Pre-Transplant Multiparameter Flow Cytometric Measurable Residual Disease in Acute Myeloid Leukemia As an Independent Prognostic Factor for Survival after Myeloablative Allogeneic Hematopoietic Cell Transplantation

Background Growing evidence suggests that multiparameter flow cytometric (MFC) measurable residual disease (MRD) assessment in acute myeloid leukemia (AML) should be used for risk stratification prior to allogeneic hematopoietic stem cell transplantation (alloHCT). We investigated the significance of MFC-MRD status in AML patients treated with myeloablative conditioning (MAC) alloHCT at transplant centers within the Polish Adult Leukemia Group (PALG). Methods All consecutive patients who underwent MAC alloHCT from HLA identical siblings (SD), family haploidentical donor, matched (10/10) unrelated donor (MUD), or mismatched (9/10) unrelated donor (mMUD) between January 2014 and May 2022 and for whom pre-transplant MFC-MRD was assessed were included in the analysis. MFC-MRD assessment has been routinely performed on bone marrow aspirates obtained before alloHCT since 2014 or as part of the PALG-AML1/2016 study protocol (NCT03257241) since 2018. Flow cytometry laboratory protocols were modified and unified by the cooperating laboratories according to the European LeukemiaNet recommendations in 2018. Residual disease at a level ≥0.1% was considered MRD(+). Results We analyzed 302 adult patients (median age 44 years, range 18-68) with AML in first (88%) or subsequent CR (12%) after intensive chemotherapy, including 81 patients (27%) transplanted between 2014-2017 and 221 (73%) transplanted between 2018-2022. Positive MRD status before alloHCT was detected in 143/302 pts (47%). The European LeukemiaNet (ELN) cytogenetic/genetic risk (Dohner 2010) was available in 290 of the 302 patients and was intermediate-II (Int-II)/high in 22%. Patients received MAC based on i.v. busulfan (84%), treosulfan (1%) or total body irradiation (15%) and graft from either SD (29%), haploidentical donor (9%), MUD (55%), or mMUD (7%). GvHD prophylaxis consisted of calcineurin inhibitor (CNI) combined with MTX (with or without ATG) or was based on post-transplant cyclophosphamide in transplants from haploidentical donor or mMUD. After the median follow-up of 30 months (range, 1-97), the 3-year LFS and OS for MRD(-) vs MRD(+) patients was 69% vs 54% (log-rank p=0.007), and 77% vs 62% (log-rank p=0.013), respectively. In multivariate Cox model, the independent adverse prognostic factors for LFS were MRD positive status (HR 1.53, 95%CI 1.03-2.25; p=0.0332), Int-II/high ELN risk (HR 1.87, 95%CI 1.24-2.84; p=0.0031), and age ≥ 45 years (HR 1.65, 95%CI 1.12-2.45; p=0.011). The same factors independently influenced OS [HR 1.59, 95%CI 1.01-2.48; p=0.0411; HR 1.67, 95%CI 1.03-2.71; p=0.0363, and HR 1.79, 95%CI 1.42-2.79; p=0.0109, respectively]. When the patients were classified according to the number of identified independent risk factors, the 3-year LFS and OS for patients with 0, 1, 2 and 3 risk factors was 75%, 68%, 51%, and 27% (p<0.0001) (Figure 1), and 81%, 77%, 57%, and 44%, respectively (p=0.0006). Conclusions Our findings confirm that pre-transplant residual disease at a ≥0.1% level assessed by MFC is an independent poor prognostic factor for both LFS and OS in AML patients treated with MAC alloHCT. As predicted, outcomes after MAC alloHCT were also independently influenced by ELN cytogenetic/genetic risk and age. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal