Translational Prostate Cancer Group, Hutchison MRC Research Centre, University of Cambridge, Hills Road, Cambridge CB2 0XZ, UK.
J Cancer Res Clin Oncol. 2013 Jan;139(1):1-24. doi: 10.1007/s00432-012-1351-7. Epub 2012 Nov 28.
Tissue biomarkers could pivotally improve clinical outcome prediction following prostate cancer therapy. Clinically, prostate cancer is managed by diverse treatment modalities whose individual influence on a biomarker's predictive ability is not well understood and poorly investigated in the literature.
We conducted a systematic review to assess the predictive value of biomarkers in different treatment contexts in prostate cancer.
A literature search was performed using the MeSH headings "prostate neoplasms" and "biological markers". Rigorous selection criteria identified studies correlating expression with clinical outcomes from primary androgen deprivation therapy (ADT), radical prostatectomy and radiotherapy (± neoadjuvant ADT).
Of 10,668 studies identified, 481 papers matched initial inclusion criteria. Following rescreening, 384 studies identified 236 individual tissue biomarkers, of which 29 were predictive on multivariate analysis in at least 2 independent cohorts. The majority were only tested in surgical cohorts. Only 8 predictive biomarkers were tested across all 3 treatments with Ki67 identified as universal predictive marker. p16 showed potential for treatment stratification between surgery and radiotherapy but needs further validation in independent studies.
Despite years of research, very few tissue biomarkers retain predictive value in independent validation across therapy context. Currently, none have conclusive ability to help treatment selection. Future biomarker research should consider the therapy context and use uniform methodology and evaluation criteria.
组织生物标志物可以显著改善前列腺癌治疗后的临床预后预测。临床上，前列腺癌通过多种治疗方式进行管理，但其对生物标志物预测能力的个体影响在文献中尚未得到很好的理解和研究。
我们进行了一项系统评价，以评估生物标志物在前列腺癌不同治疗环境中的预测价值。
使用 MeSH 主题“前列腺肿瘤”和“生物标志物”进行文献检索。严格的选择标准确定了与原发性去势治疗（ADT）、根治性前列腺切除术和放疗（±新辅助 ADT）中表达与临床结局相关的研究。
在确定的 10668 项研究中，有 481 篇论文符合初始纳入标准。经过重新筛选，有 384 项研究确定了 236 个单独的组织生物标志物，其中 29 个在至少 2 个独立队列的多变量分析中具有预测性。大多数研究仅在手术队列中进行了测试。只有 8 个预测性生物标志物在所有 3 种治疗方法中进行了测试，Ki67 被确定为普遍的预测性标志物。p16 显示出在手术和放疗之间进行治疗分层的潜力，但需要在独立研究中进一步验证。
尽管经过多年的研究，很少有组织生物标志物在独立验证中在治疗背景下具有预测价值。目前，没有一种生物标志物具有明确的帮助治疗选择的能力。未来的生物标志物研究应考虑治疗背景，并使用统一的方法和评估标准。
