Monitoring of Incipient Haematological Relapse by CD34⁺ Donor Chimerism Determination in Patients with Acute Myelogenous Leukaemia and Myelodysplastic Syndrome after Allogeneic Hematopoietic Cell Transplantation (HCT) Following Low Dose TBI-Based Conditioning Regimens.

Introduction: Relapse of the malignant disease remains the major challenge after reduced intensity conditioning (RIC) HCT. Here we investigate the outcome of haematological relapse after low dose TBI-based conditioning regimens and the possibility to monitor minimal residual disease in patients with AML with the aim to prevent haematological relapse.

Patients and methods: From 7/98 – 12/06, 156 consecutive pts [86 m/70 f; median age 61 (range 21–75) y] with CD34⁺ AML, n=138 and high-risk MDS, n=18 received RIC-HCT. Stage of disease at HCT was CR1, n= 99 (63%), CR2, n=18 (12%), and >CR2, n=39 (25%). HCT were performed from matched related donors (n=40) or matched unrelated donors (n=116) after 200cGy TBI ± fludarabine 30 mg/m²/day on 3 days followed by mycophenolate mofetil and cyclosporine. Chimerism was monitored at days (d) 28, 56, 84, and at 3 months interval thereafter in unsorted as well as flow-sorted CD3⁺ and CD34⁺ bone marrow cells by FISH for the XY chromosome in gender mismatched (100–300 IP) or PCR based analysis of polymorphic micro satellite regions in gender matched HCT. A decrease of >5% CD34⁺ donor chimerism (“CD34⁺ relapse”) was followed by a taper of immunosuppression. Haematological relapse was defined as increase of blasts to >5%.

Results: Of the 156 patients, 141 pts engrafted as shown by donor T-cell chimerism. OS, RI, and NRM at 5 y were 38±0,05%, 47±0,06%, and 30±0,05% respectively. From the 141 patients, 48 had hematological relapse. Treatment of relapse included reduction of immunosuppression, low dose AraC and intensive chemotherapy ± mobilized buffy coat. CR was achieved in 50% of the patients. Reduction of immunosuppression and/or DLI (p=0,04), amount of overall donor chimerism at diagnosis (p=0,01), the interval from HCT to relapse (p=0,02), and the occurrence of GvHD (p=0,05) were factors associated with CR in the multivariate analysis. Survival of patients with haematological relapse, however, was poor (3,6±0.03% at 5a). All patients with haematological relapse within 100 d died despite therapy. Survival was independently influenced by the time of relapse after HCT (<100 d vs. >100 d; p=0.001), by the myeloblast count in the marrow (p=0.01) and by the type of relapse (haematological vs. “CD34 relapse” only). All patients with haematological relapse and available CD34⁺ donor chimerism (38/48 patients) had a decrease of CD34⁺ chimerism a median of 17 (range 0–181) d before the diagnosis of haematological relapse. Decrease of CD34⁺ chimerism without total chimerism decrease was observed in 12 additional patients (8,7%) a median of 149 (range 56–852) d after HCT. CD34⁺ donor chimerism amounted 84 (0–92)% before and returned to 100 (93–100)% after reduction of immunosuppression. Survival of these patients was 90±9% at 4 y.

Conclusions: We conclude that relapse after low-dose TBI based conditioning has a poor outcome with survival of <10% at 5 years in patients with AML. All patients with haematological relapse within 100 days died despite therapy. Determination of CD34⁺ donor chimerism allows early detection of relapse in patients with AML and can be used to tailor the immunosuppressive therapy after HCT.