<scp>IgG</scp> ‐Related Neutrophil Activation Is a Novel Mechanism That Defines a New Type of Immediate Hypersensitivity Reactions ( <scp>IHR</scp> ) to Beta‐Lactam

We appreciate the interest shown by Dr. Blanca and colleagues [1] to our recent publication [2]. We fully agree on the need to accurately phenotype patients in clinical studies of immediate hypersensitivity reactions (IHRs) to beta-lactams (BLs). They typically occur within the first hour but may be observed within six hours after drug exposure [3, 4]. In our study [2], all reactions occurred less than one hour after BL exposure and were typical of IHRs. BLs are the drugs most frequently involved in IgE-mediated IHRs [5] whose diagnosis is based on clinical history, skin testing and, to a lesser extent, in vitro assays of serum-specific IgE (sIgE). When this assessment is negative or inconclusive, a drug provocation test (DPT) is necessary to establish a firm diagnosis [3]. Emerging evidence challenges the univocal role of the IgE-mediated pathogenic mechanism in IHRs, highlighting the need for new biomarkers to investigate the full spectrum of drug-induced hypersensitivity [6]. Our study [2] relied on the careful assessment of 26 cases of BL-IHRs with positive DPTs, no evidence of an IgE-mediated IHR ascertained by negative skin tests (STs), undetectable sIgE and high BL-sIgG. STs are a first-line diagnostic tool in guidelines [3, 4, 7], with sensitivity and negative predictive value higher than those of sIgE assays [3, 7]. Furthermore, serum blocking IgG may cause false-negative results in sIgE assays in individuals with sIgE above 0.20 kUA/L, but not in those with lower levels [8]. In our study [2], most patients had sIgE concentrations below 0.20 kUA/L, indicative of a very marginal influence of blocking IgG. We performed BL-sIgE assays for amoxicillin, ampicillin, penicillin V and penicillin G using the ImmunoCAP, which is commercially available (supplementary table 1 of the study [2]). In all cases, BL-sIgE levels were very low (< 0.35 kUA/L) or undetectable. Blanca et al. [1] questioned the sensitivity of the ImmunoCAP in quantifying BL sIgE. In fact, highly sensitive in-house assays may detect BL-sIgE in subjects who tested negative for commercial assays [9, 10]. However, these in-house techniques are neither standardised nor indicated for routine diagnostics [7, 11]. As pointed out by Blanca et al. [1], initially negative diagnostic tests may become positive after a DPT. In our study [2], STs were repeated 4 weeks after positive DPTs and remained negative. The specific circumstances in which a high level of BL-sIgG may be clinically relevant remain unclear. A previous experimental study [12] demonstrated that sIgG can compete with sIgE for binding to the Fc receptor and trigger anaphylaxis through IgG-mediated activation of effector cells such as neutrophils and monocytes. No sIgG difference has been reported between BL-allergic patients with urticaria and those with anaphylaxis, and BL-sIgG can be detected in exposed tolerant individuals [13, 14]. However, ST-negative and DPT-positive patients present levels of BL-sIgG higher than those of ST-positive patients, suggesting a possible pathogenic role of sIgG in some cases [13]. In addition to the traditional Gell and Coombs classification of IHR (type I to IV) and the extended classifications of non-immediate reactions commented by Blanca and colleagues [1], recent precision medicine approaches have led to the development of new classifications based on endotypes and phenotypes [15, 16]. These new classifications distinguish IHRs caused by an IgE/IgG specific immune response from those caused by a non-antigenic response, such as the direct interaction with the mas-related G-protein coupled receptor member X2 (MRGPRX2), the complement activation (CARPA), the inhibition of cyclooxygenase-1 or the cytokine release reactions (CRRs). Our data on patients with positive DPTs suggest adding the reactions related to IgG-dependent activation of neutrophils as a new subset in the revised classification of immediate allergic reactions to drugs and elastase and other components of neutrophil extracellular traps (NETs) as new biomarkers of this specific endotype. The IgE-independent mechanism was ascertained in most cases by negative STs, low or undetectable sIgE and limited increase of tryptase. The involvement of neutrophil activation was evidenced by the increased release of elastase and other components of NETs. CXCR2, a chemokine involved in neutrophil activation, was highly correlated with neutrophil elastase, but we found no increase in cytokines involved in CRR endotype. In our opinion, the evaluation of patients with perioperative reactions produces uncertain conclusions that do not allow the identification of new endotypes of IHRs, as they may be influenced by conditions other than the culprit drug exposure, such as the administration of other drugs, perioperative stress, pathological background of the surgery etc. Compared to our patients positive to DPTs and negative STs, our 8 cases with perioperative IHRs had more severe reactions, lower levels of elastase, higher levels of tryptase and positive STs, despite low sIgE and high sIgG. A previous study [17] reported increased markers of neutrophil activation in a series of patients with perioperative IHRs against neuromuscular blocking agents (NMBAs). We disagree with Blanca et al. [1] to consider that our study corroborated the results of the aforementioned study [17], where cases had high sIgG and sIgE against NMBA, but were not assessed by STs, making the discrimination of cases with and without IgE-mediated IHRs very uncertain. Indeed, NMBA-sIgE can be frequently detected despite negative STs, as they reflect the immune sensitization rather than mast cell and basophil activation, as demonstrated by two other population studies from our group [18, 19]. Surprisingly, only 28% of patients with perioperative anaphylaxis had detectable NMBA-sIgE, a percentage considerably lower than the 88%–94% reported in previous studies, in which the level of NMBA sIgE was determined by a more sensitive assay, called QAS-IgE [17, 18, 20]. NMBA-sIgG correlated closely with NMBA-sIgE and had concentrations above 0.20 kUA/L, suggesting a possible influence of blocking IgG antibodies on IgE binding and mast-cell reactivity [8]. These considerations suggest that there was no evidence of neutrophil activation as a specific endotype independent from IgE-mediated mechanisms in the two series of subjects. In conclusion, IgG-related neutrophil activation is a novel specific pathogenic mechanism evidenced in our cases with positive DPTs despite negative STs and BL sIgE and no elevation of tryptase, but elevated elastase and other components of NETs. We propose to include it as a new endotype of BL IHRs in the revised classification of immediate allergic reactions. The use of biomarkers of neutrophil-driven responses may be clinically relevant in DPTs, in particular when conventional IgE-mediated mechanisms cannot be demonstrated. J.-L.G. main drafting of the manuscript. A.R. and J.-J.L. drafting of specific parts. R.-M.G.-R. and A.G. writing – review and editing. All authors Critical revision of the manuscript. The study was funded by the EU EuroNanoMed research project DrNanoDall, the research project FHU ARRIMAGE and the French PIA project “Lorraine Université d'Excellence”, reference ANR-15-IDEX-04-LUE. This work was also supported by grants from the Instituto de Salud Carlos III (PI21/00969), FEDER Thematic Networks and Cooperative Research Centres RETICS ARADyAL (RD16/0006/0033) and RICORS Red de Enfermedades Inflamatorias (REI) (RD21/0002/0008). Grants from SEAIC, Luis Alvarez 2022 IDIPAZ Foundations and the XV Call for Research Project Development Grants from the Alfonso X el Sabio University Foundation. This work was supported by the European Commission, Agence Nationale de la Recherche (ANR-15-IDEX-04-LUE), Instituto de Salud Carlos III (PI21/00969), European Regional Development Fund (RD16/0006/0033, RD21/0002/0008). The authors declare no conflicts of interest. Data are available upon request to the corresponding author.