Background：

Acute myeloid leukemia (AML) is an aggressive hematologic malignancy characterized by accumulation and expansion of immature myeloid cells in the bone marrow, peripheral blood and other tissues. Despite the progress made in mechanisms and approval of new therapeutics, the overall prognosis remains poor, with a 5-year survival rate of less than 30%.The ELN 2022 guidelines categorize patients with AML into three groups based on fusion genes, genetic mutations, and cytogenetic abnormalities, while excluding other characteristics of patients such as the age of onset, sensitivity to chemotherapy, and relevant biochemical indicators.

Study Design and Methods:

We retrospectively collected the clinical data of 718 patients with newly diagnosed primary AML (excluding acute promyelocytic leukemia) treated at our center between January 1st 2019 and April 30th 2023. The 718 patients were randomly divided into training (603 patients) and validation (115 patients) sets.The diagnosis and treatment strategies were based on European Leukemia Net (ELN) criteria and National Comprehensive Cancer Network (NCCN) guideline. Composite complete remission (CRc) rate (CR+CR with incomplete recovery [CRi]) were assessed at the completion of induction or re-induction therapy.Overall survival (OS) was measured from the day of diagnosis until death or the last follow-up date.

Results:

The median (range) follow-up duration for the training and validation sets was 22.8 (0.830-55.530) and 21.37(1.43-54.93) months, respectively. There were no significant differences in the clinical characteristics at the onset of the disease and induction regimen between the two sets. The 1-year, 2-year, and 3-year OS rates were 79.9%, 61.9%, and 51.9% in the training set and 76.1%, 59.0%, and 56.5% in the validation set, respectively.

Cox regression analysis was performed on the training set,and the patient's age of onset, lactate dehydrogenase (LDH) level at initial diagnosis, CBFB::MYH11 gene fusion, KMT2A rearrangement, TP53 mutation (VAF > 10%), DNMT3A mutation, RUNX1 mutation, and achievement of CRc in the first course of treatment were considered to be independent predictors and were used to build the model, named as ZJ-AML model. The C-index in the training set was 0.748 (95% CI 0.713-0.783), the ROCs at 1-year, 2-year, and 3-year time points were 0.794, 0.757, and 0.764, respectively,while the C-index in the validation set was 0.825 (95% CI 0.774-0.876), and the ROCs at 1-year, 2-year, and 3-year time points were 0.906, 0.839, and 0.870, respectively,indicating a good discrimination in two sets.

We assigned values to the variables according to the coefficients of the variables in multivariate Cox regression in the training set to calculate the risk score and divide the whole cohort into three risk groups.As seen in the figure, the low-risk population in the ZJ-AML model had more optimized survival than the ELN 2022 low-risk population, while the high-risk population had less optimized survival than the ELN 2022 high-risk population.

According to the risk grouping in this model, OS was analyzed using transplantation status in patients under 60 years of age in the study population. In the low risk group, there was no significant difference in survival time between the two groups,while patients underwent transplantation both in the intermediate and high risk groups showed significant superior survival than those without transplantation .

We divided the first-course regimens into three groups: regimen 1 (intensive chemotherapy combined with venetoclax regimens), regimen 2 (intensive chemotherapy), and regimen 3 (venetoclax based low-intensity chemotherapy).The subgroup analysis by induction regimen revealed that while ELN 2022 could only distinguish the survival outcomes of patients receiving intensive chemotherapy for induction, ZJ-AML model could distinguish the survival outcomes across the three regimens.

Conclusion:

The ZJ-AML prognosis prediction model in this study may offer more optimized discrimination compared to ELN 2022 stratification, particularly for patients with AML receiving combination therapy with targeted BCL-2 inhibitors.

Disclosures

No relevant conflicts of interest to declare.

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