Alloimmunization in autoimmune hemolytic anemia patient: The differential adsorption approach.

Patients of β-thalassemia major are dependent on regular blood transfusions for their entire lifetime. Development of antibodies against red blood cell (RBC) antigen which may be alloantibody or autoantibody, several times as a result of frequent red cell component transfusions, further complicates the subsequent transfusion therapy. Among the autoantibodies, warm-reactive autoantibodies are commoner and interfere in the pretransfusion testing. These RBC autoantibodies present in patient's serum potentially react with all the cells of antibody identification panel giving "pan-reactive" picture and making alloantibody identification complex. In this report, we present our approach in a thalassemia patient who presented with warm-type autoimmune hemolytic anemia, low hemoglobin of 5.8 g/dl, and three significant alloantibodies (anti-D, anti-S, and anti-Jkb) which were masked by pan-reactive warm autoantibody(s). Differential adsorption was used to unmask underlying alloantibodies. We suggest that differential adsorption procedure is an effective and efficient method for autoantibody adsorption, detection, and identification of masked alloantibody(s), especially in patients with low hemoglobin and history of recent blood transfusion.

Autoimmune hemolytic anemia (AIHA) is characterized by abnormality in the immune system which results in the formation of autoantibodies directed against the patient's self-antigens and evidence of anemia. Many a times, anemia is severe enough to warrant red cell component transfusion. Warm-reactive autoantibodies interfere in the pretransfusion testing. These red blood cell (RBC) autoantibodies present in patient's serum potentially react with all the cells of antibody identification panel giving “pan-reactive” picture and making alloantibody identification complex. However, identifying the alloantibodies in these cases is important to prevent the occurrence of severe hemolytic reaction. In untransfused patients, the estimated incidence of alloantibody formation after blood transfusion ranges from 0.5% to 1%.[1] In chronically transfused patients, this risk increases to 20%–60%.[2] Likewise in AIHA patients, over one-third to one-half of patients have underlying alloantibody(s).[3] Patients of β-thalassemia major are dependent on regular blood transfusion for entire lifetime. The development of antibodies against RBC antigen which may be alloantibody or autoantibody further complicates the transfusion therapy. It is in these patients with autoantibodies that the transfusion requirements are also high. Many a times, these patients do not get appropriately matched blood units and get access to only partially compatible or incompatible blood for transfusion. The problem is that autoantibodies mask underlying alloantibodies and failure to recognize alloantibody(s) may cause hemolytic transfusion reactions which may be at times even life threatening and also limit the availability of subsequent safe transfusion(s).

Finding the compatible blood units in thalassemia patients with existing antibody(s) is a tedious and complex process and needs immunohematological expertise and specialized reagents in a well-equipped immunohematology (IH) laboratory. In this report, we present our approach in a thalassemia patient presented with 5.8 g hemoglobin with AIHA.

Case Report

A 23-year-old male known case of β-thalassemia major presented with severe anemia and was denied compatible blood at other hospitals before being referred to our hospital. He was admitted in hemato-oncology unit for blood transfusion since his hemoglobin was 5.8 at admission and had marked pallor and other symptoms of anemia. Antibody screen was pan reactive. With this “clinical information,” the sample was sent to our IH reference laboratory for workup.

Initial immunohematology workup

Forward blood grouping was AB positive, while in reverse grouping, there was 3+ agglutination reaction with all reagent cells (A, B, and O cells). Antibody screen was repeated and found to be pan reactive with evidence of hemolysis. Auto-control and direct antiglobulin test (DAT) was 3+ positive. This picture of anemia with positive DAT and auto-control was suggestive of possible AIHA. Blood grouping was done by conventional tube technology (CTT), and DAT was done on polyspecific antihuman globulin (AHG) column agglutination card (Ortho Clinical Diagnostics; Mumbai, India).

Repeat forward grouping (warm saline washes)

Forward grouping in CTT was repeated after washing the red cells of the patient three times with warm normal saline as per Departmental Standard Operating Procedure. The blood group of the patient was now confirmed as O Rh D negative (instead of initial false-positive typing as AB positive).

Monospecific direct antiglobulin test

DAT was repeated using monospecific card (IgG, C3d, and Control; Ortho Clinical Diagnostics; Mumbai, India) to identify the type of sensitization. Results showed IgG positive and C3d negative. Control was negative which validated the results. This picture was suggestive of warm type of AIHA.

Direct antiglobulin test-IgG dilution and IgG subclasses

On IgG dilution studies, anti-IgG titer of 1000 was identified. This titer was clinically relevant indicating risk of hemolysis and necessity to do IgG subclasses.

Patient's red cells were treated to identify responsible IgG subclass for sensitization. IgG1 and IgG3 were positive.

Elution

Acid elution (Bag Systems; Germany) was performed to free the IgG antibody from sensitized red cells. Positive and negative controls were used in parallel to validate the results. Eluate showed reactivity with all 11 reagent red cells (pan reactive) suggesting autoantibody. Last wash was negative validating the elution and wash process.

Differential adsorption

As the patient's hemoglobin was low and with history of recent transfusion, differential adsorption was done to adsorb autoantibodies and to reveal underlying masked alloantibody(s).

Source of cells - The cells used for adsorption were obtained from departmental rare donor registry and donor red cell inventory.

Treatment - Untreated red cells were used for adsorption.

Methods - For this, patient's serum sample was divided into three aliquots of 1 ml each. Each aliquot was adsorbed using three different donors’ cells (R1R1, R2R2, and rr). Among the three cells, one was negative for K, Fyb, M, and S; another negative for Jka, and the third negative for Jkb as shown [Table 1]. Adsorption was done at 37°C in incubator for 1 h with intermittent agitation. A total of four sets of cells were used to fully adsorb autoantibody [Figure 1]. Following adsorption, 11 cell antibody identification panels were performed separately on each serum aliquot (R1R1, R2R2, and rr), and the reaction pattern was compared to reveal underlying alloantibodies, if any [Table 1].

Table 1

Warm autoantibody with underlying alloantibodies

Figure 1

Differential adsorption technique for detecting alloantibodies in the serum of a patient with warm-reacting autoantibodies.

Results - In “R1R1” - the reaction pattern of 11 cell identification panels matched the pattern with S antigen suggesting “Anti-S” alloantibody as R1R1 cells were negative for S antigen.

In “R2R2” - the reaction pattern of 11 cell identification panels matched the pattern again with S antigen suggesting “Anti-S” alloantibody, but R2R2 cells were positive (heterozygous state) for S antigen.

In “rr” - In 11 cell identification panels, all cells were 3 + reactive except cell no. 10 which was negative suggesting alloantibody. As “anti-S” antibody was confirmed in R1R1 cells, S-negative select cells were selected from the same 11 cell panels and 22 cell extended panels. These S-negative cells were treated with “rr” adsorbed serum and reaction pattern showed match reactivity with D and Jkb antigens suggesting “anti-D and anti-Jkb” antibodies. The rr cells were negative for S, D, and Jkb; thus, all three alloantibodies were not adsorbed while autoantibody got adsorbed.

Now, the question was why the R2R2 being positive for S antigen, still showing reaction matching with S suggesting anti-S. To resolve, these titers were done for all three alloantibodies (anti-D, anti-S, and anti-Jkb).

Titration

Alloantibodies (anti-D, anti-S, and anti-Jkb)

Adsorbed rr plasma was used for titration as rr plasma was having all three alloantibodies with no autoantibody. In this case for titration of specific antibody, we needed select cells which were positive for one antigen while negative for other two antigens. For titration of anti-D, selected cell was chosen from in-date antibody identification panel which was D antigen positive while S and Jkb antigen negative (cell no. 4 - R0r) [Table 1]. For titration of anti-S, selected cell was chosen again from sample in-date antibody identification panel which was S antigen positive while D and Jkb antigen negative (cell no. 9 - rr) [Table 1]. Moreover, for titration of anti-Jkb, selected cell (no. 21) from panel B (extended cell panel) was used which was Jkb antigen positive while D and S antigen negative [Table 2].

Table 2

Selected S antigen-negative red blood cells to confirm and exclude red blood cell alloantibodies

Titers were found to be 64 (1:64) for anti-D, 128 (1:128) for anti-S, and 16 (1:16) for Anti-Jkb.

Autoantibody

Unadsorbed normal patient's plasma was used for titration of autoantibodies. In this, we needed a cell which was negative for antigens against which alloantibodies (anti-D, anti-S, and anti-Jkb) are directed so that only autoantibody will react not the alloantibody. Thus, cell no. 10 from in-date identification panel was chosen as it was a single cell which was negative for all three antigens (D, S, and Jkb) [Table 1].

Finding compatible blood units

Two units were found compatible after typing 14 O Rh D negative units considering S antigen and Jkb antigen-negative frequency as 0.43 and 0.33, respectively.[4]

Transfusion

Units were transfused to the patient under observation, and no reaction was observed. The patient was started on high-dose intravenous methylprednisolone to counteract rapidly evolving severe hemolysis.

Discussion

In our case, a thalassemia major patient had formed three clinically significant alloantibodies which were masked by autoantibody(s). Patients with warm autoantibodies have a higher rate of alloimmunization.[1567] Branch and Petz indicate that approximately 32% of patients with warm autoantibodies have clinically significant alloantibodies.[3] The transfusion management of thalassemia patient remains challenging as these patient populations have higher rates of alloimmunization and are at high risk for hemolytic transfusion reactions. Development of warm AIHA (WAIHA) in these patients makes the picture more complex as it may delay blood transfusion and may affect patients’ clinical condition. These warm autoantibodies interfere and create more of a serologic dilemma than cold autoantibodies (agglutinins). While many of the times, reaction of cold autoantibodies is avoided by omitting room temperature testing but with WAIHAs, both clinically significant alloantibodies and autoantibodies react best at the AHG phase; therefore, masking all alloantibodies, therefore, more complicated and time-consuming procedures for resolving the problems may have to be used. These patients present with severe anemia and ongoing hemolysis. Due to lack of proper immunohematological services, these patients with warm autoantibodies frequently receive transfusion with incompatible blood.

Many patients with warm autoantibodies never require transfusion; they are managed by medical treatment. Sometimes, however, because of symptomatic anemia, transfusion becomes necessary. The patient in this case report was known case of thalassemia developed WAIHA presented with symptomatic anemia. In these cases, especially when the patient has history of transfusion or pregnancy, the primary concern will be to ensure identification of masked alloantibodies. In these cases, authors recommend the following approach:

If there is no recent history of transfusion and patient hemoglobin is good enough to get enough quantity of red cells, prepare autologous cells to be used for autoadsorption procedures

If the warm autoantibody shows reaction pattern to single antibody specificity (for, e.g., anti e), test the patients sample with panel of selected cells negative for corresponding antigen (e antigen here) and positive for other clinically significant RBC antigens

If enough red cells are not available, determine the RBC phenotype of the patient using monoclonal antisera which do not require AHG phase testing (using antisera requiring AHG phase testing will not give reliable results due to the presence of warm autoantibodies). Select phenotypically similar donor from inventory and perform adsorption. However, most of the times, these patients have history of multiple transfusions making phenotyping of patients unreliable. Thus, performing autoadsorption becomes difficult

If autoadsorption is not possible, differential allogenic adsorption can be performed by selecting three R1R1, R2R2, and rr donor units; one negative for other common significant antibodies (Jk, Fy, and S). This procedure will remove autoantibody efficiently as enough red cells will be available for sufficient number of adsorptions. Ficin or ZZAP treatment of the allogeneic adsorbing donor cells will be helpful as it increases antibody uptake. The number of adsorptions procedures required to get rid of all autoantibody(s) depends on the its amount present in the serum. Usually, if the strength of reaction of auto-control is (3+ to 4+) then more than two adsorptions may be required. Although differential allogenic adsorption is effective in removing autoantibodies, there are certain limitations too like alloantibody to a high-incidence RBC antigen will get adsorbed by allogeneic cells but not with autoadsorption as antigen will not be present on the patient's own cells. With the exception of possible adsorption of an antibody to a high-incidence antigen, allogenic adsorption seems to better alternative to autoadsorption in these anemic and multitransfused patients.

Many centers practice giving least incompatible/best-matched blood to these patients without performing adsorption procedures. This practice puts the patient at risk; however, it seems to be better than “doing nothing.”

In our case, if we had provided least incompatible blood, it could have resulted in severe hemolytic transfusion reaction. Allogenic adsorption is therefore good alternative compared to “least incompatible” blood; however, challenge is finding R2R2 donor unit from the inventory which needs typing of sometimes more than 100 donor units. However, once identified, the whole unit or part of it can be used to solve these complex problems for over 1 month. This is what we practice at our center.

In our case, R2R2 being positive for S antigen, adsorbed R2R2 serum was showing reaction that matches with S suggesting anti-S. This was because of two reasons, first high titer of anti-S 128 (1:128) and heterozygous expression of S antigen in R2R2 cells as small “s” antigen was positive in all three cells (R1R1, R2R2, and rr) used for alloadsorption.

In this case, we have used DAT dilution, and IgG subclasses as these helps decide the prognosis of these patients and also guide in treatment. Numbers of IgG molecules per red cell affect the red cell destruction in AIHA. The DAT IgG-dilution guides whether to proceed for IgG subclasses or not; if anti-IgG titer is 30 (1:30) or less, it is not relevant and determination of IgG subclasses is not required. However, if the titer is more than 100 (1:100), determination of IgG subclasses is significant as it indicates the risk of hemolysis.[7] The risk of hemolysis also depends on IgG subclasses involved. IgG1 and IgG3 may activate complement while IgG2 and IgG4 rarely do so.

Conclusion

Differential adsorption procedure is an effective and efficient method for autoantibody adsorption, detection, and identification of masked alloantibody(s).

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.