BONE Marrow Chimerism and Measurable Residual Disease Analysis Interplay: Impact on Prognosis of ACUTE Myeloid Leukemia Undergoing STEM Celltransplatation

Measurable residual disease and chimerism analysis help to identify patients with acute myeloid leukemia (AML) at higher risk of relapse after allogeneic SCT (allo HSCT). Moreover, our growing understanding of relapse biology in AML is leading to development of therapeutic

interventions to diminish risk of relapse.

To describe the utility of chimerism and measurable residual disease analysis to predict relapse in patients with AML after allo-HSCT.

We retrospectively collected data from AML patients in remission with or without hematologic recovery treated with allo-HSCT in our centre between 2010 and 2021. MRD was analyzed both by RT-PCR in bone marrow aspirates (BM) and/or peripheral blood (PB) (NMP1, RUNX1-

RUNX1T1, CBFB-MYH11, KMT2A-MLLT3, WT1) and by flow cytometry (FCM) in BM at pre-HSCT (within 30 days), at day +30 and +90 post-transplant. MRD positivity was defined as above cut- off level MRD detection by either FCM or RT-PCR in a given timepoint. Whole BM cellularity was analyzed by STR at days +30 and +90 post-transplant. Information regarding prophylaxis (Pr) or pre-emptive (PE) strategies used to prevent AML relapse was also collected.

Data from 147 patients were analyzed. Median follow-up was 50 months (IQR 14-48). Median age was 47 years (IQR 38-57). 53% of patients were male. 43% of patients were classified as adverse risk according to ELN 2017 risk stratification. 13% of patients had complex karyotype.

42 patients were NPM1 positive and 45 patients FLT3-ITD positive. 21 patients received some kind of relapse preventive strategy (18 PE, 4 PR; one of them received both).

47% haploidentical, 24% MSD, 20% MUD, 7% UCB and 1% MMUD. Stem cell source was PB in 92%, BM in 1% and UCB 7%. GVHD prophylaxis was PT-Cy in 77%; MTX-CNI 15% and prednisone-CNI 7%. Myeloablative conditioning was used in 56% of patients.

76 patients (52%) were MRD+ pre-transplant. Immunosuppression withdrawal started earlier in post-transplant period in patients pre-transplant MRD+ than in those pre-transplant MRD- (73.1 vs 56.8 days, p=0.018). No differences in 5-year-OS or 5-year-RFS were found according to pre-transplant status.

33 patients out of 145 evaluable patients (23%) were MRD+ at day +30. MRD+ patients showed significantly lower 5y-RFS than those MRD- (48 vs 68%, p=0.03) and a trend towards lower 5y-OS (54 vs 70%, p=0,071). According to MRD status at day +90, we found differences between those MRD+ and those MRD- in 5y-OS (54 vs 77%, p=0.015) and 5y-RFS (45 vs 74%, p=0.001).

Chimerism analysis at day +30 comparison between those in mixed chimerism (MC) and those in complete chimerism (CC) only found a trend to worse 5y-OS (57% vs 71%, p=0,14) and 5y- RFS (50 vs 58%, p=0,077). Besides, BM chimerism analysis at day +90 found that those in MC

compared to those in CC had worse 5y-OS (45% vs 82%, p<0.001) and worse 5y-RFS (35 vs 74.3%, p<0.001). No differences were found between patients MRD+ compared to MRD- if CC was achieved at day +90 in terms of 5y-OS (76.9 vs 86, p=0.7) or 5y-RFS (76.9 vs 78.7%, p=0.9). However, if chimerism status at day +90 was MC, MRD+ compared to MRD- showed worse 5y-OS (27.3 vs 66.7%, p=0.017) and worse 5y-RFS (9.1 vs 66%, p<0.001).

MRD and chimerism interplay in determining relapse risk. In our experience, pre-transplant MRD positivity was not predictive for relapse, possibly due to quick use of relapse mitigating strategies in those patients. However, MRD positivity at day +90 and, mainly, its combination with chimerism analysis was found to be more accurate in outcome prediction in AML transplanted patients.

Garcia-Sanz:Janssen: Consultancy, Honoraria; BMS/Celgene: Consultancy, Honoraria; Novartis: Consultancy, Honoraria; Amgen: Consultancy, Honoraria; MSD: Honoraria; Takeda Pharmaceutical: Consultancy, Honoraria. Kwon:Gilead-Kite: Honoraria, Research Funding, Speakers Bureau; Pfizer: Speakers Bureau; Sanofi: Honoraria; Jazz: Speakers Bureau.