Incidence of Adult T-Cell Leukemia/Lymphoma among Human T-Lymphotropic Virus Type I Carriers in Saga, Japan1

Using a population-based cancer registry, we tabulated 69 definite adult I -IT11leukemia/lymphoma cases (36 males and 33 females) and 2.20 expected cases (0.95 for males and 1.25 for females) diagnosed from 1981 to 1983 in Saga, Japan. The number of human T-lymphotropic virus type I carriers was computed by applying sexand age-specific antihuman T-lymphotropic virus type I antibody positive rates among blood donors at the blood center in 1986 to the whole population of Saga Prefecture in 1982. The age-specific incidence rates among male human T-lymphotropic virus type I carriers from 40 to 79 yr of age per 100,000 were significantly higher than those of female carriers ( /' < 0.05), and the rates from 60 to 69 yr of age were the highest in both sexes. The annual crude incidence rates among carriers were 115.9 for males and 66.4 for females. The summary incidence rates with 95% confidence intervals were 115.9 (58.4 to 193.0) for males and 65.9 (30.0 to 115.9) for females. The cumulative risks were 4.5% (0.8 to 11.0) for males and 2.6% (0.3 to 7.0) for females. These morbidity figures were assumed to be underestimated partly due to the newly proposed clinical entity of adult T-cell leukemia/lymphoma. INTRODUCTION ATL3 or ATLL is a new clinical entity proposed by Takatsuki et al. in 1977 (1). By using the cell lines established beforehand (2), the associated virus was detected by means of the passive immunofluorescence method (3). Later the DNA sequences were clarified (4), and the ATL virus was verified to be identical to the one isolated from the lymphocytes of a patient with cutaneous T-cell lymphoma in the United States (5). In the meantime, the virus was named HTLV-I, and it was proved that the virus is directly associated with ATLL. However, the precise carcinogenesis of ATLL is still unknown. There are several epidemiological studies dealing with the morbidity of (or mortality from) ATLL in Japan (6-9); how ever, we considered it to be important to calculate more accurate and reliable incidence rates and cumulative risks of ATLL by sex with confidence intervals among HTLV-I carriers in an area with moderate endemicity of the virus by using a popula tion-based local cancer registry in Saga Prefecture, located in Kyushu, the southern part of Japan with a population of 880,000. PATIENTS AND METHODS Patients. We first abstracted 256 reports of malignant lymphomas and other lymphoid malignancies (including case reports submitted by Received 4/7/88; revised 7/29/88. 9/19/88; accepted 9/29/88. 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. ' Supported in part by a Grant-in-Aid for Cancer Research from the Ministry of Education (61010059). Japan. 2To whom requests for reprints should be addressed. ' The abbreviations used are: ATL. adult T-cell leukemia: ATLL, adult T-cell leukemia/lymphoma: HTLV-I. human T-lymphotropic virus typ«I; HAM, HTLV-I associated myelopathy. physicians, abstracted information from medical records at major hos pitals by the registry staff, and death certificates whose addresses are located in Saga) newly diagnosed between 1981 and 1983 from the data of the Prefectura! Cancer Registry (or Annual Cancer Survey, formerly). Then we sent questionnaires to the attending physicians inquiring whether they adopted the definite and/or differential diagnostic proce dures, such as detections of monoclonal integration of proviral DNA, histolÃ3gica! and/or cytological diagnoses, examinations of surface marker of lymphocytes, tests of anti-HTLV-I antibody, or specific clinical signs and symptoms (10). When the histolÃ3gica! and/or cyto logical specimens were kept, microscopic examinations were carried out for reconfirmation by pathologists and hematologists, and the final diagnoses were made. Consequently, one male anti-HTLV-I antibody negative ATLL (11) and one female smoldering ATL (or pre-ATL) (10) with positive antibody were not included. Next we reviewed 121 autopsy cases described in the Annual of the Pathological Autopsy Cases in Japan, which were reported from the institutes in Saga and in the neighboring prefectures from 1981 to 1984. Definite ATLL cases whose addresses were located in Saga were tabulated into our study. Possible and/or suspected cases were referred to respective pathologists to obtain final diagnoses. On the basis of the formation above, we compiled 69 definite incident ATLL cases (36 males and 33 females) diagnosed between 1981 and 1983 (Table 1). For a male case with probable ATLL based upon microscopic diagnosis and with positive T-cell marker but without confirmation of proviral DNA, the case was judged as probable and counted as 0.75. This inference was also based on the proportion of lymphoid T-cell versus B-cell malignancies in Kyushu (12). For one female case with possible ATLL based upon microscopic diagnosis with positive T-cell marker but without other diagnostic procedures, it was regarded as possible and counted as 0.5. For one male case and six female cases described as malignant lymphoma, not otherwise specified, the estimated number of ATLL patients was calculated by applying the number of these cases to the proportion of the number of definite ATLL cases to definite non-ATLL patients among malignant lymphomas in the respective age groups. In all, 36.95 ATLL patients among males and 34.25 cases among females were tabulated in the present study. HTLV-I Carriers. In order to calculate population at risk or the number of HTLV-I carriers in Saga Prefecture, sexand age-specific anti-HTLV-I antibody positive rates among blood donors at the Saga Red Cross Blood Center during the fiscal year of 1986 (from April 1986 to March 1987) were tabulated. At the blood center, the antibody to HTLV-I was screened at the final serum dilution (or titer) of 23 by the gelatin particle agglutination test (13), which is highly sensitive but less specific. In other words, there is little false negative but moderate false positive at this titer. Because the population seropositive rate at the titer of 26 or over by the gelatin particle agglutination method is reported to be practically equivalent to that by the passive immunoflu orescence method (14), this cutoff titer was utilized to calculate the number of HTLV-I carriers. As the seropositive rates for the elderly groups (aged 65 or over) were not available from the center, the seroconversion rates (or hazard functions) in these age groups surveyed among the general population in a neighboring prefecture (10) were used for extrapolating the prevalence curves in Saga. Applying these HTLV-I prevalence rates to the whole population in 1982, we computed the sexand age-specific number of HTLV-I carriers (Table 2).