Suez Canal University Medical Journal Vol. 19 (1), 2016 Pages 60-67

Autoimmune in Chronic Liver Disease Patients with Hepatitis C Virus at Suez Canal University Hospital

Nashaat M. Soliman1*, Fikry Gobran2, Fadia M. Attia2, Nashwa R. Elessawy2

Departments of 1Infectious and Endemic Diseases, 2Clinical , Faculty of Medicine, Suez Canal University, Ismailia, Egypt

Abstract

Background: Chronic hepatitis C (CHC) continues to be a public health problem in Egypt with data suggesting that its incidence may be increasing. Hepatitis C virus (HCV) infec- tion has been associated with a variety of extra-hepatic manifestations, including auto- immune disorders. Aim: To assess the prevalence of autoimmune neutropenia (AIN) in chronic liver disease (CLD) patients with HCV. Patients and Methods: A case-control de- sign was used to fulfill the study objective. It was conducted on 60 subjects, divided into 2 groups: group 1 (30 HCV-infected patients with low total leucocytic count [TLC] as a study group) and group 2 (30 HCV-infected patients with normal TLC as a control group). AIN was evaluated by granulocyte immuno-fluorescence test (GIFT) to detect auto-antibodies against neutrophils in HCV-infected patients presented with neutro- penia. Results: The mean age of the first group was 49.9±11.6 years, while the mean age of the second group was 52.7±11.3 years. Male gender was found in 53.3% of the low TLC group compared to 63.3% in normal TLC group; Positive GIFT was found in 13.3% of study group; while none of the control group has positive GIFT. The significant variables among patients with low TLC included low ANC (OR: 5.0, 95% CI: 1.5-16.5, low HB (OR: 7.0, 95% CI: 1.4-35.5), and low PLT count (OR: 15.5, 95% CI: 3.8-63.3). Conclusion: AIN is sig- nificantly higher among CHC patients with low TLC compared to normal TLC patients.

Keywords: low total leucocytic count, Granulocyte Immuno-fluorescence Test

Introduction among 15- to 59-year-old population(2). The viremic population of Egypt was HCV is a serious health prob- estimated at over 6 million(3). Approx- lem affecting over 130 million people imately 85% of the HCV-infected popu- worldwide (Ref). Infection with the lation develops severe complications, HCV affects 2-3% of the world’s popula- such as CHC, cirrhosis, diabetes, and tion(1). Egypt has the highest recorded (HCC). About prevalence of HCV in the world, reach- 10–30% of CHC patients develop liver ing 14.7% for HCV-antibody positivity cirrhosis. (4, 5) CHC is reported to be the

*Corresponding Author: [email protected]

61 AIN in CHC patients

major cause of 50-76% of all liver can- divided into two forms. In primary AIN, cers globally and also the reason for neutropenia is usually the sole hemato- two-third of all cases of liver transplan- logic abnormality and it is more com- tation. (6) Liver mortality in Egypt mon in children. Secondary AIN, which reaches 40,000 per year, making 10% of is more prevalent in adults, is associat- total mortality, and comes second af- ed with underlying autoimmune dis- ter heart diseases(7). Primarily, HCV in- eases, malignancies, , partic- fection appears to be confined to the ularly viral, neurological diseases or liver, however, a wide variety of extra- drug exposure(12). Diagnosis of AIN is hepatic disease manifestations have based on the presence of neutropenia been reported to be associated with and demonstration of anti-neutrophil the infection. Extra-hepatic manifesta- antibodies. The GIFT is a common tions are mainly due to associated au- method that can be used to test for toimmune disorders. A widely postu- antibodies against neutrophils in both lated, but controversial, hypothesis is the patient's serum and directly on au- that extra-hepatic disease manifesta- tologous cells(13). tions are caused by extra-hepatic tro- pisms of HCV, particularly lympho- Patients and Methods tropism. The latter is thought to be involved in the production of auto- This case-control study was carried antibodies(8). Persistent HCV infection out at Internal Medicine and Clinical is responsible for the production of a Pathology Departments, Suez Canal variety of auto-antibodies including University Hospitals, Ismailia, Egypt. It non-organ-specific auto-antibodies and included 60 CHC patients. They were organ-specific auto-antibodies, as a classified into two groups; group 1 of virus-induced autoimmune phenome- 30 CHC patients with leucopenia non. The diversity of auto-antibodies in which considered as the study group the sera of patients with HCV-related and group 2 of 30 CHC patients with CLD has been shown(9,10). Some auto- normal TLC which considered as the antibodies in CHC infection have bio- control group. Data were collected chemical, histological, or genetic char- from CHC patients using interview acteristics, while other auto-antibodies questionnaire. The recorded data in- may predict the response to anti-viral cluded; patients’ age, gender, Child- treatments, concomitant disorders, or Pugh score, and laboratory parame- prognosis in patients with HCV-related ters [hemoglobin (HB), platelets (PLT) CLD(11). AIN in adults are a heterogene- count, TLC, absolute neutrophil count ous group of diseases with clinical (ANC), total and direct bilirubin, Ala- manifestations varying from being nine transaminase (ALT), Aspartate asymptomatic to having infectious transaminase (AST), albumin, total complications with considerable mor- proteins, and prothrombin time (PT)]. bidity and mortality. They are charac- GIFT is an indirect immunefluore- terized by auto-antibodies directed scence assay for detection of granu- against neutrophils, resulting in de- locyte auto-antibodies in peripheral struction of neutrophils. AIN can be blood. Freshly air-dried peripheral Soliman NM et al. 62

blood slides were individually the initial result validated by a second wrapped in domestic aluminum foil or review. In cases where the results paraffin frozen within 2 hours of col- were not clear, a second individual lection, transported on dry ice and performed independent analysis of stored at -25ºC. Frozen slides were re- the slides. If the quality of the slide viewed within 2–3 weeks, but were was inadequate (because of sample stable for up to 3 months. A peripher- preparation, storage or transport), al blood sample from normal person these observations were recorded was frozen at the same time as each and reported as unsuitable for analy- sample and used as a negative con- sis. trol. Two patient slides and the nor- mal control slide were warmed to Results room temperature and fixed with 1% para-formaldehyde in phosphate The age ranged from 24-75 years. The buffered saline (PBS) for 5 min. Fixed mean age was 49.9±11.5 yrs versus neutrophils were incubated with the 52.7±11.3, in the study and the control 1:2 dilution of patient serum at 37ºC groups respectively. About 53.3% for 30 min. Known positive samples were males in the study group com- were analyzed with each sample pared to 63.3% in the control group. batch. Slides were washed with PBS The groups were matched fro age and and incubated with fluorescein gender without statistical significant isothiocyanate (FITC) conjugated to differences between them (p>0.05). human immunoglobulins IgG and in- There were insignificant differences cubated for further 30 min at room between the two groups regarding temperature in the dark. The dilution liver function tests and Child-Pugh of the fluorescein conjugated was de- scoring (p>0.05) (Table 1). Regarding termined by titration for each batch hematological parameters, there was of antibody (different dilutions of the statistically significant higher mean fluorescein conjugated were applied ANC, HB and PLT count in the control to peripheral blood control slides in- group compared to the study group cubated with known positive serum, (p<0.01) (Table 1). According to GIFT the dilution that gave the best fluo- results, the study group had signifi- rescence was selected). The slides cantly higher incidence of AIN com- were subsequently rinsed and cov- pared to the control group (13.3% vs. ered with a PBS–glycerol mountant. 0.0%, p=0.0002) (Table 2). The signifi- The slides were examined microscop- cant variables among patients with ically using epifluorescent ultraviolet low TLC included low ANC, low HB, illumination. A qualitative increase in and low PLT count. cytoplasmic fluorescence of patient neutrophils over controls indicated Discussion the presence of auto-antibodies(14). Positive and negative controls were HCV infection is a common cause of run with each batch. The whole slide progressive liver disease. In addition, was reviewed microscopically with chronic HCV infection has been associ- 63 AIN in CHC patients

ated with a variety of extra-hepatic vere neutropenia and a paucity of se- manifestations, including autoimmune vere bacterial infections. Anti- disorders. HCV may skew the immune neutrophil antibodies are well recog- response toward the production of nized causes of neutropenia, produc- auto-antibodies. AIN is a relatively be- ing both quantitative and qualitative nign disorder characterized by neutro- defects in neutrophils and increased phil specific auto-antibodies, often se- risk for infection(15).

Table 1: Demographic, clinical and laboratory characteristics of the studied populations Variables Study group Control group p-value (n=30) (n=30) Age (yrs) 49.90±11.56 52.70±11.30 0.347 Gender 0.432 Male 16 (53.3%) 19 (63.3%) Female 14 (46.7%) 11 (36.7%) Child-Pugh score 0.094 A 4 (13.3%) 0 (0.0%) B 17 (56.7%) 17 (56.7%) C 9 (30.0%) 13 (43.3%) T. bilirubin (mg/dL) 2.69±2.82 3.65±4.05 0.096 D. bilirubin (mg/dL) 1.25±0.64 1.68±0.623 0.589 ALT (U/L) 31.43±19.97 35.60±24.55 0.375 AST (U/L) 56.27±43.72 61.43±35.11 0.391 Albumin (mg/dL) 2.65±0.77 2.71±0.61 0.784 T. proteins (mg/dL) 6.18±1.23 6.65±0.66 0.135 PT (seconds) 19.72±4.99 19.63±6.41 0.882 TLC (cell/mm3) 2.44±0.64 7.11±2.36 0.001** ANC (cell/mm3) 1.26±0.22 4.81±1.99 0.001** HB (gm/dL) 8.64±1.71 9.83±1.73 0.017* PLT count (cell/mm3) 64.04±38.68 128.87±90.43 0.001** Data are presented as Mean ± SD ; *Significant p-value ≤0.05, **highly significant p- value ≤0.01, Alanine transaminase (ALT), Aspartate transaminase (AST), prothrombin time (PT), total leucocytic count (TLC), absolute neutrophil count (ANC), hemoglobin (HB), platelets (PLT). ments with an absent or decreased Autoimmune neutropenia is believed number of mature neutrophils. In to be triggered by a viral infection (as some cases, mature forms of neutro- HCV) and, in some individuals, by im- phils are seen. This suggests that, in munization. When identified in chronic most cases, neutropenia is due to in- benign neutropenia, antibodies most creased consumption of neutrophils in commonly include immunoglobulin G the peripheral circulation and in the (IgG) antibodies against neutrophil tissues that result from phagocytosis glycosylated isoforms. Bone marrow of the antibody-coated neutrophils(16). examination typically demonstrates This case-control study was carried out increased cellularity of myeloid ele- aiming to detect presence of AIN in Soliman NM et al. 64

chronic liver disease patients with HCV. patients with autoimmune neutro- When analyzing the results of the la- penia, hematological evaluation boratory tests, our data showed a sta- demonstrates a TLC count that is ei- tistically significant difference be- ther decreased or within the reference tween study and control group regard- range and a neutrophil count less than ing ANC, HB and PLT count (p<0.05). In 1000/µ L (absolute neutropenia).

Table 2: Incidence of autoimmune neutropenia (AIN) ac- cording to granulocyte immuno-fluorescence test (GIFT) Study group Control group Variables p-value (n=30) (n=30) AIN 0.0002** Present 4 (13.3%) 0 (0.0%) Absent 26 (86.7%) 30 (100.0%) **highly significant p-value ≤0.01.

Performing sequential CBC counts with significantly lower levels of PLT count differential to document chronicity is compared to the control group important because most neutropenia (64.04±38.68 versus 128.87±90.43, re- in infants resolves with recovery from spectively, p< 0.05). Thrombocytope- an acute infection. In individuals with nia is a common hematological toxicity autoimmune neutropenia, the ANC of- among patients with chronic liver dis- ten remains less than 500(17, 18). Our da- ease, occurring in 15% up to 70% of pa- ta revealed that the study group had tients with cirrhosis(19)

Table 3: Odds ratio and 95% confident interval of the studied varia- bles among patients with low and normal TLC Variables Odds ratio 95% CI p-value Age (<40 years) 2.8 0.50-15.7 0.42 Gender (female) 1.5 0.54-4.2 0.43 Total bilirubin 1.0 0.12-8.3 0.70 Direct bilirubin 0.75 0.04-15.0 0.71 ALT 1.08 0.50-2.3 0.50 AST 0.73 0.41-1.3 0.28 Albumin 1.0 0.04-24.5 0.79 Total proteins 0.86 0.044-16.9 0.73 PT 1.2 0.31-4.6 0.79 ANC 5.0 1.5-16.5 0.006** HB 7.0 1.4-35.5 0.0098** PLT count 15.5 3.8-63.3 <0.0001** *Significant p-value ≤0.05, **highly significant p-value ≤0.01, Alanine trans- aminase (ALT), Aspartate transaminase (AST), prothrombin time (PT), total leucocytic count (TLC), absolute neutrophil count (ANC), hemoglobin (HB), platelets (PLT). Thrombocytopenia can increase the sive or surgical procedures, causing risk of bleeding associated with inva- delays or cancellations and preventing 65 AIN in CHC patients

initiation of selected planned thera- to detect positive anti-neutrophil anti- pies(20). The pathophysiology of bodies from peripheral blood of HCV thrombocytopenia in CLD is complex patients. Sensitivity for antibody de- and multi-factorial. Hypersplenism tection varies depending on the test. caused by portal hypertension and Indirect GIFT is more sensitive than subsequent PLT sequestration were monoclonal antibody-specific immobi- thought to be the primary pathogenic lization of granulocyte antigens mechanisms; however, splenic PLT (MAIGA) in detection of autoimmune pooling does not account for throm- neutropenia(26). Antibody specificity is bocytopenia in all patients; hyper- assessed by testing a panel of splenism is not always observed in pa- neutrophils from donors of known tients with cirrhosis and portal hyper- human neutrophil antigen (HNA) tension(21). Some studies in patients genotypes. Up to now, the combina- with CHC suggest that HCV may cause tion of GIFT and granulocyte direct bone marrow suppression and agglutination (GAT) has been the best decreased thrombopoiesis with re- means of detection, and this has duced hepatic production of the become the standard approach in thrombopoietin (thrombopoietic most laboratories. GIFT is considered growth factor)(22). Other pathogenic to be the most sensitive of the two mechanisms contribute to the de- methods(27,28). In our study, study creased numbers of circulating PLT in- group were found to have significantly clude increased PLT destruction medi- lower levels of ANC. Positive GIFT was ated by immune mechanisms involving 13.3% in study group versus 0.0% in the anti-platelet auto-antibodies and PLT- control group with statistically signifi- associated immune complexes(23,24). cant difference. The positive GIFT indi- Our data reported that about 13.3% of cates presence of auto-antibodies study group are GIFT positive, while against neutrophils in CLD patients there were no patients in the control with HCV. In Jeffery et al(29) study, pa- group had positive GIFT. There was tients with HCV infection can develop statistically significance difference be- peripheral blood cell count abnormali- tween the two groups regarding GIFT. ties that are commonly attributed to There is no statistically significance be- decreased thrombopoietin levels and/ tween two GIFT results regarding gen- or autoimmune mechanisms. This was der and Child-Pugh classifications. So, emphasized in this study, where GIFT results are independent of the cytopenias were occurred due to auto- degree of liver fibrosis, cirrhosis, and immune responses in the absence of splenomegaly. In Lane et al.(25) study, splenomegaly and/ or antiviral medica- bone marrow immunofluorescenece tions, which were excluded from the test (BMIFT) used to demonstrate au- study. The results of neutrophil anti- to-antibodies to granulocytes and their bodies may or may not be positive. precursors on fresh-frozen bone mar- Lalezali and colleagues(30) demonstrat- row slides. It used to differentiate AIN ed antibodies in 98% of patients with from other causes of childhood neu- chronic neutropenia, whereas Jonsson tropenia. GIFT was used in this study and Buchanan(31) demonstrated a posi- Soliman NM et al. 66

tive result in only 13 of 28 patients tion: some current issues. J Hepatol. studied. In conclusion, AIN is signifi- 2004; 40(2):341-52. cantly higher among CHC patients with 9. Galossi A, Guarisso R, Bellas L, Puoti low TLC compared to normal TLC pa- C. Extra-hepatic manifestations of tients independent of the degree of chronic HCV infection. Journal of Gastrointestinal and Liver Disease. liver fibrosis, cirrhosis, and spleno- 2007; 16(1): 65–73. megaly. 10. Himoto T, Masaki T. Extra-hepatic Manifestations and Auto-antibodies References in Patients with Hepatitis C Virus In- fection. Clin Dev Immunol. 2012; 1. Alter MJ. Epidemiology of hepatitis 2012: 871401. C virus infection. World J Gas- 11. Himoto T, Nishioka M. Auto- troenterol. 2007; 13(17): 2436-2441. antibodies in hepatitis C virus- 2. Benova L, Awad SF, Miller FD, Abu- related chronic liver disease. Hepati- Raddad LJ. Estimation of Hepatitis C tis Monthly. 2008; 8(4):295–303. Virus Infections Resulting From Ver- 12. Mojtaba A, Brian C, Edmund K. Auto- tical Transmission in Egypt. immune neutropenia in adults. Hepatology. 2015; 61: 834–842. Autoimmun Rev. 2009; 9: 62–66. 3. Bruggmann P, Berg T, Ovrehus AL, 13. Fromont P, Prié N, Simon P, Moreno C, Brandao-Mello CE, Cesbron-Gautier A, et al. Granulo- Roudot-Thoraval F. Historical epi- cyte antibody screening: evaluation demiology of hepatitis C virus (HCV) of a bead-based assay in compari- in selected countries. J Viral Hepat. son with classical methods. Trans- 2014; 21(Suppl 1): 5-33. fusion. 2010; 50(12): 2643-2648. 4. Badar S, Khubaib B, Idrees M, et al. 14. Lalezari P, Jiang AF, Yegen L, Association of Hepatitis C Virus with Santorineou M. Chronic autoim- Insulin Resistance: Evidences from mune neutropenia due to anti-NA2 Animal Studies and Clinical Studies. antibody. N Engl J Med. 1975; Hepat Mon. 2012; 11-15. 293(15):744-747. 5. El-Ezawy H, Hemida M, El-Shiekh M. 15. Capsoni F, Sarzi-Puttini P, Zanella A. Association of Insulin Resistance, Primary and secondary autoim- Diabetes Duration and Diabetes mune neutropenia. Arthritis Res Treatment with Risk for Hepato- Ther. 2005; 7(5): 208–214. cellular Carcinoma. J Am Sci. 2011; 16. Lee CY, Herant M, Heinrich V. Tar- 7(11): 95-101. get-specific mechanics of phagocy- 6. El-Zanaty F, Way A. Egypt Demo- tosis: protrusive neutrophil res- graphic and Health Survey 2008. ponse to zymosan differs from the Cairo: Ministry of Health, El-Zanaty uptake of antibody-tagged patho- and Associates, and Macro Interna- gens. J Cell Sci. 2011; 124(7): 1106– tional, 2009. 1114. 7. El-Sayed M. The New National Strat- 17. Inoue S. Autoimmune and Chronic egy on Viral Hepatitis. Presentation Benign Neutropenia. eMedicine. in a seminar hosted by Al-Ahram 2009; 13. Science Clubs, 1 June 2014. 18. Denic S, Showqi S, Klein C, Takala 8. Agnello V, De Rosa FG. Extra-hepatic M, Nagelkerke N, Agarwal MM. disease manifestations of HCV infec- Prevalence, phenotype and inher- itance of benign neutropenia in 67 AIN in CHC patients

Arabs. BMC Blood Disord. 2009; 26. Bux J, Behrens G, Jaeger G, Welte 27: 9-13. K. Diagnosis and clinical course of 19. Giannini EG. Review article: autoimmune neutropenia: analy- thrombocytopenia in chronic liver sis of 240 cases. Blood. 1998; disease and pharmacologic trea- 91(1):181-186. tment options. Alim Pharmacol 27. Bierling P, Bux J, Curtis B, et al, Ther. 2006; 23: 1055–1065. ISBT Working Party on Granulo- 20. Dieterich DT, Spivak JL. Hemato- cyte Immunobiology. Recom- logic disorders associated with mendations of the ISBT Working hepatitis virus infection and their Party on Granulocyte Immuno- management. Clin Infect Dis. biology for leucocyte antibody- 2003; 37: 533–541. screening in the investigation and 21. Adinolfi LE, Giordano MG, An- prevention of antibody-mediated dreana A, et al. Hepatic fibrosis transfusion-related acute lung in- plays a central role in the patho- jury. Vox San. 2009; 96:266-269. genesis of thrombocytopenia in 28. Fromont P, Prié N, Simon P, et al. patients with chronic viral hepati- Granulocyte antibody screening: tis. Br J Haematol. 2001; 113: 590– evaluation of a bead-based assay 595. in comparison with classical 22. Wang C, Yao W, Wang A, Chang T, methods. Transfusion. 2010; Chou P. Strong association of 50(12): 2643-2648. hepatitis C virus (HCV) infection 29. Jeffery MK, Bob G, Elizabeth MB, and thrombocytopenia: implica- et al. Bone marrow biopsy in pa- tions from a survey of a commu- tients with hepatitis C virus infec- nity with hyperendemic HCV in- tion: spectrum of findings and di- fection. Clin Infect Dis. 2004; 39: agnostic utility. Am J Hematol. 790–796. 2010; 85:106–110. 23. Doi T, Homma H, Mezawa S, et al. 30. Lalezari P, Khorshidi M, Petrosova Mechanisms for increment of M. Autoimmune neutropenia. J platelet associated IgG and plate- Pediatr. 1986; 109(5):764-769. let surface IgG and their implica- 31. Jonsson OG, Buchanan GR. Chron- tions in immune thrombocytope- ic neutropenia during childhood. nia associated with chronic viral A 13-year experience in a single liver disease. Hepatol Res. 2002; institution. Am J Dis Child. 1991; 24: 23–33. 145(2):232-235. 24. Kajihara M, Kato S, Okazaki Y, et al. A role of autoantibody- mediated platelet destruction in thrombocytopenia in patients with cirrhosis. Hepatology. 2003; 37: 1267–1276. 25. Lane SW, Hassell P, Kennedy GA, Fung YL, Williams BA. Characteri- zation of the bone marrow im- munofluorescence test in child- hood autoimmune neutropenia. Int J Lab Hematol. 2009; 31(5): 567-571.