Importance  The role of epidermal growth factor receptor (EGFR) inhibition in chemoradiation strategies in the nonoperative treatment of patients with esophageal cancer remains uncertain.

Objective  To evaluate the benefit of cetuximab added to concurrent chemoradiation therapy for patients undergoing nonoperative treatment of esophageal carcinoma.

Design, Setting, and Participants  A National Cancer Institute (NCI) sponsored, multicenter, phase 3, randomized clinical trial open to patients with biopsy-proven carcinoma of the esophagus. The study accrued 344 patients from 2008 to 2013.

Interventions  Patients were randomized to weekly concurrent cisplatin (50 mg/m²), paclitaxel (25 mg/m²), and daily radiation of 50.4 Gy/1.8 Gy fractions with or without weekly cetuximab (400 mg/m² on day 1 then 250 mg/m² weekly).

Main Outcomes and Measures  Overall survival (OS) was the primary endpoint, with a study designed to detect an increase in 2-year OS from 41% to 53%; 80% power and 1-sided α = .025.

Results  Between June 30, 2008, and February 8, 2013, 344 patients were enrolled. This analysis used all data received at NRG Oncology through April 12, 2015. Sixteen patients were ineligible, resulting in 328 evaluable patients, 159 in the experimental arm and 169 in the control arm. Patients were well matched between the treatment arms for patient and tumor characteristics: 263 (80%) with T3 or T4 disease, 215 (66%) N1, and 62 (19%) with celiac nodal involvement. Incidence of grade 3, 4, or 5 treatment-related adverse events at any time was 71 (46%), 35 (23%), or 6 (4%) in the experimental arm and 83 (50%), 28 (17%), or 2 (1%) in the control arm, respectively. A clinical complete response (cCR) rate of 81 (56%) was observed in the experimental arm vs 92 (58%) in the control arm (Fisher exact test, P = .66). No differences were seen in cCR between treatment arms for either histology (adenocarcinoma or squamous cell). Median follow-up for all patients was 18.6 months. The 24- and 36-month local failure for the experimental arm was 47% (95% CI, 38%-57%) and 49% (95% CI, 40%-59%) vs 49% (95% CI, 41%-58%) and 49% (95% CI, 41%-58%) for the control arm (HR, 0.92; 95% CI, 0.66-1.28; P = .65). The 24- and 36-month OS rates for the experimental arm were 45% (95% CI, 37%-53%) and 34% (95% CI, 26%-41%) vs 44% (95% CI, 36%-51%) and 28% (95% CI, 21%-35%) for the control arm (HR, 0.90; 95% CI, 0.70-1.16; P = .47).

Conclusions and Relevance  The addition of cetuximab to concurrent chemoradiation did not improve OS. These phase 3 trial results point to little benefit to current EGFR-targeted agents in an unselected patient population, and highlight the need for predictive biomarkers in the treatment of esophageal cancer.

Trial Registration  clinicaltrials.gov Identifier: NCT00655876

重要性  表皮生长因子受体 (EGFR) 抑制在食管癌患者非手术治疗的放化疗策略中的作用仍不确定。

目的  评价西妥昔单抗联合同步放化疗对食管癌非手术治疗患者的益处。

设计、设置和参与者  美国国家癌症研究所 (NCI) 赞助的多中心 3 期随机临床试验向经活检证实的食道癌患者开放。该研究从 2008 年到 2013 年招募了 344 名患者。

干预  患者被随机分配到每周同时使用顺铂 (50 mg/m ² )、紫杉醇 (25 mg/m ² ) 和每日 50.4 Gy/1.8 Gy 分次的放射治疗，联合或不联合每周一次的西妥昔单抗（第 1 天 400 mg/m ²然后250 mg/m ²每周）。

主要结果和措施  总生存期 (OS) 是主要终点，一项研究旨在检测 2 年 OS 从 41% 增加到 53%；80% 功率和 1 面 α = .025。

结果  2008 年 6 月 30 日至 2013 年 2 月 8 日期间，共有 344 名患者入组。该分析使用了截至 2015 年 4 月 12 日在 NRG Oncology 收到的所有数据。16 名患者不符合条件，导致 328 名可评估患者，159 名在实验组，169 名在对照组。患者在治疗组之间的患者和肿瘤特征非常匹配：263 名（80%）患有 T3 或 T4 疾病，215 名（66%）N1 和 62 名（19%）患有腹腔淋巴结受累。任何时间 3、4 或 5 级治疗相关不良事件的发生率在实验组分别为 71 (46%)、35 (23%) 或 6 (4%)，83 (50%)、28 (17 %) 或 2 (1%) 在控制臂中，分别。实验组的临床完全缓解 (cCR) 率为 81 (56%)，而对照组为 92 (58%)（Fisher 精确检验，P = .66)。对于任一组织学（腺癌细胞或鳞状细胞），治疗组之间的 cCR 均未见差异。所有患者的中位随访时间为 18.6 个月。实验组 24 个月和 36 个月的局部失败率为 47% (95% CI, 38%-57%) 和 49% (95% CI, 40%-59%) vs 49% (95% CI, 41对照组为 %-58%) 和 49% (95% CI, 41%-58%) (HR, 0.92; 95% CI, 0.66-1.28; P  = .65)。实验组的 24 个月和 36 个月 OS 率为 45%（95% CI，37%-53%）和 34%（95% CI，26%-41%）与 44%（95% CI，36对照组为 %-51%) 和 28% (95% CI, 21%-35%) (HR, 0.90; 95% CI, 0.70-1.16; P  = .47)。

结论和相关性  在同步放化疗中加用西妥昔单抗并未改善 OS。这些 3 期试验结果表明，当前 EGFR 靶向药物在未经选择的患者群体中几乎没有益处，并强调了在食管癌治疗中对预测性生物标志物的需求。

试验注册  临床试验.gov 标识符：NCT00655876