Updated threshold dose‐distribution data for sesame

To the Editor, Sesame is classified as a “major” food allergen for which mandatory disclosure is required. Understanding reaction thresholds and how these vary within the allergic population is crucial in providing appropriate dietary advice to patients, providing guidance to the food industry, and informing dosing regimens for oral food challenges (FC). However, the largest data series used to derive a threshold dosedistribution for sesame included blinded challenge data from just 40 individuals.1 Data from lowdose, open FC can be used to supplement that from blinded FC, reducing uncertainty in estimating threshold dosedistributions for allergenic foods which otherwise lack sufficient data.2 We, therefore, undertook a systematic search of the literature and performed dosedistribution modelling of individual patient FC data (including open FC) to update estimated eliciting doses for sesame. Eleven studies were included (Table S1), representing data from 246 positive FC. The discrete and cumulative eliciting dose predicted to provoke reactions in 5% of the sesameallergic population (ED05) were 2.4 (95% CI 1.0– 7.7) and 2.5 (95% CI 0.9– 9.5) mg sesame protein, respectively. Dosedistributions are shown in Figure 1 and Table S1. These estimates are reassuringly similar to those previously reported,1 only with much greater precision reflecting the increased number of datapoints (Table 1). Furthermore, these estimates were robust at sensitivity analyses when excluding data from unblinded food challenges or studies with a significant proportion of “first dose reactors” (Table 1). With this analysis, the dataset for sesame is now similar to that used to inform eliciting doses for other food allergens, and sufficient to inform public policy despite the potential limitations of analyses using FC data.1– 3 The CODEX committee of the Food and Agricultural Organization of the United Nations and the World Health Organization recently commissioned an Expert Consultation which recommended the inclusion of sesame as a global “priority” allergen.4 The data presented here will be used to inform a reference dose which might be recommended to guide the use of precautionary allergen (“may contain”) labelling. Given that ED values remain robust at sensitivity analysis when limited to blinded FC in the ED01ED10 range, we recommend using ED values based on the blinded FC dataset for risk assessment and risk management purposes, to maintain consistency with approaches for other food allergens.5 A strength of this dataset is the inclusion of cohorts spanning four of the six global CODEX regions. These data were mostly generated from FC using ground sesame or tahini and may not be directly


Updated threshold dose-distribution data for sesame
To the Editor, Sesame is classified as a "major" food allergen for which mandatory disclosure is required. Understanding reaction thresholds and how these vary within the allergic population is crucial in providing appropriate dietary advice to patients, providing guidance to the food industry, and informing dosing regimens for oral food challenges (FC). However, the largest data series used to derive a threshold dose-distribution for sesame included blinded challenge data from just 40 individuals. 1 Data from low-dose, open FC can be used to supplement that from blinded FC, reducing uncertainty in estimating threshold dose-distributions for allergenic foods which otherwise lack sufficient data. 2  Eleven studies were included (Table S1), representing data from 246 positive FC. The discrete and cumulative eliciting dose predicted to provoke reactions in 5% of the sesame-allergic population (ED 05 ) were 2.4 (95% CI 1.0-7.7) and 2.5 (95% CI 0.9-9.5) mg sesame protein, respectively. Dose-distributions are shown in Figure 1 and Table S1.
These estimates are reassuringly similar to those previously reported, 1 only with much greater precision reflecting the increased number of datapoints (Table 1). Furthermore, these estimates were robust at sensitivity analyses when excluding data from unblinded food challenges or studies with a significant proportion of "first dose reactors" (Table 1).
With this analysis, the dataset for sesame is now similar to that used to inform eliciting doses for other food allergens, and sufficient to inform public policy despite the potential limitations of analyses using FC data. [1][2][3] The CODEX committee of the Food and Agricultural Organization of the United Nations and the World Health Organization recently commissioned an Expert Consultation which recommended the inclusion of sesame as a global "priority" allergen. 4 The data presented here will be used to inform a reference dose which might be recommended to guide the use of precautionary allergen ("may contain") labelling. Given that ED values remain robust at sensitivity analysis when limited to blinded FC in the ED 01 -ED 10 range, we recommend using ED values based on the blinded FC dataset for risk assessment and risk management purposes, to maintain consistency with approaches for other food allergens. 5 A strength of this dataset is the inclusion of cohorts spanning four of the six global CODEX regions. These data were mostly generated from FC using ground sesame or tahini and may not be directly baked into the surface. 6 This would be equivalent to an ED25 level of exposure, implying tolerance in ~25% of sesame-allergic individuals.
It is, therefore, unclear whether baked sesame seeds are tolerated due to the low level of allergen exposure, the lower bioavailability of sesame seed protein with this form of consumption, or both.
Finally, these data confirm that a semi-log dosing regimen for FC (as recommended by PRACTALL) is appropriate for sesame. Tahini is commonly used for the higher doses used at sesame-FC; however, the strong taste can create difficulties, particularly in younger children. Our data indicate that a top dose of 1 g protein (around 4 g of tahini paste, approximately 1 teaspoon) will cause objective symptoms in ~93% of sesame-allergic individuals (Table S1), and thus, inform the risk of a false negative challenge in someone unable to ingest a higher dose at FC.

FU N D I N G I N FO R M ATI O N
RLP and JK receive research support from the National Health and

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request, but may be subject to non-disclosure agreements.
TA B L E 1 Doses of sesame protein predicted to cause a reaction in 1% (ED 01 ), 5% (ED 05 ) and 10% (ED 10 ) of the sesame-allergic population (together with 95% confidence intervals) calculated using both discrete and cumulative dosing schemes a Note: Discrete dosing schemes are reported as the mg protein amount of each separate dose within a food challenge when determining the individual NOAEL and LOAEL. Cumulative dosing schemes are reported as the cumulative sum of all prior doses within a food challenge when calculating the individual NOAEL and LOAEL. Population dose-distributions were determined using "Stacked Model Averaging" as previously described. E2 a Left-censoring of data occurs when participants react to the first dose of the challenge protocol, and is more likely to occur in those studies with a higher initial challenge dose. All doses are presented as mg sesame protein.