第一部分：II~III期结直肠癌根治术后吻合口复发与病理切缘距离相关性研究

背景

吻合口复发（anastomotic recurrence, AR）是一类特殊的局部复发，发生率约为0.4%至4.2%，目前对其危险因素和预后意义尚不明确。尽管普遍认为AR与切缘距离不足有关，但临床对最佳切缘距离尚无共识，且仍有很多患者根治性术后会出现AR。因此有必要通过大规模、多中心队列研究来明确高危人群，评估最佳手术切缘距离，并分析AR患者的肿瘤学预后。

方法

本研究共收集了6019例2009年1月至2019年12月期间接受根治性手术的病理II~III期结直肠癌患者，患者人群来自于中国国家结直肠癌队列研究数据库和中国医学科学院北京协和医院数据库。详细收集了患者远近切缘距离、局部复发情况等临床病理资料。采用Cox模型进行单因素和多因素分析，Kaplan–Meier曲线用于绘制复发后的生存曲线。

结果

本研究最终纳入4087例II~III期结直肠癌患者进行分析，其中AR率为1.9%，非吻合口局部复发（non-anastomotic local recurrence, NAR）率5.2%。多因素分析显示pT4期、淋巴结清扫个数不足及较高的CEA水平显著增加AR风险；而NAR的独立危险因素为术中出血量增加、低分化、pT4期及pN1/2期。进一步分析手术切缘距离与AR的关系显示，对于结肠癌患者，当近端切缘距离≥5cm（对于pT0/1/2/3期患者为≥4cm）或远端切缘距离≥4cm时，AR风险最低。而对于直肠癌患者，当近端切缘距离≥3cm或远端切缘距离≥2cm（对于pT4期患者为≥3cm）时，AR风险最低。AR和NAR患者的生存分析显示，与 NAR 患者相比，AR组患者接受挽救性根治手术的比例相对较高，其复发后中位总生存时间（overall survival, OS）为40.5个月，高于NAR组的28.5个月，但差异无统计意义（P=0.203）。进一步分析发现，在NAR患者中，接受挽救性根治手术的患者3年OS率可达86.1%，显著优于AR组（51.2%）及未接受根治性手术NAR组（38.2%）（P值均<0.05）。

结论

1.pT分期、淋巴结清扫个数、术前CEA水平与II~III期结直肠癌患者AR具有显著相关性。

2.对于T4期患者，适当增加手术切缘距离可以最大程度降低AR风险。在结肠癌患者中，当近端切缘距离≥5cm时术后AR率最低；在直肠癌患者中，当远端切缘距离≥3cm时术后AR率最低。

3. AR和NAR患者复发后的OS相当，但实现挽救性根治手术的患者复发后OS更好，提示能否达到根治性切除是局部复发患者长期生存的关键。

第一部分：II~III期结直肠癌根治术后吻合口复发与病理切缘距离相关性研究

背景

吻合口复发（anastomotic recurrence, AR）是一类特殊的局部复发，发生率约为0.4%至4.2%，目前对其危险因素和预后意义尚不明确。尽管普遍认为AR与切缘距离不足有关，但临床对最佳切缘距离尚无共识，且仍有很多患者根治性术后会出现AR。因此有必要通过大规模、多中心队列研究来明确高危人群，评估最佳手术切缘距离，并分析AR患者的肿瘤学预后。

方法

本研究共收集了6019例2009年1月至2019年12月期间接受根治性手术的病理II~III期结直肠癌患者，患者人群来自于中国国家结直肠癌队列研究数据库和中国医学科学院北京协和医院数据库。详细收集了患者远近切缘距离、局部复发情况等临床病理资料。采用Cox模型进行单因素和多因素分析，Kaplan–Meier曲线用于绘制复发后的生存曲线。

结果

本研究最终纳入4087例II~III期结直肠癌患者进行分析，其中AR率为1.9%，非吻合口局部复发（non-anastomotic local recurrence, NAR）率5.2%。多因素分析显示pT4期、淋巴结清扫个数不足及较高的CEA水平显著增加AR风险；而NAR的独立危险因素为术中出血量增加、低分化、pT4期及pN1/2期。进一步分析手术切缘距离与AR的关系显示，对于结肠癌患者，当近端切缘距离≥5cm（对于pT0/1/2/3期患者为≥4cm）或远端切缘距离≥4cm时，AR风险最低。而对于直肠癌患者，当近端切缘距离≥3cm或远端切缘距离≥2cm（对于pT4期患者为≥3cm）时，AR风险最低。AR和NAR患者的生存分析显示，与 NAR 患者相比，AR组患者接受挽救性根治手术的比例相对较高，其复发后中位总生存时间（overall survival, OS）为40.5个月，高于NAR组的28.5个月，但差异无统计意义（P=0.203）。进一步分析发现，在NAR患者中，接受挽救性根治手术的患者3年OS率可达86.1%，显著优于AR组（51.2%）及未接受根治性手术NAR组（38.2%）（P值均<0.05）。

结论

1.pT分期、淋巴结清扫个数、术前CEA水平与II~III期结直肠癌患者AR具有显著相关性。

2.对于T4期患者，适当增加手术切缘距离可以最大程度降低AR风险。在结肠癌患者中，当近端切缘距离≥5cm时术后AR率最低；在直肠癌患者中，当远端切缘距离≥3cm时术后AR率最低。

3. AR和NAR患者复发后的OS相当，但实现挽救性根治手术的患者复发后OS更好，提示能否达到根治性切除是局部复发患者长期生存的关键。

第二部分：单细胞测序揭示结直肠癌吻合口复发癌的免疫特征

背景

关于结直肠癌吻合口复发（anastomotic recurrence, AR）的机制研究较为有限，尤其缺乏通过前瞻性队列、利用挽救性手术标本结合单细胞RNA测序（single-cell RNA sequencing, scRNA-seq）技术深入解析AR机制的研究。既往研究将AR发生机制归因于肿瘤细胞的腔内植入或异时性致癌作用，但缺乏细胞与分子层面的直接证据支撑。本研究创新性针对结直肠癌AR进行scRNA-seq分析的研究，旨在通过系统性分析AR患者肿瘤微环境特征，揭示潜在致病机制，为优化结直肠癌患者的治疗、监测、干预提供理论依据与实践指导。

方法

自2023年2月起，本研究前瞻性纳入在中国医学科学院肿瘤医院接受挽救性根治手术的AR患者，收集其新鲜肿瘤组织标本。同时，根据AR患者原发肿瘤病理分期，配对收集原发结直肠癌患者（primary tumor, PT）新鲜肿瘤组织样本。对AR及PT患者新鲜肿瘤组织进行scRNA-seq测序分析，旨在系统比较两组患者的肿瘤微环境组成及细胞异质性特征，深入探索AR发生的潜在机制及其与原发肿瘤的生物学差异。

结果

本研究共收集了10例结直肠癌患者新鲜肿瘤组织进行了scRNA-seq测序分析，其中包括5例AR患者及5例PT患者。通过对AR和PT中细胞亚群进行对比分析，发现T细胞在AR癌中相对比例上升。进一步构建T细胞亚群的细胞通讯网络，发现T细胞群在AR与PT中均表现出密切的细胞间信号交互。考虑到滤泡辅助性T细胞（T follicular helper cell, Tfh）在调控适应性免疫应答，尤其是B细胞活化与分化中的核心作用，本研究进一步聚焦于Tfh的功能特征。在分析生发中心B细胞（Germinal Center B cell）与不同B细胞亚群的互作关系中发现，在PT样本中，B细胞亚群（B-14）与其他B细胞的交互显著活跃，显著富集于“CD22-PTPRC”、“MIF-（CD74+CXCR4）”以及“MIF-（CD74+CD44）”等经典细胞通讯通路中。而在AR样本中，尽管生发中心B细胞在“CD22-PTPRC”通路中仍存在一定程度的交互，但在“MIF-相关”通路中的富集程度明显减弱。上述结果提示，Tfh细胞可能通过调节B细胞亚群之间的通讯差异，在AR与PT的肿瘤微环境中发挥不同的免疫调控作用。因此，本研究进一步深入分析Tfh在AR与PT中的功能状态和分子特征，以期揭示其在结直肠癌吻合口复发发生中的潜在机制。

结论

1. AR与PT癌中T淋巴细胞的组成比例存在明显差异，其中Tfh在调控肿瘤相关免疫反应中发挥关键作用。

2. Tfh细胞在AR中与抗原递呈相关的信号通路中表现出较弱的细胞通讯活性，提示其在B细胞辅助及抗原递呈功能方面可能存在功能障碍。

3. 相比于PT癌，Tfh细胞在AR中的功能在肿瘤经典调控中更为富集，可能在肿瘤的发展过程中发挥重要调控作用。

Abstract

Part I: Association between pathological resection margin and anastomotic recurrence after radical surgery for stage ii-iii colorectal cancer

Background

Anastomotic recurrence (AR) is a special type of local recurrence with an incidence of approximately 0.4% to 4.2%. Currently, its risk factors and prognostic significance are still unclear. Although it is generally believed that AR is related to insufficient surgical margin distance, there is no clinical consensus on the optimal surgical margin distance, and many patients still experience AR after radical surgery. Therefore, it is necessary to conduct large-scale, multicenter cohort studies to identify high-risk groups, evaluate the optimal surgical margin distance, and analyze the oncological prognosis of AR patients.

Method

A total of 6019 patients with pathological stage II-III colorectal cancer who underwent radical surgery from January 2009 to December 2019 were collected in this study. The patient population was from the China National Colorectal Cancer Cohort Study Database and the Peking Union Medical College Hospital Database of the Chinese Academy of Medical Sciences. The clinical pathological data of the patients, such as the distance between the distal and proximal margins and local recurrence, were collected in detail. The Cox model was used for univariate and multivariate analysis, and the Kaplan–Meier curve was used to draw the post-recurrence survival curve of the patients.

Results

A total of 4087 patients with stage II-III colorectal cancer were included in this study, with an AR rate of 1.9% and a non-anastomotic local recurrence (NAR) rate of 5.2%. Multivariate analysis showed that pT4 stage, inadequate lymph node dissection, and higher CEA levels significantly increased the risk of AR; independent risk factors for NAR were increased intraoperative blood loss, poor differentiation, pT4 stage, and pN1/2 stage. Further analysis of the relationship between pathological surgical margin distance and AR showed that for patients with colon cancer, the AR risk was lowest when the proximal margin distance was ≥5cm (≥4cm for pT0/1/2/3 stage) or the distal margin distance was ≥4cm. For patients with rectal cancer, the AR risk was lowest when the proximal margin distance was ≥3cm or the distal margin distance was ≥2cm (≥3cm for pT4 stage). The survival analysis results of AR and NAR patients showed that compared with NAR patients, AR patients had a higher rate of salvage radical surgery, with an overall survival (OS) of 40.5 months after median recurrence, which was higher than the 28.5 months of the NAR group, but the difference was not statistically significant (P=0.203). In addition, the 3-year OS of NAR patients who underwent salvage radical surgery was as high as 86.1%, which was significantly better than that of the AR group (51.2%) and the NAR group who did not receive radical surgery (38.2%) (P<0.05).

Conclusion

1. pT stage, number of lymph node dissection, and preoperative CEA level are significantly correlated with AR in patients with stage II-III colorectal cancer.

2. For patients with T4 stage, appropriately increasing the surgical margin distance can minimize the risk of AR. In patients with colon cancer, the postoperative AR rate is lowest when the proximal margin distance is ≥5cm; in patients with rectal cancer, the postoperative AR rate is lowest when the distal margin distance is ≥3cm.

3. The OS of AR and NAR patients after recurrence is comparable, but patients who achieve salvage radical surgery have better OS after recurrence, suggesting that whether radical resection can be achieved is the key to long-term survival of patients with local recurrence.

Part II: Single-cell sequencing reveals immune characteristics of anastomotic recurrence in colorectal cancer

Background

There are limited studies on the mechanism of anastomotic recurrence (AR) in colorectal cancer, especially those based on prospective cohorts, using salvage radical surgery specimens combined with single-cell RNA sequencing (scRNA-seq) technology to deeply analyze the mechanism of AR recurrence. Previous studies attributed the mechanism of AR to the intraluminal implantation of tumor cells or metachronous carcinogenesis, but lacked direct evidence at the cellular and molecular levels. This study is the first to conduct scRNA-seq analysis on colorectal cancer AR. It aims to reveal the potential pathogenic mechanism by systematically analyzing the differences in tumor microenvironment characteristics between AR patients, and to provide theoretical basis and practical guidance for optimizing the treatment, monitoring, and intervention of colorectal cancer patients.

Method

Since February 2023, this study prospectively enrolled AR patients who underwent salvage radical surgery at the Cancer Hospital of the Chinese Academy of Medical Sciences, and collected fresh tumor tissue specimens during surgery. At the same time, according to the pathological stage of the primary tumor of AR patients, fresh tumor tissue samples from primary colorectal cancer patients (primary tumor, PT) were collected in pairs. scRNA-seq sequencing analysis was performed on fresh tumor tissues of AR and PT patients, aiming to systematically compare the tumor microenvironment composition and cellular heterogeneity characteristics of the two groups of patients, and to deeply explore the potential mechanism of AR occurrence and its biological differences from primary tumors.

Results

In this study, fresh tumor tissues from 10 patients with colorectal cancer were collected for scRNA-seq sequencing analysis, including 5 AR patients and 5 PT patients. By comparing the cell subsets in AR and PT, it was found that the relative proportion of T cells in AR cancer increased. The cell communication network of T cell subsets was further constructed, and it was found that T cell populations showed close cell-to-cell signaling interactions in both AR and PT. Considering the core role of T follicular helper cells (Tfh) in regulating adaptive immune responses, especially B cell activation and differentiation, this study further focused on the functional characteristics of Tfh. In the analysis of the interaction between germinal center B cells and different B cell subsets, it was found that in PT samples, the interaction between B cell subsets (B-14) and other B cells was significantly active, and was significantly enriched in classic cell communication pathways such as "CD22-PTPRC", "MIF-(CD74+CXCR4)" and "MIF-(CD74+CD44)". In AR samples, although there was still a certain degree of interaction between germinal center B cells in the "CD22-PTPRC" pathway, the enrichment in the "MIF-related" pathway was significantly weakened. The above results suggest that Tfh cells may play different immunoregulatory roles in the tumor microenvironment of AR and PT by regulating the differences in communication between B cell subsets. Therefore, this study further analyzed the functional status and molecular characteristics of Tfh in AR and PT, in order to reveal its potential mechanism in the occurrence of anastomotic recurrence in colorectal cancer.

Conclusion

1. There are significant differences in the composition ratio of T lymphocytes between AR and PT samples, among which Tfh plays a key role in regulating tumor-related immune responses;

2. Tfh cells show weaker cell communication activity in the signaling pathway related to antigen presentation in AR, suggesting that they may have functional disorders in B cell assistance and antigen presentation functions;

3. Compared with PT samples, the function of Tfh cells in AR is more enriched in the classical regulation of tumors and may play an important regulatory role in the development of tumors.