Background The goal of this study was to perform a systematic review to examine the efficacy and safety of various salvage therapy regimens on patients with relapsed/refractory PTCL. Method The electronic searches were performed using PubMed, Cochrane Library, EMBASE, and Web of Science from inception through June 2015, with search terms related to relapsed/ refractory PTCL, salvage chemotherapy regimens, and clinical trials. An eligible study met the following inclusion criteria: (1) Patients had refractory or relapsed PTCL; (2) drug regimens were used for salvage therapy; (3) the study was a clinical trial; (4) the study reported on a series of at least 10 patients of PTCL. Results Of 35 records identified, a total of 14 studies were eligible for systematic reviews, and 12 different salvage regimens were investigated. A total of 618 relapsed/refractory PTCL patients were identified. The ORRs ranged from 22% for those treated with lenalidomide to 86% for those with brentuximab vedotin. By the three most frequent subtypes, the ORRs ranged from 14.2% to 71.5% for patients with the PTCL-NOS subtype, 8% to 54% for AITL subtypes, and 24% to 86% for the ALCL subtype. The medians of DOR, PFS, and OS ranged from 2.5 to 16.6 months, 2.6 to 13.3 months, and 3.6 to 14.5 months, respectively. The most frequently reported grade 3 or 4 adverse events (AEs) were hematological AEs, such as neutropenia and thrombocytopenia. Conclusion The efficacy of salvage therapy regimens is highly diverse for patients with relapsed/refractory PTCL; this heterogeneity in therapeutic effects might be due to the diversity in mechanisms, PTCL subtype distribution, and/or numbers/profiles of prior therapy. Comparative studies with matched pair analysis are warranted for more evidence of the salvage treatment effect on relapsed or heavily pretreated patients with PTCL.