BACKGROUND: Pre-transplantation (18)F-fluorodeoxyglucose (FDG) PET-negativity is one of the strongest predictors of outcome after high-dose therapy and autologous stem-cell transplant (HDT/ASCT) for patients with relapsed or refractory Hodgkin's lymphoma. In this study, we assessed the feasibility and activity of PET-adapted salvage therapy with brentuximab vedotin, followed by augmented ifosfamide, carboplatin, and etoposide (ICE). METHODS: In this non-randomised, open-label, single-centre, phase 2 trial, we enrolled patients with relapsed or refractory Hodgkin's lymphoma who had failed one previous doxorubicin-containing chemotherapy regimen. All patients received weekly infusions of 1·2 mg/kg brentuximab vedotin on days 1, 8, and 15 for two 28 day cycles. After completion of brentuximab vedotin treatment, patients received a PET scan. Patients who achieved PET-negative status (a Deauville score of 1 or 2) proceeded directly to HDT/ASCT; those with persistent abnormalities on PET received two cycles of augmented ICE (augICE; two doses of ifosfamide 5000 mg/m(2) in combination with mesna 5000 mg/m(2) continuous infusion over 24 h, days 1 and 2; one dose of carboplatin AUC 5, day 3; three doses of etoposide 200 mg/m(2) every 12 h, day 1) before consideration for HDT/ASCT. Only patients with persistent abnormalities on PET after brentuximab vedotin received augICE; however, all patients in the intention-to-treat population were assessed for the primary outcome, which was the proportion of patients who were PET-negative after brentuximab vedotin alone or brentuximab vedotin followed by augICE. This study is registered with ClinicalTrials.gov, number NCT01508312, and is no longer accruing patients. FINDINGS: Between Jan 5, 2012, and Oct 4, 2013, we enrolled 46 patients. One patient was deemed ineligible, and not evaluable, before treatment initiation owing to having nodular, lymphocyte-predominant Hodgkin's lymphoma and thus 45 patients received treatment. After brentuximab vedotin, 12 patients (27%, 95% CI 13-40) were PET-negative and proceeded to HDT/ASCT. 33 (73%, 95% CI 60-86) patients were PET-positive after brentuximab vedotin; one PET-positive patient withdrew consent, therefore 32 PET-positive patients received augICE, 22 (69%, 95% CI 53-85) of whom were PET-negative. Overall, 34 patients (76%, 95% CI 62-89) achieved PET-negativity. All 44 patients who completed treatment as per protocol proceeded to receive HDT/ASCT. Brentuximab vedotin was well tolerated and associated with few grade 3-4 adverse events including hyperglycaemia (two [4%] patients, grade 3), nausea (one [2%], grade 3), hypoglycaemia (one [2%], grade 3 and one [2%], grade 4), and hypocalcaemia (one [2%], grade 3 and one [2%], grade 4). INTERPRETATION: PET-adapted sequential salvage therapy with brentuximab vedotin followed by augICE resulted in a high proportion of patients with relapsed or refractory Hodgkin's lymphoma achieving PET-negativity, and therefore could optimise the chance of cure after HDT/ASCT. FUNDING: Seattle Genetics.