Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Feb 27;101(5):961–971. doi: 10.1002/ajh.70262

Genetic Contribution to Asthma Informs Acute Chest Syndrome Pathophysiology and Risk Stratification

Sara El Aouhel 1, Vanessa Bellegarde 2, Stennio Da Silva Faria 1, Tristan St‐Laurent 1, Estelle Lecluze 1, Anne‐Laure Pham Hung d’Alexandry d’Orengiani 3,4, Frédéric Galactéros 3,4, Pablo Bartolucci 3,4, Marc‐André Legault 1,5, Guillaume Lettre 2,4,6, Thomas Pincez 1,4,7,8,✉
PMCID: PMC13055117  PMID: 41757739

ABSTRACT

Acute chest syndrome (ACS) is a severe complication of sickle cell disease (SCD) occurring in ~50% of patients, some presenting frequent episodes. We lack tools to identify patients at high risk of ACS occurrence or frequent episodes. Epidemiological studies have found an association between asthma and ACS, but whether this link is causal is unclear. We used polygenic scores (PGS) to analyze whether the genetic propensity for asthma was associated with ACS and could be used to stratify the risk of ACS. We identified that PGS for asthma (PGSasthma) was associated with ACS rate (number of episodes per year), but not ACS occurrence, in both the CSSCD (n = 1278) and the GEN‐MOD (n = 406) prospective SCD cohorts, independently of fetal hemoglobin (HbF) (β = 0.17, standard error = 0.06, p = 0.006). This effect was most pronounced in patients with low HbF levels. Combining PGSasthma and HbF identified a population at high risk of frequent ACS after a first episode: individuals within the highest PGSasthma quintile and the lowest HbF quintile. Finally, we found that the genetic correlation between these two conditions only partially overlapped. This suggests that genetically determined asthma is not the unique contributor to the genetic propensity for ACS and that other genetic determinants may also play a role. In sum, our results suggest that patients with a high genetic propensity for asthma are prone to frequent ACS if not protected by high HbF levels. Combining PGSasthma and HbF may allow identifying patients at high risk of frequent ACS after a first episode for personalized management.

Keywords: acute chest syndrome, asthma, polygenic score, risk stratification, sickle cell disease

1. Introduction

Sickle cell disease (SCD) is one of the most frequent monogenic diseases worldwide. SCD is due to a variant in the β‐globin gene leading to an abnormal hemoglobin—hemoglobin S—which polymerizes while deoxygenated [1]. The resulting hemolysis and vaso‐occlusion cause several complications, including pain crisis, stroke, early organ damage, and acute chest syndrome (ACS) [2]. SCD is highly heterogeneous in terms of clinical severity and there is currently no tool to stratify the patients according to the risk of complications [3].

ACS is a frequent complication in SCD, especially in young children, but can occur throughout life and some patients may present frequent episodes [3]. ACS manifests with respiratory symptoms and a new infiltrate in a chest X‐ray, and can progress to life‐threatening respiratory failure. Despite a great improvement in management in the last decade, ACS remains potentially life‐threatening, particularly in adults [4, 5]. Some patients will experience frequent ACS but the cause of these recurrent events is unclear [6]. ACS is a heterogeneous syndrome defined as the occurrence of a new pulmonary consolidation associated with respiratory symptoms [7]. Various non‐mutually exclusive triggers can lead to ACS, including infection, vaso‐occlusive pain crisis, fat embolism, and asthma [8]. Several studies identified an epidemiological link between asthma and ACS. Children and adults with asthma have an increased risk of ACS [9, 10, 11, 12, 13]. Asthma is considered a frequent trigger of ACS in children [14]. However, the causality of the association and the pathophysiological link between these conditions are unclear.

Genetic susceptibility to asthma has been extensively investigated by genome‐wide association studies (GWAS), identifying many loci associated with asthma risk [15, 16, 17]. The genetic susceptibility to ACS is much less known, and the few loci found have not been replicated [18, 19]. Whether the genetic susceptibility of asthma and ACS is similar is unknown. Moreover, polygenic scores (PGS) that summarize genetically determined susceptibility to asthma have allowed for the quantification of asthma risk [15]. It is unknown whether asthma PGS can identify patients with SCD at high risk of ACS occurrence or frequent episodes. Identifying populations at high risk for ACS would allow tailoring their management.

Here, we leveraged large asthma GWAS from non‐SCD individuals and genotyped cohorts of children and adults with SCD to investigate the link between asthma and ACS. We hypothesized that the knowledge gained on asthma genetics can be used to improve our understanding and risk modeling of ACS. We aimed to assess whether asthma was causally associated with ACS and which pathophysiological component was involved. We also aimed to evaluate the usefulness of PGS for asthma (PGSasthma) in the risk stratification of ACS in patients with SCD.

2. Methods

2.1. Populations

We used the main genotyped dataset from the Cooperative Study of Sickle Cell Disease (CSSCD) as our discovery cohort [20]. The CSSCD was a US, multicenter, prospective study of the natural history of SCD that enrolled participants between 1978 and 1988. The main genotyped dataset included 1278 patients (mean age at inclusion, 14 ± standard deviation [SD] 12 years) of whom 435 (34%) were 18 or older, and 662 (52%) were female (Table S1). ACS was defined as the occurrence of a new infiltrate on chest X‐ray (or by a perfusion defect on a lung radioisotope scan but < 1% of ACS were diagnosed using this method) [21]. A total of 484 (38%) experienced an ACS. The mean ACS rate (defined as the number of ACS per year of study) was 0.13 ± 0.29. As replication cohorts, we used the adult Genetic Modifier (GEN‐MOD, n = 406) cohort and a smaller subset of the CSSCD that has been genotyped separately, herein called CSSCD2 (n = 261). GEN‐MOD is an ongoing prospective cohort of adults with SS SCD followed at Henri‐Mondor Hospital in Creteil, France [22, 23]. Within this cohort, ACS is defined as “the appearance of an auscultatory abnormality (crepitants and/or bronchial breathing) and/or chest pain and an infiltrate on chest film and/or thoracic computed‐tomography scan but excluding atelectasia” [24]. No participant of the CSSCD received hydroxyurea. In the GEN‐MOD cohort, 242 patients (60%) received hydroxyurea during the prospective follow‐up. The studies were approved by the institutional ethics committee, and we collected data according to the Helsinki Declaration. DNA genotyping and data processing have been described elsewhere [23, 25].

2.2. PGSasthma

For this study, we selected a recently published PGSasthma (n = 3972 232 variants, PGS Catalog number PGS004725) [26] built using PRSmix+ from 47314 individuals (n = 8043 cases) [27]. PRSmix+ combines several PRS published in the PGS Catalog or independent biobanks for the same trait to improve prediction accuracy. As a result, both adults and children may have been included in the initials PGS that were part of the PGSasthma generated using PRSmix+. PRSmix+ also considers genetically correlated traits to increase the sample size. The prediction gain associated with PRSmix+ was demonstrated in two adult cohorts: individuals of European ancestry in the All of Us Research Program, and individuals of South Asian ancestry in the Gene & Health Biobank. Since most PGS published to date were generated using European ancestry individuals [28, 29], PRSmix+ mainly included data from this population [27]. To account for the fact that most patients with SCD are of African ancestry, we validated our results by using a PGS generated exclusively from individuals of African ancestry. We used the PGS developed by the Global Biobank Meta‐analysis Initiative (GBMI) from six biobanks with a total of 32658 African ancestry individuals (n = 5051 cases) (PGS Catalog number 005121) [15].

We used plink2 to compute PGSasthma for each cohort [30], and then we normalized the PGSasthma and used the obtained z‐scores throughout the study.

We computed the effect of the PGSasthma on ACS rate using quasi‐Poisson regression in a model including age, sex, fetal hemoglobin (HbF), and the first 10 principal components. We used quasi‐Poisson to be consistent with previous studies investigating ACS and pain rates [31]. However, we also performed a sensitivity analysis using more conservative linear regression. For linear regression, we applied inverse normal transformation to normalize the ACS rate. We also tested the association between PGSasthma and other complications using quasi‐Poisson (pain rate) or logistic (stroke, osteonecrosis, priapism, leg ulcer, and death) regression with the same covariates.

2.3. PGS Partitioning

We used a recently published partitioned PGS (pPGS) for asthma to analyze the different pathophysiological mechanisms of asthma. pPGS separates the single‐nucleotide polymorphisms (SNPs) of a PGS based on the cell type or pathophysiological pathway likely involved. The pPGS was built using cell epigenomic data (H3K4me2 marks determined by chromatin immunoprecipitation followed by DNA sequencing (ChIP‐seq)) and identified five clusters with different cell type compositions. These clusters were associated with different asthma endophenotypes, particularly the one that includes eosinophils, which was associated with eosinophilic asthma [32]. We tested the association between the five pPGS clusters and asthma, as described for PGSasthma.

2.4. GWAS and SNPs‐Specific Analyses

We performed a GWAS for ACS rate on the CSSCD using plink2 after adjusting ACS rate for sex and age and applying inverse normal transformation [30]. We tested an additive model using linear regression adjusted for the first 10 principal components. To investigate the association of ORMDL3 gene with ACS, we also focused on the SNPs located within the 17q21 locus and associated with asthma in Open Targets and GWAS catalog [26, 33]. We performed the same process as the GWAS for these SNPs.

2.5. Genetic Correlation

To compare genetic susceptibility to asthma and ACS, we used the African ancestry GWAS results of the GBMI, from which was derived the PGS005121 described above [15].

We used genome‐wide pairwise‐association signal sharing (GPS) to analyze the genetic correlation between asthma and ACS GWAS. GPS has been developed for rare diseases in which the classical methods based on linkage disequilibrium are not suitable (heritability of ACS in the CSSCD cohort: h 2, −0.74; SE, 0.55) [34]. GPS is a nonparametric test of the independence of the p‐values obtained by the two GWAS to compare. We used the gps_cpp package to compute GPS and generated a control distribution using 1000 permutations to obtain p‐value [35].

We analyzed the distance between the statistically suggestive SNPs (p < 1 × 10−6) found in the two GWAS using bedtools [36]. We computed the correlation between the −log10 of the p‐values using Spearman correlation coefficient.

2.6. Statistical Analyses

We compared continuous and categorical variables using nonparametric Fisher's exact and Wilcoxon‐Mann–Whitney tests, respectively. We used a binomial test to assess enrichment in the direction of effect. All tests were two‐sided, and a statistical significance threshold was set to p < 0.05. We corrected for multiple testing the analysis of the ORMDL3 locus with the Benjamini–Hocheberg method, using a 0.05 false discovery rate, and considered a q‐value < 0.05 as statistically significant [37]. All statistical analyses were performed using RStudio (version 1.2.5033) or GraphPad Prism (version 9.2.0, GraphPad Software LLC, CA).

3. Results

3.1. Genetically Determined Asthma Risk Modulates the Rate of ACS

We used a PGSasthma built from a large multi‐ancestry population [27]. In the CSSCD cohort, PGSasthma was associated with ACS rate, independently of HbF (adjusted β = 0.17, standard error [SE] = 0.06, p = 0.006) (Table 1). Of note, the association was specific to ACS rate as we found no association with ACS occurrence in a logistic regression model (adjusted β = 0.07, SE = 0.06, p = 0.24). We found no association of PGSasthma with pain rate, stroke, osteonecrosis, priapism, leg ulcers, and death (p ≥ 0.17, Table S2). We integrated HbF into our model, as this is the major modifier of SCD complications, including ACS. We noted that the association was stronger in a model with HbF than without (adjusted β = 0.13, SE = 0.06, p = 0.02). This is evocative of a differential effect depending on HbF level, despite the interaction term between HbF and PGSasthma not being significant (p = 0.28). We previously shown that the effect of a PGS for HbF depended on the HbF level and was higher in individuals with low HbF levels [38]. We thus separated our cohort into subgroups with HbF < and ≥ 5% (n = 513 and n = 625, respectively). We found that the effect of PGSasthma was higher in patients with HbF < 5% than in the whole cohort (adjusted β = 0.24, SE = 0.08, p = 0.005), whereas we found no association in patients with HbF ≥ 5% (adjusted β = 0.08, SE = 0.09, p = 0.37). We chose a HbF threshold of 5% to obtain relatively balanced subgroups, but the effect of HbF may be continuous rather than dichotomous. Indeed, we noticed a more pronounced effect of PGSasthma on ACS rate among patients with HbF < 2% than among those with HbF between 2% and 5% (adjusted β = 0.50, SE = 0.21, p = 0.02, and adjusted β = 0.20, SE = 0.01, p = 0.04, respectively).

TABLE 1.

Association of the PGSasthma with ACS rate.

Quasi‐Poisson Linear regression
Beta SE p‐Value Beta SE p‐Value
CSSCD (n = 1278), including 513 with HbF < 5% and 625 with HbF ≥ 5%
Whole cohort 0.17 0.06 0.006 0.02 0.009 0.02
HbF < 5% 0.24 0.08 0.005 0.04 0.02 0.02
HbF ≥ 5% 0.08 0.09 0.37 0.007 0.009 0.46
GENMOD (n = 317), including 137 with HbF < 5% and 180 with HbF ≥ 5%
Whole cohort 0.51 0.17 0.002 0.17 0.06 0.009
HbF < 5% 0.73 0.37 0.05 0.19 0.09 0.03
HbF ≥ 5% 0.38 0.21 0.07 0.13 0.09 0.16
CSSCD2 (n = 261), including 120 with HbF < 5% and 141 with HbF ≥ 5%
Whole cohort −0.06 0.11 0.60 −0.03 0.01 0.33
HbF < 5% 0.11 0.18 0.54 0.02 0.09 0.98
HbF ≥ 5% −0.24 0.16 0.13 −0.12 0.09 0.21
Open in a new tab

Note: We tested the effect of PGSasthma z‐scores on ACS rate. We adjusted all models on age, sex, HbF and the 10 first principal component. We applied inverse normal transformation to normalize ACS rate in the linear regression models. HbF, fetal hemoglobin; SE, standard error.

We then aimed to validate our findings in other cohorts. In the adult GENMOD cohort (n = 261), we found similar effects of the PGSasthma on ACS rate (adjusted β = 0.51, SE = 0.17, p = 0.003, adjusted on HbF). We also confirmed the association was stronger in patients with HbF < 5% than in those with HbF > 5% (adjusted β = 0.73, SE = 0.37, p = 0.05 and adjusted β = 0.38, SE = 0.21, p = 0.07, respectively). We found the same direction of effect in the smaller CSSD2 cohort for HbF < 5% (n = 120) but without significant association, possibly due to a lack of power.

In a sensitivity analysis, we used the more conservative linear regression approach on inverse transformation‐normalized ACS rates. We found similar effects for all the models in each cohort (Table 1). In a second sensitivity analysis, we analyzed only children (n = 843) from the CSSCD. Indeed, since the GEN‐MOD cohort includes only adults and the CSSCD includes 34% of adults, our results may be driven by adult patients and not applicable to children. We confirmed that, among children in the CSSCD, PGSasthma was associated with ACS rate, independently of HbF (adjusted β = 0.23, SE = 0.08, p = 0.005).

In sum, PGSasthma is associated with ACS rate independently of HbF and this association is stronger in the subgroup of patients with baseline HbF < 5%.

3.2. PGSasthma Allows Identifying ACS Rate Subgroups

We then asked whether combining PGSasthma and HbF would allow stratifying the ACS rate (Figure 1A). As PGSasthma was associated with ACS rate and not occurrence, we analyzed the risk of ACS recurrence (i.e., new episode after a first ACS) by studying the subgroup with at least one ACS. We separated the patients from the CSSCD based on PGSasthma and HbF quintiles. We found that among patients within the lowest HbF quintile, those within the top PGSasthma quintile had a higher ACS rate than those within the lowest PGSasthma quintiles (median 0.35 vs. 0.23, p = 0.049) (Figure 1B). In the lowest HbF and lowest PGSasthma quintiles group, 2/21 patients (9.5%) had an ACS rate > 0.5 compared to 10/26 (38%) in the lowest HbF and top PGSasthma quintiles group (p = 0.04) (Figure 1C). Within the highest HbF quintile, there was no difference in ACS rate between patients within the highest and the lowest PGSasthma quintiles (median 0.24 vs. 0.27, p = 0.24) (Figure 1B). We found similar results based on quartiles: among patients within the lowest HbF quartile, those within the top PGSasthma quartile had a higher ACS rate than those within the lowest PGSasthma quartile (median 0.34 vs. 0.22, p = 0.049).

FIGURE 1.

FIGURE 1

Open in a new tab

Combination of PRS for asthma and HbF identify a subgroup at high risk for asthma. (A) ACS rate according to z‐score of PGSasthma and HbF value. Higher ACS rates are shown in red and larger circles. (B) ACS rate according to HbF quintile and PGSasthma quintile (1 being the lowest and 5 the highest quintile). (C) Frequency of ACS rate value in the subgroup of patients within low HbF quintile and high or low PGSasthma quintile. ACS, acute chest syndrome, HbF, fetal hemoglobin; PRS, polygenic risk score. [Color figure can be viewed at wileyonlinelibrary.com]

In sum, combining HbF and PGSasthma allows identification of a subgroup of patients at high risk of ACS recurrence.

3.3. Genetically Mediated Contribution of Asthma on ACS Could Be Mediated by Lymphocytes

To gain insight into the pathophysiological mechanisms underlying the effect of asthma on ACS, we used pPGS which was built by assigning asthma GWAS loci to the likely affected cell types based on the intersection between the GWAS loci and epigenomic H3K4me2 marks obtained by ChIP‐seq [32]. The pPGS identified five clusters with enrichment in CD4+ T‐helper lymphocytes (cluster 1), neutrophils, eosinophils and mast cells (cluster 2), CD8+ T cells and B cells (cluster 3), lung epithelial cells (cluster 4) mast cells and activated group 2 innate lymphoid cells (cluster 5). This pPGS is relevant to identify which cell types were involved in the genetically mediated link between asthma and ACS rate.

In the CSSCD, we found that the pPGS for clusters 1 and 3, the two clusters that included lymphocytes, were associated with ACS rate (Table 2). The effects of the two clusters were independent, as confirmed in a multivariate model including the five pPGS clusters. We found the same results for pPGS in the CSSCD using linear regression (Table S3). However, we did not replicate our findings in the GEN‐MOD cohort (Cluster 1: adjusted β = 0.17, SE = 0.16, p = 0.30, Cluster 3: adjusted β = −0.04, SE = 0.16, p = 0.80).

TABLE 2.

Association of the different cluster of the pPGS for asthma with ACS rate.

Single‐cluster analysis All‐clusters analysis
Beta SE p‐value Beta SE p‐value
Cluster 1: CD4+ T‐helper lymphocytes 23.35 9.79 0.02 0.013 0.06 0.02
Cluster 2: neutrophils, eosinophils and mast cells −0.01 0.06 0.84 −0.01 0.06 0.82
Cluster 3: lymphocytes, especially CD8+ T cells and B cells 0.13 0.06 0.03 0.13 0.06 0.03
Cluster 4: lung epithelial cells 0.04 0.05 0.43 0.04 0.05 0.56
Cluster 5: mast cells and activated group 2 innate lymphoid cells −0.06 0.06 0.25 −0.07 0.06 0.24
Open in a new tab

Note: We tested the effect of the clusters of the pPGS for asthma z‐scores on ACS rate in the CSSCD using quasi‐Poisson regression. We adjusted all models on age, sex, HbF and the 10 first principal component. We tested each cluster independently (left) and all together (right). HbF, fetal hemoglobin; SE, standard error.

As cluster 3 includes the well‐known asthma risk locus 17q21 which contains the ORMDL3 gene, we tested the association between SNPs at this locus and ACS rate and found no association (Table S4).

3.4. Asthma and ACS Are Only Partially Overlapping Genetic Conditions

Having shown that PGSasthma was associated with ACS rate, we finally investigated to what extent the genetic propensity for asthma and ACS rate were similar. Such a genetic correlation analysis requires full GWAS results for both traits. This prevents the use of PGSasthma as a comparator, since only the data on the SNPs included in the PGS are available. To limit ancestry‐related bias, we used the African‐ancestry GWAS of asthma from the GBMI [15]. We confirmed that the PGS derived from this GWAS was associated with ACS rate in the CSSCD, validating the results found with PGSasthma (adjusted β = 0.14, SE = 0.10, p = 0.01) [39]. For the GWAS of ACS rate, we performed a GWAS for ACS rate in the CSSCD (Figure S1) [15].

We first compared the overall genetic correlation using GPS and found no global correlation between the two GWAS (GPS = 1.97, p = 0.77) (Figure 2A).

FIGURE 2.

FIGURE 2

Open in a new tab

Genetic correlation between asthma and ACS. (A) Miami plot of the GWAS for asthma and ACS. (B) Comparison of p‐values (upper panel) or effect sizes (lower panel) between asthma and ACS GWAS considering the SNPs statistically suggestive association (< 1 × 10−6) in the asthma (left panel) or ACS (right panel) GWAS. The black line in the upper panel represents the linear regression line. ACS, acute chest syndrome. [Color figure can be viewed at wileyonlinelibrary.com]

We then focused on the statistically suggestive SNPs (defined as p < 1 × 10−6) from the two GWAS (n = 124 in the ACS GWAS and n = 149 in the asthma GWAS). We found that the closest SNPs were separated by 3.0 M base pairs, thus in distinct loci unlikely to be in linkage disequilibrium. We found no positive correlation between the two GWAS. In fact, the p‐values were inversely correlated when considering the statistically suggestive SNPs from the asthma GWAS (r − 0.26, 95% confidence interval [CI] −0.31 to −0.10, p = 0.001) (Figure 2B). The direction of the effects was discordant between the GWAS with a significant depletion for both GWAS (considering the statistically suggestive SNPs from the asthma GWAS: 37% with concordant effect, p = 8.6 × 10−4; considering those from the ACS GWAS: 36% with concordant effect, p = 0.001).

Thus, despite PGSasthma being associated with ACS rate, we found no overall genetic correlation between ACS and asthma, and the most significant loci were different. This suggests that ACS and asthma genetics are only partially shared: some genetic determinants are specific to asthma, while others are specific to ACS (Figure 3).

FIGURE 3.

FIGURE 3

Open in a new tab

Genetic relationship between asthma and ACS. Based on our results, we show how genetic propensity for asthma and ACS rate are related with only partial overlap. Our data do not allow us to precisely define the extent of the overlap nor the specific components. Created in BioRender. Pincez, T. (2025) https://BioRender.com/lf2x0fx. [Color figure can be viewed at wileyonlinelibrary.com]

4. Discussion

Using PGS, we showed that asthma was causally associated with ACS rate, independently of the major genetic modifier HbF. This effect was mainly identified in patients with low HbF levels. The combination of HbF and PGSasthma allowed identification of a subgroup of patients with a high risk of ACS recurrence. Epigenomic‐based pPGS suggested that the lymphocyte‐mediated contribution to asthma was the main driver of ACS risk. Finally, we found that the genetic propensity for asthma and ACS only partially overlaps with unique genetic determinants for each condition.

ACS is driven by vaso‐occlusion within the pulmonary microvasculature [40]. This vaso‐occlusion leads to hypoxia, endothelial injury, vasoconstriction, and eventually pulmonary parenchyma infarction [40, 41]. As a result, local acidosis and inflammation develop. All these consequences of vaso‐occlusion will tend to exacerbate it. This vicious cycle explains the tendency of ACS to spontaneously worsen. Clinical deterioration can be rapid in patients, and ACS is a significant cause of death in SCD [42, 43]. Multiple triggers can contribute to pulmonary vaso‐occlusion [40]. In the patients presenting initially with vaso‐occlusive pain episode, fat embolism or microthrombi from the site of initial vaso‐occlusion can lead to a secondary occurrence of ACS. Infections are another common trigger, especially in children, with both viral and bacterial agents being involved. The role of asthma as a trigger is supported by many epidemiological studies that identified a higher risk for ACS in children but also adults [9, 10, 11, 12, 13]. Most studies also identified a higher risk of vaso‐occlusive pain episodes [13, 44, 45]. As ACS and vaso‐occlusive pain episodes are risk factors for mortality in SCD [42, 46], asthma has also been associated with an increased mortality risk [47, 48]. Our data show that, in addition to being a marker of severity, asthma is causally associated with the repetition of ACS. The fact that we found an association between PGSasthma and ACS rates, but not with occurrence, suggests that the most striking consequence of asthma is SCD is the repetition of ACS. Indeed, around half of patients with SCD will experience at least one ACS in their lifetime, but only a minority will experience several episodes [3]. In the CSSCD, among the 538 patients with ACS, 443 (82%) experienced only one episode [6]. Thus, most ACS are likely caused by a transient trigger, whereas patients with asthma are prone to ACS all their lifetime and will develop several episodes.

The association between PGSasthma and ACS rate, but not occurrence, also provides meaningful insight into an unexplained observation in ACS. Several multicenter studies identified that children hospitalized for an ACS before the age of 4 had a distinct lung evolution than those hospitalized for an ACS at an older age [11, 48]. Early onset ACS was associated with a high risk of ACS recurrence, asthma, and hospitalization for vaso‐occlusive pain or ACS. DeBaun and Strunk hypothesized that early onset ACS resulted in an asthma‐like syndrome, as can occur after viral infection at a young age [7]. This asthma‐like syndrome would further predispose to regional hypoxia and inflammation upon infection or other airway triggers. Our data supports this hypothesis by showing that a genetically determined risk of asthma results in a high risk of recurrence. This suggests that patients with a high propensity for asthma are more likely to develop hypoxia and inflammation upon common airway triggers.

Asthma pathophysiology is complex and heterogeneous across individuals [49], involving several components, such as atopy, bronchial hyperresponsiveness, and airway inflammation and remodeling [50]. The various components involved drive different endotypes with specificity in terms of pathophysiological mechanisms. The Global Initiative for Asthma defines Type 2 inflammation phenotype characterized by interleukin (IL) 4, IL 5, IL 13, and eosinophils, and non‐type 2 inflammation characterized by neutrophils [51]. Asthma GWAS identified many loci involving several immune‐related pathways, such as IL 33 and IL1R1 [52]. Epigenomic‐derived pPGS showed that this genetic propensity for asthma affected different immune cell types and was associated with distinct phenotypes [32]. Our data suggest that lymphocytes could be the main cell types involved in the increased rate of ACS associated with asthma. These results require confirmation, as we could not replicate our findings in the GEN‐MOD cohort, possibly due to the smaller sample size. However, they are consistent with the known pathophysiology of ACS, in which lymphocyte‐associated inflammation is known to play a role, whereas the role of eosinophils is not established [40, 53].

We showed that the genetically determined propensity for asthma, represented by PGSasthma, increased ACS rate. However, overall and local genetic correlation of statistically suggestive SNPs identified no association, as did the analysis of ORMDL3 locus. This contrast helps to define the relationship between asthma and ACS genetics. The validated association between PGSasthma and ACS rate shows that there are shared genetic features between ACS and asthma. We acknowledge that this association would have resulted in a significant, even mild, genetic correlation with a very large sample size. However, our SCD cohort's sample size was large enough to detect a moderate‐to‐high correlation with the GPS approach [34, 35]. As we have confirmed the association with ACS rate of the PGS derived from the GWAS used for genetic correlation, this lack of genetic correlation is not due to different data sets. Moreover, this result is consistent with local genetic correlation analyses of statistically suggestive SNPs and ORMDL3 locus association analyses. The results are also consistent with the pPGS analysis, which suggests that the impact of genetically mediated asthma risk on ACS may be mediated by lymphocytes. Collectively, these results suggest that the genetic overlap between asthma and ACS is only partial. While some of the genetic propensity for asthma appears to increase ACS risk, some components may not be involved. Similarly, some of the genetic propensity for ACS may be unrelated to asthma. This is consistent with the previous identification of SNPs within COMMD7 as associated with ACS rate but not identified in asthma GWAS [19]. To summarize, patients with a high propensity for asthma are prone to develop multiple ACS but several other mechanisms and triggers can lead to ACS in patients with a low propensity for asthma. Asthma is a heterogeneous disease, and the underlying pathophysiology may differ across patients [49, 50]. Larger SCD cohorts will further elucidate the shared and distinct genetic components, as well as the extent of the overlap.

We found that the association of PGSasthma with ACS was mainly in patients with low HbF levels. This suggests that patients with high HbF levels may have protection against the increased risk of ACS associated with a genetic propensity for asthma. HbF is known to be the main modifier of SCD [54]. Higher HbF levels have been associated with a reduced risk of most SCD complications and mortality [6, 42, 46, 55]. However, how HbF interacts with other risk factors remains poorly known. Our results suggest that the consequences on hypoxia and inflammation that can occur in patients with a high propensity for asthma depend on HbF levels.

The high heterogeneity of SCD is poorly understood and a major clinical challenge [1]. Some patients rarely experience any complications, while others have many, as mentioned earlier regarding ACS [6]. Therefore, risk stratification is a major need in SCD care to identify patients requiring specific management or therapy intensification. Treatment options are expanding for SCD, but their role and the patients who would benefit from them are not yet clear [56]. Risk stratification can help identify high‐risk patients for inclusion in clinical trials, an area of active research. With the notable exception of transcranial Doppler to stratify the risk of stroke [57, 58], no tools currently exist in clinical practice. Despite its association with most complications, HbF is not effective for risk stratification, as most patients have low levels of HbF [54], impeding the identification of a high‐risk population. Moreover, the cellular distribution of HbF among erythrocytes varies among patients, limiting the individual correlation between HbF level and SCD severity [54, 59]. PGS have been shown to be effective at identifying individuals at high risk of unfavorable outcomes in many settings, including common conditions such as diabetes but also rare conditions such as telomere biology disorders [60, 61]. In SCD, we have previously shown that a PGS for HbF improved risk modeling for pain rates [38]. However, the clinical usefulness of PGS in ACS remains to be clarified. Here, we showed that the combination of HbF and PGSasthma allowed us to identify a subpopulation at high risk of ACS recurrence after a first episode. Given the potential severity of ACS, this population is at high risk of significant morbidity and mortality. Our results suggest that, after a first episode, PGSasthma could be used to propose treatment intensification in selected patients. While SCD‐specific treatments should be considered, asthma‐directed therapies may also warrant considerations. Further studies should investigate the optimal strategy for this high‐risk population.

Our study has several limitations that should be considered regarding both the SCD cohorts and the asthma GWAS data set used. Our primary SCD dataset, the CSSCD, was derived from a cohort of patients recruited between 1978 and 1988. None of them received hydroxyurea, and medical practices were different at that time. However, since we have confirmed our findings in an external, contemporary cohort from a different geographical region, with a different ACS definition, and which used hydroxyurea for many patients, it is unlikely that these factors led to a false‐positive association. Further studies are needed to determine if hydroxyurea affects the magnitude of the association between the genetic propensity for asthma and ACS. Asthma phenotype and genetic propensity may differ between adults and children. We have shown that our findings were applicable to adults and children with SCD. Although children‐onset and adult‐onset asthma share many genetic similarities, they also have specificities in loci and effect sizes [62, 63]. Our two asthma GWAS used mainly adult cohorts, and further research is needed to clarify the relationship between SCD and children‐onset and adult‐onset asthma. Finally, although we have validated our results using the PGS derived from the African ancestry GBMI, the high diversity among African ancestry individuals may have led to some residual population stratification that was not considered in the adjustment for principal components. Our findings need to be replicated in a single large cohort that includes both SCD and non‐SCD individuals from a similar geographical region. However, such a cohort does not currently exist.

In conclusion, this study paves the way for the use of PGS as a stratification tool in SCD. Specifically, patients with a high genetic propensity for asthma being at higher risk of recurrent ACS if not protected by high HbF levels may benefit from individualized therapy.

Conflicts of Interest

Thomas Pincez: Research funds from Biossil Inc. Pablo Bartolucci: Consultant for ADDMEDICA, NOVARTIS, ROCHE, GBT, Bluebird, EMMAUS, HEMANEXT, AGIOS, Lecture fees for NOVARTIS, ADDMEDICA, JAZZPHARMA, Steering committee for NOVARTIS, ROCHE, ADDMEDICA, PFIZER, Research support from ADDMEDICA, foundation Fabre, NOVARTIS, Bluebird, EMMAUS, Cofounder and CSO of INNOVHEM. The other authors have no conflicts of interest.

Supporting information

Table S1: Characteristics of the cohorts included.

Table S2: Association of the polygenic risk score for asthma and complications.

Table S3: Association of the different cluster of the pPGS for asthma with ACS rate.

Table S4: Association of the 17q21 locus with ACS.

Figure S1: Quantile‐quantile plot for the GWAS of ACS rate in the CSSCD. We used a linear regression model on normalized ACS rate, adjusted for age, sex, and the first 10 principal components. Lambda gc = 1.011.

AJH-101-961-s001.docx (51.6KB, docx)

Acknowledgments

We thank all participants who contributed data to this study. This work was funded by the Canadian Institutes of Health Research (PJT #186159) and the Canada Research Chair program (G. Lettre). M‐A. Legault receives funding from the Institut de Valorisation des Données (IVADO).

Data Availability Statement

The CSSCD genetic dataset is available on the database of Genotypes and Phenotypes (dbGaP: https://www.ncbi.nlm.nih.gov/gap/), accession phs000366.v1.p1. The GEN‐MOD dataset has not been deposited in a public repository because data are not public but are available from the corresponding author on request.

References

  • 1. Ware R. E., de Montalembert M., Tshilolo L., and Abboud M. R., “Sickle Cell Disease,” Lancet 390, no. 10091 (2017): 311–323, 10.1016/S0140-6736(17)30193-9. [DOI] [PubMed] [Google Scholar]
  • 2. Kato G. J., Piel F. B., Reid C. D., et al., “Sickle Cell Disease,” Nature Reviews Disease Primers 4, no. 1 (2018): 1–22, 10.1038/nrdp.2018.10. [DOI] [PubMed] [Google Scholar]
  • 3. Castro O., Brambilla D. J., Thorington B., et al., “The Acute Chest Syndrome in Sickle Cell Disease: Incidence and Risk Factors. The Cooperative Study of Sickle Cell Disease,” Blood 84, no. 2 (1994): 643–649. [PubMed] [Google Scholar]
  • 4. Chaturvedi S., Ghafuri D. L., Glassberg J., Kassim A. A., Rodeghier M., and DeBaun M. R., “Rapidly Progressive Acute Chest Syndrome in Individuals With Sickle Cell Anemia: A Distinct Acute Chest Syndrome Phenotype,” American Journal of Hematology 91, no. 12 (2016): 1185–1190, 10.1002/ajh.24539. [DOI] [PubMed] [Google Scholar]
  • 5. Nouraie M., Darbari D. S., Rana S., et al., “Tricuspid Regurgitation Velocity and Other Biomarkers of Mortality in Children, Adolescents and Young Adults With Sickle Cell Disease in the United States: The PUSH Study,” American Journal of Hematology 95, no. 7 (2020): 766–774, 10.1002/ajh.25799. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Vichinsky E. P., Neumayr L. D., Earles A. N., et al., “Causes and Outcomes of the Acute Chest Syndrome in Sickle Cell Disease,” New England Journal of Medicine 342, no. 25 (2000): 1855–1865, 10.1056/NEJM200006223422502. [DOI] [PubMed] [Google Scholar]
  • 7. DeBaun M. R. and Strunk R. C., “The Intersection Between Asthma and Acute Chest Syndrome in Children With Sickle‐Cell Anaemia,” Lancet 387, no. 10037 (2016): 2545–2553, 10.1016/S0140-6736(16)00145-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Vichinsky E. P., Neumayr L. D., Earles A. N., et al., “Causes and Outcomes of the Acute Chest Syndrome in Sickle Cell Disease. National Acute Chest Syndrome Study Group,” New England Journal of Medicine 342, no. 25 (2000): 1855–1865, 10.1056/NEJM200006223422502. [DOI] [PubMed] [Google Scholar]
  • 9. Boyd J. H., Macklin E. A., Strunk R. C., and DeBaun M. R., “Asthma Is Associated With Acute Chest Syndrome and Pain in Children With Sickle Cell Anemia,” Blood 108, no. 9 (2006): 2923–2927, 10.1182/blood-2006-01-011072. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Knight‐Madden J. M., Forrester T. S., Lewis N. A., and Greenough A., “Asthma in Children With Sickle Cell Disease and Its Association With Acute Chest Syndrome,” Thorax 60, no. 3 (2005): 206–210, 10.1136/thx.2004.029165. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. DeBaun M. R., Rodeghier M., Cohen R., et al., “Factors Predicting Future ACS Episodes in Children With Sickle Cell Anemia,” American Journal of Hematology 89, no. 11 (2014): E212–E217, 10.1002/ajh.23819. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Bernaudin F., Strunk R. C., Kamdem A., et al., “Asthma Is Associated With Acute Chest Syndrome, but Not With an Increased Rate of Hospitalization for Pain Among Children in France With Sickle Cell Anemia: A Retrospective Cohort Study,” Haematologica 93, no. 12 (2008): 1917–1918, 10.3324/haematol.13090. [DOI] [PubMed] [Google Scholar]
  • 13. Knight‐Madden J. M., Barton‐Gooden A., Weaver S. R., Reid M., and Greenough A., “Mortality, Asthma, Smoking and Acute Chest Syndrome in Young Adults With Sickle Cell Disease,” Lung 191, no. 1 (2013): 95–100, 10.1007/s00408-012-9435-3. [DOI] [PubMed] [Google Scholar]
  • 14. Vichinsky E. P., Styles L. A., Colangelo L. H., Wright E. C., Castro O., and Nickerson B., “Acute Chest Syndrome in Sickle Cell Disease: Clinical Presentation and Course. Cooperative Study of Sickle Cell Disease,” Blood 89, no. 5 (1997): 1787–1792. [PubMed] [Google Scholar]
  • 15. Tsuo K., Zhou W., Wang Y., et al., “Multi‐Ancestry Meta‐Analysis of Asthma Identifies Novel Associations and Highlights the Value of Increased Power and Diversity,” Cell Genomics 2, no. 12 (2022): 100212, 10.1016/j.xgen.2022.100212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Kim K. W. and Ober C., “Lessons Learned From GWAS of Asthma,” Allergy, Asthma & Immunology Research 11, no. 2 (2018): 170–187, 10.4168/aair.2019.11.2.170. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Han Y., Jia Q., Jahani P. S., et al., “Genome‐Wide Analysis Highlights Contribution of Immune System Pathways to the Genetic Architecture of Asthma,” Nature Communications 11, no. 1 (2020): 1776, 10.1038/s41467-020-15649-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Pincez T., Ashley‐Koch A. E., Lettre G., and Telen M. J., “Genetic Modifiers of Sickle Cell Disease,” Hematology/Oncology Clinics of North America 36, no. 6 (2022): 1097–1124, 10.1016/j.hoc.2022.06.006. [DOI] [PubMed] [Google Scholar]
  • 19. Galarneau G., Coady S., Garrett M. E., et al., “Gene‐Centric Association Study of Acute Chest Syndrome and Painful Crisis in Sickle Cell Disease Patients,” Blood 122, no. 3 (2013): 434–442, 10.1182/blood-2013-01-478776. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Farber M. D., Koshy M., and Kinney T. R., “Cooperative Study of Sickle Cell Disease: Demographic and Socioeconomic Characteristics of Patients and Families With Sickle Cell Disease,” Journal of Chronic Diseases 38, no. 6 (1985): 495–505, 10.1016/0021-9681(85)90033-5. [DOI] [PubMed] [Google Scholar]
  • 21. Castro O., Brambilla D. J., Thorington B., et al., “The Acute Chest Syndrome in Sickle Cell Disease: Incidence and Risk Factors. The Cooperative Study of Sickle Cell Disease,” Blood 84, no. 2 (1994): 643–649. [PubMed] [Google Scholar]
  • 22. Bartolucci P., Brugnara C., Teixeira‐Pinto A., et al., “Erythrocyte Density in Sickle Cell Syndromes Is Associated With Specific Clinical Manifestations and Hemolysis,” Blood 120, no. 15 (2012): 3136–3141, 10.1182/blood-2012-04-424184. [DOI] [PubMed] [Google Scholar]
  • 23. Ilboudo Y., Bartolucci P., Rivera A., et al., “Genome‐Wide Association Study of Erythrocyte Density in Sickle Cell Disease Patients,” Blood Cells, Molecules & Diseases 65 (2017): 60–65, 10.1016/j.bcmd.2017.05.005. [DOI] [PubMed] [Google Scholar]
  • 24. Bartolucci P., Habibi A., Khellaf M., et al., “Score Predicting Acute Chest Syndrome During Vaso‐Occlusive Crises in Adult Sickle‐Cell Disease Patients,” eBioMedicine 10 (2016): 305–311, 10.1016/j.ebiom.2016.06.038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Bae H. T., Baldwin C. T., Sebastiani P., et al., “Meta‐Analysis of 2040 Sickle Cell Anemia Patients: BCL11A and HBS1L‐MYB Are the Major Modifiers of HbF in African Americans,” Blood 120, no. 9 (2012): 1961–1962, 10.1182/blood-2012-06-432849. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Sollis E., Mosaku A., Abid A., et al., “The NHGRI‐EBI GWAS Catalog: Knowledgebase and Deposition Resource,” Nucleic Acids Research 51, no. D1 (2023): D977–D985, 10.1093/nar/gkac1010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Truong B., Hull L. E., Ruan Y., et al., “Integrative Polygenic Risk Score Improves the Prediction Accuracy of Complex Traits and Diseases,” Cell Genomics 4, no. 4 (2024): 100523, 10.1016/j.xgen.2024.100523. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Martin A. R., Kanai M., Kamatani Y., Okada Y., Neale B. M., and Daly M. J., “Clinical Use of Current Polygenic Risk Scores May Exacerbate Health Disparities,” Nature Genetics 51, no. 4 (2019): 584–591, 10.1038/s41588-019-0379-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Ding Y., Hou K., Xu Z., et al., “Polygenic Scoring Accuracy Varies Across the Genetic Ancestry Continuum,” Nature 618, no. 7966 (2023): 774–781, 10.1038/s41586-023-06079-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Purcell S., Neale B., Todd‐Brown K., et al., “PLINK: A Tool Set for Whole‐Genome Association and Population‐Based Linkage Analyses,” American Journal of Human Genetics 81, no. 3 (2007): 559–575, 10.1086/519795. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Lettre G., Sankaran V. G., Bezerra M. A. C., et al., “DNA Polymorphisms at the BCL11A, HBS1L‐MYB, and Beta‐Globin Loci Associate With Fetal Hemoglobin Levels and Pain Crises in Sickle Cell Disease,” Proceedings of the National Academy of Sciences of the United States of America 105, no. 33 (2008): 11869–11874, 10.1073/pnas.0804799105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Stikker B., Trap L., Sedaghati‐Khayat B., et al., “Epigenomic Partitioning of a Polygenic Risk Score for Asthma Reveals Distinct Genetically Driven Disease Pathways,” European Respiratory Journal 64, no. 2 (2024): 2302059, 10.1183/13993003.02059-2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Ochoa D., Hercules A., Carmona M., et al., “The Next‐Generation Open Targets Platform: Reimagined, Redesigned, Rebuilt,” Nucleic Acids Research 51, no. D1 (2023): D1353–D1359, 10.1093/nar/gkac1046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Li Y. R., Li J., Zhao S. D., et al., “Meta‐Analysis of Shared Genetic Architecture Across Ten Pediatric Autoimmune Diseases,” Nature Medicine 21, no. 9 (2015): 1018–1027, 10.1038/nm.3933. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Willis T. W. and Wallace C., “Accurate Detection of Shared Genetic Architecture From GWAS Summary Statistics in the Small‐Sample Context,” PLoS Genetics 19, no. 8 (2023): e1010852, 10.1371/journal.pgen.1010852. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Quinlan A. R. and Hall I. M., “BEDTools: A Flexible Suite of Utilities for Comparing Genomic Features,” Bioinformatics 26, no. 6 (2010): 841–842, 10.1093/bioinformatics/btq033. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Benjamini Y. and Hochberg Y., “Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing,” Journal of the Royal Statistical Society. Series B, Statistical Methodology 57, no. 1 (1995): 289–300. [Google Scholar]
  • 38. Pincez T., Lo K. S., D'Orengiani A. L. P. H. D., et al., “Variation and Impact of Polygenic Hematological Traits in Monogenic Sickle Cell Disease,” Haematologica 108, no. 3 (2023): 870–881, 10.3324/haematol.2022.281180. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Tsuo K., Zhou W., Wang Y., et al., “Multi‐Ancestry Meta‐Analysis of Asthma Identifies Novel Associations and Highlights the Value of Increased Power and Diversity,” Cell Genomics 2, no. 12 (2022): 100212, 10.1016/j.xgen.2022.100212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Gladwin M. T. and Vichinsky E., “Pulmonary Complications of Sickle Cell Disease,” New England Journal of Medicine 359, no. 21 (2008): 2254–2265, 10.1056/NEJMra0804411. [DOI] [PubMed] [Google Scholar]
  • 41. Gladwin M. T. and Rodgers G. P., “Pathogenesis and Treatment of Acute Chest Syndrome of Sickle‐Cell Anaemia,” Lancet 355, no. 9214 (2000): 1476–1478, 10.1016/S0140-6736(00)02157-7. [DOI] [PubMed] [Google Scholar]
  • 42. Platt O. S., Brambilla D. J., Rosse W. F., et al., “Mortality in Sickle Cell Disease. Life Expectancy and Risk Factors for Early Death,” New England Journal of Medicine 330, no. 23 (1994): 1639–1644, 10.1056/NEJM199406093302303. [DOI] [PubMed] [Google Scholar]
  • 43. Darbari D. S., Kple‐Faget P., Kwagyan J., Rana S., Gordeuk V. R., and Castro O., “Circumstances of Death in Adult Sickle Cell Disease Patients,” American Journal of Hematology 81, no. 11 (2006): 858–863, 10.1002/ajh.20685. [DOI] [PubMed] [Google Scholar]
  • 44. Cohen R. T., Madadi A., Blinder M. A., DeBaun M. R., Strunk R. C., and Field J. J., “Recurrent, Severe Wheezing Is Associated With Morbidity and Mortality in Adults With Sickle Cell Disease,” American Journal of Hematology 86, no. 9 (2011): 756–761, 10.1002/ajh.22098. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Glassberg J. A., Chow A., Wisnivesky J., Hoffman R., Debaun M. R., and Richardson L. D., “Wheezing and Asthma Are Independent Risk Factors for Increased Sickle Cell Disease Morbidity,” British Journal of Haematology 159, no. 4 (2012): 472–479, 10.1111/bjh.12049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Platt O. S., Thorington B. D., Brambilla D. J., et al., “Pain in Sickle Cell Disease,” New England Journal of Medicine 325, no. 1 (1991): 11–16, 10.1056/NEJM199107043250103. [DOI] [PubMed] [Google Scholar]
  • 47. Boyd J. H., Macklin E. A., Strunk R. C., and DeBaun M. R., “Asthma Is Associated With Increased Mortality in Individuals With Sickle Cell Anemia,” Haematologica 92, no. 8 (2007): 1115–1118. [DOI] [PubMed] [Google Scholar]
  • 48. Vance L. D., Rodeghier M., Cohen R. T., et al., “Increased Risk of Severe Vaso‐Occlusive Episodes After Initial Acute Chest Syndrome in Children With Sickle Cell Anemia Less Than 4 Years Old: Sleep and Asthma Cohort,” American Journal of Hematology 90, no. 5 (2015): 371–375, 10.1002/ajh.23959. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Kuruvilla M. E., Lee F. E. H., and Lee G. B., “Understanding Asthma Phenotypes, Endotypes, and Mechanisms of Disease,” Clinical Reviews in Allergy and Immunology 56, no. 2 (2019): 219–233, 10.1007/s12016-018-8712-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Holgate S. T., Wenzel S., Postma D. S., Weiss S. T., Renz H., and Sly P. D., “Asthma,” Nature Reviews Disease Primers 1, no. 1 (2015): 1–22, 10.1038/nrdp.2015.25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51. Global Initiative for Asthma , “Global Strategy for Asthma Management and Prevention,” 2023.
  • 52. Stikker B. S., Hendriks R. W., and Stadhouders R., “Decoding the Genetic and Epigenetic Basis of Asthma,” Allergy 78, no. 4 (2023): 940–956, 10.1111/all.15666. [DOI] [PubMed] [Google Scholar]
  • 53. Bhasin N. and Sarode R., “Acute Chest Syndrome in Sickle Cell Disease,” Transfusion Medicine Reviews 37, no. 3 (2023): 150755, 10.1016/j.tmrv.2023.150755. [DOI] [PubMed] [Google Scholar]
  • 54. Steinberg M. H., “Fetal Hemoglobin in Sickle Cell Anemia,” Blood 136, no. 21 (2020): 2392–2400, 10.1182/blood.2020007645. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55. Ohene‐Frempong K., Weiner S. J., Sleeper L. A., et al., “Cerebrovascular Accidents in Sickle Cell Disease: Rates and Risk Factors,” Blood 91, no. 1 (1998): 288–294. [PubMed] [Google Scholar]
  • 56. Salinas Cisneros G. and Thein S. L., “Recent Advances in the Treatment of Sickle Cell Disease,” Frontiers in Physiology 11 (2020): 11, 10.3389/fphys.2020.00435. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57. Adams R. J., McKie V. C., Hsu L., et al., “Prevention of a First Stroke by Transfusions in Children With Sickle Cell Anemia and Abnormal Results on Transcranial Doppler Ultrasonography,” New England Journal of Medicine 339, no. 1 (1998): 5–11, 10.1056/NEJM199807023390102. [DOI] [PubMed] [Google Scholar]
  • 58. Verduzco L. A. and Nathan D. G., “Sickle Cell Disease and Stroke,” Blood 114, no. 25 (2009): 5117–5125, 10.1182/blood-2009-05-220921. [DOI] [PubMed] [Google Scholar]
  • 59. Khandros E., Huang P., Peslak S. A., et al., “Understanding Heterogeneity of Fetal Hemoglobin Induction Through Comparative Analysis of F and A Erythroblasts,” Blood 135, no. 22 (2020): 1957–1968, 10.1182/blood.2020005058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60. Ortega H. I., Udler M. S., Gloyn A. L., and Sharp S. A., “Diabetes Mellitus Polygenic Risk Scores: Heterogeneity and Clinical Translation,” Nature Reviews. Endocrinology 21, no. 9 (2025): 530–545, 10.1038/s41574-025-01132-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61. Poeschla M., Arora U. P., Walne A., et al., “Polygenic Modifiers Impact Penetrance and Expressivity in Telomere Biology Disorders,” Journal of Clinical Investigation 135, no. 16 (2025): e191107, 10.1172/JCI191107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62. Pividori M., Schoettler N., Nicolae D. L., Ober C., and Im H. K., “Shared and Distinct Genetic Risk Factors for Childhood‐Onset and Adult‐Onset Asthma: Genome‐Wide and Transcriptome‐Wide Studies,” Lancet Respiratory Medicine 7, no. 6 (2019): 509–522, 10.1016/S2213-2600(19)30055-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63. Ferreira M. A. R., Mathur R., Vonk J. M., et al., “Genetic Architectures of Childhood‐ and Adult‐Onset Asthma Are Partly Distinct,” American Journal of Human Genetics 104, no. 4 (2019): 665–684, 10.1016/j.ajhg.2019.02.022. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Table S1: Characteristics of the cohorts included.

Table S2: Association of the polygenic risk score for asthma and complications.

Table S3: Association of the different cluster of the pPGS for asthma with ACS rate.

Table S4: Association of the 17q21 locus with ACS.

Figure S1: Quantile‐quantile plot for the GWAS of ACS rate in the CSSCD. We used a linear regression model on normalized ACS rate, adjusted for age, sex, and the first 10 principal components. Lambda gc = 1.011.

AJH-101-961-s001.docx (51.6KB, docx)

Data Availability Statement

The CSSCD genetic dataset is available on the database of Genotypes and Phenotypes (dbGaP: https://www.ncbi.nlm.nih.gov/gap/), accession phs000366.v1.p1. The GEN‐MOD dataset has not been deposited in a public repository because data are not public but are available from the corresponding author on request.

Articles from American Journal of Hematology are provided here courtesy of Wiley

RESOURCES