Institut Gustave Roussy, Villejuif, France.
J Clin Oncol. 2013 Jul 1;31(19):2428-36. doi: 10.1200/JCO.2012.47.3314. Epub 2013 May 20.
Several prognostic scoring systems have been proposed for chronic myelomonocytic leukemia (CMML), a disease in which some gene mutations-including ASXL1-have been associated with poor prognosis in univariable analyses. We developed and validated a prognostic score for overall survival (OS) based on mutational status and standard clinical variables.
We genotyped ASXL1 and up to 18 other genes including epigenetic (TET2, EZH2, IDH1, IDH2, DNMT3A), splicing (SF3B1, SRSF2, ZRSF2, U2AF1), transcription (RUNX1, NPM1, TP53), and signaling (NRAS, KRAS, CBL, JAK2, FLT3) regulators in 312 patients with CMML. Genotypes and clinical variables were included in a multivariable Cox model of OS validated by bootstrapping. A scoring system was developed using regression coefficients from this model.
ASXL1 mutations (P < .0001) and, to a lesser extent, SRSF2 (P = .03), CBL (P = .003), and IDH2 (P = .03) mutations predicted inferior OS in univariable analysis. The retained independent prognostic factors included ASXL1 mutations, age older than 65 years, WBC count greater than 15 ×10(9)/L, platelet count less than 100 ×10(9)/L, and anemia (hemoglobin < 10 g/dL in female patients, < 11g/dL in male patients). The resulting five-parameter prognostic score delineated three groups of patients with median OS not reached, 38.5 months, and 14.4 months, respectively (P < .0001), and was validated in an independent cohort of 165 patients (P < .0001).
A new prognostic score including ASXL1 status, age, hemoglobin, WBC, and platelet counts defines three groups of CMML patients with distinct outcomes. Based on concordance analysis, this score appears more discriminative than those based solely on clinical parameters.
几种预后评分系统已被提出用于慢性髓单核细胞白血病（CMML），在单变量分析中，一些基因突变，包括 ASXL1，与不良预后相关。我们开发并验证了一种基于突变状态和标准临床变量的总生存（OS）预后评分。
我们对 312 例 CMML 患者的 ASXL1 及多达 18 个其他基因（包括表观遗传（TET2、EZH2、IDH1、IDH2、DNMT3A）、剪接（SF3B1、SRSF2、ZRSF2、U2AF1）、转录（RUNX1、NPM1、TP53）和信号（NRAS、KRAS、CBL、JAK2、FLT3）调节剂）进行了基因分型。将基因型和临床变量纳入 OS 的多变量 Cox 模型中，并通过自举法进行验证。使用该模型的回归系数开发了一种评分系统。
ASXL1 突变（P<0.0001）和在较小程度上 SRSF2 突变（P=0.03）、CBL 突变（P=0.003）和 IDH2 突变（P=0.03）在单变量分析中预测 OS 不良。保留的独立预后因素包括 ASXL1 突变、年龄大于 65 岁、白细胞计数大于 15×10(9)/L、血小板计数小于 100×10(9)/L 和贫血（女性患者血红蛋白<10g/dL，男性患者血红蛋白<11g/dL）。由此产生的五参数预后评分将患者分为三组，中位 OS 分别为未达到、38.5 个月和 14.4 个月（P<0.0001），并在 165 例患者的独立队列中得到验证（P<0.0001）。
一种包含 ASXL1 状态、年龄、血红蛋白、白细胞和血小板计数的新预后评分可将 CMML 患者分为三组，具有不同的结局。基于一致性分析，该评分似乎比仅基于临床参数的评分更具鉴别力。
