Hepatitis B and C Viruses, Human T-Cell Lymphotropic Virus Types I and II, and Leukemias: A Case-Control

The relationship between acute myeloid leukemia (AML), acute lymphocytic leukemia, chronic myeloid leukemia (CML), and refractory anemia with excess of blasts (RAEB) and antibodies to human T-cell lymphotropic virus types I and II (HTLV-I and HTLV-II), and hepatitis B virus and hepatitis C virus (HCV) was investigated in a multicenter case-control study. There were 431 cases enrolled in the study at the time of diagnosis of hematological malignancies, and 862 controls ages 15 years or older were recruited in three hospitals. Antibodies to HTLV-I and HTLV-II, antibody to HCV, hepatitis B surface antigen, and antibody to hepatitis B core antigen were assayed. All cases and controls were negative for HTLV-I antibodies; one case (1 of 431; 0.2%), and one control (1 of 862; 0.1 %) were found positive for HTLV-II antibodies. A nonsignificant excess of risk for hepatitis B surface antigen was present among RAEB cases (odds ratio, 2.40; 95% confidence interval, 0.46-12), CML Received 8/7/95: revised I 1/30/95; accepted I 1/30/95. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. I Preliminary data from this study were presented at the 86th Annual Meeting of American Association for Cancer Research, held in Toronto. Canada, in March 1995 (Abstract 1666). 2 This study was partially supported by the Italian Ministry of Health. Grant 5(X).4IRSC/70.lS/T/1886: by the Consiglio Nazionale delle Ricerche Bilateral Project Italia-USA. Grants 87.00186.04. 88.00600.04. and 89.02995.04; and by Progetto Finalizzato ACRO Consiglio Nazionale delle Ricerche, Grant 92.022 19.39. ‘ To whom requests for reprints should be addressed, at Istituto Superiore di Sanit#{224}. Laboratorio di Epidemiologia e Biostatistica. Reparto di Epidemiologia Clinica. Viale Regina Elena 299. 00161 Rome. Italy. Phone: 06-4450607; Fax: 06-4456686. 4 The members of the Italian Study group are as follows: M. Trematerra and U. Villano )Laboratorio di Virologia, Istituto Superiore di Sanit#{224}. Rome, Italy); L. Femgno. G. lantosca, and R. Da Cas (Istituto Superiore di Sanit#{224}, Rome, Italy); C. Bemasconi (Divisione di Ematologia. Istituto San Matteo di Pavia, Pavia, Italy); S. Merantc (Divisione di Ematologia, Pavia, Italy); S. Tura and R. Colombini )lstituto di Ematologia Seragnoli. Bologna. Italy); and A. Capobianchi. M. Rolli. P. P. Petasecca Donati. and S. Santilli (Istituto di Ematologia. Rome. Italy. (odds ratio, 2.70; 95% CI, 0.86-8.43), and between antibody to hepatitis B core antigen and AML (odds ratio, 1.40; 95% CI, 0.93-2.10). A weak, nonsignificant association was present between AML, acute lymphocytic leukemia, RAEB, and antibody to HCV. These preliminary results suggest a possible association (elevated odds ratios) between hepatitis B virus, AML, RAEB, and CML. However, because all confidence intervals overlapped the null value, these findings need to be confirmed in larger case-control studies. Introduction Viruses have long been considered potential risk factors for leukemias ( 1 ). Animal experiments have established causal relationships between viruses and leukemia and lymphoma (2, 3). Human T-cell HTLV-l5 has been implicated in the etiology of ATL in many parts of the world, particularly in the Caribbean (4). There is close agreement worldwide between areas of HTLV-I endemicity and ATL occurrence. Recently, Manns et a!. (5), in a case-control study performed in Jamaica and Trinidad and Tobago, found that HTLV-I was associated with non-Hodgkin’s lymphoma but not with other hematobogical malignancies (ALL and chronic lymphocytic leukemia or AML and CML). An association between HTLV-II and malignancy has been suggested by the isolation and detection of HTLV-II DNA in malignant cells from hairy-cell leukemia patients, and by its ability to transform T lymphocytes in vitro (6-8); however, HTLV-II has not yet been shown to be the cause of any disease (9). A causal role of hepatitis viruses, particularly non-Anon-B virus, has been hypothesized for aplastic anemia, and case reports suggest an association between non-A-non-B hepatitis and leukemias (10-12). Recent advances in the molecular biology of HBV have identified viral genome or proteins not only in hepatocytes but also in a number of extrahepatic sites, such as lymphoblastic cells, lymph nodes, and vascular elements in the liver ( 13). The permissiveness of these extrahepatic cells for viral replication might also promote tumor development in hematopoietic cells ( I 3). It has been shown that HCV is also a lymphotropic virus. Positive and negative stranded (replicative) forms of HCV have been observed in peripheral blood mononuclear cells of patients with chronic liver disease ( 14). Furthermore, recent reports describe the presence of HCV infection in patients presenting 5 The abbreviations used are: HTLV-I, lymphotropic virus type I; ATL. adult T-cell leukemia/lymphoma; ALL. acute lymphocytic leukemia; AML, acute myeloid leukemia; CML, chronic myeloid leukemia; RAEB, refractory anemia with excess of blasts; HBV, hepatitis B virus; HCV. hepatitis C virus: HBsAg. hepatitis B surface antigen; anti-HBc. antibody to hepatitis B core antigen; OR. odds ratio; CI, confidence interval. on October 29, 2017. © 1996 American Association for Cancer Research. cebp.aacrjournals.org Downloaded from 228 Short Communication: HBV, HCV, HTLV-I, HTLV-II, and Leukemias 1gM monocbonal gammopathies (15) and in patients affected by mixed cryoglobulinernia-related lymphoproliferative disorders (low-grade non-Hodgkin’s lymphomas; Ref. 16). These observations suggest that HCV infection may play a role in both liver disease and lymphoproliferative disorders. Within a multicenter case-control study of risk factors for acute leukemias ( I 7), we investigated the relationships between AML, ALL, CML, and RAEB; antibodies to HTLV-I and HTLV-II; and HBV and HCV. Materials and Methods This study was conducted between November 1 , 1986, and March, 3 1 , 1990, in Rome, Bologna and Pavia, Italy. Recruitment of Study Subjects. The method of recruitment of study subjects and data collection is described in detail elsewhere (17). Briefly, cases were 15 years or older with newly diagnosed AML, ALL, CML, or RAEB. Diagnostic criteria were based on the revised French-American-British classification of bone marrow aspirates for acute leukemias and RAEB, whereas diagnosis for CML was based on typical dinical and cytogenetic laboratory features. Controls were recruited in the region of the three hospitals during the study period among outpatients without hematobogical malignancies who were seen in the same hospitals at which cases had been identified. The control group was selected by taking the first five outpatients in Rome and the first three in Bologna and Pavia, seen on a random day each week. Controls having platelet disorders, leukocytosis, leukopenias, or monocbonal gammopathies of undetermined significance were excluded from the study because of possible shared risk factors with the case diseases. Subjects evaluated in the study hospitals for chronic hepatitis and hyperbilirubinemias were also excluded. A standard, precoded questionnaire was administered to both cases and controls to collect data on medical history and behavioral and environmental exposure. In particular, questions were included regarding iv. drug use, homosexual or bisexual activity, and receipt of blood products. Subjects who had received blood product transfusions (i.e. , blood and pooled plasma products) were excluded from the study. Laboratory Methods. All sera were stored at 20#{176}Cand subsequently tested. HBV markers, HBsAg, anti-HBc were tested, in serum, by immunoenzymatic assays (Abbott Laboratories, North Chicago, IL). Cutoff of positivity was calculated according to manufacturer’ s instructions. Anti-HCV prevalence was calculated using a secondgeneration ELISA (Ortho Diagnostic, Raritan, NJ). Repetitively, reactive sera were further tested by third-generation supplemental RIBA (RIBA III, Ortho Diagnostic), which includes structural (C22) and nonstructural (CbOO, C33, NS5) viral proteins. Sera were considered anti-HCV positive when two or more bands were shown by RIBA. Only 1 band (indeterminate result) was observed in 4.2 and 4.4% of anti-HCV ELISA reactive sera from cases and controls, respectively. No bands were detected in 18.3 and 6. 1 % of anti-HCV ELISA reactive sera from cases and controls, respectively. All RIBA indeterminate (3 cases and 5 controls) and nonreactive samples were excluded from analysis. Coded samples were screened using an HTLV-I viral lysatelHTLV-I/HTLV-II recombinant protein-based enzymeimmunoassay (HTLV-1/HTLV-II ELISA, Diagnostic Biotechnobogy, Singapore). Samples were confirmed as seropositive if antibodies against gag (p 19 or p 24) and env (gp 46) gene products were present by Western blot (HTLV blot 2.3, Diagnostic Biotechnology, Singapore). Anti-HTLV-I or HTLV-II reactive profiles were differentiated, based on reactivity to either HTLV-Ior HTLV-II-recombinant gp46 protein. Statistical Analysis. Adjusted ORs for age, sex, education, history of hospitalization, and area of residence and 95% CIs were calculated by polychotomous logistic regression ( 18). Age was included as a continuous variable in the logistic regression model.