Smith B Douglas, Liu Jun, Latremouille-Viau Dominick, Guerin Annie, Fernandez Daniel, Chen Lei
a Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins , Baltimore , MD , USA ;
b Harvard University , Cambridge , MA , USA ;
Curr Med Res Opin. 2016 May;32(5):817-27. doi: 10.1185/03007995.2016.1140030. Epub 2016 Feb 10.
Objective Though the median age at diagnosis is 64 years, few studies focus on elderly (≥65 years) patients with chronic myeloid leukemia (CML). This study examines healthcare outcomes among elderly Medicare beneficiaries with CML who started nilotinib or dasatinib after imatinib. Research design and methods Patients were identified in the Medicare Research Identifiable Files (2006-2012) and had continuous Medicare Parts A, B, and D coverage. Main outcome measures Treatment patterns, overall survival (OS), monthly healthcare resource utilization and medical costs were measured from the second-line tyrosine kinase inhibitor (TKI) initiation (index date) to end of Medicare coverage. Results Despite similar adherence, dasatinib patients (N = 379) were more likely to start on the recommended dose (74% vs. 53%; p < 0.001), and to have dose reductions (21% vs. 11%, adjusted hazard ratio [HR] = 1.94; p = 0.002) or dose increases (9% vs. 7%; adjusted HR = 1.81; p = 0.048) than nilotinib patients (N = 280). Fewer nilotinib patients discontinued (59% vs. 67%; adjusted HR = 0.80; p = 0.026) or switched to another TKI (21% vs. 29%; adjusted HR = 0.72; p = 0.044) than dasatinib patients. Nilotinib patients had longer median OS (>4.9 years vs. 4.0 years; p = 0.032) and 37% lower mortality risk than dasatinib patients (adjusted HR = 0.63; p = 0.008). Nilotinib patients had 23% fewer inpatient admissions, 30% fewer emergency room visits, 13% fewer outpatient visits (all p < 0.05), and lower monthly medical costs (by $513, p = 0.024) than dasatinib patients. Limitations Lack of clinical assessment (disease phase and response to first-line therapy) and retrospective nature of study (unobservable potential confounding factors, non-randomized treatment choice). Conclusions In the current study of elderly CML patients, initiation of second-line TKIs frequently occurs at doses lower than the recommended starting doses and, despite this, many patients require dose adjustments. Here, nilotinib patients required fewer dose adjustments than dasatinib patients. Further research focusing on elderly CML patients is warranted in order to help define future best clinical practices.
目的 尽管慢性髓性白血病（CML）的诊断中位年龄为64岁，但很少有研究关注老年（≥65岁）CML患者。本研究调查了伊马替尼治疗后开始使用尼罗替尼或达沙替尼的老年医疗保险受益CML患者的医疗保健结局。研究设计与方法 在医疗保险研究识别档案（2006 - 2012年）中识别患者，这些患者持续享有医疗保险A、B和D部分的覆盖。主要结局指标 从二线酪氨酸激酶抑制剂（TKI）开始使用（索引日期）至医疗保险覆盖结束，测量治疗模式、总生存期（OS）、每月医疗资源利用情况和医疗费用。结果 尽管依从性相似，但达沙替尼组患者（N = 379）更有可能起始使用推荐剂量（74%对53%；p < 0.001），并且与尼罗替尼组患者（N = 280）相比，更有可能出现剂量减少（21%对11%，调整后风险比[HR] = 1.94；p = 0.002）或剂量增加（9%对7%；调整后HR = 1.81；p = 0.048）。与达沙替尼组患者相比，尼罗替尼组患者中断治疗（59%对67%；调整后HR = 0.80；p = 0.026）或换用另一种TKI（21%对29%；调整后HR = 0.72；p = 0.044）的情况较少。尼罗替尼组患者的中位OS更长（>4.9年对4.0年；p = 0.032），死亡风险比达沙替尼组患者低37%（调整后HR = 0.63；p = 0.008）。与达沙替尼组患者相比，尼罗替尼组患者的住院次数减少23%，急诊就诊次数减少30%，门诊就诊次数减少13%（均p < 0.05），每月医疗费用更低（低513美元，p = 0.024）。局限性 缺乏临床评估（疾病阶段和对一线治疗的反应）以及研究的回顾性性质（潜在混杂因素不可观察，治疗选择非随机）。结论 在当前对老年CML患者的研究中，二线TKI的起始使用剂量经常低于推荐起始剂量，尽管如此，许多患者仍需要调整剂量。在此，尼罗替尼组患者比达沙替尼组患者需要更少的剂量调整。有必要针对老年CML患者开展进一步研究，以帮助确定未来最佳临床实践。
