National assessment strategy for adults and children with a history of vaccine allergy shows low utility of COVID-19 vaccine skin testing

Following initial reports of anaphylaxis linked to the Pfizer/BioNTech COVID-19 vaccine (Pfizer vaccine),1 vaccine excipient polyethylene glycol (PEG) was the suspected candidate causing these reactions. Later on, polysorbate 80 (PS-80) was included due to theoretical concerns about cross-reactivity between PEG and PS-80 contained in viral vector vaccines.2 The UK and USA Medicines and Healthcare Products Regulatory Agency initially recommended halting mRNA vaccinations in patients with “a history of a significant allergic reaction to a vaccine, medicine, or food or who have been advised to carry an adrenaline autoinjector”3 resulting in governments worldwide establishing triage systems to survey vaccine safety. These initial recommendations were extremely broad and were quickly revised. A nationwide strategy in Singapore was put in place to evaluate individuals deemed at risk of vaccine allergy prior to receiving mRNA COVID-19 vaccination through a Ministry of Health (MOH) sanctioned vaccine allergy clinics nationwide. This initially included all patients with a history of anaphylaxis to food or drugs but was later revised to include only those with a history of vaccine allergy. Criteria for referral were any prior vaccine allergy including to non-COVID-19 vaccines. At inception of this referral pathway, there was limited universal experience with skin testing using mRNA-based COVID-19 vaccines, and there remains no consensus among allergists regarding the true sensitivity and specificity of these tests.2, 4 This approach was particularly important for pediatric patients where vaccine hesitancy was greater and prior allergy history was often used to defer vaccination.5 Currently, there are no countries that restrict COVID-19 vaccination for anyone with allergies unrelated to the COVID-19 vaccines. This is a prospective observational study of vigilance undertaken in adults and children who were referred to the Adult and Paediatric Allergy & Immunology units of two tertiary academic medical centers between April 2021 and November 2021 for evaluation of suitability to receive the Pfizer vaccine. This study was approved by the institutional ethics committee (DSRB Reference Number 2021/00375), and informed consent was obtained from all patients. Basic demographics, personal and family allergy history, nature of reported prior vaccine allergy, and tolerance of subsequent vaccines were collected. Patients with histories consistent with an IgE-mediated allergic reaction underwent skin prick (SPT) and intradermal tests (IDT) with the Pfizer vaccine and representative drugs to screen for sensitization to known excipients. SPTs (1:1 concentration) and IDTs (1:100 and 1:10 concentrations) to the Pfizer vaccine, PS-20 (Havrix® Hepatitis A vaccine), PEG-3350/PS-80 (Medroxyprogesterone Acetate), and PS-80 (Optive® Advanced Eyedrops) were performed according to Banerji et al.6 Patients with negative skin tests were offered a two-step graded provocation test to the Pfizer vaccine. The provocation test was performed in two steps: step 1 administered 0.1 mL, step 2 administered the remaining 0.2 mL, with a cumulative dose of 0.3 mL (full vaccine dose). Patients were observed for 1 h in between doses, and for 2 h after the last dose. Patients with positive skin tests were counseled to choose between a two-step provocation test to the Pfizer vaccine or vaccination with Sinovac, the only non-mRNA COVID-19 vaccine alternative available in Singapore at the time. A safety phone call was made 48–72 h after provocation to monitor for any delayed reactions. Of a total of 635 patients referred for evaluation, 43.0% were deemed to be eligible for direct vaccination in the community upon review of the referral information. These included patients whose reaction to the index vaccine were consistent with side effects (e.g., fever and myalgia) rather than allergy. Another 23.1% were cleared by an allergist for vaccination after consultation. These patients had symptoms inconsistent with an allergy or had subsequently tolerated the same vaccine or other vaccines containing the excipients of interest. The remaining 31.2% required closer evaluation (Figure 1). It is important to note that cross-reactivity between PS-80 and PEG is theoretical and remains unproven, thus tolerating a vaccine containing one does not accurately allow us to conclude that the individual would tolerate both. Of the patients undergoing further evaluation, we enrolled 190 individuals (Table 1). Patient age ranged from 12 to 83 years with a median age in the pediatric group (defined as age < 18 years) of 14 years (n = 39) and 46 years in adults (n = 151). Only children 12 years and above were eligible for COVID-19 vaccination during this period. Most patients reported symptoms consistent with a type 1 hypersensitivity reaction to the vaccines, including urticaria or angioedema (56.33%) or anaphylaxis (9%) based on physician reporting on the national electronic database and patient history taken during the specialized allergy consultation. All SPTs to the Pfizer vaccine were negative. Fourteen patients had an immediate positive IDT [1:10 Pfizer vaccine (five adults, three children) and 1:10 PS-80 (five adults, two children)] and three had delayed positive IDTs (48 h after testing). Seven patients had only one positive skin test, while seven others had several positive results. There was no relationship between positive skin tests and family history of allergy, personal medical history, type of reaction to vaccines, or other demographic variables. Patients with negative skin tests (n = 109) were offered provocation tests. Of these, seven patients reported mild reactions (nausea and urticaria being the most common). One patient developed anaphylaxis7 characterized by generalized urticaria, wheezing, and shortness of breath, 5 min after the first dose of the graded challenge, requiring treatment with one dose of intramuscular adrenaline. Eight of the 17 patients with positive skin tests (including seven positives to the Pfizer vaccine) tolerated the provocation tests with the Pfizer vaccine without any immediate or delayed reactions. Seven patients chose non-mRNA alternatives (Sinovac–CoronaVac was the only non-COVID-19 mRNA vaccine available at the time). Two patients deferred COVID-19 vaccination. At a safety phone call made 48–72 h of post vaccination, mild reactions were reported in 13.7% of patients, 18% for children and adults, respectively. No severe reactions were reported in either age group. The proportion of pediatric patients with skin test positivity and tolerance of graded challenge is similar in adults and children. While there was one anaphylactic episode in the adult group, none of the children developed a reaction during graded challenge. To our knowledge, allergy referrals in patients with suspected prior non-COVID-19 vaccine allergies was not routinely done in other countries. We demonstrate that within our study population, (i) skin testing with vaccine and excipients prior to vaccination with mRNA COVID-19 vaccines was not a useful predictor of mRNA vaccine tolerance and provocation tests should be recommended as gold standard; (ii) a history of reported vaccine allergy did not correlate with excipient skin test positivity, even in those with convincing type 1 hypersensitivity reactions, and (iii) a history of prior vaccine allergy did not predict successful vaccination with the mRNA COVID-19 vaccine. Our findings were consistent with previous reports2, 8 demonstrating poor correlation between allergy skin-testing to excipients and mRNA COVID-19 vaccination tolerance. This could be in part due to an irritant concentration of the Optive Advanced eye drops used to test for sensitization to PS-80 and low cross-reactivity between PEG and PS-80. Furthermore, other nonallergic mechanisms could play a role in the vaccine adverse events reported such as complement activation-related pseudo allergic reaction (CARPA). If CARPA is responsible for a subset of the adverse reactions seen, this could explain why most patient with a previous reaction subsequently tolerate the same vaccine. We are limited by the patient-reported nature of the past vaccine allergies, which may inadvertently have included patients with nonallergic reactions. Careful history was taken to exclude patients with inconsistent allergy symptoms. Patients who were cleared for direct vaccination after a review of history alone received their vaccinations at the community vaccine centers and could not be followed up as they were not recruited into the study. However, none of these patients were referred back to us for COVID-19 vaccine-related reactions. Although the Brighton criteria7 has been used to define anaphylaxis in vaccine-related reactions, the criteria has been reevaluated as the inclusion of subjective symptoms could account for higher reports of adverse events following immunization.9 This is not applicable to the anaphylactic reaction within our study as it presented with objective symptoms but could have played a role in overestimating the severity of the initial vaccine allergic reaction for which the patients were referred to the vaccine clinic. Our study shows that prior vaccine allergy history is not a risk factor for vaccination with COVID-19 mRNA vaccines, a concern both among physicians and patients, which is still a cause of hesitancy for vaccination in the general population. The strengths of our study include the prospective recruitment and follow-up of patients undergoing allergy review, work up and provocation tests in the pediatric population, providing a comprehensive overview of the utility of such an allergy referral workflow in predicting COVID-19 mRNA vaccine allergies. Our study is the first to describe a systematic allergy testing approach in subjects under 18 years of age. A previously published series of three patients also demonstrates that children with immediate first-dose reactions to mRNA COVID-19 can safely receive their second injection in a graded challenge without the need for premedication.10 The subjects presented with cutaneous symptoms after vaccination which were managed with oral antihistamines. Contrary to our study, cutaneous allergy tests where not performed; the authors performed basophil activation tests which did not predict the outcome. Another study in adolescents and young adults with a history of acute lymphoblastic leukemia and allergy to PEG-asparaginase retrospectively analyzed data from 32 patients who received the Pfizer-BioNTech vaccine in a specialized vaccination clinic without any allergic reaction.11 No allergy tests or allergy visits were performed for these patients in order not to delay vaccination. Robust data on pediatric population are still lacking, but these studies and our own data show that prevaccination testing is of limited use in this age group. There have been little pediatric data regarding potential vaccine allergy and the utility of testing in children published so far, necessitating extrapolation of adult data to the pediatric context. Our study includes allergy testing in pediatric subjects, a strategy which has not been done in other countries. This initial screening program for adolescents helped reassure the public of the continued safety of COVID-19 vaccination in children. As we demonstrated a low utility of allergy testing both in the adult and adolescent age groups with suspected past vaccine allergies prior to mRNA COVID-19 vaccination, along with rising evidence on the limitations of allergy testing to predict COVID-19 vaccine allergy, government policy on the eligibility for vaccination with mRNA COVID-19 vaccines in these patients was revised,12 allowing them to receive full dose vaccinations without prior testing or graded challenges. When the COVID-19 vaccination drive was opened up to children under the age of 12 years in Singapore, they were able to follow this revised vaccination guidance. Currently, COVID-19 vaccination is freely available in Singapore for all age groups 6 months and above. Referral to the allergy specialist is indicated only in patients with a history of prior reaction to mRNA COVID-19 vaccine or its excipients. Overall, we do not recommend excipient testing in either adult or pediatric patients receiving the mRNA COVID19 vaccination, supporting the validity of common recommendation across the age groups, and we conclude that a history of prior vaccine allergy should not preclude vaccination with the mRNA COVID-19 vaccine. Carmen Riggioni was involved in Conceptualization (lead), Recruitment(equal), Writing-review (lead), editing (equal); Kee Fong Phang: Recruitment(equal), review & editing (equal); Sen Hee Taywas involved in Recruitment(equal), review & editing (equal); Hugo Van Bever was involved in Recruitment(equal), review & editing (equal); Serene Xu was involved in Recruitment(equal), review & editing (equal); Sylvia Oh was involved in Recruitment(equal), review & editing (equal); Angela Bueno was in involved in Recruitment(equal), review & editing (equal); Ruth Chua was involved in Recruitment(equal), review & editing (equal); Si Hui Goh was involved in review & editing (equal), Kok Wee Chong was involved in review & editing (equal), Elizabeth Huiwen Tham was involved in Conceptualization (lead), Recruitment(equal), Writing-review (lead), editing (lead); Amelia Santosa was involved in Conceptualization (lead), Recruitment(equal), Writing-review (lead), editing (lead). All authors have no conflicts of interest to disclose. The peer review history for this article is available at https://publons.com/publon/10.1111/pai.13923. The peer review history for this article is available at https://publons.com/publon/10.1111/pai.13923.