Youth that are victims of commercial sexual exploitation of children (CSEC) have a host of clinical problems and often run away from home, residential care, and treatment, which complicates and limits treatment effectiveness. No research to date has attempted to predict running away in CSEC victims. The present study aimed to 1) characterize a clinically referred sample of girls who were victims of CSEC and compare them to other high-risk girls (i.e., girls who also have a history of trauma and running away, but deny CSEC); and 2) examine the utility of using the Youth Level of Service/Case Management Inventory (YLS/CMI) to predict future running away. Data were collected from de-identified charts of 80 girls (mean age = 15.38, SD = 1.3, 37.9% White, 52.5% CSEC victims) who were referred for psychological assessment by the Department of Child Services. Girls in the CSEC group were more likely to have experienced sexual abuse (χ = 6.85, p = .009), an STI (χ = 6.45, p = .01), a post-traumatic stress disorder diagnosis (χ = 11.84, p = .001), and a substance use disorder diagnosis (χ = 11.32, p = .001) than high-risk girls. Moderated regression results indicated that YLS/CMI scores significantly predicted future running away among the CSEC group (β = 0.23, SE = .06, p = .02), but not the high-risk group (β = -.008, SE = .11, p =.90). The YLS/CMI shows initial promise for predicting future running away in girls who are CSEC victims. Predicting running away can help identify those at risk for and prevent running away and improve treatment outcomes. We hope current findings stimulate future work in this area.