longitudinal changes in acrylamide content between matched breakfast cereal products from 2018 and 2025. All statistical procedures were conducted using SPSS version 29 (SPSS, Chicago, IL) and the R software (version 4.3.2), with implementation via the following packages: rstatix (for rank-based tests and effect size estimation), exactRankTests (for exact non-parametric inference), stats (base functions), and ggstatsplot, ggplot2, and ggpubr (for graphical representation of results).

3. Results and discussion

3.1. Acrylamide in breakfast cereals: Current levels and influencing factors

Acrylamide concentrations in the analyzed breakfast cereal samples ranged from below the limit of quantification (LOQ, 15 μg/kg) to 569 μg/kg, with a mean value of 114 μg/kg (CI95 %: 87–140 μg/kg) and a median of 83 μg/kg (Fig. 1). Two samples presented acrylamide levels below the LOQ. One sample (337 μg/kg) was classified as an outlier, and another (569 μg/kg) as an extreme value. Overall, the acrylamide concentrations observed were within the range reported in recent European studies, such as those conducted in Germany (Lipinski et al., 2024), which documented levels between < 23 μg/kg and 556 μg/kg in commercially available breakfast cereals. Additional studies from other countries have shown comparable variability in acrylamide levels: 41–362 μg/kg (Basaran & Sadighara, 2024), < 27–643 μg/kg (Anwarul Hasan et al., 2022), 33–744 μg/kg (Nica-Badea, 2022), and 82–486 μg/kg (Merhi et al., 2020).

Fig. 1.

Acrylamide content in breakfast cereals grouped by (a) the type of breakfast cereal and (b) the major cereal.

Acrylamide formation in cereal products is primarily driven by factors related to formulation and processing. Key determinants include the type of cereal grain used, which affects the natural content of asparagine and reducing sugars, as well as added ingredients such as sugars or honey. Thermal processing parameters, especially baking, extrusion, toasting, and puffing, play a major role due to their direct impact on Maillard reaction pathways (FDE (FoodDrinkEurope), 2019; Stadler & Gökmen, 2024).

In a first approximation, samples were grouped into seven format categories based on their physical structure and processing method: rolled cereals (n = 8), extruded cereals (n = 13), granola (n = 3), flakes (n = 19), biscuits (n = 4), sticks (n = 4), and puffed cereals (n = 9) (Fig. 1a). Acrylamide levels varied markedly across these groups. Rolled cereals exhibited the lowest concentrations (LOQ–58 μg/kg; mean: 31 μg/kg), followed by extruded cereals (24–186 μg/kg; mean: 60 μg/kg) and granola (32–118 μg/kg; mean: 78 μg/kg). Intermediate to high values were found in flakes (33–255 μg/kg; mean: 104 μg/kg), biscuits (57–182 μg/kg; mean: 105 μg/kg), and sticks (23–226 μg/kg; mean: 151 μg/kg). Puffed cereals showed the highest acrylamide levels of all categories (122–569 μg/kg; mean: 284 μg/kg). The diversity in acrylamide levels across cereal types can be explained by their underlying processing technologies and thermal intensity (Croisier et al., 2021; Lipinski et al., 2024). The type and purpose of processing technologies for ready-to-eat breakfast cereals are extensively described by Ng (2023). Rolled cereals are produced by steaming whole grains followed by mechanical flattening. The process involves moderate temperatures and minimal Maillard reaction, resulting in low acrylamide levels. Extruded cereals are formed by forcing grain-based dough through high-pressure, high-temperature conditions (typically 120–180 °C), leading to partial gelatinization and significant browning reactions. Granola is baked at moderate temperatures (130–160 °C), usually with added oils and sweeteners, which can promote acrylamide formation. Flakes undergo several stages, including cooking, drying, rolling, and toasting. The final toasting step can reach high surface temperatures and contributes to acrylamide variability within this group. Biscuits are typically rich in sugar and fat and are baked at 180–200 °C. The combination of composition and temperature explains their moderately high acrylamide levels. Sticks are often extruded and toasted or baked, and their elongated shape with increased surface area can enhance thermal reactivity. Puffed cereals are produced via oven puffing or gun puffing. Oven puffing involves rapid heating at atmospheric pressure to vaporize internal moisture. Gun puffing cereal grains to a sudden pressure drop after being superheated, causing explosive expansion. Both methods apply short (30–90 s) but intense heat (200–300 °C), which strongly favors acrylamide formation (Konkubaeva et al., 2025; Ng, 2023).

A Kruskal-Wallis test for independent samples revealed a significant (p < 0.0001) and large effect size (ε² = 0.385) of cereal shape on the acrylamide levels (Fig. 1a). Subsequent post-hoc pairwise comparisons using Bonferroni correction due to the heterogeneity of the sample groups identified statistically significant differences between puffed cereals and both rolled and extruded cereals (p-adjusted < 0.05). No other pairwise comparisons were significant after correction for multiple testing, indicating that puffed cereals stand out with distinct acrylamide profiles. These results are consistent with previous studies (e.g., Lipinski et al., 2024), reinforcing the link between processing conditions and acrylamide formation in breakfast cereal matrices.

In addition to processing conditions, the type of cereal used in the formulation also plays a critical role in acrylamide formation, due to the natural presence of its main precursors: asparagine and reducing sugars. Asparagine content has been identified as the limiting precursor in cereal-based matrices. Cereals such as rice tend to have low levels of asparagine and reducing sugars, and are therefore expected to yield lower acrylamide formation. Rye, wheat, and spelt typically contain higher concentrations, leading to greater acrylamide potential under similar processing conditions (Žilić et al., 2020; Stockmann et al., 2018, Mesias & Morales, 2025). In this study, acrylamide content varied significantly by grain type (Fig. 1b). Contrary to expectations, the mean acrylamide concentration in rice-based samples was relatively high (120 μg/kg). This group included puffed, extruded, and flaked products, underscoring the substantial influence of processing intensity and formulation beyond the raw material itself. Maize-based cereals exhibited the lowest average acrylamide levels (54 μg/kg), with extruded corn cereals showing values around 39 μg/kg, and corn flakes reaching 69 μg/kg. Oat-based cereals presented intermediate values, averaging 71 μg/kg, although concentrations varied widely depending on processing, from non-detectable levels in rolled oats to 187 μg/kg in extruded oat products. The single rye-based sample showed a relatively low acrylamide level (41 μg/kg); however, as this product was a rolled cereal, likely subjected to mild thermal treatment, it cannot be considered representative of rye-based products in general. Among the cereal types, spelt and wheat exhibited the highest acrylamide concentrations. Spelt-based cereals averaged 140 μg/kg, with values ranging from 31 μg/kg in rolled products to 337 μg/kg in puffed formats. Wheat-based cereals had the highest overall mean (202 μg/kg), with a range extending from 33 μg/kg in biscuits to 569 μg/kg in puffed products. These differences were statistically significant and reflect both the higher asparagine content of wheat and the processing techniques used. Finally, cereals formulated with mixed grains showed intermediate values, ranging from non-detectable acrylamide levels in rolled products to 182 μg/kg in biscuits.

The dataset evidenced distinct patterns in the use of cereal grains across breakfast cereal formats. Maize appears to be predominantly utilized in the production of flakes (n = 9) and extruded cereals (n = 6), indicating its suitability for both flaking and extrusion technologies. This observation is supported by global industry data, which identifies maize as the most widely employed grain in breakfast cereal manufacturing, accounting for approximately 36.7 % of ingredient usage in 2024 (Mordor Intelligence, 2025). The technological adaptability of maize is further supported by its low fibre content and favorable starch gelatinization properties, which make it especially compatible with extrusion-based processes. Wheat, by contrast, is represented more broadly across formats, with presence in flakes (n = 5), biscuits (n = 2), sticks (n = 4), and puffed cereals (n = 7), reflecting its versatility in high-temperature processing. Puffed cereals are commonly produced from wheat, rice, or spelt due to their ability to expand under extreme thermal conditions. These formats are typically manufactured through extrusion and rapid expansion techniques. Oats are primarily associated with rolled cereals (n = 3), granola (n = 2), and to a lesser extent with extruded and flaked formats, consistent with their use in minimally processed or baked products. Rolled cereal formats are typically produced from oats, rye, or wheat, given these grains' structural tolerance to steaming and flattening without fragmentation. Spelt and rice, though less frequent, are each linked to select formats, spelt with puffed and flaked cereals, and rice with extruded and flaked types. Rye was observed exclusively in the rolled cereal category (n = 1), suggesting limited integration into industrial processing variants.

The use of whole grains versus refined grains can significantly influence acrylamide formation in cereal products. Whole grains generally contain higher levels of free asparagine and reducing sugars, particularly in the bran and germ fractions, which are largely removed during refining. Consequently, products made with whole grains are typically expected to exhibit higher acrylamide levels after thermal processing compared to those made with refined flours (Žilić et al., 2020). However, this trend was not observed in the present study, as whole grain samples did not show significantly higher acrylamide concentrations (Table 1). Similarly, no significant differences were found when samples were grouped by fibre or protein content according to the nutritional profile declared by the manufacturers of breakfast cereals (Table 2). Statistically significant differences were only observed in the “source of fiber” and “gluten-free” categories, although the limited number of samples in these groups may have affected the robustness of the results (Table 1). Neither the presence of honey nor the intended target consumer group had a significant effect on acrylamide levels. However, a clear association was observed with sugar content: samples with higher sugar levels tended to exhibit higher acrylamide concentrations. In fact, a positive and significant correlation was found between acrylamide and sugar content in the breakfast cereals (p = 0.014). This finding aligns with the established role of reducing sugars as key precursors in acrylamide formation during thermal processing (Stadler & Gökmen, 2024). The classification of high protein content, sugar-free, and the presence of sweeteners was not considered due to the low number of samples in these groups.

Table 1.

Average acrylamide content (μg/kg), minimum, maximum and median in cereals samples grouped according to different factors.

Factor	n	Mean ± SD	Minimum	Maximum	Median
Type of grain
Refined	30	125 ± 127a	24	569	55
Wholegrain	30	102 ± 70a	< LOQ	255	89
Source of protein
< 12 % total energy	46	112 ± 106a	< LOQ	569	84
> 12 % total energy	14	118 ± 94a	23	337	97
Low sugar content
< 5 %	20	73 ± 74a	< LOQ	337	47
> 5 %	40	134 ± 109b	23	569	114
Without added sugar
Yes	18	76 ± 83a	< LOQ	337	40
No	42	129 ± 106b	23	569	101
Source of fibre
< 3 %	5	49 ± 34a	24	107	32
> 3 %	55	119 ± 105b	< LOQ	569	88
High fibre content
< 6 %	26	109 ± 24a	24	569	51
> 6 %	34	117 ± 15a	< LOQ	337	104
Gluten free
Yes	55	44 ± 33a	< LOQ	95	33
No	5	120 ± 104b	< LOQ	569	88
Presence of honey
Yes	13	151 ± 159a	31	569	88
No	47	103 ± 79a	< LOQ	337	82
Target consumer
Children	15	152 ± 156a	24	569	56
General population	45	101 ± 75a	< LOQ	337	85

Analyses were performed in duplicate. Data are means ± S.D. Different letters within a factor indicate statistically significant differences (p < 0.05). LOQ: Limit of quantitation (15 μg/kg).

Table 2.

Nutritional profile of breakfast cereals (per 100 g), as declared by the manufacturer.

	Mean ± SD	Minimum	Maximum
Energy (kcal)	375 ± 24	296	429
Proteins (g)	9.0 ± 2.8	4.5	16.0
Carbohydrates (g)	73 ± 12	35	88.8
Sugars (g)	13 ± 12	0.3	40.9
Fats (g)	3.3 ± 2.9	0.6	13.3
Saturated fats (g)	0.6 ± 0.4	0.1	2.2
Fibre (g)	8.5 ± 7.4	0.1	40.0
Salt (g)	0.6 ± 0.5	0.0	1.8

3.2. Evolution of breakfast cereal formulation and acrylamide content

Breakfast cereal consumption contributes significantly to the intake of essential nutrients such as vitamins and minerals in both children and adults, largely due to the widespread practice of fortification. However, despite their nutritional benefits, breakfast cereals can also contain considerable amounts of added sugars and salt, components associated with adverse health effects, including dental caries, obesity, type 2 diabetes, and cardiovascular diseases (Croisier et al., 2021). In response to increasing public health concerns, global initiatives aligned with World Health Organization (WHO) recommendations have been launched to improve food environments, particularly those shaping children's dietary habits. Within this framework, the European Commission has promoted innovation, product reformulation, and healthier marketing practices to support the development of improved food options with lower sodium, sugar, and saturated fatty acid content, while also highlighting dietary fibre and whole grains in cereal-based products (Santos et al., 2021).

This initiative is reflected in the current trend of the Spanish market. According to the latest Spanish Food Consumption Report (2024) (Mercasa, 2025), breakfast cereals targeted at adults have become the leading market segment, accounting for nearly 60 % of total cereal sales. This shift reflects evolving consumer preferences, with increasing demand for gluten-free, whole grain, sugar-free, and high-fibre products. This trend is also evident in the findings of the present study. Compared to previous assessments conducted by our research group, breakfast cereals marketed on the Spanish market exhibit lower levels of sugar (13.0 g/100 g in 2025 vs. 15.8 g/100 g in 2018) and saturated fat (0.6 g/100 g in 2025 vs. 0.8 g/100 g in 2018) (Table 2). In addition, a higher prevalence of products with elevated fibre content has been observed. In 2025, 67 % of the analyzed cereals contained more than 5 % fibre, compared to 57 % in 2018 (Mesías et al., 2019) and only 22 % in 2006 (Rufián-Henares et al., 2006). Furthermore, 50 % of the breakfast cereals sampled were formulated with whole grains. These results are consistent with those reported by Croisier et al. (2021), who observed an increase in the proportion of whole grain breakfast cereals containing ≥8 g of whole grain per serving, from 67 % to 74 %, between 2013 and 2020 in products marketed in Sydney.

Along with changes in nutritional composition, the most significant aspect of the current market is the profile of the main grains included in breakfast cereals. While in 2006 breakfast cereals were formulated only with rice, corn, and wheat (Rufián-Henares et al., 2006), the current samples include other grains such as spelt, oats, and rye, a trend that was already widespread by 2018 (Mesías et al., 2019). This diverse and expanding consumption pattern underscores the importance of continuous monitoring of breakfast cereals, evaluating them not only from a nutritional standpoint but also regarding the presence of processing contaminants such as acrylamide.

3.3. Benchmark compliance of breakfast cereals according to EU regulation

According to Commission Regulation (EU) 2017/2158, breakfast cereals are classified into three groups based on the major cereal used in their formulation. A benchmark level of 300 μg/kg has been established for breakfast cereals containing bran, whole grains, and gun-puffed grains (Group 1), as well as those formulated with wheat and rye (Group 2). Breakfast cereals primarily formulated with maize, oats, spelt, barley, or rice fall under Group 3, with a benchmark level of 150 μg/kg (EC, 2017). In the present study, samples were classified accordingly (Fig. 2): 39 products were assigned to Group 1 and 21 to Group 3, with no samples meeting the criteria for Group 2. Within Group 1, samples were further subdivided into whole grain cereals (G1-wholegrain, n = 30) and puffed cereals (G1-puffed, n = 9). When assessed on a sample-by-sample basis, all products met the benchmark thresholds, with the exception of three puffed cereal samples from Group 1. The observed compliance with EU benchmark levels for acrylamide in breakfast cereals can be attributed to several key factors related to both raw material composition and processing conditions (FDE, 2019). Cereals classified in Group 1, which contain bran, whole grains, and rye or wheat, tend to have higher levels of free asparagine and dietary fibre, both of which can promote acrylamide formation. This explains the generally higher benchmark threshold established for this group (300 μg/kg) compared to Group 3 cereals (150 μg/kg) (EC, 2017).

Fig. 2.

Acrylamide content in breakfast cereals (BC-all) and grouped according to the classification indicated in the Regulation (EU) 2017/2158 for Group 1 (BC-G1), Group 3 (BC-G3), and Group 1 divided in wholegrain (BC-G1-wholegrain) and puffed (BC-G1-puffed).

To assess compliance more rigorously, one-sample Wilcoxon signed-rank tests were conducted to compare acrylamide concentrations in each regulatory group against their respective benchmark levels. In Group 1 (benchmark threshold = 300 μg/kg), acrylamide levels were significantly lower than the reference value (p < 0.001; n = 39), with a large effect size (r = 0.784; Supplementary Fig. 1a). Similarly, Group 3 products (benchmark threshold = 150 μg/kg) exhibited acrylamide concentrations significantly below the regulatory threshold (p < 0.001; n = 21), also with a large effect size (r = 0.877; Supplementary Fig. 1b). These results indicate strong compliance with EU safety standards and underscore the effectiveness of current mitigation strategies in reducing acrylamide content across both benchmark groups.

Compared to previous years in the Spanish market, a progressive decrease in acrylamide levels has been observed: from a mean of 292 μg/kg in 2006 (Rufián-Henares et al., 2006), to 118 μg/kg in 2019 (Mesías et al., 2019), and 114 μg/kg in 2025 (Fig. 3). Moreover, the 95th percentile has decreased from 515 μg/kg in 2006 to 331 μg/kg in 2019 and 262 μg/kg in 2025. It is important to highlight that although the average values for 2019 and 2025 are similar, the current data set shows a narrower range of variability, with values ranging from <15 μg/kg to 337 μg/kg, compared to a wider range of 27–482 μg/kg in 2019. While 85 % of breakfast cereals sampled in 2019 complied with the respective benchmark levels, the present study shows an improvement, with 95 % of samples falling below the established regulatory thresholds, indicating increased compliance with European safety standards. Similar results have been reported for breakfast cereals marketed in Turkey, where only 15 % of the products exceeded the benchmark set by the European Commission. (Basaran & Sadighara, 2024).

Fig. 3.

Overall mean acrylamide levels in breakfast cereal samples from 2006 to 2025, including standard deviations and trends in the 95th percentile. Different letters indicate significant differences between years (p < 0.05).

Mitigation strategies implemented in the cereal industry, including the use of asparaginase enzymes to reduce asparagine content in raw materials, optimization of baking temperatures and times, and selection of cereal varieties with lower precursor levels, have been effective in decreasing acrylamide concentrations. Moreover, differences in cereal matrix composition affect heat and mass transfer during processing; for example, Group 3 cereals often have higher moisture content and different starch structures, which can limit acrylamide formation by altering reaction kinetics. The progressive decline in acrylamide levels observed from 2006 to 2025 reflects the combined effect of these mitigation measures alongside regulatory pressures encouraging continuous improvement. Notably, the narrower variability range in recent data suggests greater control and consistency in processing conditions. These mechanistic insights not only reinforce the regulatory compliance demonstrated by our results but also highlight the importance of tailored strategies depending on cereal type and formulation to effectively manage acrylamide risk.

3.4. Paired sample-based comparison of acrylamide content in recurrent commercial cereal products

To assess changes in acrylamide content between 2018 and 2025 in breakfast cereals from the same commercial brand, a Wilcoxon Signed-Rank Test was performed. The test evaluates whether the median difference between paired observations significantly differs from zero, making it particularly suitable for matched samples. It is important to note that samples were not strictly matched by lot, formulation, or production code, but were instead considered “paired” based on continuity in brand identity and product label across both years. Thus, 34 paired samples, selected from a total of over 60 collected observations, were included in the analysis. The paired sample size (n = 34) was adequate for statistical inference. However, given the presence of outliers in both the 2018 and 2025 datasets, a more conservative and robust rank-based method was adopted. This approach is less sensitive to extreme values and well-suited for ordinal comparisons of paired differences.

As illustrated in Fig. 4, the median acrylamide content in breakfast cereals was 83 μg/kg in 2018 and 86 μg/kg in 2025. Although this overall shift is not statistically significant, it is noteworthy that cereals with higher acrylamide levels in 2018 tended to show substantial reductions in 2025. In contrast, cereals with initially lower acrylamide levels exhibited modest changes. The Wilcoxon Signed-Rank Test reported a test statistic of 336.00, a p-value of 0.521, and an effect size (r) of 0.11, indicating a small effect and confirming that the mean values show an overall reduction although there is no statistically significant change in the median acrylamide concentration between 2018 and 2025 for breakfast cereals.

Fig. 4.

Paired comparison of acrylamide content in breakfast cereal samples from 2018 and 2025 using the Wilcoxon Signed-Rank Test (n = 34).

A more detailed analysis of the results revealed that, among the 34 paired observations, 15 exhibited positive ranks, indicating a net increase in acrylamide content from 2018 to 2025, while 19 showed negative ranks, suggesting a reduction (Supplementary Fig. 2a). The sum of positive ranks was +721, whereas the sum of negative ranks was −1073, reflecting a stronger overall tendency toward decreased acrylamide levels across the dataset. Although this directional imbalance suggests a downward trend, the effect was not statistically significant, as indicated by the overall p-value. The net positive differences in acrylamide content ranged from +8 to +107 μg/kg, with a mean of +48 μg/kg, while the negative differences ranged from −182 to −2 μg/kg, with a mean of −56 μg/kg. Interestingly, the five brands with the highest acrylamide concentrations in 2018 experienced the greatest reductions by 2025. These particular samples had previously exceeded the benchmark threshold of 300 μg/kg, the highest reference value established for this food category. In contrast, three products showed acrylamide increases of approximately 100 μg/kg, which appeared to be associated with the incorporation of higher fibre content in their reformulated versions. Nevertheless, their final acrylamide concentrations remained below the benchmark threshold.

An evaluation of relative differences in acrylamide content between 2025 and 2018 for each sample provided additional insights (Supplementary Fig. 2b). Among the positive rank subgroup (n = 15), relative increases ranged from 11.28 % to 73.47 %, with a mean increase of 38.40 % compared to the original acrylamide levels. In the negative rank subgroup (n = 19), relative decreases ranged from −291.49 % to −4.33 %, with a mean reduction of −75.96 %. Overall, the paired samples exhibited a mean relative reduction of −25.5 % in acrylamide content in 2025 compared to 2018. Although these changes were not statistically significant, the observed downward trend, particularly among products with initially high acrylamide concentrations, may reflect again the impact of mitigation strategies or modifications in processing practices implemented in recent years.

Further analysis of the dataset revealed that, among the 34 matched commercial cereal samples, 22 (64.7 %) were non–store-brand products, while 12 (35.3 %) corresponded to store-brand cereals. Notably, several store-brand products exhibited substantial reductions in acrylamide content, often exceeding 50 %, particularly when initial concentrations were high in 2018. This pattern may reflect proactive reformulation efforts by retailers to comply with benchmark threshold or address growing consumer health concerns. Statistical comparisons between the two groups supported this trend: store-brand cereals showed a mean absolute reduction of −31.4 μg/kg and a mean relative reduction of −68.5 %, while non–store-brand products exhibited a smaller mean absolute change of −4.0 μg/kg and an average relative change of +1.2 %. These differences in relative acrylamide reduction were statistically significant according to the Mann–Whitney U test (U = 70.0, p = 0.0113), suggesting that store-brand products may have implemented more effective or recent mitigation strategies during the 2018–2025 period. This trend may be partly explained by differences in reformulation timelines. Large, well-established commercial brands likely adopted acrylamide mitigation strategies prior to 2018, driven by greater technological capacity and reputational concerns, resulting in relatively low acrylamide levels by that time. In contrast, store-brand products, often produced by third-party manufacturers, may have implemented such strategies more recently, aligning with evolving regulatory guidance or retail-driven quality standards (FDE, 2019). This hypothesis is supported by the fact that many non–store-brand samples exhibited relatively low initial acrylamide levels in 2018, possibly indicating that mitigation efforts had been deployed earlier in response to previous regulatory pressure or internal quality standards.

3.5. Evaluation of acrylamide exposure from Spanish commercialised breakfast cereals

Exposure to acrylamide from breakfast cereals in the Spanish population was estimated using the total per capita consumption of this food category reported by the Spanish Ministry of Agriculture, Fishery and Food (1.51 kg/person/year) (MAPA, 2025) alongside the acrylamide content measured in the present samples. Considering the minimum and maximum acrylamide concentrations observed across all samples, estimated exposure ranged from <0.06 to 2.35 μg/person/day, with a mean value of 0.47 μg/person/day. These results are slightly lower than previous estimates from 2019, where the average contribution of breakfast cereals to dietary acrylamide exposure in Spain was calculated as 0.54 μg/person/day (range: 0.05–2.94 μg/person/day) (Mesías et al., 2019).

Assuming an average body weight (bw) of 70 kg (EFSA, 2012), the mean daily acrylamide intake from breakfast cereals for the general population was estimated at 0.007 μg/kg bw, with a range of 0.001 to 0.033 μg/kg bw. In comparison, higher exposure levels have been reported for breakfast cereal consumption in the Turkish population, ranging from 0.03 to 0.20 μg/kg bw/day (Basaran & Sadighara, 2024).

Risk characterization for acrylamide in breakfast cereals was conducted taking MOE values of 125 and 10,000 as values indicating no concern for neurotoxicity and carcinogenicity in people, respectively. MOE was calculated as the BMDL value divided by respective total acrylamide exposure; 430 μg/kg bw/day was considered as the BMDL₁₀ value for neurotoxicity, and 170 μg/kg bw/day as the value for carcinogenicity, as dictated in EFSA opinion on acrylamide (EFSA, 2005). In this respect, mean MOE value of 64,179 was obtained for neurotoxicity (range: 53,750–13,030 for minimum and maximum exposure), indicating no health concern. When comparing with the BMDL₁₀ for carcinogenicity, a mean value of 25,373 was calculated (range: 212,500–5151 for minimum and maximum exposure). In this case, maximum exposure involves a MOE below the safety limit of 10,000, suggesting that it should be considered from a public health point of view. These findings underscore the critical importance of controlling both the formulation and, especially, the thermal processing conditions of breakfast cereals to prevent the formation of high acrylamide levels. Effective mitigation strategies focused on ingredient selection and optimized processing are essential to minimize acrylamide content and thus reduce potential health risks associated with chronic exposure to this contaminant. Ensuring such control is vital for safeguarding consumer health while maintaining the nutritional quality of breakfast cereals.

4. Conclusions

This study provides a comprehensive evaluation of acrylamide levels in breakfast cereals marketed in Spain over the last two decades, demonstrating a significant overall reduction in acrylamide concentrations, with 95 % of samples from 2025 complying with the benchmark levels established by European regulations. The data confirm that cereal type and processing conditions critically influence acrylamide formation, as evidenced by the higher concentrations observed in puffed cereals and those formulated with wheat and spelt. These results highlight the need to consider both raw material composition and thermal processing intensity when developing mitigation strategies.

While reformulation efforts aimed at improving nutritional profiles, such as increasing fibre content, reducing sugars and saturated fats, and incorporating more whole grains, were observed, their impact on acrylamide levels was inconsistent, suggesting that nutritional improvements alone may not directly translate to lower acrylamide formation. The paired analysis of products from 2018 and 2025 further confirms a general downward trend in acrylamide, especially notable in private-label brands, indicating that targeted industry initiatives can effectively reduce contaminant levels. Dietary exposure assessments suggest low risk for the general population. However, habitual consumers of breakfast cereals could approach exposure levels associated with carcinogenic risk, underscoring the importance of continued vigilance.

Future research should focus on elucidating the mechanistic pathways governing acrylamide formation in different cereal matrices and processing conditions to refine mitigation approaches further. Additionally, exploring consumer behaviour and exposure patterns will be essential to accurately assess public health implications. Ongoing monitoring and regulatory review are recommended to ensure continued compliance and protect consumer safety, alongside promoting industry innovation in ingredient selection and processing technologies.

CRediT authorship contribution statement

Marta Mesías: Writing – original draft, Validation, Supervision, Software, Resources, Project administration, Investigation, Funding acquisition, Data curation, Conceptualization. Alicia García: Formal analysis. Francisco J. Morales: Writing – review & editing, Validation, Supervision, Software, Resources, Project administration, Investigation, Funding acquisition, Data curation, Conceptualization.

Funding

This research was funded by project ACRYREVAL (PID2022-137697NB-I00), funded by MCIN/AEI/10.13039/501100011033 and by “ERDF A way of making Europe”. This work is related to COST Action ACRYRED, CA21149, supported by COST (European Cooperation in Science and Technology).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary material

Data availability

Data will be made available on request.