Integrating Radiosensitivity and Immune Gene Signatures for Predicting Benefit of Radiotherapy in Breast Cancer.
Integrating Radiosensitivity and Immune Gene Signatures for Predicting Benefit of Radiotherapy in Breast Cancer.
复制标题
DOI：
10.1158/1078-0432.ccr-18-0825
复制
发表时间：
2018-10-01
期刊：
影响因子：
--
通讯作者：
Li R
中科院分区：
文献类型：
--
作者：
Cui Y;Li B;Pollom EL;Horst KC;Li R
乳腺癌是一种异质性疾病，并非所有患者对辅助放疗的反应相同。需要预测性生物标志物来选择将从治疗中受益的患者，并使其他人免受辐射的毒性和负担。我们首先训练和测试了一种内在的放射敏感性基因特征，以预测三个队列948例患者放疗后的局部复发。接下来，我们通过在129名患者中最大化治疗相互作用效应，开发了一种基于抗原加工和呈递的免疫特征。为了测试其预测价值，我们在一个独立的验证队列中匹配了接受或不接受放疗的患者的临床病理因素，包括年龄，ER状态，HER2状态，分期，放疗，化疗和手术。疾病特异性生存期（DSS）是主要终点。我们的验证队列包括1，439名患者。根据放射敏感性特征进行匹配和分层后，接受放射治疗的患者的DSS优于未接受放射治疗的患者（风险比[HR]=0.68，P=0.059，n=322），而在放射抵抗组中观察到相反的趋势（HR=1.53，P=0.059，n=202）。同样，免疫有效组接受放疗的患者DSS显著更好（HR=0.46，P=0.0076，n=180），免疫缺陷组DSS无差异（HR=1.27，P=0.16，n=348）。两个特征均预测放疗获益（P相互作用=0.007和0.005）。放射敏感性和免疫特征的整合进一步将患者分为三组，接受或不接受放射治疗的患者结局不同（P相互作用=0.003）。拟议的签名有可能选择最有可能从放射治疗中获益的患者。
Breast cancer is a heterogeneous disease and not all patients respond equally to adjuvant radiotherapy. Predictive biomarkers are needed to select patients who will benefit from the treatment and spare others the toxicity and burden of radiation. We first trained and tested an intrinsic radiosensitivity gene signature to predict local recurrence after radiotherapy in three cohorts of 948 patients. Next, we developed an antigen processing and presentation-based immune signature by maximizing the treatment interaction effect in 129 patients. To test their predictive value, we matched patients treated with or without radiotherapy in an independent validation cohort for clinicopathologic factors including age, ER status, HER2 status, stage, hormone-therapy, chemotherapy, and surgery. Disease specific survival (DSS) was the primary endpoint. Our validation cohort consisted of 1,439 patients. After matching and stratification by the radiosensitivity signature, patients who received radiotherapy had better DSS than patients who did not in the radiation-sensitive group (hazard ratio [HR]=0.68, P=0.059, n=322), while a reverse trend was observed in the radiation-resistant group (HR=1.53, P=0.059, n=202). Similarly, patients treated with radiotherapy had significantly better DSS in the immune-effective group (HR=0.46, P=0.0076, n=180), with no difference in DSS in the immune-defective group (HR=1.27, P=0.16, n=348). Both signatures were predictive of radiotherapy benefit (Pinteraction=0.007 and 0.005). Integration of radiosensitivity and immune signatures further stratified patients into three groups with differential outcomes for those treated with or without radiotherapy (Pinteraction=0.003). The proposed signatures have the potential to select patients who are most likely to benefit from radiotherapy.