OBJECTIVE To determine the rate of concordance of the Microhemagglutination Assay for Antibodies to T. pallidum (MHA-TP) and the Fluorescent Treponemal Antibody-Absorption test (FTA-ABS) prior to therapy in patients with early stage syphilis and to assess the incidence of and associated risk factors for seroreversion of these treponemal specific tests during the first year after therapy for early syphilis. METHODS Multicenter, prospective, cohort treatment study of patients with early syphilis. METHODS Five hundred twenty-five patients were enrolled in a study to evaluate the response of early syphilis to either benzathine penicillin 2.4 million units intramuscularly once or this therapy plus amoxicillin 2 g and probenecid 500 mg orally both three times daily for 10 days. Serologic and clinical follow-up was conducted at intervals over 1 year. MHA-TP and FTA-ABS tests were performed on serologic specimens from each patient visit. RESULTS Enrollment specimens showed 5% discordant MHA-TP and FTA-ABS results with 85% of these demonstrating a nonreactive MHA-TP. This occurred most commonly in primary syphilis. In patients who had a 1-year serologic follow-up with FTA-ABS or MHA-TP, seroreversion occurred in 9% and 5% of cases, respectively. No association between HIV-seropositivity and TST seroreversion was demonstrated. CONCLUSIONS The MHA-TP may be less sensitive than the FTA-ABS for identifying patients with primary syphilis. Treponemal specific tests may become nonreactive during the first year after therapy for early syphilis.