This study was set up to demonstrate whether prognostic classification based on the secondary age-adjusted International Prognostic Index (sAA-IPI) for recurring aggressive non-Hodgkin lymphoma (NHL) or the prognostic score for recurring Hodgkin lymphoma (HL) can be improved by including the midtreatment results of fluorine-18-fluorodeoxy-glucose-positron emission tomography (FDG-PET). Clinical data on patients with recurring lymphoma who were treated with second-line chemotherapy (DHAP-VIM-DHAP) followed by autologous stem cell transplantation (ASCT) were collected and combined with the results of FDG-PET performed before and after 2 cycles of reinduction chemotherapy. PET responses after 2 courses were scored as complete remission (CR), partial remission (PR), or no response (NR). A multivariate analysis was performed to design a predictive model. The number of patients (101 of 117) included those (78 patients with aggressive NHL and 23 patients with HL) that could be analyzed according to protocol. Of these, 80 patients were chemosensitive and 77 received transplants. Both secondary clinical risk score (P<.001) and FDG-PET response (P<.001) were independent predictive factors for the total evaluable group of patients with lymphoma and for patients with NHL alone. The combined use of the clinical risk score and FDG-PET response after 2 chemotherapy courses identified at least 4 categories of patients with a failure-free survival varying between 5% to 100% after transplantation (P<.001). These data indicate that the secondary clinical risk score in conjunction with FDG-PET response provides a more accurate prognostic instrument for the outcome of second-line treatment at least in patients with recurring NHL.