Allergy to PEG (polyethylene glycol) – sensitivity of basophil activation test with COVID-19 mRNA-vaccine BNT162B2

ABSTRACT

Basophil activation test (BAT) with COVID-19 mRNA vaccine seems particularly suitable for detecting sensitization to polyethylene glycol (PEG) in patients with PEG allergy. It was the aim of this study to determine the cutoffs for BAT using BNT162B2 (Comirnaty®) in a larger group of PEG allergic patients and controls. 10 PEG allergic patients and 10 controls were studied. BAT was performed using anti-CCR3 for basophil identification and anti-CD63 to assess basophil activation. Incubations with BNT162B2 at four different concentrations were performed. Basophil activation was significantly higher in PEG allergic patients compared to controls at the higher concentrations used. ROC curves showed best results with a sensitivity of 60% and specificity of 100% with a cutoff of 5% CD63⁺ basophils at a concentration of 4.5 µg/ml. Controls showed no positive results. In our group of PEG allergic patients, a concentration of 4.5 µg/ml BNT162B2 with a cutoff of 5% CD63⁺ basophils was the most suitable condition for identifying patients with a sensitization to PEG. Allergological work-up of PEG allergic patients including BAT with PEGylated lipid nanoparticles might play a role in the future when these substances will be used for other vaccines and cancer immunotherapies.

KEYWORDS:

• Basophil activation test
• polyethylene glycol allergy
• BNT162B2
• PEGylated lipid nanoparticles
• CD63

Since the beginning of the COVID-19 vaccination campaign, there have been discussions about the extent to which COVID-19 vaccines can trigger allergic reactions. To date, mRNA COVID-19 vaccine reactions have not been proven to be IgE-mediated and the mechanisms are still not entirely clear. Since polyethylene glycol (PEG), which is found on PEGylated lipid nanoparticles (PEG-LNP) of the mRNA vaccines, was suspected as a trigger from the beginning, investigations were carried out especially in PEG allergic patients. Positive results were obtained here, which finally supported an IgE-mediated mechanism of PEG allergy. Since the basophil activation test (BAT) is suitable as an in vitro test for detecting IgE-mediated drug allergies, it was used in allergology work-up and interestingly showed the highest activation values with a mRNA vaccine (BNT162B2, Comirnaty®) compared to other PEG-containing drugs in this group of patients.^1,2 BAT with a mRNA vaccine thus seems particularly suitable for detecting sensitization to PEG in patients with PEG allergy. However, thresholds for a positive result were not always specified or used similarly to other drug allergies because the number of patients investigated was small.^2,3 This study aimed to determine the cutoffs for BAT using BNT162B2 in a larger group of PEG allergic patients and controls.

10 PEG allergic patients (2 males, 8 females, 28 to 79 years) presenting in the year 2021 and 2022 for allergological work up and 10 controls (4 males, 6 females, 23 to 62 years) were studied (Tables 1 and 2).

Table 1. Characteristics and results of basophil activation test in PEG allergic patients.

Patient No. (age, sex)	Pat. 1 (73 y, male)	Pat. 2 (77 y, female)	Pat. 3 (59 y, female)	Pat. 4 (28 y, female)	Pat. 5 (59 y, female)	Pat. 6 (43 y, female)	Pat. 7 (66 y, female)	Pat. 8 (47 y, male)	Pat. 9 (79 y, female)	Pat. 10 (40 y, female)
History Skin test results	Year 2018: 30 min after taking a colon cleansing solution (contains PEG 3350) palmar and plantar itching and erythema, generalized itch, dizziness	Year 2013: Anaphylactic reaction to Multinorm® Husten-Brauseta-bletten® (contain macrogol)	February 2021: Minutes after Moviprep® (contains PEG 3350) swelling of back of hands, generalized rash, vomiting, nausea, diarrhea	Year 2015: After intake of WickMediNait® (contains PEG 6000) respiratory distress, generalized rash Year 2016: After intake of Movicol® (contains PEG 3350) generalized swelling and rash, loss of consciousness	Year 2002: Traumeel® and Zeel® ➔ unconscious-ness Year 2005: Triamcinolone, prednisolone ➔ anaphylactic reaction grade III-IV Year 2008: Wick MediNait® (contains PEG 6000) ➔ anaphylactic reaction	May 2021: Immediately after intake (150 ml) of PLENVU® (contains PEG 3350) before colonoscopy feeling of heat, tingling, constriction of food, trachea, shortness of breath, palpitations, dizziness, vomiting	July 2004: Mecaine and triamcinolone: After 10 min urticaria, dizziness, loss of consciousness June 2005: Klean-Prep® (contains PEG 3350): After 15 min dizziness, nausea, circulation problems	Several weeks before (2021): Intake of Moviprep® (contains PEG 3350): After 30 min severe agitation, flushing, repeated vomiting, itching	April 2015: Paclitaxel (contains PEG): Problems with circulation, clouding of conscious-ness, abdominal cramps, skin rash. Year 2015: After intake of Moviprep® (contains PEG 3350): urticaria Year 2016: Triamhexal®: hypertension, itching, heat sensation	Year 2013: After intake of Moviprep® (contains PEG 3350): Urticaria, rhinitis
	Positive: Spt: PEG 3350, 4000 IDT: PEG 3350, 4000, 6000, Comirnaty®; after IDT palmar and plantar erythema, generalized itch	Positive in the year 2013: Macrogol (PEG 3350) (Negative in the year 2021: All tested PEGs and PEG-containing vaccines)	Positive: SPT: PEG 4000 1%, PEG 4000 10%, PEG 6000 10%, dizziness 15 minutes after skin test	Positive in the year 2017: Macrogol (PEG 3350) Year 2021: IDT: Comirnaty® (SPT: Negative: All tested PEGs and PEG-containing vaccines)	Positive before year 2021: PEG 6000 Positive in the year 2021: IDT: Comirnaty®, Spikevax® (SPT: Negative: All tested PEGs and PEG-containing vaccines)	Positive: IDT: PEG 6000 0.01% and 0.1% (SPT: Negative: All tested PEGs and PEG-containing vaccines)	Positive: SPT: PEG 2000	Positive: IDT: Comirnaty® 10%, Spikevax® 10% (SPT: Negative: All tested PEGs and PEG-containing vaccines)	Positive: IDT: Comirnaty® 0.1%, 0.01% (SPT: Negative: All tested PEGs and PEG-containing vaccines)	Positive: IDT: Comirnaty® 10%, Spikevax® 10% (SPT: Negative: All tested PEGs and PEG-containing vaccines)
Provocation tests	Provocation test not performed	Provocation test not performed	Provocation test not performed	Provocation test not performed	Provocation test not performed	Provocation test not performed	35 min after intake of PEG 2000 generalized itching, urticaria	Provocation test not performed	After intake of PEG 4000: Generalized itching After intake of PEG 3350: Tingling of fingertips, urticaria, abdominal cramps	Provocation test not performed
COVID-19 vaccination	1^st (fractio-nated) and 2^nd vaccination with Vaxzevria® (contains no PEG) tolerated	unknown	Vaccination with COVID-19 Vaccine Janssen® (contains no PEG) tolerated	Fractionated vaccination with Vaxzevria® (contains no PEG) tolerated	Vaccination with COVID-19 Vaccine Janssen® (contains no PEG) tolerated	Vaccination with Spikevax® (contains PEG) and Vaxzevria® (contains no PEG) tolerated	Fractionated vaccination with Spikevax® (contains PEG) tolerated	Vaccination with Spikevax® (contains PEG) tolerated	Vaccination with Vaxzevria® (contains no PEG) tolerated	unknown
Basophil activation test (% CD63^+ basophils)
Negative control	1.6	4.5	3.2	0.2	0.8	1.8	0.6	1.6	2.4	0
Positive control 1 (anti-FcεRI)	90.0	90.4	16.0	90.2	80.6	96.4	61.8	89.2	91.8	34.8
Positive control 2 (fMLP)	23.2	29.2	14.4	43.7	73.5	42.0	22.5	2.2	43.8	15.8
Comirnaty®, BNT162B2
22.5 µg/ml	20.0	9.4	5.6	17.4	4.2	1.3	1.4	2.8	30.2	0.3
4.5 µg/ml	34.4	16.9	7.8	37.4	9.8	4.1	1.3	3.0	37.6	0.6
0.9 µg/ml	18.4	21.3	7.4	6.6	4.0	2.5	0.3	5.0	37.6	0.7
0.18 µg/ml	2.2	n.d.	3.6	2.0	2.0	2.2	0.3	2.0	34.8	0.6

IDT, intradermal test; n.d. = not done; SPT, skin prick test.

Table 2. Characteristics and results of basophil activation test in controls.

Control No. (age, sex)	Control 1 (61 y, female)	Control 2 (30 y, male)	Control 3 (29 y, male)	Control 4 (29 y, female)	Control 5 (59 y, female)	Control 6 (26 y, female)	Control 7 (23 y, male)	Control 8 (26 y, male)	Control 9 (62 y, female)	Control 10 (56 y, female)
History	Vaccinated with Comirnaty®, COVID-19 infection	No vaccination, COVID-19 infection	Vaccinated with Comirnaty®, COVID-19 infection	Vaccinated with Comirnaty®, COVID-19 infection	Vaccinated with Comirnaty®, COVID-19 infection	No vaccination, COVID-19 infection	Vaccinated with Comirnaty®, COVID-19 infection	Vaccinated with Comirnaty®, COVID-19 infection	Vaccinated with Comirnaty®, no COVID-19 infection	Vaccinated with Comirnaty®, no COVID-19 infection
Basophil activation test (% CD63^+ basophils)
Negative control	0.4	3.9	0.4	0.6	2.9	0.2	0.7	0.7	2.2	1.0
Positive control 1 (anti-FcεRI)	91.2	50.8	73.7	15.8	91.0	78.0	83.3	93.2	90.8	82.2
Positive control 2 (fMLP)	47.7	68.7	44.9	9.5	20.6	2.0	9.7	92.2	4.5	11.8
Comirnaty®, BNT162B2
22.5 µg/ml	1.0	4.4	1.3	1.3	5.8	0.0	0	0.9	2.0	0.6
4.5 µg/ml	1.1	3.8	0.9	0.3	4.8	0.2	0.3	0.5	1.6	1.0
0.9 µg/ml	1.0	3.4	1.4	0.7	4.0	2.4	0.5	0.2	0.3	1.3
0.18 µg/ml	0	4.4	0.3	0.9	4.2	0.4	1.2	0.3	n.d.	3.2

n.d. = not done

The Flow CAST (BÜHLMANN Laboratories AG, Schönenbuch, Switzerland) was used for quantitative measurement of in vitro basophil activation, as previously described⁴: Venous blood was collected in 10-mL EDTA tubes from each subject. For each subject and allergen, polystyrene tubes were prepared with 50 μL of allergen at the defined concentration and diluted in 100 μL of stimulation buffer (containing heparin, Ca²⁺, and IL-3 [2 ng/mL]). An anti-FcεRI mAb and anti – N-formyl-methionyl-leucyl-phenylalanine were used as positive controls. Background values were assessed with 50 μL of stimulation buffer. We added and gently homogenized 50 μL of each subject’s whole blood (with EDTA), added 20 μL of staining reagent (anti-CD63–fluorescein isothiocyanate and anti-CCR3–phycoerythrin mAbs), and mixed and incubated the solution at 37°C for 25 minutes in an incubator. Two milliliters of lysing buffer stopped the stimulation within 5 minutes of incubation in the dark at room temperature, followed by centrifugation of the solution for 5 minutes at 460 g. The supernatant was decanted, and stained cells were washed with 300 μL of washing buffer. Cells were resuspended by means of gentle vortexing and analyzed by using BD FACSCanto II (Becton Dickinson Immunocytometry System; Becton Dickinson, Heidelberg, Germany) with BD FACSDiva software. Basophils were gated as low side scatter CCR3⁺/side scatter^low. Anti-CCR3 was used as a selection marker to separate basophils from other leukocytes. Upregulation of the basophil activation marker CD63 was calculated based on the percentage of CD63⁺ cells compared with the total number of identified basophils. In each assay a minimum of 300 events (ie, basophils) were recorded. Details of the flowcytometric data acquisition can be found in Supplementary Methods and Fig. S1.

BNT162B2 was used at four different end concentrations (0.18 µg/ml; 0.9 µg/ml; 4.5 µg/ml; 22.5 µg/ml corresponding to concentrations after reconstitution of 0.8 µg/ml; 4 µg/ml; 20 µg/ml; 100 µg/ml).

Basophil activation with BNT162B2 (mean ± SD in % CD63⁺ basophils) was significantly higher in PEG allergic patients compared to controls at the higher concentrations used: Concentrations of 0.9 µg/ml/4.5 µg/ml/22.5 µg/ml induced a significant (p < .01) upregulation of 10.95 ± 11.7/15.55 ± 15.26/9.26 ± 10.03 in PEG allergic patients vs. 1.52 ± 1.3/1.46 ± 1.53/1.73 ± 1.90 in controls (Figure 1). ROC curves showed best results with a sensitivity of 60% and specificity of 100% with a cutoff of 5% CD63⁺ basophils at a concentration of 4.5 µg/ml (Figure 2).

Figure 1. Basophil activation with different concentrations of BNT vaccine in PEG allergic patients and controls (*p < .01).

Figure 2. Receiver operator characteristic curve of BNT vaccine (4.5 µg/ml).

In our group of PEG allergic patients, a concentration of 4.5 µg/ml BNT with a cutoff of 5% CD63⁺ basophils was the most suitable condition for identifying patients with a sensitization to PEG. This cutoff was the same or lower than thresholds used in the literature with lower numbers of patients.^1,2 This study also showed that the sensitivity was in a similar range compared to other drugs.³ It remains unclear why the test was negative in some patients despite positive provocation tests (e.g. patient 7). One could assume a decreasing sensitization over time as has also been described for skin tests or a non-IgE-mediated mechanism.⁵

This manuscript only contains data on BAT in PEG-allergic patients, not in patients with proven hypersensitivity reactions to PEG-containing vaccines. Since the vast majority of people who reacted to the first vaccination tolerated the second vaccination, an IgE-mediated allergy to PEG-containing vaccines appears to be extremely rare.⁴ There are some data in the literature that reported hypersensitivity reactions to PEG-containing vaccines are not IgE-mediated and rather pseudo-allergic reactions via the activation of anaphylatoxins C5a. None of these tested patients (n = 5) in this study demonstrated a positive BAT to the BNT162B2 vaccine.⁶

In the manuscript of Labella et al. 5 out of 10 controls with COVID-infection (3 non-vaccinated and 2 vaccinated) showed a positive result in the BAT.⁷ Also other investigators presented results of 6 controls including 3 with a polymerase chain reaction confirmed SARS-CoV-2 infection in the 3 months before sampling with no reactivity to BNT162B2.¹ Our negative results in 8 controls with previous COVID-19 infection (2 non-vaccinated and 6 vaccinated) are consistent with the latter results.

Although PEG allergy has a low prevalence, allergological work-up of PEG allergic patients including BAT with PEG-LNPs might play a role in the future when these substances will be used not only for vaccination, but also for cancer immunotherapies.⁸

Author contribution

BE conceived, designed and supervised the experiments. BE had full access to all of the data in the study, is the guarantors of this study and wrote the paper. SM, UD, TB and KB have made contributions to the allergology work-up of the patients, data interpretation and revision of the manuscript. All authors agree to be accountable for all aspects of the work.

Acknowledgments

We thank Franziska Martin for her excellent technical support.

Disclosure statement

BE reports non-financial support from Bühlmann Laboratories outside the submitted work. KB received lecture fees from ALK, Bencard and ThermoFisher and reports non-financial support from Bühlmann Laboratories outside the submitted work. The rest of the authors declare that they have no relevant conflicts of interest.

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website at https://doi.org/10.1080/21645515.2024.2312600.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.