Clinical Characteristics and Outcomes of Infection with Human T-Cell Lymphotropic Virus (HTLV) in a Non-Endemic Area: A Single Institution Study

Abstract

Introduction: The first human oncogenic retrovirus discovered, human T cell lymphotropic virus (HTLV) remains poorly understood. Etiologically associated with adult T-cell leukemia (ATL), HTLV type 1 (HTLV-1) causes lifelong latent infection in CD4+ T cells and is estimated to affect between 5 to 20 million individuals globally. Despite being closely related, HTLV-2 infection has not been casually linked to clinical disease. In recent years, there has been an increased incidence of HTLV in the United States, however, affected patients are often under- and mis-diagnosed. While great strides have been made in understanding the virology, epidemiology, and clinical course of HTLV-1 and HTLV-2 infection, much of our current knowledge of HTLV and HTLV-related diseases is based on research in endemic areas, particularly Japan. We sought to characterize the epidemiology, clinical features, comorbidities, and survival of HTLV-1- or HTLV-2-positive individuals identified in a non-endemic area, using patient information from our academic institution. Methods: Our study was a single institution, retrospective case-control analysis of HTLV-1 or HTLV-2 patients between 1998 and 2020. We utilized two HTLV-negative controls, matched for age, sex, and race, for each one HTLV-positive case. Hematologic, neurologic, and rheumatologic covariates, including viral co-infection, opportunistic illnesses, malignancy, and clinical characteristics, were evaluated as variables affecting overall survival (OS). Results: We identified 38 cases of HTLV infection, of whom 23 were HTLV-1 and 15 were HTLV-2 positive. Median follow up was 39.7 months and median OS was 67.2 months for HTLV-seropositive patients, compared to 49.6 months and 131.2 months, respectively, for HTLV-seronegative patients. HTLV-1 positive patients had lower median OS compared to HTLV-2 patients (47.7 months vs. 77.36 months). During follow-up, 21 of 38 HTLV-seropositive patients died, compared to 33 of 76 HTLV-seronegative patients who died. Underlying reasons for HTLV testing were grouped into eight mutually exclusive categories: transplant evaluation, neurologic deficit(s), hematologic disorder, dermatologic, gastrointestinal, lymphadenopathy, infection, and other/unknown. Notably, the majority (53.9%) of patients in our control group received HTLV testing for transplant evaluation testing, compared to 23.7% of HTLV-seropositive patients. In evaluation of co-morbidities associated with HTLV, hepatitis C seropositivity was higher in HTLV-seropositive patients compared to controls (OR 16.8, 95% CI = 3.4 - 81.8, p < 0.001). There were no cases of HIV infection in our study. In Kaplan Meier analysis, hepatitis C and HTLV co-infection resulted in decreased 1-year and 3-year OS, compared to no infection, hepatitis C infection alone, or HTLV infection alone (Figure 1). Similarly, patients with any cancer diagnosis and HTLV infection had worse 1-year OS compared to patients with cancer or HTLV alone. There was no significance in 1-year OS with other risk factors, including CMV/HTLV co-infection or blood stream bacterial/HTLV co-infection. In univariate analysis, the hazard for 1-year all-cause mortality was increased among patients with HTLV-seropositivity (unadjusted HR 3.4, 95% CI 1.4- 8.2, P = 0.006), ATL (unadjusted HR 4.8, 95% CI 1.8 - 13.3, P = 0.08), AML (unadjusted HR 4.3, 95% CI 1.6-11.9, P = 0.01), and hepatitis C infection (unadjusted HR 3.2, 95% CI 1.2-8.5, P = 0.03) (Table 1). When corrected, multivariate analysis showed that HTLV seropositivity was no longer associated with 1-year all-cause mortality; however association with AML (adjusted HR 10.0, 95% CI 2.8-35.2, P = 0.0003) and hepatitis C infection (adjusted HR 7.0, 95% CI 1.5 - 32.0, P = 0.01) remained significant. Conclusions: HTLV-seropositivity was not associated with increased 1-year mortality in multivariate analysis. Our study is limited by our small patient sample size, as well as the biased patient control population due to HTLV testing. These studies warrant continued evaluation of clinical characteristics and outcomes of HTLV infection in non-endemic regions. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal