Posttransfusion Alloimmunization Following Corticosteroid Treatment

INTRODUCTION

Blood transfusions are a common medical procedure, and like any other procedure, they carry risks. Alloimmunization is a primary concern in transfusion medicine and is defined as the formation of alloantibodies in response to foreign antigens encountered during exposure events, including transfusion and pregnancy.^1,2 Alloimmunization complicates blood provision because each time an alloantibody is made, it reduces the chances of finding compatible blood and increases the likelihood of potential hemolytic transfusion reactions (HTRs).^3,4 HTRs can range from subclinical to lethal in severity and account for about 5% of serious transfusion-related adverse reactions.⁵ As the name implies, HTRs result from hemolysis of donor red blood cells (RBCs) within the recipient. This can trigger other mechanisms, such as cytokine storm, shock, disseminated intravascular coagulation, and cause death.⁶

Each time patients are transfused, they are at risk of becoming alloimmunized. This is especially true for patients with conditions that require frequent packed RBC (PRBC) transfusions, such as sickle cell disease.^1,7 There are hundreds of unique antigens present on the surface of RBCs that can induce antibody development, so mitigating alloimmunization to these antigens in patients who are frequently transfused can help minimize the risk of HTRs.⁴

Antibodies, also known as immunoglobulins, play a crucial role in the immune system by binding to foreign antigens.^8,9 Although the exact mechanisms of why some patients make alloantibodies when others do not are currently unknown, it is accepted that patients in proinflammatory states are more likely to become alloimmunized than patients in noninflammatory states.^3,10 Because inflammation may augment antibody formation, immunosuppressive therapy may offer protection against alloantibody production.

Because inflammation promotes alloimmunization, anti-inflammatory agents such as corticosteroids, which exert immunosuppressive effects, may reduce the risk of alloimmunization.¹⁰ Immunosuppressants are a classification of medication that work by inhibiting the immune system, and these drugs are commonly used in organ transplantation to help prevent organ rejection by the recipient’s body.¹¹ In fact, use of immunosuppressive therapy, primarily rituximab, was recommended by the American Society of Hematology (ASH) in patients deemed at high risk for acute HTRs or severe, life-threatening delayed HTRs; however, these guidelines are conditional because there is little evidence about its effects.⁷ Among the various immunosuppressive agents, corticosteroids represent a widely used class with well-established anti-inflammatory mechanisms.

Corticosteroids are used to treat inflammatory disorders and allergies by inhibiting cytokine transcription after binding to intracellular glucocorticoid receptors. This nonspecific inhibition of cytokine transcription enables a wide range of applications.¹² Corticosteroids bind to glucocorticoid receptors in the cytoplasm before entering the nucleus to exert anti-inflammatory effects.¹³ Corticosteroids can be categorized according to the duration of action: short acting, intermediate acting, and long acting. The most common short-acting corticosteroid is hydrocortisone, and the most common long-acting corticosteroid is dexamethasone; prednisone, prednisolone, and methylprednisolone are all common intermediate-acting corticosteroids.¹⁴

If corticosteroids are found to have a protective effect against first-time alloimmunization, then future research could be conducted to determine if corticosteroids can be used to prevent alloimmunization in patients who are frequently transfused and are at a higher risk of having been previously alloimmunized. A 2014 study in the Netherlands by Zalpuri et al is the primary evidence supporting the use of immunosuppressive drugs to reduce the risk of alloimmunization.^15⇓-17 The researchers concluded that there was a lower risk of alloimmunization with immunosuppressant usage in the 30 days preceding the date of the transfusion, but there have been few attempts to extend the observation of the protective effect of immunosuppressants—specifically corticosteroids—in other settings.¹⁵

The purpose of this study was to investigate the relationship between corticosteroid use and first-time alloimmunization in nonpregnant adults undergoing PRBC transfusions. The relationship was investigated by comparing the demographic characteristics of patients who received corticosteroid treatment and those who did not, comparing alloimmunization between these groups, and evaluating the effects of different corticosteroid treatments on alloimmunization. It was hypothesized that corticosteroid exposure within 30 days before transfusion would be associated with a lower risk of first-time alloimmunization.

METHODS

This was an observational, retrospective cohort study spanning January 1, 2014, to December 31, 2024. The study was conducted using data collected from a multisite academic health system located along the Texas Gulf Coast. The health system transfuses approximately 12 000 blood products per year. Over the period, 708 eligible patients were identified, and all were included in the study. This study was reviewed through an expedited review procedure and approved under protocol 24-0404 by the University of Texas Medical Branch at Galveston institutional review board in accordance with 45 Code of Federal Regulations 46.110(a)-(b)(1).

RESULTS

All statistical analyses were performed using SAS Studio 3.82 on SAS 9.4 (SAS Institute Inc, Cary, NC), with statistical significance set at P less than .05. Normality assumptions for age were assessed using the Shapiro–Wilk test to determine whether to use a parametric or nonparametric test. Assumptions of multicollinearity and outliers were checked for the logistic regression models.

The Wilcoxon rank-sum test revealed no significant difference in age between those who received corticosteroid treatment and those who did not (P = .806); age distributions did not differ significantly (Table 1). Chi-square revealed no significant association between sex and treatment (P = .364) or between ethnicity and treatment (P = .414; Table 1). Fisher exact test indicated no significant association between race and treatment (P = .667; Table 1). Fisher exact also revealed no significant association between treatment and alloimmunization (P = .098), with an OR indicating imprecision and no clear evidence of an association between corticosteroid exposure and first-time alloimmunization (Table 2).

Sparse outcome data limited class-specific corticosteroid analyses. Among patients not receiving corticosteroid treatment, only 20 developed first-time alloimmunization (Table 3). Among patients receiving corticosteroid treatment, first-time alloimmunization occurred in only 1 patient, observed in the intermediate-acting corticosteroid group. Corticosteroid classification was not significantly associated with alloimmunization (Table 3). Given the sparse class-specific event counts, exploratory Firth’s penalized logistic regression was performed using binary corticosteroid exposure to assess whether the observed association changed after limited adjustment. All 3 models converged successfully, and the estimated association was consistent across the treatment-only; treatment and age; and treatment, age, and ethnicity models (Table 4). In each model, the treatment-term Wald P value remained nonsignificant, and the 95% CIs were wide and crossed 1.0, indicating no statistically significant association between corticosteroid exposure and first-time alloimmunization in these exploratory analyses (Table 4).

DISCUSSION

Alloimmunization remains a significant concern in transfusion medicine because it complicates future transfusions with the possibility of HTRs. Mitigating alloantibody formation can help minimize the risk of HTRs.⁴ The current study evaluated whether corticosteroid exposure was associated with reduced first-time alloimmunization. Overall, the point estimates suggest a possible reduction of first-time alloimmunization, but the association was not statistically significant. In the 2020 ASH guideline, the use of immunosuppressive therapy, primarily rituximab, was recommended for patients with sickle cell disease deemed to be at high risk for acute HTRs or severe, life-threatening delayed HTRs.^7,16 The use of immunosuppressive therapy to prevent alloimmunization is primarily supported by the Zalpuri et al 2014 case-referent study, in which researchers concluded there was a lower risk of alloimmunization with immunosuppressant usage in the 30 days preceding the date of the transfusion.^4,10,15,17

This study aimed to further explore the findings of Zalpuri et al by conducting a similar investigation in a different setting and to extend the original study by examining the protective effects of corticosteroids specifically in cases of first-time PRBC alloimmunization.¹⁵ If future studies demonstrate effectiveness, corticosteroids could present a less costly preventive strategy than monoclonal immunosuppressive therapy. A comparison between the unit prices of employer-sponsored insurance (ESI) and Medicare, based on 2020 pricing, showed rituximab (a monoclonal antibody immunosuppressant) to be $1122.90 on ESI and $936.30 on Medicare, whereas dexamethasone (a corticosteroid immunosuppressant) was $0.50 on ESI and $0.10 on Medicare.¹⁸

The original Zalpuri et al study reported a lower risk of alloimmunization among patients receiving immunosuppressive medications, with an adjusted relative rate of 0.55.¹⁵ The current study reported an OR of 0.181 (95% CI, 0.004–1.149), which is directionally compatible with the findings in the Zalpuri et al study but imprecise, with wide CIs.¹⁵ The results from the exploratory Firth’s penalized logistic regression yielded stable but not significant treatment estimates across all the models. This direction was broadly consistent with the results of Zalpuri et al, but in the present study, the adjusted findings remained imprecise and generated only further hypotheses.¹⁵ Further studies can be done to investigate the protective effect of corticosteroids in cases of repeat alloimmunization.

This study included only patients aged 18 years and older. It was conducted within a single health care system, which may limit generalizability to pediatric populations and to institutions with different transfusion practices or patient case mix. Additionally, patients without repeat antibody testing were excluded because they could not be classified according to alloantibody development. Excluded patients were not characterized demographically because the institutional statistician did not provide these data to the principal investigator. This method may produce a nonrepresentative cohort, potentially introducing bias. Although the point estimate for corticosteroid exposure (OR) was below 1.0, the wide CI precluded definitive conclusions about the direction or magnitude of the association. This uncertainty is likely driven by the low number of alloimmunization events observed, which reduces statistical power and yields unstable effect estimates.

Corticosteroid exposure may also be affected by confounding by indication because patients receiving corticosteroids often differ clinically from those who do not (eg, inflammatory disease burden, illness severity, or transfusion context), and residual confounding may remain despite adjustment. Exposure misclassification is possible if corticosteroids were administered outside the captured record, and the 30-day exposure window may not fully account for heterogeneity in dose, route, or timing relative to transfusion, potentially diluting a biologically meaningful effect. Additional studies are needed with larger cohorts, more events, and more granular characterization of immunosuppression to better define whether corticosteroid exposure influences alloantibody formation after transfusion.

In conclusion, corticosteroid exposure within 30 days before transfusion was not significantly associated with first-time alloimmunization in this cohort. Although the point estimate suggested a possible reduction in risk, the small number of alloimmunization events resulted in wide CIs and limited precision, and exploratory Firth’s penalized logistic regression likewise failed to identify a significant adjusted association. These findings suggest that, if present, any protective effect of corticosteroids could not be demonstrated in this study and should be interpreted cautiously. Larger studies with more events and more detailed characterization of corticosteroid exposure are needed to better determine whether pretransfusion corticosteroid exposure influences alloantibody formation.