Three groundnut cultivars (SAMNUT-23, SAMNUT-24, and SAMNUT-26) were purchased from Sokoto Central Market, Sokoto State, Nigeria, on 10th February, 2025. The seeds were examined and authenticated by a taxonomist in the Department of Botany, Usmanu Danfodiyo University, Sokoto. As part of the authentication procedure, voucher specimens of the authenticated cultivars were documented under the reference number “UDUH/ANS/0982”, dated 13th February, 2025, and deposited in the university’s herbarium for future reference.

The rate of groundnut consumption in Sokoto was assessed using a structured questionnaire and an oral interview of willing and consenting individuals. The ethical approval (ethical no: SKHREC/086/2025) for the study was granted by the Sokoto Health Research Ethics Committee, Ministry of Health Sokoto, Sokoto State, Nigeria. All procedures were conducted in accordance with the ethical principles outlined in the Declaration of Helsinki for medical research involving human subjects, first adopted in 1964 and most recently amended in October 2024. The assessment was conducted using 500 adults and 300 children not less than 4 years, randomly selected within Sokoto metropolis. Consent to engage the children was obtained from their parents or guardians. Also, prior to the evaluation, each adult participant signed a consent form acknowledging their understanding of the nature of the research. The information on groundnut intake was collected using a 24-hour dietary recall method for 7 consecutive days (1st–7th September, 2025) weekends excluded.

The groundnut seeds were manually sorted to remove broken seeds, insect-infested kernels, and other extraneous materials. Sensory evaluation of the samples was conducted as outlined by Wakhu-Wamunga and Wamunga [20]. A panel of 100 individuals with equal representation of males and females (50 males and 50 females) assessed the samples based on taste, aroma, flavor, and texture using a nine-point hedonic scale, where 1 = dislike extremely, 2 = dislike very much, 3 = dislike moderately, 4 = dislike slightly, 5 = neither like nor dislike, 6 = like slightly, 7 = like moderately, 8 = like very much, and 9 = like extremely. Prior to the evaluation, each participant signed a consent form acknowledging their understanding of the nature of the samples to be tested. The panel for the sensory evaluation consisted exclusively of individuals aged eighteen and above. During the assessment, each panelist was provided with a white tray containing the samples served in disposable plastic bowls, along with bottled water for rinsing the mouth before and between tastings. To ensure blinding, four-digit random codes were assigned to the samples, and the order of sample presentation was randomized for each panelist.

The best-performing cultivar obtained from sensory evaluation was subjected to ten different pretreatment methods to identify the best performing method for reducing aflatoxin and heavy/trace metal levels. These methods include roasting, soaking in various media (10% hot saline, 10% normal saline, distilled water, and hot distilled water), combinations of soaking and roasting, as well as oven drying [13, 21]. Roasting was carried out manually in a clean pan at approximately 150°C for about 15 minutes with continuous stirring until the groundnuts developed a characteristic golden-brown colour. For soaking treatments, the groundnuts were immersed in the respective solutions for 4 hours. The hot saline solution consisted of 10% NaCl solution maintained at approximately 60°C during soaking, while the normal saline solution consisted of 10% NaCl at room temperature.

The effects of these pretreatment methods on nutritional composition (protein quality, carbohydrates, and lipids) and anti-nutritional factors (ANFs) were evaluated in our previous study and reported [15]. This present study focuses on sensory evaluation and subsequent safety assessments. Figure 1 depicts the flowchart of the pretreatment processes.

SAMNUT-23 (the best performing cultivar) was analyzed for the presence of oxalate and phytate (anti-nutrients). The analysis was conducted according to the procedures described by Ijarotimi et al. [22]. All analyses were carried out in triplicate.

The method outlined by Nurudeen et al. [23] was used to determine the levels of heavy/trace metals in the sample. Exactly 5 g of the pre-treated SAMNUT-23 was digested in 20 mL of aqua regia (1:3 HCl/HNO₃) until the cessation of brown fume evolution. Five grams (5 g) of the dried sample was weighed in a porcelain crucible and heated at 600°C for 2 hours in a muffle furnace (Gallenkamp, Model OV 880, London, England). The burnt sample was allowed to cool for 1 hour and then digested in 20 mL of aqua regia. The solution was heated on an electric hotplate until the brown smoke vanished. The solution was then filtered using Whatman No. 1 filter paper and made up to 100 mL with distilled water. The solution was then subjected to atomic absorption spectrophotometric (Shimadzu UK) analysis operated according to the manufacturer’s guidelines to obtain the concentrations of heavy/trace metals [Pb, Cd, Cr, nickel (Ni), and zinc (Zn)]. Appropriate blanks and calibration standards were run to ensure accuracy and precision. The analysis was carried out in triplicate.

The mole of phytate or oxalate and heavy/trace metals ratio was determined as previously described [22, 24]. The molar ratio of the phytate or oxalate and heavy/trace metals was obtained by dividing the weight of phytate or oxalate and heavy/trace metals with its atomic weight (phytate: 660 g/mol, oxalate: 126 g/mol, Zn: 65.38 g/mol; Cr: 52 g/mol, Cd: 112.41 g/mol, Ni: 58.69 g/mol, and Pb: 207.20 g/mol) as shown in Equation 1.

(1) p h y t a t e / o x a l a t e : m e t a l   r a t i o   =   m o l e s   o f   p h y t a t e   o r   o x a l a t e m o l e s   o f   h e a v y / t r a c e   m e t a l s

Aflatoxin levels were evaluated using high-performance liquid chromatography (HPLC) following the procedure described by Nurudeen et al. [23]. Chromatographic separation was performed using a uBondapak C18 reversed-phase column (250 × 4.6 mm, 5 μm particle size). The mobile phase consisted of acetonitrile and water in a 70:30 (v/v) ratio, filtered through a 0.45 μm membrane, and degassed using ultrasonic treatment before use. Methyl acetate was added to stabilize the mobile phase. The flow rate was set at 2 mL/min, with an injection volume of 5 μL. Specific detection wavelengths were used for identifying the various aflatoxins (B1, B2, G1, and G2) in the groundnut samples. The identity of each analyte was confirmed by comparing its retention time and UV spectrum with those of certified reference standards. Triplicate analysis was run on each sample.

The human health risk assessment procedures used to estimate cancer and non-cancer risks were adapted from the methodology described by Yahaya et al. [16]. The non-cancer risks were obtained from estimated daily intake (EDI), hazard quotient (HQ), and hazard index (HI) of the heavy/trace metals in the groundnuts (Equations 2, 3, and 4, respectively). The cancer risk was estimated from Equation 5.

(2) E D I   m g / k g / d a y   =   C i   ×   G I R A B W

(3) H Q   =   C i   ×   G I R A B W   ×   O R F D

(4) H I   =   ∑ n i n   -   1 H Q

(5) E C R   =   E D I   ×   C S F

Where Ci is the concentration of the individual heavy/trace metals in the groundnut samples, GIR represents groundnut ingestion rate in kg·person^–1day^–1, ABW indicates average body weight, and ORFD means oral reference dose (mg·kg^–1day^–1). ΣHQ is the summation of the HQs, while CSF means cancer slope factor (mg·kg^–1day^–1). HQ and HI values above 1 were considered toxic.

The exposure levels of consumers to aflatoxin were determined as described by Adetunji et al. [12]. The quantity of aflatoxins (obtained using questionnaires described earlier) in the groundnut samples was multiplied by the average consumption rate of the nuts in kg/person/day, which was then divided by the ABW of adults in kg. The exposure assessment [probable daily intake (PDI)] was thus calculated using Equation 6. In addition, HQ was computed using Equation 7 while HI was calculated using Equation 8.

(6) P D I   μ g · k g – 1 · b w · d a y – 1   =   C a   ×   A C G A B W

(7) H Q   =   P D I O R F D

(8) H I   =   ∑ n i n   -   1 H Q

Where PDI is probable daily intake for aflatoxins (µg·kg^–1·bw·day^–1); Ca is concentration of aflatoxins in the groundnuts (µg/kg); ACG is average consumption of groundnuts (kg/day); ABW is body weight for an adult (kg).

The risk of aflatoxin exposure from the groundnut was determined based on the margin of exposures (MOEs) [12, 25]. This was calculated by dividing the benchmark dose lower limit (BMDL) for aflatoxins 170 µg·kg^–1·bw·day^–1 by aflatoxin exposure (PDI) as shown in Equation 9.

(9) M O E   =   B M D L P D I

In cases where MOEs were lower than 10,000, a public health concern is indicated which implied that aflatoxin exposures above 0.017 µg·kg^–1·bw·day^–1 (as obtained by dividing 170 µg·kg^–1·bw·day^–1 by 10,000) represented a risk of public health concern [12, 25].

The concentrations of heavy/trace metals and aflatoxin in the samples were presented as mean ± standard deviation (SD) using GraphPad Prism 6.01 (GraphPad Software, Inc., San Diego, California, United States). The means obtained were compared with standard values using Student’s t-test and one-way analysis of variance (ANOVA) for multiple comparisons at p < 0.05. The GraphPad was also used to calculate the EDI, PDI, MOE, and HQ values of the heavy metals and aflatoxin. The charts were drawn using Microsoft Excel 2010.

Table 1 shows the demographic information of the adult respondents, including their age, gender, educational status, marital status, and ethnicity. Of the 500 adults that responded, males had the highest participants with 288 (57.6%) individuals while females were 212 (42.4%) in number. Age class 31–35 years had the highest respondents, having 214 (42.8%) individuals while the least fall within the age class 41 and above, comprising 11 respondents (2.2%). In terms of education, 38 respondents (7.6%) had no formal education, 54 (10.8%) respondents had primary education, while 144 (28.8%) and 264 (52.8%) had secondary and tertiary education, respectively. Ethnically, the Hausa group constituted the majority, comprising 250 respondents (50%), followed by the Yoruba ethnic group, with 145 respondents (29%), Fulani with 58 respondents (11.6%), and Nupe and Igbo with 25 (5%) and 22 (4.4%) respondents, respectively. The majority of the respondents identified as married, having 470 (94%) individuals, while 30 respondents (6%) identified as single.

Table 2 presents the demographic information of the child respondents. The age distribution indicates that the highest number of participants falls within the 6–8 years age group, comprising 128 children (42.67%), followed by the 8–10 years age group with 103 respondents (34.33%). The 10–12 years age group accounted for 66 children (22%), while the least represented group was the 4–5 years age class with only 3 children (1%). In terms of gender, the study included equal representation of males and females, with 150 male respondents (50%) and 150 female respondents (50%). Regarding current educational status, all the children were enrolled in schools. Ethnic distribution shows that the Hausa group constituted the majority, comprising 205 children (68.33%), followed by the Yoruba with 81 respondents (27%) and the Fulani with 14 respondents (4.67%). The parental status of the children revealed that none of the respondents were orphans, as all 300 children (100%) reported having both parents alive.

Table 3 presents the ABW and corresponding consumption rate of SAMNUT-23 groundnuts among adults and children in Sokoto State. The results indicated that the mean body weight of adults was 60 kg, while that of children was 22.11 kg. The average groundnut intake by adults was 52 g (0.052 kg), while 19.49 g (0.0195 kg) was recorded for children.

The sensory qualities of the three groundnut cultivars are displayed in Figure 2. The hedonic scale test revealed that SAMNUT-23 had the highest acceptability rating score, having 7.57 compared to 6.57 for SAMNUT-24 and 6.43 for SAMNUT-26. Moreover, SAMNUT-23 had the highest sensory scores in terms of color (8.14), taste (6.43), and texture (7.29). The nutritional and ANFs of these cultivars were comprehensively evaluated in our previous study [15], and these data, combined with the current sensory evaluation, were considered in the selection process. For clarity and focus, only sensory results are presented in the current study, as these were the criteria for selection of the most acceptable sample for further studies.

Figure 3 shows the sensory qualities of SAMNUT-23 subjected to various pretreatment methods. Compared to untreated SAMNUT-23, there were significant differences (p < 0.05) across appearance, aroma, taste, texture, and overall acceptability. In terms of appearance, the roasted-only SAMNUT-23 had the most appealing color and visual quality, recording the highest score of 7.46. In contrast, the sample soaked in normal water had the least favorable appearance score of 6.06. The combination of normal saline soaking followed by roasting produced the most desirable aroma (7.13), followed by hot saline plus roasting treatment with a score of 7.06. Meanwhile, hot water soaking and normal water soaking recorded the lowest aroma scores of 5.00 and 5.13, respectively. The most preferred taste was observed in normal saline + roasting (8.00) and hot saline +roasting (7.66). Conversely, hot water soaking had the lowest taste score (4.53). Samples subjected to roasting alone and hot saline + roasting shared the highest texture scores (7.33). In contrast, normal water soaking (5.33) and hot water soaking (5.66) recorded the lowest texture ratings, reflecting a softer or less firm consistency due to moisture absorption without adequate drying. In terms of overall acceptability, the hot saline plus roasting and normal saline plus roasting treatment showed no significance difference (p < 0.05) and were the most preferred by the panelists, with an overall acceptability score of 7.93 and 7.86 respectively. Hot water soaking yielded the lowest overall acceptability score of 5.26.

Furthermore, among all pretreatment methods examined for sensory parameters, the combination of normal saline soaking followed by roasting emerged as the best pretreatment method for enhancing the sensory quality of SAMNUT-23, offering superior scores across taste, aroma, texture, and overall acceptability. Similarly, roasting alone also produced favorable results, particularly for appearance and texture. Hot water soaking was the least effective method, resulting in consistently low scores across nearly all sensory attributes. It is noteworthy that the effects of these pretreatments on nutritional composition and ANFs were comprehensively evaluated in our previous findings [15], and these data, combined with the current sensory evaluation, were considered in the selection process. However, for clarity, only sensory results are presented in the current study, as these were the criteria for selection and further evaluation.

Based on its superior sensory attributes and acceptable nutritional profile, as previously reported by Yusuf et al. [15], normal saline + roasting was selected for further analysis. Figure 4 shows the levels of aflatoxins in SAMNUT-23 subjected to normal saline and roasting. A significant decrease (p < 0.05) was observed in the levels of aflatoxin B1 (0.71 µg/kg), B2 (0.18 µg/kg), and G2 (0.47 µg/kg) for the SAMNUT-23 treated with a combination of normal saline + roasting when compared with untreated SAMNUT-23, which has concentrations of 1.92 µg/kg, 0.76 µg/kg, and 0.89 µg/kg for aflatoxin B1, B2, and G2, respectively. However, a significant increase (p < 0.05) in aflatoxin G1 (0.36 µg/kg) was observed in the normal saline + roasting treated SAMNUT-23 when compared to the untreated SAMNUT-23 (0.09 µg/kg).

Table 4 shows the concentrations of oxalate and phytate in treated and untreated SAMNUT-23. The oxalate content of the untreated sample was 0.008 ± 0.001 mg/g, but decreased to 0.002 ± 0.001 mg/g after treatment. The phytate content of the untreated sample was 4.6 ± 0.4 mg/g but decreased to 1.3 ± 0.4 mg/g after pretreatment.

Table 5 shows the concentrations of Pb, Cd, Cr, Ni, and Zn in the treated and untreated SAMNUT-23. Pb was below detection levels (BDL) in either the untreated or treated samples. Cd levels were low in both untreated (0.01 ± 0.00 mg/kg) and treated (0.01 ± 0.00 mg/kg) SAMNUT-23 samples, with no significant difference between them. The concentration of Cr was significantly reduced from 0.05 ± 0.01 mg/kg in the untreated to 0.02 ± 0.01 mg/kg in the treated SAMNUT-23. A significant difference (p < 0.05) was also noticed in Ni concentrations between the untreated (0.04 ± 0.02 mg/kg) and treated (0.03 ± 0.01 mg/kg) SAMNUT-23, while Zn was notably reduced from 0.70 ± 0.00 mg/kg in the untreated to 0.43 ± 0.00 mg/kg in the treated SAMNUT-23. Noteworthily, all the analyzed heavy metals were within the permissible limits set by FAO/WHO [26].

The phytate-to-heavy metal molar ratios indicated generally reduced chelation capacity after treatment. The phytate:Cd ratio decreased from 65.96 in the sample to 19.70 in the treated sample. Phytate:Ni decreased from 9.18 to 3.89, and phytate:Cr from 7.18 to 4.00, while phytate:Zn showed a slight reduction (0.66 to 0.29). In the oxalate-to-heavy metal molar ratios, significant reductions were also observed following treatment. A significant decrease was noted in oxalate:Cd ratio (0.80 to 0.16), oxalate:Cr (0.09 to 0.03), oxalate:Zn (0.008 to 0.002), and oxalate:Ni (0.11 to 0.03) ratios. All showed significant decrease (p < 0.05) between the raw and treated SAMNUT-23.

Table 6 reveals that the EDI of aflatoxins (B1, B2, G1, and G2) from both untreated and treated SAMNUT-23 was within the WHO tolerable daily intake (TDI) limits. For children that consumed the untreated SAMNUT-23, the EDI values were B1: 0.002 µg/kg/day, B2: 0.0007 µg/kg/day, G1: 0.00008 µg/kg/day, and G2: 0.0008 µg/kg/day. After pretreatment, these values were reduced to B1: 0.0006, B2: 0.0002, G1: 0.0003, and G2: 0.0004 µg/kg/day. A similar reduction trend was observed in adults, with treated SAMNUT-23 having markedly lower values (B1: 6.15 × 10^–4 µg/kg/day; B2: 1.56 × 10^–4 µg/kg/day).

The HQ and HI of aflatoxin from SAMNUT-3 by children and adults are shown in Table 7. The HQ for all aflatoxin types in both children and adults was less than the threshold of 1. For children that consumed untreated SAMNUT-23, the highest HQ was observed in aflatoxin B2 (0.07), followed by B1 (0.02). Treated SAMNUT-23 further lowered these values (B2: 0.02; B1: 0.006). The overall HI for children was 0.099 for the untreated and 0.033 for treated SAMNUT-23, while for adults, the HI was 9.09 × 10^–2 for the untreated and 2.89 × 10^–2 for treated SAMNUT-23.

The MOE of B1, B2, G1, and G2 in SAMNUT-23 were shown in Table 8. The MOE values were generally high, indicating low risk levels. For children consuming untreated SAMNUT-23, the MOE for aflatoxin B1 was 8.5 × 10⁴, while a significant increase was observed in the treated SAMNUT-23 (2.83 × 10⁵). Adults showed even higher MOE values for treated SAMNUT-23 (B1: 2.76 × 10⁵, B2: 1.09 × 10⁶). Interestingly, the MOE values were greater than 10,000, indicating that both untreated and treated SAMNUT-23 fall within acceptable safety margins.

Table 9 reveals that the EDI of Cd, Cr, Ni, and Zn was higher in untreated SAMNUT-23 when compared to the treated SAMNUT-23 in both children and adults. However, it is worth noting that EDI values for the analyzed metals in both untreated and treated SAMNUT-23 remained within the FAO/WHO [26] TDI limits, indicating no immediate concern of excessive intake.

Table 10 shows that in children, the HQ for Cd in untreated SAMNUT-23 (1.02) exceeded the safe limit (≥ 1), while other metals (Cr, Ni, Zn) were within the threshold (≤ 1). The HI for untreated SAMNUT-23 in children was 1.04, slightly above the safety threshold, suggesting potential health risk. Interestingly, in the treated SAMNUT-23, all HQ values were within the safety threshold (< 1), and the HI reduced to 0.103, indicating safety. In adults, HQ values for all metals were < 1, with HI values of 0.123 (untreated) and 0.105 (treated), both within safe limits.

Table 11 revealed that ECRs for Cd, Cr, and Ni were higher in untreated SAMNUT-23 than in treated SAMNUT-23 for both children and adults. Children consuming untreated SAMNUT-23 had ECR values of 6.36 × 10^–4 (Cd), 2.25 × 10^–5 (Cr), and 3.61 × 10^–6 (Ni), whereas lower values (5.82 × 10^–5, 1.11 × 10^–5, and 2.33 × 10^–6, respectively) were in the treated SAMNUT-23. Interestingly, for adults, treated groundnuts also had lower ECR (Cd = 5.71 × 10^–5, Cr = 1.09 × 10^–5, Ni = 2.28 × 10^–6). ECR values for Zn were not available (N/A) in all cases.

In conclusion, this study demonstrated that both varietal selection and processing methods play critical roles in determining the sensory quality, safety, and nutritional suitability of groundnuts for therapeutic food applications. Among the three cultivars evaluated, SAMNUT-23 consistently exhibited superior sensory attributes, including taste, texture, aroma, color, and overall acceptability, confirming it as the most preferred cultivar. This superiority is likely linked to its favorable biochemical composition and aligns with existing evidence on the influence of genetic variability on groundnut quality. Pretreatment methods significantly influenced sensory outcomes, with normal saline soaking combined with roasting emerging as the most effective approach. This method achieved the highest overall acceptability by enhancing flavor, aroma, texture, and visual appeal, demonstrating that consumer acceptability is not solely dependent on nutritional composition but also on processing-induced sensory improvements. Although hot water soaking combined with roasting showed slightly better nutritional advantages, the saline-roasting method provided a more balanced profile suitable for practical utilization, particularly in RUTF formulations for children with SAM. Importantly, the selected treatment also improved safety parameters by significantly reducing aflatoxin levels and lowering ANFs such as phytate and oxalate, thereby enhancing mineral bioavailability. Heavy metal concentrations remained within permissible limits, with Pb undetected, further supporting the safety of SAMNUT-23. Health risk assessments confirmed that both treated and untreated samples posed no significant risk to adults, while treatment effectively eliminated potential non-ECRs in children. Overall, the findings establish SAMNUT-23 treated with normal saline soaking followed by roasting as a safe, nutritionally adequate, and highly acceptable candidate for RUTF production. This integrated approach highlights the importance of combining optimal cultivar selection with appropriate processing techniques to achieve both safety and consumer acceptability in food product development.