Minimal Residual Disease After Allogeneic Stem Cell Transplantation: A Comparison Among Multiparameter Flow Cytometry, Wilms Tumor 1 Expression and Chimerism Status (Complete vs Low-Level Mixed Chimerism) in Patients with Acute Leukemia

Relapse represents the main cause of treatment failure after allogeneic stem cell transplantation (allo-SCT). Thus, monitoring of minimal residual disease (MRD) in allografted patients allows an early detection of recurrence and a subsequent intervention prior to florid relapse. Multiparameter Flow Cytometry (MFC), chimerism, cytogenetics and molecular analysis have been widely used for this purpose, although the gold standard needs to be established yet. The evaluation of fusion transcripts represents the most sensitive method but more than 65% of patients do not demonstrate a molecular target. WT1 mRNA is over expressed in > 90% of patients with acute myeloid leukemia (AML) but it is not so expressed in acute lymphoblastic leukemia (ALL). On the MFC analysis instead, the complexity of maturational patterns and phenotypic shifts after therapy may underestimate residual leukemic cells. Finally, previous papers reported that a state of mixed chimerism (MC) in RT-PCR may identify higher risk of relapse in patients with AML. However, the test has a low sensitivity (from 0,1% to 1%) and recipient cells are not necessarily linked to disease recurrence. Hypothesizing the presence of blast cells in the low-level mixed chimerism (MC, 1%<MC<5% of autologous cells) but not in the complete chimerism (CC, no evidence of autologous cells) status, we used these two ranges as markers of positive and negative MRD, respectively. The aim of our study was twofold. Firstly, to assess the overall agreement among chimerism, MFC and WT1 mRNA methods in monitoring MRD. Secondly, to investigate whether such methods were associated to patient' s relapse-free survival.

Fresh BM samples from 24 patients (17 AML and 7 ALL) in both morphological CR and CC or low-level MC status after allo-SCT were investigated. MRD with MFC, WT1 mRNA expression and chimerism analysis was evaluated at different time points: +1, +3, +6,+12,+18,+24 months after allo-SCT. The immunophenotypic analysis was performed using a six-color combinations and acquiring 250.000 events. RQ- PCR to test WT1 mRNA expression was made according to the standardized and quality-controlled method. Chimerism studies were performed with a multiplex amplification of 16 (STR). The agreement between two methods in monitoring MRD after allo-SCT was assessed by Kappa statistic. Moreover, time-to-event analyses were performed using Cox proportional-hazard models with time-dependent covariates. Risks were reported as hazards ratios (HR) along with their 95% confidence interval (95% CI).

Comparisons among results of MFC, RQ-PCR for WT1 mRNA and Chimerism were performed in all 67 serial samples obtained from 24 patients. A significant moderate agreement between MFC and WT1 mRNA evaluations was found (k= 0.463, p<0.001) as well as fair agreement between chimerism and MFC (k= 0.284, p=0.009) and chimerism and WT1 mRNA (k= 0.197, p=0.073). These results suggested a concordance among the three investigated techniques. In particular, the low-level MC would well detect the presence of leukemic cells, since the proportion of positive samples for MFC was not statistically different to the proportion of positive samples for WT1 mRNA within samples with low-level MC. Indeed, among 12 samples with low-level MC, 9 samples (75.0%) were also positive for MFC and 7(58.3%) were also positive for WT1 mRNA. Cut-off of 0.1% and 0.01% for MFC, 83.5 × 10⁴ ABL for WT1 mRNA and 96%-99% of donor cells for chimerism were selected. The median follow-up times for relapse-free was 12.8 mths and the overall estimated relapse-free survival after 36 mths was 66.5%. At the end of follow- up, 5 patients relapsed and 4 patients died. Although not significant, the detection of a positive MRD for all methods were associated to a higher incidence of relapse than the negative MRD. Similar HRs were observed in the analysis of MFC(HR = 6.55, 95%CI= 0.71–60.17, p=0.096) and WT1 mRNA (HR= 7.17, 95% CI=0.77–66.42, p=0.083) whilst a slighter HR was found for the chimerism analysis (HR= 0.40, 95%CI= 0.04–4.44, p=0.456).

According to our study MFC, WT1 mRNA and CC or low-level MC displayed an overall agreement in monitoring MRD. In particular, the agreement on the low-level MC may suggest the presence of leukemic cells. The detection of a positive MRD by MFC and WT1 mRNA similarly identified higher risk of relapse in patients with acute leukemia (AL) undergone to allo-SCT.

No relevant conflicts of interest to declare.