Real-world practice-based prognostic model for higher-risk myelodysplastic syndromes treated with azacitidine monotherapy: The Kyoto prognostic scoring system.

ABSTRACT

The prognostic impact of patient-related factors, including age, nutritional parameters, and inflammation status, in higher-risk myelodysplastic syndromes (HR-MDS) has been largely unexplored. This multicenter retrospective study aimed to establish a real-world practice-based prognostic model for HR-MDS by considering both disease- and patient-related parameters in 233 patients treated with AZA monotherapy at seven institutions. We found that anemia, presence of circulating blasts in peripheral blood, low absolute lymphocyte count, low total cholesterol (T-cho) and albumin serum levels, complex karyotype, and del(7q) or - 7 were poor prognostic factors. Therefore, we developed a new prognostic model called the Kyoto Prognostic Scoring System (KPSS) by incorporating the two variables with the highest C-indexes (complex karyotype and serum T-cho level). The KPSS classified patients into the following three groups good (0 risk factors), intermediate (1), and poor (2). Median overall survival for these groups was 24.4, 11.3, and 6.9, respectively (p < 0.001). The discriminatory power of the KPSS was higher than that of the traditional International Prognostic Scoring System. In conclusion, we identified several nutritional parameters with prognostic relevance in patients with HR-MDS and generated a prognostic model consisting of complex karyotype and serum T-cho level that enabled excellent risk stratification.