Infection with human T-cell lymphotropic virus type I and II (HTLV-I/II) has been described among different populations worldwide(1,2). The first study of HTLV-I infection in Brazil was described by Kitagawa et al.(3). They reported the seroprevalence of antibodies to HTLV-I in Japanese immigrants living in this country for 50 years. This study disclosed that 10% of issei and nissei from the Okinawa islands were HTLV-I positive, and all of 51 immigrants from other parts of Japan were negative(3). Since then, several reports have described the occurrence of HTLV-I and HTLV-II infections among urban and isolated populations in Brazil. Most studies performed with blood donors after mandatory tests to screen HTLV-I/II antibodies revealed a different prevalence between Northern and Southern parts of the country, ranging from 0.17 to 1.8%(4-6). HTLV-II positivity was also found in Indian communities in the Amazon region and in intravenous drug abusers in the urban areas of Recife, São Paulo, and Rio de Janeiro(7-10).
In endemic areas, HTLV-I infection is associated with adult T-cell leukemia/lymphoma (ATLL), a myelopathy or tropical spastic paraparesis (HAM/TSP) diseases that affect sphincter tone and lower limbs, and other immune-mediated diseases such as uveitis and Sjögren syndrome(1,11,12). The mechanisms responsible for the development of ATLL or other HTLV-I-associated diseases are still the subject of research, but it is clear that this retrovirus produces lymphotropic and neurologic damage(13).
Although ATLL has been identified in Rio de Janeiro, epidemiologic data on this disease have been hampered by difficulties in the diagnosis. Some ATLL cases were confused with other lymphoproliferative disorders, particularly with cutaneous T-cell lymphoma or pure forms of T-cell lymphoma(12), because serologic tests for HTLV-I antibodies are not performed regularly in assessing mature lymphoproliferative diseases.
In this study, we determined the seroprevalence rate of HTLV-I in patients with hematologic and malignant conditions and in blood donors and patients exposed to multiple blood transfusions. Our data are representative and estimate the prevalence of HTLV-I among those with hematologic disorders and the occurrence of ATLL among those with lymphoid malignancies in Rio de Janeiro, Brazil.
MATERIAL AND METHODS
Informed consent was obtained from all blood donors and from all patients and their parents enrolled in this study. From July 1992 to September 1994, we recruited 2430 individuals for serologic analysis. They were tested for HTLV-I/II antibodies in two hospitals: the Cancer Hospital (CH-INCA) and Instituto Estadual de Hematologia do Rio de Janeiro (HEMORIO). Samples from hematologic patients were taken consecutively and with consent, as part of diagnostic investigations in both centers. Samples from cancer patients were taken at random in the outpatient department of CH-INCA. Blood donors' samples were taken during 1 week in the blood bank of HEMORIO before HTLV-I screening became mandatory in Brazil. The families of the HTLV-infected individuals were recruited with their consent. For each case we recorded age, sex, and ethnic origin, and from all patients, the clinical and demographic features were provided by the referring physicians. Those with hematologic diseases without previous blood transfusions were classified according to clinicopathologic criteria, as follows: T-cell leukemia and lymphoma (n = 152 cases), B-cell leukemia and lymphoma (n = 250 cases), myeloproliferative disorders (n = 67 cases), and Hodgkin's disease (n = 41 cases).
All individuals included as multitransfused patients had received at least 3 units of blood products before implementation of donor screening for HTLV-I/II antibodies in the blood banks. They had hemoglobinopathies(n = 215), hemophilia (n = 61), leukemia (n= 32), or severe aplastic anemia (n = 7) and had received mainly packed red blood cells and platelets when appropriate for their clinical conditions.
The diagnosis of hematologic disease was based on clinical, morphologic, and immunologic classification of the patients according to each group of disease and to the FAB criteria described elsewhere(14). The criteria for ethnic groups considered whites, the Brazilians of European descent, and nonwhites, all blacks and mulattoes, as well as mixed races of whites, Amerindian, or blacks. No person of Asian origin was included in this study.
Serum samples separated from whole clotted blood were frozen and stored at −20°C until testing. The samples were screened by enzyme-linked immunosorbent assay (ELISA, Cambridge Biotech Corporation, Worcester, MA, U.S.A.) and particle agglutination (Serodia Fujirebio, Japan) according to the manufacturer's instructions. Samples repeatedly reactive were also analyzed by Western blot (HTLV blot 2.3 Diagnostic Biotechnology Ltd., Singapore) that incorporates a recombinant envelope protein (rgp21) common to HTLV-I and HTLV-II and two polypeptides specific to HTLV-I(MTA-1) or HTLV-II (K55) to differentiate the two types of virus. A specimen was considered positive for HTLV-I and HTLV-II when reactive to gag (p24) and env (gp46 or rgp21). Viral type was defined as HTLV-I or HTLV-II when reactive to recombinant gp46-I or gp46-II, respectively. A sample was considered indeterminate when the reactivity did not meet the positive criteria.
Eighty-six specimens were specifically collected for polymerase chain reactions (PCR) analysis, specified as 10 samples with negative serology, 40 with positive, and 30 with indeterminate Western blot results. There were 34 cases of T-cell diseases, 39 samples from polytransfused patients, 10 from blood donors, and 3 from family members of the HTLV-infected individuals. PCR amplification was performed with primer pairs for pol (SK110/111) and for tax (SK43/44) regions. All samples were run in duplicate, and PCR products were analyzed by liquid hybridization according to Rios et al.(15). Twenty-four specimens of T-cell disease included in this study were described elsewhere(16).
Statistical analysis (Mantel-Haenszel chi-square test and Fisher's exact test) were performed on the EPIINFO version 5.01b from a database bank generated on DBase II plus software. The statistically significant level (p value) was set to 0.05, odds ratios (ORs), and 95% confidence intervals (CIs) for the comparison between T-cell diseases and other hematologic diseases. Age younger than 20 years, cancer and other hematologic diseases without exposure to blood transfusion, male sex. and white race were the reference categories for ORs(17).
We tested blood samples from 510 patients with hematologic diseases without previous exposure to blood transfusions, 351 exposed to multiple blood transfusions, 260 with cancer, 109 members of 36 families of HTLV-infected individuals, and 1200 blood donors. The demographic characteristics of the studied population and the overall results of HTLV tests are shown in Table 1. There were 1688 male and 742 female subjects, with a 2:1 male-female ratio in the total group, between the ages of 2 and 92 years, with 136 persons younger than 21 and 25 persons older than 71 years of age.
Among those with hematologic diseases without previous blood transfusions, 46 samples were positive for HTLV-I. Among those with T-cell diseases, 44 samples were HTLV-I positive, resulting in a high seroprevalence ratio of 28.9% (CI = 21.88 to 36.84) in this subgroup, and all patients fit the criteria of ATLL. The seroprevalence of HTLV-I antibodies in B-cell diseases was 0.4%(1 of 250). The one HTLV-I-positive B-cell malignancy was a 70-year-old white man. born in Rio de Janeiro, whose lymph node biopsy disclosed a small, well-differentiated lymphoma, and molecular analysis of lymphoid cells showed no HTLV-I proviral monoclonal integration.
None of those with myeloid leukemia had antibodies to HTLV-I. Among those with Hodgkin's disease, a 71-year-old white man was diagnosed with Hodgkin's disease with mixed cellularity and with clinically advanced disease (stage IVB). The serology tests performed during this study demonstrated HTLV-I antibodies, but unfortunately, other studies could not be performed to elucidate the cellular origin of his disease and a possible association to HTLV-I. He died during treatment of resistant disease.
Among multitransfused patients, 32 samples (9.1%) were typed as HTLV-I positive, 3 (0.9%) as HTLV-II positive, and 1 (0,2%) with double reactivity for HTLV-I and HTLV-II. In blood donors, a concordant result of ELISA and particle agglutination disclosed 10 of 1200 individuals with HTLV antibodies, but a seropositivity of 0.4% for HTLV-I antibodies (95% CI = 0.15 to 1.03) and 0.1% for HTLV-II was found after confirmatory testing with Western blot and PCR assays. HTLV-I antibodies were seen in 3 (1.1%) of 235 patients with cancer. They were a 47-year-old mulatto woman with breast cancer, a 27-year-old mulatto woman with parotid adenocarcinoma(treated by surgery and asymptomatic during the time of this study), and a 28-year-old man with anal carcinoma who had coinfection with HTLV-I and HIV.
In family members of HTLV-infected individuals, the HTLV-I seroprevalence ratio was 27.5% (95% CI = 6.65 to 13.07). There was increased HTLV-I infection with age, particularly for subjects 41 years and older(data not shown) and a relatively increased rate of HTLV-I infection in females (9.3% versus 3%; p < 0.001 in the unadjusted analysis and 0.002 adjusting for age). These data may result from mother-to-child and sexual modes of HTLV-I transmission. Thirty-nine (4.0%) of 981 white patients were HTLV positive, compared with 76 (5.2%) of 1449 in nonwhite patient. There were no significant differences regarding ethnicity in the total group of individuals analyzed, but in the subgroup of those with T-cell diseases, HTLV-I positivity appeared to be more prevalent among nonwhite patients (p < 0.001).
The mean age of the HTLV-I positive group (n = 116) was 42 years, compared with 35 years for the HTLV-negative population and patients with T-cell diseases, and those with HTLV-I were more likely to be between the ages of 31 and 50 years.
As shown in Table 2, the risk of HTLV-I infection was 49.9- and 44.4-fold higher in those with T-cell diseases(p < 0.0001) and in family members of HTLV-I-positive patients, respectively, compared with the total number of patients with cancer and those with other hematologic diseases. In multitransfused patients, the risk of HTLV-I infection was 14.9-fold higher than for those with other hematologic diseases without exposure to blood transfusion.
Previous reports from serologic studies in Brazil of neurologic patients and blood donors suggested that HTLV-I may be endemic in some areas of the country(18-20). HTLV-I is clearly associated with human diseases, and it is transmitted sexually, through breast milk, and by blood transfusions(21,22). To prevent the dissemination of this virus through blood transfusions, serologic tests to detect HTLV-I/II antibodies in blood donors have become mandatory procedures, especially in places where HTLV-I and HTLV-II are endemic. Counseling programs for HTLV-infected individuals have been set up to try to break the virus transmission cycle.
In our study, we first evaluated those at risk for acquiring HTLV-I and HTLV-II infections among healthy individuals and among hematologic and nonhematologic patients to estimate the rate of HTLV infection in Rio de Janeiro. We confirmed a seropositivity of 0.4% of HTLV-I antibodies for blood donors, a rate that was in agreement with a previous report(6). This rate is relatively low compared with the studies of seropositivity rate of HTLV-I in Jamaica or in Japan. We believe that our data regarding healthy individuals probably underestimate the rate in the general population, because in Brazil, most blood donors are male and young, but HTLV-I infection is more frequent among women and individuals older than 40 years of age(24).
We also found an increased risk of HTLV-I and II infections among multitransfused patients. The observed seroprevalence in this group was 9.1% and 0.8 for HTLV-I and HTLV-II, respectively. The transmission of HTLV-I and HTLV-II by blood products was observed only in recipients of cellular blood components, and the efficiency was also correlated to the potential risk of infection associated with viral load. In our group, the patients with aplastic anemia and acute leukemia who received platelets and fresh red blood cells were at more risk for HTLV-I/II infection than those exposed to plasma products or cellular blood units older than 5 days. Three patients in this subgroup developed HAM/TSP 12 months after seroconversion, and these findings are similar to those described previously in HTLV-I/II-endemic and -nonendemic areas(25,26).
We also found an increased seroprevalence rate for HTLV-I among those with lymphoid malignancies, mainly T-cell diseases. The systematic screening of HTLV-I in these cases disclosed 44 patients with ATLL, further segregated into distinct clinical subgroups such as acute, chronic, smoldering, and lymphoma types. Sixteen patients had typical acute ATLL with abnormal lymphocyte counts, flower cells, and hypercalcemia at the time of the serologic study. However, in patients with the diagnosis of lymphoma or cutaneous T-cell lymphoma without the classic signs or symptoms of ATLL, the detection of HTLV-I antibodies was important in characterizing 28 more cases of ATLL. HTLV-I proviral DNA was detected in the lymphoid cells of all of these patients(16).
The epidemiologic studies performed in endemic areas have indicated that ATLL develops mainly in individuals infected at birth, but a blood transfusion has been identified as significant risk factor for the development of HAM/TSP(27,28). We have documented that T-cell diseases in Rio de Janeiro are highly likely to be associated with HTLV-I infection compared with other diseases such as solid cancers, B-cell leukemia, and myeloid leukemia, and the high prevalence of HTLV-I antibodies among the family members of the infected individuals corroborates the previous reports of intrafamilial transmission of HTLV(23).
The geographic variations in HTLV-I positivity around the world involve a complex mixture of the effects of social behaviors and biologic factors. Although ethnic composition together with standard lifestyles has been thought to influence the risk of HTLV-I infection(1), we have not found significant differences among total groups of whites and nonwhites. However, in the subgroup of those with T-cell diseases, nonwhite patients had a higher rate of ATLL. The possible explanation is that nonwhite individuals have some genetic susceptibility that, associated with some environmental cofactors, may help the virus to transform normal to malignant cells.
In conclusion, we have confirmed the presence of HTLV-I and HTLV-II in Rio de Janeiro, Brazil, in healthy individuals and in patients exposed to multiple blood transfusion. Our data emphasized the importance of using a serologic screening test for HTLV antibodies to differentiate ATLL from other lymphoproliferative disorders. To evaluate the risk for ATLL among HTLV-I carriers, some sequential epidemiologic studies need to be done in Brazil to understand the complex mixture of behavior of patients and the biology of this retrovirus.
Acknowledgments: We gratefully acknowledge the staff of the Hematology and Oncology Service of Cancer Hospital(INCA) and Instituto de Hematologia for providing patient samples for these studies. We also acknowledge the efforts of the blood bank personnel and the blood donors who made this study possible. This study was partially supported by a grant from the ECC. Some data were presented (abstracts E-123 and E-138) at the Sixth International Conference in Human Retrovirology: HTLV, May 14-19, 1994, Absecon, NJ, U.S.A.
1. Blattner WA. Epidemiology of HTLV-I and associated diseases. In: Blattner WA (ed). Human retrovirology: HTLV
. New York: Raven Press, 1990:251-65.
2. Levine PH, Jaffe ES, Manns A, Murphy EL, Clark J, Blattner WA. Human T-cell leukemia virus type I and adult T-cell leukemia/lymphoma outside Japan and the Caribbean Basin. Yale J Biol Med
3. Kitagawa T, Fujishita M, Taguchi H, Miyoshi I Tadokoro H. Antibodies to HTLV-I in Japanese immigrants in Brazil. JAMA
4. Ferreira O, Vaz RS, Carvalho MB, et al. Human T-lymphotropic virus type I and II infections and correlation with risk factors in blood donors from São Paulo, Brazil. Transfusion
5. Cortes E, Detels R, Aboulafia D, et al. HIV-1, HIV-2, and HTLV-I infection in high-risk groups in Brazil. N Engl J Med
6. Matutes E, Schulz T, Andrada Serpa MJ, Campos-Araujo AQ, Pombo de Oliveira MS. Report of the Second International Symposium on HTLV in Brazil. Leukemia
7. Maloney EM, Biggar RJ, Neel JV, et al. Endemic human T cell lymphotropic virus type II infection among isolated Brazilian Amerindians. J Infect Dis
8. Nakauchi CM, Maryama K, Kanzaki LI, et al. Prevalence of HTLV-I antibody among two distinct ethnic groups inhabiting the Amazon region of Brazil. Rev Inst Med Trop Sao Paulo
9. Ishak R, Harrington WJ, Azevedo VN, et al. Identification of human T-cell lymphotropic virus type IIa infection in the Kayapo, an indigenous population of Brazil. AIDS Res Hum Retroviruses
10. Eiraku N, Novoa P, Ferreira MC, et al. Identification and characterization of a new and distinct molecular subtype of human T-cell lymphotropic virus type 2. J Virol
11. Yoshida M, Seiki M, Yamaguchi K, Takatsuki K. Monoclonal integration of HTLV in all primary tumors of adult T-cell leukemia suggests causative role of HTLV in the disease. Proc Natl Acad Sci USA
12. Gessain A, Barin F, Vernant JC, Gout O, Maurs L, Calender A, de The G. Antibodies to human T-lymphotropic virus type I in patients with tropical spastic paraparesis. Lancet
13. Franchini G. Molecular mechanisms of human T-cell leukemia/lymphoma virus type I infection. Blood
14. Catovsky D, Foa R. The lymphoid leukemias
. London: Butterworths, 1990.
15. Rios M, Kabbaz RF, Kaplan JE, Hall WW, Kessler D, Bianco C. Transmission of human T cell lymphotropic virus (HTLV) type II by transfusion of HTLV-I screened blood products. J Infect Dis
16. Pombo de Oliveira MS, Matutes E, Schulz T, et al. T-cell malignancies in Brazil. Clinicopathological and molecular studies of HTLV-I positive and negative cases. Int J Cancer
17. Fleiss JL. Statistical methods for rates and proportions
. New York: John Wiley & Sons, 1981:14-15.
18. Takayanaqui OM, Cantos JL, Jardim E. Tropical spastic paraparesis in Brazil (letter). Lancer
19. Araujo AQC, Ali A, Newell A, Dagleish AG, Rudge P. HTLV-I infection and neurological disease in Rio de Janeiro. J Neurol Neurosurg Psychiatry
20. Moreira ED, Ribeiro TP, Swanson P, et al. Seroepidemiology of human T-cell lymphotropic virus type I/II in Northeastern Brazil. J Acquir Immune Defic Syndr
21. Proietti FA, Lima-Martins MVC, Passos VMA, Brener E. Carneiro-Proietti ABF. HTLV-I/II seropositivity among eligible blood donors from Minas Gerais State, Brazil [letter]. Vox Sang
22. Gabbai AA, Bordini JO, Vieira-Filho JPB, et al. Selectivity of human T lymphotropic virus type-1 (HTLV-1) and HTLV-2 infection among different populations in Brazil. Am J Trop Med Hyg
23. Tajima K, Tominaga S, Suchi T. Epidemiological analysis of the distribution of antibody to adult T-cell leukemia virus associated antigen: possible horizontal transmission of adult T-cell leukemia virus. Jpn J Cancer Res
24. Murphy E. The epidemiology of HTLV-I: modes of transmission and their relation to patterns of seroprevalence. In Blattner WA, ed. Human retrovirology: HTLV
. New York: Raven Press, 1990:295-305.
25. Osame M, Izumo S, Igata A. Blood transfusion and HTLV-I associated myelopathy. Lancet
26. Gout O, Baulac M, Gessain A, et al. Rapid development of myelopathy after HTLV-I infection acquired by transfusion during cardiac transplantation. N Engl J Med
27. Manns A, Murphy EL, Wilks R, et al. Detection of early human T-cell lymphotropic virus type I antibody patterns during seroconversion among transfusion recipients. Blood
28. Manns A, Wilks RJ, Murphy EL, et al. A prospective study of transmission by transfusion of HTLV-I and risk factors associated with seroconversion. Int J Cancer