Serological analysis of autoantibody mimicking anti-Rh(e) specificity: a case report

Abstract

Mimicking antibodies are described as autoantibodies with the “wrong” specificity, possessing both specificity and panreactivity. We report the case of a female patient with suspected hemolytic anemia in whom an autoantibody mimicking anti-Rh(e) specificity was found in both plasma and red blood cell samples. Mimicking antibodies are not usually identified in the clinic; however, it is still necessary to determine the specificity of mimicking antibodies to avoid antigen-positive blood transfusions and achieve maximum transfusion compatibility.

Introduction

Mimicking antibodies are described as autoantibodies with the “wrong” specificity, showing not only apparent specificity but also panreactivity in red blood cell (RBC) panel testing. Mimicking antibodies are generally considered to be uncommon in routine blood bank serology. In most patients, autoantibodies react with all RBCs resulting in agglutination; however, some patients have autoantibodies with no apparent specificity in RBC panel tests, representing the most striking feature of mimicking antibodies.^1–4

In this study, we report the discovery of autoantibodies mimicking anti-Rh(e) specificity in a patient with suspected autoimmune hemolytic anemia. The mimicking antibody was detected in the patient’s plasma and RBC eluate. In addition, we performed major crossmatching to select appropriate donor RBCs for transfusion.

Case report

An approximately 60-year-old woman presented to the hospital with chest tightness, shortness of breath, fatigue, dizziness, and tinnitus. Her past medical history included myasthenia gravis and interstitial pneumonia and her current medications included pyridostigmine and hormones. She showed generalized jaundice and examination revealed severe anemia (hemoglobin 37 g/L) and high bilirubin, indicating hemolytic anemia, requiring transfusion. No further laboratory evaluation of the anemia was performed. She had a history of pregnancy, no history of blood transfusions, and no known history of RBC alloimmunization.

Antibody screening and identification

Routine inpatient laboratory examinations, including RBC type and antibody screening, revealed type O, D+ and positive antibody screening. Further antibody screening was carried out using antibody screening cells (Bioxun Biotech, Changchun, China) and 11-cell (REAGENS, Budapest, Hungary) or 16-cell (Sanquin, Amsterdam, Netherlands) antibody identification panels. The corresponding cell suspension and plasma were added into different media, including saline, polybrene (Baso, Zhuhai, China), and gel Coombs cards (DiaMed, Cressier FR, Switzerland), respectively. For the saline method, the cells and plasma were mixed in the tube and centrifuged at 1000 × g for 15 s, and for the polybrene method, the cells and plasma were mixed in the tube followed by the addition of low ionic medium and polybrene, and the supernatant was removed by centrifugation at 1000 × g for 10 s. The results were observed after adding the resuspending solution. For the gel Coombs card method, the cells and plasma were added to the cards, mixed slightly, and incubated at 37°C for 15 minutes and the results were observed after centrifugation at 1000 × g for 10 minutes. The results of the saline method were negative, but the polybrene method revealed I (±), II (1+), III (1+^W), suggesting the presence of immunoglobulin (Ig)G (Table S1). An 11-cell antibody identification panel was used and serological tests were performed using gel testing, according to the manufacturer’s instructions. The antibody identification panel showed panagglutination in all 11 cells. Notably, the agglutination intensity of cell 2 (1+) was weaker than that of the other 10 cells (Table 1). The reaction pattern showed suspected anti-Rh(e) specificity in addition to panreactivity.

Table 1.

Patient’s antibody identification using an 11-cell identification panel

sno	Rh-Hr						Kell		MNS				P	Lewis		Duffy		Kidd		Di	Kell		Luther		Result
	D	C	E	c	e	Cw	K	k	M	N	S	s	P1	Lea	Leb	Fya	Fyb	JKa	Jkb	Dia	Kpa	Kpb	Lua	Lub
1	+	+	0	0	+	0	0	+	+	+	0	+	+	0	+	0	+	+	0	+	0	+	0	+	2+
2	+	0	+	+	0	0	0	+	+	0	+	0	+	0	+	+	0	+	0	0	0	+	0	+	1+
3	0	+	+	+	+	0	0	+	0	+	+	0	+	+	0	+	0	+	+	0	0	+	0	+	2+
4	0	0	0	+	+	0	0	+	0	+	+	+	0	0	+	0	+	+	+	0	0	+	0	+	2+
5	0	0	0	+	+	0	+	0	+	0	+	+	+	0	+	+	+	+	0	0	0	+	0	+	2+
6	+	0	0	+	+	0	0	+	0	+	0	+	+	0	+	+	0	+	0	0	0	+	0	+	2+
7	+	+	0	0	+	+	+	+	+	0	+	0	0	0	0	+	0	0	+	0	0	+	0	+	2+S
8	0	0	0	+	+	0	0	+	+	0	0	+	0	0	+	0	+	0	+	0	0	+	+	+	2+S
9	+	+	+	+	+	0	0	+	+	+	0	+	0	0	0	0	+	+	0	0	0	+	0	+	2+
10	+	+	0	0	+	0	0	+	+	0	+	+	+	0	+	0	+	0	+	+	+	+	0	+	2+
11	0	0	0	+	+	0	0	+	+	+	0	+	+	+	0	+	+	0	+	0	0	+	0	+	2+S

+, Presence of antigen; 0, absence of antigen; sno, serial number; S, strength. Agglutination score ranged from 0 to 4+.

Direct antiglobulin test

We performed a direct antiglobulin test using the tube method. The patient’s RBCs were washed three times with saline, diluted in saline to concentrations of 2%–5%, and reacted with anti-IgG, anti-C3d, and anti-IgG+C3d reagents (Bioxun Biotech). The agglutination intensity was observed after centrifugation at 1000 × g for 15 s. The results were positive (1+) for anti-IgG+C3d and anti-IgG, but negative for anti-C3d. Similarly, the results were positive (3+) for anti-IgG+C3d using the column agglutination technique. These results suggested that RBCs were sensitized by IgG.

Adsorption-elution technique

To identify the specificity of the antibodies bound to RBCs, the patient’s RBCs were treated using an acid elution method (BX-1 sample releaser, Bioxun Biotech), according to the manufacturer’s recommendations. Eluates of the patient’s RBCs and Rh(e)-negative RBCs adsorbed with patient plasma were tested using a 16-cell antibody identification panel for the patient’s RBC eluate to improve discrimination ability. The antibody in the eluate showed panagglutination. The agglutination intensities of cells 3, 6, and 12 were 1+^S, 2+, and 2+, respectively, which were significantly weaker than that of the other panel cells (3+) (Table S2). Consistent with the results of plasma antibody identification, eluate antibody identification showed anti-Rh(e) specificity and panreactivity. The alloantibody was serologically specific and was only adsorbed from the plasma by RBCs expressing the corresponding antigen, while the mimicking antibody was adsorbed regardless of the presence or absence of the corresponding RBC antigen. To exclude the possibility of autoantibodies combined with alloantibodies, Rh(e)-negative RBCs (type O, ccDEE) were incubated with the patient’s plasma and the eluate of adsorbed Rh(e)-negative RBCs was used for antibody identification. The agglutination intensity of cells 3, 6, and 12 were all 1+, which was significantly weaker than that of the other panel cells (2+) (Table S2). Antibody identification consistently showed anti-Rh(e) specificity and panreactivity. These results demonstrated the presence of autoantibody mimicking anti-Rh(e) specificity in the patient.

Major crossmatch test

Rh(e)-negative (donor 1 and 2) and Rh(e)-positive (donor 3) RBCs from type O donors were screened for reaction with the patient’s plasma for major crossmatching, using the saline, polybrene, and gel Coombs card methods. The agglutination intensities of donors 1 and 2 RBCs were 1+, which was weaker than that of donor 3 and the patient’s RBCs (auto control) (Table 2). According to the principle of compatible blood transfusion, Rh(e)-negative RBCs were recommended for transfusion.

Table 2.

Major crossmatching.

Donor RBCs	Result
	IS	Poly	Coombs
Donor 1(O, ccDEE)	0	1+	1+
Donor 2(O, ccDEE)	0	1+	1+
Donor 3(O, CCDee)	0	2+	2+
Auto control(O, CCDee)	0	2+	2+

IS, immediate spin; Poly, polybrene test. Agglutination score ranged from 0 to 4+.

In our case, consistent ABO and Rh serological phenotype blood was transfused on the second day, but routine blood examination on the next day showed a poor transfusion effect, despite the use of glucocorticoids (Figure 1). After confirmation of the mimicking antibody, Rh(e)-negative donor RBCs were selected because of their weaker agglutination in the major matching test and were eventually used for clinical transfusion. Type O ccDEE RBCs were then transfused on the sixth day. The patient’s hemoglobin was significantly elevated, reaching 52 g/L for the first time since hospitalization (Figure 1). There was no evidence of hemolytic transfusion reaction and the desired effect of blood transfusion could be achieved.

Figure 1.

Hemoglobin trend in patient since admission. Red dots indicate blood transfusion that day. RBC, red blood cell.

Discussion

In the current case, an antibody with anti-Rh(e) specificity was detected in both the patient’s plasma and RBC eluate. This antibody reacted with all RBCs on the RBC panel, showing panagglutination. The patient had no history of blood transfusion and her RBCs were type CCDee, and we therefore suspected mimicking antibody rather than alloantibody combined with autoantibody. We used Rh(e)-negative RBCs to adsorb antibodies from the patient’s plasma, and the resulting RBC eluate was specific for anti-Rh(e), confirming the presence of an autoantibody mimicking anti-Rh(e) specificity in the patient’s plasma.

Issitt et al. ⁵ reported that partially adsorbed autoantibodies apparently mimicking specificity occurred in 21.0% (29/138) of patients with warm autoantibodies in the ZZAP adsorption method. Jang et al. ⁶ found that 26.8% (19/71) of patients with warm autoantibodies had autoantibodies with mimicking specificity by the dilution technique. The specificity of mimicking antibodies is easily masked by their panreactivity and may thus be ignored in routine antibody identification for blood bank serology. Mimicking antibodies are difficult to identify in the clinic. Most mimicking antibodies are autoantibodies, but they can be present regardless of the presence of the relevant antigen in the patient’s RBCs.^5,7 If a relevant antigen is negative, mimicking antibodies may be mistaken for alloantibody co-existing with an autoantibody. If a mimicking antibody is suspected, more tests are therefore required to assess its serological and clinical significance.

In the event of mimicking antibodies in the patient’s plasma, complete crossmatching cannot be achieved because of panreactivity. In clinical practice, an antigen-avoidance transfusion strategy is adopted to transfuse the corresponding antigen-negative RBCs, to achieve maximum blood transfusion compatibility, which is also applicable to patients with mimicking antibodies.^2,8 In our case, transfusion of matched ABO and Rh serological phenotype blood showed a poor transfusion effect, and Rh(e)-negative donor RBCs were eventually used for clinical transfusion. Type O ccDEE RBCs were then transfused and the patient’s hemoglobin level increased, with no evidence of hemolytic transfusion reaction. Homotypic blood transfusion should thus be carefully considered in patients with mimicking antibodies, and the antigen-avoidance transfusion strategy applies to such patients. However, Rh non-homotypic transfusion increases the risk of alloimmune responses. Attention should be paid to the timely monitoring of the post-transfusion serological pattern, i.e., antibody identification is recommended after transfusion, and changes in antibody response patterns and intensity should be checked to prevent missing the detection of alloantibodies.