Association of IFN‐γ +874 T/A Polymorphism With COVID‐19 Susceptibility, Severity, and Inflammatory Response: A Case–Control Study From Southern Iran

ABSTRACT

Background

Significant interpatient variability in Coronavirus Disease 2019 (COVID‐19) manifestations highlights the influence of host genetic factors on disease outcomes. This study examines its association with COVID‐19 susceptibility, severity, and inflammatory markers in an Iranian population.

Methods

A retrospective case–control analysis was performed on 210 COVID‐19 patients and 210 age‐ and sex‐matched control subjects to evaluate whether the Interferon‐gamma (IFN‐γ) gene polymorphism (+874 T/A) is associated with disease severity.

Results

Patients had more comorbidities (e.g., diabetes 26.2% vs. 14.3%; hypertension 29.5% vs. 16.7%) and pronounced paraclinical abnormalities, including elevated Neutrophil‐to‐Lymphocyte Ratio (NLR) (8.1 vs. 1.9), C‐reactive protein (CRP) (45.5 vs. 3.2 mg/L), and lymphopenia (1.1 vs. 2.1 × 10³/μL; all p < 0.001). The TA genotype and A allele increased disease susceptibility (Odds Ratio (OR) =2.05 (95% CI: 1.32–3.19), p = 0.002; OR = 1.58 (95% CI: 1.11–2.25), p = 0.01, respectively). The TA genotype independently predicted severe disease (aOR = 2.85, p = 0.001), and both the TA and AA genotypes were linked to higher IL‐6 levels (p < 0.001). TT carriers had shorter hospitalizations (p < 0.01). Combining TA/AA with NLR improved severity prediction (AUC = 0.82 (95% CI: 0.76–0.88)).

Conclusion

The IFN‐γ +874 TA genotype was independently associated with severe COVID‐19. The TA genotype appears to confer increased risk of severe disease, whereas the AA genotype may be linked to a milder clinical presentation despite elevated inflammatory markers. Integration of genetic and hematological parameters may enhance risk stratification; however, further validation in larger and diverse populations is warranted.

Graphical abstract illustrating the association between IFN‐γ +874 T/A polymorphism and COVID‐19 outcomes. TA and AA genotypes are linked to increased susceptibility, higher inflammatory markers (notably IL‐6), greater disease severity, and longer hospitalization. Integration of IFN‐γ genotypes with NLR improves prognostic accuracy for risk stratification and clinical decision‐making.

Abbreviations

ACE2

angiotensin‐converting enzyme 2

ARDS

Acute respiratory distress syndrome

ARMS‐PCR

amplification‐refractory mutation system polymerase chain reaction

AUC

area under the curve

CI

confidence interval

COVID‐19

coronavirus disease 2019

CRP

C‐reactive protein

ELISA

enzyme‐linked immunosorbent assay

ESR

erythrocyte sedimentation rate

HBV

hepatitis B virus

HCV

hepatitis C virus

HLA

Human leukocyte antigen

HWE

Hardy–Weinberg equilibrium

IFN‐γ

interferon‐gamma

IL‐6

interleukin‐6

LDH

lactate dehydrogenase

NLR

neutrophil‐to‐lymphocyte ratio

OR

odds ratio

PBMC

peripheral blood mononuclear cell

PCR

polymerase chain reaction

ROC

receiver operating characteristic

SARS‐CoV‐2

severe acute respiratory syndrome coronavirus 2

SNP

single‐nucleotide polymorphism

TMPRSS2

transmembrane protease, serine 2

WHO

World Health Organization

1. Introduction

The clinical manifestations of COVID‐19 vary widely, ranging from cases with no symptoms to severe respiratory distress and dysfunction of multiple organs [1]. The significant variability in disease severity, beyond established risks like age, sex, and comorbidities, points to a role for host genetics, including genes involved in viral entry and immune regulation [2, 3]. A critical feature of severe COVID‐19 is a dysregulated immune response, characterized by a “cytokine storm” where pro‐inflammatory signaling goes awry. IFN‐γ is a key cytokine in this process; it is essential for antiviral defense but can also cause tissue damage when overproduced [4]. The +874 T/A polymorphism (rs2430561) regulates IFN‐γ levels, with the T allele promoting higher production and the A allele associated with lower expression [5]. The rs2430561 single nucleotide polymorphism (SNP) is located within the first intron of the IFNG gene and influences transcriptional activity through differential binding of nuclear transcription factors. Specifically, functional in vitro studies by Pravica et al. demonstrated that the T allele provides a unique binding site for the nuclear factor kappa B (NF‐κB), driving higher IFN‐γ production, whereas the A allele lacks this binding affinity and correlates with reduced cytokine secretion [6]. Variations in IFN‐γ expression have been subsequently implicated in susceptibility to a range of inflammatory, infectious, and pulmonary diseases, including tuberculosis, chronic hepatitis C, asthma, and acute respiratory infections [7, 8].

These findings highlight the functional significance of this polymorphism in modulating host immune responses. Although the A allele has been linked to other diseases, such as chronic Hepatitis B Virus (HBV) susceptibility in different populations, its role in COVID‐19 within this group remains unclear [9].

In the context of viral infections, host SNPs have emerged as important determinants of disease susceptibility, inflammatory dysregulation, and clinical severity [10]. Large‐scale genetic studies in COVID‐19 have successfully identified variants in interferon signaling pathways, innate immune receptors, and inflammatory mediators as major contributors to heterogeneous clinical outcomes [11, 12].

Given the central role of IFN‐γ in antiviral immunity and immune regulation, the functional +874 T/A polymorphism represents a biologically plausible candidate variant for influencing COVID‐19 susceptibility and disease progression. Therefore, the present study investigates the potential link between the IFN‐γ +874 T/A genetic variation and both the likelihood of contracting COVID‐19 and the severity of its clinical outcomes. It also examines associated inflammatory biomarkers, including serum IL‐6 concentrations and the NLR, in individuals from southern Kerman Province, Iran. Insights from this research could enhance understanding of genetic contributions to disease progression and support the development of tailored therapeutic and preventive measures. Considering the established immunomodulatory role of IFN‐γ and the functional relevance of rs2430561, the present study investigates the potential link between the IFN‐γ +874 T/A genetic variation.

2. Materials and Methods

2.1. Study Design and Participants

This retrospective case–control investigation leveraged a biorepository of 420 peripheral blood mononuclear cell (PBMC) specimens. PBMC samples were collected at hospital admission under aseptic conditions. Approximately 3 mL of blood was drawn into EDTA‐containing tubes for hematological analysis, and 5 mL was collected into plain tubes for serum separation. Serum samples were allowed to clot at room temperature and were subsequently centrifuged at 3000 rpm for 10 min. The separated serum was either analyzed immediately or stored at −20°C until biochemical evaluation. Hematological parameters were measured using an automated hematology analyzer (K21, SYSMEX, Japan). The NLR was calculated by dividing the absolute neutrophil count (×10⁹/L) by the absolute lymphocyte count (×10⁹/L) derived from routine baseline automated complete blood count (CBC) profiles at admission. Biochemical and colorimetric assays, including CRP and other routine parameters, were performed using commercial diagnostic kits (Pars Azmom, PARS, Iran) according to the manufacturer's instructions on an automated chemistry analyzer (BS‐480, Mindray, China). Serum IL‐6 levels were quantified using a commercially available Diaclone ELISA kit (Besançon, France; Cat. No. 950.030) following the protocol provided by the manufacturer. The participants were equally divided between 210 individuals with laboratory‐confirmed COVID‐19 and 210 healthy controls, meticulously matched for age and sex, all drawn from a population in southern Kerman Province, Iran. The case population was constituted from consecutive admissions to the tertiary referral teaching hospital in Kerman, wherein severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) infection was definitively diagnosed via reverse‐transcription polymerase chain reaction (RT‐PCR) assay of nasopharyngeal swab specimens. Blood samples were obtained at the time of hospital admission and prior to initiation of COVID‐19‐specific treatments to avoid potential confounding effects of immunomodulatory therapies on inflammatory marker levels. An a priori power calculation was conducted utilizing G‐Power software (v. 3.1) to determine sample size adequacy. This calculation was predicated on effect sizes derived from preceding genetic association studies in the context of viral infections. The parameters were calibrated to ascertain an odds ratio of 2.0 with a statistical power of 80% and an alpha level of 0.05 denoting significance. Patients were recruited upon hospital admission, a period corresponding to a significant pandemic wave dominated by the Omicron variant of concern, to ensure our samples reflected a defined and critical phase of the outbreak, similar to the approach used in other contemporary studies [2, 3]. Data collection from electronic health records for all participants was completed by July 2024. The primary follow‐up period for clinical outcomes was the duration of hospitalization. Case eligibility was restricted to adult inpatients (≥ 18 years) with RT‐PCR‐confirmed SARS‐CoV‐2 infection. To minimize the potential confounding effect of population stratification, a strict ancestral screening protocol was applied: only individuals who self‐reported as belonging to the local population of Southern Kerman, and whose parents and grandparents were native to the same geographical region for at least three generations, were eligible for enrollment. Exclusion criteria, applied uniformly to both cases and controls, comprised pre‐existing immunodeficiency disorders (e.g., HIV/AIDS or primary immunodeficiency), a diagnosis of active malignancy, the presence of a concurrent active bacterial or fungal infection at admission, or insufficient clinical or genetic data. Control subjects were healthy individuals recruited from the same geographical area as the case samples to represent the source population. All participants were recruited from the same regional population in southern Kerman Province to minimize potential population stratification; however, detailed ethnicity data were not systematically recorded. Eligibility was contingent upon the absence of a clinical history of COVID‐19 and a negative SARS‐CoV‐2 RT‐PCR test at the time of recruitment, thereby ensuring they constituted a representative sample of the underlying source population. To enhance recruitment efficiency and ensure regional matching, a hospital‐based control group was selected. They were selected from the same hospital's visitor population and from individuals attending routine health check‐ups at the affiliated clinic during the same period (April–June 2024). The use of hospital visitors was intended to enhance recruitment efficiency while maintaining regional matching. To minimize confounding, a matched case–control design was implemented. Controls were individually matched to cases in a 1:1 ratio, with exact matching on sex and age within a predefined tolerance of ±5 years.

Study subjects were categorized into four strata defined by the World Health Organization (WHO) severity classification. The strata consisted of: mild (n = 40), presenting with symptoms in the absence of pneumonia; moderate (n = 80), exhibiting pneumonia without concomitant hypoxemia; severe (n = 75), demonstrating hypoxemia necessitating oxygen support; and critical (n = 15), involving the most severe manifestations, including acute respiratory distress syndrome (ARDS) or multi‐organ failure. For multivariate logistic regression, the primary outcome of “severe/critical disease” was defined as a composite of the severe and critical categories. The selection of healthy controls from the matching community and hospital visitor pool required a negative SARS‐CoV‐2 PCR test, no verifiable history of COVID‐19, and no recent symptoms of infection. The study was initiated only after receiving ethical clearance from the Kerman University of Medical Sciences Institutional Review Board. All individuals provided written informed consent before participating (Ethics Code: IR.JMU.REC.1402.120).

2.2. Data and Sample Collection

Demographic details, medical history, comorbidities (defined as a documented diagnosis in the medical record or active treatment for the condition), and clinical parameters were recorded from electronic health records and hospital charts. Clinical symptoms at admission (e.g., fever, cough, and dyspnea) were documented. A detailed summary of clinical symptom frequency stratified by disease severity is provided in Table S2. Vaccination history was not consistently available for all participants during the study period and therefore was not incorporated into multivariable modeling.

A comprehensive panel of laboratory tests was conducted on blood samples from all study participants. This included a complete blood count, a suite of inflammatory biomarkers (C‐reactive protein, erythrocyte sedimentation rate, ferritin, D‐dimer, and lactate dehydrogenase), and standard clinical chemistry panels to assess liver and kidney function. Quantification of serum interleukin‐6 (IL‐6) was performed using a commercial solid‐phase sandwich Enzyme‐Linked Immunosorbent Assay (ELISA) kit from Diaclone (Besançon, France; Cat. No. 950.030). The assay methodology employs a capture antibody highly specific for human IL‐6 pre‐coated onto the microtiter plate, with simultaneous incubation of samples and a biotinylated anti‐IL‐6 secondary antibody, followed by streptavidin‐HRP conjugate and TMB substrate for colorimetric detection. According to the manufacturer's validated specifications, the minimum detectable dose of IL‐6 is 2 pg/mL, determined by adding three standard deviations to the mean optical density obtained from 40 replicate determinations of the zero standard. In a panel of 20 healthy human sera tested by the manufacturer, all IL‐6 values fell below this analytical detection limit, establishing the expected normal physiological range as below 2 pg/mL. The analytical acceptability of this assay is supported by an intra‐assay coefficient of variation (CV) of 3.6% and an inter‐assay CV of 7.7%, both of which are within accepted standards for cytokine quantification, with a mean spike recovery of 103%. The assay recognizes both natural and recombinant human IL‐6 with no observed cross‐reactivity against other tested cytokines, including IL‐1α, IL‐1β, IL‐10, IL‐12, IFN‐γ, IL‐4, TNF‐α, IL‐8, and IL‐13. All serum samples in the present study were stored at −20°C and analyzed within 2 weeks of collection with no freeze–thaw cycles, a protocol consistent with the manufacturer's stability data confirming no significant loss of IL‐6 reactivity after up to five freeze–thaw cycles or storage at recommended temperatures. Concurrently, chest computed tomography (CT) scans were interpreted by a board‐certified radiologist to identify pulmonary manifestations characteristic of COVID‐19, such as ground‐glass opacities and consolidation. The total duration of each patient's hospitalization was recorded as a clinical endpoint for subsequent survival analysis.

For the genetic component of the study, a separate 6 mL whole blood sample was collected from each participant into heparinized vacuum tubes. Peripheral Blood Mononuclear Cells (PBMCs) were isolated immediately via density gradient centrifugation. In this procedure, the blood was diluted with phosphate‐buffered saline (PBS) and layered over a Ficoll‐Paque solution (Bahar‐Afshan, Iran). Following centrifugation, the buffy coat containing the PBMCs was carefully harvested. The isolated cells were then washed twice with PBS to remove platelets and residual plasma, and the final cell pellet was resuspended in 200 μL of PBS. A Neubauer hemocytometer was used to determine total cell count, and aliquots containing approximately 2.1 million cells were cryopreserved at −80°C for long‐term storage pending downstream molecular analysis.

2.3. Genotyping

Genomic Deoxyribonucleic Acid (DNA) was isolated from PBMCs with the Favorgen commercial kit (Taiwan), adhering to the protocols provided by the manufacturer. We assessed the concentration and purity of the DNA by measuring its absorbance at 260 nm and 280 nm using a Nanodrop ND 1000 spectrophotometer (Thermo Scientific, USA). The DNA extracts were then stored at −80°C until further use. To verify DNA integrity, we amplified the beta‐globin gene by PCR.

For genotyping the IFN‐γ +874 T/A polymorphism (rs2430561), we employed an Amplification‐Refractory Mutation System PCR (ARMS‐PCR) approach with primers specific to each allele (detailed in Table 1). The PCR was carried out in a Thermal Cycler (Eppendorf Mastercycler) using a pre‐mixed PCR Master Mix from Amplicon (Odense, Denmark). The PCR conditions were as follows: initial denaturation at 95°C for 5 min, followed by 35 cycles of 95°C for 30 s, annealing at 60°C for 40 s, extension at 72°C for 40 s, and a final extension at 72°C for 5 min. PCR products were visualized on a 2% agarose gel stained with Safe Stain. A representative agarose gel electrophoresis image is provided in Figure S1, showing the 262 bp (IFN‐γ) and 110 bp (beta‐globin) bands for TT, TA, and AA genotypes. Amplification of the beta‐globin gene served as an internal control for each reaction. We separated the resulting PCR products by electrophoresis through a 2% agarose gel and documented them with a UVITEC imaging system (Cambridge). Amplification of the beta‐globin gene served as an internal control for each reaction.

TABLE 1.

Primers for IFN‐γ +874 A/T genotyping.

Target	Primer type	5′‐Sequence‐3′	Product size
Beta‐globin	Forward	ACACAACTGTGTTCACTAGC	110 bp
	Reverse	CAACTTCATCCACGTTCACC
IFN‐γ +874 T/A	Reverse (Common)	TCAACAAAGCTGATACTCCA	262 bp
	Forward (A allele)	TTCTTACACACAAAATCAAATCA
	Forward (T allele)	TTCTTACACACAAAATCAAATCT

To ensure genotyping accuracy, a random subset of samples (10%) was re‐genotyped, yielding a 100% reproducibility rate. Furthermore, all laboratory staff were blinded to the clinical data of the participants throughout the analysis to prevent any potential bias.

2.4. Statistical Analysis

Data analysis was performed using SPSS (version 27, IBM Corp., USA) for statistical procedures and GraphPad Prism (version 10, GraphPad Software, USA) for generating graphical representations. Statistical significance was defined as a two‐sided p‐value of less than 0.05.

The distribution of continuous variables was assessed, and comparisons were made using the student's t‐test for normally distributed data or the Mann–Whitney U test for non‐parametric data. Associations involving categorical variables were examined with the chi‐square test. To quantify genetic associations with disease susceptibility, we calculated odds ratios (ORs) alongside their 95% confidence intervals (CIs). Although a matching strategy was used for participant recruitment, we opted for an unmatched analysis to prevent potential overmatching. While conditional logistic regression is often recommended in individually matched designs, we applied unconditional logistic regression with covariate adjustment for age and sex to preserve statistical efficiency, given the ±5‐year matching tolerance. This approach allowed us to explicitly control for the original matching variables age and sex in our subsequent multivariate models.

We constructed a multivariate logistic regression model to pinpoint factors independently associated with the development of severe or critical illness. This model was adjusted for several pre‐specified confounders known from the existing literature to influence COVID‐19 severity: age greater than 60 years, male sex, a history of diabetes or hypertension, and a neutrophil‐to‐lymphocyte ratio (NLR) exceeding 5. The > 60 years age threshold was selected as it represents a well‐established risk factor, and an NLR > 5 is a recognized indicator of systemic inflammation linked to poorer outcomes in COVID‐19 patients. This NLR threshold was selected based on previously published COVID‐19 prognostic study [13] rather than ROC‐derived optimization within the present dataset. Sensitivity analyses using alternative NLR cutoffs (e.g., 3, 7) yielded similar trends.

For the outcome of time‐to‐discharge, we employed Kaplan–Meier survival analysis, and the significance of differences between groups was tested with the log‐rank test. The discriminatory power of predictive models was evaluated and compared by analyzing Receiver Operating Characteristic (ROC) curves. The statistical significance of the difference between area under the curve (AUC) values for the combined model versus individual parameters was assessed using the DeLong test.

Our analysis followed a complete‐case principle for each specific test. Formal statistical assessment of missingness patterns (e.g., testing for missing completely at random) was not performed. Individuals missing data for key variables relevant to a particular analysis (such as IL‐6 levels) were excluded from that specific assessment. The genotype distribution in the control population was verified for conformity to Hardy–Weinberg Equilibrium (HWE).

“Additionally, a post hoc power analysis was performed to verify the statistical strength of our primary and secondary outcomes.” Based on the enrolled sample size (n = 210 cases and n = 210 controls) and the observed control genotype frequencies, the study achieved a post hoc statistical power of 87.2% to detect the observed odds ratio (OR = 2.05) for susceptibility (TA vs. TT genotype) at a two‐sided alpha level of 0.05. However, for secondary stratified subgroup analyses across clinical severity grades, the estimated post hoc power was lower (54.5%). Serum IL‐6 quantification was available for 180 out of 210 patients (85.7%). Chi‐square analysis confirmed that the 30 missing samples due to random laboratory constraints were evenly distributed across genotypes (p > 0.05) and severity categories (p > 0.05), demonstrating a “Missing Completely at Random” pattern with no systematic bias.

3. Results

3.1. Demographic and Clinical Characteristics

This study screened 245 confirmed COVID‐19 patients and 250 control candidates for eligibility. After applying exclusion criteria—which primarily concerned incomplete data or participant withdrawal, 35 patients and 40 controls were excluded. The final matched‐pair analysis therefore included 210 subjects per group, revealing statistically significant disparities between them. Patients and controls were matched for age (52.1 ± 8.7 vs. 49.3 ± 9.1 years, p = 0.15) and sex (60% vs. 58.6% male, p = 0.78). Critically, patients had a higher comorbidity burden, including diabetes (26.2% vs. 14.3%, p = 0.002) and hypertension (29.5% vs. 16.7%, p = 0.001). The most common clinical symptoms at admission among COVID‐19 patients were fever (78.1%), cough (71.4%), and dyspnea (55.2%), while gastrointestinal symptoms such as diarrhea (12.4%) and vomiting (8.6%) were less frequent. The frequency and distribution of all documented clinical symptoms across severity groups are detailed in Table S2. ARDS was present in all 15 patients within the critical category, consistent with the WHO definition applied. Paraclinical assessment showed markedly altered biomarkers in the patient group, including elevated NLR (8.1 ± 4.5 vs. 1.9 ± 0.7, p < 0.001), CRP (45.5 ± 38.2 vs. 3.2 ± 2.1 mg/L, p < 0.001), ferritin (680 ± 450 vs. 85 ± 45 ng/mL, p < 0.001), and D‐dimer (1.8 ± 1.5 vs. 0.3 ± 0.1 mg/L, p < 0.001). Hematological profiles were also significantly deranged, featuring lymphopenia (1.1 ± 0.5 vs. 2.1 ± 0.6 × 10³/μL, p < 0.001), neutrophilia (6.9 ± 2.8 vs. 3.8 ± 1.1 × 10³/μL, p < 0.001), and thrombocytopenia (185 ± 95 vs. 250 ± 50 × 10³/μL, p < 0.001) (Table 2). All data presented in Table 2 are from participants with complete data for each variable; there were no missing data for the demographic, comorbidity, or basic laboratory parameters reported in this table. No significant associations were observed between the IFN‐γ +874 T/A genotype and the presence of specific comorbidities (diabetes or hypertension; p > 0.05), suggesting that the observed genetic effect on disease severity operates independently of these pre‐existing conditions.

TABLE 2.

Comprehensive demographic and clinical profile of COVID‐19 patients and healthy controls.

Parameter	COVID‐19 patients (n = 210)	Healthy controls (n = 210)	p	Reference range
Demographics
Age (Years, Mean ± SD)	52.1 ± 8.7	49.3 ± 9.1	0.15	—
Male/Female, n (%)	126 (60%)/84 (40%)	123 (58.6%)/87 (41.4%)	0.78	—
Disease severity (WHO classification), n (%)
Mild	40 (19.0%)	—	—	—
Moderate	80 (38.1%)	—	—	—
Severe	75 (35.7%)	—	—	—
Critical	15 (7.1%)	—	—	—
Comorbidities, n (%)
Diabetes mellitus	55 (26.2%)	30 (14.3%)	0.002	—
Hypertension	62 (29.5%)	35 (16.7%)	0.001	—
Complete blood count (CBC)
WBC (×103/μL)	8.9 ± 3.1	6.5 ± 1.5	< 0.001	4.0–11.0
Lymphocyte count (×103/μL)	1.1 ± 0.5	2.1 ± 0.6	< 0.001	1.2–3.5
Neutrophil count (×103/μL)	6.9 ± 2.8	3.8 ± 1.1	< 0.001	2.0–7.0
NLR (Neutrophil‐to‐lymphocyte ratio)	8.1 ± 4.5	1.9 ± 0.7	< 0.001	< 3.0
Platelet count (×103/μL)	185 ± 95	250 ± 50	< 0.001	150–400
Hemoglobin (g/dL)	13.2 ± 1.9	14.1 ± 1.5	0.001	M: 13.5–17.5, F: 12.0–15.5
Inflammatory and Biochemical markers
CRP (mg/L)	45.5 ± 38.2	3.2 ± 2.1	< 0.001	< 5.0
ESR (mm/h)	45 ± 25	12 ± 8	< 0.001	M: < 15, F: < 20
Ferritin (ng/mL)	680 ± 450	85 ± 45	< 0.001	30–300
D‐Dimer (mg/L FEU)	1.8 ± 1.5	0.3 ± 0.1	< 0.001	< 0.5
LDH (U/L)	350 ± 120	190 ± 40	< 0.001	140–280
Serum creatinine (mg/dL)	1.2 ± 0.5	0.9 ± 0.2	< 0.001	0.7–1.3
Blood urea nitrogen (BUN, mg/dL)	28 ± 15	15 ± 5	< 0.001	8–20
AST (SGOT, U/L)	40 ± 25	25 ± 8	< 0.001	< 40
ALT (SGPT, U/L)	38 ± 22	23 ± 9	< 0.001	< 40
Serum albumin (g/dL)	3.5 ± 0.6	4.4 ± 0.3	< 0.001	3.5–5.2
Blood gas analysis (Critical patients, n = 15)
PaO2/FiO2 ratio	185 ± 75	—	—	> 300
pH	7.32 ± 0.08	—	—	7.35–7.45

Note: NLR was calculated using absolute counts (×10³/μL) of neutrophils divided by absolute counts (×10³/μL) of lymphocytes, obtained from complete blood count analysis performed on an automated hematology analyzer (K21, SYSMEX, Japan). All data presented are from participants with complete data for each variable; there were no missing data for the demographic, comorbidity, or basic laboratory parameters reported in this table. p‐value < 0.05 was considered statistically significant.

3.2. Association With Susceptibility

The genotype distribution in the control group conformed to Hardy–Weinberg equilibrium (p > 0.05). In addition, the genotype distribution of the IFN‐γ + 874 T/A polymorphism in the healthy control group was in perfect agreement with the Hardy–Weinberg Equilibrium (p = 0.74, X ² = 0.11). The frequency of the TA genotype was higher among patients compared with controls (52.4% vs. 40.5%; OR = 2.05 (1.32–3.19); p = 0.002), and a similar pattern was observed for the A allele (45.2% vs. 34.5%; OR = 1.58 (1.11–2.25); p = 0.01) (Table 3). These findings indicate that individuals carrying the TA genotype or the A allele have a greater likelihood of developing COVID‐19. Conversely, although the AA genotype exhibited a tendency toward elevated risk, this association did not reach statistical significance (p = 0.15).

TABLE 3.

Association of IFN‐γ + 874 T/A genotypes and alleles with COVID‐19 susceptibility.

Genotype/Allele	Patients (n = 210)	Controls (n = 210)	p	OR (95% CI)
TT	60 (28.6%)	95 (45.2%)	1 (Reference)	—
TA	110 (52.4%)	85 (40.5%)	0.002	2.05 (1.32–3.19)
AA	40 (19%)	30 (14.3%)	0.15	1.56 (0.85–2.87)
Allele T	230 (54.8%)	275 (65.5%)	1 (Reference)	—
Allele A	190 (45.2%)	145 (34.5%)	0.01	1.58 (1.11–2.25)

Note: The genotype distribution in the control group conformed to Hardy–Weinberg equilibrium (χ ² = 2.30, df = 1, p = 0.129). p‐values were calculated using the chi‐square test with the TT genotype or T allele as the reference category. Data are presented as $n\left(\%\right)$ for categorical variables and mean $\pm$ SD or median (IQR) for continuous variables. Categorical (qualitative) variables were compared using the Chi‐square (${\chi }^2$) test or Fisher's exact test. Continuous (quantitative) variables were compared using the independent samples $t$ ‐ test [or Mann–Whitney U test for non‐normally distributed data]. A two‐tailed $p$‐value $<0.05$ was considered statistically significant.

Abbreviations: CI, confidence interval; OR, odds ratio.

3.3. Association With Disease Severity and Clinical Parameters

The TA genotype correlated with critical (OR = 6.10, p = 0.03), severe (OR = 2.24, p = 0.04), and moderate (OR = 2.01, p = 0.02) disease; the AA genotype (vs. TT) was associated with mild disease (OR = 2.85 (1.10–7.36), p = 0.03) (Table 4). Notably, the AA genotype was significantly associated with the mild disease phenotype. Analysis of inflammatory marker levels by genotype subgroup revealed that the AA genotype was also linked to lower platelet counts (p = 0.04), while the TA genotype showed a trend for elevated creatinine (p = 0.08). A comprehensive analysis of all inflammatory and biochemical markers stratified by genotype is provided in Table S3. No significant association was observed between IFN‐γ genotype and invasive mechanical ventilation (IMV) requirement (p > 0.05).

TABLE 4.

Association of IFN‐γ + 874 T/A genotypes with COVID‐19 severity.

Genotype	Controls (n = 210)	Critical (n = 15)	P/OR (95% CI)	Severe (n = 75)	P/OR (95% CI)	Moderate (n = 80)	P/OR (95% CI)	Mild (n = 40)	P/OR (95% CI)
TT	95 (45.2%)	3 (20%)	1 (Reference)	20 (26.7%)	1 (Reference)	25 (31.3%)	1 (Reference)	12 (30%)	1 (Reference)
TA	85 (40.5%)	9 (60%)	0.03/6.10 (1.15–32.3)	40 (53.3%)	0.04/2.24 (1.02–4.90)	45 (56.3%)	0.02/2.01 (1.10–3.68)	16 (40%)	0.35/1.49 (0.64–3.47)
AA	30 (14.3%)	3 (20%)	0.25/2.11 (0.59–7.50)	15 (20%)	0.18/1.75 (0.77–3.97)	10 (12.5%)	0.89/0.95 (0.41–2.18)	12 (30%)	0.03/2.85 (1.10–7.36)

Note: aOR, adjusted odds ratio for age, sex, hypertension, and diabetes. All p‐values and ORs are derived from univariate analysis with the TT genotype as the reference category. Statistically significant associations (p < 0.05) are indicated in bold. Adjusted model: Values (aORs) were calculated using multivariable logistic regression analysis after adjusting for potential confounding variables, including age, biological sex, and presence of baseline comorbidities (e.g., diabetes, hypertension).

Abbreviations: aOR, adjusted odds ratio; CI, confidence interval; OR, odds ratio.

3.4. Multivariate Analysis for Severe Disease

Multivariate logistic regression analysis demonstrated that individuals carrying the TA genotype had a significantly higher likelihood of developing severe or critical forms of COVID‐19 (adjusted OR = 2.85; 95% CI: 1.52–5.35; p = 0.001). This association remained robust after controlling for potential confounders, including age above 60 years, male sex, diabetes mellitus, hypertension, and elevated neutrophil‐to‐lymphocyte ratio (Table 5).

TABLE 5.

Multivariate logistic regression analysis for independent predictors of severe COVID‐19.

Predictor variable	Adjusted odds ratio (aOR)	95% confidence interval	p
IFN‐γ genotype (TA vs. TT)	2.85	1.52–5.35	0.001
IFN‐γ genotype (AA vs. TT)	1.90	0.88–4.10	0.10
Age (> 60 years vs. ≤ 60)	3.42	1.95–6.00	< 0.001
Gender (Male vs. Female)	1.78	1.05–3.02	0.03
Diabetes mellitus (Yes vs. No)	2.25	1.30–3.90	0.004
Hypertension (Yes vs. No)	1.95	1.15–3.31	0.01
NLR (> 5 vs. ≤ 5)	4.10	2.40–7.02	< 0.001

Note: The multivariate model included all variables listed in the table simultaneously. The primary outcome was defined as a composite of severe and critical disease according to WHO classification. The NLR threshold of > 5 was selected based on previously published COVID‐19 prognostic studies [13]. Sensitivity analyses using alternative NLR cutoffs (3 and 7) yielded similar trends. Statistically significant associations (p < 0.05) are indicated in bold.

Abbreviations: aOR, adjusted odds ratio; CI, confidence interval; NLR, neutrophil‐to‐lymphocyte ratio.

3.5. Comparison of Inflammatory Biomarkers Across IFN‐γ (+874 T/A) Genotypes

To explore the functional association between genetic variation and inflammatory response, serum interleukin‐6 (IL‐6) concentrations and NLR were evaluated. IL‐6 measurements were obtained for 180 of the 210 patients (85.7%), while data for the remaining 30 individuals were unavailable owing to limited sample quantity or minor technical constraints (TT: n = 55, TA: n = 95, AA: n = 30). Notably, individuals carrying the TA or AA genotypes exhibited markedly higher mean IL‐6 concentrations (78.5 ± 35.6 pg/mL and 85.3 ± 40.1 pg/mL, respectively) compared with those possessing the TT genotype (45.2 ± 20.1 pg/mL; p < 0.001). In addition, elevated IL‐6 concentrations demonstrated a statistically significant positive association with disease severity, aligning with the established role of cytokine hyperactivation in severe COVID‐19 pathology.

3.6. Survival and Predictive Performance

Survival analysis using the Kaplan–Meier method demonstrated that individuals carrying the TT genotype experienced a notably shorter hospitalization period compared with those harboring the TA or AA genotypes (log‐rank test, p < 0.01) (Figure 1). To further examine the joint prognostic contribution of genetic background and inflammatory status, a receiver operating characteristic (ROC) curve was generated. For ROC modelling, TA and AA were grouped as non‐TT genotypes to increase statistical power, although clinical severity was predominantly associated with the TA genotype. When the high‐risk genotype group (TA) was combined with an elevated NLR > 5, the model achieved an area under the ROC curve of 0.82 (95% CI: 0.76–0.88) for identifying severe or critical COVID‐19 cases—outperforming each individual parameter alone (AUC for genotype = 0.67; AUC for NLR = 0.75) (Figure 2). The improvement in AUC for the combined model compared with NLR alone was statistically significant (DeLong test, p = 0.04).

FIGURE 1.

Kaplan–Meier survival curve for time to discharge. Kaplan–Meier curves depict the probability of remaining hospitalized over time for COVID‐19 patients according to IFN‐γ +874 T/A genotype. The x‐axis represents the duration of hospitalization in days, and the y‐axis represents the cumulative proportion of patients remaining in the hospital. Tick marks indicate censored observations (discharge events). The TT genotype group demonstrated a significantly shorter median time to discharge compared with the TA and AA genotype groups (log‐rank test, p < 0.01). Median hospital stay and interquartile range for each genotype group are shown in the inset table. This survival analysis was performed on all 210 COVID‐19 patients with complete hospitalization data.

FIGURE 2.

Comparative ROC curves for predicting severe/critical COVID‐19. Receiver operating characteristic (ROC) curves comparing the discriminatory performance of three models for identifying severe or critical COVID‐19 cases (composite outcome): (1) NLR alone (AUC = 0.75, 95% CI: 0.67–0.83), shown as the dashed line; (2) high‐risk genotype alone (TA/AA vs. TT; AUC = 0.67, 95% CI: 0.59–0.75), shown as the dotted line; and (3) the combined model integrating high‐risk genotype (TA/AA) and elevated NLR (> 5) (AUC = 0.82, 95% CI: 0.76–0.88), shown as the solid line. The diagonal reference line (AUC = 0.50) represents a non‐discriminatory test. The improvement in AUC for the combined model compared with NLR alone was statistically significant (DeLong test, p = 0.04). AUC, area under the curve; CI, confidence interval; NLR, neutrophil‐to‐lymphocyte ratio; ROC, receiver operating characteristic.

3.7. Comparison With Other Populations and Clinical Outcomes

Population allele frequencies for rs2430561 were verified using the ALFA (Allele Frequency Aggregator) database and previously published association studies, ensuring that the comparative values reported in Table S1 are consistent with publicly available genetic epidemiology resources. Because rs2430561 lies within a repetitive CA‐repeat region of intron 1, it is not consistently captured by short‐read sequencing datasets such as the 1000 Genomes Project. Therefore, population frequency comparisons were primarily based on the ALFA database and previously published genotyping studies. The A allele frequency in our Iranian control group was similar to that observed in South Asian populations (36%–42%) but higher than in European (approximately 43%–48%) and African (approximately 24%–30%) populations. Regarding clinical outcomes, only two patients (0.95%) died during hospitalization, precluding survivor versus non‐survivor analysis. IMV was required in 10 patients (4.8%), and no significant association was observed between IFN‐γ genotype and IMV requirement (p > 0.05).

4. Discussion

This research undertook a case–control investigation in southern Iran to elucidate the influence of the IFN‐γ +874 T/A single nucleotide polymorphism on susceptibility to SARS‐CoV‐2 infection and subsequent clinical outcomes. The participants comprised 210 confirmed COVID‐19 patients and an equivalent number of demographically matched healthy controls. Genetic analysis demonstrated that possession of the heterozygous TA genotype or the A allele significantly heightened susceptibility to infection. Multivariate analysis established the TA genotype as an independent determinant of disease severity, conferring a 2.85‐fold increased likelihood of progression to severe or critical illness. In contrast, the homozygous AA genotype was significantly correlated with mild disease phenotype (Table 4). Phenotypically, carriers of the risk‐associated genotypes presented with significantly elevated serum IL‐6 and extended hospitalization durations. Direct measurement of IFN‐γ was not performed; consequently, the functional implications of these polymorphisms were extrapolated from previously documented correlations between this genotype and its cytokine expression profile. Therefore, the proposed mechanistic interpretation should be considered hypothesis‐generating, and future studies incorporating direct quantification of circulating IFN‐γ are required to confirm this biological link.

The association between the TA/AA genotypes and elevated IL‐6 levels (p < 0.001) is consistent with the concept of a cytokine network in which IFN‐γ dysregulation amplifies downstream inflammatory cascades. Other proinflammatory cytokines, such as TNF‐α and IL‐1β, were not measured in this study but may also be influenced by the IFN‐γ genotype, given the well‐established role of IFN‐γ signaling in priming macrophages and synergistically upregulating the transcription of these downstream inflammatory mediators [14, 15].

The clinical features observed in the patient group, such as common respiratory manifestations (fever, cough, and shortness of breath), together with hematological changes, including reduced lymphocyte counts, increased neutrophil levels, and elevated inflammatory indicators (CRP, D‐dimer, and LDH) [13, 16, 17], aligned with established profiles of hospitalized COVID‐19 patients, confirming the population's representativeness. These findings were consistent with other Iranian studies reporting severe outcomes in males and comorbid individuals [18], high CRP in inpatients [19], and similar abnormalities in Northern Iran [20], thereby validating the clinical context.

The link between the A allele (often tied to lower IFN‐γ production) and increased COVID‐19 risk seems paradoxical. However, this aligns with findings from the SARS‐CoV‐1 outbreak, where the same allele was a risk factor [21]. This paradox, aligning with the “Goldilocks” principle of immunity, where a balanced response is optimal, may be explained by a dysregulated immune response. An initially weak interferon reaction may fail to control the virus, leading to a pathological inflammatory cascade [22]. In this temporal framework, a delayed or insufficient early interferon response may permit enhanced viral replication, which subsequently drives compensatory hyperinflammatory pathways, including excessive IL‐6 production. In our dataset, this context dependence was evident, as the AA genotype correlated with milder clinical presentation, while the TA genotype predominated among severe and critical cases. This model accounts for why AA/TA genotypes can be associated with either mild illness or severe hyperinflammation [23], while the TA genotype (intermediate‐to‐high IFN‐γ) predicts severe disease. This apparent paradox can be rationalized through the biphasic role of IFN‐γ in COVID‐19 pathogenesis. Although the AA genotype is traditionally linked to lower baseline IFN‐γ production, this lower expression may prevent the immune system from transitioning into an uncompensated hyper‐inflammatory state. In contrast, intermediate‐to‐high producers (TA and TT carriers) may undergo a dysregulated, persistent interferon response that drives downstream cytokine cascades, as evidenced by the significantly higher IL‐6 levels and NLR observed in our TA cohort. By keeping interferon‐mediated cellular recruitment in check, the AA genotype appears to act as a regulatory buffer against immunopathology. However, because direct serum IFN‐γ quantification was not performed in this study, this kinetic model remains a mechanistic hypothesis that requires direct proteomic confirmation in future studies. The polymorphism's effect is context‐dependent; in an Iranian HCV study, the TA genotype correlated with the highest IFN‐γ levels [24], suggesting that in COVID‐19, this genotype might drive a dysregulated hyperinflammatory response, as seen with high IL‐6, rather than a simple deficiency. This contrasts with the traditional view of the A allele solely conferring a “low‐producer” phenotype and highlights the genotype‐specific and disease‐specific nature of this polymorphism's effect [25]. Our finding supports two potentials, non‐exclusive pathways to severe disease: an initially insufficient interferon response that fails to control viral replication, or a subsequent dysregulated and excessive inflammatory reaction. This was corroborated by clinical outcomes, as Kaplan–Meier analysis demonstrated a significantly longer duration of hospitalization for patients with TA/AA genotypes. Multivariate logistic regression analysis demonstrated that the TA genotype independently predicted progression to advanced stages of COVID‐19, yielding an adjusted odds ratio of 2.85 after accounting for potential confounding factors such as age, sex, diabetes, hypertension, and increased NLR. Regarding the relationship between the IFN‐γ variant and the NLR, we hypothesize that the TA genotype leads to a dysregulated, delayed, or excessive IFN‐γ response, which may promote prolonged neutrophil activation and lymphocyte apoptosis, thereby elevating NLR. Although direct IFN‐γ measurement was not performed, the strong association between the TA genotype and both high IL‐6 and high NLR supports this inflammatory pathway. Direct causal inference, however, requires further experimental validation.

Beyond establishing genetic susceptibility, evaluating the practical clinical utility of the IFN‐γ + 874 T/A polymorphism in risk stratification provides key prognostic insights for patient management. Specifically, when integrating host germline genetics with dynamic inflammatory markers such as NLR, our combined predictive model demonstrated superior prognostic accuracy (AUC = 0.82) compared to established clinical parameters alone. The clinical relevance of this polymorphism is further emphasized by its association with prolonged hospitalization, with the TT genotype linked to a significantly shorter time to discharge. Moreover, integrating the high‐risk genetic profile (TA/AA genotypes) with the NLR yielded a powerful predictive model for severe disease, outperforming either marker alone. This combined genetic and clinical approach aligns with the development of multi‐parameter prognostic models for COVID‐19 [26]. However, internal validation procedures such as bootstrapping or cross‐validation were not performed; therefore, the possibility of model overfitting cannot be excluded.

Our findings differ from those of a recent study conducted in northwestern Iran by Seyyed Rezaei et al., which did not identify a significant association between the IFN‐γ +874 T/A polymorphism and COVID‐19 severity [27].

This discrepancy highlights the complex and potentially population‐specific nature of genetic associations [28]. Differences in ethnicity, regional genetic backgrounds, environmental factors, or viral strains between the northwestern and southern regions of Iran could account for these varying results. Furthermore, the study by Seyyed Rezaei et al. [27] focused exclusively on critically ill ICU patients, whereas our study enrolled patients across all severity strata (mild to critical), which may also influence the genetic associations observed. Our study was conducted during a period dominated by the Omicron variant in Iran. Omicron is generally associated with less severe lower respiratory tract involvement compared to Delta or earlier strains; the persistent, robust association of the IFN‐γ +874 TA genotype with severe disease (OR = 2.85) underscores that host genetic risk factors remain a critical determinant of pathogenesis regardless of viral lineage. It is plausible that in the context of more virulent ancestral variants, which trigger faster viral replication and higher baseline antigen loads, the dysregulated hyperinflammatory response in TA/AA carriers might be even more severe. Therefore, while our specific odds ratios reflect the Omicron clinical landscape, the fundamental biological principle of IFN‐γ polymorphism‐driven susceptibility is highly likely to be generalizable across different SARS‐CoV‐2 lineages. In addition, recent large‐scale genome‐wide association studies and international consortia analyses have demonstrated that COVID‐19 susceptibility and severity are influenced by multiple genetic loci across interferon signaling, inflammatory, and host defense pathways. These studies underscore the polygenic architecture of COVID‐19 and support the importance of integrating candidate gene findings, such as the IFN‐γ +874 polymorphism, within a broader genetic context [29].

Furthermore, our study enrolled patients across all severity strata (mild to critical), whereas the aforementioned study focused exclusively on critically ill ICU patients, which may also influence the genetic associations observed. Furthermore, the risk associated with the IFN‐γ polymorphism presents a contrast to protective variants identified in other immune pathways, such as the Type I interferon receptor gene IFNAR2, illustrating the distinct roles these interferon pathways play in SARS‐CoV‐2 pathogenesis [30]. This study focused on a single functional polymorphism within IFNG. Broader pathway‐based analyses including additional interferon‐related or innate immune genes may provide a more comprehensive understanding of genetic susceptibility.

The clinical applicability of this genetic marker was enhanced through its integration with a routine biomarker. A model combining the high‐risk genotypes (TA/AA) with an elevated NLR yielded a powerful predictor for severe/critical disease. This combined model demonstrated superior predictive performance compared to the genetic variant or NLR alone, aligning with the paradigm of multi‐parameter risk stratification for COVID‐19 prognostication [13, 26]. This approach of combining a key genetic marker with a clinical biomarker represents a step towards the emerging concept of polygenic risk scores (PRS) for COVID‐19, demonstrating the potential to identify high‐risk individuals more accurately.

This research affirms that a multi‐parameter approach, integrating genetic and clinical data, is vital for accurate COVID‐19 prognostication, a concept supported by Balzanelli et al. [31]. Our findings reveal that the IFN‐γ +874 TA/AA genotypes and A allele confer a higher risk of severe outcomes, underscoring the significant role of host genetics. This contrasts with studies on other immune pathways, such as TLR4, where different genetic variants are protective [32, 33, 34, 35, 36]. These findings should be interpreted cautiously, as causal mechanisms remain to be validated through direct cytokine quantification and replication in larger, multi‐ethnic populations.

Population comparisons indicate that the A allele frequency in the Iranian control group (34.5%) falls within the range reported for other global populations, although variation between populations has been observed. Estimates from the Allele Frequency Aggregator (ALFA) database suggest approximate A‐allele frequencies of about 43%–48% in European populations, 36%–42% in South Asian populations, and 24%–30% in African populations. These differences emphasize the importance of conducting genetic association studies in diverse populations, as allele frequencies and linkage patterns may vary substantially between ethnic groups. Such variation can influence both statistical power and the interpretation of genetic risk factors.

This research has some limitations. Its single‐region design in Southern Iran necessitates confirmation in larger, multi‐ethnic populations to ensure broader applicability. A principal limitation is that influence of the IFN‐γ +874 genetic variant on cytokine production was inferred from genotype using established literature, rather than through direct measurement of serum IFN‐γ levels in the patients themselves. Furthermore, the observation that the intermediate‐expression TA genotype conferred a greater risk of severe disease than the low‐expression AA genotype is intriguing and merits deeper investigation into the underlying immune mechanisms. As a hypothesis‐driven candidate gene study, formal multiple testing corrections (e.g., Bonferroni) were omitted to minimize Type II errors. While our primary multivariate outcomes were highly robust (p = 0.001), exploratory subgroup findings carry an increased risk of Type I errors and should be interpreted with caution until validated in independent cohorts. Finally, the limited number of participants, particularly within the critical patient samples (n = 15), may reduce the statistical power available for conducting more detailed subgroup analyses. This limited sample size in certain strata is reflected by wider confidence intervals, suggesting reduced precision of some effect estimates. Although our cohort was adequately powered (87.2%) to identify primary genetic susceptibility risks, the statistical power for secondary subgroup stratifications—such as evaluating independent risks within specific COVID‐19 severity grades—was limited due to the smaller sample sizes in individual clinical arms. Consequently, these stratified findings should be interpreted as exploratory and require validation in larger multi‐center cohorts. Additionally, while we reported on missing data for IL‐6, the potential impact of other missing data was not explored through sensitivity analyses. The absence of formal gene–gene interaction analyses and the lack of systematic recording of detailed clinical symptom profiles across all severity groups represent additional limitations. We recognize the possibility of selection bias, given that our study population was drawn from one tertiary care center, which might limit the generalizability of our findings to the broader population of individuals affected by COVID‐19. Additionally, formal assessment of ethnic background or genetic ancestry markers was not performed; therefore, residual population stratification cannot be entirely excluded. Although controls were recruited from the same catchment area and screened for active infection, hospital‐based recruitment may not fully represent the general population structure. Lastly, although controls were matched to cases by age and sex, conducting an unmatched analysis for the primary association may theoretically allow some residual confounding. Lack of comprehensive vaccination data may represent a potential unmeasured confounder. Although COVID‐19 vaccines are known to induce Th1‐biased immune responses and may transiently increase IFN‐γ production, our study focused on a germline genetic variant rather than cytokine quantification. Therefore, vaccination would not affect genotype distribution, although it could theoretically influence inflammatory marker levels. Nonetheless, the potential influence of these matched variables was accounted for in the multivariable analysis evaluating disease severity. The use of hospital‐based controls, while efficient, may introduce selection bias if the health profiles of hospital visitors differ systematically from the general population; this was a considered trade‐off for feasibility and regional matching.

5. Conclusion

The results of this study suggest that the IFN‐γ +874 polymorphism is associated with variability in COVID‐19 clinical outcomes in the studied population, with the heterozygous TA genotype linked to increased disease severity. When combined with inflammatory biomarkers such as NLR and IL‐6, genetic information may help improve the assessment of disease risk. Continued investigation of immune‐related genetic variants may enhance our understanding of host–pathogen interactions and contribute to more personalized approaches for the management of infectious diseases.

Author Contributions

Nasir Arefinia, Emad Behboudi designed the experiments; Nasir Arefinia, Bahman Aghcheli performed experiments and collected data; Seyed Mohammad Ali Hashemi and Zohreh‐al‐sadat Ghoreshi discussed the results and strategy; Emad Behboudi supervised, directed, and managed the study; Nasir Arefinia, Bahman Aghcheli, Seyed Mohammad Ali Hashemi, Zohreh‐al‐sadat Ghoreshi, and Emad Behboudi approved of the final version to be published.

Funding

The authors have nothing to report.

Ethics Statement

This study received approval from the Ethics Committee of Kerman University of Medical Sciences (Ethics Code: IR.JMU.REC.1402.120). All participants provided written informed consent, in accordance with the principles outlined in the Declaration of Helsinki.

Consent

Informed consent was obtained from all individual participants included in the study.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Representative agarose gel electrophoresis of ARMS‐PCR products for IFN‐γ +874 T/A genotyping. Representative 2% agarose gel electrophoresis image stained with Safe Stain showing the amplification products of the Amplification‐Refractory Mutation System PCR (ARMS‐PCR) for IFN‐γ +874 T/A (rs2430561) genotyping. Lane Ladder: 100 bp DNA ladder; Lane 1–2: TT genotype showing the 262 bp IFN‐γ‐T allele specific band+ The 110 bp beta‐globin internal control band; Lanes 3–4: AT genotype showing the 262 bp A + T allele‐specific band+ The 110 bp beta‐globin internal control band; Lanes 5–6: AA genotype showing the 262 bp A allele‐specific band+ The 110 bp beta‐globin internal control band. The 110 bp beta‐globin band served as an internal amplification control for each reaction to verify DNA integrity and successful PCR amplification. PCR conditions: initial denaturation at 95°C for 5 min; 35 cycles of 95°C for 30 s, 60°C for 40 s, 72°C for 40 s; and final extension at 72°C for 5 min. Lanes: M: DNA ladder; 1–2: TT genotype (262 bp); 3–4: AT genotype (262 bp); 5–6: AA genotype (262 bp).

Table S1: Comparison of IFN‐γ +874 A Allele Frequency in the present study with Global populations.

Table S2: Clinical Symptoms at Admission Stratified by COVID‐19 Severity Group.

Table S3: Inflammatory and Biochemical Markers Stratified by IFN‐γ +874 T/A Genotype in COVID‐19 Patients.

Data Availability Statement

The data supporting this study's findings are available from the corresponding author upon reasonable request.