Determination of the true prevalence of infection with the human T-cell lymphotropic viruses (HTLV-I/II) may require a combination of biomolecular and serological analyses.

ABSTRACT

Infection with the human T-cell lymphotropic virus types I and II (HTLV-I/II) usually is determined by tests that detect antibodies to the viral structural proteins. However, recent studies revealed that patients with mycosis fungoides have proviral DNA sequences related to the HTLV transactivating-transforming gene tax, without having antibodies to the virus. These results raised the possibility that the prevalence of HTLV infection in the general population of the United States also may be underestimated. To reassess the prevalence of HTLV-I/II infection effectively, a population at increased risk for infection (i.e., a cohort of injection drug users [IDUs]) was studied. Paired sera and peripheral blood mononuclear cells (PBMCs) from 81 IDUs were subjected to testing by Western blot analysis for antibodies to the viral structural proteins gag and env and by polymerase chain reaction (PCR) Southern analysis to detect gag, pol and tax proviral DNA sequences. Western blot assays showed 1 of 81 IDUs to be positive for HTLV-I, 14 to be positive for antibodies to HTLV-II, and 3 to be HTLV-serotype indeterminate. When whole-cell lysates of PBMCs from these individuals were subjected to PCR and Southern analysis. 39 of 81 were found to have HTLV-related sequences. A total of nine IDUs were found to be infected with HTLV-I, a figure nearly 10 times higher than that estimated by serology alone. Bio-molecular analysis showed HTLV-II-specific proviral sequences in 21 IDUs. Three individuals were seropositive for HTLV-II but lacked PCR evidence of gag, pol and tax sequences. Thus, the overall prevalence of HTLV infection among this cohort was 59% (43 of 81) (i.e., more than twice the frequency predicted by serology, 18 of 81 or 22%). These results indicate that it may be necessary to incorporate biomolecular as well as serological methodologies to identify all persons infected with these retroviruses.