Natural history of sesame allergy in pediatric patients: Insight from a retrospective analysis

Abstract

Background

Sesame allergy (SA) is a growing concern because of its association with severe reactions and the limited knowledge of long‐term outcomes.

Objective

This retrospective study aimed to identify the risk factors influencing persistent SA (PSA) in children to improve management and select suitable candidates for oral immunotherapy (OIT).

Methods

We analyzed the electronic medical records of 84 children with confirmed SA, as defined by consistent clinical reactions and immunoglobulin E (IgE)‐mediated sensitization. Patients were followed for a median (IQR) of 56.5 (46.0–82.5) months.

Results

Most participants were male (72.6%) with concurrent food allergies (71.4%). They experienced a median (IQR) of 3.0 (2.0–3.7) allergic episodes, with 46.4% experiencing at least one anaphylactic reaction. PSA was observed in 82.1% (69/84) of the patients. A larger skin prick test (SPT) wheal size at the first reaction (adjusted OR = 1.79, CI:1.05–3.04; p = .03) and allergic reaction grade≥2 (adjusted OR = 19.93, CI:1.37–289.13; p = .02) were independent risk factors for persistence. A 3‐fold increase in the likelihood of persistence was observed in patients with SPT results greater than 6.7 mm at first reaction compared with those with results less than 6.7 mm during follow‐up (HR = 3.08; CI:1.17–8.12; p = .02). Patients with sustained or increased SPT wheal size (93% remained allergic) and specific IgE (95% remained allergic) at the final visit were more likely to have PSA, whereas those with decreased levels (37% and 39% developed natural tolerance, respectively) were less likely to have resolved SA.

Conclusions

This study identified novel risk factors for PSA, including SPT wheal size at the first reaction, reaction severity, and sustained sensitization markers. These findings can inform management strategies and the selection of OIT candidates. Further long‐term studies are crucial to elucidate the natural history of SA across populations and to evaluate early interventions, such as OIT.

Abbreviations

AD

atopic dermatitis

AUC

area under the curve

CI

confidence interval

CSE

commercial sesame extract

FA

food allergy

HR

hazard ratio

IgE

immunoglobulin E

IQR

interquartile ranges

OFC

oral food challenge

OIT

oral immunotherapy

PFA

persistent food allergy

PSA

persistent sesame allergy

ROC

receiver operating characteristic

RSA

resolved sesame allergy

SA

sesame allergy

Ses i 1

sesamum indicum 1

SPT

skin prick test

ssIgE

sesame‐specific IgE

tIgE

total IgE

U.S.

United States

Key message.

Natural tolerance proportion in patients with a confirmed sesame allergy (SA) is determined to be 18%. SPT results are an independent predictor of SA persistence, but sesame‐specific IgE levels are not. Increasing levels of sensitizations served as better predictors of persistence than decreasing levels in predicting resolution.

1. INTRODUCTION

Sesame allergy (SA) is gaining recognition as a prevalent food allergy (FA) with substantial regional variations, ¹ potentially linked to ethnicity and cultural factors. For example, the reported prevalence of SA in children is 0.7% in Israel, ² and 0.8% in Australia. ³ In contrast, physician‐confirmed SA in children in the United States (U.S.) is less common, with a prevalence of 0.12%. ⁴ However, a study highlighting the importance of ethnicity revealed that individuals of Middle Eastern and North African descent in the U.S. were more likely to develop SA compared with other food allergies. ⁵ This trend was not observed in other ethnic groups. ⁵

The significance of SA is underscored by its ranking as the third most common immunoglobulin E (IgE)‐mediated FA among Israeli children, ⁶ and the fourth among Australian ⁷ and Turkish children. ⁸ Documented fatalities highlight the severity of SA, ⁹ with strict avoidance remaining the primary management approach, while oral immunotherapy (OIT) shows emerging promise. ¹⁰ Exciting data suggests that early introduction of OIT may lead to allergy resolution. ¹¹ Unlike more frequently outgrown allergies like cow's milk and hen's eggs, SA displays a relative persistence similar to that of tree nuts and peanuts. ¹²

Limited data regarding the natural history of SA are available, suggesting that 20%–32% of children may achieve tolerance over time. ¹³ , ¹⁴ , ¹⁵ However, persistent SA (PSA) predictors remain inconclusive, with studies reporting conflicting results for factors such as age at onset, reaction severity, coexisting food allergies, sex, and skin prick test (SPT) wheal size. ¹³ , ¹⁴ , ¹⁵ This variability highlights the potential influence of exposomal and ethnic factors, along with their inherent characteristics, on PSA. The currently available data on SA resolution stem primarily from studies conducted in specific regions, leaving a gap in our comprehensive understanding of the factors influencing the natural history of SA. This includes identifying patients eligible for OIT and improving the overall management of SA.

This study aimed to investigate the clinical characteristics of patients with confirmed SA, determine the proportion of resolution and persistence, and identify the clinical and laboratory parameters that may predict PSA.

2. METHODS

2.1. Study design and participants

This retrospective longitudinal study was conducted at Hacettepe University Ihsan Dogramacı Children's Hospital, Pediatric Allergy Department, a referral center in Turkey. Approval was obtained from the university ethics committee before the study began (SBA 23/090). Approximately 1000 pediatric patients with FAs are regularly followed at our center, with electronic records of their current status. We included patients who were monitored for SA in our department between January 2005 and December 2023. There were 175 patients with sesame sensitization, of whom 143 were diagnosed with IgE‐mediated SA. Twenty‐three patients with SA were excluded from the study because they were seen only once or did not have a detailed medical history. There were 120 with IgE‐mediated SA with at least 2 years of follow‐up data. Subsequently, 36 patients without a history of sesame exposure at the final visit or within 1 year of the final visit were excluded from the analysis to assess persistence or resolution. Ultimately, the study population comprised 84 patients with IgE‐mediated SA, of whom data were collected (Figure 1).

FIGURE 1.

Flowchart of patient selection.

In Turkey, sesame seeds are the fourth most common cause of FA in infants ⁸ and one of the most important food allergens in children. ¹⁶ Sesame is typically consumed as a seed on bakery products or as tahini in our country. Tahini, or sesame paste, is a common ingredient in Middle Eastern culinary traditions, typically served either on its own or incorporated into dishes such as hummus or baba ghanoush. It can also be present in certain cookies, candies, and desserts like halva.

IgE‐mediated SA was diagnosed via three routes: (1) Early‐onset moderate to severe atopic dermatitis (AD) (AD subgroup). Patients with early‐onset, moderate‐to‐severe and persistent AD despite appropriate skin care were evaluated for SA. These patients had a positive SPT with commercial sesame extract (CSE) or tahini and/or elevated sesame‐specific IgE (ssIgE) and exhibited clear improvement in AD symptoms following sesame elimination. (2) During evaluation for other FAs (other FA subgroup). Patients in the other FA subgroup were tested by a SPT using CSE if they had not yet consumed sesame products at the time of diagnosis. Sesame elimination was recommended only for patients with a result of ≥8 mm wheal diameter of CSE SPT. An 8 mm cut‐off was established for peanuts allergy in a previous study ¹⁷ and resulted in a 95% positive predictive value for SA. ¹⁸ Patients with a sensitization level of less than 95% PPV (3–7 mm) were considered for oral food challenge (OFC). (3) Through post‐reaction diagnostics (reactive subgroup). These patients diagnosed with SA after sesame‐related IgE‐mediated reaction.

For cases where SA was diagnosed through sesame sensitization, individuals were included in the study if they experienced a reaction due to exposure to any sesame product during the follow‐up period. This was considered to be the first reaction in these patients. Sesame‐related IgE‐mediated reactions relied on a positive sesame OFC or a consistent history (physician‐confirmed) of IgE‐mediated allergy following the ingestion of any sesame product. ¹⁹ This was complemented by IgE‐mediated sensitization indicators, such as a positive SPT with CSE or tahini and/or elevated ssIgE.

The follow‐up period was defined as the interval between the diagnosis of SA (first visit) and the final visit at which persistence or resolution was evaluated. To ensure an accurate assessment of the natural history of SA, we included only patients with SA with over 2 years of follow‐up period.

The primary endpoint of this study was to identify patients with SA resolution and persistence during the follow‐up period. Resolution was determined through a negative OFC at the final visit and a consistent history of regular consumption. Persistence was defined as a positive OFC result at the final visit or a consistent history of IgE‐mediated reactions reported by the patients or their families to sesame products during the last year of the follow‐up period. Patients who consumed minimal quantities of sesame or its products without experiencing any reaction were categorized as unknown unless they underwent an OFC thereafter. The IgE‐mediated allergic reactions at the first and last reactions were based on objective clinical symptoms. The patient selection process is illustrated in Figure 1 and the study design is illustrated in Figure S1.

The secondary endpoint was to determine the risk factors for PSA and the predictive values of sensitization levels to distinguish between persistent and resolved SA cases. The following information was obtained from the patients' electronic records: sex; type of diagnosis; age at diagnosis; age at first reaction; symptoms of the first reaction; sesame form (sesame seeds or tahini) responsible for the first reaction; amount of sesame product consumed at the first reaction; number of allergic reactions to sesame; number of anaphylactic reactions to sesame; severity of allergic reactions; SPT results; ssIgE results; age at the final visit; symptoms of the last reaction; OFC outcome based on the final visit; and atopic comorbidities. Coexisting FAs were diagnosed based on objective IgE‐mediated clinical reactions confirmed by a physician or through an OFC following ingestion of the suspected foods, like sesame. Inconsistencies were resolved through telephone interviews, and the reported reactions were cross‐checked with medical records. Anaphylaxis was diagnosed according to the European Academy of Allergy and Clinical Immunology guidelines. ²⁰ The severity of the allergic reactions was graded using the severity grading system for acute allergic reactions. ²¹

2.2. Skin prick and specific immunoglobulin tests

SPTs were performed using a commercial sesame allergen extract (ALK‐Abello; Hørsholm, Denmark) and/or raw tahini (Koska; Istanbul, Turkey, containing 230 mg protein/g). Raw tahini was applied without dilution using the prick‐to‐prick technique. Saline and histamine were used as the negative and positive controls, respectively. A positive SPT result was defined as a mean wheal diameter of 3 mm or greater than the control.

The ImmunoCAP system (Thermo Fisher Scientific, Waltham, MA, USA) was used to measure ssIgE levels, with a positive result considered 0.1 kU_A/L or greater based on the current evidence. ²² Patients with multiple FAs were also tested by molecular analysis using a commercial multiplex array (ALEX, ² Macro Array Diagnostics, Vienna, Austria) ²³ with a positive cutoff of ≥0.3 kU_A/L. All ssIgE levels in our study were measured using ImmunoCAP, while Ses i 1 sIgE levels were obtained using a commercial multiplex microarray (ALEX2). Total IgE (tIgE) levels, measured by both ImmunoCAP and the multiplex microarray, were used to calculate the ssIgE/tIgE ratio and the Ses i 1 sIgE/tIgE ratio, respectively.

All reported sensitization levels were recorded during the follow‐up period. Sensitization levels representing the first reaction were documented within 1 year of the first reaction time, and sensitization levels representing persistence and resolution were documented at the final visit.

2.3. OFC protocol

In line with the practical allergy (PRACTALL) consensus, a standardized protocol was implemented for the open OFC. ²⁴ To assess tolerance, patients who underwent the OFC with tahini were included, as our previous study suggested that a specific quantity of sesame seeds was likely tolerated by patients with SA. ¹⁹ However, if the OFCs using sesame seeds were performed to evaluate the potential tolerability and yielded positive results, then those patients were also included in the study. The initial dose consisted of 50 mg of tahini (11.5 mg of sesame protein), followed by incremental increases in dose every 20 min (50 mg, 100 mg, 250 mg, 500 mg, 1000 mg, 2000 mg, 4000 mg, and 4000 mg of tahini). Among children under 3 years of age, the interval between doses was 30 min, and the maximum amount of tahini to be administered at the last dose was 2000 mg. The sesame challenge was deemed positive when objective evidence of an allergic reaction was observed. The test was considered negative if the patient showed no signs or symptoms of allergies within 2 h after taking the last dose and subsequently reported regular tahini and/or halva consumption without allergic reactions. Written informed consent was obtained from the families and patients before the OFC.

2.4. Statistical analysis

Statistical analyses were performed using IBM SPSS Statistics, version 25.0 (International Business Machines Corporation, Armonk, NY, USA). Non‐normally distributed data were presented as medians and interquartile ranges (IQR). The chi‐square or Fisher's exact test was used to determine the differences in nominal variables. The numerical data for the two independent groups were compared using the non‐parametric Mann–Whitney U test. The Wilcoxon signed‐rank and McNemar's tests were used to compare two dependent samples. Multivariable logistic regression analysis was performed using forced entry method. Log‐transformed ssIgE levels were utilized for logistic regression analysis. Receiver operating characteristic (ROC) curve analyses were used to assess the predictive levels of ssIgE, ssIgE/tIgE ratio, and SPT wheal size to distinguish between resolved and persistent patients. The time to resolution of SA was analyzed using follow‐up time as the time metric. A proportional hazards regression model was fitted to examine the effect of covariates on the hazard function. This model was established according to significance of factors and the adequacy of model fit. P values of <0.05 were considered statistically significant.

3. RESULTS

A total of 84 patients with SA were included in this study, and 72.6% (n = 61) of them were male. Most patients (71.4% [n = 60]) had a concurrent FA, most commonly tree nuts (57.1% [n = 48]) and seeds (25.0% [n = 21]) (poppy, sunflower, pumpkin, and mustard), at the final visit. Sixty‐nine (82.1%) patients had a previous diagnosis of AD, 48 (57.1%) had current asthma, and 37 (44.0%) had current allergic rhinitis. The median (IQR) follow‐up time for SA was 56.5 (46.0–82.5) months, ranging from 24 to 208 months, with a median (IQR) age of 65.0 (58.0–96.0) months at the final visit. For patients diagnosed through sensitization, the median follow‐up time after first immediate reaction was 38 months (IQR: 27–48 months). All participants had documented sesame‐related allergic reactions, with a median (IQR) of 3.0 (2.0–3.7) episodes per person. Notably, 46.4% (n = 39) of the patients experienced at least one anaphylactic episode. Additionally, 41.6% (n = 35) of the patients exhibited at least one positive OFC result during the follow‐up period.

Among the 84 patients, 24 did not seek medical attention for their first sesame‐related reaction, and the incidents were not formally documented in the healthcare system; therefore, their first sesame‐related reaction data could not be included in the analysis. Sesame seeds accounted for 16.6% of the first reactions (n = 60), while tahini accounted for 83.3%. The first reaction resulted in adrenaline administration in 16.6% (10/60) of the patients. Nine patients experienced their first allergic reaction during the OFC, and 56.8% (n = 29/51) of the first reactions to sesame exposure outside of the hospital led to emergency room visits. Skin/mucosal reactions were the most common first reactions, representing 85% of cases; gastrointestinal and respiratory reactions were reported in 31.7% and 11.6% of patients, respectively (Figure 2A). Of the 60 patients with first reaction data, 53.3% experienced grade 1 reactions, 33.3% grade 2, 11.7% grade 3, and 1.7% grade 5 reactions (Figure 2A). The patient demographics and clinical characteristics are shown in Table S1.

FIGURE 2.

Grade and symptom rates of the first (A) and last reactions (B).

3.1. SA diagnosis

Thirty‐two (38%) patients were in the AD subgroup, 40 (48%) in the reactive subgroup, and 12 (14%) in the other FA subgroup. Median (IQR) age, SPT wheal diameter with CSE, and ssIgE level at diagnosis were 10.0 months (6.0–13.0 months), 6.0 mm (4.8–8.0 mm), and 5.6 kU_A/L (1.0–14.9 kU_A/L), respectively (Table S1). The median (IQR) age of the first reaction in the AD (36.0 months [24.0–51.0 months]) and other FA (36.0 months [23.0–105.0 months]) subgroups was significantly higher than in the reactive subgroup (12.0 months [9.5–22.0 months]) (p < .01 for each, Table S2).

3.2. SA outcome

Our final evaluation revealed that 69 patients (82.1%) exhibited PSA, while 15 (17.8%) developed natural tolerance at a median follow‐up time of 56.5 months (Figure S2), which was supported by a consistent history of regular sesame consumption without allergic reactions. Among the patients with PSA, 19 had a positive OFC result, and 50 reported a consistent history of IgE‐mediated allergic reactions to sesame products during the last year of follow‐up. Notably, all patients with resolved sesame allergy (RSA) demonstrated negative OFC outcomes. This suggests that none of the patients with RSA had reintroduced sesame products into their diet before the challenge.

The median follow‐up time was 58.0 months (IQR:47.0–83.5 months) for patients with PSA and 54.0 months (IQR:36.0–60.0) for patients with RSA. Two patients with PSA had follow‐up periods of 24 and 26 months, while all other patients with PSA had follow‐up periods exceeding 33 months. The age of patients with RSA ranged from 3.2 to 18 years.

The most recent sesame‐related reactions in patients with persistent allergy are shown in Figure 2B. Among 69 patients with PSA, 50.7% had grade 1 reactions, 27.5% had grade 2, 15.9% had grade 3, and 5.8% had grade 4 at the last reactions. There were no significant differences in the rates of cardiovascular and respiratory symptoms between the first and last reactions. During the follow‐up period, 49 patients (49/84, 58.3%) experienced physician‐confirmed grade 2 or higher severe allergic reaction to sesame.

No significant difference was observed in the persistence proportions at the final visit among the diagnostic subgroups, with proportions of 90.6% (n = 29/32) in the AD subgroup, 77.5% (n = 31/40) in the reactive subgroup, and 75.0% (n = 9/12) in the other FA subgroups.

3.3. Persistent sesame allergy predictors

Patients with PSA had a significantly higher rate of grade 2 or higher severe reactions (the most severe reactions experienced during the follow‐up period) than patients with RSA (Table 1). In addition, patients with PSA exhibited significantly higher levels of sensitization, as documented by the SPT with CSE and tahini, as well as the ssIgE levels during their first reaction, compared with patients with RSA. No significant differences were found between the two groups regarding sex, concomitant FAs (based on the last visit), atopic comorbidities, age at diagnosis, and age at final visit (Table 1).

TABLE 1.

Comparison of patients with persistent and resolved sesame allergy.

Number of patients, (%)	Persistent sesame allergy	Resolved sesame allergy	p value
	69 (82.1)	15 (17.9)
Male, n (%)	50 (72.4)	11 (73.3)	.945
Diagnosed with			.276
Atopic dermatitis, n (%)	29 (42.0)	3 (20.0)
Reaction to sesame, n (%)	31 (44.9)	9 (60.0)
Reaction to other foods, n (%)	9 (13.0)	3 (20.0)
Coexisting food allergies, n (%)	50 (72.4)	10 (66.6)	.652
Current asthma, n (%)	41 (59.4)	7 (46.6)	.366
Ever atopic dermatitis, n (%)	56 (81.1)	13 (86.6)	1.000
Current allergic rhinitis, n (%)	33 (47.8)	4 (26.6)	.161
Parental atopic history, n (%)	45 (65.2)	6 (40.0)	.070
Number of reactions with sesame, median (IQR)	3.0 (2.0–4.0)	2.0 (2.0–2.0)	.001
Anaphylaxis with sesame, n (%)	37 (53.6)	2 (13.3)	.005
Ever adrenaline administration related to sesame allergy, n (%)	26 (37.6)	1 (6.6)	.030
Grade of systemic reaction (most severe experienced during follow‐up period) ≥ Grade 2, n (%)	44 (63.7)	5 (33.3)	.030
Follow‐up time (mo), median (IQR)	58.0 (47.0–83.5)	54.0 (36.0–60.0)	.195
Age at diagnosis (mo), median (IQR)	9.0 (6.0–12.5)	12.0 (8.0–18.0)	.094
Total IgE at diagnosis (kU/L, ImmunoCAP), n, median (IQR)	43, 225.0 (49.0–399.0)	10, 53.5 (34.5–176.7)	.067
Sesame‐specific IgE at diagnosis (kUA/L), n, median (IQR)	43, 8.5 (1.4–15.0)	9, 1.8 (0.3–9.8)	.056
Sesame‐specific IgE / total IgE at diagnosis (%), n, median (IQR)	43, 4.1 (1.1–7.8)	9, 1.0 (0.4–4.9)	.113
Skin prick test with commercial sesame extract at diagnosis (mm), n, median (IQR)	52, 6.7 (5.0–8.5)	14, 5.5 (4.1–6.1)	.050
Age at first reaction (mo), n, median (IQR)	45, 24.0 (11.0–36.0)	15, 18.0 (12.0–24.0)	1.000
Total IgE at first reaction (kU/L, ImmunoCAP), n, median (IQR)	38, 290.0 (102.7–438.7)	13, 54.0 (25.2–181.0)	.004
Sesame‐specific IgE at first reaction (kUA/L), n, median (IQR)	38, 6.2 (2.0–15.7)	12, 1.5 (0.3–3.2)	.009
Sesame‐specific IgE/total IgE at first reaction (%), n, median (IQR)	38, 2.4 (0.8–5.5)	12, 1.4 (0.7–3.9)	.481
Skin prick test with commercial sesame extract at first reaction (mm), n, median (IQR)	38, 8.2 (6.7–10.2)	15, 6.0 (5.5–7.0)	.001
Skin prick test with tahini at first reaction (mm), n, median (IQR)	27, 10.0 (9.0–13.0)	8, 7.5 (5.2–10.5)	.032
Age at final visit (mo), n, median (IQR)	69, 71.0 (58.0–96.0)	15, 65.0 (48.0–82.0)	.416
Total IgE at final visit (kU/L, ImmunoCAP), n, median (IQR)	68, 371.5 (204.0–921.0)	15, 111.0 (21.0–483.0)	.013
Sesame spesific IgE at final visit (kUA/L), n, median (IQR)	67, 6.2 (2.5–26.0)	15, 0.5 (0.1–2.1)	<.001
Sesame‐specific IgE/total IgE at final visit (%), n, median (IQR)	67, 1.8 (0.6–5.6)	15, 0.7 (0.2–0.9)	.002
Total IgE at final visit (kU/L, multiplex array), n, median (IQR)	28, 232.5 (122.8–719.8)	3, 158.1 (20.0‐N/A)	.124
Ses i 1 at final visit (kUA/L, multiplex array), n, median (IQR)	28, 10.4 (3.5–28.6)	3, 0.6 (0.5‐N/A)	.009
Ses i 1/total IgE at final visit (%), n, median (IQR)	28, 2.9 (1.4–6.9)	3, 0.4 (0.1‐N/A)	.140
Skin prick test with commercial sesame extract at final visit (mm), n, median (IQR)	67, 9.0 (6.5–11.0)	15, 3.0 (3.0–5.0)	<.001
Skin prick test with tahini at final visit (mm), n, median (IQR)	60, 10.0 (7.6–14.3)	12, 4.0 (3.2–6.1)	<.001

Note: The chi‐square or Fisher's exact test were used to determine the differences in nominal variables. The non‐parametric Mann–Whitney U test was used to compare numerical data between the groups with persistent and resolved sesame allergy. Bold values indicate statistical significance with a p‐value of < .05.

Abbreviations: IgE, immunoglobulin E; IQR, interquartile range; mo, month.

In multivariate analysis, incorporating age at diagnosis, CSE SPT and ssIgE at first reaction, concomitant FAs, parental atopic history, and grade of the most severe reaction: CSE SPT and the presence of any reaction with grade≥2 emerged as risk factors for PSA (Table 2). A grade 2 or higher severe reaction was linked to a 19.9‐fold heightened risk of PSA (95% confidence interval [CI]:1.37–289.13, p = .02). A 1‐mm increase in the CSE SPT at the first reaction was associated with a 1.7‐fold escalation in the risk of PSA (95% CI:1.05–3.04, p = .03).

TABLE 2.

Predictive risk factors for persistent sesame allergy.

	Multivariate
	VIF	aOR	95% Cl	p value
SPT with commercial sesame extract at first reaction (mm)	1.14	1.79	1.05–3.04	.03
Grade of the allergic reaction ≥2	1.38	19.93	1.37–289.13	.02
Age at diagnosis (mo)	1.06	0.89	0.77–1.03	.12
Sesame‐specific IgE at first reaction (kUA/L, log‐transformed)	1.46	0.91	0.41–2.05	.83
Parental atopic history	1.28	4.90	0.56–43.02	.15
Coexisting food allergies	1.30	0.55	0.05–6.06	.62

Note: Potential predictive risk factors for persistent sesame allergy were evaluated using multivariable logistic regression analysis forced entry method. Log‐transformed sesame‐specific IgE levels were used for logistic regression analysis. Bold values indicate statistical significance with a p‐value of < .05.

Abbreviations: aOR, adjusted odds ratio; CI, confidence interval; IgE, immunoglobulin E; mo, month; SPT, skin prick test; VIF, variance inflation factor.

Furthermore, the proportional hazards regression model revealed that CSE SPT level at the first reaction is a significant predictor of persistence throughout the follow‐up period from the first sesame‐related reaction (Table S3, Figure 3). Patients with SPT results greater than 6.7 mm at the first reaction had a 3‐fold higher probability of PSA compared with those with results less than 6.7 mm (HR = 3.08; 95% CI = 1.17–8.12; p = .02). Cut‐off values were established using ROC analysis, in which the diagnostic tests at the first reaction were evaluated to predict the final reaction (Figure S3). Grade of allergic reaction ≥2 and ssIgE levels at first reaction were not associated with persistence according to proportional hazards regression analysis (Table S3).

FIGURE 3.

The probability of persistent sesame allergy according to skin prick test results at the first reaction. CSE, commercial sesame extract. Kaplan–Meier analysis was performed to demonstrate resolution of sesame allergy based on the skin prick test results with CSE at the first reaction over the follow‐up period since the first reaction.

3.4. Alterations in SPT and sesame‐specific IgE results

The CSE SPT wheal size significantly decreased at the final visit in patients who developed tolerance compared with their first reaction (p < .001). The median change in CSE SPT results from the first reaction to the final visit was a 0.8 mm increase in children with PSA and a 3 mm decrease in children with RSA (Figure 4). Among the patients whose SPT results decreased between their first reaction and final visit (n = 38), 63% remained allergic to sesame. Conversely, among those with no change and increased SPT results (n = 3 and n = 12, respectively), 93% remained allergic to sesame. While ssIgE levels significantly decreased in patients with RSA from the first reaction to the final visit (p < .05), no significant change was observed in the ssIgE levels in children with PSA (p = .22) (Figure 4). Among the patients with decreased ssIgE levels between their first reaction and final visit (n = 28), 61% remained allergic to sesame. Conversely, among those with no change and increased ssIgE levels (n = 2 and n = 19, respectively), 95% remained allergic to sesame.

FIGURE 4.

The alteration in median (IQR) sesame sIgE levels (A) and CSE SPT results (B) between the first reaction and final visit in patients with resolved and persistent sesame allergy. CSE, commercial sesame extract; FR, first reaction; FV, final visit; sIgE, specific immunoglobulin E. The Wilcoxon signed‐rank test was used to compare CSE SPT and ImmunoCAP ssIgE results between the first reaction and final visit.

At the final visit, patients with PSA had significantly higher median (IQR) levels of SPT results with CSE and tahini, as well as significantly higher median (IQR) levels of ssIgE, ssIgE/tIgE ratio and Ses i 1 sIgE than those with RSA (Table 1; Figure S4). ROC curve analysis demonstrated that the SPT wheal size for CSE and tahini and the ssIgE levels significantly predicted persistence (p < .001 for each). The optimal cutoff values for persistence were determined to be 5.7 mm for CSE SPT (area under the curve [AUC] = 0.93, 95% CI:0.86–0.99, sensitivity = 79.3% specificity = 91.7%), 6.7 mm for tahini SPT (AUC = 0.92, 95% CI:0.84–1.00, sensitivity = 94.5%, specificity = 84.3%), and 2.1 kU_A/L for ssIgE (AUC = 0.84, 95% CI:0.72–0.96, sensitivity = 82.8%, specificity = 75.0%) (Figure S5).

4. DISCUSSION

A noteworthy characteristic of SA is its common association with anaphylaxis, ¹ , ²⁵ and SA is a significant risk factor for severe allergic reactions. ²⁶ Consistently, approximately half of our patients had a history of anaphylaxis to sesame throughout the follow‐up periods. Furthermore, among 51 patients, 56.8% were admitted to the emergency department for their first reaction, and 17% were administered adrenaline. A previous study at our center identified sesame as the fourth most common food causing severe anaphylaxis, at a rate of 5.6%. ²⁷ A recent Canadian study reported that sesame accounted for 4% of food‐induced anaphylaxis in children, with 66.2% of the reactions classified as moderate to severe. ²⁸ These findings emphasize the crucial nature of SA, which is increasingly prevalent, ¹ and its considerable potential burden on the healthcare system.

Skin signs and symptoms were the most prevalent IgE‐mediated symptoms in our study, consistent with a questionnaire‐based U.S. study. ⁴ This U.S. study reported a lower incidence of gastrointestinal symptoms in SA than in other common FAs (reported vomiting: 8.3% vs. 33.4%). ⁴ However, our study revealed a 32% prevalence of gastrointestinal reactions for the first and 39% for the last reaction, similar to the results for the other most common food allergens in the study by Warren et al. ⁴ No significant differences were observed in the rates of severe reactions, including cardiovascular and respiratory symptoms, between the first and last reactions.

Our study revealed an 18% resolution proportion at a median age of 5.4 years. In Israel, two studies involving 45 and 30 pediatric patients with confirmed SA reported resolution proportions of 20% and 30%, respectively, at a mean age of 4.4 and 2.8 years. ¹⁵ , ²⁹ Another study conducted in the same region in 190 children reported a resolution proportion of 32% at a mean age of 3.5 years. ¹⁴ A multicenter French study involving a small pediatric group (n = 14) reported a resolution proportion of 21%. ³⁰ Among these studies, Aaronov et al. ²⁹ assessed persistence using a questionnaire. However, in our study, PSA was determined based on a positive OFC or a physician‐diagnosed sesame‐related reaction. Studies by Mahlab‐Guri ¹⁴ and Cohen et al. ¹⁵ focused on resolution, but persistence was not explicitly defined. Our study's relatively low resolution proportion might be attributed to our center's tertiary referral structure. When presenting retrospective real‐life data, one must consider that the patients' parents and doctors might have avoided OFC because of various factors (such as a high level of sensitization or a previous severe reaction), potentially underestimating the true prevalence of PSA. Obtaining the actual proportion of resolution requires data on confirmed cases based on objective clinical reactions and positive diagnostic tests, which were the conditions fulfilled in our study. We also demonstrated that the likelihood of tolerance decreased during follow‐up, suggesting persistence into adulthood. However, the successful completion of the OFC at 18 years of age in one of our patients underscores the importance of ongoing tolerance monitoring. Neverthless, the patient might have outgrown the SA earlier, but this was detected at the age of 18 years because of the retrospective design of the study. In addition, families' early preferences for the OFC of high‐nutrient foods such as milk, eggs, and wheat might have resulted in the detection of RSA from the age of 3 years despite earlier recovery.

Different clinical and allergen sensitization characteristics have been identified as persistent FA (PFA) predictors. ³¹ SPT results and food‐specific IgE levels are commonly used assays for detecting allergen sensitization, with elevated levels in these tests serving as biomarkers of PFA. ¹² , ³¹ However, the data on PSA are limited. A study conducted in Israel reported that 15% of patients with SA exhibited a negative SPT with CSE during their first reaction, with only the tahini SPT results showing a significant difference between persistent and tolerant patients. ¹⁴ The authors noted that natural tolerance developed in 43% of individuals with a CSE SPT level under 7 mm in their first reaction. ¹⁴ Serological tests could not be evaluated in this study because of the limited number of patients with ssIgE measurements. ¹⁴ A previous study with a small cohort of patients (n = 45) demonstrated no difference in SPT results between children with PSA and RSA. ¹⁵

In our study, patients with PSA had significantly higher levels of SPT with CSE and ssIgE as measured by ImmunoCAP than those with RSA at their first reaction. However, the SPT with CSE alone has been identified as an independent risk factor for predicting PSA. This outcome contrasts with that of previous studies, indicating that serum‐specific IgE levels can serve as potential biomarkers for persistent milk, egg, and peanut allergies. ³² , ³³ , ³⁴ Notably, Peters et al. found that high levels of specific IgE to peanuts predicted persistent allergy, whereas Ara h 2 peanut allergen levels did not. ³⁴ Previous studies have shown that the SPT is a more accurate predictor of SA than ssIgE. ¹⁹ , ³⁵ Our results support these findings for PSA. To the best of our knowledge, this is the first study to examine the significance of ssIgE in PSA. Further studies with diverse populations and allergen molecules are needed to clarify our findings.

One of the clinical risk factors identified for PFA is the age at allergy diagnosis. ³¹ An early‐onset allergy has been associated with an increased likelihood of resolution for SA ¹⁴ but persistence for milk allergy. ³⁶ In our study, SA diagnosis involved three methods, revealing variations in the age of the first reaction across diagnostic subgroups. Patients in the AD and other FA subgroups had their first sesame‐related reaction at a significantly older age (median age, 3 years) than those diagnosed after a direct sesame reaction (median age, 1 year). It is important to consider the potential emergence of allergic reactions after initiating elimination diets for patients in the AD group. ³⁷ , ³⁸ , ³⁹ Therefore, thoroughly evaluating the relationship between specific foods and AD is crucial to avoid unnecessary elimination diets. The median age at SA diagnosis is reported to be 1 year in Israel ¹⁵ and 3.5 years in the U.S. ⁴ In our study, the median age of the first reaction was 1 year in the reactive subgroup, indicating that sesame products are incorporated into the diet of our population at an early age, reflecting the influence of Middle Eastern culture. ⁴⁰

We identified a grade 2 or higher reaction as a risk factor for PSA in logistic regression analysis, but Cox regression analysis did not confirm this finding. Mahlab‐Guri et al. observed higher resolution rates in patients with mild reactions. ¹⁴ A U.S. study encompassing the eight most common FAs, including sesame, revealed that children with PFA had a higher rate of severe reactions to relevant foods than those who outgrew the allergy. ¹³ However, no significant differences were observed in the logistic regression model. ¹³ Further studies are needed to elucidate the immune mechanisms underlying these observations.

Coexisting FAs, particularly to tree nuts, have been reported as risk factors for PSA. ¹³ , ¹⁴ However, we did not identify coexisting food and nut allergies as risk factors for PSA. Traditionally, children allergic to peanuts or a single tree nut in early childhood develop multiple allergies to nuts and seeds over time. ³¹ , ⁴¹ In a multicenter study, coexistent nut allergies were reported in 60% of patients with SA, ⁴¹ and similarly, 58% of our patients with PSA had nut allergies. The coexistence of nut and sesame allergies, which tend to be persistent, highlights the importance of FA prevention strategies. A United Kingdom‐Israeli questionnaire‐based study revealed a higher prevalence of sesame and peanut allergies among Jewish children in the United Kingdom than in Israel, possibly related to the earlier introduction of sesame and peanuts into the diet in Israel. ⁴² Longitudinal studies are required to assess the influence of the early introduction of nuts and sesame into the diet to prevent potential multiple nut and sesame seed allergies.

Frequently, the decreasing levels of allergy tests align with a clinical scenario indicating the resolution of allergies. ¹² However, our findings indicate that increased SPT and ssIgE levels are stronger markers for predicting PSA, whereas decreased levels have limited predictive value for tolerance development. Based on the final visit, ROC analysis revealed optimal thresholds for CSE and tahini SPT to predict sesame‐related reactions: 5.7 mm (sensitivity: 79%, specificity: 91%) and 6.7 mm (sensitivity: 94%, specificity: 83%), respectively. For ssIgE, the determined threshold was 2.1 kU_A/L with 83% sensitivity and 75% specificity. This aligns with a study on a limited number of tolerant patients (n = 5), suggesting a threshold of 2.0 kU_A/L for considering tolerance development. ⁴³ Although our results provide valuable insights into distinguishing persistence, it is crucial to note that SPT and ssIgE thresholds can be influenced by factors such as age, population, and geographic region. ⁴⁴

The strengths of our study include its conduction in a diverse population, being the first to evaluate ssIgE levels as a potential risk factor for PSA, assessing alterations and final results in ssIgE and SPT levels to predict persistence and resolution, and determining the first and last sesame‐related reactions. However, major limitations include the retrospective nature of the study, which involved missing data, unbalanced sample sizes, and within‐group variations, all of which might have significantly affected the analysis. Goldberg et al. determined that using Ses i 1 IgE and basophil activation test together provided a strong predictive value for the diagnosis of SA. ⁴⁵ However, we were unable to evaluate Ses i 1 IgE levels as a potential predictive risk factor for PSA, as our patients did not have recorded Ses i 1 IgE levels at the time of first reaction or diagnosis. Our study relied on history‐based reactions and the use of open‐label OFCs, and not all of the first and last reactions observed in our patient cohort were confirmed by OFC. However, this mirrors real‐life scenarios where tolerance is not universally confirmed through OFCs, and only consistent reactions confirmed by physicians are considered. Open‐label OFCs, which are more prevalent in clinical practice, ⁴⁶ were used given the infrequency of physicosomatic reactions during early childhood. We were unable to assess certain previously documented risk factors, such as eczema severity or concurrent food sensitization at diagnosis, as in some previous studies. This limitation could be attributed to potential type 2 errors owing to the moderate size of the study population.

In conclusion, the proportion of persistence in children with confirmed SA was 82%. The independent risk factors for PSA included the severity of the reaction and SPT results. SPT results that were greater than 6.7 mm were associated with an increased likelihood of persistence during the follow‐up period. In addition, increasing levels of sensitizations served as better predictors of persistence than decreasing levels in predicting resolution. Current evidence indicates that sesame OIT is effective and safe and should be considered a viable treatment option for SA. ¹⁰ , ⁴⁷ Utilizing predictive risk factors for PSA may facilitate the optimization of healthcare and the selection of candidates for early OIT, which can ultimately enhance the quality of life of patients and their parents.

AUTHOR CONTRIBUTIONS

Alp Kazancioglu: Conceptualization; writing – original draft; methodology. Melike Ocak: Writing – review and editing. Umit Murat Sahiner: Writing – review and editing. Ozge Soyer: Writing – review and editing. Bulent Enis Sekerel: Conceptualization; supervision; writing – review and editing; visualization; methodology.

FUNDING INFORMATION

This study was not supported by any sponsor or funder.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest in relation to this study.

Supporting information

Data S1.

Kazancioglu A, Ocak M, Sahiner UM, Soyer O, Sekerel BE. Natural history of sesame allergy in pediatric patients: Insight from a retrospective analysis. Pediatr Allergy Immunol. 2025;36:e70022. doi: 10.1111/pai.70022