第一部分 基于肿瘤相关中性粒细胞和中性粒细胞胞外诱捕网构建结直肠癌预后模型

摘要

研究背景和目的 在全球范围内，结直肠癌（Colorectal cancer，CRC）位居癌症发病率第三位，且其年发病率呈现持续增长趋势。在CRC的发生及进展过程中，肿瘤微环境（Tumor microenvironment，TME）始终扮演着重要角色。肿瘤相关中性粒细胞（Tumor associated neutrophils，TANs）作为TME中的一个关键构成要素，其在CRC肿瘤微环境中发挥的功能受到广泛关注。有趣的是，TANs内部存在异质性，并且具有显著的功能可塑性，根据不同的信号表达，TANs可以极化为抑制肿瘤的“N1”表型或促进肿瘤的“N2”表型。TANs还可释放中性粒细胞胞外诱捕网（Neutrophil extracellular traps，NETs），这是一种包含DNA和组蛋白复合物的网状结构，其在炎症反应、感染性免疫应答及某些肿瘤发展过程中的作用已得到证实。有研究认为TANs可以通过释放NETs来参与肿瘤的进展，从而导致较差的生存预后，然而并未考虑TANs亚型的影响。因此，本研究通过利用NETs相关基因构建预后模型，并且进一步探讨N2-TAN与NETs之间的联系及其在CRC患者临床预后中的潜在价值，为揭示TANs在CRC肿瘤微环境中扮演的角色提供新的见解。

材料和方法 在这项研究中，我们从TCGA数据库和GEO数据库中检索了CRC患者的转录组数据和临床信息。我们还收集并汇总了与NETs相关的基因，基于这些基因，使用一致性聚类方法，按照基因表达富集程度将患者分为两个亚组。接下来，我们获取了两个亚组之间的差异表达基因（Differentially expressed gense，DEGs），通过单因素Cox分析筛选出与预后相关的基因，随后利用最小绝对收缩和选择算子（Least absolute shrinkage and selection operator，Lasso）回归分析和多因素Cox分析构建了一个NETs评分模型（NETscore），并通过中位值将人群分为NETscore高分组和低分组。随后利用Kaplan-Meier生存分析和受试者工作特征（Receiver operating characteristic，ROC）曲线分析评估NETscore区分患者总生存期（Overall survival，OS）的能力。更进一步，我们探索了两组人群中CRC肿瘤免疫微环境的景观和免疫检查点相关基因的表达。通过分析两组人群的微卫星状态、肿瘤突变负荷和免疫表型评分，以评估不同人群对免疫疗法的应答差异。接下来，我们利用结直肠癌组织芯片进行实验验证。本研究中HOXC6是NETscore的构建基因之一，瓜氨酸组蛋白H3（H3Cit）是构成NETs的重要组成部分。所以我们通过免疫组化定量NETscore（HOXC6）与NETs（H3Cit），并分析其相关性。另一方面，采用免疫荧光共定位标记N2-TAN，分析其与NETs的关系。我们还对CD8+T细胞这一特异性免疫细胞进行定量分析。最后，分析上述指标与OS的关系，评估其临床预后价值。

研究结果 我们的研究构建了一个由11个NETs相关基因组成的NETscore。通过内部验证和外部验证均证明NETscore高分组的OS远低于低分组（p<0.001），并且NETscore具有良好的预测预后的能力。我们进一步联合其他临床变量构建了一个诺模图，并在验证集中取得了理想的预测效能。研究NETscore与免疫细胞浸润的关系，结果提示中性粒细胞主要富集在NETscore高分组，而CD8+T细胞主要富集在NETscore低分组，并且低分组的免疫评分显著高于高分组（p<0.05）。NETscore与肿瘤突变负荷呈负相关，并且微卫星状态高度不稳定患者在低分组占比远高于高分组，低分组人群的免疫表型评分也显著高于高分组（p<0.05）。进一步的免疫组化结果显示，HOXC6与H3Cit表达呈正相关，且均与较差的预后有关（p<0.05）。通过免疫荧光共定位，结果提示H3Cit显著集中于N2-TAN高浸润组（p<0.001）。生存分析结果提示N2-TAN是影响OS的独立风险因素（p<0.05），并且N2-TAN与CD8+T细胞的比值（N2-TAN/CD8）能更好地对患者预后进行区分（p=0.001）。 研究结论 本研究通过构建NETscore全面分析了NETs与CRC患者的临床预后、免疫浸润和临床病理特征的关系, 并进一步阐明了NETscore预测患者生存的能力。通过更深层次研究肿瘤微环境中TANs与NETs的关系，提示H3Cit可能主要由N2-TAN产生，并且揭示了N2-TAN显著的预后价值，而且与CD8+T细胞的结合，可以更好地预测患者生存。本研究还探究了NETscore可以预测人群对免疫治疗的敏感性。这些发现突出了NETs和N2-TAN的关键临床意义，并为指导CRC患者的个性化免疫治疗策略提供了新的思路。

 第二部分 直肠神经内分泌瘤G2期患者的临床病理特征与预后分析

摘要

研究背景和目的 神经内分泌肿瘤（Neuroendocrine neoplasm，NEN）是一类罕见肿瘤，在消化道神经内分泌肿瘤中，直肠神经内分泌瘤（Rectal neuroendocrine tumor，R-NET）的发病率较高，并且其预后相对较好。然而R-NET G2因其发病率低，研究相对匮乏，目前尤其缺乏针对R-NET G2的临床病理特征和生存预后的研究，所以在诊治过程中尚未明确统一的意见。本研究旨在总结和分析R-NET G2的临床病理特征，研究其影响预后的因素，并总结相关治疗经验。

材料和方法 对中国医学科学院肿瘤医院自2003年1月至2023年9月收治的经病理学诊断确诊为R-NET G2患者的临床资料进行了回顾性分析。通过电子病历系统收集患者基本信息包括：性别、年龄、现病史、诊治经过、手术方式、肿瘤TNM分期、病理结果、术后辅助治疗情况等。通过门诊记录和电话方式对患者进行随访，来确认患者生存状态。我们采用Fisher精确检验与Kaplan-Meier曲线对病理特征与预后间的关系进行了分析。

研究结果 本研究共纳入22例符合条件的患者，患者年龄32-76岁，中位年龄为52岁。其中男性12例，女性10例。复习其临床病理资料并对其进行随访，其中1例患者因术后间隔不足1月，未进行随访。其余21例患者获得随访，随访时间为6-98月，中位随访时间为42月，随访期间5例患者因肿瘤进展而死亡。整体患者的1、3、5 年肿瘤特异性生存率分别为100%、92.9%、69.6%。随访的21例患者中19例接受手术治疗，4例患者因肿瘤肝转移术后肿瘤进展，药物治疗不理想而死亡，5例患者术前/术后出现远处转移，现带瘤生存，同时行巩固性内科治疗，其余10例术后患者目前无病生存；2例患者因无法手术而行全程内科治疗，其中1例随访95个月患者仍带瘤生存，另外1例患者在初诊25个月后死于肿瘤进展。对患者进行亚组生存分析发现，肿瘤直径≥2cm，T分期为T2-3，伴有淋巴结转移或远处转移的患者预后较差。此外，与无远处转移的患者相比，伴有肝转移患者的原发瘤直径较大，T分期较高（P<0.05）。

研究结论 相比于R-NET G1，R-NET G2恶性程度更高，并且具有较高的转移倾向。临床医生应根据肿瘤大小、侵犯深度、淋巴结状态、存在远处转移以及远处转移的部位和程度等因素制定相应的诊疗策略。对于未发生远处转移的R-NET G2，需综合考虑肿瘤的大小、浸润程度和淋巴结状态，选择适当的治疗方式。若肿瘤已出现远处转移，则需要通过多学科协作的综合治疗方式，根据具体情况，个体化地选择治疗方案。

Part I Construction of a prognostic model for colorectal cancer based on tumor-associated neutrophils and neutrophil extracellular traps

Abstract

Background and Purpose  Globally, colorectal cancer (CRC) ranks third in cancer incidence, with its annual incidence showing a continuous upward trend. Throughout the development and progression of CRC, the tumor microenvironment (TME) plays a crucial role. Tumor-associated neutrophils (TANs), as a key component of the TME, have garnered extensive attention for their function within the CRC tumor microenvironment. Interestingly, TANs exhibit significant functional plasticity, capable of polarizing into a tumor-suppressive "N1" phenotype or a tumor-promoting "N2" phenotype in response to microenvironmental signals. TANs can also release neutrophil extracellular traps (NETs), a web-like structure composed of DNA and histone complexes, whose role in inflammatory responses, infectious immune responses, and the development of tumors has been confirmed. Some studies have suggested that TANs can participate in tumour progression by releasing NETs, leading to a poorer survival prognosis, however the effect of TANs subtype was not considered. This study aims to further explore the relationship between N2-TAN and NETs and their potential value in the clinical prognosis of CRC patients by constructing a prognostic model based on NETs-related gene expression, providing new insights into the role of TANs in the TME of CRC.

Materials and Methods In this study, we retrieved transcriptomic data and clinical information of CRC patients from the TCGA and GEO databases. We also collected and summarized genes related to NETs and, used a consensus clustering method to divide patients into two subgroups according to the enrichment degree of gene expression. Next, we obtained differentially expressed genes (DEGs) between two subgroups and screened for prognostic related genes through univariate Cox analysis and constructed a NETs scoring model (NETscore) using the least absolute shrinkage and selection operator (Lasso) regression analysis and multivariate Cox analysis. Patients were divided into high and low NETscore groups based on the median value. The surival differences among groups were analyzed using the Kaplan-Meier survival analysis. The capability of NETscore to differentiate patient overall survival (OS) was assessed using the receiver operating characteristic (ROC) curve analysis. Further, we explored the landscape of the CRC tumor immune microenvironment and the expression of immune checkpoint-related genes in the two groups. By analyzing the microsatellite status, tumor mutation burden, and immune phenotype scores of the two populations, we evaluated the differences in response to immunotherapy among different populations. Subsequently, we used colorectal cancer tissue microarrays for experimental validation. In this study, HOXC6 was one of the genes constructing the NETscore, and citrullinated histone H3 (H3Cit) was a crucial component of NETs. Therefore, we quantified NETscore (HOXC6) and NETs (H3Cit) using immunohistochemistry and analyzed their correlation. On the other hand, N2-TAN was labeled using immunofluorescence colocalization to analyze its relationship with NETs. We also quantified CD8+ T cells, a specific immune cell. Finally, the relationship between the above indicators and OS was analysed to assess their clinical prognostic value.

Results  Our study constructed a NETscore composed of 11 NETs-associated genes. Both internal and external validations demonstrated that the prognosis of the high NETscore group was significantly worse than that of the low score group (p<0.001), and NETscore had a good predictive ability for prognosis. A nomogram was further constructed by combining other clinical variables, achieving ideal predictive performance in the validation set. The study of the relationship between NETscore and immune cell infiltration suggested that neutrophils were mainly enriched in the high NETscore group, while CD8+ T cells were mainly enriched in the low NETscore group, and the immune score of the low group was significantly higher than that of the high group (p<0.05). NETscore was negatively correlated with tumor mutation burden, and patients with high microsatellite instability were significantly more prevalent in the low group than in the high group, with the immune phenotype score of the low group also being significantly higher than that of the high group (p<0.05). Immunohistochemistry results showed that the expression of HOXC6 and H3Cit, both associated with poorer prognosis (p<0.05). Immunofluorescence colocalization indicated that H3Cit was significantly concentrated in the high N2-TAN infiltration group (p<0.001). Survival analysis suggested that N2-TAN was an independent risk factor for OS (p<0.05) and that the ratio of N2-TAN to CD8+ T cells (N2-TAN/CD8) better differentiated patients' prognosis (p=0.001).

Conclusions  This study comprehensively analyzed the relationship between NETs and the clinical prognosis, immune infiltration, and clinicopathological characteristics of CRC patients by constructing the NETscore, further elucidating the ability of NETscore to predict patient survival. By further studying the relationship between TAN and NETs in the tumor microenvironment, it suggested that H3Cit may be mainly generated by N2-TAN and revealed the significant prognostic value of N2-TAN and that the combination with CD8+ T cells may better predict patient survival. This study also explored that NETscore can predict the sensitivity of the population to immunotherapy. These findings highlight the critical clinical significance of NETs and N2-TAN and provide new insights for guiding the personalized immunotherapy strategies for CRC patients.

Part II Clinicopathological features and prognosis of rectal neuroendocrine tumor with grade 2

Abstract

Background and Purpose  Neuroendocrine neoplasms (NEN) are a rare category of tumors, among which rectal neuroendocrine tumors (R-NET) have a higher incidence rate in gastrointestinal neuroendocrine tumors, and their prognosis is relatively better. However, due to the low incidence of R-NET G2, research is relatively scarce, especially lacking studies on the clinicopathological features and survival prognosis of R-NET G2, hence there is no clear consensus in the diagnostic and treatment process. This study aims to summarize and analyze the clinicopathological characteristics of R-NET G2, study the factors affecting prognosis, and summarize related treatment experiences.

Material and Methods  A retrospective analysis was conducted on the clinical data of patients diagnosed with R-NET G2 at the Cancer Hospital of the Chinese Academy of Medical Sciences from January 2003 to September 2023. Basic information of patients including gender, age, medical history, diagnosis and treatment process, surgical methods, tumor TNM staging, pathological results, and postoperative adjuvant treatment were collected through the electronic medical record system. Patients were followed up through outpatient records and telephone to confirm their survival status. The relationship between clinicopathological characteristics and relative prognosis was analyzed using Fisher's exact test and the Kaplan-Meier curve.

Results  A total of 22 eligible patients were included in this study, with ages ranging from 32 to 76 years, and a median age of 52 years. There were 12 males and 10 females. Clinical and pathological data were reviewed, and patients were followed up, except for one patient who was not followed up due to less than one month post-operation interval. The remaining 21 patients were followed up, with a follow-up period of 6-98 months and a median follow-up period of 42 months. During the follow-up period, 5 patients died due to tumor progression. The overall 1, 3, and 5-year tumor-specific survival rates were 100%, 92.9%, and 69.6%, respectively. Among the 21 followed-up patients, 19 underwent surgical treatment, 4 of whom died of tumor progression after liver metastasis and unsatisfactory drug treatment, 5 had distant metastasis before/after surgery and are currently living with tumors undergoing consolidation medical treatment, and the remaining 10 postoperative patients are currently living without disease; 2 patients underwent full-course medical treatment due to inoperability, one of whom is still living with the tumor after 95 months of follow-up, and the other died of tumor progression 25 months after diagnosis. Subgroup survival analysis showed that patients with tumor diameter ≥2cm, T stage of T2-3, with lymph node metastasis or distant metastasis had worse prognosis. Moreover, compared to patients without distant metastasis, those with liver metastasis had larger primary tumor diameter and higher T stage (P<0.05).

Conclusions  Compared to R-NET G1, R-NET G2 has a higher malignancy level and a higher tendency to metastasize. Clinicians should formulate corresponding diagnostic and treatment strategies based on factors such as tumor size, depth of invasion, lymph node status, presence of distant metastasis, and the location and extent of distant metastasis. For R-NET G2 without distant metastasis, the tumor size, degree of infiltration, and lymph node status should be comprehensively considered to choose the appropriate treatment method. If distant metastasis has occurred, a multidisciplinary comprehensive treatment approach should be adopted to select the treatment plan individually based on the specific situation.