分子检测技术在血液中检测和量化肿瘤DNA的新兴技术，俗称为液体活检，受到了极大的热情追捧。然而，人们对于这项技术的临床效用与检测指征之间的关系却付之闲章，而这些检测指征包括一系列不同的临床情况，每一种情况都具备独特的挑战。本文考虑了5种循环肿瘤DNA（ctDNA）血液检测的指征：（1）对于治疗选择，ctDNA检测可以识别患有可用于靶向治疗的癌症患者的遗传改变，但大多数患者没有可靶向的改变；（2）对于对治疗的反应，癌症手术后残余肿瘤DNA的缺失可能会减少辅助化疗的使用，但实际上这是否会发生还不清楚；（3）对于癌症治疗后的疾病监测，ctDNA检测可以在症状出现之前发现癌症复发，但早期干预可能对生存率没有影响；（4）对于疑似癌症的诊断，ctDNA检测能够识别一些有症状的癌症，但它们是否能够增加常规诊断评估的价值尚不明确；（5）对于癌症筛查，多癌种检测可以检测到多种癌症，但其对早期肿瘤的敏感性较低，因此对于筛查是否能帮助患者延长寿命或提高生活质量还存在疑问。循环肿瘤DNA检测在多种指征上得到了推广。目前有众多研究正在进行，但对于其对患者中心结果的影响的随机临床试验很少。虽然这些检测具备在选定指征上改善护理的潜力，但这一点必须得到证实，因为它们会给患者增加成本、复杂性和意外的不良反应。

There has been great enthusiasm for the emerging technology of molecular-based tests to detect and quantify tumor DNA circulating in the bloodstream, colloquially known as a liquid biopsy. However, less attention has been given to how their clinical utility depends on the indication for testing, which includes a range of clinical situations, each presenting unique challenges.Five indications for circulating tumor DNA (ctDNA) blood testing were considered. (1) For therapy selection, ctDNA tests can identify genetic alterations in patients with cancer amenable to targeted therapy, but most patients do not have a targetable alteration. (2) For response to therapy, the absence of residual tumor DNA following cancer surgery could reduce the use of adjuvant chemotherapy, but it is unclear that this will happen in practice. (3) For disease surveillance following cancer treatment, ctDNA tests may well detect cancer recurrence before symptoms appear, yet earlier intervention may have no effect on mortality. (4) For diagnosis of suspected cancer, ctDNA tests are able to identify some symptomatic cancers, but how they add to the conventional diagnostic evaluation is unknown. (5) For screening for cancer, multicancer tests can detect many types of cancer, but their low sensitivity for early-stage tumors raises questions as to whether screening can help patients live longer or live better.Circulating tumor DNA tests are being promoted for multiple indications. Numerous studies are ongoing, but randomized clinical trials of their effect on patient-centered outcomes are rare. While these tests have the potential to improve care in selected indications, this must be proven, as they will add cost, complexity, and unintended adverse effects for patients.