Summary:Relapse postautograft in acute myeloid leukaemia (AML), may in part arise from leukaemia cells present in the bone marrow (BM) inoculum, and the level of minimal residual disease (MRD) in BM harvests used for autografting may therefore be clinically important.We have used the WT1 transcript as a marker of MRD, which was quantitated by RQ-PCR, in the BM harvests of 24 patients receiving an ABMT for AML. ABL was used as a control gene with WT1 level being normalised to 10 5 copies of ABL per sample. Median WT1 level was 651 copies (range ¼ 113-32 700) for the 13 patients with relapse-free survival (RFS) of less than 5 years, and 174 (range ¼ 0-1900) for patients with RFS of over 5 years postautograft (Po0.04). The RFS was 10.5 months for patients with WT1 level of 42000 copies (n ¼ 5), and has not yet been reached for patients with WT1 levelo2000 (n ¼ 21), at a median follow-up of 92 months (Po0.05). We show that elevated levels of MRD in BM harvests are associated with a higher relapse risk in patients autografted for AML. Autologous bone marrow transplantation (ABMT), as consolidation therapy, can prolong disease-free survival (RFS) for patients with acute myeloid leukaemia (AML) in first CR. 1-3 However, relapse remains the main cause of treatment failure, and may in part arise from leukaemia cells present in the bone marrow (BM) inoculum. 4 Thus, the level of minimal residual disease (MRD) in BM harvests may be relevant in determining the risk of relapse. Methods of detection of MRD with increased sensitivity, such as RQ-PCR are now available and quantitation of WT1 is useful in patients who lack specific fusion transcripts. This gene is overexpressed in over 90% of leukaemia cells and several studies have shown that quantitation of WT1 transcripts for patients in remission has prognostic value. [5][6][7][8] This study was undertaken to measure MRD levels in BM harvests used for autografting in AML, through WT1 quantitation, and to correlate these results with clinical outcome.
Materials and methods
PatientsBM harvest samples were available from 24 patients who had received an ABMT as consolidation treatment for AML between 1987 and 2000 (Table 1). There were 14 female and 10 male patients, with a median age of 42.5 years (range 16-55). Cytogenetics analysis at diagnosis showed two patients to be in a poor prognostic and 17 in a standard prognostic group using UK MRC criteria. 9 Five patients were in a favourable prognostic group (two with t(15;17), two with inv16, and one with t(8:21)) because the autograft was performed before risk stratification by cytogenetic analysis came into practice. Induction and consolidation treatments were given as per UK MRC 10 and 12 trials and responding patients were autografted after two courses of consolidation treatment. All patients were in morphological and cytogenetic remission at the time of BM harvest and also at autografting. A total of 18 patients were in 1st CR, five in 2nd and one in 3rd CR. Conditioning consisted either of standard Cy/TBI, Bu/Cy, or high-d...