Abstract

Abstract 3541

Poster Board III-478

Introduction

Quantitative analysis of chimerism after allogeneic hematopoïetic stem cell transplantation (allo-HSCT) for acute leukemia is routinely used to monitor the kinetic of engrafment. Usually, chimerism is assessed by variable number tandem repeat (VNTR) or short tandem repeat (STR) amplification by polymerase chain reaction (PCR), which have a sensitivity of 1 to 5 %. Many studies have shown that these methods were not sensitive enough to predict relapse. Insertion/Deletion (InDel) polymorphism analysis by real-time quantitative PCR (InDel-QPCR) is a more sensitive method and thus it should be able to predict relapse earlier.

Patients and methods

We conducted a retrospective unicentric study including all consecutive patients transplanted for acute leukemia from May 2004 to January 2009 in the Pitié Salpétrière Hospital (Paris France). Seventy four patients (53 acute myeloblastic leukemia and 21 acute lymphoblastic leukemia) were included. Median age was 41 years (18-67). Conditionning regimen was myeloablative in 51 patients (69%). The donor was a HLA-identical sibling for 31 patients. The source of stem cell was bone marrow in 35 patients (47%), peripheral blood in 35 (47%) and umbilical cord blood in 4 (6%). Sixty four patients (86%) were in complete remission (CR) at the time of transplantation. InDel-QPCR was performed every month in peripheral blood samples with a reproducible sensitivity of 0.1% at least. An increasing mixed chimerism was defined as a one log increase between two successive chimerism assays when recipient rate was inferior to 0.1% and as any increase beyond this limit (0.1%).

Results

In this 74 patients population, overall survival was 65% at one year, with a median follow up of the survivors of 651days (129-1809). Thirty six patients developed an acute GVHD (2-4) and 21 a chronic GVHD. Eighteen patients (24%) presented a cytological relapse, at a median time of 203 days (63-1001) after transplant. In two patients, the quantification of recipient DNA rate was constantly superior to 1% at each time point post transplant and these patients presented an early relapse (at 3 and 4 months after transplant). In the 72 remaining patients, DNA rate was inferior to 1%: in 63 patients inferior to 0.1% and in 41 patients inferior to 0.01%. Among the 62 patients who presented an increasing mixed chimerism (IMC) at one point, 18 relapsed. Among the 24 patients who presented an IMC at two successive points, 16 relapsed. In univariate analysis, factors associated with higher risk of relapse were the status of disease at transplant (refractory disease+CR2 versus CR1, p=0.002, hazard ratio=9.54) and 2 successive IMC (p=0.0001, hazard ratio=8.74 (CI: 3.33-22.9). In multivariate analysis both factors remained significantly correlated with the incidence of relapse (status of disease p=0.003, hazard ratio=4.24(CI: 1.6-11.2) and two IMC p<0.0001, hazard ratio=9.20(CI: 3.44-24.57)).

The median interval between the first IMC and the diagnosis of relapse was 45 days (7-154). At the time of the first IMC, the peripheral blood cell count was normal in all except one patient who had persistent post-transplant thrombocytopenia.

Conclusion

Analysis of chimerism by using In/Del polymorphism is a sensitive technique with a reproducible detection threshold inferior to 0.1%. The detection of two successive IMC is highly predictable of relapse in patients transplanted for acute leukemia. Therefore, the detection of an increasing chimerism should incite to perform a rapid control, in order to make a precocious diagnosis of relapse and to provide an early therapeutic intervention. This analysis could help to monitor minimal residual disease in post transplant patients without a more specific molecular marker of malignancy.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.