Pharmacogenomics of Anti-Inflammatory Therapy: Toward Personalized Use of Nonsteroidal Anti-Inflammatory Drugs, Corticosteroids, and Biologics

1. Introduction

For several decades, individualized prescribing has been recognized as a fundamental strategy for maximizing the efficacy, tolerability, and safety of pharmacological therapies. Drug response is influenced by a wide range of patient-specific factors, including age, sex, disease phenotype and severity, environmental exposures, dietary habits, organ function, and developmental stage. These determinants can substantially affect both pharmacokinetics and pharmacodynamics. Consequently, reliance on standard dosing regimens without adequate consideration of individual variability may lead to clinically relevant deviations from expected therapeutic outcomes, including therapeutic failure, excessive drug exposure, or an increased risk of adverse effects [1]. This principle is well established in routine clinical practice. Dose adjustment is commonly required in patients with renal impairment to prevent drug accumulation and toxicity [2], in those with advanced liver disease because of altered drug metabolism and clearance [3], and in neonates and young infants, in whom maturational differences in absorption, distribution, metabolism, and excretion significantly influence drug disposition [4,5].

More recently, advances in pharmacogenomics have further expanded the concept of precision medicine by demonstrating that genetic variability represents a major determinant of interindividual differences in drug response. Polymorphisms in genes encoding drug-metabolizing enzymes, transporters, pharmacological targets, receptors, and downstream signaling pathways may markedly influence both therapeutic efficacy and susceptibility to toxicity. In selected clinical settings, inherited genetic factors account for a substantial proportion of variability in drug response, in some cases approaching the majority of observed interindividual differences [6].

Anti-inflammatory therapies, including nonsteroidal anti-inflammatory drugs (NSAIDs), corticosteroids, and biologic agents, are central to both adult and pediatric clinical practice because of their widespread use in acute and chronic inflammatory disorders. Despite their established therapeutic value, these drugs are characterized by considerable variability in clinical response and safety. A relevant proportion of patients fail to achieve adequate disease control, require treatment escalation or switching, or experience adverse drug reactions. Increasing evidence indicates that genetic variability contributes meaningfully to this heterogeneity. Variants affecting drug metabolism, transport, immune regulation, inflammatory pathways, and pharmacological targets may modulate treatment efficacy, drug exposure, immunogenicity, and the risk of toxicity.

In this context, pharmacogenomic approaches offer a promising strategy to optimize anti-inflammatory therapy. The identification of clinically relevant genetic variants may support more precise drug selection, individualized dosing, early recognition of patients at increased risk of adverse events, and improved prediction of therapeutic response. This approach may be particularly relevant in pediatrics, where genetic determinants interact with developmental changes in drug disposition and immune function. The aim of this narrative review is therefore to critically evaluate the role of genetic determinants in modulating the effects of anti-inflammatory drugs, with particular attention to NSAIDs, corticosteroids, and biologic therapies, and to explore the potential integration of pharmacogenomics into routine clinical practice.

2. Methods

A structured literature search was conducted in PubMed and Embase to identify relevant publications addressing the pharmacogenomics of anti-inflammatory therapies. The search included studies published between January 1, 2000, and January 31, 2026, and was complemented by a manual review of relevant surveillance reports and guidance documents from the European Centre for Disease Prevention and Control, the UK Health Security Agency, and the Centers for Disease Control and Prevention. The search strategy combined keywords and Medical Subject Headings (MeSH) related to pharmacogenomics, pediatrics, anti-inflammatory drugs, nonsteroidal anti-inflammatory drugs (NSAIDs), corticosteroids, and biologic therapies. Additional terms, including drug metabolism, genetic polymorphisms, adverse drug reactions, pharmacokinetics, pharmacodynamics, and precision medicine, were used to broaden retrieval and ensure comprehensive coverage.

Eligible publications included clinical studies, phase II–III clinical trials, randomized controlled trials, observational and multicenter studies, systematic reviews, meta-analyses, practice guidelines, and relevant narrative reviews. Only English-language publications were considered. Articles were screened for relevance according to their focus on the influence of genetic variability on drug efficacy, safety, pharmacokinetics, pharmacodynamics, and interindividual treatment response in the context of anti-inflammatory therapy.

Particular attention was given to studies evaluating clinically relevant genetic determinants, including variants in genes encoding drug-metabolizing enzymes, especially members of the cytochrome P450 family, pharmacological targets such as cyclooxygenases and the glucocorticoid receptor, drug transporters, cytokines, immune-regulatory pathways, and molecules involved in biologic drug response and immunogenicity. Because the review also addresses pediatric implications, studies examining developmental pharmacology and the interaction between ontogeny and pharmacogenomic variability were included when relevant. Reference lists of selected articles were manually screened to identify additional pertinent publications.

Data were synthesized qualitatively. Priority was given to findings with clinical relevance, biological plausibility, consistency across studies, and potential implications for individualized therapy. Given the heterogeneity of study designs, populations, interventions, genetic variants, and outcome measures, a formal meta-analysis was not performed.

3. Nonsteroidal Anti-Inflammatory Drugs

3.1. General Characteristics

NSAIDs are among the most widely used pharmacological agents in both adult and pediatric populations worldwide [7,8]. They are primarily prescribed for the symptomatic treatment of fever, pain, and inflammation, and represent first-line therapy in several acute and chronic inflammatory conditions. Their extensive use reflects their broad availability, rapid onset of action, and generally favorable efficacy profile when administered at recommended doses. NSAIDs account for a substantial proportion of global drug consumption; it has been estimated that they represent almost 20% of all prescription drug use and that approximately 30 million people worldwide take NSAIDs daily [9].

NSAIDs are also frequently used in children and adolescents. In a combined primary care database study conducted in Italy, the Netherlands, and the United Kingdom, NSAIDs were the tenth most frequently prescribed drugs among children aged 2–11 years, with 33 users per 1000 person-years, and the sixth most frequently prescribed drugs among adolescents aged 12–18 years, with 57 users per 1000 person-years [10]. Among available agents, ibuprofen remains the most commonly prescribed NSAID in pediatric practice, followed by diclofenac and naproxen, reflecting their broad clinical indications, established antipyretic and analgesic efficacy, and widespread accessibility [7].

The therapeutic effects of NSAIDs are mainly mediated through inhibition of cyclooxygenase enzymes, COX-1 and COX-2, which catalyze the conversion of arachidonic acid into prostaglandins and thromboxanes. These lipid mediators play central roles in inflammation, pain sensitization, fever generation, platelet aggregation, vascular homeostasis, renal blood flow regulation, and gastrointestinal mucosal protection. Differences in selectivity for COX-1 and COX-2 partly explain variability in both efficacy and toxicity across NSAID classes. Nonselective NSAIDs inhibit both COX isoforms, whereas COX-2 selective inhibitors were developed to reduce gastrointestinal toxicity while preserving anti-inflammatory efficacy. However, suppression of prostacyclin production by COX-2 inhibition may contribute to an increased cardiovascular risk in susceptible individuals.

Although NSAIDs are generally safe when used appropriately, they are associated with clinically relevant adverse effects, particularly in the setting of high doses, prolonged exposure, dehydration, comorbidities, concomitant nephrotoxic or anticoagulant drugs, or underlying renal, gastrointestinal, or cardiovascular risk factors [7,11,12]. The most common toxicities involve the gastrointestinal tract, kidney, cardiovascular system, and immune-mediated hypersensitivity pathways. Gastrointestinal adverse events range from dyspepsia to ulceration and bleeding; renal toxicity may include reduced glomerular filtration, sodium and water retention, hypertension, and acute kidney injury; and hypersensitivity reactions may manifest as urticaria, angioedema, bronchospasm, or systemic reactions [12].

Importantly, both therapeutic response and susceptibility to adverse events vary substantially among individuals receiving the same NSAID at comparable doses. This interindividual variability reflects the interaction of clinical factors, such as age, inflammatory status, comorbidities, organ function, and concomitant medications, with genetic determinants that influence NSAID pharmacokinetics and pharmacodynamics. Pharmacogenomics has emerged as a major contributor to this variability, particularly through polymorphisms affecting drug-metabolizing enzymes, transporters, and pharmacological targets [7,8]. In pediatric populations, genetic variability is further modulated by developmental changes in drug absorption, distribution, metabolism, and excretion, making genotype–phenotype relationships potentially age-dependent [13,14].

3.2. Genetic Determinants of NSAID Metabolism

The cytochrome P450 2C9 enzyme, encoded by the CYP2C9 gene, is the principal hepatic enzyme involved in the metabolism of several commonly used NSAIDs, including ibuprofen, diclofenac, naproxen, celecoxib, meloxicam, flurbiprofen, and lornoxicam [8]. CYP2C9 contributes to oxidative metabolism and clearance of these agents, thereby influencing systemic drug exposure, elimination half-life, and the risk of concentration-dependent toxicity.

The CYP2C9 gene is highly polymorphic, with more than 60 identified variant alleles and multiple suballeles, many of which have clinically relevant functional consequences [15]. Allelic frequencies differ substantially across populations, contributing to interindividual and interethnic variability in NSAID disposition and clinical outcomes. Functionally, CYP2C9 alleles are commonly categorized as normal-function, decreased-function, or no-function alleles. CYP2C9*1 is considered the reference normal-function allele, whereas variants such as *2, *5, *8, and *11 are associated with reduced enzymatic activity, and alleles such as *3, *6, and *13 may result in markedly reduced or absent function [16].

Carriers of decreased- or no-function CYP2C9 alleles metabolize CYP2C9-dependent NSAIDs more slowly, resulting in decreased clearance, higher plasma concentrations, prolonged elimination, and increased systemic exposure [8]. In adults, these pharmacokinetic changes have been associated with a greater risk of dose-related NSAID toxicity, particularly gastrointestinal bleeding, renal adverse events, and, for selected agents, cardiovascular complications [7,8]. Conversely, altered metabolism may also affect therapeutic response, since insufficient or excessive exposure can both compromise the benefit–risk balance of NSAID therapy.

Ibuprofen provides a clear example of the functional impact of CYP2C9 variability. Kirchheiner et al. demonstrated that individuals homozygous for CYP2C93 had approximately a 50% reduction in clearance of both racemic ibuprofen and its active S-enantiomer compared with CYP2C91/*1 carriers [17]. Similarly, García-Martín et al. reported a gene–dose relationship, with ibuprofen metabolic clearance progressively decreasing across CYP2C9 genotypes: 4.43 L/h in *1/*1 carriers, 3.26 L/h in *1/*2 carriers, 2.91 L/h in *1/*3 carriers, 2.05 L/h in *2/*2 carriers, 1.83 L/h in *2/*3 carriers, and 1.13 L/h in *3/*3 carriers [18]. These findings confirm that CYP2C9 polymorphisms can substantially modify NSAID pharmacokinetics and support the biological rationale for genotype-guided prescribing.

Although pediatric data remain more limited than adult data, available evidence suggests that CYP2C9 genotype may also influence NSAID pharmacokinetics and clinical response in children [19]. For example, in African American children and adolescents with sickle cell disease, approximately 30% were classified as intermediate metabolizers because of reduced-function CYP2C9 alleles, most commonly *1/*8. This finding is clinically relevant because standard dosing in intermediate or poor metabolizers may increase the risk of drug accumulation and adverse events, particularly when NSAIDs are administered repeatedly or in the context of dehydration, vaso-occlusive crises, or renal vulnerability.

However, the translation of CYP2C9 pharmacogenomics into pediatric practice is complicated by ontogeny. CYP2C9 expression is developmentally regulated. Enzyme levels are estimated to be approximately 1–2% of adult activity during early fetal life, increase to around 30% by late gestation, and rise further after birth [20]. Considerable interindividual variability persists during the first months of life, whereas from approximately 5 months of age through adolescence, CYP2C9 activity approaches adult levels, with reduced variability [20]. This developmental pattern implies that the phenotypic consequences of reduced-function CYP2C9 alleles may be attenuated in neonates, in whom baseline enzyme activity is already low, but may become increasingly relevant in older infants, children, and adolescents as hepatic metabolic capacity matures [13,14].

Genotype-guided NSAID prescribing therefore represents a plausible strategy for improving safety in patients at increased risk of toxicity, particularly those requiring prolonged or high-dose therapy, those receiving NSAIDs with longer elimination half-lives, or those with additional clinical risk factors [8,21]. Celecoxib, for example, has a longer half-life of approximately 11–16 hours, compared with shorter half-lives for agents such as ibuprofen, flurbiprofen, and lornoxicam [22]. In individuals carrying loss-of-function CYP2C9 alleles, higher plasma concentrations of CYP2C9-metabolized NSAIDs are expected, supporting the use of lower starting doses, careful titration, selection of alternative agents, and intensified monitoring for adverse events [23,24].

Current pharmacogenomic recommendations suggest that, for patients with markedly reduced CYP2C9 activity, clinicians may consider initiating therapy at a reduced dose or selecting an NSAID less dependent on CYP2C9 metabolism [8,23]. Potential alternatives include agents with minimal in vivo dependence on CYP2C9, such as aspirin, ketorolac, metamizole, sulindac, etoricoxib, parecoxib, or valdecoxib, although the clinical appropriateness of each option must be evaluated according to age, indication, availability, contraindications, and safety profile. While routine pre-prescription pharmacogenomic testing for all NSAID users is not yet standard practice, targeted or pre-emptive testing may be particularly valuable in selected high-risk populations.

The feasibility and clinical value of pre-emptive pharmacogenomic approaches have been demonstrated in implementation studies. A large multicenter European study using a 12-gene pharmacogenetic panel, including CYP2C9, enabled genotype-guided prescribing and reduced clinically relevant adverse drug reactions compared with standard care [25]. Similarly, institutional pharmacogenomics programs in the United States have incorporated CYP2C9 results into electronic clinical decision support systems to guide dosing for ibuprofen, celecoxib, and meloxicam [26]. Real-world evaluations of NSAID pharmacogenomic alerts have shown moderate clinician uptake, with common responses including drug discontinuation, dose modification, or selection of an alternative analgesic [27]. These findings highlight both the promise of pharmacogenomic implementation and the practical challenges of integrating genetic information into routine prescribing workflows.

3.3. Genes Affecting Cyclooxygenase Activity and NSAID Pharmacodynamics

In addition to genes involved in drug metabolism, genetic variation in pharmacodynamic pathways may influence NSAID efficacy and toxicity. The prostaglandin-endoperoxide synthase genes PTGS1 and PTGS2 encode COX-1 and COX-2, respectively, the principal pharmacological targets of NSAIDs. These enzymes catalyze the conversion of arachidonic acid into prostaglandin H2, the precursor of prostaglandins, prostacyclin, and thromboxane. Through these mediators, COX enzymes regulate inflammation, pain, fever, platelet aggregation, vascular tone, gastrointestinal mucosal integrity, and renal perfusion.

Functional polymorphisms in PTGS1 and PTGS2 may alter enzyme expression, activity, or inducibility, thereby modifying the pharmacodynamic effects of NSAIDs. Variability in these genes may influence analgesic and anti-inflammatory efficacy, platelet inhibition, cardiovascular risk, gastrointestinal toxicity, and hypersensitivity susceptibility [7]. For example, the PTGS1 variant rs10306114 has been associated with altered platelet inhibition and variability in aspirin responsiveness, potentially contributing to residual thrombotic risk among aspirin-treated patients [28].

PTGS2 polymorphisms have also been implicated in variability in NSAID response. The functional promoter variant −765G>C, also known as rs20417, disrupts a binding site for the Sp1 transcription factor and has been associated with reduced COX-2 transcription. This reduction may decrease downstream prostaglandin and prostacyclin production. In large prospective cohorts, carriers of the minor allele showed lower thromboxane and prostacyclin metabolite levels and a reduced risk of major cardiovascular events, particularly among aspirin users [29]. However, reduced COX-2 activity may also influence analgesic response. In pharmacogenetic studies, rs20417 CC carriers demonstrated diminished pain relief with ibuprofen, suggesting that genetic reduction in COX-2 expression may alter the pharmacological effect of COX inhibition [30].

Other PTGS2 variants further illustrate the complexity of NSAID pharmacodynamics. The rs689466 polymorphism, located in the promoter region, has been associated with susceptibility to inflammatory bowel disease and early-onset inflammatory phenotypes [31]. The rs5275 variant has been linked to gene–environment interactions involving NSAID exposure, diet, and colorectal cancer risk. These findings suggest that PTGS polymorphisms may influence not only drug response but also baseline inflammatory risk, disease phenotype, and the balance between therapeutic and adverse effects.

From a safety perspective, genetic variability in prostaglandin pathways may contribute to heterogeneity in cardiovascular and gastrointestinal toxicity, particularly with COX-2 selective inhibitors such as celecoxib. Reduced prostacyclin production, altered thromboxane–prostacyclin balance, and differences in platelet function may all influence cardiovascular risk. Similarly, genetically mediated changes in prostaglandin synthesis may affect gastrointestinal mucosal protection and renal hemodynamic responses. Although adult studies provide evidence supporting the role of PTGS variants in modulating NSAID efficacy and toxicity, pediatric evidence remains limited.

In children, the clinical relevance of PTGS polymorphisms is plausible but still insufficiently defined. NSAID response in pediatric populations may be influenced by developmental regulation of prostaglandin synthesis, age-related differences in COX expression, renal maturation, and immune pathway development. NSAID-related hypersensitivity, which is a clinically relevant problem in children, has been linked in part to altered arachidonic acid metabolism and imbalance between prostaglandin and leukotriene pathways [32,33]. Since COX enzymes play a central role in these pathways, genetic variants affecting COX expression or activity may contribute to susceptibility to hypersensitivity reactions, although further pediatric studies are needed to clarify these associations.

3.4. Transporter Genes and NSAID Disposition

Drug transporters represent another important component of NSAID pharmacogenomics. Among these, ATP-binding cassette subfamily B member 1, encoded by ABCB1, has received particular attention [12]. ABCB1 encodes P-glycoprotein, an efflux transporter expressed in several tissues relevant to drug disposition, including the intestinal epithelium, hepatobiliary canaliculi, renal tubules, placenta, and blood–brain barrier. By limiting intestinal absorption, promoting biliary and renal excretion, and restricting tissue penetration, P-glycoprotein may influence both systemic exposure and organ-specific NSAID concentrations.

Common ABCB1 polymorphisms, including C3435T, G2677T/A, and C1236T, have been associated with altered P-glycoprotein expression or function [12]. Reduced transporter activity may increase systemic concentrations or tissue exposure to P-glycoprotein substrates, potentially enhancing analgesic efficacy but also increasing the risk of adverse effects, including central nervous system symptoms, renal accumulation, or other dose-related toxicities. Conversely, increased P-glycoprotein activity could reduce bioavailability or tissue penetration, potentially contributing to reduced efficacy.

Although the clinical relevance of ABCB1 variation for NSAID therapy remains less well established than that of CYP2C9, transporter polymorphisms may be particularly important when combined with metabolic variants, high-dose or prolonged NSAID exposure, renal impairment, or polypharmacy. Future studies should evaluate transporter and metabolic gene variants together, rather than in isolation, to better capture the polygenic nature of NSAID response.

3.5. Clinical Implications of NSAID Pharmacogenomics

Table 1 summarizes the main pharmacogenomic determinants of response to NSAIDs.

The pharmacogenomic determinants of NSAID response highlight the limitations of a uniform prescribing approach. CYP2C9 genotype has the strongest current evidence for clinical implementation, particularly for NSAIDs that are predominantly metabolized by this enzyme. Patients carrying decreased- or no-function alleles may require lower starting doses, slower titration, closer monitoring, or selection of an alternative agent. Such considerations are especially relevant in individuals with additional risk factors for adverse events, including older age, renal impairment, gastrointestinal disease, cardiovascular risk, concomitant anticoagulant or antiplatelet therapy, dehydration, or need for repeated or long-term NSAID use.

In pediatrics, the potential value of NSAID pharmacogenomics is particularly compelling but remains incompletely defined. Children differ from adults not only in body size but also in organ maturation, enzyme ontogeny, inflammatory physiology, and disease indications for NSAID use. Therefore, pharmacogenomic interpretation in children should account for both genotype and developmental stage. A reduced-function CYP2C9 allele may have different clinical implications in a neonate, infant, school-aged child, or adolescent. Similarly, variants affecting prostaglandin pathways may interact with developmental differences in renal blood flow, immune response, and inflammatory regulation.

At present, routine pharmacogenomic testing before NSAID administration is not universally recommended. However, targeted testing may be considered in selected clinical scenarios, including patients with a history of NSAID-related toxicity, those requiring long-term or high-dose treatment, those prescribed NSAIDs with substantial CYP2C9 dependence, and individuals already undergoing pre-emptive pharmacogenomic panel testing for other indications. Integration of pharmacogenomic information into electronic prescribing systems, together with clinician education and decision support, will be essential for translating genetic findings into safer and more effective NSAID therapy.

Overall, NSAID pharmacogenomics illustrates the broader promise of precision anti-inflammatory therapy. Variants in CYP2C9, PTGS1, PTGS2, ABCB1, and related pathways contribute to variability in drug exposure, therapeutic response, and adverse-event risk. While CYP2C9 currently represents the most actionable marker, future implementation will likely require multigene models that incorporate metabolic, transporter, pharmacodynamic, developmental, and clinical variables. Such approaches may improve the ability to identify patients most likely to benefit from NSAIDs, those requiring dose adjustment, and those for whom alternative analgesic or anti-inflammatory strategies may be safer.

4. Corticosteroids

Since the earliest clinical use of corticosteroids, it has been evident that patients with apparently similar diseases may show markedly different therapeutic responses and adverse-effect profiles, even when treated with identical doses and durations. This interindividual variability has been documented across several inflammatory and immune-mediated conditions and across different age groups, with particularly strong evidence in both pediatric and adult asthma. Asthma represents one of the most extensively studied models of corticosteroid responsiveness. In children and adults treated with inhaled corticosteroids (ICS), clinical outcomes range from near-complete disease control to persistent symptoms, impaired lung function, and recurrent exacerbations despite apparently appropriate therapy [34].

Several clinical and biological factors contribute to this heterogeneity. Poor adherence [35], incorrect inhaler technique or suboptimal drug delivery [36], and comorbidities [37] may reduce the apparent effectiveness of corticosteroid treatment. In addition, baseline disease severity, reduced pulmonary function, inflammatory phenotype, and aeroallergen sensitization have all been associated with differential treatment responsiveness [38,39,40]. These observations indicate that variability in corticosteroid response is not random, but reflects the complex interaction between clinical phenotype, inflammatory endotype, environmental exposures, and underlying molecular mechanisms. Accordingly, corticosteroid therapy should increasingly be guided by individualized patient characteristics, with the aim of maximizing efficacy while minimizing treatment failure and drug-related toxicity. Within this framework, pharmacogenomics has emerged as a promising approach to clarify the molecular determinants of corticosteroid response and to identify patients more likely to benefit from treatment or to develop adverse effects [41].

4.1. Genetic Determinants of Corticosteroid Response

The glucocorticoid receptor, encoded by NR3C1, is central to corticosteroid pharmacodynamics and represents one of the most biologically plausible determinants of interindividual variability in corticosteroid efficacy and safety. After ligand binding, the glucocorticoid receptor translocates to the nucleus, where it modulates the transcription of numerous genes involved in inflammation, immunity, metabolism, and cellular homeostasis [42]. Genetic variation in NR3C1 may affect receptor expression, ligand-binding affinity, nuclear translocation, isoform balance, and transcriptional activity. In particular, alterations in the relative expression of the transcriptionally active GRα isoform and the dominant-negative GRβ isoform may influence tissue sensitivity to glucocorticoids.

Rare inactivating NR3C1 mutations may cause primary generalized glucocorticoid resistance, also known as Chrousos syndrome, a condition characterized by impaired glucocorticoid receptor-mediated transcription despite elevated circulating cortisol concentrations [43]. Clinically, this may result in hypercortisolemia without typical Cushingoid features and may be accompanied by manifestations of mineralocorticoid and androgen excess, including hypertension, hypokalemia, and hyperandrogenism. For example, a heterozygous missense mutation in the DNA-binding domain of NR3C1 (c.1330T>G; p.Phe444Val) has been shown to markedly reduce transcriptional activity, resulting in clinical glucocorticoid resistance and biochemical hypercortisolemia without classical Cushingoid manifestations [44]. Other pathogenic NR3C1 variants have similarly been associated with severe resistance phenotypes, including hypertension, androgen excess, and failure to suppress cortisol after dexamethasone administration [45].

Conversely, gain-of-function or functionally altered variants may lead to increased glucocorticoid sensitivity. A well-described adult case identified combined GRβ variants, A3669G and G3134T, associated with exaggerated transcriptional responses to glucocorticoids and clinical hypersensitivity to both endogenous and exogenous steroids [46]. These observations highlight the bidirectional nature of NR3C1-driven variability, which may range from glucocorticoid resistance to glucocorticoid hypersensitivity.

Beyond rare mutations, common NR3C1 polymorphisms contribute to more subtle but clinically relevant differences in corticosteroid response and toxicity. In children with steroid-treated nephrotic syndrome, specific single nucleotide polymorphisms, including rs10482634, have been associated with steroid resistance and with differential adverse effects, such as growth impairment and Cushingoid features during glucocorticoid therapy [44]. Other variants, including the BclI polymorphism and related haplotypes, have been linked to differences in prednisone sensitivity, including faster or delayed remission of proteinuria. In pediatric asthma, NR3C1 variants have been associated with differential responses to ICS, with some genotypes correlating with improved lung function and symptom control, and others with poorer clinical outcomes despite treatment [50]. Adult asthma studies have also reported associations between NR3C1 variants, including rs4585488 and rs4607376, and disease control or bronchodilator responsiveness during corticosteroid therapy [51]. Collectively, these findings suggest that NR3C1 variation exists along a functional continuum, ranging from rare, highly penetrant mutations causing overt glucocorticoid resistance or hypersensitivity to common polymorphisms that modulate treatment response and adverse-effect risk.

Genes encoding proteins that regulate glucocorticoid receptor folding, trafficking, and signaling also influence corticosteroid responsiveness. Among these, FKBP5 is one of the most extensively studied. FKBP5 encodes FK506-binding protein 51, a co-chaperone of the glucocorticoid receptor complex that reduces receptor sensitivity and delays nuclear translocation. Genetic variation in FKBP5, particularly the rs1360780 polymorphism, has been associated with impaired glucocorticoid receptor-mediated feedback at both systemic and tissue levels. In a controlled clinical study including 68 patients with major depression and 87 healthy controls, carriers of the FKBP5 risk allele T showed significantly reduced suppression of cortisol and adrenocorticotropic hormone after dexamethasone administration, indicating impaired hypothalamic–pituitary–adrenal axis feedback [47]. Increased FKBP5 expression, whether genetically determined or induced by glucocorticoid exposure, may further delay negative feedback and reduce glucocorticoid receptor sensitivity, resulting in relative hypercortisolemia and functional glucocorticoid resistance [48].

STIP1, another component of the glucocorticoid receptor heterocomplex, has also been implicated in corticosteroid response. STIP1 encodes stress-induced phosphoprotein 1, a co-chaperone involved in receptor maturation and intracellular trafficking. In a well-characterized cohort of 382 adult patients with asthma treated with ICS, several STIP1 single nucleotide polymorphisms, including rs4980524, rs6591838, rs2236647, and rs2236648, were significantly associated with lung function outcomes [49]. These variants correlated not only with baseline FEV₁, as observed for rs4980524 and rs2236647, but also with the percentage change in FEV₁ after 4 and 8 weeks of corticosteroid treatment, particularly for rs6591838 and rs2236647. These findings indicate that STIP1 variation may influence dynamic treatment response rather than simply reflecting baseline disease severity. Haplotype analyses further supported associations between STIP1 genetic profiles and corticosteroid-induced improvement in pulmonary function [49]. Pediatric evidence, although more heterogeneous, is consistent with a role for this pathway. In a cohort of 263 children with asthma, the STIP1 rs2236647 polymorphism was associated with increased disease susceptibility, supporting a potential contribution of STIP1-mediated glucocorticoid receptor regulation in early-life airway disease [50].

Downstream effector genes involved in corticosteroid-induced transcriptional regulation provide an additional layer of pharmacogenomic variability. Variants in genes such as GLCCI1, DUSP1, and HDAC1/2 may influence the magnitude of corticosteroid-mediated repression of pro-inflammatory pathways. Among these, GLCCI1 has received particular attention because variants in this gene have been repeatedly associated with reduced short-term response to intranasal and inhaled corticosteroids in respiratory disease cohorts [41]. These findings suggest that variability in transcriptional effector pathways may contribute to incomplete suppression of inflammation despite adequate corticosteroid exposure.

Pharmacokinetic genes may further modulate corticosteroid response by influencing systemic exposure and tissue concentrations. Polymorphisms in CYP3A4 and CYP3A5, which contribute to the metabolism of several corticosteroids, may alter plasma concentrations of prednisone, prednisolone, methylprednisolone, and dexamethasone [41]. Similarly, variation in transporter and detoxification genes, including ABCB1 and GSTP1, may affect drug distribution, cellular efflux, oxidative stress responses, and susceptibility to dose-related adverse effects [41]. These mechanisms may contribute to variability in both therapeutic efficacy and toxicities such as hyperglycemia, infection, adrenal suppression, growth impairment, and Cushingoid features.

Finally, genes involved in immune regulation and disease endotypes may indirectly determine corticosteroid sensitivity by shaping baseline inflammatory pathways. Variants in TBX21, FCER2, ORMDL3, and other immune-related genes may influence airway inflammation, allergic sensitization, IgE regulation, and susceptibility to exacerbations [41]. As a result, these genes may help distinguish patients whose inflammatory profile is highly corticosteroid-responsive from those with steroid-refractory disease driven by alternative immune pathways.

Table 2 shows the main pharmacogenomic determinants of corticosteroid response.

Overall, corticosteroid pharmacogenomics reflects the combined influence of receptor-level variation, chaperone-mediated signaling, downstream transcriptional responses, pharmacokinetic determinants, and disease-specific immune pathways. Although no single genetic marker currently explains the full spectrum of corticosteroid response, accumulating evidence supports the clinical relevance of multigene models that integrate pharmacogenomic, developmental, and phenotypic data. Such approaches may be particularly valuable in pediatrics, where genetic determinants interact with growth, maturation of drug-metabolizing systems, evolving immune responses, and age-specific vulnerability to corticosteroid toxicity.

5. Biologics

Inflammatory diseases arise from dysregulated interactions between innate and adaptive immunity, in which cytokine signaling, immune-cell activation, and inflammasome-mediated pathways play central pathogenic roles. Inflammasomes are intracellular multiprotein complexes that assemble in response to pathogen-associated or danger-associated signals. Their activation leads to caspase-1–dependent maturation and secretion of interleukin (IL)-1β and IL-18, as well as pyroptotic cell death, thereby amplifying local and systemic inflammation and contributing to tissue injury [52].

Conventional anti-inflammatory agents, including NSAIDs and corticosteroids, remain widely used across inflammatory and immune-mediated diseases. However, their clinical utility may be limited by incomplete efficacy, non-specific immunosuppression, and off-target toxicity in a substantial proportion of patients. Increasing understanding of disease-specific inflammatory pathways has enabled the development of mechanism-directed therapies that selectively target key immune mediators [53]. These include monoclonal antibodies and receptor antagonists directed against tumor necrosis factor-alpha (TNF-α), IL-6, IL-1, IL-17, IL-23, and other cytokine pathways, as well as small-molecule inhibitors such as Janus kinase inhibitors. Such agents have substantially improved outcomes in several conditions, including juvenile idiopathic arthritis, inflammatory bowel disease, psoriasis, rheumatoid arthritis, autoinflammatory syndromes, and severe asthma [54,55].

Despite these advances, biologic therapies are characterized by considerable interindividual variability in efficacy, durability of response, and safety [11,56]. Approximately 20–40% of patients fail to achieve adequate disease control after treatment initiation, a phenomenon referred to as primary non-response, and up to 40% of initial responders lose response over time, commonly because of secondary non-response. Mechanisms underlying this variability include differences in disease endotype, inflammatory pathway dominance, drug pharmacokinetics, immunogenicity, anti-drug antibody formation, concomitant immunomodulatory therapy, and genetic background. In this context, pharmacogenomics may help identify patients most likely to respond to a given biologic agent, those at increased risk of treatment failure, and those who may benefit from intensified monitoring or combination therapy [57,58].

5.1. Cytokine and Immune-Regulatory Gene Variants

Polymorphisms in cytokine genes and their receptors may influence baseline inflammatory set-points and modify pharmacodynamic responses to targeted therapies. Variants in genes such as TNF, IL6, and IL1B can alter cytokine expression or signaling intensity, potentially affecting both disease activity and response to cytokine blockade. Among these, TNF promoter variants have been extensively investigated as predictors of response to anti-TNF therapy, particularly in psoriasis, psoriatic arthritis, rheumatoid arthritis, and inflammatory bowel disease.

Several TNF promoter polymorphisms, including −308G>A, rs1800629; −238G>A, rs361525; −857C>T, rs1799724; and −1031T>C, rs1799964, have been evaluated in relation to anti-TNF efficacy. Systematic reviews and meta-analyses have reported heterogeneous findings, with associations varying according to ancestry, disease type, treatment outcome definition, and study design. Some European cohorts have identified associations between specific alleles and improved clinical response, whereas several Asian cohorts have failed to replicate these findings [59]. This variability suggests that the predictive value of individual TNF variants may be context-dependent and influenced by population-specific linkage patterns, environmental exposures, disease heterogeneity, and publication bias.

Genetic variation in immune-regulatory loci beyond TNF may also influence biologic response. Variants in TNFAIP3, located at 6q23.3, and IL12B, including rs2546890, have been investigated as candidate predictors of response to biologic therapy. TNFAIP3 encodes A20, a key negative regulator of nuclear factor-kappa B signaling and inflammatory activation. Some studies have linked TNFAIP3 polymorphisms, such as rs610604 and rs6920220, with improved response to anti-TNF therapy, whereas others have not confirmed these associations [60]. Similarly, variants in IL12B, a gene involved in IL-12 and IL-23 signaling, may influence pathways targeted by biologics used in psoriasis, inflammatory bowel disease, and related immune-mediated disorders [60]. However, available data remain inconsistent, and the clinical application of these markers is not yet established.

5.2. Fc Gamma Receptor Polymorphisms and Monoclonal Antibody Response

Fc gamma receptor (FCGR) polymorphisms represent another biologically plausible source of variability in response to monoclonal antibody therapies. These receptors regulate interactions between the Fc portion of immunoglobulin G antibodies and immune effector cells, thereby influencing antibody-dependent cellular cytotoxicity, immune-complex clearance, drug distribution, and possibly immunogenicity. Genetic variants in FCGR2A and FCGR3A may alter receptor affinity for IgG subclasses and modulate the effector functions of therapeutic monoclonal antibodies [61].

Associations between FCGR variants and biologic response have been reported across rheumatologic, dermatologic, and gastroenterologic indications. In particular, FCGR2A and FCGR3A polymorphisms have been linked to variable responses to anti-TNF monoclonal antibodies and other antibody-based therapies [61]. These findings provide a mechanistic rationale for altered therapeutic efficacy, differences in drug clearance, and variability in immune-mediated effects. However, as with cytokine-gene variants, the magnitude and consistency of these associations differ across studies, and further validation is required before routine clinical implementation.

5.3. HLA Variants, Immunogenicity, and Loss of Response

Human leukocyte antigen (HLA) alleles have been associated with both disease susceptibility and differential response to biologic therapies. Their role is particularly relevant in relation to biologic immunogenicity, as HLA molecules determine antigen presentation and may influence the development of anti-drug antibodies. Anti-drug antibodies can reduce circulating drug concentrations, increase drug clearance, impair therapeutic efficacy, and contribute to infusion or injection reactions.

The most robust pharmacogenomic evidence in this field concerns HLA-DQA1*05, primarily studied in inflammatory bowel disease. Carriage of HLA-DQA1*05 has been associated with accelerated anti-drug antibody formation, lower anti-TNF drug trough levels, earlier treatment discontinuation, and increased risk of secondary non-response [62]. In a large pharmacogenomic analysis, HLA-DQA1*05:01 was associated with a significantly shorter time to secondary non-response in patients treated with infliximab, whereas HLA-DQA1*05:05 showed a similar association in patients treated with adalimumab [63]. In contrast, these alleles did not predict secondary non-response to ustekinumab or vedolizumab, suggesting that their effect is more specific to anti-TNF monoclonal antibodies than to biologics with different mechanisms of action [63].

Importantly, concomitant immunomodulator therapy, such as azathioprine or methotrexate, appears to attenuate the risk of secondary non-response in carriers of relevant HLA-DQA1*05 subtypes, consistent with suppression of anti-drug antibody formation [63]. These findings support HLA-DQA1*05 genotyping as a potential tool for stratifying immunogenicity risk at the time of anti-TNF initiation. In clinical practice, identification of high-risk carriers may help guide decisions regarding combination therapy, closer therapeutic drug monitoring, or selection of biologics with lower immunogenic potential.

Other HLA variants may also contribute to biologic response, although evidence is less mature. Single nucleotide polymorphisms in the HLA-DRB9 region, including rs2395185, have been associated with primary non-response and long-term treatment failure in pediatric inflammatory bowel disease cohorts [64]. In rheumatoid arthritis, the HLA-E*01:01 allele has been associated with better European Alliance of Associations for Rheumatology responses to anti-TNF therapy compared with HLA-E*01:03 [65]. These observations suggest that HLA-related pharmacogenomic effects may extend beyond immunogenicity and may influence broader immune mechanisms involved in biologic response.

5.4. Clinical Implications

Table 3 illustrates the main pharmacogenomic determinants of response to biologic therapies.

Pharmacogenomics offers a promising framework for improving the precision of biologic therapy. Unlike small-molecule drugs, biologics are less commonly affected by classical drug-metabolizing enzyme polymorphisms. Instead, genetic determinants of biologic response are more closely related to immune regulation, cytokine signaling, antigen presentation, Fc receptor function, and immunogenicity. Among currently studied markers, HLA-DQA1*05 has the strongest evidence for potential clinical translation, particularly for predicting anti-drug antibody formation and secondary loss of response to anti-TNF monoclonal antibodies in inflammatory bowel disease.

Nevertheless, most candidate biomarkers remain insufficiently validated for routine use. Many reported associations are limited by small sample sizes, heterogeneous disease populations, variable response definitions, differences in ancestry, and inconsistent adjustment for clinical factors such as drug levels, concomitant immunomodulators, disease severity, and adherence. Therefore, future implementation will likely require integrated predictive models rather than single-marker approaches. These models should combine pharmacogenomic data with therapeutic drug monitoring, inflammatory biomarkers, disease phenotype, age, prior biologic exposure, and concomitant therapy.

In pediatric populations, pharmacogenomic approaches may be particularly valuable because early optimization of biologic therapy can prevent cumulative inflammatory damage, reduce corticosteroid exposure, and improve long-term outcomes. However, pediatric-specific validation is essential, as genetic effects may interact with developmental immune maturation, growth, body composition, and age-related differences in drug clearance and immunogenicity.

Overall, biologic pharmacogenomics is an evolving but clinically important field. Current evidence supports a role for variants in cytokine genes, immune-regulatory loci, FCGR genes, and especially HLA alleles in modulating response, immunogenicity, and durability of biologic therapy. The integration of these markers into precision medicine strategies may enable more rational biologic selection, improved prediction of treatment failure, reduced immunogenicity, and more individualized use of concomitant immunomodulation.

6. Conclusions

The implementation of pharmacogenetic testing in routine clinical care has reached actionable maturity for selected gene–drug pairs, particularly when robust evidence links genotype to drug exposure, therapeutic response, or clinically relevant toxicity. Among anti-inflammatory therapies, the strongest current example is represented by CYP2C9-guided NSAID prescribing. The Clinical Pharmacogenetics Implementation Consortium provides evidence-based genotype-to-dosing recommendations for CYP2C9 and several NSAIDs, including celecoxib, ibuprofen, flurbiprofen, lornoxicam, and meloxicam. These recommendations support dose reduction, careful titration, enhanced monitoring, or selection of alternative agents in patients carrying reduced-function alleles, with the aim of limiting excessive drug exposure and reducing the risk of gastrointestinal, renal, and cardiovascular adverse events. Accordingly, pre-emptive or reactive CYP2C9 testing may be integrated into analgesic stewardship pathways, particularly for patients requiring prolonged or high-dose NSAID therapy, those with comorbidities that increase NSAID-related risk, or those with a previous history of adverse reactions.

For biologic therapies, pharmacogenomic implementation is less advanced but rapidly evolving. Large pharmacogenomic studies have identified HLA-DQA1*05 carriage as a reproducible predictor of anti-drug antibody formation, reduced drug trough concentrations, and accelerated secondary non-response to anti-TNF monoclonal antibodies, particularly infliximab and adalimumab. Importantly, concomitant immunomodulator therapy appears to attenuate this risk in allele carriers, supporting the potential use of HLA-DQA1*05 genotyping to guide decisions on combination therapy at treatment initiation. This approach may help identify patients who require closer therapeutic drug monitoring, early optimization of dosing, or consideration of biologics with lower immunogenic potential.

By contrast, pharmacogenomic markers of corticosteroid response, including variants in NR3C1, FKBP5, STIP1, GLCCI1, and pharmacokinetic genes such as CYP3A4, CYP3A5, and ABCB1, remain promising but are not yet sufficiently standardized for routine implementation. Current evidence supports their biological relevance and potential contribution to variability in efficacy and toxicity, but findings are often heterogeneous across diseases, populations, and outcome definitions. Further validation in large, prospective, ancestry-diverse, and pediatric-inclusive cohorts is required before these markers can be translated into formal prescribing recommendations.

Overall, pharmacogenomics offers a clinically meaningful opportunity to improve the precision of anti-inflammatory therapy. Its greatest near-term value lies in identifying patients at increased risk of toxicity or treatment failure and in supporting individualized decisions on drug selection, dose adjustment, monitoring intensity, and concomitant therapy. However, the successful integration of pharmacogenomics into clinical practice will require more than genetic testing alone. Implementation should be supported by electronic decision tools, clinician education, therapeutic drug monitoring when appropriate, cost-effectiveness analyses, and clear pathways for interpreting results in relation to age, comorbidities, organ function, disease phenotype, and concomitant medications.

In pediatric populations, this integrated approach is particularly important because genetic determinants interact with developmental changes in drug metabolism, immune maturation, growth, and vulnerability to adverse effects. Future research should therefore prioritize pediatric-specific evidence, multigene predictive models, and prospective trials evaluating whether pharmacogenomic-guided anti-inflammatory therapy improves clinical outcomes. As evidence continues to mature, pharmacogenomics is likely to become an increasingly important component of personalized anti-inflammatory treatment in both adults and children.