Molecular Taxonomy Classification for Transplant Decision Making of Myelodysplastic Syndromes

Background: Myelodysplastic syndrome (MDS) is a heterogenous group of hematological malignancy and patients are at risk of developing acute myeloid leukemia (AML). The only curative treatment for MDS patients is hematopoietic stem cell transplantation (HSCT). The molecular analysis for HSCT patients (Yoshizato T et al., Blood, 2017) has indicated patients with TP53 mutations or complex-karyotype (CK) alone benefit from HSCT and a recent paper (Versluis J et al., 2023) has suggested this is the case regardless of TP53 allelic status. Although these previous studies have demonstrated the impact of single genetic abnormalities on the prognosis of HSCT patients, the impact of molecular classification of MDS still remains unclear. A recent study has grouped recurrent gene mutations to form a molecular taxonomy classification for MDS (Bernard E et al., Blood, 2024). However, the prognostic implications of the molecular taxonomy on HSCT patients have not been investigated. Here, we aim to elucidate which molecular taxonomy groups benefit from HSCT.

Method: We re-analyzed a published dataset (Yoshizato T et al., Blood, 2017) of 797 myeloid neoplasm patients who underwent HSCT. Genetic abnormalities for these patients were analyzed through targeted-capture sequencing of 69 myeloid driver genes, along with 1,158 SNP-probes to identify copy-number alterations. Results from the targeted-capture sequencing analysis were used to group the patients into the molecular taxonomy groups.

Result: Our cohort predominantly consists of relatively younger patients diagnosed with higher-risk MDS as all patients were eligible for and received HSCT. Of the 797 patients, 63% were male, and the median age at diagnosis was 51 yrs (range: 12-65) and the median age at transplantation was 53 yrs (range: 16-66). The median time from diagnosis to transplant was 9.4 months (range: 0.2-363). The diagnosis of these patients at the time of transplant was 30.4% low-risk MDS, 35.5% high-risk MDS, 23.7% sAML and 5.8% MDS/MPN.

The molecular taxonomy groups in our cohort were as follows: 3.6% DDX41, 6.1% AML-like, 17.2% TP53-CK, 3.9% der(1;7), 7.5% -7/SETBP1, 1.1% del(5q), 0.5% EZH2-ASXL1, 2.4% IDH-STAG2, 1.0% bi-TET2, 1.0% U2AF1¹⁵⁷, 6.4% U2AF1³⁴, 1.4% SRSF2, 0.9% ZRSR2, 3.0% SF3B1, 6.0% CCUS-like, 5.0% mNOS, and 28.4% no event, which might be attributed from the smaller number of analyzed gene in our study compared to the original taxonomy paper. A noticeable difference between the frequency of each group in the published molecular taxonomy groups vs. our cohort is der(1;7) (0.5% vs. 3.9%), del(5q) (6.9% vs. 1.1%), EZH2-ASXL1 (0.5% vs. 2.4%), bi-TET2 (12.8% vs. 1.0%), U2AF1³⁴ (2.1% vs. 6.4%), and SF3B1 (14.1% vs. 3.0%). The high frequency of der(1;7) in our cohort is consistent with our previous findings (Okuda R et al., ASH 2023), as this unbalanced translocation has been found to be more frequent in Asian MDS compared to Caucasian MDS. The underrepresentation of bi-TET2 and SF3B1 may be due to the high proportion of high-risk MDS cases compared to the molecular taxonomy paper.

In survival analysis, the two-year OS rates were as follows: 56.62% for DDX41, 46.1% for AML-like, 24.7% for TP53-CK, 50.5% for der(1;7), 58.9% for -7/SETBP1, 37% for del(5q), 0% for EZH2-ASXL1, 57.4% for IDH-STAG2, 54.0% bi-TET2, 14.6% for U2AF1¹⁵⁷, 44.3% for U2AF1³⁴, 45.5% for SRSF2, 85.7% for ZRSR2, 68.2% for SF3B1, 60.5% for CCUS-like, 58.8% for mNOS, and 67.2% for no event. In a multivariate analysis conducted on TP53 mutation status, RAS pathway gene mutation status, and taxonomy groups, TP53-CK and U2AF1³⁴ group showed a significant impact on OS (p<0.001 and p<0.022, respectively), and it was found the TP53 mutation status and RAS pathway mutations were the two factors that impact OS in HSCT patients (p<0.001), as previous studies have indicated.

Conclusion: Through a large-scale genomic analysis of HSCT patients, we demonstrated the distribution of each molecular taxonomy group in an HSCT cohort and uncovered a group of patients who may benefit from HSCT. We observed a noticeable difference in the frequency of several molecular subgroups probably due to the difference in the proportion of ethnicity and high-risk diseases. Our analysis indicates that as previous studies have indicated, TP53 and RAS pathway mutations are the two key gene mutations that impact OS in HSCT patients. Taxonomy group, U2AF1³⁴, is a group of MDS that may predict poorer survival in HSCT patients.