Objective: To evaluate the factors affecting the degree of radical resection and the prognosis of patients with locally recurrent rectal cancer (LRRC). Methods: A retrospective case-control study was performed. Clinical data of 111 patients with LRRC undergoing operation at the General Surgery Department of Peking University First Hospital from January 2009 to August 2019 were analyzed retrospectively. The "Peking University First Hospital F typing" was performed according to the preoperative images of the pelvic involvement. The pelvis was assigned into four directions: the front wall, lateral sides of the pelvic wall and the sacrum. According to the degree of pelvic wall involvement, F typing included F0 type (no involvement of the pelvic wall, the cancer only involved the adjacent organs or invaded conteriorly the urinary tract, genital organs or small intestine), F1 type (cancer involved the pelvic wall in one direction, such as the sacrum, or one side of the pelvic wall), F2 type (cancer involved the pelvic wall in two directions) and F3 type (cancer involved the pelvic wall in three directions). Case inclusion criteria: (1) LRRC was confirmed by imaging and pathological examination of samples (puncture or endoscopic biopsy); (2) complete clinical and follow-up data; (3) informed consent of patient. Those with dysfunction of heart, lung, etc., intolerance of operation, F3 type indicated by image, and distant metastasis were excluded. The degree of radical resection was evaluated according to the postoperative pathological results. Patients were followed up every 12 months and related examinations were arranged. The univariate analysis of radical resection was performed by χ(2) test, and the multivariate analysis was performed by logistic methods. The survival rate was calculated by Kaplan-Meier method and the survival curve was drawn. The survival rate was compared by log-rank test. Cox proportional hazards model was used to analyze the factors affecting the prognosis of patients with LRRC. Results: A total of 111 patients were included in this study. Of 111 patients, 59 were male and 52 were female; recurrent age of 36 cases was ≥ 65 years old; CEA level of 48 cases was ≥15 μg/L. According to the "Peking University First Hospital F typing", 70 cases were F0 type, 38 F1 type and 3 F2 type. Surgical procedures were abdominoperineal resection (n=28), posterior pelvic exenteration (n=32), and total pelvic exenteration (n=51, including 1 case of TPE combined with sacrectomy). According to the postoperative pathological results, R0, R1 and R2 resections were 83, 20 and 8 cases, respectively. Univariate analysis showed that the degree of radical resection was associated with the secondary surgical procedure, F typing and lymph node metastasis (all P<0.05). Multivariate analysis showed that F typing (F1-F2) was an independent risk factor for non- R0 resection (OR=37.256, 95%CI:8.572 to 161.912, P<0.001). The morbidity of operative complications was 22.5% (25/111); the perioperative mortality was 1.8% (2/111); the local recurrence rate after the second operation was 37.8% (42/111). The 3- and 5-year overall survival rates were 41.2% and 21.9% respectively. The 3-year survival rates of patients with and without postoperative chemotherapy were 52.7% and 32.4% respectively (P=0.005). The 3-year survival rates of patients with lower (<15 μg/L) and higher CEA level (≥15 μg/L) were 52.9% and 24.3% respectively (P<0.001). The 3-year survival rates of patients with R0, R1 and R2 resection were 49.8%, 21.3% and 8.5% respectively (P=0.002). The 3-year survival rates of patients with F0, F1 and F2 type were 52.7%, 22.0% and 0 respectively (P<0.001). Cox analysis confirmed that the degree of radical resection (HR=2.088, 95%CI:1.095 to 3.979, P=0.025), the CEA level before the secondary operation (HR=1.857, 95%CI:1.157 to 2.980, P=0.010) and postoperative chemotherapy (HR=1.826, 95%CI:1.137 to 2.934, P=0.013) were independent factors affecting the prognosis. Conclusions: The indication of LRRC surgical treatments must be strictly limited. Evaluation of the fixation site to the pelvic wall is helpful for improving the rate of R0 resection. Lower preoperative CEA level, radical resection and postoperative chemotherapy are protective factors of prolonged overall survival time of patients with LRRC.
目的： 探讨影响局部复发直肠癌（LRRC）肿瘤手术根治性和预后的相关因素。 方法： 采用回顾性病例对照研究。回顾性分析2009年1月至2019年8月北京大学第一医院手术治疗的LRRC患者病例资料。术前根据患者的影像检查结果对盆壁的累及情况行"北大医院F分型"，即将骨盆划分为前方、两侧盆壁和骶骨4个方向，根据盆壁累及程度分为F0（无累及盆壁，肿瘤累及临近脏器或向前方侵犯泌尿、生殖器官或小肠）、F1（肿瘤累及一个方向的盆壁，如骶骨、两侧盆壁之一）、F2（累及两个方向的盆壁）和F3（累及3个方向的盆壁）。病例纳入标准：（1）经影像学及活检病理学检查（穿刺或肠镜活检）确诊为直肠癌术后局部复发；（2）随访资料和临床资料完整；（3）获得患者的知情同意书。排除因心肺等功能不全无法耐受手术治疗、影像学检查结果提示为F3或有远处转移的患者。根据病理结果对肿瘤的手术根治性进行评价。患者术后每12个月进行随访。影响肿瘤根治性的单因素风险分析采用χ(2)检验，多因素风险分析采用Logistic分析。采用Kaplan-Meier法计算生存率并绘制生存曲线，生存率的比较采用Log-rank检验。应用Cox比例风险模型对于预后进行多因素回归分析。 结果： 共计入组111例LRRC患者，其中男性59例，女性52例，复发年龄≥65岁者36例，癌胚抗原水平≥15 μg/L者48例。根据"北大医院F分型"结果，F0、F1和F2患者分别有70例、38例和3例。行腹会阴联合切除术28例，后盆腔脏器切除术32例，全盆腔脏器切除术51例（其中联合骶尾骨切除1例）。根据术后病理评价，R(0)、R(1)和R(2)切除分别为83例、20例和8例。单因素分析提示，LRRC的手术根治性与LRRC手术方式、北大医院F分型以及淋巴结是否转移密切相关（均P<0.05）。多因素分析提示，北大医院F分型为F1~2是不能完成R(0)切除的独立危险因素（OR=37.256，95%CI：8.572~161.912，P<0.001）。全组手术并发症发生率为22.5%（25/111），围手术期病死率1.8%（2/111），LRRC术后局部复发率为37.8%（42/111）。全组患者术后3年和5年生存率分别为41.2%和21.9%，其中LRRC术后化疗组与未予化疗组的3年生存率分别为52.7%与32.4%（P=0.005）；癌胚抗原水平<15 μg/L与≥15 μg/L者的3年生存率分别为52.9%与24.3%（P<0.001）；R(0)、R(1)和R(2)切除患者的3年生存率分别为49.8%、21.3%和8.5%（P=0.002）；F0、F1和F2的3年生存率分别为52.7%、22.0%和0（P<0.001）。多因素分析提示，手术根治性（HR=2.088，95%CI：1.095~3.979，P=0.025）、术前癌胚抗原水平（HR=1.857，95%CI：1.157~2.980，P=0.010）和是否接受术后辅助化疗（HR=1.826，95%CI：1.137~2.934，P=0.013）是影响LRRC患者预后的独立因素（均P<0.05）。 结论： LRRC手术必须严格限制手术适应证。评估复发肿瘤累及盆壁的情况有利于提高手术根治性。术前较低的癌胚抗原水平、提高肿瘤手术根治性和术后辅助化疗是延长LRRC患者总生存期的关键因素。.
Keywords: Curative resection; Local recurrence; Multi-disciplinary treatment; Rectal neoplasms; Total pelvic exenteration.