An Informative Prognostic Ability of New Scoring Model in Myelodysplastic Syndrome: Short Telomere Length and Somatic Mutation Integrated with IPSS-R

• Published in: Blood Vol. 132; no. Supplement 1; p. 3089
• Main Authors: Park, Hee Sue, Park, Junseo, Im, Kyongok, Kim, Sung-Min, Yun, Jiwon, Jeong, Dajeong, Shin, Dong-Yeop, Hwang, Sang Mee, Yoon, Sung-Soo, Lee, Dong Soon
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 29.11.2018
• ISSN: 0006-4971; 1528-0020
• DOI: 10.1182/blood-2018-99-113243

Background: Molecular genetic alterations and short telomere length have reported as affect clinical outcome of patients with myelodysplastic syndrome (MDS). Recently, gene mutation was suggested as a new variable in new developed prognostic scoring model integrated IPSS-R. In addition, shorted telomere length reflected the genetic instability and closely associated various clinical fatal conditions. Here we proposed a new scoring system that encompass gene variations, telomere lengths and IPSS-R together. Methods: In 153 patients diagnosed with MDS in Seoul National University Hospital, G-banding, fluorescence in situ hybridization (FISH), targeted capture sequencing for 88 hematopoiesis-related genes, and measurement of telomere length (TL) were performed. Kaplan-Meier survival analysis with the log-rank test and Cox proportional hazards regression analysis were used to develop a new prognostic system using Mathematica. Results: A total of 378 candidate mutations in 72 genes were found. Mutation of TP53, ASXL1, EZH2 showed significant correlation with adverse prognosis among 12 genes with frequency over 5% or more (ASXL1, U2AF1, TP53, RUNX1, TET2, DNMT3A, SRSF2, BCOR, EZH2, SF3B1, STAG2, and WT1). Patients with telomere length <5.37 showed an adverse survival. We developed a new scoring model incorporating the weighted coefficients of these variables: Age x 0.017 + IPSS-R score x 0.220 + ASXL1 mutation x 0.406 + TP53 mutation x 0.897 + EZH2 mutation x 0.706 + Telomere length (less than 5.37) x 0.548. The age and IPSS-R score were used as a continuous variable. The presence of gene mutations and TL below 5.37 was scored as 1. According to this new scoring system, patients were divided into four groups: low score cutoff (<1.5, n=16), intermediate-1 (1.5-2.0, n=27), intermediate-2 (2.0-2.5, n=39), high (≥2.5, n=45). The median OS was 150, 73, 67 and 36 months for low, intermediate-1, intermediate-2, and high, retrospectively (p<0.001). Meanwhile, according to conventional IPSS-R scoring system, the median OS was 97, 78, 79, 63, and 34 months for very low, low, intermediate, high and very high, retrospectively (p <0.001). The mutation of ASXL1, TP53, EZH2, RUNX1, SFSR2 and STAG2 were observed more frequency in intermediated-2 and high risk group in new scoring model. The C-index increased in new scoring models (0.79) compared with the existing IPSS-R (0.69). In addition, we assessed new scoring model excluding telomere length. The score cut off was slightly changed in low score cutoff (<1.7, n=29), intermediate-1 (1.7-2.0, n=50), intermediate-2 (2.2-2.7, n=34), high (≥2.5, n=40). The C-index was 0.78 in this model. Conclusion: We proposed new model combined TL, molecular variations integrated existing IPSS-R. This new model showed more informative prognostic ability than existing IPSS-R. And short telomere length makes new molecular scoring system a powerful prognostic model. No relevant conflicts of interest to declare.