Adam Roman, Díez-González Laura, Ocaña Alberto, Šeruga Boštjan, Amir Eitan, Templeton Arnoud J
Faculty of Medicine, University of Basel, Basel, Switzerland.
Traslational Research laboratory, Albacete University Hospital, and Regional Biomedical Research Center, Castilla La Mancha University, Albacete, Spain.
Exp Mol Pathol. 2017 Jun;102(3):455-474. doi: 10.1016/j.yexmp.2017.05.010. Epub 2017 May 12.
Telomere length (TL) has been associated with several health conditions including cancer. To quantify the effect of TL on outcomes in malignancies and explore the role of type of TL measurement we conducted a librarian-led systematic search of electronic databases identified publications exploring the prognostic role of TL on cancer outcomes. Overall survival (OS) was the primary outcome measure while other time-to-event endpoints were secondary outcomes. Data from studies reporting a hazard ratio (HR) with 95% confidence interval (CI) and/or p-value were pooled in a meta-analysis. HRs were weighted by generic inverse variance and computed by random effects modeling. All statistical tests were two-sided. Sixty-one studies comprising a total of 14,720 patients were included of which 41 (67%) reported OS outcomes. Overall, the pooled HR for OS was 0.88 (95%CI=0.69-1.11, p=0.28). Long (versus short) telomeres were associated with improved outcomes in chronic lymphatic leukemia (CLL) and urothelial cancer (HR=0.45, 95%CI=0.29-0.71 and HR=0.68, 95%CI=0.46-1.00, respectively), conversely worse OS was seen with hepatocellular carcinoma (HR=1.90, 95%CI=1.51-2.38). Pooled HRs (95% CI) for progression-free survival, relapse/disease-free survival, cancer-specific survival, and treatment-free survival were 0.56 (0.41-0.76), 0.76 (0.53-1.10), 0.72 (0.48-1.10), and 0.48 (0.39-0.60), respectively. There was substantial heterogeneity of tissues and methods used for TL measurement and no clear association between TL and outcome was identified in subgroups. In conclusion, there is inconsistent effect of TL on cancer outcomes possibly due to variable methods of measurement. Standardization of measurement and reporting of TL is warranted before the prognostic value of TL can be accurately assessed.
端粒长度（TL）已与包括癌症在内的多种健康状况相关联。为了量化TL对恶性肿瘤预后的影响，并探讨TL测量类型的作用，我们在图书馆员的带领下对电子数据库进行了系统检索，确定了探索TL对癌症预后作用的出版物。总生存期（OS）是主要结局指标，而其他事件发生时间终点则是次要结局。将报告了风险比（HR）及95%置信区间（CI）和/或P值的研究数据纳入荟萃分析。HR采用通用逆方差加权，并通过随机效应模型计算。所有统计检验均为双侧检验。纳入了61项研究，共14720例患者，其中41项（67%）报告了OS结局。总体而言，OS的合并HR为0.88（95%CI=0.69-1.11，P=0.28）。长端粒（与短端粒相比）与慢性淋巴细胞白血病（CLL）和尿路上皮癌的预后改善相关（HR分别为0.45，95%CI=0.29-0.71和HR=0.68，95%CI=0.46-1.00），相反，肝细胞癌的OS较差（HR=1.90，95%CI=1.51-2.38）。无进展生存期、复发/无病生存期、癌症特异性生存期和无治疗生存期的合并HR（95%CI）分别为0.56（0.41-0.76）、0.76（0.53-1.10）、0.72（0.48-1.10）和0.48（0.39-0.60）。用于TL测量的组织和方法存在很大异质性，在亚组中未发现TL与结局之间有明确关联。总之，TL对癌症结局的影响不一致，可能是由于测量方法不同。在能够准确评估TL的预后价值之前，有必要对TL的测量和报告进行标准化。
