Positron emission tomography using [(18)F]fluorodeoxyglucose (FDG-PET) has emerged as the standard response assessment tool in frontline therapy for classical Hodgkin lymphoma (cHL). The ability of FDG-PET to predict outcomes in patients with relapsed cHL treated with modern standard salvage chemotherapy and autologous stem cell transplantation (ASCT) remains uncertain. Forty-six patients with relapsed/refractory cHL treated from 2001 to 2007 with standard salvage/ASCT therapy had FDG-PET available for blinded review. The results of pre-ASCT FDG-PET interpreted by the international harmonization project (IHP) criteria were compared with published prognostic models for prediction of event-free survival (EFS) and overall survival (OS). Overall, 3-year EFS was 62% and OS was 78%, with a median follow-up of 38 months. Pre-ASCT FDG-PET response significantly predicted 3-year EFS in FDG-PET-negative (82%) versus FDG-PET-positive (41%) patients (P = .02). A trend was observed for 3-year OS comparing FDG-PET-negative (91%) versus -positive (64%) patients (P = .08). Multivariate analysis demonstrated the independent prognostic significance of pre-ASCT FDG-PET for EFS with a hazard ratio (HR) of 3.2 (confidence interval [CI] 1.1-9.0, P = .03). Pre-ASCT FDG-PET scans predict EFS in patients with relapsed cHL patients treated with modern salvage/ASCT therapy and warrant prospective evaluation.