Primary CNS lymphoma (PCNSL) treatment includes two phases: induction and consolidation. Induction consists of high-dose-methotrexate-based polychemotherapy for most patients, with regimen and dose variations according to patient characteristics and country. Several strategies have been proposed for the consolidation phase, with whole-brain irradiation (WBRT) the most common. However, some authorities recommend avoiding WBRT because of its related risk of severe neurotoxicity. The most relevant alternatives to WBRT are high-dose chemotherapy supported by autologous stem cell transplantation (HDC/ASCT) or non-myeloablative chemotherapy, the former supported by several single-arm phase II trials. Moreover, HDC/ASCT is the only strategy that is assessed in comparison with WBRT in ongoing randomized trials. The rationale for the use of HDC/ASCT in PCNSL patients is based on the fact that the delivery of high doses could achieve therapeutic drug concentrations in the brain and cerebrospinal fluid, and that non-cross resistant drugs used for conditioning (e.g. alkylating agents) could favour the elimination of residual chemoresistant lymphoma cells. Worldwide experience with HDC/ASCT is limited to few single-arm phase II trials, but overall results are encouraging, mostly when thiotepa/containing conditioning regimens are used, both in newly-diagnosed and relapsed patients. However, several questions on efficacy and feasibility of HDC/ASCT, as well as the best candidates for this strategy, the optimal conditioning regimen, the best time for response assessment, and acute and late effects remain unanswered. In this review, we critically analyze reported studies on HDC/ASCT in PCNSL, and discuss its current role and future perspectives in treating this aggressive malignancy.