Objective: To evaluate the prognostic value of JAK2, MPL and CALR mutations in Chinese patients with primary myelofibrosis (PMF).
Methods: Four hundred and two Chinese patients with PMF were retrospectively analyzed. The Kaplan-Meier method, the Log-rank test, the likelihood ratio test and the Cox proportional hazards regression model were used to evaluate the prognostic scoring system.
Results: This cohort of patients included 209 males and 193 females with a median age of 55 years (range: 15- 89). JAK2V617F mutations were detected in 189 subjects (47.0% ), MPLW515 mutations in 13 (3.2%) and CALR mutations in 81 (20.1%) [There were 30 (37.0%) type-1, 48 (59.3%) type-2 and 3 (3.7%) less common CALR mutations], respectively. 119 subjects (29.6%) had no detectable mutation in JAK2, MPL or CALR. Univariate analysis indicated that patients with CALR type-2 mutations or no detectable mutations had inferior survival compared to those with JAK2, MPL or CALR type- 1 or other less common CALR mutations (the median survival was 74vs 168 months, respectively [HR 2.990 (95% CI 1.935-4.619),P<0.001]. Therefore, patients were categorized into the high-risk with CALR type- 2 mutations or no detectable driver mutations and the low- risk without aforementioned mutations status. The DIPSS-Chinese molecular prognostic model was proposed by adopting mutation categories and DIPSS-Chinese risk group. The median survival of patients classified in low risk (132 subjects, 32.8% ), intermediate- 1 risk (143 subjects, 35.6%), intermediate- 2 risk (106 subjects, 26.4%) and high risk (21 subjects, 5.2%) were not reached, 156 (95% CI 117- 194), 60 (95% CI 28- 91) and 22 (95% CI 10- 33) months, respectively, and there was a statistically significant difference in overall survival among the four risk groups (P<0.001). There was significantly higher predictive power for survival according to the DIPSS-Chinese molecular prognostic model compared with the DIPSS-Chinese model (P=0.005, -2 log-likelihood ratios of 855.6 and 869.7, respectively).
Conclusion: The impact of the CALR type- 2 mutations or no detectable driver mutation on survival was independent of current prognostic scoring systems. The DIPSS- Chinese molecular prognostic model based on the molecular features of Chinese patients was proposed and worked well for prognostic indication.
目的: 评价JAK2、MPL和CALR基因突变在中国原发性骨髓纤维化（PMF）患者中的预后意义。
方法: 回顾性分析402例PMF患者的临床资料及JAK2、MPL和CALR基因突变，应用Kaplan-Meier、Log-rank和Cox回归模型进行相关预后因素分析。
结果: 402例PMF患者中男209例，女193例，中位年龄55(15~89）岁。JAKV617F基因突变189例（47.0%），MPLW515基因突变13例（3.2%），CALR基因突变81例（20.1%）[1型突变30例（37.0%），2型突变48例（59.3%），少见型突变3例（3.7%）]，119例（29.6%）未检测到JAK2、MPL和CALR基因突变。单因素分析显示，2型CALR突变或未检测到JAK2、MPL和CALR基因突变的患者中位生存期（74个月）短于检测到JAK2、MPL或1型及少见型CALR基因突变的患者（168个月）[HR=2.990（95% CI 1.935~4.619），P <0.001]。因此，将基因突变类型分为预后不良组（2型CALR突变和未检测到JAK2、MPL和CALR基因突变）和预后良好组（JAKV617F、MPLW515、1型及少见型CALR基因突变）。经多因素分析提出了中国PMF患者的动态国际预后积分系统（DIPSS-Chinese）分子预后积分系统，各参数赋值如下：DIPSS-Chinese低危组0分，中危组1分，高危组2分，2型CALR突变或未检测到JAK2、MPL和CALR基因突变积1分。402例患者中，低危组（0分）132例（32.8%），中危-1组（1分）143例（35.6%），中危-2组（2分）106例（26.4%），高危组（3分）21例（5.2%）。低危组中位生存时间未达到，中危-1组为156(95% CI 117~194）个月，中危-2组为60(95% CI 28~91）个月，高危组为22（95% CI 10~33）个月，总生存差异具有统计学意义（P<0.001）。DIPSS-Chinese分子预后积分系统较DIPSS-Chinese有更准确的预测能力（−2 log似然比分别为855.6和869.7，P=0.005）。
结论: 在中国PMF患者中，2型CALR突变和未检测到JAK2、MPL和CALR突变是独立的不良预后因素，DIPSS-Chinese分子预后积分系统有更好的预测价值。