Human T-lymphotropic virus 1 (HTLV-1) infection is associated with HTLV-associated myelopathy/tropical spastic paraparesis (HAM/TSP), which affects approximately 5% of carriers. High proviral load is a risk marker for HAM/TSP, although there is an overlap of proviral load levels in peripheral blood between asymptomatic carriers and HAM/TSP patients. In this study, receiver operating characteristic curve analysis was used to define a set point of HTLV-1 proviral load that better indicates an increased risk for HAM/TSP. Proviral load was quantified in 75 asymptomatic carriers and 78 HAM/TSP patients in a Brazilian cohort. The cut-off of proviral load was defined as 114 copies/104 cells, with 78.2% sensitivity to identify true HAM/TSP patients. The mean proviral load levels were not significantly different between males and females with the same clinical status, and there was no significant correlation between proviral load and age at blood sampling, age at the onset of illness, or duration of disease. In HAM/TSP patients, proviral load was significantly higher in wheelchair-bound patients than in individuals able to walk without support and in those with the worst spinal cord injuries. Follow-up of HTLV-1-infected individuals showed that proviral load was more stable in asymptomatic carriers than in HAM/TSP patients. In a cohort study, periodically quantifying proviral load in asymptomatic carriers is necessary to identify those at risk for developing neurological disease, and it is necessary for HAM/TSP patients to monitor spinal injury and progression to walking disability. The measure of proviral load in clinical practice implicates the definition of the cut-off of proviral load and its validation during follow-up. J. Med. Virol. 84:664–671, 2012. © 2011 Wiley Periodicals, Inc.