Abstract
Morbidity associated with hepatitis C virus (HCV) infection can involve a variety of extrahepatic conditions, including lichen planus (LP) and sialadenitis, predominantly or exclusively involving the oral region, conditions which have been largely neglected in reviews. The literature suggests that HCV-infected patients may frequently have Sjögren-like sialadenitis with mild clinical symptoms, whereas oral LP may be significantly associated with HCV infections in Southern Europe and Japan but not in Northern Europe. These geographical differences could be related to immunogenetic factors such as the HLA-DR6 allele, significantly expressed in Italian patients with OLP and HCV. Analysis of experimental data suggests that HCV could be involved in the pathogenesis of both these diseases. Moreover, parotid lymphoma may arise in patients with sialadenitis, mainly with type II cryoglobulinemia. Little attention has been paid to oral health needs in HCV-infected patients and the variable effect of interferon-alpha therapy on oral tissues. Further research is needed, because of the potentially great influence of oral diseases possibly linked to HCV on the quality of life of millions of patients.
(I) Introduction
Hepatitis C virus (HCV) is an enveloped, single-stranded positive-sense RNA virus that was isolated in 1989 from a chimpanzee chronically infected by contamination with a human factor VIII concentrate (Choo et al., 1989). Although details of HCV replication are not known, it is thought to take place in the cytoplasm, where, among other products, negative-sense viral RNA (vRNA), replicative double-stranded forms, and non-structural proteins are synthesized. Their presence can thus be used as evidence that virus multiplication, as opposed to passive transportation, is occurring (Negro et al., 1999). HCV has an extremely variable genome, and six distinct genotypes and multiple subtypes have been identified (Pileri et al., 1998). Furthermore, sequence variants forming a quasi-species may circulate within an individual, possibly as a consequence of ongoing immune surveillance and viral mutations (Toyoda et al., 1998). HCV is presently considered the main etiologic agent of both blood-borne and sporadic non-A non-B hepatitis, and is one of the major causes of chronic liver disease worldwide.
The overall estimated prevalence of HCV infection is 3%, representing approximately 170 million infected people worldwide. However, there is great geographic variation in the prevalence of infection. On the basis of studies among blood donors (Wasley and Alter, 2000), the lowest prevalence of anti-HCV antibodies (0.01-0.1%) is in the United Kingdom and Scandinavia, followed by slightly higher rates (0.2-0.05%) in Western Europe, North America, most areas of Central and South America, Australia, and South Africa; intermediate rates (1-5%) are reported in Brazil, Eastern Europe, the Mediterranean area, the Middle East, the Indian subcontinent, and parts of Africa and Asia, and the highest HCV prevalence is reported in Egypt (17-26%). The natural history of HCV infection is difficult to assess because of the usually silent onset of the acute phase and the few symptoms seen during the early stages of chronic infection. Acute infection leads to chronic infection in the majority of persons (up to 80%), of whom 20% will eventually develop cirrhosis (Di Bisceglie, 1998). Once cirrhosis is established, the risk of the subject’s developing hepatocellular carcinoma is approximately 1 to 4% per year (Colombo et al., 1991).
Morbidity associated with HCV infection is due not only to the sequelae of chronic liver disease, but also to a variety of extrahepatic manifestations (Table 1). Most of these are immune-mediated, possibly as a result of virus-dependent proliferation of monoclonal or polyclonal lymphocytes (Bonkovsky and Mehta, 2001). One approach used to study the pathogenesis of HCV-associated disease is to follow its replicative pattern in infected tissue and to establish anatomo-clinical correlations. A series of extrahepatic cell types possibly supporting HCV replication has been proposed, including peripheral blood mononuclear cells, pancreas, thyroid, adrenal gland, kidney, lung, and spleen, and gastric, oral mucosa, and skin cells (De Vita et al., 2000; Laskus et al., 2000; Nagao et al., 2000c). However, in most cases, links to the various extrahepatic manifestations remain to be proved. Notably, two of the most frequently reported extrahepatic manifestations of HCV infection, lichen planus (LP) and sialadenitis, involve the oral region predominantly or exclusively. Some reviews on non-hepatic diseases associated with HCV infection, focused on its oral manifestations, are available (Lodi and Porter, 1997; Lodi et al., 1998; Roy and Bagg, 1999).
(II) Sjögren-like Sialadenitis
Cases of chronic hepatitis C with mixed cryoglobulinemia and signs of Sjögren’s syndrome (SS) first suggested a possible link between SS and HCV infection. This was reinforced by a French study finding that 57% of HCV-associated chronic liver disease patients exhibited a grade 3 or 4 sialadenitis (according to Chisholm and Mason’s classification, 1968; Haddad et al., 1992). A subsequent study noted that, in contrast to SS, lymphocytic infiltration in HCV-infected patients was pericapillary rather than periductal, with no destruction of the salivary gland (SG) ducts, and that lymphocytic capillaritis resembled an early stage of disease (Pawlotsky et al., 1994a). Further data (Almasio et al., 1992; Haddad et al., 1992; Guisset et al., 1993; Pawlotsky et al., 1994a; Pirisi et al., 1994; Poet et al., 1994; Boscagli et al., 1996; Taliani et al., 1997; Cacoub et al., 1999, 2000; Verbaan et al., 1999; Coates et al., 2000; Henderson et al., 2001; Loustaud-Ratti et al., 2001; Ferreiro et al., 2002) have shown that up to 80% of HCV-infected individuals may have some salivary or lachrymal abnormality, frequently represented by histological signs of mild sialadenitis, whereas clinical evidence of dry mouth and mainly of dry eyes is often absent (but may be underevaluated) (Table 2). Several studies seem indeed to indicate that this sialadenitis may be significantly different from that of SS. There is no female predominance, no specific antinuclear (SS-A [anti-RO] and SS-B [anti-LA]) antibodies, a less frequent association with the HLA-DR3 allele, milder histopathology (with a CD8+ rather than a CD4+ T-cell predominance), and fewer clinical symptoms (Pawlotsky et al., 1994a; Pirisi et al., 1994; Scott et al., 1997).
HCV may be present in the saliva of 83% of patients with HCV-associated sialadenitis (Jorgensen et al., 1996) and is also detectable in tears in concentrations higher than serum (Feucht et al., 1994). Much of the work on sialadenitis in HCV infection is not from dental institutions, and thus there are few available data on the prevalence and effect of hyposalivation in such patients. In contrast, from 0 to 19% of patients with frank SS can be HCV-infected, the frequency varying with the geographical region, the HCV test used, and the inclusion criteria (Table 3). False-positive immunoenzymatic (ELISA) HCV tests have been reported when hyper-γ-globulinemia (frequently observed in SS) is present (Marson et al., 1991; Vitali et al., 1992), and thus confirmatory tests are clearly recommended so that bias can be avoided. Indeed, several European (mainly Italian and Spanish) authors have reported prevalence of HCV antibodies in SS primary patients ranging from 3 to 75% using second-generation ELISA, from 14 to 19% using third-generation ELISA, and from 5 to 19% using second-generation confirmatory immunoblot assay (RIBA) (Ramos-Casals et al., 2001). Conversely, lower HCV prevalence (0-1% using a RIBA-2 technique) has been reported in USA and UK studies (King et al., 1994; Marrone et al., 1995; Porter et al., 1996). Notably, the number of patients fulfilling the European criteria is five- to ten-fold greater than those fulfilling the Fox criteria, and in the latter classification, hepatitis C patients are considered as distinct from SS (Roy and Bagg, 1999; Fox et al., 2000) (Table 3). This problem has been highlighted by Laustaud-Ratti et al. (2001), who reported that the prevalence of SS in a cohort of 45 French HCV-infected patients increased from 8% to 38% according to the Fox and the European criteria, respectively.
Controversy #1
HCV-infected patients may frequently have Sjögren-like sialadenitis with mild clinical symptoms, although we have few data on lachrymal and salivary function in these patients.
(III) Pathogenesis of HCV-related Sialadenitis
Viral factors
An early study suggested that the presence of sialadenitis was not dependent on the HCV genotype, since this lesion was found in patients infected with each of the three main HCV serotypes present in developed countries (Pawlotsky et al., 1994b). Other data confirm the lack of relationship between the presence of sialadenitis and particular genotypes (Loustaud-Ratti et al., 2001). It has been reported that patients with HCV-RNA in the saliva are more likely to complain of xerostomia (Roy et al., 1998), but whole salivary flow is not associated with the presence of the virus in the saliva (Ferreiro et al., 2002).
Cryoglobulinemia
The term ‘cryoglobulinemia’ refers to several syndromes characterized by the presence of abnormal proteins (cryoglobulins) that precipitate from cooled serum (Brouet et al., 1974). After the identification of HCV, it became evident that cryoglobulinemia is the most frequent extrahepatic manifestation of the virus (Dammacco et al., 2001). In most HCV patients, cryoprecipitates are composed of polyclonal IgGs against HCV and monoclonal IgM rheumatoid factors (type II), or of polyclonal IgG and IgM (type III) (Brouet et al., 1974). According to the composition of the cryoprecipitates, both these cryoglobulinemias are designated as mixed cryoglobulinemia (MC). About 15% of patients with MC have SS (Gorevic et al., 1980), whereas 80 to 100% have HCV infection (Dammacco et al., 2001); HCV is uncommon in SS patients without mixed cryoglobulinemia (King et al., 1994; Verbaan et al., 1999). However, not all the HCV-positive patients with evidence of salivary gland abnormalities have detectable serum cryoglobulinemia (Pawlotsky et al., 1994a; Loustaud-Ratti et al., 2001).
Direct infection of salivary glands
HCV is distantly related to flaviviruses (Houghton et al., 1991), which are able to infect the salivary glands of their arthropod vector. HCV antigens have been immunohistochemically detected in SG epithelial cells (De Vita et al., 1995) but not invariably (Verbaan et al., 1999). Whereas SGs of patients with chronic HCV infection but without clinical and histological signs of sialadenitis were not infected (Taliani et al., 1997), positive- and negative-strand HCV-RNA has been detected in minor SGs of patients with sialadenitis and chronic hepatitis C by PCR and in situ hybridization (Takamatsu et al., 1992; Biasi et al., 1995; Arrieta et al., 2001). In particular, HCV seems to infect and replicate in epithelial cells of the SG acini. However, there is no correlation between the percentage of infected SG epithelial cells and the serum HCV-RNA titer, and the infected cells did not show any differences with respect to unaffected ones (Arrieta et al., 2001).
Transgenic animal model
Recently, an animal model of transgenic mice carrying the HCV envelope genes E1 and E2 has been constructed (Koike et al., 1997). The mice developed an exocrinopathy involving the SGs and lachrymal glands (LGs) in 84% of cases. Initially, pericapillary lymphocytes were found, but soon focal infiltrates of small lymphocytes appeared, closely resembling the Chisholm and Mason grade 3 or 4 sialadenitis noted in humans (Haddad et al., 1992). Nests of lymphatic infiltrates were also noted in the LGs, but they occurred later and were less extensive than those found in the SGs. This model clearly suggests a direct role of the viral proteins in the pathogenesis of HCV-related sialadenitis. Moreover, because lymphocytic capillaritis preceded sialadenitis, this may reflect the pathological sequence in Sjögren-like sialadenitis occurring in human patients. The model also predicts that xerophthalmia, rarely reported in HCV-infected patients, would be a late development. The pathogenesis of this sialadenitis in transgenic mice is unclear, but it seems unlikely to be induced by an immune reaction against ductal cells expressing viral antigens, since only one out of 20 transgenic mice showed a weak antibody reaction to E1 protein. Alternative explanations include the induction of interferon-γ or interleukin 2 by HCV proteins or the induction of an immunological disturbance by the transgene.
Controversy #2
Analysis of experimental data suggests that HCV proteins play a direct role in the pathogenesis of sialadenitis, although it seems unlikely that sialadenitis is induced by direct infection of salivary glands or by an immune reaction against ductal cells expressing viral antigens.
(IV) Lichen Planus
Lichen planus (LP) is a chronic inflammatory disease that affects skin and mucous membranes of squamous cell origin. The oral form of LP (OLP) seems more common, chronic, and recalcitrant than the cutaneous type, persisting for more than 20 years without spontaneous remission (Scully et al., 2000). OLP is unlikely to be caused by a single antigen, given that studies of T-cell receptor variable region genes from lesional OLP T-cells have not revealed the use of a restricted number of different variable region genes (Thomas et al., 1997). Probably, OLP is the common outcome of the influence of a limited range of extrinsic antigens, altered self-antigens, or superantigens. In a minority of patients, precipitating factors have been identified. These include dental materials (mainly dental restorative materials such as amalgam), drugs (such as non-steroidal anti-inflammatory drugs and angiotensin-converting enzyme inhibitors), stress, trauma, and infectious agents (including herpes simplex virus I, herpes virus 6, cytomegalovirus, human papilloma virus, Epstein-Barr virus, and Helicobacter pylori and hepatitis viruses) (Scully et al., 2000). The frequent association of LP with chronic liver disease (CLD) is well-known (Table 4), especially in Mediterranean patients with oral erosive LP (Gandolfo et al., 1992b), but no hypothetical pathogenic correlation could be found until sensitive hepatitis C virus (HCV) assays became available. The risk of chronic liver disorders in LP patients is in fact independent of age, sex, and alcohol consumption, and is still significantly high after adjustment for a positive hepatitis B surface antigen (HBsAg) reaction (GISED, 1990). Markers of past hepatitis B virus (HBV) infection (antibodies to hepatitis B surface antigen [HBsAb] and to hepatitis B core antigen [HBcAbIgG]) have been reported in 21 to 30% of Spanish and Italian patients having LP (Ayala et al., 1986; del Olmo JAet al., 1990; Carrozzo et al., 1996), but this prevalence is near the average figure in the Mediterranean area. There are also a few reports of mainly skin lichenoid eruption following administration of different HBV vaccines (Rebora et al., 1999). Nevertheless, the majority of patients with both LP and CLD are not HBV-infected (del Olmo et al., 1990; Carrozzo et al., 1996). On the other hand, Wilson’s disease, haemochromatosis, and alpha-1-antitrypsin deficiency have rarely been related to LP (Rebora, 1992), whereas the association of LP with primary biliary cirrhosis is mostly due to the administration of penicillamine treatment (Graham-Brown et al., 1982; Powell et al., 1982; Oleaga et al., 1995).
In contrast to the Mediterranean data, retrospective and prospective studies of Scandinavian and British OLP patients have failed to show any significant correlation with liver diseases (Mobacken et al., 1984; Scully et al., 1985; el-Kabir et al., 1993). However, since 1991, more than 80 case reports world-wide have supported the link between LP and HCV infection (Mokni et al., 1991; Agner et al., 1992; Divano et al., 1992; Rebora, 1994; Cayla et al., 1993; Sassigneux et al., 1993; Strumia et al., 1993; Cecchi et al., 1994; Gandolfo et al., 1994; Jubert et al., 1994; Amichai et al., 1994; Benchikhi et al., 1994; Daoud et al., 1995; Hyrailles et al., 1995; Mouly et al., 1995; Bellman et al., 1996; Jauregui et al., 1996; Tanei et al., 1997; Schissel and Elston, 1998; Toure et al., 1998; Pellicano et al., 2000; Calista and Landi, 2001). Moreover, Epstein-Barr virus and the recently discovered viruses, hepatitis G virus and TTV, are not associated with LP (Pedersen, 1996; Nagao et al., 1997a; Lodi et al., 2000; Bez et al., 2001; Rodriguez-Inigo et al., 2001). HCV-associated hepatic disease may precede LP onset or may be diagnosed together with it. Apparently, there are no significant differences in the histopathological characteristics specific to OLP or in the ratio of T- and B-cells among infiltrating lymphocytes regardless of the presence or absence of HCV infection (Kirby et al., 1998; Nagao et al., 2000a).
Several controlled experiments have confirmed that HCV is the main correlate of liver disease in patients with LP (Carrozzo et al., 1996; del Olmo et al., 2000) and have suggested that HCV could be involved in the development of LP, especially the oral variety (Table 5). Analyses of early data suggested that mainly oral erosive OLP may be linked to HCV infection (Gandolfo et al., 1994; Carrozzo et al., 1996), but a recent controlled study reported that the reticular and plaque clinical forms are prevalent as well (Mignogna et al., 2000). Almost all data from Italy, Spain, Brazil, Japan, and all but one study in the USA support the existence of a relationship between LP and HCV infection (Divano et al., 1992; Gandolfo et al., 1994; Rebora, 1994; Bellman et al., 1995; Gimenez-Arnau et al., 1995; Nagao et al., 1995a; Schmitt et al., 1995; Tanei et al., 1995; Carrozzo et al., 1996; Sanchez-Perez et al., 1996; Serpico et al., 1997; Bagan et al., 1998; Mignogna et al., 1998; Chuang et al., 1999; Rossi and Colasanto, 2000; Beaird et al., 2001; Chainani-Wu et al., 2001; Eisen, 2002; Figueiredo et al., 2002). In France, the picture appears quite complicated: Low prevalence of HCV infection ranging from 3.8 to 4.9% was found in OLP patients from northern France (Strasbourg) and Ile-de-France (Paris) (Cribier et al., 1994; Dupin et al., 1997), whereas a study from southern France (Nice) reported 29% of HCV-positives in 28 patients with erosive OLP (Dupond et al., 1998). There are also controversial data from Germany and Turkey (Imhof et al., 1997; Grote et al., 1998; Ilter et al., 1998; Kirtak et al., 2000; Erkek et al., 2001), whereas investigations in England and Holland (Ingafou et al., 1998; Tucker and Coulson, 1999; Roy et al., 2000; van der Meij and van der Waal, 2000) did not find HCV antibodies in LP cases, possibly reflecting the low incidence in those populations.
On the other hand, the very high prevalence of HCV infection among Japanese LP patients (38 to 62%), mainly those coming from southern regions (Kyushu) (Nagao et al., 1995a; Tanei et al., 1995) was probably influenced by the prevalence of HCV in the general population, since 7.9% in their sixth decade were HCV-infected (Hayashi et al., 1994). Curiously, a small-scale study from Egypt, which has the highest reported prevalence of HCV infection in the general population, did not report a significant association between LP and HCV (Ibrahim et al., 1999). The few studies investigating the frequency of LP among HVC-positive subjects showed that from 1.6 to 20% of patients with HCV-related chronic hepatic disease may have LP (Pawlotsky et al., 1994a; Dupin et al., 1997; Grote et al., 1998; Coates et al., 2000; Nagao et al., 2000b, 2002; Henderson et al., 2001; Mignogna et al., 2001). Because these prevalences are generally higher than expected, OLP would probably be more easily identified in HCV-infected patients in countries with low levels of HCV, such as the UK (Carrozzo, 2001). Well-designed prospective studies, mainly from countries with low prevalence of HCV infection, are clearly needed.
Controversy #3
LP, mainly of the oral cavity, appears to be significantly associated with HCV infection in southern Europe and Japan but not in Northern Europe. However, this difference seems only partially related to the level of endemism of HCV in the general population.
(V) Pathogenesis of HCV-related LP
(A) Viral factors
The role of particular HCV genotypes in the pathogenesis of HCV-related OLP is ruled out by the observation that LP can be associated world-wide with the same genotypes commonly found in patients without LP (Pawlotsky et al., 1995b; Lodi et al., 1997b), though mainly genotype 1b seems associated with LP, and it appears to be uncommon in the UK (Harris et al., 1999). Studies have also shown no differences in serum levels or HCV-RNA levels between HCV-infected patients with LP and those without (Nagao et al., 1996a).
(B) Molecular mimicry
A constant feature in LP patients with HCV infection is the presence of polyclonal hypergammaglobulinemia (Carrozzo et al., 1996; Lodi et al., 1997a; Nagao et al., 1997b). This does not seem to be caused by an elevated frequency of non-organ-specific autoantibodies (Carrozzo et al., 1999), although anti-epithelial antibodies have been detected with significantly higher frequency in patients with HCV-related OLP than in those without HCV infection (Lodi et al., 1997a). In this study, however, 62% of the OLP-HCV+ve patients with circulating anti-epithelial auto-antibodies had been treated with alpha interferon (IFN-α) for more than 6 months (Lodi et al., 1997a), and Fleishmann et al. (1996) have shown, in vivo, that low-dose IFN-α for a period of 12 months can induce anti-epidermal auto-antibodies. Indeed, most OLP-HCV+ve cases do not have this type of auto-antibody (Carrozzo et al., 1999), and their pathological significance is probably negligible in the absence of α-IFN therapy. For unknown reasons, the effect of α-IFN therapy for HCV infection in patients with LP differs markedly from case to case. α-IFN has been reported to have no influence (Pawlotsky et al., 1995a), to ameliorate (Doutre et al., 1992, 1996; Strumia et al., 1993; Hildebrand et al., 1995; Pedersen, 1998; Nagao et al., 1999), or to trigger or worsen LP lesions, mainly in cases with oral involvement (Agner et al., 1992; d’Agay-Abensour et al., 1992; Cayla et al., 1993; Protzer et al., 1993; Sassigneux et al., 1993; Heintges et al., 1994; Papini et al., 1994; Perreard et al., 1994; Barreca et al., 1995; Fornaciari et al., 1995; Nunez et al., 1995; Areias et al., 1996; Nagao et al., 1996b; Schlesinger et al., 1997; Dalekos et al., 1998; Varela et al., 2000; Guijarro et al., 2001). The adjuvant use of ribavirin with α-IFN also seems to increase the risk of adverse cutaneous reactions, often of a lichenoid type (Sookoian et al., 1999; Manjón-Haces et al., 2001), whereas data on any effect on the oral mucosa are still lacking. In any case, molecular mimicry between the virus and host epitopes is unlikely to be active in LP, and hypergammaglobulinemia should rather be linked to cryoglobulinemia (Carrozzo et al., 1999).
(C) Genetic factors
Interestingly, geographic heterogeneity in the prevalence of HCV infection similar to that observed in OLP patients was also found in patients with other extrahepatic abnormalities linked to HCV infection, such as serum autoantibodies, porphyria cutanea tarda (PCT), and lymphoma (Lenzi et al., 1991; McColl et al., 1997; Lamoril et al., 1998). This striking fact suggests the existence of possible genetic differences among different populations. Indeed, it has been reported that PCT susceptibility is different in British and Italian patients, being correlated with mutation in the human leukocyte antigen (HLA)-linked hemochromatosis gene C 282Y in the former and to the H63D gene and HCV in the latter (Elder and Worwood, 1998). Most idiopathic LP is related worldwide to the HLA-DR1 (DRB1*0101) allele (La Nasa et al., 1995), whereas secondary LP as well as OLP is not. HCV-related OLP appears to be associated with the HLA-DR6 allele in Italy (Carrozzo et al., 2001), and this could partially explain the peculiar geographic heterogeneity of the association between HCV and LP.
(D) Direct infection of LP lesions
There are few data on the localization of HCV antigens in LP. Two studies used various immunohistochemical techniques to look for HCV antigens in paraffin-embedded sections of cutaneous LP, either frozen (Sansonno et al., 1995) or formalin-fixed (Boyd et al., 1998), but all samples failed to take up the stain. However, all but two of the samples studied were from persons not infected with HCV. Conversely, analysis of some data suggests that OLP lesions may contain both genomic and antigenomic HCV-RNA (Arrieta et al., 2000; Nagao et al., 2000c; Carrozzo et al., 2002), whereas no HCV intermediate RNA was detected in skin specimens of LP, although HCV-RNA may be occasionally detected in lesional skin (Mangia et al., 1999; Erkek et al., 2001). Both in situ hybridization and extractive PCR techniques revealed the presence of replicative intermediate HCV-RNA in OLP specimens (Arrieta et al., 2000; Nagao et al., 2000c; Carrozzo et al., 2002). Using sensitive extractive PCR, investigators detected positive and negative strands in 82 to 93% and 21 to 36% of the OLP tissue specimens, respectively (Nagao et al., 2000c; Carrozzo et al., 2002). Moreover, sequence analysis suggests a possible compartmentalization of HCV in the oral mucosa (Carrozzo et al., 2002). However, HCV is probably unlikely to cause direct damage to epithelial cells in OLP lesions, since it was also found in normal mucosa (Arrieta et al., 2000). The lympho-mononuclear infiltrate typically found in oral lichen lesions rather suggests that the progressive destruction of the oral mucosa lining is due to local immune aggression. Recent data have shown that HCV-specific T-cells can be found in the oral mucosa of patients with chronic hepatitis C and LP (Pilli et al., 2001). In situ and peripheral blood-derived T-cell lines were able to proliferate and to produce gamma interferon upon stimulation with structural and non-structural HCV antigens, and showed the same specificity. However, T-cell clones present in the oral mucosa showed a TCR-Vβ chain usage different from those circulating in the peripheral blood. Furthermore, tetramer assays with four different HCV antigens showed that the frequency of HCV-specific CD8+ T-cells was higher in mucosa tissue than in the blood (Pilli et al., 2001).
Controversy #4
Immunogenetic factors partially explain the geographic heterogeneity of the association between HCV and LP. Analysis of experimental data strongly suggests that HCV is involved in the pathogenesis of OLP via local induction of an immune response specific for HCV epitopes.
(VI) Other Diseases
Some cases of parotid gland lymphoma in Italian patients with HCV infection have been reported (De Vita et al., 1995; Luppi et al., 1996; Ascoli et al., 1998), mainly associated with type II cryoglobulinemia or SS. The occurrence of B-cell non-Hodgkin’s lymphoma is a complication of SS and, at least in Italy, the USA, and Japan, of chronic HCV infection (Ferri et al., 1994; Luppi et al., 1996; Zuckerman et al., 1997). Lymphomas occurring in both diseases share several characteristics: predominance of low-grade, marginal zone histological type, frequency of mucosal localization, possible transformation into a large B-cell lymphoma, association with asymptomatic low-level cryoglobulinemia (De Vita et al., 1997; Mariette, 2001). It has been proposed that, in both diseases, the first event of lymphomagenesis is chronic stimulation at the site of the disease of polyclonal B-cells capable of secreting rheumatoid factor (De Vita et al., 1997; Mariette, 2001).
Oral verrucous and squamous cell carcinomas have been reported in HCV-infected patients with or without OLP (Nagao et al., 1995b, 1996c, 1999; Carrozzo et al., 1997; Porter et al., 1997; Cervoni, 1998; Lo Muzio et al., 1998), although the epidemiological relevance of this observation is unclear. HCV prevalence is generally not increased in patients with potentially malignant oral lesions such as leukoplakia (Carrozzo et al., 1996; Nagao et al., 1997c), and HCV is a common cause of liver cirrhosis, which may itself represent an independent risk factor for the development of oral cancer (Sorensen et al., 1998). In contrast, both positive and negative HCV-RNA strands were detected in oral cancer tissues (Nagao et al., 2000c), but further data are required for these preliminary observations to be confirmed and their significance clarified.
Conclusions
Based on findings in the literature, OLP and sialadenitis may be significantly associated with HCV infection, and the virus may be involved in the pathogenesis of both diseases, probably via an immunological pathway still to be defined. Why some patients tolerate HCV in the oral mucosa without developing histological lesions, and why others will develop OLP or sialadenitis apparently independently of the local viral load is, at present, unknown. Parotid lymphoma may arise in patients with Sjögren-like sialadenitis with mainly type II cryoglobulinemia, whereas the link between oral carcinoma and HCV is weak and possibly influenced by the presence of liver cirrhosis. Little attention has been paid to the prevalence and the effect of sialadenitis-induced hyposalivation on oral health in HCV-infected patients and to the variable effect of IFN-α (with and without ribavirin) therapy on oral tissues. Because of the high global prevalence of HCV infection and the potentially great influence that oral diseases possibly linked to this virus may have on the quality of life of millions of patients, further research is needed, to explore the oral condition of patients with HCV infection.
Footnotes
Acknowledgements
We would like to thank Dr. Francesco Negro, Department of Gastroenterology and Hepatology and of Clinical Pathology, University Hospital, Geneva, Switzerland, for his helpful comments. This work was supported by the M.U.R.S.T. (ex quota 60%) and the Department of Biomedical Sciences and Human Oncology, University of Turin.