The Proportional Association between WT1 mRNA Level in Peripheral Blood before Allogeneic Hematopoietic Cell Transplantation and Risk of Mortality in Acute Myeloid Leukemia Not in Remission

INTRODUCTION

The prognosis for allogeneic hematopoietic cell transplantation (allo-HCT) is relatively poor in patients with acute myeloid leukemia (AML) who are not in remission (non-CR). Wilms' tumor-1 gene messenger ribonucleic acid (WT1 mRNA), which is overexpressed in almost all AML, has been used as an indicator of monitoring minimal residual disease (MRD) and has been regarded as one of the prognostic factors in AML patients with complete remission (CR). However, the association between the expression levels of WT1 mRNA in peripheral blood (WT1) before allo-HCT and risk of mortality in AML patients in non-CR remains quite elusive. We therefore assessed the effect of the pre-transplant WT1 level on survival after allo-HCT in AML patients in non-CR.

METHODS

We retrospectively analyzed data from consecutive patients undergoing allo-HCT at our institution between November 2007 and January 2017. We excluded patients who did not have an available WT1 value within 28 days before the start of conditioning.

RESULTS

A total of 125 AML patients, including 46 non-CR patients (36.8 %), were eligible for this study. The median age was 48 years (range: 17-68 years) and the number of males was 69 (55.2%). Forty-two patients (33.6%) had AML with myelodysplasia-related changes. This study included 34 patients (27.2%) who had undergone multiple transplantation. Myeloablative and reduced-intensity conditioning were used in 73 (58.4%) and 52 (41.6%) patients, respectively. In 63 patients who received transplants from either a related peripheral blood stem cell or bone marrow (BM) donor, four pairs (6.3%) were serologically mismatched for one antigen of HLA and 45 pairs (71.4%) received transplants from HLA haploidentical donors. In 32 patients who received transplants from an unrelated BM donor, seven pairs (21.9%) were serologically mismatched for one HLA antigen. Twenty-nine patients (23.2%) received unrelated cord blood transplantation.

Among 46 non-CR patients, the median follow-up period among survivors was 1,092 days (range: 688-2,607 days) following allo-HCT. Receiver operating characteristic analysis for prediction of composite events including any cause of death or subsequent transplantation indicated that the best cutoff value for WT1 was 5,000 copies/ μg RNA in non-CR AML patients. Estimated two-year, event-free survival of non-CR patients with WT1 levels of less than 5,000 copies/ μg RNA was significantly higher than that of those with WT1 levels of 5,000 copies/ μg RNA or more (41.2% [95% confidence interval (CI), 18.6-62.6] vs 10.3% [95% CI, 2.6-24.3]; P=0.006). On multivariate analysis of the non-CR patients, WT1 5,000 copies/ μg RNA or more was significantly associated with increased risk of mortality (hazard ratio (HR), 2.7; 95% CI, 1.3-5.5; P=0.008). Log₁₀-transformed WT1 (continuous scale) was also significantly related to increased risk of mortality (HR, 1.5; 95% CI, 1.0-2.1; P=0.037).

In the entire cohort, log₁₀-transformed WT1 (divided into four categories) before allo-HCT was found to be the most powerful prognostic factor of survival among remission status, cytogenetic risk group, and revised disease risk index, applying Akaike's information criterion. Moreover, under the assumption of a non-linear model, as the log₁₀-transformed WT1 level rose, the HR of mortality increased monotonically; a non-linear fit was obtained using a restricted cubic spline function (three knots; Figure 1).

CONCLUSION

Our data showed that WT1 levels obtained before allo-HCT monotonically increased HR of mortality in a non-linear manner regardless of CR status. Of importance is the finding that WT1 levels before allo-HCT were proportionally correlated with probability of survival after allo-HCT in non-CR AML patients. This suggests that the level of expression of WT1 mRNA in peripheral blood might reflect tumor burden and that there may be a WT1 threshold in non-CR patients for benefiting from allo-HCT. Further prospective studies of WT1 are necessary to determine acceptable or feasible levels of disease activity for receiving allo-HCT in relapsed or refractory AML patients.

View largeDownload slide

View largeDownload slide

Close modal