From July 1997 through June 2002, the German CLL Study Group conducted a prospective multicenter trial to assess the feasibility and efficacy of early SCT in patients with poor-risk chronic lymphocytic leukemia (CLL). The protocol comprised optional cytoreduction with CHOP, fludarabine, or FC; PBSC mobilization using the Dexa-BEAM regimen; and myeloablative therapy with TBI/CY followed by reinfusion of purged stem cells. Inclusion criteria were age <61 years, stage Binet B/C or poor-risk stage A as defined by short lymphocyte doubling time plus elevated STK, and one line of pretreatment or less. Endpoints were feasibility (as defined by the proportion of patients able to successfully undergo all study procedures); toxicity; efficacy; and quality of life (QOL) as defined by the Spitzer QOL Index.Results: As of July 2004, 128 of 179 eligible patients were evaluable. Median age was 51 (27–60) years. Binet stages were poor-risk A 15%; B 61%; C 24%. An unmutated VH rearrangement was present in 72%. Eighty-six percent of the patients were chemotherapy naive at study entry. Best documented response during the pretransplant phase was complete remission (CR) in 28% and partial remission (PR) in 64% of the patients, giving an overall response rate of 92%. Immunomagnetically purged grafts were obtained in 83% of the patients in whom mobilization was attempted, containing 4.6 (1.2–22.1) CD34+ cells per kg body weight. Altogether, 98 of 128 patients (77%) proceeded to SCT. Reasons for exclusion from SCT were disease progression, mobilization failure, toxicity during the mobilization phase, or patient's refusal. At 3 months post SCT CR was reported in 78% and PR in 21%. With a follow-up time of 36 (3–77) months, median progression-free survival of all 128 patients intended to treat was 59 months (54 months for those with unmutated VH). 4-year overall survival of all 128 patients was 84% (95%CI 74–94). Grade 3/4 non-hematological toxicity was mainly due to infections, which occurred in 21% of the patients (18% pretransplant; 7% post transplant). Six complications were fatal (4 pretransplant, 2 post transplant), translating into a 5% probability of treatment-related death. A single case of treatment-related myelodysplasia / AML has been reported to date. One year after SCT, QOL index had the maximum score of 10 points in 48 of 57 evaluable patients (84%). In comparison to prestudy levels, QOL had increased in 35%, remained similar in 50%, and decreased in 15% of the patients.Conclusions: Early sequential high-dose therapy including SCT for poor-risk CLL is feasible and has promising efficacy. Whereas the transplant-related toxicity appears to be acceptable, future studies should aim at replacing Dexa-BEAM by a similarly effective but less toxic mobilization regimen.