Although more than 50% of Hodgkin's disease patients are cured with conventional chemotherapy, many will relapse and eventually die from their disease. Many efforts have been made to identify poor prognostic factors that could be useful in selecting high-risk patients in 1st CR who may benefit from high-dose chemo/radiotherapy. However, the role of early transplantation in 1st CR remains unclear. We have retrospectively analyzed the results obtained with this procedure in 22 hospitals belonging to the Spanish GEL/TAMO cooperative group. Twenty-seven patients, of whom 19 were males, underwent autologous transplantation for Hodgkin's disease in 1st CR between January 1987 and January 1996. Remission had been achieved after one (n = 22) or two (n = 5) lines of treatment. Twenty-four patients had advanced stage disease, 12 patients bulky mediastinal disease, nine bone marrow involvement and 18 had extranodal disease. Peripheral blood was used as the source of hematopoietic stem cells in 15 patients, BM in nine, and both in three. All but three patients received chemotherapy-based conditioning regimens (16 CBV, four BEAM and four BEAC), while three were conditioned with CY and TBI. There were no transplant-related deaths. Median (range) times to recover >0.5 x 10(9)/l neutrophils and >50 x 10(9)/l platelets were 14 (8-56) days and 16 (8-240) days, respectively. With a median follow-up of 30 (8-66) months, 21 patients are alive and in continuous CR. Four patients who relapsed after transplant at 8, 17.5, 22 and 26 months achieved a second CR with conventional chemotherapy; one patient relapsed 92 months post-transplant and died 5 months afterwards. Another patient died 30.5 months post-transplant from a secondary malignancy. In conclusion, high-dose therapy in poor prognosis Hodgkin's disease in 1st CR was well tolerated with no transplant-related mortalities. Although the follow-up of this series is relatively short, our results seem promising. Nevertheless, late relapses can occur, and the role of this procedure vs conventional treatment in very high-risk patients should be assessed in prospective randomized studies.