According to the current European Leukemia Network guidelines (ELN 2022) for diagnosis and treatment of acute myeloid leukemia (AML) the prognostic risk stratification is based on molecular and cytogenetic characterization. Due to the small number of prospective Minimal Residual Disease (MRD) driven studies, the MRD role in AML risk stratification is not clearly defined.

The aim of this study was to evaluate the effect of MRD in a real-life setting. We retrospectively analyzed 124 patients with Favorable (FAV) or Intermediate (INT) risk AML treated with intensive chemotherapy from January 2017 to June 2023 at the Hematological Centers of the REV-network.

MRD was tested by RT-PCR. The MRD negativity was defined as undetectable MRD level, MRD Low Positive (MRD Low+) as MRD < 10-3, and MRD High Positive (MRD

High+) was set at level ≥ 10-3. Univariate statistical analysis was performed with χ2-test, log-rank test for survival outcomes (OS and PFS), and Grey-test for cumulative incidence of relapse (CIR). Cox regression analysis and Fine-Gray proportional hazard regression model were used in multivariate analysis. The statistical significance was set at ≤ 0.05.

Patient characteristics were: male/female ratio 59/65, median age 60 years (range 20-74), ECOG 0 (range, 0-4), Charlson Comorbidity Index modified sec. Deyo 2 (range, 0-5). Fifty- nine patients (48%) had FAV-risk and 65 (52%) INT-risk AML, with the following cytogenetic/molecular features: normal karyotype (75%), core binding factor alterations (6%; 4 RUNX1-RUNX1T1, 4 CBFB-MYH11), NPM1 mutations (73%), FLT3-ITD mutations (40%), FLT3-TKD mutations (7%), and CEBPA bZIP mutations (1%). MRD was evaluable in 80 patients (65%) and in 44 MRD was unknown (MRDu).

OS and PFS of the whole cohort were 69% and 60% at 2 years (2y), and 65% and 60% at 3 years; 2y-CIR was 28%. FAV and INT patients did not show significant differences in clinical presentation, complete response (CR) rate, MRD negativity rate, OS, PFS and CIR.

In univariate analysis, factors associated with higher 2y-OS were post induction CR (76% vs 8%, p<0.0001), MRD negativity after 2 chemotherapy cycles (2CT MRD- 100% vs MRD Low+81% vs MRD High+ 62% vs MRDu 57%, p=0,009), and with End of Treatment (EoT) MRD negativity (EoT MRD- 97% vs MRD Low+ 76% vs MRD High+ 47% vs MRDu 50%, p<0,0001). In multivariate analysis, EoT MRD High+ and EoT MRDu retained a significant negative impact on OS (HR 3,5, p=0,0001 and HR 3,2, p<0,0001).

In univariate analysis, 2y-PFS was negatively influenced by: age > 60 (50% vs 69% p=0.02), PLT < 69 x 103/mL (53% vs 67%, p=0,04), 2CT MRD High+ (MRD- 100% vs MRD Low+ 73% vs MRD High+ 37% vs MRDu 50%, p=0,006) and EoT MRD High+ (MRD- 87% vs MRD Low+ 65% vs MRD High+ 30% vs MRDu 44%, p<0,0001). Multivariate analysis confirmed EoT MRD High+ and EoT MRDu as negative predictors for PFS (HR 2,8 p<0,0001 and HR 2,3 p=0,006).

CIR correlated with GB count > 31,3 x 103/mL (37% vs 19%, HR 2,09, p=0,03) and with EoT MRD+ regardless of MRD value (38% vs 21%, HR 1,9, p=0,04).

A stratified analysis was performed to evaluate the HSCT-effect depending on MRD status. HSCT positively affected outcomes in patients 2CT MRD High+ (PFS and OS, 50% vs 25%, p=0,04), 2CT MRD+(Low/High) (2y OS 74% vs 56%, p=0,04; 2y PFS 70% vs 54%, p=0,04), and in EoT MRD+ (2y OS 87% vs 52%, p=0,04; 2y PFS 73% vs 40%, p=0,04). HSCT+ and HSCT- patients did not show different outcomes for MRD-, MRD Low+ and MRDu status.

In HSCT+ subgroup, 2y OS and PFS were higher in patients with pre HSCT MRD- (100% vs 66%, p=0,04; 00% vs 62%, p=0,04) and with CR1 vs CR>2 status at transplant (81% vs 62%, p<0,0001; 80% vs 12%, p<0,0001).

In our real-life subset of FAV-INT ELN 2022 AML patients, MRD status is the main prognostic factor. This result highlights the relevance of improving MRD evaluation in INT risk patients for driving therapeutic decisions.

Despite the relevant negative prognostic impact of pre HSCT MRD positivity, allograft is still necessary to ensure the best survival for MRD+ patients. In MRD negative patients, on the contrary, performing HSCT does not affect the outcome.

Given the retrospective design and the relatively small number of patients, further studies in a larger population could be useful to confirm our findings, as well as for answering the main open question: is allogenic transplantation still a cornerstone in intermediate risk AML patients who reach MRD negativity?

Disclosures

No relevant conflicts of interest to declare.

This content is only available as a PDF.
2024
Sign in via your Institution