Allogeneic Transplanst for ACUTE Myeloid Leukemia: Cut Off Levels of WT1 Expression in 207 Patients and Pre-Emptive Therapy of Relapse

Leukemia relapse remains a significant problem in patients with AML undergoing an allogeneic stem cell transplant(HSCT). Wilms Tumour 1 (WT1) expression has been shown to be a sensitive marker of minimal residual disease (MRD), both in patients after induction chemotherapy, as well as in patients undergoing an allogeneic HSCT.

Hypotheses. The present study had 2 hypotheses: (1) WT1 expression in marrow cells of AML patients post-HSCT, will predict leukemia relapse and (2) WT1 based pre-emptive immunotherapy (IT) such as abrupt cyclosporin discontinuation and/or donor lymphocyte infusion (DLI), will prevent leukemia relapse.

Patients. Bone marrow WT1 expression, was monitored in 207 patients with acute myeloid leukemia (AML) before and monthly after an allogeneic HSCT, until day +150, and then at every other outpatient access. Eligible for IT were patients without acute or chronic GvHD, with increased WT1 expression and a a marrow in hematologic remission. The trigger for IT was 180 WT1 copies in a first group of 122 patients (group A): this was based on the fact that WT1 expression in normal bone marrow is up to 180 copies . In a subsequent group of 85 patients (group B) the cut off for IT, was 100 copies, due to the fact that a first analysis of group A had shown 100 copies to be an earlier predictor of relapse (BJH 2013; 160: 503). DLI were given in escalating doses, starting at 1x105 CD3+ cells/kg in alternative donor grafts and at 1x106/kg in HLA identical grafts. DLI were escalated ½ log every month, in the absence of GvHD, to a maximum dose of 1x107/kg. Sixtyfour patients were eligible for IT, but only 35 received IT: reasons for non intervention were ongoing GHD, unavailable donor and delay in WT1 results.

Results-Hypothesis N.1.

Following transplantation, WT1 expression, was highly predictive of leukemia relapse: 12 relapses in 99 patients with WT1 < 100 copies /104 abl (12%); 19 relapses in 55 patients with WT1 between 101 and 180 copies (35%) and 37 relapses in 53 patients with WT1 >180 copies (70%) (p<0.0001). The median interval between WT1 positivity and relapse was 75 days in group A and 60 days in group B.

Results-Hypothesis N.2.

35 patients received pre-emptive immune intervention, 17 in group A and 18 in group B. The latter had more patients beyond first remission at transplant (56% vs 23%) , more myeloablative regimens (100% vs 65%) and more family haploidentical donors (72% vs 6%); age was comparable. The risk of relapse was 13/17 (76%) for group A and 3/18 (17%) for group B (p<0.001), despite the larger proportion of patients beyond CR1 at transplant. GvHD following DLI occurred in 15% of patients. DLI-related mortality was 0%. The overall 3 year survival for patients in group A and B was 69% vs 47% (p=0.3). The relapse risk in patients of group A eligible but not receiving IT (n=21) was 74%; in group B (n=8) it was 50%.

In conclusion, WT1 expression post-transplant is a strong predictor of leukemia relapse in patients with AML, and can be used to trigger pre-emptive immunotherapy, in approximately 50% of eligible patients. IT triggered at a WT1 cut-off level of 100 copies in bone marrow cells, is more effective, as compared to180 copies, in preventing leukemia relapse.