Original Article

Access this article online

Quick Response Code:
Autoantibodies mimicking
alloantibodies: A case series unveiling
the dilemmas of transfusion
Soma Agrawal, Mohit Chowdhry, Shiva Prasad Gajullupalli, Muthukumaravel
Website:
www.ajts.org

DOI:
Abstract:
10.4103/ajts.ajts_161_20 INTRODUCTION: Autoimmune hemolytic anemia is characterized by increased red cell destruction and/
or decreased red cell survival due to autoantibodies directed against self‑antigens on red cells. Since
autoantibodies react with self and nonself red blood cells (RBCs), they tend to mask the underlying
clinically significant alloantibodies and many a times mimic a specific pattern like alloantibodies.
MATERIALS AND METHODS: We discuss three immune hematological cases of warm autoantibodies.
Antibody screening was performed by solid‑phase red cell adherence (SPRCA) technique on a fully
automated platform NEO Iris (Immucor Inc., USA). In case of a positive antibody screen, antibody
identification was performed using SPRCA, NEO Iris (Immucor Inc., USA). Alloadsorption for adsorbing
the autoantibodies was done using in‑house prepared allogenic packed RBCs – R1R1, R2R2, and rr.
RESULTS: All cases had warm autoantibody with a broad specificity against self‑Rh antigens. Anti
“C” and Anti “e” antibodies were identified in case 1 and autoanti “e” antibody in cases 2 and 3. Case
3 had underlying alloanti “E” along with autoanti “e” which posed a transfusion challenge.
CONCLUSION: Our case series highlights the importance of detecting the nature of the antibody
whether it is alloantibody or autoantibody with antigen specificity. This would help in selecting
appropriate antigen negative blood units for transfusion purpose.
Keywords:
Adsorption, autoantibody, autoimmune hemolytic anemia, Rh antibodies

Introduction the underlying clinically significant

alloantibodies and many a times mimic a

A utoimmune hemolytic anemia (AIHA)

is characterized by increased red cell
destruction and/or decreased red cell
specific pattern like alloantibodies. There
are very few studies reporting the frequency
and specificity of the autoantibodies with
survival due to autoantibodies directed mimicking specificity.[3] We, hereby, discuss
against self‑antigens on red cells.[1] The three such cases in our setting.
incidence of the disease has been reported
to vary from 1 in 80,000 to 100,000/year.[2] Materials and Methods
Department of Transfusion AIHAs are divided into warm, cold, or mixed
Medicine, Indraprastha autoantibody. [1] Warm autoantibodies The cases were identified and evaluated in
Apollo Hospitals,
are more reactive at 37°C, whereas cold the department of transfusion medicine in
New Delhi, India
autoantibodies react best at 0–5°C.[3] a tertiary care center in India. As a protocol,
Address for all patient samples were subjected for “Type
correspondence: Since autoantibodies react with all self and Screen” policy at our center. Blood
Dr. Soma Agrawal, and nonself RBCs, they tend to mask grouping was performed on fully automated
Department of Transfusion
Medicine, Indraprastha immunohematology system solid‑phase red
Apollo Hospitals, This is an open access journal, and articles are cell adherence (SPRCA) technology, NEO
New Delhi ‑ 110 076, India. distributed under the terms of the Creative Commons Iris (Immucor Inc., Norcross, GA, USA) using
E‑mail: doctorsoma86@ Attribution‑NonCommercial‑ShareAlike 4.0 License, which
gmail.com allows others to remix, tweak, and build upon the work
non‑commercially, as long as appropriate credit is given and How to cite this article: Agrawal S, Chowdhry M,
Submitted: 06‑10‑2020 the new creations are licensed under the identical terms. Gajullupalli SP, Muthukumaravel. Autoantibodies mimicking
Revised: 04‑07‑2022
alloantibodies: A case series unveiling the dilemmas of
Accepted: 31‑07‑2022 transfusion. Asian J Transfus Sci 2023;17:58-62.
Published: 12-12-2022 Forreprintscontact:WKHLRPMedknow_reprints@wolterskluwer.com

58 © 2022 Asian Journal of Transfusion Science | Published by Wolters Kluwer - Medknow

 Agrawal, et al.: Autoantibodies mimicking alloantibodies

commercially available antisera (Immucor Inc., Norcross, Results

GA, USA). Antibody screening was performed on NEO
Iris using 4‑cell commercial panel (Capture‑R ready Case 1
screen, Immucor Inc., Norcross, GA, USA). Any screen An 11‑year‑old girl was admitted to our center with
positive was further evaluated for antibody identification. a 1½‑month history of jaundice in September 2019.
RBC antibody identification was done with a commercial She had a history of urinary tract infection at the age
16‑cell identification panel (Ready‑Id, NEO Iris, Immucor, of 6 months. During admission, she informed of no
USA) and/or with 11‑cell panel (ID‑DiaPanel, Biorad, history of transfusion. The patient was hemodynamically
Switzerland) in indirect antiglobulin test (IAT) phase. stable. The complete blood count revealed hemoglobin:
A 10‑cell commercial panel by the conventional tube 7.4 g/dL, total leukocyte count: 11.3 × 10 3/L, and
technique (Immucor, USA) was also used wherever platelet count: 403 × 109/L. RBC indices were as follows:
necessary. Direct antiglobulin test (DAT) and autocontrol mean corpuscular volume 113.2 fL, mean corpuscular
were done by column agglutination technology (CAT) hemoglobin (MCH) 36.2 pg, and MCH concentration
using polyspecific (immunoglobulin G [IgG] and C3d) 32.0 g/dL. The corrected reticulocyte count was
Coombs’ reagent (Biorad, Switzerland) and on SPRCA 12% (range: 0.5%–2.5%). Other pertinent investigations
using monoclonal IgG (NEO Iris, Immucor, USA). Red were as follows: total bilirubin was 6 mg/dL and direct
cell antigen phenotyping for “C,” “c,” “E,” “e,” and fraction was 0.5 mg/dL, normal complement C3 level
“K” was done on fully automated immunohematology 101.9 mg/dL (normal range: 90–180 mg/dL), elevated
system SPRCA technology, NEO Iris (Immucor Inc., lactate dehydrogenase 935 U/L (normal range: up to
Norcross, GA, USA) using commercially available 225 U/L), creatinine 0.36 mg/dL (range: 0.3–0.7 mg/dL),
antisera (Immucor Inc., Norcross, GA, USA). As per the and albumin 4.34 mg/dL (range: 3.4–5.2 mg/dL).
manufacturer’s instruction, these antisera are low protein
monoclonal blends manufactured from a blend of IgM Anticipating a need of transfusion, we received sample
antibodies of their respective specificities and they do for blood grouping, antibody screening, and DAT.
not enhance agglutination of Ig‑coated red blood cells On visual inspection, the plasma revealed evidence
and hence can be used to type DAT positive red cells. of hemolysis. Blood group of the patient was O Rh
Alloadsorption for adsorbing the autoantibodies was D positive. DAT was positive (4+) with both the
performed using in‑house prepared allogenic packed techniques (CAT and SPRCA).
RBCs (DCe/DCe (R1R1), DcE/DcE (R2R2), and dce/
dce (rr)). Allogenic cell panels (R1R1, R2R2, and rr) are RBC antibody screening was pan positive (3+) on SPRCA
regularly prepared as per the recommendations of the with a positive autocontrol. RBC antibody identification
American Association of Blood Banks.[4] Alloadsorption on 16‑cell identification panel in IAT phase was pan
of the serum was done based on the technique described positive (3+).
by Issitt et al.[5] An equal volume of patient’s plasma and
packed RBCs R1R1, R2R2, and rr RBCs were mixed and Her Rh antigen phenotyping was D+, C+, c+, E−, e+,
incubated at 37°C for 1 h. Usually, a total of 3 adsorptions and K−. Autoadsorption was not attempted owing to the
were done. After the third adsorption, the adsorbed limited patient sample volume. Alloadsorption was done
plasma was tested for the presence of antibody using using in‑house prepared allogenic packed RBCs R1R1,
a commercial 3‑cell panel (ID‑DiaCell I‑II‑III, Biorad, R2R2, and rr. After the third adsorption, the adsorbed
Switzerland) along with the necessary controls. If the plasma was tested for the presence of antibody using a
screen showed a positive reaction, an extended antigen commercial 3‑cell panel. The 3‑cell panel was negative
panel of 11 cells (ID‑DiaPanel, Biorad, Switzerland) was with R1R1 and was positive with R2R2 and rr, indicating
used to identify the antibody specificity. Both antibody the possible presence of anti‑C and anti‑e antibodies.
screening and antibody identification on 3‑cell and Adsorbed plasma of R2R2 and rr was tested with 10‑cell
11‑cell panel, respectively, were done by CAT using commercial panel by the conventional tube technique,
polyspecific (IgG and C3d) Coombs’ reagent (Biorad, which showed pattern suggestive of anti‑C and anti‑e.
Switzerland). The patient’s phenotype (positive for “C” antigen and
“e” antigen) and no history of transfusion in the presence
Cross‑matches between patient’s serum and packed of positive autocontrol further indicated the presence of
red cell (PRC) units were performed between patient’s autoantibody. The antibodies reacted only at 37°C and
adsorbed plasma and PRC units by CAT using in AHG phases. Thus, the overall workup confirmed
polyspecific Coombs’ reagent. the presence of warm autoantibody mimicking anti‑C
and anti‑e.
Informed consent was taken to authorize the use of
patient’s information who were included in this case R1R1, R2R2, and rr PRC units were cross‑matched with
series. the patient serum and adsorbed plasma. Adsorbed
Asian Journal of Transfusion Science - Volume 17, Issue 1, January-June 2023 59
 Agrawal, et al.: Autoantibodies mimicking alloantibodies

plasma with R2R2 and rr was found compatible with Autoadsorption was not attempted owing to the limited
PRC units lacking C antigen and e antigen (i.e., R2R2 patient sample volume. Alloadsorptions of the plasma
phenotype). However, the patient was managed were done. Antibody screening with adsorbed plasma
conservatively and transfusion was not needed. was positive. After crossing out all other antigens,
adsorbed plasma from R1R1 and rr cells revealed a
Case 2 pattern confirming the presence of alloantibody against
A 51‑year‑old female was admitted elsewhere with “E” antigen when tested with 11‑cell panel. However,
stroke in the later part of October 2019. Her blood the R2R2 adsorbed plasma showed a pattern suggesting
samples were sent to our department for blood grouping the presence of autoantibody with “anti‑e” specificity
and antibody screening, anticipating blood transfusion. on 11‑cell panel. The antibody reacted only at 37°C and
She was never been transfused previously. She had two in AHG phase.
living children (P2L2). No other lab reports or patient
details were available to us. The baseline hemoglobin level was 6.7 g/dL. He
received 2 units of “E” negative PRC and the
She was typed as group “O” having Rh and Kell posttransfusion hemoglobin was 7.2 g/dL. He was
phenotype of D+, C+, c+, E−, e+, and K−. RBC antibody given dexamethasone and rituximab 600 mg after the
screening was pan positive (3+) in IAT phase on SPRCA PRC transfusion. He was subsequently transfused with
with a positive autocontrol. DAT was positive (3+) with another unit of “E” negative PRC and was discharged
both (CAT and SPRCA) the techniques. with a hemoglobin of 7.8 g/dL. He was advised to
continue the steroid treatment in a tapering dose for
RBC antibody identification was done with a commercial the next 4 days in addition to his regular medications
16‑cell identification panel in IAT phase and was for his underlying disease conditions. All the three
positive with varying strengths (1+, 2+, and 3+) with transfusions were uneventful. Thereafter, he was
all the e + cells except for one which was negative for advised to get admitted after 3 weeks for the second
e antigen, suggesting the presence of autoantibody dose of rituximab.
with anti‑e specificity. To reconfirm, 11‑cell panel on
CAT was also tested with the sample, which revealed Discussion
a clear pattern for the presence of autoantibody with
anti‑e specificity. The antibodies reacted only at 37°C AIHA is a rare clinical disorder characterized by
and in AHG phases. Alloadsorption was performed shortening of red cell survival due to of the presence of
as described previously.[5] The antibody screening was autoimmune antibodies and requires efficient immune
negative with adsorbed plasma on 3‑cell panel on CAT. hematological support.[6] Transfusing AIHA patient is a
The PRC units with R2R2 phenotype were compatible challenge to the immune hematologist.
after AHG cross‑match using the patient’s neat and
adsorbed plasma. Thus, the overall workup confirmed Warm autoantibodies react with self‑antigen optimally
the presence of autoantibody against “e” antigen. The at 37°C and are present in serum of about 80% of
patient was managed conservatively and did not require patients with warm AIHA.[7] In AIHA, most of the
any transfusion. warm antibodies are pan‑agglutinins in nature
and lack any apparent specificity (based on weak/
Case 3 negative reaction with Rh null red cells).[4] However,
A 67‑year‑old male was admitted at our tertiary care it has been observed that a significant proportion
center in November 2019 for chemotherapy for B‑cell of these antibodies are directed against antigens of
chronic lymphocytic leukemia. He was a known Rh system.[8,9] Further, it was established that absent
hypertensive with coronary artery disease and Type II or weak reactivity of autoantibodies with Rh null
diabetes mellitus. His blood samples were sent to our red cells does not indicate specificity to Rh complex
department for blood grouping, antibody screening, because Rh null cells might have other membrane
and identification anticipating future blood transfusion. abnormalities. [6] Race and Sanger in 1954 were
the first to report autoantibodies having a clearly
He had received one unit of PRC a year ago. He was defined Rh specificity, for example, anti “e”. [6,10]
typed Group “O” having Rh and Kell phenotype of D+, The autoantibodies having an apparent and relative
C+, c+, E−, e+, and K−. DAT by CAT and SPRCA was specificity for a single antigen (e.g., anti “e”) along
positive (4+). RBC antibody screening was pan positive with an evidence of hemolysis required compatible
with variable strengths (3 + and 4+) on SPRCA with a antigen negative unit for transfusion.[6]
positive autocontrol (4+). RBC antibody identification
in IAT phase was positive with varying strengths (2+, In the first case, we found autoantibodies mimicking
3+, and 4+). against two Rh antigen specificities, i.e., anti‑“C” and
60 Asian Journal of Transfusion Science - Volume 17, Issue 1, January-June 2023
 Agrawal, et al.: Autoantibodies mimicking alloantibodies

anti‑“e.” The prevalence of “C” and “e” antigens is very autoantibodies were completely adsorbed by e‑positive
high in Indian population as reported previously. In an and e‑negative cells equally.
erstwhile study at our center, the Indian donors (n = 3073)
were typed and prevalence of “C” and “e” antigens Blood transfusion for patients with AIHA presents
was found to be 87% and 98% respectively.[11] After a unique set of potential problems because of the
extensive literature search, we did not come across any masking effect of the presence of RBC alloantibodies
autoantibodies with specificity against both anti‑“e” and by autoantibodies, the need for complex pretransfusion
anti‑“C”. In a patient diagnosed with primary sclerosing immune hematological workup, and the challenge to
cholangitis associated with AIHA, autoantibody with find a compatible unit.[15]
anti‑C specificity was identified.[12] The said patient
was transfused with two units negative for C antigen Selecting the red cell unit for transfusion was highly
which were crossmatch compatible. Issitt et al. reported challenging for the third case due to the presence of
specificity of autoantibodies among 87 patients with alloantibody against ‘E’ antigen with concomitant
AIHA warm autoantibodies; only 4 cases were reported presence of autoantibody of anti‑e specificity. This
to have anti‑“e” or anti‑“c” antibodies.[5] posed a dilemma in selecting antigen negative unit
for the patient. In case series report by Yürek et al.,
Procuring “e” and “C” negative unit (R2R2) usually is a there were no cases that definitively demonstrated
difficult task in a country like ours wherein the resources significant exacerbation of hemolysis in patients
are not widely available. The importance becomes high with true AIHA.[16] Since our patient had transfusion
in situations with time constraints. The risk–benefit requirement, appropriate “E” negative PRC units were
ratio needs to be weighed when comparing the time transfused. All the three transfusions went uneventful.
required to search for such antigen negative units and The patient was also started on corticosteroids and
time the patient can sustain without transfusion support. rituximab medication. More recently, rituximab is being
Anticipating such situations and many others, our center used as a second‑line therapeutic approach following
has a set protocol to perform the extended Rh and Kell corticosteroid therapy, instead of the traditional
phenotyping of all the blood donors as a routine to find secondary approach of splenectomy. [15] Patients
compatible antigen negative units during immediate/ demonstrated higher initial clinical response rates and
urgent requirements. relapse‑free survival over a 3‑year time period, when
compared to glucocorticoid therapy alone.[15] Our patient
Although we detected the autoantibody specificity was discharged with a stable hemoglobin of 7.8 g/dL
against Rh, the presence of anti‑Ce (anti‑RH7) could and was advised for admission after 3 weeks for the
not be ruled out. Ce (RH7) is considered to be a cis gene second dose of rituximab.
product. This compound antigen can form antibody
and few reports of this alloantibody causing hemolytic Rh genotyping facilities were not available at our center
disease of the new born and delayed type hemolytic and therefore could not be performed in any of the cases.
transfusion reactions have been mentioned.[13]
Conclusion
The second case which came was for the presence of
autoantibody with anti‑e specificity. A similar case The autoantibodies present in a patient may seldom
was discussed by Pahuja and Verma, in a 2½‑year‑old mimic a specific antibody. It is worthwhile to perform
boy diagnosed of AIHA with anti “e” specificity.[6] In advanced immune‑hematological workup to determine
most patients with warm AIHA, RBC autoantibodies the exact specificity of this antibody. This would not
react with all RBCs (pan‑reactive). Infrequently, these only help in providing antigen negative blood to the
autoantibodies do have apparent specificity (patient’s patient but also help to determine the exact nature of
RBCs may or may not contain the antigen) which the antibody.
disappears following adsorption with antigen positive
or negative cells.[4] Subramaniyan and Veerasamy also Financial support and sponsorship
explained a case where in they had an autoantibody Nil.
mimicking anti‑C specificity.[14] After adsorption with
R1R1 and rr RBCs, the adsorbed plasma was tested for Conflicts of interest
the presence of anti‑C using an RBC antibody screening There are no conflicts of interest.
panel. Antibodies were completely adsorbed from the
plasma with C‑negative as well as C‑positive cells.[14] This References
was similar to the case in discussion, where adsorption
with R1R1, R2R2, and rr RBCs revealed complete 1. Gehrs BC, Friedberg RC. Autoimmune hemolytic anemia. Am J
adsorption of anti‑e antibody, suggesting that the Hematol 2002;69:258‑71.

Asian Journal of Transfusion Science - Volume 17, Issue 1, January-June 2023 61

 Agrawal, et al.: Autoantibodies mimicking alloantibodies

2. Duffy TP. Autoimmune hemolytic anemia and paroxysmal Med Inst Mex Seguro Soc 2005;43:107‑11.
nocturnal hemoglobinuria. In: Simon TL, Dzik WH, Synder EL, 10. Race RR, Sanger R. Blood Groups in Man. 2nd ed. Oxford: Blackwell
Stowell CP, Strauss RG, editors. Rossi's Principles of Transfusion Scientific Publications; 1954. p. 361.
Medicine. 3rd ed. Philadelphia, USA: Lippincott Williams and 11. Makroo RN, Bhatia A, Gupta R, Phillip J. Prevalence of Rh, Duffy,
Wilkins Publication; 2002. Kell, Kidd & MNSs blood group antigens in the Indian blood
3. Jang MJ, Cho D, Park KU, Yazer MH, Shin MG, Shin JH, et al. donor population. Indian J Med Res 2013;137:521‑6.
Autoantibodies with mimicking specificity detected by the 12. Bajpai M, Maheshwari A, Gupta S, Bihari C. Autoanti‑C in a
dilution technique in patients with warm autoantibodies. Ann patient with primary sclerosing cholangitis and autoimmune
Lab Med 2013;33:343‑8. hemolytic anemia: A rare presentation. Immunohematology
4. Fung MK. Technical Manual. 18 th ed. Bethesda: American 2016;32:104‑7.
Association of Blood Banks; 2014. p. 435. 13. Molthan L, Matulewicz TJ, Bansal‑Carver B, Benz EJ. An
5. Issitt PD, Combs MR, Bumgarner DJ, Allen J, Kirkland A, immediate hemolytic transfusion reaction due to anti‑C and a
Melroy‑Carawan H. Studies of antibodies in the sera of patients delayed transfusion reaction due to anti‑Ce+e: Hemoglobinemia,
who have made red cell autoantibodies. Transfusion 1996;36:481‑6. hemoglobinuria and transient impaired renalfunction. Vox Sang
6. Pahuja S, Verma D. Autoimmune hemolytic anemia caused by anti 1984;47:348‑53.
“e”: A challenge: A case report with review of literature. Asian J 14. Subramaniyan R, Veerasamy M. Red cell autoantibody mimicking
Transfus Sci 2017;11:195‑8. anti‑C specificity: A rare manifestation. Rev Bras Hematol
7. Petz LD, Garratty G. Acquired Immune Hemolytic Anemias. Hemoter 2017;39:91‑2.
New York: Churchill Livingstone; 1980. 15. Simon TL, McCullough J, Snyder EL, Solheim BG, Strauss RG.
8. Weiner W, Battey DA, Cleghorn TE, Marson FG, Meynell MJ. Rossi’s Principles of Transfusion Medicine. 5th ed. West Sussex,
Serological findings in a case of haemolytic anaemia, with some UK: Wiley & Blackwell; 2016. p. 148.
general observations on the pathogenesis of this syndrome. Br 16. Yürek S, Mayer B, Almahallawi M, Pruss A, Salama A. Precautions
Med J 1953;2:125‑8. surrounding blood transfusion in autoimmune haemolytic
9. Garratty G. What do you do when all units are incompatible? Rev anaemias are overestimated. Blood Transfus 2015;13:616‑21.

62 Asian Journal of Transfusion Science - Volume 17, Issue 1, January-June 2023