Prognostic Impact of Physiological B-Cell Precursors (hematogones) in the Bone Marrow As Determined by Immunophenotyping in the Post-Transplant Period in Patients with AML

Abstract 1406

Aiming to improve post-transplant monitoring strategies for patients with acute myeloid leukemia (AML) or myelodysplastic syndromes (MDS), approaches to monitor the minimal residual disease load are at present limited to distinct genetic subgroups, e.g. NPM1mut AML. In the conservative treatment setting, increased levels of physiologic B-cell precursors (>0.01% CD10+/CD19+ cells; “hematogones”) as determined by multiparameter flow cytometry (MFC) in the bone marrow (BM) were suggested to be favorable in AML pts (Chantepie et al., Blood, 2011). We investigated whether the prognostic impact of physiologic B-cell precursors can be confirmed in pts with myeloid malignancies in the transplant context.

We quantified CD10+/CD19+ B-cell precursors by MFC in the post-transplant period in 76 pts with AML/MDS who received allogeneic hematopoietic stem cell transplantation (HSCT) at Hamburg University, and performed correlation with survival (41 m/35 f; 18–71 yrs; de novo AML: n=51; s-/t-AML: n=11; MDS: n=13). The majority (n=84, 73%) received reduced intensity conditioning (RIC), mainly based on the FLAMSA regimen. Hematogones were evaluated by MFC on days +30 (d30) and +100 (d100) post-HSCT including the following antibody panel: CD19+, CD10+, CD34+, TdT+.

Median levels of CD10+/CD19+ cells were 0,06% on day 30 (mean, 0.25%; range, 0–2.4%) and 1.55% on day 100 (mean, 2.13%; range, 0–12.4%). OS from HSCT did not differ on the basis of the CD10+/CD19+ cells measured on d30 (mean±SD=61±6 weeks, for CD10+/CD19+ cells below the mean of 0.25% vs 74±13 weeks for CD10+/CD19+ cells above the mean, RR=0.73, p=0.685). Estimated 1-year-OS±SE was 63±10% for CD10+/CD19+ cells below the mean of 0.25% versus 71±17% for CD10+/CD19+ cells above the mean, respectively. EFS did not show a significant difference, either. Mean CD10+/CD19+ cells on day 100 associated positively with 1-year-OS. In the group below the mean CD10+/CD19+ cells of 2.13%, 1-year-OS±SE was estimated to be 69±9% (median±SEM=89±10 weeks) which was worse when compared to the group above the mean with a 1-year-OS±SE of 83±15% (median±SEM=96±8 weeks; RR=0.124, p=0.046). The effect of low versus high CD10+/CD19+ cells measured at day 100 on EFS was even more pronounced. 1-year-EFS was significantly worse with 60±10% in the group with lower CD10+CD19+ counts on day 100 versus 95±4% in the group with higher counts (RR=0.095, 95%CI=0.012–0.728, p=0.024).

In univariate analysis, disease stage, remission status pre-HSCT, cytogenetic risk group and cGVHD had significant influence on OS and/or EFS. By multivariate analysis, CD10+/CD19+ cells on day 100 remained significantly associated with improved EFS (RR=0.998, p=0.001) but significance of the association with OS was lost (RR=0.145, p=0.07). Other factors significantly associated with superior OS and EFS were occurrence of cGVHD (OS: RR=0.997, p=0.049, EFS: RR=0.996, p=0.001), and low/intermediate cytogenetic risk group (OS: RR=0.995, p=0.005, EFS: RR=0.997, p=0.03).

To the best of our knowledge, an association of CD10+/CD19+ cells measured by MFC with survival has not been performed after HSCT in pts with AML and MDS. Our study is the first one showing a favorable prognostic impact of increased CD10+/CD19+ cells 100 days after HSCT. It seems that the level of physiological precursors contributes to the prognostic parameters in patients with myeloid malignancies undergoing HSCT. In more detail, a proportion of more than 2.13% of hematogones at day 100 was associated with improved OS and EFS. The association remained positive in multivariate Cox regression analysis, when other parameters such as pre-transplant permission status, disease stage and cGvHD were included. In conclusion, the hematogones level at day 100 seems to be an independent prognostic parameter which should be further evaluated in pts with myeloid malignancies undergoing HSCT. Considering the experience needed for the determination of leukemia associated immunophenotypes by MFC, or the lack of molecular markers suitable for MRD monitoring in many AML and MDS patients, the introduction of hematogones measurement at defined time points may contribute to the identification of high-risk patients in the transplant context.