Immunological disorders in C virus chronic hepatitis

Nephrol Dial Transplant (1996) 11 [Suppl 4]: 31-35

Nephrology Dialysis Transplantation

Immunological disorders in C virus chronic hepatitis J. Cosserat, P. Cacoub1 and O. Bletry Department of Internal Medicine, CMC Foch, 40 Rue Worth, 92151 Suresnes Cedex; department of Internal Medicine, Hopital Pitie-Salpetriere, 83, Boulevard de 1' Hopital, 75651 Paris Cedex 13, France Abstract. Hepatitis C virus infection can be accompanied by a number of systemic, non-specific or autoimmune disorders and by extra-hepatic biological abnormalities. Their exact prevalence remains to be determined, together with their association with certain pathologies such as non-Hodgkin lymphoma. The most frequent biological peculiarity is cryoglobulinaemia, found in more than 50% of patients. It is only symptomatic in less than one-third of cases (as purpura, Raynaud's syndrome, neuropathy or renal failure), and could be the origin of benign lymphoproliferative haematological pathology, then of a malignant one (non-Hodgkin lymphoma). Rheumatoid arthritis may develop in 20-25% of cases. Periarteritis nodosa is a rare but possible occurrence. Autoimmune disorders proper include hepatitis with anti-LKMl antibodies, thyroiditis, skin manifestations, complex connectivitis and Gougerot-Sjogren's syndrome. This syndrome is histologically detected in 15-50% of patients. It affects women preferentially and is usually not accompanied by anti-SSA or anti-SSB antibodies. Thyroidites are different from those occurring under Interferon-a therapy, and should be systematically investigated before initiating that therapy. Lichen planus can be associated to HCV infection. The proven induction of autoimmune disorders by Interferon therapy requires that immune disorders be fully assessed before treatment initiation, and close monitoring for their occurrence is necessary throughout the treatment course. TSH and thyroid hormone in particular should be closely monitored.

Introduction It is only 5 years since it was established that HCV is associated to a number of immunological disorders, and literature is already profuse on the topic. It is also often contradictory between studies, in particular because the first HCV screenings, mainly using first generation ELISA, often were false positives. Pawlotsky et al's 1994 case-control study [1] provided a quite accurate account of the magnitude of the problem. In the 61 consecutively recruited HCVinfected patients, there were proportions of 70% posit-

ive rheumatoid factors, 36% positive cryoglobulins, 21% positive anti-smooth muscle antibody, 13% high titre anti-nuclear antibody, and these differences were greatly significant in comparison to the HCV-negative control group. In that study, there was no significant increase in antithyroid antibodies and anti-LKMl antibodies, in contradiction with results from other studies. A more recent study by the same team confirmed that 41% of the patients examined had at least one auto-antibody and that the three most frequent HCV serotypes in France also expose to autoimmune manifestations. In fact a distinction has to be made between the high prevalence of biological or pathological abnormalities (with regards to salivary glands) in prospective studies, and the scarcity of their clinical expression. For that reason, biological and histological disorders are distinguished in this study for each auto-immune manifestation. Non specific dysimmune manifestations These are studied in detail in another section of this monograph, and are therefore only summarized at this stage. Biological abnormalities

Beside the rheumatoid factor present in 26-70% of patients according to series, cryoglobulinaemia is the most important. Since 1990, and above all 1991, Italian and French teams [2,3] have warned that mixed cryoglobulinaemias could be associated with active HCV infections. It has now been well established that antibodies to HCV are present in 50-75% of patients with essential mixed cryoglobulinaemia: 52% in a series involving more than 100 patients in a survey conducted by one of our team [3]. There are several arguments suggesting that HCV is directly responsible for cryoglobulin production: concomitant increase in transaminases, viral RNA concentration 20-1000 times greater in precipitates than in sera [2,4], presence of HCV capside proteins in the cryoprecipitate [5]. Tackling the problem the other way, Lunel et al. [5] found cryoglobulinaemia (always mixed) in 54% of

© 1996 European Renal Association-European Dialysis and Transplant Association

J. Cosserat, P. Cacoub and O. Bletry

32

patients from a cohort of 127 untreated chronic hepatitis C, whereas this abnormality was only found in 25% of non HCV-related chronic liver diseases. Clinical abnormalities

There are four types of clinical abnormalities: clinical manifestations of cryoglobulinaemias, non-Hodgkin lymphomas, rheumatological patterns resembling rheumatoid arthritis, systemic vasculitis resembling periarteritis nodosa. Only 27% of the mixed cryoglobulinaemias found in a little more than half HCV-positive patients are symptomatic [5] and are presenting with: purpura (33%), Raynaud's syndrome (28%), peripheral neuropathy (28%) and/or, renal failure (17%). When these mixed symptomatic cryoglobulinaemias are treated with interferon a, there is a parallel decrease in viral RNA and cryoglobulinaemia (in one of two cases) but clinical and biological relapse is almost systematic upon interferon withdrawal [6]. The involvement of the C virus in the genesis of non-Hodgkin lymphomas (NHL) is under debate. Ferri [7] compared the prevalence of HCV infection in 50 patients with type B-NHL relatively to patients with Hodgkin's disease. It was 30% in the former group, 3% in the latter and 1-3% in the Italian general population. Only 8% of these patients had cryoglobulinaemia. There were no serological markers of active virus C infection. These were intermediate malignancy lymphomas. The pathophysiological hypothesis implicates the lymphotropism of HCV and its capacity to induce benign lymphoproliferative disease (cryoglobulinaemia), and possibly a malignant one (NHL) under certain genetic, immunological and environmental conditions [7]. Pozzato [8] performed bone marrow biopsies in 31 patients with cryoglobulinaemia, 84% of whom had HCV+ RNA and he found low grade NHL in 38.7% of cases, and reactive hyperplasia in 35.5% of cases [8]. De Vita [9] isolated HCV RNA strands from the RNA of a parotid lymphomatous tumour. McColl put all these results to question when he studied 38 Scottish patients with NHL (including 35 type B) and 35 patients with LLC. None was infected by HCV. The identical prevalence of NHL in Italy and Scotland should encourage this author to investigate another interpretation to that association [10]. Several French teams have drawn the attention of the American Rheumatology Conference in 1994 on the prevalence of polyarthritis simulating rheumatoid arthritis in HCV-infected patients: 20-25% according to that study. In half of cases, this pseudo-polyarthritis is accompanied by carpal tunnel syndrome and palmar tenosynovitis; these patients get relief from an association of steroids and anti-malaria drugs. To our knowledge, no case of polyarteritis nodosa in the course of a known hepatitis C infection has been described to date. In contrast, anti-HCV antibodies have been detected in 5-20% of patients with polyarter-

itis nodosa, depending on the screening method used [11,12]. Under positive serology clearly confirmed by RIBA 2, prevalence was between 5 and 12%. In the same cohort of patients, Cacoub et al. [11] found 12% of anti-HCV antibodies, 8% HBs antigen-positives and 20% HBc antibodies. The prevalence of anti-HCV antibodies was significantly greater in patients with polyarteritis nodosa than in those with systemic lupus erythematosus (12% vs 2%; P=0.03). Autoimmune manifestations

Autoimmune hepatitis (AIH) and HCV The relationship between autoimmune hepatitis and HCV infection is still to be clarified. It is worth reminding that since Homberg's description in 1987 [13] two major types of AIH can be distinguished: • Type I AIH, which essentially affects young women and are accompanied by antinuclear antibodies (Ab), anti-smooth muscle Ab (in fact anti-actin antibodies) and major hypergammaglobulinaemia, usually in excess of 30 g/1. • Type II AIH, identifiable by the presence of antiliver-kidney microsome antibodies (anti LKM1) which in fact are directed against a cytochrome (more specifically cytochrome p 450 II D6); hypergammaglobulinaemia is generally more modest, in the order of 20 g/1. Some of those patients have developed a yet more specific Ab called anti-liver cytosol 1 (LCI). Biological abnormalities

It is worth stressing again the fact that many studies are biased according to the type of serological tests used and to the geographical origin of the patients. The prevalence of anti-HCV antibodies is thus much greater in Italy than in the UK [14] and ELISA tests are frequently falsely positive because of AIH hypergammaglobulinaemia [15]. • During hepatitis C: anti-smooth muscle antibodies are present in 5-20% of patients [16]; in Pawlotsky's study [1], high titre (> 1/160) anti-smooth muscle antibodies were found in 7% of patients with chronic hepatitis C versus 3% in chronic hepatitis B patients, although that difference was not significant. On the other hand, anti-LKMl antibodies are present in 5-6% of chronic hepatitis C patients [1,16] but only in those patients (not in controls nor in chronic hepatitis B patients). As a rule, carriers of HCV and anti LKM1 antibodies have no anti-LCI antibodies. It is thought that LKM1 antibody production by HCVinfected individuals is linked to an analogy in nucleotide sequences between the HCV genome and a part of the cytochrome p450 II D6 gene [16]. • There is no HCV infection in patients with type I autoimmune chronic hepatitis. • Sixty to 90% of anti-LKMl carriers have positive

Immunological disorders in C virus chronic hepatitis

serology. In Lunel et al.'s study [17] involving 83 patients with an anti-LKMl antibody, HCV serology was positive by ELISA in 77% of cases and 49% by immunoblot, the latter patients being all positive for HCV viral sequences. Most patients with anti-LKMl antibody and anti HCV antibody are also carriers of an antibody directed against a fusion protein (GOR 47-1) isolated in a chimpanzee infected by non-A, non-B hepatitis. In Michel's study [18], of the 29 patients with an anti-LMKl antibody, 14 were seropositive for HCV and 11 had an anti-GOR antibody. Clinical consequences

33

the general population, although these results need to be confirmed on larger GS series. It has to be noted that patients with GS and true seropositivity are usually free of anti-SS A and anti-SS B antibodies, which are one of the immunological criteria of primary or secondary GS [1,21]. Clinical abnormalities

Clinical abnormalities are less frequent and above all more discreet that biological disorders. In Haddad et al.'s study [19], only 35% of patients followed up for chronic hepatitis C had dry mouth and none suffered from dry eyes. In Almazio et aL's study [22], Schirmer's test was positive in 14% of chronic hepatitis C as compared with 41% of primary biliary cirrhosis. Women with a mean age of 50 years are more susceptible to HCV-associated GS (sex ratio «4/l); these syndromes can be either primary or secondary, but as already stated they are not usually accompanied by anti-SS A or anti-SS B antibodies. The interval between the beginning of liver disease and the occurrence of dryness syndrome would seem to be more than 15 years.

It can be concluded from the various studies that there is not link between type I AIH and HCV infection. Within type II AIH, which are by definition accompanied by anti-LKMl antibodies, two groups of patients are to be distinguished: • Young women carriers of anti-LCI but no antiHCV antibodies, and who improve under steroid therapy. • Older men or women (40-70 years) who exhibit seropositivity for HCV by immunoblot, who develop anti-GOR antibodies but no anti-LCI antibodies, and whose liver disease will improve under Autoimmune tbyroiditis and HCV interferon treatment. This is more likely a special type of chronic hepatitis C than a true AIH. Associations between these two conditions can only be explored in patients who have not been treated with Interferon, because of the high incidence of the thyroid Gougerot-Sjogren syndrome (GS) and HCV disorders induced by that cytokine. Pathological abnormalities

The Gougerot-Sjogren syndrome is pathologically characterized by lymphocyte infiltrates in accessory salivary glands, and their importance is graded as four classes according to Chrisholm's classification; grades 3 or 4 are required to establish a GS diagnosis. Haddad's pilot study [19] involved labial biopsy specimens from 28 patients with HCV-related chronic hepatitis or cirrhosis and from 20 necropsic controls. Chisholm grade 3 or 4 was noted in 57% of HCV + patients and in only 5% of controls. In Pawlotsky's case-control study [1], 49 out of 61 patients had labial biopsy and Chisholm grade 3 or 4 was found in 14% of them; the lymphocyte infiltrate tended to be pericapillary rather than pericanalar. Biological abnormalities

The reverse approach which consisted in looking for HCV antibodies in patients with histologically confinned GS was performed as early as 1992. Hypergammaglobulinaemia, frequent in GS, explains the risk of overestimating HCV prevalence in patients consulting for a dryness syndrome. In Mariette et al.'s study [20], 15 of the 20 primary GS studied had anti-HCV antibodies detected by ELISA although only two were still positive under RIBA2. This constitutes a 10% prevalence, greater than that of HCV in

Biological disorders

There again, only the ELISA-positive confirmed by immunoblot are to be taken into account, otherwise the dysthyroiditis-HCV association would be overestimated. Paterson et al. [23] found 14% of thyroid abnormalities in a population of patients with chronic hepatitis C. Tran et al. [24] refined those results by exploring 72 cases of chronic hepatitis C: there were no antithyroid antibodies in men whereas those Ab were present in 31% of the 29 women (anti-thyroglobulin and/or anti-microsome and anti-thyroxidase). These figures were higher than in Pawlotsky's study [1], who found only 7% of anti-thyroid antibodies among 61 patients; but the latter study did not specify whether anti-thyroid antibodies were only present in the 26 women, and there was no screening for antithyroperoxidase antibodies. Conversely, anti-HCV antibody screening during Hashimoto's thyroiditis is very disappointing. Quaranta et al. in a preliminary study had found 6.8% anti-HCV antibodies in patients with anti-thyroid antibodies but these results were not confirmed by RIBA. In an unpublished study conducted in the Rhone-Alpes region of France (Bisot-Locard, Bernet, Rousset et al), 216 Hashimoto's sera were tested: only one was positive by ELISA (and confirmed so by RIBA 2).

34

Clinical disorders

In view of this latter figure, clinical disorders should be investigated in HCV-infected populations. In Pawlotsky 's study [1], only one patient out of 61 had clinical thyroiditis. In contrast, in Tran's study [24] four of the nine women with anti-thyroid antibodies (among the 29 tested) evolved to authentic Hashimoto's syndrome (i.e. 13.8% of the female population in the study). Lichen planus and HCV

J. Cosserat, P. Cacoub and O. Bletry

study on hepatites B treated by Interferon, autoantibodies occurred in 87% of patients; these were mainly anti-nuclear, anti-smooth muscle and antithyroid antibodies. The prevalence of clinically expressed true autoimmune diseases was assessed as 19% in a series of 135 patients treated for carcinoid tumours. Dysthyroiditis is the most frequent. The risk of inducing or exacerbating autoimmune diseases with Interferon has two consequences: • HCV-associated dysimmune disorders have to be assessed before IFN treatment. • treating active chronic hepatitis C by IFN may aggravate some possibly associated autoimmune diseases; in particular, close monitoring of TSH and thyroid hormones is paramount.

Lichen planus is a chronic skin-mucosal disease of middle-age adults; it is characterized by sub-epidermal lymphocyte infiltrate. Its autoimmune origin is likely and would implicate T cell-mediated immunity. It was established 10 or 15 years ago that lichen References planus is associated with chonic liver diseases which 1. Pawlotsky M, Ben Yahia M, Andre C et al. Immunological have been found in 35% of patients in some series. disorders in C virus chronic active hepatitis: a prospective case Italian and French teams [25] have shown that chronic control study. Hepatology 1994; 19: 841-848 hepatitis C (evidenced by viral RNA in serum) was the 2. Cacoub P, Musset L, Lunel-Fabiani F et al. Mixed cryoglobulimost frequent. nemia and hepatitis C virus. Am J Med 1994; 96: 124-132 Lichen planus may follow chronic hepatitis C or be 3. Casato M, Pucillo LP, Lagana B et al. Cryoglobulinemia and hepatitis C virus. Lancet 1991; 337: 1047-50 concomitant. 4. Agnello V, Chung R, Kaplan L. A role for hepatitis C virus This association was confirmed by Pawlotsky et al. 's infection in type II cryoglobulinaemia. N Eng J Med 1992; 19: case-control study [1], who found 5% lichen planus in 1490-1495 their 61 patients, whereas the general population pre- 5. Lunel F, Musset L, Cacoub P et al. Cryoglobulinemia in chronic liver diseases: role of hepatitis C virus and liver damage. valence is less than 1%. Gastroenterology 1994; 106: 1291-1300 This association poses therapeutical problems that 6. Misiani R, Bellavita P, Fenili D et al. Interferon alpha-2a have not been resolved: it could be expected that therapy in cryoglobulinemia associated with hepatitis C virus. treating HCV infection by Interferon would reduce the N Engl J Med 1994; 330: 751-756 7. Feni C, Carraciolo F, Zigneco AL et al Hepatitis C virus skin-mucosal lesions but cases have been reported infection in patients with non-Hodgkins lymphoma. Br where lichen lesions were enhanced by Interferon. J Haematol 1994; 88: 392-394 Other dermatological manifestations are associated 8. Pozzato G, Mazzaro C, Crovatto M et al. Low-grade malignant to HCV. Some have immunological origin, such as the lymphoma, hepatitis C virus infection and mixed cryoglobulinemia. Blood 1994; 84: 3047-3053 vasculitis of mixed cryoglobulinaemias. Others have 9. De Vita S, Sansonno D, Dolcetti R et al. Hepatitis C virus no known immunological substratum, like late skin within malignant lymphoma lesion in the course of type II mixed porphyria tardea associated with anti-HCV antibodies cryoglobulinemia. Blood 1995; 86: 1887-1892 in more than 80% of cases. 10. McCole M, Campbell Tait R. Hepatitis C virus infection in Complex connectivitis Some HCV-infected patients exhibit several associated autoimmune signs, such as for instance GoujerotSjogren syndrome and type II AIH. A more original association is that of incomplete CREST syndrome with anti-centromeric antibodies, GS and hepereosinophilia. We observed this association twice (communication to the SNFMI Conference, Paris 9—10 December 1992); Pawlotsky et al. [1] also reported a CREST syndrome but did not specify if other autoimmune manifestations were associated. Interferon-induced dysimmune complications Predictable from experimental facts (aggravation of murine lupus), induction of autoimmune disorders by IFN has been confirmed in man [26]. In a prospective

patients with lymphoproliferative disorders. Br J Haematol 1996; 92: 766 11. Cacoub P, Lunel-Fabiani F, Le Thi Huong Du. Polyarteritis nodosa and hepatitis C virus infection. Ann Intern Med 1992; 116:605-606 12. Carson CW, Conn DL, Czaja AJ, Wright TL, Brecher ME. Frequency and significance of antibodies to hepatitis C virus in polyarteritis nodosa. J Rheumatol 1993; 20: 304-309 13. Homberg JC, Abuaf N, Bernard O et al. Chronic active hepatitis associated with anti liver/kidney microsome antibody type 1: a second type of 'autoimmune' hepatitis. Hepatology 1987; 7: 1333-1339 14. Lenzi M, Johnson PJ, McFarlane JG et al Antibodies to hepatitis C vims in autoimmune liver disease: evidence for geographical heterogeneity. Lancet 1991; 338: 277-80 15. McFarlane IG, Smith HM, Johnson PJ et al Hepatitis C virus antibodies in chronic active hepatitis: pathogenetic factor or false positive result? Lancet 1990; 335: 754-757 16. Lunel F. Hepatite C et anomalies immunologiques. Gastroenterol Clin Biol 1994; 18: 829-838 17. Lunel F, Abuaf N, Frangeul L et al Liver/kidney microsome antibody type 1 and hepatitis C virus infection. Hepatology 1992; 16: 630-636 18. Michel G, Ritter A, Gerken G et al Anti-GOR and hepatitis C virus in autoimmune liver disease. Lancet 1992; 339: 267-269

Immunological disorders in C virus chronic hepatitis

35

19. Haddad J, Deny P, Munz-Gotheil C et al Lymphocytic sialadenBeaugrand M. Latent autoimmune thyroid disease in patients itis of Sjogren's syndrome associated with chronic hepatitis C with chronic HCV hepatitis. / Hepatol 1992; 16: 244-245 virus liver disease. Lancet 1992; 339: 321-323 24. Tran A, Quaranta JF, Benzaken S et al. High prevalence of 20. Mariette X, Zerbib M, Jaccard A et al Hepatitis C virus and thyroid autoantibodies in a prospective series of patients with Sj5gren's syndrome. Arthritis Rheum 1993; 36: 280-281 chronic hepatitis C before interferon therapy. Hepatology 1993; 18: 253-257 21. Jorgensen C, Perney P, Legouffe MC et al Association between hepatitis C virus and Sjogren's syndrome. Arthritis Rheum 1994; 25. Jubert C, Pawlotsky JM, Pouget F et al Lichen planus and hepatitis C virus-related chronic active hepatitis. Arch Dermatol 37: S 207 1994; 130: 73-76 22. Almazio P, Provenzano G, Scimemi M et al Hepatitis C virus and Sjdgren's Syndrome. Lancet 1992; 339: 989-990 26. Bletry O, Papo T. Interferon alpha and gamma: indications dans les maladies systemiques. Ann Med Interne 1993; 144: 557-562 23. Paterson D, Hartmann DJ, Duclos-Vallee JC, Jouanolle H,