A Case-Control Study on a Novel DNA Virus (TT Virus) Infection and Hepatocellular Carcinoma

Transcription

1 A Case-Control Study on a Novel DNA Virus (TT Virus) Infection and Hepatocellular Carcinoma ALESSANDRO TAGGER, 1 FRANCESCO DONATO, 2 MARIA LISA RIBERO, 3 GIORGIO BINELLI, 4 UMBERTO GELATTI, 2 GIUSEPPE PORTERA, 1 ALBERTO ALBERTINI, 2 MICHELE FASOLA, 1 ROBERTA CHIESA, 2 GIUSEPPE NARDI, 2 AND THE BRESCIA HCC STUDY* We performed a case-control study to evaluate the association of a new human DNA virus named TT virus (TTV) with hepatocellular carcinoma (HCC). We recruited 174 subjects hospitalized for HCC (84% males; mean age: 64 years) and 118 patients hospitalized for non-liver diseases in Brescia, northern Italy, as controls (94% males; mean age: 66 years). TTV DNA was found in serum by polymerase chain reaction (PCR) in 26 cases (15%) and 11 controls (9.3%) (P G.1). TTV group 2 infection was identified in 16 cases (61.5%) and 4 controls (36.4%) (P G.1) using a type-specific PCR method. Sequence analysis of 222 nt of TTV DNA demonstrated that the remaining 10 cases and 7 controls were all infected by group 1. The odds ratio (OR) for TTV-DNA positivity, adjusted for demographic variables, hepatitis B surface antigen (HBsAg), hepatitis C virus (HCV) RNA, and heavy alcohol intake was 1.8 (95% CI: ; P G.1). The OR did not change when the analysis was restricted to 14 HCC cases and 56 controls who were negative for each known risk factor for HCC (OR 1.7; 95% CI: ). TTV-DNA positivity was not associated with transfusion history. The prevalence of TTV DNA was higher among HCC cases positive for HBsAg (10 of 38 [26.3%]) than among those positive for HCV RNA (8 of 62 [12.9%]) or negative for hepatitis B virus (HBV), HCV, and hepatitis G virus (HGV) infections (5 of 62 [8.1%]) (P.02). This study does not support the hypothesis of an association between TTV infection and HCC. (HEPATOLOGY 1999;30: ) Abbreviations: TTV, TT virus; HBV, hepatitis B virus; HCV, hepatitis C virus; HCC, hepatocellular carcinoma; HGV, hepatitis G virus; PCR, polymerase chain reaction; OR, odds ratio; HBsAg: hepatitis B surface antigen. From the 1 Istituto di Virologia and 3 Istituto di Igiene dell Università di Milano, 2 Cattedra di Igiene dell Università di Brescia, and 4 Dipartimento di Scienze Fisiologiche, Biochimiche e Cellulari dell Universitàdi Sassari, Italy Received November 20, 1998; accepted April 13, *Participants in the Brescia HCC Study: L. Bettini and M. G. De Tavonatti (I Medicina), G. Pelizzari (II Medicina), E. Radaeli and L. Biasi (III Medicina), R. Farfaglia (II Chirurgia), M. Puoti (II Malattie Infettive), M. Favret (II Anatomia Patologica) and F. Bonetti (I Anatomia Patologica), C. Leali (Dermatologia) at the Spedali Civili di Brescia; N. Portolani and M. Ronconi (Cattedra di Chirurgia Generale II, Università di Brescia); M. Ghirardi (Cattedra di Chirurgia Generale I, Università di Brescia); A. Salmi and G. Lanzani (Medicina), A. Savio (Anatomia Patologica), and M. Garatti (Chirurgia) at the Ospedale S. Orsola di Brescia; C. Scarcella (ASL di Brescia). The sequences reported in this article have been deposited in GenBank, accession numbers: AF AF Address reprint requests to: Francesco Donato. M.D., Cattedra di Igiene, Università di BresciaVia Valsabbina Brescia, Italy. donato@med.unibs.it; fax: Copyright 1999 by the American Association for the Study of Liver Diseases /99/ $3.00/0 A new unenveloped DNA virus that resembles viruses of the Parvoviridae family was isolated recently from the serum of a patient with posttransfusion hepatitis of unknown etiology, using representational difference analysis, and designated TT virus (TTV) after the patient s initials. 1 TTV was successfully detected in serum and liver tissues of patients with acute and chronic liver diseases. 1-3 TTV-associated posttransfusion hepatitis was found among patients undergoing surgery with blood transfusion. 4 TTV contamination was found in a high proportion of blood products. 5 Use of blood products was associated with an increased risk of TTV infection, 3,6,7 suggesting a parenteral route of transmission. Furthermore, excretion of TTV in feces was demonstrated, suggesting a fecal-oral route. 8 It is still unknown whether TTV can determine liver disease in humans. TTV DNA was found in patients with various liver diseases including cryptogenic fulminant hepatitis failure. 1-3,5,6 Titers of TTV DNA in serum were found to be closely related with aminotransferase levels in patients with posttransfusion non A-G hepatitis, 1 and TTV DNA was detected in liver tissues from all the hepatitis patients with TTV DNA in serum, 2 thus suggesting that the virus can cause acute or chronic liver disease. Most cross-sectional studies found higher prevalence of TTV infection in subjects with liver disease with respect to healthy persons. 2,3,6,9,10 Four TTV subgroups have been identified in Japanese patients and designated G1a, G1b, G2a, and G2b, and primers capable of amplifying TTV DNA of any genotype have been developed to evaluate the association of some of them with the severity of the liver disease. 2 At present, most information on TTV prevalence comes from case series, usually small in size, in which selection bias cannot be excluded. The case-control study design is appropriate for investigating whether an association between TTV infection and liver disease really exists when also taking into account known determinants of liver disease, such as hepatitis B (HBV) and C viruses (HCV) and alcohol intake. Furthermore, the association of TTV subgroups with liver disease can also be assessed using this study design. To this end, we performed a hospital-based case-control study in the province of Brescia, an area in northern Italy with one of the highest incidence rates of hepatocellular carcinoma (HCC) in Europe PATIENTS AND METHODS This research is part of a large case-control study performed to investigate the role of viral and nonviral agents in determining HCC. The design of the study and some results regarding the role of HBV, HCV and its genotypes, alcohol drinking, and GB virus/hepatitis G virus (HGV infection) have been reported in detail in previous

2 HEPATOLOGY Vol. 30, No. 1, 1999 TAGGER ET AL. 295 articles. 12,13 Briefly, 174 subjects admitted to the 2 main hospitals in the province of Brescia with first diagnosis of HCC between January 1995 and July 1996 were enrolled as cases. As controls, we examined 118 subjects not affected with liver diseases or neoplasms and admitted to the departments of Ophthalmology, Dermatology, Urology, Cardiology, and Internal Medicine in the same hospitals to which the cases were admitted. The research was approved by local Ethics Committees, and the patients gave their informed consent to participate. All subjects were interviewed on their past alcohol consumption, and heavy alcohol intake was defined as having drunk more than 80 g of ethanol a day for at least 10 years. A 10-mL blood sample was taken from all subjects, and the serum was separated and stored at 80 C. Serological markers of current or past infection by HBV, HCV, and HGV were investigated as described previously. 12,13 For detection of TTV DNA, nucleic acids were obtained from 100 µl of serum by treatment with proteinase K and sodium dodecyl sulfate, followed by extraction with phenol and chloroform as described by Nishizawa et al. 1 Nucleic acids were dissolved in 20 µl of Tris-HCl buffer (10 mmol/l, ph 8.0) supplemented with 1 mmol/l ethylenediaminetetraacetic acid, heated at 95 C for 15 minutes, and cooled immediately on ice. TTV DNA was determined in a half portion of the extract by polymerase chain reaction (PCR) with nested primers, conserved between the variants of TTV described to date. 1,2,8,14,15 Primer sequences were derived from the published gene sequences after multiple sequence alignment using the Pile Up program of GCG (Fig. 1). The first round of PCR used primers T50 (5 -ATTCACAGA- CAGAGGAGAAGG-3, nt 1896 to 1916, and TR430 (5 -GACTGT- GCTAAAGCCTCTAG-3, nt 2276 to 2257). Amplification conditions were 94 C for 60 seconds, 55 C for 60 seconds, and 72 C for 90 seconds, for 30 cycles. Five microliters of amplified DNA was subjected to a second round of PCR with primers T72 (5 - AACATGCTGTGGATAGACTGG-3, nt 1918 to 1938) and TR341 (5 -TCCTGGCATTTTACCATTTCCAAAG-3, nt 2187 to 2163) for 30 cycles as before. The PCR product (224 bp of length) was analyzed by agarose gel electrophoresis and DNA stained with ethidium bromide. TTV infection of group 2 (G2) was identified by a second round of PCR with primers T72 and type-specific antisense primer TR242 (5 -GATAACACATCTRCAGTTGTG-3, nt 2088 to 2068, R G/A). Type-specific primer TR 242 was designed after alignment of TTV G2 sequences collected worldwide 1,2,8,14-17 (Fig.2). Its specificity was checked by direct sequence analysis of nested PCR products of 224 bp (nt 1939 to 2162). PCR samples were gel-purified and sequenced on an Applied Biosystems ABI 373 automated sequencer using the PRISM dye terminator cycle sequencing kit (PE Applied Biosystems, Foster City, CA). Sequences were analyzed using the Sequencer 3.0 analysis program (Gene Codes Corporation, Ann Arbor, MI). For phylogenetic analysis, 23 sequences, 222 nt long (the last two 3 nt were removed), taken from 13 HCC cases (BsHCC) and 10 controls (BsCO), were compared together with 44 TTV sequences of the same length, sampled from GenBank. 1-2,14-17 The whole set of 67 sequences was resampled by bootstrapping 1,000 times and then analyzed using the PHYLIP, v.3.5c, package. 18 Sequence divergence was estimated by the Kimura 2-parameter method, with a transition-transversion ratio of 2:0. Phylogenetic analysis of the TTV sequences was performed using UPGMA (unweighted pair group method with arithmetic mean) of PHYLIP. The same analysis was also performed by the Neighbour-Joining method, with identical results. The final unrooted tree was obtained using the CONSENSE program. Comparison between proportions was performed using the 2 test and the exact methods when necessary. The odds ratios (ORs) and their 95% CIs were computed as relative risk estimates by unconditional logistic regression analysis using the maximum-likelihood method. 19 The association of TTV with HCC was evaluated after adjusting for possible confounding by demographic variables, namely sex, age, and area of residence (city of Brescia vs. rest of the province), and by hepatitis B surface antigen (HBsAg) and HCV- RNA seropositivity and heavy alcohol intake. As a result of the small number of cases and controls positive for TTV DNA, both asymptotic and exact methods were used. Because the OR estimates were similar using the 2 methods, only those based on asymptotic methods are shown. All the statistical tests were performed using the value of 0.05 to reject the null hypothesis, although values between 0.05 and 0.1 are reported. The analyses were performed using the BBMDP/Dynamic computer program. 20 RESULTS A total of 174 HCC cases and 118 controls were included in the analysis. Their distribution according to sex, age, and residence is shown in Table 1. The mean ages were 63.8 years (SD: 8.0) for cases and 65.7 years (SD: 8.1) for controls. TTV DNA was found in 15% (26 of 174) of HCC cases and 9.3% (11 of 118) of controls (P.1). No differences were found in TTV prevalence by sex, age, or residence in both cases and controls. TTV group 2 was identified in 16 of the 26 cases (61.5%) and 4 of the 11 controls (36.4%) (P.1) using FIG. 1. Nucleotide sequences and positions of the primers published and those used in this study, according to the numeration of TTV prototype sequence TA278. A total of 15 TTV isolates, including N22 clone 1 and 2 complete coding sequences from Japan (TA278) 2 and China (CHN1) of subtype 1a, are shown for comparative purposes. 14,15

3 296 TAGGER ET AL. HEPATOLOGY July 1999 FIG. 2. Nucleotide sequence and position of TTV group 2 specific antisense primer (nt ), according to the sequence of TTV prototype isolate (TA278) and in comparison with 34 sequences taken from GenBank 1,2,8,14-17 and 6 sequences of this study (BsHCC76, BsCO1551, BsHCC37, BsCO122, BsCO77, BsHCC30). Attribution of the TTV group and subgroup is given in parentheses. a type-specific PCR method. The sequence analysis was performed in 3 cases and 3 controls among those positive for group 2 and in all HCC cases (n 10) and controls (n 7) among the others. All the 6 subjects with group 2 TTV infection were confirmed by sequence analysis as G2 isolates. The other 17 subjects were all infected with TTV group 1: 6 had TTV subgroup G1a and 11 subgroup G1b isolates. The phylogenetic tree obtained from the UPGMA grouping of the 1,000 pairwise distance matrices is shown in Fig. 3. Both HCC cases and controls of TTV group 1 belong to clusters different from the Japanese isolates both in the G1a and the G1b subgroups, as indicated by significant branches of the tree after bootstrap analysis. HCC cases and controls of TTV group 2, with the exception of BsHCC30, cluster together with G2a, G2b, and G2c subgroups. The prevalence of TTV among total subjects and according to positivity for HBsAg, HCV, and HGV RNA is shown in Table 2. Among HCC cases, the lowest prevalence was found among patients negative for HBsAg, HCV, and HGV RNA (8.1%) as compared with those positive for HBsAg only (26.3%) or HCV RNA only (16.7%) ( 2 test for comparison between the three groups: P.02). Among the controls, 9.9% of subjects negative for HBV, HCV, and HGV infection were infected with TTV; among the others, 1 of the 2 with HGV RNA and none of those with positivity for HBsAg or HCV RNA were infected with TTV. The OR for TTV DNA adjusted for age, sex, residence, HBsAg, HCV RNA, and alcohol intake was 1.8 (95% CI: ; P.1). When the 2 TTV groups were considered separately, the OR for group 1 positivity (OR 1.85, 95% CI: ) was almost exactly the same as the OR for group 2 positivity (OR 1.73, 95% CI: ). The ORs for the other risk factors investigated, adjusted for each other and for demographic variables, were: 9.0 (95% CI: ) for HBsAg, 29.4 ( ) for HCV RNA, and 6.2 ( ) for heavy alcohol intake. Forty-eight of the 62 HCC cases and 45 of the 101 controls negative for HBsAg, HCV RNA, and HGV RNA had a history of heavy alcohol intake. Therefore, 14 cases ( cryptogenic HCC cases) and 56 controls were negative for each risk factor investigated and also for hemochromatosis and other known causes of liver disease. Among them, 1 case (7.1%) and 5 controls (8.9%) were positive for TTV DNA (P.1). The OR for TTV DNA among subjects negative for HBsAg, HCV RNA, HGV RNA, and heavy alcohol intake, computed using the exact logistic regression methods, and adjusted for age, sex, and residence, was 1.7 (95% CI: ; P 1). Transaminase mean serum levels were not associated with TTV infection among controls. Alanine transaminase and aspartate transaminase values were 22.4 and 22.7 IU/L among TTV-DNA positive and 27.1 and 25.9 IU/L among TTV-DNA negative subjects, respectively. Furthermore, no differences in mean transaminase values were found between TTV-DNA positive and negative subjects when considering only subjects negative for HBsAg, HCV RNA, and heavy alcohol intake. Transfusion history was claimed by 18.3% of total subjects (cases: 23.3%; controls: 14.4%). No association was found between TTV-DNA positivity and transfusion when considering cases and controls together: 22.2% of subjects with, and 17.9% of those without, TTV DNA had transfusion history (P.1). No differences were observed when considering cases and controls separately: among HCC cases: 28.6% of TABLE 1. Demographic Characteristics of Cases and Controls Cases Controls Demographic Characteristics No. (%) No. (%) Sex Male 146 (83.9) 111 (94.1) Female 28 (16.1) 7 (5.9) Age (yr) (4.0) 2 (1.7) (20.7) 20 (16.9) (47.7) 53 (44.9) (27.6) 43 (36.4) Residence City of Brescia 77 (44.3) 74 (62.7) Rest of the province 97 (55.7) 44 (37.8) Total 174 (100) 118 (100)

4 HEPATOLOGY Vol. 30, No. 1, 1999 TAGGER ET AL. 297 FIG. 3. Unrooted phylogenetic tree (UPGMA) of a 222-nt region (positions , according to the sequence of TTV prototype isolate TA278) from 67 isolates. Bootstrap values 800/1,000 are reported at the nodes. Accession numbers for the 44 sequences taken from GenBank are: AB (TA278), AB (G977801), AB (G103301), AB (G102001), AB (G104901), AB (G105001), AB (G88801), AB (TY96117), AB (COL22), AB (COL7), AB (COL15), AB (N22), AB (PT3), AB (TX011), AB (TS003), AB (NA004), AB (CMR12), AB (CMR809), AB (JaM18), AB ( JaM21), AB ( JaM28), AB (JaHCC55), AB ( JaHCC59), AB ( JaBD28), AB (JaBD43), AB ( JaBD98), AB ( JaNBNC10), AF (BT/38), AF (B/1), AF (BT3), AF (L/2), AF (BT/ 13), AF (P/1), AF (G18), AF (G23), AF (CHN1), AF (Alb16), AF (Bhu96), AF (Ger926), AF (Ind37), AF (Ita18), AF (Ita23), AF (Ita32), AF (Dan5); and for the 23 sequences originated from our HCC cases (BsHCCX) and controls (BsCOX) are: AF AF subjects with and 22.9% of those without TTV infection had a history of transfusion (P.1), the corresponding proportions among controls being 18.2% and 14.0%, respectively (P.1). Transfusion history was found to be associated with HCV infection (31.3% and 15.9% of HCV-RNA positive and negative subjects, respectively; P.04) and not with either TABLE 2. Distribution of HCC Cases and Controls According to TTV-DNA Positivity and HBsAg, HCV-RNA, and HGV-RNA Status HCC Cases Controls HBsAg, HCV-RNA, and HGV-RNA Status TTV DNA / total (%) TTV DNA / total (%) Negative for each 5/62 (8.1) 10/101 (9.9) HBsAg only 10/38 (26.3) 0/8 HCV RNA only 8/62 (12.9) 0/6 HGV RNA only 1/6 (16.7) 1/2 (50) HBsAg, HCV RNA, HGV RNA 1/3 (33.3) 0/0 HBsAg, HCV RNA, HGV RNA 1/1 (100) 0/1 HBsAg, HCV RNA, HGV RNA 0/2 0/0 Total 26/174 (15.0) 11/118 (9.3) current HBV infection (22.3% and 17.4% among HBsAgpositive and -negative subjects, respectively; P.1) or past HBV infection (21.3% and 13.8% among anti-hbc positive and negative subjects, respectively; P.1). DISCUSSION Most cases of chronic hepatitis, cirrhosis, and HCC in developed countries are caused by HBV or HCV infections and heavy alcohol intake. A relatively small proportion of liver diseases is of unknown etiology. The recently discovered HGV seems to have no actual role in causing acute or chronic liver disease, 21 whereas the role of the more recently discovered TTV is still to be defined. We found that 15% of the HCC cases as compared with 9.3% of the controls had TTV DNA in their serum. Among HCC cases, TTV-DNA positivity was less common than HBsAg and HCV-RNA positivities, whereas among controls, it was equal to or more common than the other 2 viral marker positivities. The prevalence of TTV among controls in our study is below the 10% value found among blood donors and

5 298 TAGGER ET AL. HEPATOLOGY July 1999 healthy controls in Europe and the United States (lowendemicity area), 5,6,9 whereas a 10% to 40% prevalence has been found in blood donors and healthy subjects in Asia (intermediate-endemicity area), 2,7,10,22 and a prevalence up to 80% has been observed in rural people living in Africa, Australia, and Central and South America (high-endemicity area). 22 A single study on 100 healthy adults from Japan 23 reports a 92% prevalence of TTV infection using a PCR with newly designed primer pair (T801 and T935) from a 5 end region of the prototype TTV isolate TA278, compared with the 23% prevalence obtained by PCR with the published NG061 and NG063 primer pair. 2 However, we detected no additional TTV-DNA positive subjects among 50 HCC cases and 50 controls tested using the single-step 55-cycle PCR with T801 and T935 primers, as suggested by the authors. 23 In this study, the proportion of subjects positive for TTV-DNA infection was only slightly higher among HCC cases than controls, and from that, a weak association (OR 1.8) was found between HCC and TTV infection. However, though this OR estimate was not statistically significant (P.1), the possibility of an association between TTV and HCC cannot definitely be excluded on the basis of our data. However, should this association exist, it would seem to be weaker than that found for the well-recognized causes of HCC. Similarly, no significant association was found between TTV infection and HCC among subjects negative for HBsAg, HCV RNA, and history of heavy alcohol intake. Overall, these findings are similar to those from other studies performed among subjects with chronic liver disease or HCC, including cryptogenic liver disease, and healthy controls. 3,6 Among the total HCC cases recruited in this study, the vast majority had a well-defined risk factor for liver disease, and only 14 of 174 cases (8.0%) were negative for each factor, 1 of which (7.1%) being positive for TTV DNA. The prevalence of TTV among cryptogenic HCC cases in our study is similar to the 8.7% found among subjects with cryptogenic chronic liver disease in Spain 9 and lower than the 39% to 47% observed among patients with non B-C chronic liver disease in Japan. 2,24 These findings suggest that TTV may only play a limited role in this epidemiological scenario, should it cause liver disease. A bias in our results regarding the association between HCC and TTV is unlikely, because the HCC series was representative of the total incident cases that occurred in this area during the study period, and the control series was representative of the general population in the same area, as discussed previously. 12,13 In our case-control study, we evaluated the presence of TTV DNA in serum at first diagnosis of the disease (cases) or at recruitment (controls). However, the duration of TTV- DNA persistence in patients with chronic liver disease is still an unresolved question. To investigate this issue, one of us performed a prospective study among 15 patients undergoing orthotopic liver transplantation for HCV-related HCC who had TTV infection: 2 had recurrence of TTV infection and 13 acquired de novo TTV infection after orthotopic liver transplantation. 25 Among the patients with TTV infection, 9 showed persistence of TTV throughout the post orthotopic liver transplantation follow-up of 40 to 75 months, whereas the remaining 6 cleared the virus either spontaneously or after treatment with interferon or a combination of interferon and ribavirin in 9 to 41 months. We observed similar findings in other patients undergoing OLT for HBV-related (n 2) or not B-C related HCC (n 3). These findings are in agreement with those of other studies showing a 2- to 6-year persistence of TTV in serum among patients with chronic HCV infection 26 and among patients with hemophilia 7 or -thalassemia. 27 The extremely high prevalence of TTV viremia found in indigenous rural populations also suggests that the infection can be persistent. 22 We found TTV group 2 infection in 61.5% of the cases and 36.4% of the controls, and group 1 in the remaining subjects, with no significant difference between cases and controls, suggesting that there is probably no association between TTV group and HCC in our population. This TTV group distribution is similar to that found among German patients with chronic liver disease 16 and different from that observed in various populations in Japan, where group 2 was identified in a minority of subjects. 2 Overall, the prevalence of groups 1 and 2 seems very high worldwide. 17 However, differences regarding the distribution of TTV groups appear evident: phylogenetic analysis demonstrates a significant branching of both G1a and G1b types, which split HCC cases and controls into 2 groups, each comprising isolates of distinct geographic origin, one European and the other Asiatic. TTV group 2 appears divided into 2 subtypes, the first including subgroups G2a and G2b, 2 and the second comprised of subgroup 2c. 8,16 Eleven samples, including 1 of our HCC cases, BsHCC30, form heterogeneous groups, including the TTV sequences formerly belonging to groups G2, G4, G5, and G6. 15,17 Some considerations should be made on the possible route of transmission of TTV. An association between HCV infection and the presence of TTV DNA was found in one study. 28 On the contrary, we found a higher TTV-DNA prevalence among cases positive for HBsAg (26.3%) than those positive for HCV RNA (12.9%) or negative for both (8.1%). The lack of association between TTV DNA and HCV RNA suggests that the parenteral route is not as fundamental as it is for HCV infection. In effect, a history of transfusion was not associated with TTV positivity, nor was it associated with HBsAg or anti-hbc positivity, whereas it was related to HCV-RNA positivity. Taken together, these findings suggest that TTV shares with HBV routes of transmission other than transfusion. Moreover, the frequency of TTV viremia in individuals at high risk for sexual and parenteral infection (intravenous drug users, prostitutes, and homosexual men) was low and not significantly different from that found among low-risk controls, in a study in Edinburgh. 29 Nonparenteral or community-acquired ways of transmission of TTV have been hypothesized on the basis of the increase in prevalence of TTV-DNA viremia by age in the general population. 29,30 The hypothesis of transmission of TTV through personal contact is also supported by the finding of TTV fecal excretion from infected individuals. 8 In conclusion, this study does not support the hypothesis of an association between TTV infection and HCC. However, the role of TTV in determining HCC, should it have any, seems much less important than that of established risk factors such as HBV, HCV, and heavy alcohol intake. Acknowledgment: The authors are grateful to the staff in the Departments of Ophthalmology, Dermatology, Urology, Cardiology, and Internal Medicine at the Spedali Civili and S. Orsola hospitals in Brescia for their help in recruiting the controls. The cooperation of the Centre for the Study of Surgical Liver Diseases of the University of Brescia, funded

6 HEPATOLOGY Vol. 30, No. 1, 1999 TAGGER ET AL. 299 by Ente Universitario Lombardia Orientale (EULO), is also appreciated. REFERENCES 1. Nishizawa T, Okamoto H, Konishi K, Yoshizawa H, Miyakawa Y, Mayumi M. A novel DNA virus (TTV) associated with elevated transaminase levels in posttransfusion hepatitis of unknown etiology. Biochem Biophys Res Commun 1997;241: Okamoto H, Nishizawa T, Kato N, Ukita M, Ikeda H, Iizuka H, Miyakawa Y, et al. Molecular cloning and characterization of a novel DNA virus (TTV) associated with posttransfusion hepatitis of unknown etiology. Hepatol Res 1998;10: Naoumov NV, Petrova EP, Thomas MG, Williams R. Presence of a newly described human DNA virus (TTV) in patients with liver disease. Lancet 1998;352: Fujiwara T, Iwata A, Iizuka H, Tanaka T, Okamoto H. Transfusion transmitted virus. Lancet 1998;352: Simmonds P, Davidson F, Lycett C, Prescott LE, MacDonald DM, Ellender J, Yap PL, et al. Detection of a novel DNA virus (TTV) in blood donors and blood products. Lancet 1998;352: Charlton M, Adjei P, Poterucha J, Zein N, Moore B, Therneau T, Krom R, et al. TT-Virus infection in North American blood donors, patients with fulminant hepatic failure, and cryptogenic cirrhosis. HEPATOLOGY 1998; 28: Sumazaki R, Yamada-Osaki M, Kajiwara Y, Shirahata A, Matsui A. Transfusion transmitted virus. Lancet 1998;352: Okamoto H, Akahane Y, Ukita M, Fukuda M, Tsuda F, Miyakawa Y, Mayumi M. Fecal excretion of a nonenveloped DNA virus (TTV) associated with posttransfusion non-a-g hepatitis. J Med Virol 1998;56: Gimenz-Barcons M, Ampurdanès S, Soler M, Forn X, Soguero C, Guilera M, Sanchez A, et al. Prevalence of TT virus infection among patients with liver diseases [Abstract]. HEPATOLOGY 1998;28:368A. 10. Tanaka H, Okamoto H, Luengrojanakul P, Chainuvati T, Tsuda F, Tanaka T, Miyakawa Y, et al. Infection with an unenveloped DNA virus (TTV) associated with posttransfusion non-a to G hepatitis in hepatitis patients and healthy blood donors in Thailand. J Med Virol 1998;56: Chiesa R, Donato F, Tomasoni S, Portolani N, Favret M, Nardi G. Primary liver cancer in a high-incidence area in North Italy: etiological hypothesis arising from routinely collected data. Eur J Epidemiol 1995;11: Donato F, Tagger A, Chiesa R, Ribero ML, Tomasoni V, Fasola M, Gelatti U, et al. Hepatitis B and C virus infection, alcohol drinking and hepatocellular carcinoma: a case-control study in Italy. HEPATOLOGY 1997;26: Tagger A, Donato F, Ribero ML, Chiesa R, Tomasoni V, Portera G, Gelatti U, et al. A case-control study on GB virus-c/hepatitis G virus infection and hepatocellular carcinoma. HEPATOLOGY 1997;26: Takahashi K, Ohta Y, Mishiro S. Partial 2.4-kb sequences of TT virus (TTV) genome from eight Japanese isolates: diagnostic and phylogenetic implications. Hepatol Res 1998;12: Viazov S, Ross RS, Niel C, de Oliveira JM, Varenholz C, da Villa G, Roggendorf M. Sequence variability in the putative coding region of TT virus: evidence for two rather than several major types. J Gen Virol 1998;79: Höhne M, Berg T, Müller R, Schreier E. Detection of sequences of TT virus, a novel DNA virus, in German patients. J Gen Virol 1998;79: Tanaka Y, Mizokami M, Orito E, Ohno T, Nakano T, Kato T, et al. New genotypes of TT virus (TTV) and a genotyping assay based on restriction fragment length polymorphism. FEBS Lett 1998;437: Felsenstein J. PHYLIP (Phylogeny Inference Package) version 3.5c. Department of Genetics, University of Washington, Seattle, Hosmer DW, Lemeshow S. Applied logistic regression. New York: John Wiley and Sons, Dixon WJ, Ed. BMDP Statistical Software Manual. Los Angeles: University of California Press, Theodore D, Lemon SM. GB virus C, hepatitis G virus, or human orphan flavivirus? HEPATOLOGY 1997;25: Prescott LE, Simmonds P. Global distribution of transfusion-transmitted virus. N Engl J Med 1998;339: Takahashi K, Hoshino H, Ohta Y, Yoshida N, Mishiro S. Very high prevalence of TT virus (TTV) infection in general population of Japan revealed by a new set of PCR primers. Hepatol Res 1998;12: Ikeda H, Takasu M, Inoue K, Okamoto H, Miyakawa Y, Mayumi M. Infection with an unenveloped DNA virus (TTV) in patients with acute or chronic liver disease of unknown etiology and in those positive for hepatitis C virus RNA. J Hepatol 1999;30: Tagger A. TTV infection in OLT recipients [Abstract]. Second Europe and Japan Meeting on Hepatitis Research. Venice, June 24-25, Irving WL, Ball JK, Berridge S, Curran R, Grabowska AM, Jameson C, Thomson BJ, et al. Chronic infection with TT virus: frequency, effect on chronic HCV infection, and virus evolution. [Abstract]. HEPATOLOGY 1998;28:362A. 27. Prati D, Lin YH, De Mattei C, Liu JK, Farma E, Ramaswamy L, Zanella A, et al. A prospective study on a TT virus infection in transfusiondependent patients with beta-thalassemia. Blood 1999;93: Gerolami V, Halfon P, Chambost H, Thuret I, Halimi G. Transfusion transmitted virus. Lancet 1998;352: MacDonald DM, Scott GR, Clutterbuck D, Simmonds P. Infrequent detection of TT virus infection in intravenous drug users, prostitutes, and homosexual men. J Infect Dis 1999;179: Yamada-Osaki M, Sumazaki R, Noguchi E, Shibasaki M, Matsui A. Transfusion transmitted virus. Lancet 1998;352: