Validation of the Newly Proposed MD Anderson Prognostic Risk Model for Patients with Myelodysplastic Syndromes

Abstract

AbstractAbstract 444 Background:The international prognostic scoring system (IPSS) is the most widely used clinical tool for risk stratification and tailoring treatment in myelodysplastic syndromes (MDS). Despite its utility, the IPSS has several limitations. The IPSS was developed using outcomes of untreated primary MDS patients at time of diagnosis, and does not account for patient age, performance, and degree of cytopenia. The recently reported MD Anderson risk model (MDAS) addresses many of the limitations of IPSS (Kantarjian et al, CANCER September 15, 2008 / Volume 113 / Number 6). We validated this new risk model in a large external single institution cohort of patients. Methods:Data were collected retrospectively from Moffitt Cancer Center (MCC) MDS database and chart review of patients with MDS. The primary objective was to validate the new risk model calculated at time of initial presentation MCC. The MDAS was calculated as published based on age, performance status, blast%, degree of thrombocytopenia, cytogenetics, white blood cell count, and prior history if transfusion.Patients were divided into four risk groups: low (0-4 points), int-1 (5-6 points), int-2 (7-8 points), and high risk (≥ 9 points). All analyses were conducted using SPSS version 15.0. (SPSS Inc, Chicago, IL). The Kaplan–Meier method was used to estimate median overall survival. Log rank test was used to compare Kaplan–Meier survival estimates between two groups. Cox regression was used for multivariable analysis. Results:Between January 2001 and December 2009, 844 patients were captured by MCC MDS database. The median age was 69 years, MDS subtypes were coded as Refractory anemia (RA) 98 (12%), refractory anemia with ring sideroblasts (RARS) 76 (9%), MDS with del(5q) 20 (2.4%), refractory cytopenia with multi-lineage dysplasia (RCMD) 96 (11%), refractory anemia with excess blasts (RAEB) 255 (30%), therapy related MDS 22 (2.6%), and MDS-nos 275 (33%). IPSS risk groups were low risk in 158 (18.7%), intermediate-1 (int-1) 362 (42.9%), intermediate-2 (int-2) 168 (19.9%), high risk 45 (5.3%), and missing in 111 (13.2%).Based on the new risk model 169 patients (20%) were low risk, 250 (29.6%) int-1, 182 (21.6%) int-2, 135 (16%) high risk, and 94 (11.1%) were unknown. The median OS for all patients was 36 months (95% CI 31.5–40.5 mo). Age, IPSS risk group, serum ferritin, and RBC transfusion dependence were all significant prognostic factors in univariable analysis. The median OS was 92 mo (95%CI 68.1–115.9 mo), 49 mo (95%CI 40.4–57.6 mo), 26 mo (95%CI 21.2–30.8 mo), and 15 month (95%CI 11.8–42.1 mo) respectively for patients with low, int-1, int-2 and high risk patients according to MDAS. (Figure-1) (P < 0.005).In patients with low/int-1 IPSS risk group the median OS according to MDAS was 92 mo (95%CI 68.3–115.7 mo), 49 mo (95%CI 49.3–58.7 mo), 28 mo (95%CI 20.7–35.3), and 19 mo(95% CI 9.9–28.1 mo) respectively for patients with low, int-1, int-2, high risk MDAS (p<0.005). In patients with int-2/high IPSS risk categories only 4 patients were reclassified as low MDAS risk and the median OS for those patients was 10 month (95% CI 0–38 mo). The median OS was 49 mo (95%CI 23.5–74.5 mo), 23 mo (95%CI 19.4–26.6 mo), 14 mo (95% CI 11.5–16.5 mo). (p<0.005).For all the patients the rate of AML transformation according to MDAS was 5.9%, 16.8%, 36.3%, and 50.4% respectively for low, int-1, int-2, and high risk MDAS groups. (p <0.005). In Cox regression analysis, higher risk MDAS predicted inferior OS (Hazard ratio (HR) 1.54 (95%CI 1.35–1.75) (p <0.005) independent of IPSS risk group (HR 1.25 95%CI 1.1–1.45) (p =0.004). Conclusion:Our data validates the prognostic value of the MDAS risk model which was predictive for overall survival and AML transformation. The MDAS complements the IPSS particularly in low/int-1 risk group by identifying patients with higher risk disease behavior and inferior outcome. The utility of this model as a treatment decision tool should be studied prospectively. [Display omitted] [Display omitted] Disclosures:No relevant conflicts of interest to declare.