Interactions Between Total Nucleated and CD34 Cell Dose and Their Impact on Outcomes After Single and Double Unrelated Cord Blood Transplantation for Patients with Hematological Malignancies. An Analysis on Behalf of Societe' Française De Greffe De Moelle Et Therapie Cellulaire (SFGM-TC) and Eurocord

Abstract

AbstractAbstract 361Pre-freezing (pf) and post-thawing (pt) total nucleated cells (TNC) is one of the most important factor for outcomes after UCBT. Its impact has been demonstrated after single UCBT (sUCBT); a minimum cell dose has been established as pfTNC of 2.5 ×107/kg, but its impact on outcomes after double UCBT (dUCBT) has not been shown. Also number of pfCD34 cells/kg is associated with outcomes after sUCBT, but only small series of patients have been analyzed. In order to evaluate the interactions between pf and pt TNC and CD34 and their impact on outcomes, we have studied separately 600 patients with hematological malignancies receiving a first sUCBT and 397 a first dUCBT in France (Table1). Methods:for all prognostic analysis pf and pt TNC and CD34 were divided into 4 categories at 25th, 50th and 75th percentiles. Results:Single UCBT: there was a highly statistical significant correlation between pf and pt TNC and CD34 (p<0.001, respectively). Median time to ANC recovery was 26 days (6-84). Cumulative incidence (CI) of ANC recovery at day 60 was 83%. In univariate analysis, the best cutoff point (associated with greater ANC recovery) of pfTNC and pfCD34 were ≥3.6 ×107/kg and '1.6 ×105/kg and ptTNC and ptCD34 were ≥3.3 ×107/kg and ≥1.3 ×105/kg. In multivariate analysis pf or pt TNC or CD34 were independently associated with ANC recovery (pfTNC HR=1.4, p=0.005; pfCD34 HR=1.3, p=0.01; ptTNC HR=1.4, p=0.01, ptCD34 HR=1.4, p<0.0001). We have not found any association of number of HLA disparities and ANC recovery. When analyzing only adults, patients receiving <2.0 ×107/kg, CI of ANC recovery was only 70% compared to 81% for ≥2.0 ×107/kg (p=0.02). CI of aGVHD at day 100 was 36% and was not associated with TNC or CD34. Estimated 2 year (y) OS was 47%, 2 y DFS was 40%. Interestingly, in a multivariate analysis, pfTNC, pfCD34 or ptCD34 were not associated with OS or DFS, but only higher ptTNC (≥3.3 ×107/kg) was associated with OS (51%vs43%, HR: 0.78, p=0.05) but not with DFS. Double UCBT: there was a highly correlation between pf and pt TNC and CD34 (p<0.001, respectively). CI of ANC recovery at day 60 was 81% in a median time of 23 days (5-64). In univariate analysis, pfTNC, pfCD34 and ptTNC were not associated with ANC recovery. However, there was an association of ptCD34 cell dose (Figure1) and ANC recovery. Chimerism data was available for 75% of the patients (n=298) during the first 100 days: 76% were full donor, 14% were mixed and 11% of the patients had autologous recovery. Autologous recovery was also associated with lower CD34 cell dose, it was 49% in patients receiving (<0.9×105/kg), and 25% for remainders (p<0.001). In multivariate analysis, ptCD34 >0.9x ×105/kg was the only independent factor associated with ANC recovery (HR=1.6, p=0.001). CI of acute GVHD at day 100 was 42% and was not associated with TNC or CD34 cell dose. At 1 y NRM was 22% and relapse incidence 26%. Estimate 1 y DFS was 50±3%. TNC or CD34 cell dose were not associated with any of the above outcomes. Only disease status at transplant impacts outcomes (64% early phase, 49% intermediate phase and 37% for more advanced disease). In conclusion, our study confirms the impact of cell dose measured by pf and ptTNC and CD34 on neutrophil recovery after sUCBT and the minimum cell dose recommended should be pfTNC≥2.5 ×107/kg and ptTNC≥2×107/kg, however only ptTNC is associated with survival in sUCBT. This is the first time that an impact of ptCD34 cell dose on neutrophil recovery after dUCBT is demonstrated and may be used to choose the best CB units. The different associations of cell dose in sUCBT and dUCBT can be explained by biological and immunological properties of other CB cells in the graft. [Display omitted] [Display omitted] Disclosures:No relevant conflicts of interest to declare.