The Cord Blood Transplantation study group conducted a prospective study of unrelated cord blood transplantation (CBT) to better define the role of this stem cell source for subjects requiring unrelated allogeneic transplantation. We report on 1 stratum of the study designated for adult subjects. The primary end point of the study was survival at 180 days. Secondary end points included engraftment, graft-versus-host disease, relapse, and long-term survival. Eligibility criteria for malignant and nonmalignant diseases were specified. Subjects with active central nervous system disease, Karnofsky performance status <70%, grade 3 or 4 or primary myelofibrosis, or suitable related donors were excluded. Enrollment required a single cord blood unit containing >10(7) nucleated cells per kilogram of recipient weight and matched at > or =4 HLA-A and -B (low or intermediate resolution) and -DRB1 (high resolution) types. Thirty-four subjects were entered, with a median age of 34.5 years (range, 18.2-55 years). Most subjects (n = 23) had a 4 of 6 match, 10 subjects had a 5 of 6 match, and 1 subject had a 6 of 6 match. Diagnoses at transplantation included acute myelogenous leukemia (n = 19), acute lymphoblastic leukemia (n = 9), chronic myelogenous leukemia (n = 3), myelodysplastic syndrome (n = 1), paroxysmal nocturnal hemoglobinuria (PNH) (n = 1), and non-Hodgkin lymphoma (n = 1); 94% were classified as poor risk according to National Marrow Donor Program criteria. Subjects received total body irradiation/cyclophosphamide (n = 27) or busulfan/melphalan (n = 7) conditioning regimens. Four subjects died before CBT and are described here but are not included in the main analysis. The cumulative incidence rates and median times to neutrophil (500/microL) and platelet (>20,000/microL) engraftment were 0.66 by day 42 (median, 31 days) and 0.35 by day 180 (median, 117 days). The cumulative incidence rate for grade II-IV GVHD was 0.34 by day 100. For the primary end point, survival at 180 days, Kaplan-Meier survival estimates were 0.30 (95% confidence interval, 0.14-0.46) by day 180 after transplantation. To date there are 2 survivors, and both are >36 months from enrollment. A retrospective analysis was performed by using high-resolution HLA-A and -B typing, which revealed that approximately one third of subjects had 1 or more additional HLA mismatches compared with results of low- or intermediate-resolution HLA typing. The findings of high treatment-related mortality and slow engraftment kinetics indicate that CBT should continue to be performed in specialized centers with a research focus on cord blood cells.