International Breast Cancer Study Group Statistical Center, Department of Biostatistics and Computational Biology, Dana-Farber Cancer Institute, Boston, MA, USA.
Breast. 2009 Oct;18 Suppl 3:S2-8. doi: 10.1016/S0960-9776(09)70265-6.
Randomized clinical trials are necessary to provide reliable evidence concerning the effectiveness and safety of adjuvant therapies for breast cancer. Such trials, however, are not sufficient to provide information needed to tailor therapies to individual patients. Trials focus on testing treatments on average for heterogeneous patient populations, while attention to the specific characteristics of the disease and the patient are needed to assess the potential benefit for the individual. While 'across the board' results are useful from a population perspective, examination of patterns of treatment response during the course of follow up and for subpopulations of patients is required to make progress and solidify consensus on how to treat individual patients. For example, for several decades it has been known that the pattern of recurrence risk from time of diagnosis is different for estrogen receptor (ER)-negative and ER-positive disease. Assuming that ER status is accurately assessed and distinguishing absence of receptors from low, intermediate and high expression cohorts, one can recognize patterns of relapse risk that are early versus later during follow up. Treatments effective against ER-negative disease reduce the risk of early relapse, while those acting on ER-positive disease demonstrate effectiveness later during the course of follow up. Another example is HER2-positive disease, where a relatively high proportion of patients tend to relapse early, and treatments such as trastuzumab that reduce the risk of early relapse have demonstrated efficacy. For premenopausal patients with ER-positive disease, ovarian function suppression and endocrine effects of chemotherapy are effective to reduce the risk of late occurring relapses. Examining the influence of patient and disease-related factors on the patterns of recurrence over time and treatment responsiveness within subpopulations in multiple randomized trials can facilitate consensus on progress that has been made and identify areas for improving the care of patients with breast cancer.
随机临床试验对于提供乳腺癌辅助治疗的有效性和安全性的可靠证据是必要的。然而，这些试验不足以提供为个体患者量身定制治疗方案所需的信息。试验侧重于针对异质患者群体平均测试治疗方法，而需要关注疾病和患者的具体特征，以评估个体的潜在获益。虽然从人群角度来看，“一刀切”的结果是有用的，但需要检查治疗反应模式在随访过程中和患者亚群中的变化，以取得进展并就如何治疗个体患者达成共识。例如，几十年来，人们已经知道从诊断时起，雌激素受体 (ER) 阴性和 ER 阳性疾病的复发风险模式是不同的。假设 ER 状态被准确评估，并且将受体缺失与低、中、高表达队列区分开来，可以识别出在随访过程中早期和晚期复发风险的模式。针对 ER 阴性疾病有效的治疗方法降低了早期复发的风险，而针对 ER 阳性疾病的治疗方法则在随访过程中晚期显示出有效性。另一个例子是 HER2 阳性疾病，其中相当一部分患者倾向于早期复发，并且已经证明能够降低早期复发风险的治疗方法，如曲妥珠单抗，具有疗效。对于 ER 阳性疾病的绝经前患者，卵巢功能抑制和化疗的内分泌作用可有效降低晚期发生复发的风险。在多个随机试验中检查患者和疾病相关因素对随时间推移的复发模式和亚群内治疗反应的影响，可以促进对已取得的进展的共识，并确定改善乳腺癌患者护理的领域。
