The methodology and significance of minimal residual disease detection after allogeneic stem cell transplantation

Abstract

Allogeneic stem cell transplantation (SCT) is the choice of therapy for leukemia patients who respond poorly to conventional chemotherapy. Despite high remission rates after SCT, relapse of the underlying disease remains one of the most frequent causes of treatment failure. Graftversus-host disease (GVHD), a major complication after SCT, is caused by the activation of alloreactive donor T-cells. Although being life threatening in its severe forms, GVHD has a protective effect called the graft-versus-leukemia effect (GVL). In order to use the GVL effect of donor T-cells, donor lymphocyte infusions (DLI) is now used as a treatment for relapse after SCT. Response to DLI is usually better when the tumor load is low. Therefore, sensitive methods to detect residual leukemic cells are needed in order to identify patients who are at the highest risk of relapse and to start immunotherapeutic interventions when the tumor load is still low. Minimal residual disease (MRD) refers to the presence of leukemic cells below the detection limit of standard morphological analysis. The most sensitive and widely used techniques for MRD detection are based on the PCR technology. The aim of this thesis is to evaluate the clinical significance of MRD detection in leukemia patients receiving SCT.

In patients with acute myeloid leukemia (AML), we evaluated the significance of mixed chimerism (MC) analysis, the detection of recipient-derived hematopoietic cells after SCT. MC analysis was performed in the leukemia-affected cell lineage to increase the specificity and sensitivity of the method. MC was detected in 14 of 30 patients. Ten of these 14 patients relapsed as compared to 2 of 16 with donor chimerism (p<0.01). MC was detected a median time of 66 (range 23-332) days before hernatological relapse.

Using immunoglobulin and T-cell receptor gene rearrangements as clonal markers, we analyzed MRD levels before and after SCT in patients with acute lymphoblastic leukemia (ALL). MRD detection before SCT was associated with increased risk of relapse. However, GVHD was shown to protect against relapse in patients with high levels of MRD. MRD detection after SCT was also associated with a high risk of relapse. Relapse occurred in 8 of 9 MRD positive patients as compared to 6 of 23 MRD negative patients (p<0.01). MRD was detected a median of 5.5 (range 0.5-30) months before relapse.

In recent years, nonmyeloablative SCT (NST) has been studied as a safer approach for patients who are not eligible for the toxic conditioning regimens given before SCT. We studied the kinetics of MRD and MC in chronic myeloid leukemia (CML) patients receiving NST and compared the results to those obtained from CML patients receiving a conventional SCT (CST). A competitive PCR approach was performed for quantitative MRD analysis of BCR-ABL transcripts. In the early posttransplant period, higher incidence and levels of MC and MRD were found in NST patients as compared to CST patients. However, molecular remissions were subsequently achieved in most NST patients.

Wilms tumor gene (WTI) has been reported as a "panleukemic" MRD marker in many studies. We wanted to evaluate WTI as a MRD marker by comparing the kinetics of WTI levels with that of BCR- ABL using realtime quantitative PCR. We found a background expression of WTI healthy controls. In addition, WTI analysis was not sensitive enough to predict relapse.

In conclusion, MRD analysis in leukemia patients provides the possibility to identify patients at high risk of relapse after SCT. Adoptive immunotherapy based on MRD results may prevent relapse and improve outcome for patients with poor prognosis.