Background T10B9.1A-31 (T 10 B 9) and Muromonab-Orthoclone OKT3 (OKT3) are both murine MAb with a narrow specificity for T lymphocytes. Over the past 10 years, we have used each for T-cell depletion (TCD) of BM. In this report we describe similarities and differences using these antibodies, as well as their effects on patient outcome. Methods We compared BM mononuclear cells (BMMC) prepared using a Cobe Spectra apheresis machine with density gradient (DG) separation to remove RBC and enrich for CD34 + cells prior to TCD. FACS and limiting dilution assays (LDA) were used to measure the efficiency of TCD, the subsets of cells removed and CD34 content. Univariate statistics were used to assess graft outcome, including GvHD, graft failure, post-transplant lymphoproliferative disease (PTLD), relapse, DFS, and TRM. Results BMMC preparation on the Cobe Spectra resulted in superior recovery of CD34 + cells. However, this method could not be used with OKT3 due to inhibition of T-cell lysis. Optimal TCD required two rounds of complement at room temperature for OKT3, compared with one or two rounds for T 10 B 9. TCR n i + T-cells, but not natural killer cells, were spared to a greater degree with T 10 B 9. Further T-cell loss occurred during culture with T 10 B 9 but not with OKT3. Overall efficiency of TCD was superior using T 10 B 9. The risk of acute GvHD was higher with OKT3-mediated TCD, independent of T-cell content, and may have led to a higher incidence of PTLD. A decreased risk of relapse for patients with high-risk disease was seen with OKT3-treated grafts, but engraftment, TRM and DFS did not significantly differ. Discussion TCD using OKT3 results in higher T-cell content and higher rates of acute GvHD and PTLD compared with T 10 B 9. Cobe Spectra cannot be used for BMMC processing with OKT3, fewer CD34 + are therefore infused. Technical, as well as biological, differences between narrow specificity MAbs can affect graft outcome.