Anti tissue transglutaminase antibody in idiopathic autoimmune haemolytic anemia

Anti tissue transglutaminase antibody in idiopathic autoimmune haemolytic anemia

Transfusion and Apheresis Science xxx (xxxx) xxx–xxx Contents lists available at ScienceDirect Transfusion and Apheresis Science journal homepage: w...

420KB Sizes 0 Downloads 21 Views

Transfusion and Apheresis Science xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Transfusion and Apheresis Science journal homepage: www.elsevier.com/locate/transci

Anti tissue transglutaminase antibody in idiopathic autoimmune haemolytic anemia ⁎

Kanjaksha Ghosha,b,c, , Kinjalka Ghoshb, Reepa Agarwalc, Kiron Shaha,b,c, Kanchan Mishraa,b,c a

Surat Raktadan Kendra & Research Centre, Udhna Magdalla Road, Nr. Chosath Joganio Mata Temple, Surat 395002, Gujarat, India Department of Clinical Biochemistry, Tata Medical Centre, Parel, Mumbai 400012, India c Jerbai Wadia Children Hospital, Parel, Mumbai 400012, India b

A R T I C LE I N FO

A B S T R A C T

Keywords: Autoimmune haemolytic anaemia Anti tissue trans glutaminase antibodyautoantibody Molecular mimcry with band 4.2 protein Redcell membrane ghost Western blot —autoantibody

Background: In idiopathic autoimmune haemolytic anaemia (AIHA haemolytic antibodies are directed to every type of red cellsWestern blot studies have shown antibody positivity towards red cell anion channel complex which also includes band 4.2 a protein with similarities to tissue trans glutaminase. Objective: Evaluation of AIHA for anti tissue transglutaminase antibody (Anti tTG). Materials & methods: Twenty three AIHA patients were tested along with routine hamatogical work up, for a series of auto antibodies and red cell eluates and serum from the patents were tested against solubilised group O red cell ghosts on western blot. Other ancillary investigations were done to rule out complications and secondary causes of haemolysis. Results: 11/23 patients (48%) were positive for anti tTG, Four, 3 and 8,7 patients were positive for anti thyroid, anti b2 glycoprotein, lupus anticoagulant and ANA respectively. One patient with anti tTG had biopsy proven celiac disease. Three patient developed DVT and all of them were lupus anticoagulant as well as b2 gp-1 antibody positive.17 had become Coombs test negative on treatment while 21/23 had positive western blot test. Discussion & conclusion: There is strong association of anti tTG antibody with idiopathic AIHA. Aetiological association of this finding needs exploration.

1. Introduction Idiopathic autoimmune haemolytic anemia is a relatively rare condition affecting 50–60 % of 1–3 patients / 100,000 populations [1] affected by AIHA. Hence this is a relatively rare disease. Many organ specific and non specific autoimmune diseases may have associated autoimmune haemolytic anemia either at the time of presentation or later during the evolution of the disease. Autoimmune thyroid disorder, SLE, primary biliary cirrhosis, reactive colitis, cryoglobulinaemias iral infe, lymphoid malignancies and other malignancies are some of the autoimmune and other disorder associated with autoimmune haemolytic anemia [2]. Celiac disease where autoantibodies against tissue transglutaminase or endomysium are demonstrated are rarely associated with Immune haemolytic anemia and only has been described as occasional case report (3) immune haemolytic anaemia along with immune thrombocytopenia ie Evan’s syndrome has also been reported in this condition but immune thrombocytopenia, type 1 diabetes and thyroid disorders are more commonly reported with celiac disease [3,4]. Though idiopathic autoimmune haemolytic anemia has been



associatd with several autoantibodies, without manifestation of disease in the target organs [5], there is very meagre data whether antitissue transglutaminase (anti tTG) antibody is associated with Idiopathic autoimmune haemolytic anemia. In the present study we followed up a cohort of patients with idiopathic autoimmune haemolytic anemia without any manifestation of other organ specific or non specific autoimmune disease to describe its possible association with anti tTG antibodies without any evidence of coeliac disease in the majority of the patients. 2. Material and methods 2.1. Patients All the patients (age 20 – 56yrs. M : F 5 : 18) presenting with clinical features of auto immune haemolytic anemia(AIHA) as evidenced by fatigue, pallor, breathlessness on exertion, jaundice with laboratory evidence of immune haemolytic anemia, reticulocytosis, splenomegaly, dark urine etc with strong direct Coombs test positivity, anemia,

Corresponding author at: National Institute of Immunohaematology, 13 h Fl, KEM Hospital Campus, Parel, Mumbai 400012, India. E-mail address: [email protected] (K. Ghosh).

https://doi.org/10.1016/j.transci.2019.06.034 Received 9 March 2019; Received in revised form 9 June 2019; Accepted 12 June 2019 1473-0502/ © 2019 Elsevier Ltd. All rights reserved.

Please cite this article as: Kanjaksha Ghosh, et al., Transfusion and Apheresis Science, https://doi.org/10.1016/j.transci.2019.06.034

Transfusion and Apheresis Science xxx (xxxx) xxx–xxx

K. Ghosh, et al.

Table 1 Spectrum of autoantibodies in patients with AIHA.

Patient (23) Normal (100) λ2 test

Anti Thyroid (%)

Anti Nuclear (%)

Anti Parietal cell (%)

Anti cardiolipin (%)

Anti β2 gp (%)

Lupus Anticoagulant (%)

Anti ttg (IgA)

4/23 (48)++ 1/100 (1) P = 0.002

7 (32) 1(1) P < 0.001

0 0 –

0 1(1) –

3(13.6) 0 P=0.003

8(32) 1(1) P < 0.001

11(18) 0(2) P < 0.001

One way ANOVA p = 0.006. ttg : tissue transglutaminase. DAT (Direct Antiglobin Test) was positive in all patients during presentation but was positive in 6/23 at the time of study. *Platelets neutralization test = 2 positive (both were β glycoprotein 1 ab positive) & had DVT (2) portal vein thrombosis (1) 6/23 Patients splenectomized. 1/23 had HLH and remitted on Rituximab + HLH protocol. 1/24 had celiac disease, growth retardation with positive intestinal biopsy and was excluded from study. Follow up period 6 months to 25 years (Median 3.2 years, Mean 3.8 yr.).

dl. All of them were started on prednisolone 1 mg/kg and after 3–4 weeks it was slowly tapered off with the objective of maintaining it less than 7.5 mg of prednisolone per dayor shifting to an alternate day schedule where feasible if it was not possible to wean off the steroid all together. Those who required prednisolone more than 10 mg /day after 12 weeks were put on additional azathioprine 2.5–3 mg/kg /day and those who still did not respond was advised splenectomy. Mostly the management of the cases followed as has been recently reviewed [1]. Patients were followed up initially every 2–4 weeks then once in 3 months. Splenectomized patients received penicillin prophylaxis as per norms and all were vaccinated against pneumococcus and haemophilus influezae at least 3 weeks prior to splenectomy. One patient who developed haemophagocytic syndrome was later treated successfully with rituximab and 1991 HLH protocol.

indirect hyperbilirubinaemia, high serum LDH level. They were evaluated for secondary causes of AIHA like drugs, lymphoma, viral infections and other established autoimmune diseases etc. Patients with identifiable cause of immune haemolytic anemia, were excluded from the study. This cohort of autoimmune haemolytic anemia was followed up and treated by one of the authors (KS) over the years. The study was sanctioned by institutional ethics committee and All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008. All patients and volunteers consented to the study. Separate consents were taken in patients undergoing small intestinal biopsy. 2.2. Investigations In addition to investigation of anemia, autoimmune nature of haemolysis (various types of antiglobulin tests). The patients were regularly investigated for renal function, liver function, thyroid function, had USG and other imaging as required and all had serological investigations for auto immune disorder i.e. anti nuclear antibody, Anti ds-DNA antibody, antiparietal cell antibody, Anti thyroid antibodies (against thyroglobulin and microsome), anticardilipin antibody, Anti β2 glycoprotein -1 antibody, Serum anti tTG IgA antibody were all measured at varying time period by ELISA assay after autoimmune haemolytic anemia was diagnosed .100 healthy volunteers were also tested side by side for normal reference ranges. Lupus anticoagulant was studied by using KCT/Russel viper venom neutralization test and was confirmed by platelet neutralisation assay. Red Cell ghost from group O, RhD negative cellswere prepared solubulised deploy merised with mercaptoethanol(Sgma) and were testedusing western blot technique with both RBC eluates from the patients or 1:10 dilution of patient sera [6]. Attempts were made to get upper small intestinal biopsy (duodenal) toevaluate for celiac disease in these patients where serum anti ttg IgA were positive. Before the intestinal biopsy was done history was taken specifically on celiac disuse as advised (NICE criteria)i.e. (1) persistent unexplained gastrointestinal symptoms such as feeling sick and being sick (ii) Faltered growth (iii) unexplained weight loss (v) unexplained anaemia (vi) Juvenile diabetes (vii) Thyroid disorder (viii) Fatigue. Intestinal biopsy specimens were stained by standard H & E & reticulin stain and was examined by an experienced histopathologist Absorption of D – Xylose was tested using standard protocol in all the patients who had anti transglutaminase antibody in the serum.

3. Results The 23 patients (additional one with known celiac disease was not analysed)with warm antibody idiopathic autoimmune haemolytic anemia were followed up between 6 month – 25 year with median follow up of 3.2 years and mean follow up 3.8 years 20/23 patients needed packed red cell transfusion, 6/22 patients needed splenectomy after failure of steroid and azathioprine therapy for at least 6 months. 4/11 patients initially presented with thrombocytopenia and 1 patient developed haemophagocytic syndrome (Pancytopenia, serum ferritin 20,000 μg/ml). During work up of these patients, a battery of autoantibodies was screened and the results are presented in Table 1. In short all were positive DAT at the time of presentation but 17/23 patient became DAT negative during the study period. 8/23 patients were positive for lupus anticoagulant but 3/8 of these were also strongly positive for β2 glycoprotein 1 antibody also and all of them showed positive platelet neutralisation test. Two of these three patients went on to develop deep venous thrombosis of the left leg and one developed portal venous thrombosis. All the three patients are now on oral anticoagulation. 7/23 were positive for ANA, however none of them were positive for dsDNA antibody and 4/23 were positive for anti thyroid antibody but none developed clinical or biochemical evidence of thyroid disease or SLE. 11/23 (48%) patients were positive for anti tissue transglutaminase actively compared to 2/100 in healthy age matched control. All the anti transglutaminase antibody positive patients were asymptomatic for the Coeliac disease related symptoms and had normal D-xylose excretion test except one who had coeliac disease (This patient was not part of the analysis). It was possible to do small intestinal biopsy only in 2 patients who were tissue transglutaminase IgA positive in the serum. One of the biopsy was diagnostic of coeliac disease but other biopsy was normal. Other patients did not agree for small intestinal biopsy. Except increase in serum bilirubin (Unconjugated 2.5–6 mg/dL) at the time of presentation and increase in LDH and SGOT (40–170 IV/L) all other liver and renal function tests were normal. Western blot was done on all the patient sera using group O RhD negative red cell ghost. 21/23 patient showed a band at 198 kd, 74, 41 and 27 K.D molecular weight (Fig. 1).

2.3. Statistics Chisquare test and one way ANOVA was used to detect the significance of the findings. 2.4. Management At the time of presentation symptomatic patients received least incompatible red cell transfusion to maintain haemoblobin above 6 gm/ 2

Transfusion and Apheresis Science xxx (xxxx) xxx–xxx

K. Ghosh, et al.

Fig. 1. Western blot analysis of ghost RBCs.

of this antibody in autoimmune haemolytic anemia patients. Though the reverse i.e. autoimmune haemolytic anemia in a patient with coeliac disease has been reported [3,4]. We could do histological examination of duodenal biopsy only on 2 patients. One of the patients who were already diagnosed with celiac disease, excluded from analysis, has histological features of coeliac disease and the other had normal biopsy. None of the serum transglutaminase antibody positive patient except the one with celiac disease has abnormal D-xylose absorption. Anti transglutaminase antibody positivity in our cases were associated with positivity of antibody to red cell membrane associated antigen of around 75Kd molecular weight (Fig. 1) and this could represent an antibody against a red cell membrane antigen with structural similarity to tissue transglutaminase. It is a curious fact that red cell membrane does have a protein in the form of band 4.2 with similar function [9]. A patient with similar finding presenting with AIHA has already been reported in the literature [7]. Authors of that paper speculated that haemolysis could be related to development of such a cross reacting antibody. This speculation lends some credence with the present findings. Band 4.2 is an important red cell membrane protein in association with band 3 protein and ankyrin. In warm antibody haemolytic anaemia IgG against this complex has been identified [10]. Absence or defective function of this protein can cause a recessive form of hereditary spherocytosis. Hence an antibody which binds to critical area of such a protein should be capable of causing haemolytic anaemia. The ELISA test which was used for anti transglutaminase antibody test in the present study contained recombinant antigen hence chances of false positive due to other contaminating antigen is unlikely [11]. Is this a false positive test or it has biological significance associated with autoimmune haemolytic anaemia?? This could not be definitely answered from the present study.We could not test for IgG anti tTG in our cases, if we could then there was likelihood that few more cases with positive serology would have been picked up. However we

However all patients with positive anti tTG were positive on western blot test.and all of them had a band at 74Kd level, 4. Discussion Fifty percent of warm antibody immuno aemolytic anemias are idiopathic in nature [1]. At the time of study some of these patients showed serological markers of both organ specific (e.g. Thyroid, Parietal cell) and or non organ specific disease serology [5]. On long follow up subsets of these patients may have clinical features of organ specific autoimmune disorders like hypo/hyperthyroidism, non organ specific autoimmune disorder like SLE and lymphoproliferative disorders. None of our patients developed any other organ related disorder. The most interesting findings of our study is the presence of a large number of patients with strongly positive anti tTG serology. However primary auto immune haemolytic anemia are not known to present with anti tissue transglutaminase IgA antibody or rarely present with such an antibody [7]. On the other hand patients with known celiac disease rarely present with immune haemolytic anemia (3,4) though they present with many other auto immune phenomenon most notably thyroid disorder and type 1 diabetes mellitus. There are several serological markers for celiac disease, anti tissue transglutaminase (anti tTG IgA) IgA, Anti endomysium antibody and antigliadin / Antideamidated gliadin antibody. Of all these antibodies anti tissue transglutaminase IgA was found to be robust and in the absence of IgA deficiency the test has a very good sensitivity and specificity for the diseases [8]. We evaluated serologically a series of autoimmune markers in a cohort of Idiopathic auto immune haemolytic anemia and the present submission is an account and analysis of that study. The most striking feature of present study was presence of anti transglutaminase antibody in a large number of patients (48%) without clinical features of celiac disease. There is paucity of data in the literature on assessment 3

Transfusion and Apheresis Science xxx (xxxx) xxx–xxx

K. Ghosh, et al.

Acknowledgement

don’t know whether this specific interaction of antibody with red cell ghost antigen can cause of haemolysis. Another weakness of our study is the retrospective nature of the study as a result we were not able to test the sera at the time of presentation of many of the cases hence we cannot say with certainly whether anti transglutaminase antibody was present in many more cases of auto immune haemolytic anemia at presentation. Our follow up of the cases are reasonably long and it seems that most of anti tTG antibody positive patient will never develop celiac disease and they really did not develop any other primary autoimmune disorder (except one HLH and another celiac disease) during the period of follow up showing that most of our cases are idiopathic AIHA. Though only 17/23 patients became DAT negative at the time of testing yet western blot was positive for 21/23 patients which included all the 11 patients with anti tTG antibody. Previous study showed that western blot test is more sensitive than positive Coombs test for auto immune haemolytic anemia and many remain positive long after DAT has become negative [6]. 3/23(13%) patient in the present series developed thrombotic episodes and all were positive both for lupus anticoagulant and for anti b2 glycoprotein -1 antibody. This finding is in line with the observation that autoimmune haemolytic anaemia patients are at increased risk of venous thrombosis [12]. Anti tTG antibody needs further evaluation as a causative antibody and a possible biomarker at least for a subset of patient with idiopathic warm antibody autoimmune haemolytic anaemia.

A portion of the paper was chronicled in American Society of Haematology meeting of Dec2016. Blood 2016 128:4807. References [1] Kalfa T. Warm antibody autoimmune hemolytic anaemia. Hematol Am Soc Hematol Educ Program 2016:690–7. [2] Liebman HA, Weitz IC, Liebman HA, Weitz IC. Autoimmune hemolytic anemia. Med Clin North Am 2017;101:351–9. [3] Miller DG. Coeliac disease with autoimmune haemolytic anaemia. Postgrad Med J 1984;60:629–30. [4] Roganovic J. Celiac disease with Evans syndrome and isolated immune thrombocytopenia in monozygotic twins: a rare association. Semin Hematol 2016;53(Suppl. 1):S61–3. [5] Blajchman MA. Tissue antibodies in idiopathic autoimmune haemolytic anaemia. Clin Exp Immunol 1971;8:741–8. [6] Bloch EM, Sakac D, Branch HA, Cserti-Gazdewich C, Pendergrast J, Pavenski K, et al. Western immunoblotting as a new tool for investigating direct antiglobulin test-negative autoimmune hemolytic anemias. Transfusion 2015;55:1529–37. [7] Ivanovski Petar, Nikolić Dimitrije, Dimitrijević Nikola, Ivanovski Ivan, Perišić Vojislav. Erythrocytic transglutaminase inhibition hemolysis at presentation of celiac disease. World J Gastroenterol 2010;28(16):5647–50. [8] Sblattero D, Berti I, Trevisiol C, Marzari R, Tommasini A, Bradbury A, et al. Human recombinant tissue transglutaminase ELISA: an innovative diagnostic assay for celiac disease. Am J Gastroenterol 2000;95:1253–7. [9] Lorand L, Graham RM. Transglutaminases: crosslinking enzymes with pleiotropic functions. Nat Rev Mol Cell Biol 2003;4:140–56. [10] Janvier D, Lam Y, Lopez I, Elakredar L, Bierling P. A major target for warm immunoglobulin G autoantibodies: the third external loop of Band 3. Transfusion 2013;53:1948–55. [11] Carroccio A, Giannitrapani L, Soresi M, Not T, Iacono G, Di Rosa C, et al. Guinea pig transglutaminase immunolinked assay does not predict coeliac disease in patients with chronic liver disease. Gut 2001;49:506–11. [12] Lecouffe-Desprets M, Néel A, Graveleau J, Leux C, Perrin F, Visomblain B, et al. Venous thromboembolism related to warm autoimmune hemolytic anemia: a casecontrol study. Autoimmun Rev 2015;14:1023–8.

Informed consent Informed consent was obtained from all patients for being included in the study.

4