Skip to main content
NCBI home page
Lippincott Open Access logoLink to Lippincott Open Access
. 2025 Jul 10;64(4):380–385. doi: 10.1097/SHK.0000000000002644

CIRCULATING INFLAMMATORY PROTEINS IN DRUG-INDUCED ANAPHYLACTIC SHOCK: EVIDENCE FROM MULTIVARIABLE MENDELIAN RANDOMIZATION AND MULTIOMICS INTEGRATION

Jinwei Dai 1,2, Nianzhe Sun 2,3, Wenye Xu 1,2, Zhihong Zuo 1,2, Xiaoyang Pang 2,4, Zhaoxin Qian 1,2
PMCID: PMC12435250  PMID: 40694027

ABSTRACT

Background: Drug-induced anaphylactic shock (DIAS) remained a critical clinical challenge due to increased drug use and novel hypersensitivity mechanisms. The role of circulating inflammatory proteins in DIAS remained unclear. Methods: We applied multivariable Mendelian randomization (MR) to explore the associations between specific inflammatory proteins and DIAS, drawing on recent findings from genome-wide association studies. Circulating inflammatory protein data were obtained from a cohort of European ancestry comprising 14,824 samples, while data on DIAS were sourced from the FinnGen consortium, including 20,806 cases and 411,845 controls. To strengthen our findings, we conducted complementary analyses such as colocalization (COLOC), enrichment studies, drug screening, and molecular docking. Results: MR analysis identified significant associations between inflammatory proteins and DIAS. CD40L exhibited a protective effect (OR = 0.69, 95% CI: 0.4951–0.9578, P = 0.027) and high colocalization probability (58%). CXCL10 (OR = 1.51, 95% CI: 1.0075–2.2549, P = 0.046) and CCL3 (OR = 2.08, 95% CI: 1.0079–4.3072, P = 0.048) significantly increased risk. Drug screening and enrichment analyses further elucidated underlying molecular mechanisms. Conclusions: This study identified novel associations between inflammatory proteins and the risk of anaphylaxis, providing insights for targeted prediction and therapeutic strategies.

KEYWORDS: Circulating inflammatory proteins, anaphylactic shock, Mendelian randomization, drug prediction, molecular docking

INTRODUCTION

With the continuous rise in global drug usage and the rapid development of novel therapeutics, anaphylactic shock induced by adverse drug reactions remains a formidable clinical challenge (1–3). Drug hypersensitivity reactions are the most common type of allergic response, and since the beginning of the 21st century, reports of such reactions have steadily increased (4). An FDA Adverse Event Reporting System (FAERS) study reported that antibiotics were implicated in 14.87% of all anaphylactic reaction cases and were responsible for 20.84% of all anaphylaxis-related deaths. Notably, from 1999 to 2019, the total number of reported adverse drug events in FAERS was 17,506,002, among which 47,496 cases (0.27%) were identified as anaphylaxis, and 2,984 of these (6.28%) resulted in death (5). However, observational studies investigating the pathophysiological mechanisms underlying drug-induced anaphylactic shock (DIAS) have yielded inconsistent results, largely due to confounding factors and reverse causation that undermined the ability of such studies to accurately reveal true causal relationships (6).

Recent evidence suggests that some circulating inflammatory proteins play a critical role in the pathophysiology of DIAS (7–11). Research indicates that these proteins are not only central to the regulation of immune responses but may also modulate an individual's drug tolerance by influencing the recruitment and activation of immune cells (9,12).

Traditional observational studies have struggled to consistently delineate the relationship between inflammatory proteins and DIAS due to issues like sample heterogeneity, confounding, and reverse causation (13). In contrast, Mendelian randomization (MR) uses single-nucleotide polymorphisms (SNPs) identified from genome-wide association studies (GWAS) as instrumental variables (IVs). These IVs are genetic variants that are robustly associated with the exposure of interest (e.g., circulating protein levels) but are not associated with confounders and influence the outcome only through the exposure. As genetic variants are randomly allocated at conception, MR approximates a “natural randomized trial,” thereby minimizing bias and improving the reliability of causal inference (14). MR has been widely applied to assess causal relationships between exposures and disease outcomes, providing more reliable insights into key pathogenic mechanisms (15). To ensure instrument validity, weak genetic instruments are typically excluded using F-statistics, and horizontal pleiotropy is evaluated using sensitivity analyses such as the MR-Egger intercept and MR-PRESSO methods.

Based on this background, we employ a multivariable Mendelian randomization approach to systematically investigate potential causal associations between inflammatory proteins and the risk of DIAS. Additionally, we integrate colocalization and bioinformatic analyses to further explore underlying biological mechanisms and identify potential therapeutic targets (16). The study aims to provide a comprehensive understanding of inflammatory pathways involved in anaphylactic shock, supporting personalized risk assessment and precision therapeutic strategies.

PATIENTS AND METHODS

Research design

This study was conducted following the Strengthening the Reporting of Observational Studies in Epidemiology for Mendelian (STROBE-MR, http://links.lww.com/SHK/C569) guidelines (17). Based on current evidence from multiple studies, we identified six circulating inflammatory proteins (10,11,18,19). We obtained genetic variants associated with both the exposures and the outcomes from publicly available genome-wide association studies (19). All study populations were of European descent, and there was no sample overlap between the exposure and outcome datasets. Our study design was based on three key assumptions: first, the genetic instruments were strongly associated with the exposures of interest; second, the genetic instruments affected the outcome exclusively through these proteins, without exerting any direct effects on the outcome; and third, the genetic instruments were independent of any potential confounding factors (20). All data used for the analyses in this study were publicly available online at the time of access and did not require approval from an ethics committee.

Source database and data collection

The GWAS data for anaphylactic shock were obtained from FinnGen, a comprehensive biobank that integrated genetic information with digital health registry data from the Finnish population (21). In this study, the outcome phenotype was specifically defined as “Anaphylactic shock due to adverse effect of correct drug or medication properly administered” (ICD-10: T88.6), as recorded in the FinnGen dataset. This dataset included a total of 414,951 individuals, comprising 336 cases and 411,845 controls. The patients' ages primarily ranged from 50 to 70 years, with the highest incidence observed in the 50 to 60 age group. The number of first-onset cases began to rise rapidly after the year 2000, peaking between 2010 and 2015. Cox regression analysis indicated that the occurrence of anaphylactic shock significantly increased mortality risk. In a 20-year follow-up, the long-term mortality risk was 22.173% for men compared to 12.907% for women (22). Circulating inflammatory proteins were selected based on previous research and clinical information derived from a genome-wide protein quantitative trait locus (pQTL) study of 91 plasma proteins, which included a large cohort of participants of European ancestry (19). Genetic association data for inflammatory protein levels were extracted from the GWAS Catalog (GCST90274772, GCST90274780, GCST90274791, GCST90274794, GCST90274814, and GCST90274825).

Instrument selection and quality control

SNPs associated with protein levels were selected based on a genome-wide significance threshold (P < 5 × 10−8). To ensure independence among instruments, clumping was performed using a linkage disequilibrium (LD) threshold of r2 < 0.001 within a 10,000-kb window (8). Instrument strength was evaluated by calculating F-statistics, where for a single instrument, the F-statistic is computed as F = (β/SE)2. SNPs with an F-statistic less than 10 were excluded to avoid weak instrument bias (23). Pleiotropy was assessed using the MR-Egger intercept test and the MR Pleiotropy Residual Sum and Outlier (MR-PRESSO) method, which identified and removed outlier SNPs to reduce bias (24).

Mendelian randomization analysis

MR analysis is an approach that uses genetic variants as instrumental variables (IVs) to determine the potential causal relationship between exposures and outcomes (14). Because genetic variants were assumed to be randomly inherited at conception and were not influenced by disease status, this method could minimize the effects of confounding and reverse causation (23). Recently, a meta-analysis of GWAS evaluated the genetic architecture underlying 91 inflammatory proteins, thereby enabling the investigation of their associations with DIAS (19). Accordingly, we employed a multivariable MR design to systematically assess the potential causal relationship between circulating cytokine levels and the risk of anaphylactic shock. Primary MR analyses included the inverse variance weighted (IVW), MR-Egger, and weighted median methods to estimate causal effects (15). Sensitivity analyses, including leave-one-out analyses and Cochran's Q tests for heterogeneity, were performed to confirm the robustness of our findings (24).

Colocalization analysis

To investigate whether anaphylactic shock and inflammatory proteins share common causal variants, we further conducted a colocalization analysis using the COLOC method (25). COLOC is a statistical approach based on GWAS summary statistics that evaluated whether two independently associated traits share a common genetic locus by calculating posterior probabilities for five hypotheses (PP.H0 to PP.H4), with PP.H4 representing the probability that both traits share the same causal variant (16).

Enrichment analysis

To investigate the functional characteristics and biological relevance of the candidate druggable genes, we conducted enrichment analyses using the R package “clusterProfiler” (version 4.10.1) (26). Specifically, we performed Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses (27).

Protein-protein interaction network construction

Protein-protein interaction (PPI) networks offer a clear and intuitive representation of the interactions among proteins encoded by critical druggable genes (28). In our study, we constructed a PPI network using the STRING database (https://string-db.org/), setting a minimum confidence score of 0.4 for interactions while keeping all other parameters at their default values.

Candidate drug prediction

The Drug Signatures Database (DSigDB; http://dsigdb.tanlab.org/DSigDBv1.0/) was an extensive resource that comprised 22,527 gene sets and 17,389 unique compounds, spanning 19,531 genes (29). In this study, we uploaded the prioritized genes to DSigDB to predict candidate drugs and assess the pharmacological activity of the target genes. The top candidates were selected based on their adjusted P values and biological relevance to anaphylactic shock. In addition, we provided a complete table of candidate drugs. The original P values provided by DSigDB reflect the significance of single tests, while the adjusted P values control the type I error rate brought about by multiple comparisons. Among the predicted candidates, AGNPC0JHFVD emerged as one of the top compounds identified by DSigDB, primarily due to its favorable binding affinity to key inflammatory proteins in our in silico screening process. As noted earlier, this compound was represented by a unique identifier rather than a traditional drug name. While it meets the computational criteria for potential efficacy, its pharmacological characteristics and mechanism of action remain to be fully elucidated. Therefore, additional experimental studies were necessary to evaluate its therapeutic potential in the context of DIAS.

Candidate drug prediction and molecular docking

Molecular docking simulations were performed to investigate the interaction between the top candidate drugs and the inflammatory proteins identified as significant in the MR analyses. The AutoDock Vina (Scripps Research, San Diego, CA) software was used for docking (30). The results of docking were evaluated and analyzed using the PLIP system (https://plip-tool.biotec.tu-dresden.de/plip-web/plip/index). Finally, the molecular docking outcomes of the two-dimensional structures were visualized using the LIGPLOT software version 4.5.3 (European Bioinformatics Institute, Cambridge, UK), and molecular docking maps were generated using PyMOL. Protein structures were obtained from PDB (https://www.pdb.org/) or AlphaFold (https://alphafold.com/), and data on drugs were retrieved from PubChem (https://pubchem.ncbi.nlm.nih.gov/) (31).

RESULT

Mendelian randomization analysis

Primary MR analyses were conducted to investigate the causal effects of six circulating inflammatory proteins on the risk of drug-induced anaphylaxis (Fig. 1, Figure S1, http://links.lww.com/SHK/C570). CD40L is significantly protective (OR = 0.69, 95% CI: 0.4951–0.9578, P = 0.027), while CXCL10 (OR = 1.51, 95% CI: 1.0075–2.2549, P = 0.046) and CCL3 (OR = 2.08, 95% CI: 1.0079–4.3072, P = 0.048) significantly increased. The other proteins (FLT3LG, IFN-γ, IL-5) also showed protective effects (ORs = 0.52–0.32, P < 0.05). Sensitivity analyses support the robustness of these findings, with leave-one-out analyses and Cochran's Q tests indicating minimal influence of individual SNPs and low heterogeneity among the instrumental variables (Fig. 2).

Fig. 1.

Fig. 1

Open in a new tab

The forest plot displays the association between various inflammatory factor levels and disease risk, calculated using the inverse variance weighted method. CI, confidence interval; nsnp, number of SNPs; OR, odds ratio. CD40L: CD40 ligand. CXCL10: C-X-C motif chemokine ligand 10. FLT3LG: Fms-related tyrosine kinase 3 ligand. IFNG: interferon gamma (IFN-γ). IL5: interleukin-5 (IL-5). CCL3: macrophage inflammatory protein-1 alpha.

Fig. 2.

Fig. 2

Open in a new tab

This figure presents the forest plots from Mendelian randomization (MR) analyses investigating the causal relationships between genetic variants and the risk of anaphylactic shock or related traits. Each panel (A–F) represents the results of MR analysis for different exposure-outcome pairs using various methods (inverse variance weighted [IVW], MR-Egger). MR, Mendelian randomization; CD40L (GCST90274772), CXCL10 (GCST90274780), FLT3LG (GCST90274791), IFNG (GCST90274794), IL-5 (GCST90274814), CCL3 (GCST90274825).

Colocalization analysis

The colocalization analysis reveals distinct patterns across the six inflammatory proteins. Notably, CD40L exhibits a PP.H4 of approximately 58.07%, providing strong evidence that the genetic architecture influencing CD40L levels was likely shared with that of anaphylactic shock (Fig. 3). In contrast, CXCL10 and CCL3 show PP.H4 values of about 34.05% and 21.13%, respectively, suggesting moderate to low probability of a shared causal variant. For FLT3LG, PP.H4 was around 31.31%, while IFN-γ and IL-5 have even lower PP.H4 values of approximately 16.08% and 17.28%, respectively, indicating that these proteins are predominantly influenced by independent genetic signals.

Fig. 3.

Fig. 3

Open in a new tab

Colocalization analysis of genetic variants associated with inflammatory proteins and DIAS. DIAS: drug-induced anaphylactic shock; COLOC: colocalization analysis; PP.H2.abf: posterior probability hypothesis 2 (association with trait 2 only); PP.H3.abf: posterior probability hypothesis 3 (distinct causal variants for trait 1 and trait 2); PP.H4.abf: posterior probability hypothesis 4 (shared causal variant for both traits); Mb (megabase), IFNG (interferon gamma), CCL3 (macrophage inflammatory protein-1 alpha), CD40L (CD40 ligand), CXCL10 (C-X-C motif chemokine ligand 10), FLT3LG (Fms-related tyrosine kinase 3 ligand), IL-5 (interleukin-5).

Enrichment analysis

Genes associated with CD40L, FLT3LG, IFN-γ, and IL-5 are significantly enriched in biological processes related to T-cell activation, antigen presentation, and immune tolerance. In contrast, genes linked to CXCL10 and CCL3 show strong enrichment in pathways governing chemotaxis and cell adhesion (Fig. 4). KEGG pathway analysis further confirmed that the target genes were significantly associated with immune-related pathways such as T-cell receptor signaling and cytokine-cytokine receptor interactions, whereas pathways involved in cellular migration and adhesion were predominantly represented among CXCL10- and CCL3-related gene sets.

Fig. 4.

Fig. 4

Open in a new tab

Enrichment analysis of key inflammatory proteins identified through multivariable Mendelian randomization. p-adjust, Benjamini–Hochberg adjusted P value; GO, gene ontology; Th1, T helper type 1 cell; KEGG, Kyoto Encyclopedia of Genes and Genomes.

Protein-protein interaction network construction

The PPI network displayed a highly interconnected structure, with CD40L emerging as the central hub exhibiting the highest degree of connectivity (Fig. S2, http://links.lww.com/SHK/C570). The network comprises 91 nodes and over 300 edges, reflecting the complex interplay among inflammatory proteins. These findings underscore CD40L's pivotal role and potential as a therapeutic target in drug-induced anaphylaxis pathways.

Candidate drug prediction and molecular docking

Candidate drug prediction and molecular docking analyses are performed to identify promising therapeutic compounds targeting the six inflammatory proteins. As summarized in Table 1 and Figure 5A, binding affinity values indicate the interaction strengths between candidate compounds and proteins, with lower values reflecting stronger affinities. Molecular docking of the top 10 candidate drugs reveals binding affinities ranging from moderate to strong (−7.0 to −8.3 kcal/mol) for inflammatory proteins, including CD40L, CXCL10, FLT3LG, IFN-γ, and IL-5 (Fig. 5B–G). Notably, the compound AGNPC0JHFVD exhibits the highest affinity with CCL3 (−7.2 kcal/mol) and moderate affinities with other proteins such as CD40L (−7.2 kcal/mol) and CXCL10 (−7.0 kcal/mol). Furthermore, atorvastatin and beclomethasone show moderate binding interactions with targets like CD40L and CXCL10. Detailed docking results, including composite scores and adjusted P values for the top 10 candidate drugs across the six target proteins, are summarized in Table 1. Additionally, several other compounds, such as atorvastatin and beclomethasone, demonstrate moderate affinities toward specific inflammatory targets, indicating their potential as therapeutic candidates.

Table 1.

Binding affinity between the top candidate drugs and the inflammatory proteins

Binding affinity/ligand CCL3 CD40LG CXCL10 FT3LG IFNG IL5
AGNPC0JHFVD −8.5 −7.2 −7 −7.5 −8.3 −8.3
dl-Mevalonic acid −4.8 −3.7 −4.3 −4.4 −4.1 −3.7
Dinoprostone −5.2 −5 −5.2 −5 −5.4 −5.3
Isoguanine −5.5 −6.1 −4.4 −4.8 −4.9 −4.7
Modrasone −6.5 −5.9 −6 −6.1 −7.4 −7
Vitamin D3 −6.9 −6.6 −6.5 −6.6 −7.4 −7.4
Atorvastatin −7.6 −5.9 −6.1 −6.7 −7 −7.5
Beclomethasone −6.9 −6.3 −6.2 −6.4 −7.3 −7.3
GSNO −5.5 −5.4 −4.9 −4.8 −5.4 −4.7
Open in a new tab

AGNPC0JHFVD is represented by a unique database identifier rather than a conventional drug name; CCL3: macrophage inflammatory protein-1 alpha; CD40LG: CD40 ligand; CXCL10: C-X-C motif chemokine ligand 10; FLT3LG: Fms-related tyrosine kinase 3 ligand; IFNG: interferon gamma; IL-5/IL5: interleukin-5; binding affinity: docking score in kcal/mol.

Fig. 5.

Fig. 5

Open in a new tab

A, Potential therapeutic compounds identified through DSigDB drug screening. B, CCL3—atorvastatin (−7.6 kcal/mol). C, IL-5—AGNPC0JHFVD (−8.3 kcal/mol). D, IFN-γ—AGNPC0JHFVD (−8.3 kcal/mol). E, CCL3—AGNPC0JHFVD (−8.7 kcal/mol). F, IL-5—atorvastatin (−7.5 kcal/mol). G, IFN-γ—beclomethasone (−7.5 kcal/mol). DSigDB: Drug Signatures Database, http://links.lww.com/SHK/C571. Combined score: composite enrichment score from DSigDB.

DISCUSSION

Circulating inflammatory proteins

The study provides multidimensional evidence to implicate circulating inflammatory proteins in the pathogenesis of DIAS. By integrating MR, colocalization analysis, functional enrichment, drug screening, and molecular docking, distinct roles for individual proteins within the inflammatory milieu are delineated. Notably, MR analyses reveal that genetically elevated levels of CD40L, FLT3LG, IFN-γ, and IL-5 are associated with a reduced risk of anaphylactic shock, whereas increased levels of CXCL10 and CCL3 confer a higher risk. These findings are consistent with the notion that inflammatory proteins did not uniformly influence disease outcomes; rather, they operate through diverse mechanisms that may be protective or deleterious depending on the context (32).

The strong protective association observed for CD40L was particularly striking. Colocalization analysis further supports a high probability (PP.H4 ≈ 58%) that the genetic determinants of CD40L levels and anaphylactic shock are shared, suggesting that CD40L may serve as a central mediator in the immune regulation underlying DIAS. CD40L is known to facilitate B-cell and dendritic cell interactions, which are essential for efficient antigen presentation and the orchestration of adaptive immune responses. Previous studies report that CD40L-CD40 signaling regulates inflammatory responses through the induction of adhesion molecules (VCAM and ICAM) and the secretion of pro-inflammatory cytokines (33). Its protective role may be attributed to the ability to enhance immune surveillance and promote a balanced immune response, thereby mitigating the hyperactivation that can lead to shock (34).

In contrast, the findings for CXCL10 and CCL3 indicate that these proteins are associated with an increased risk of anaphylactic shock. CXCL10, a chemokine known for its role in attracting Th1 cells and natural killer cells, may exacerbate anaphylactic responses by promoting excessive recruitment of immune cells to sites of inflammation (10,35). A retrospective study demonstrates that CXCL10 levels are related to and predictive of allergy development (35). Similarly, CCL3 plays an important role in recruiting inflammatory cells and amplifying inflammatory cascades. A single-cell sequencing analysis of allergic pneumonitis revealed that high expression of CCL3 correlates with the release of IL-1, IL-2, IL-12, and TNFα/NFκB (12). The colocalization analysis shows moderate posterior probabilities (PP.H4) for CXCL10, indicating limited evidence for shared causal variants (PP.H4). This suggests that the genetic variants influencing CXCL10 and CCL3 might be largely independent from those influencing anaphylactic shock. Therefore, these proteins may contribute to disease risk through distinct pathways that promote hyperinflammation (10,36).

Although MR estimates indicate that higher levels of these proteins were associated with a reduced risk of anaphylactic shock, the colocalization probabilities (PP.H4 of approximately 31% for FLT3LG, 16% for IFN-γ, and 17% for IL-5) were considerably lower compared to CD40L. These findings suggested that the genetic factors influencing these inflammatory proteins may act largely independently of those associated with anaphylactic shock. It is therefore plausible that FLT3LG, IFN-γ, and IL-5 contribute to disease risk through mechanisms distinct from the genetic basis of anaphylaxis itself. This independence indicates that their role in promoting immune responses might operate via separate inflammatory pathways.

Beyond genetic association analyses, our functional enrichment studies using GO and KEGG pathway analyses provide additional insights into the biological processes in which these proteins were involved. The genes associated with CD40L, FLT3LG, IFN-γ, and IL-5 are significantly enriched in pathways related to T-cell activation, antigen presentation, and immune tolerance. In contrast, genes linked to CXCL10 and CCL3 are enriched in pathways controlling chemotaxis and cell adhesion, which are critical in mediating inflammatory responses. These enrichment results reinforce the idea that the protective versus risk effects of inflammatory proteins in anaphylactic shock may be driven by their differential involvement in immune regulatory versus pro-inflammatory pathways (8,18,25).

Translational potential from drug screening and molecular docking

Furthermore, the candidate drug prediction and molecular docking analyses underscore the translational potential of these findings. Using the DSigDB database, several compounds with favorable binding affinities to the inflammatory proteins are identified. Among these, AGNPC0JHFVD exhibits particularly strong binding to CCL3 (−8.5 kcal/mol), while atorvastatin and beclomethasone show moderate binding with targets such as CD40L and CXCL10. These molecular docking results provide structural insights into the potential interactions between candidate drugs and their protein targets, suggesting that pharmacological modulation of these inflammatory pathways could serve as a viable strategy for the prevention or treatment of DIAS.

STRENGTH AND LIMITATIONS

Despite the insights provided, our study shares some inherent limitations common to MR and colocalization studies. First, using summary-level GWAS data may mask individual heterogeneity and limit analyses of nonlinear relationships or gene-environment interactions. Second, variability in SNP estimates can affect causal inference precision, even after excluding weak instruments. Third, residual pleiotropy may still bias results, despite sensitivity analyses. Fourth, the predominantly European (Finnish) ancestry and modest sex imbalance of the study cohort may limit the generalizability of our findings to more diverse populations. Finally, enrichment and docking analyses are predictive and therefore require experimental validation. Future research integrating larger datasets and additional omics layers could further clarify these mechanisms.

CONCLUSION

Collectively, the results support a model in which circulating inflammatory proteins play diverse roles in the etiology of anaphylactic shock. Among these proteins, CD40L emerges as a key protective factor, potentially enhancing immune regulation and promoting antigen presentation. In contrast, the pro-inflammatory mediators CXCL10 and CCL3 appear to increase the risk of anaphylactic shock by facilitating excessive immune cell recruitment and activation. The roles of FLT3LG, IFN-γ, and IL-5 are more intricate, as their beneficial effects on immune homeostasis do not appear to be directly associated with the genetic mechanisms driving anaphylactic shock. This complexity highlights the need for further investigation into their specific contributions to disease pathogenesis.

ACKNOWLEDGMENTS

The authors sincerely thank all the consortium studies and related investigators for making the summary association statistics data publicly available. In addition, the authors would like to express their gratitude to all researchers whose work contributed to the foundation of this study.

Footnotes

Disclosure: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Jinwei Dai and Nianzhe Sun contributed equally to this work as co-first authors.

Funding: Funding was provided by the National Key Research and Development Program of China (2022YFC2009800), and the Natural Science Foundation of Hunan Province (2023JJ30931).

Author contributions: J.D. and N.S. contributed equally to this work as co-first authors. They were responsible for data collection, data analysis, and manuscript drafting. Both authors played a pivotal role in the study design and interpretation of the results. W.X. contributed to data validation and critical revision of the manuscript. Z.Q. and X.P. jointly supervised the study and provided critical guidance throughout the project. They were responsible for conceptualization, funding acquisition, and final manuscript revision. All authors approved the final version of the manuscript for submission.

Data availability statement: Full per-protein GWAS summary statistics were available for download at https://www.ebi.ac.uk/gwas/ (GCST90274772, GCST90274780, GCST90274791, GCST90274794, GCST90274814, and GCST90274825). The GWAS data for anaphylactic shock were obtained from FinnGen (https://r10.finngen.fi/Anaphylactic shock due to adverse effect of correct drug or medication properly administered).

The authors report no conflicts of interest.

Supplemental digital content is available for this article. Direct URL citation appears in the printed text and is provided in the HTML and PDF versions of this article on the journal’s Web site (www.shockjournal.com).

Contributor Information

Jinwei Dai, Email: greg0453@163.com.

Nianzhe Sun, Email: sunnzh201921@sina.com.

Wenye Xu, Email: 1041965394@qq.com.

Zhihong Zuo, Email: zhihong.zuo1995@gmail.com.

Xiaoyang Pang, Email: xiaoyangpang@csu.edu.cn.

Zhaoxin Qian, Email: xyqzx@csu.edu.cn.

REFERENCES

  • 1.Blumenthal KG, Chiriac AM, Khan DA. Drug allergy: then, now, and next. J Allergy Clin Immunol Pract. 2023;11(2):429–430. [DOI] [PubMed] [Google Scholar]
  • 2.Motosue MS, Li JT, Campbell RL. Anaphylaxis: epidemiology and differential diagnosis. Immunol Allergy Clin North Am. 2022;42(1):13–25. [DOI] [PubMed] [Google Scholar]
  • 3.Castells MC. Drug allergy: phenotypes, endotypes, and biomarkers. J Allergy Clin Immunol Pract. 2017;5(3):626–627. [DOI] [PubMed] [Google Scholar]
  • 4.Altman AM Camargo CA Jr. Simons FE, et al. Anaphylaxis in America: a national physician survey. J Allergy Clin Immunol. 2015;135(3):830–833. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Yu RJ Krantz MS Phillips EJ, et al. Emerging causes of drug–induced anaphylaxis: a review of anaphylaxis–associated reports in the FDA adverse event reporting system (FAERS). J Allergy Clin Immunol Pract. 2020;9(2):819. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Castells M. Drug allergy biomarkers: are we there yet? Ann Allergy Asthma Immunol. 2023;130(2):145–146. [DOI] [PubMed] [Google Scholar]
  • 7.Luo L Zhang J Lee J, et al. Platelets, not an insignificant player in development of allergic asthma. Cells. 2021;10(8):2038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Wu X, Song Y, Wu S. Relation of 91 circulating inflammatory proteins to nonalcoholic fatty liver disease: a two-sample Mendelian randomisation study. J Cell Mol Med. 2024;28(24):e70322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Zhao AY Unterman A Abu Hussein N, et al. Peripheral blood single-cell sequencing uncovers common and specific immune aberrations in fibrotic lung diseases. bioRxiv: Prepr Serv Biol. Published online September 22, 2023:2023.9.20.558301. [Google Scholar]
  • 10.Ozga AJ Chow MT Lopes ME, et al. CXCL10 chemokine regulates heterogeneity of the CD8+ T cell response and viral set point during chronic infection. Immunity. 2022;55(1):82–97.e8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Bosco A. Novel role of the GSDMB/IFNG axis in childhood asthma. Am J Respir Crit Care Med. 2024;209(8):899–900. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Zhao AY Unterman A Abu Hussein NS, et al. Single-cell analysis reveals novel immune perturbations in fibrotic hypersensitivity pneumonitis. Am J Respir Crit Care Med. 2024;210(10):1252–1266. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Han J Pan C Tang X, et al. Hypersensitivity reactions to small molecule drugs. Front Immunol. 2022;13:1016730. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Emdin CA, Khera AV, Kathiresan S. Mendelian randomization. JAMA. 2017;318(19):1925–1926. [DOI] [PubMed] [Google Scholar]
  • 15.Sekula P Del Greco MF Pattaro C, et al. Mendelian randomization as an approach to assess causality using observational data. J Am Soc Nephrol: JASN. 2016;27(11):3253–3265. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Zhang C, He Y, Liu L. Identifying therapeutic target genes for migraine by systematic druggable genome-wide Mendelian randomization. J Headache Pain. 2024;25(1):100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Skrivankova VW Richmond RC Woolf BAR, et al. Strengthening the reporting of observational studies in epidemiology using Mendelian randomization: the STROBE-MR statement. JAMA. 2021;326(16):1614–1621. [DOI] [PubMed] [Google Scholar]
  • 18.Zhang X Wang P Dang Q, et al. Inflammatory cytokines and risk of allergic rhinitis: a Mendelian randomization study. Cytokine. 2024;177:156547. [DOI] [PubMed] [Google Scholar]
  • 19.Zhao JH Stacey D Eriksson N, et al. Genetics of circulating inflammatory proteins identifies drivers of immune-mediated disease risk and therapeutic targets. Nat Immunol. 2023;24(9):1540–1551. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Lawlor DA Harbord RM Sterne JA, et al. Mendelian randomization: using genes as instruments for making causal inferences in epidemiology. Stat Med. 2008;27(8):1133–1163. [DOI] [PubMed] [Google Scholar]
  • 21.Kurki MI Karjalainen J Palta P, et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature. 2023;613(7944):508–518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Risteys ANAPH_SHOCK_ADVER_EFFECT_CORRE_DRUG_MEDICAM_PROPE_ADMINISTE. Accessed May 5, 2025. https://risteys.finngen.fi/endpoints/ANAPH_SHOCK_ADVER_EFFECT_CORRE_DRUG_MEDICAM_PROPE_ADMINISTE
  • 23.Kuppa A Tripathi H Al-Darraji A, et al. C-reactive protein levels and risk of cardiovascular diseases: a two-sample bidirectional Mendelian randomization study. Int J Mol Sci. 2023;24(11):9129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through egger regression. Int J Epidemiol. 2015;44(2):512–525. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Cai Y Xie S Jia X, et al. Integrated analysis of Mendelian randomization and bayesian colocalization reveals bidirectional causal association between inflammatory bowel disease and psoriasis. Ann Med. 2023;55(2):2281658. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Sun Z Yun Z Lin J, et al. Comprehensive Mendelian randomization analysis of plasma proteomics to identify new therapeutic targets for the treatment of coronary heart disease and myocardial infarction. J Transl Med. 2024;22(1):404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Yu G Wang LG Han Y, et al. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS. 2012;16(5):284–287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Szklarczyk D Kirsch R Koutrouli M, et al. The STRING database in 2023: protein-protein association networks and functional enrichment analyses for any sequenced genome of interest. Nucleic Acids Res. 2023;51(D1):D638–D646. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Yoo M Shin J Kim J, et al. DSigDB: drug signatures database for gene set analysis. Bioinform (oxf Engl). 2015;31(18):3069–3071. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Morris GM Huey R Lindstrom W, et al. AutoDock4 and AutoDockTools4: automated docking with selective receptor flexibility. J Comput Chem. 2009;30(16):2785–2791. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Kim S Chen J Cheng T, et al. PubChem 2023 update. Nucleic Acids Res. 2023;51(D1):D1373–D1380. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Pan X Chiwanda Kaminga A Liu A, et al. Chemokines in non-alcoholic fatty liver disease: a systematic review and network meta-analysis. Front Immunol. 2020;11:1802. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.van Kooten C, Banchereau J. CD40-CD40 ligand. J Leukoc Biol. 2000;67(1):2–17. [DOI] [PubMed] [Google Scholar]
  • 34.Gattringer M Baranyi U Pilat N, et al. Anti-OX40L alone or in combination with anti-CD40L and CTLA4Ig does not inhibit the humoral and cellular response to a major grass pollen allergen. Clin Exp Allergy. 2016;46(2):354–364. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Huoman J Haider S Simpson A, et al. Childhood CCL18, CXCL10 and CXCL11 levels differentially relate to and predict allergy development. Pediatr Allergy Immunol. 2021;32(8):1824–1832. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Karason K Jernås M Hägg DA, et al. Evaluation of CXCL9 and CXCL10 as circulating biomarkers of human cardiac allograft rejection. BMC Cardiovasc Disord. 2006;6:29. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Shock (Augusta, Ga.) are provided here courtesy of Wolters Kluwer Health

RESOURCES