Therapy-Related Myelodysplastic Syndrome (MDS) Lacking High-Risk Karyotype Resembles De Novo Disease and Differs from Conventional Therapy-Related MDS.
Robert P Hasserjian, Yang OK Chi, Ying Hu, Michael Kluk, Eyal Attar, Sa A Wang. Massachusetts General Hospital, Boston; UMass Medical Center, Worcester, MA; UT MD Anderson Cancer Center, Houston, TX; Brigham & Women's Hospital, Boston, MA
Background: Therapy-related MDS (t-MDS) is an aggressive myeloid neoplasm and over 80% exhibit adverse cytogenetic features. According to the 2008 WHO Classification, t-MDS need not be further classified as blast count does not appear to affect prognosis, unlike de novo MDS. However, the clinical behavior of the small subset of t-MDS cases that lack adverse cytogenetic features is uncertain.
Design: We studied consecutive cases of t-MDS following cytotoxic therapy (chemotherapy+/-radiation) with low or intermediate-risk (LI) karyotype (t-MDS-LI). The authors confirmed all diagnoses of MDS by review of the slides and medical records. Cases were classified according to the 2008 WHO. Clinical features, treatment, and patient outcome were compared to control cohorts of t-MDS with high-risk (HR) karyotype (t-MDS-HR) and de novo MDS lacking history of exposure to cytotoxic agents with low or intermediate-risk karyotype (DN-MDS-LI)
Results: 53 t-MDS-LI patients were identified; the control groups comprised 34 t-MDS-HR and 159 DN-MDS-LI patients. The MDS-LI patients were younger (median age 63 vs 70 y, p<0.0001) with lower marrow cellularity (median 40% vs 60%, p<0.0001) and platelet count (median 60 vs 120, p=0.001) than the DN-MDS-LI patients; other peripheral counts, bone marrow and blood blast count, and WHO diagnosis distribution were similar in all three groups. The median overall survival (OS) of the t-MDS-LI patients was 27 m, compared to 5 m for t-MDS-HR patients (p<0.0001) and 27 m for DN-MDS-LI patients (not significant). The t-MDS-LI patients' OS was correlated with WHO diagnosis (p=0.029), bone marrow blast count (p=0.02), and hemoglobin level (p=0.036), but not platelet count, latency, or patient age; conversely, none of these parameters correlated with OS of the t-MDS-HR patient group.
Conclusions: t-MDS with LI risk karyotype has a markedly superior prognosis to t-MDS with HR karyotype and, unlike the latter, can be risk-stratified by bone marrow blast count and WHO diagnosis. These features as well as patient survival are similar to de novo MDS with LI risk karyotype. Our findings suggest that MDS that develops after cytotoxic therapies but lacks adverse cytogenetic features does not share the dismal prognosis characteristic of other t-MDS. Subclassification of these cases by WHO diagnoses based on blast percentages is warranted and may help guide clinical management.
Monday, February 28, 2011 1:00 PM
Poster Session II # 167, Monday Afternoon