Impact of Transfusion Dependency on Patients Diagnosed with Myelodysplastic Syndrome According to IPSS-M Risk Group

Introduction: The IPSS-M (Bernard NEJM Evid 2022) allows a better definition of overall survival (OS) and the risk of progression to acute myeloid leukemia (AML) in patients with myelodysplastic syndrome (MDS). Red blood cell transfusion dependency (TD) is associated with lower OS in MDS patients (Malcovati L et al; Leuk Res 2007), but it was not included in the analysis due to a low proportion of patients with these data availability. Our hypothesis is that the inclusion of TD can improve the prognostic assessment of IPSS-M.

Methods: A total of 214 patients diagnosed with MDS or myelodysplastic CMML (<13x109 leukocytes/L) in 13 centers were included in the analysis. TD was defined as receiving at least one red blood cell transfusion every 8 weeks for 4 months (Malcovati L et al; JCO 2007). Due to the limited number of patients, we established two risk categories according to IPSS-M. The low-risk group (IPSS-M-LR): IPSS-M of very low, low, and low intermediate risk, and high-risk (IPSS-M-HR): IPSS-M of high intermediate, high, and very high risk.

Results: We included 124 (57.9%) males and 90 (42.1%) females. Mean age at diagnosis was 69 years old (range: 31-94). According to ICC-2022 classification, the most frequent diagnosis were MDS with multilineage dysplasia (10.7%), MDS/AML with TP53 mutation (15%) and MDS/AML, NOS (13.1%). According to WHO-22, the most frequent entities were MDS with excess blasts type 2 (50%), followed by MDS with low blasts (12.6%) and MDS with biallelic inactivation of TP53 (11.7%) and 15% of patients had CMML according to both classifications. 23% of the patients had therapy-related MDS. According to IPSS-M, 20 (9.3%) patients had very low risk, 25 (11.7%) low risk, 20 (9.3%) moderate-low risk, 27 (12.6%) moderate-high risk, 42 (19.6%) high risk and 80 (37.4%) very high risk. After dichotomizing the patients according to IPSS-M, 29.9% were IPSS-M-LR and 70.1% were IPSS-M-HR. 58.4% had TD: 17.1% of the patients in the IPSS-M-LR group and 76% in the IPSS-M-HR group. Median follow-up among surviving patients was 19.4 months. Median OS for IPSS-M-LR MDS and IPSS-M-HR patients was (10.4 years; 95% CI 3.1-17.6) and (1.6 years; 95% CI 1.1-2.1) respectively (p<0.001). According to TD, median OS was significantly lower in TD patients compared to non-TD (NTD) patients (1.49 years vs NR; p<0.001). Analyzing the impact of transfusion dependency into IPSS-M group: IPSS-M-HR group TD patients showed significantly lower median OS compared to NTD patients (1.31 years vs. NR; p <0,001). In IPSS-M-LR group no differences were observed between TD and NTD patients OS.

In a multivariate analysis performed including age at diagnosis, gender, dichotomic IPSS-M and TD status at diagnosis, age (p <0.001), IPSS-M-HR (p 0.003; HR 4.4) and TD (p 0.006; HR 3.6) were independent factors of shorter OS.

Conclusions: Our results suggest the negative impact of TD on OS in patients with moderately high, high, and very high risk IPSS-M. We cannot conclude this for low-risk patients, possibly due to a limited number of patients and events. Since therapeutic approaches are based on the initial risk assessment, including TD status in this stratification can provide more accurate prognostic information that might potentially impact on the treatment strategy. Validation of these results in larger series is necessary.