第一部分

背景：

乳腺癌是全球第二大癌症致死因素，局部复发和远处转移是主要致死原因。其中三阴乳腺癌（TNBC）亚型组织学分级高、分期晚、远处转移率高、 淋巴细胞浸润密度高，属于侵袭力最强、预后最差的亚型。OTUD6B是卵巢肿瘤相关蛋白酶（OTU）家族成员之－，参与了包括蛋白质特异性降解、信号转导和蛋白定位等几乎所有的生理活动, 与多种肿瘤的发生发展息息相关。在这项研究中，我们使用来自一组公共数据库的数据，全面分析了OTUD6B在乳腺癌（BRCA）中的作用，其次结合OTUD6B的靶向底物叉头框3a(FOX3a)，探讨了二者在三阴性乳腺癌（TNBC）中的作用机制。

方法：

我们使用癌症基因组图谱（TCGA）数据和基因表达综合数据库（GEO）分析了BRCA和BRCA亚组中OTUD6B的mRNA表达，并通过人类蛋白免疫组化表达数据库(HPA)的免疫组织化学染色证实结果。使用R软件包分析OTUD6B与乳腺癌分期的相关性，建立ROC诊断曲线、时间依赖性生存ROC曲线和列线图模型。cBioPortal和MethSurv用于分析乳腺癌中OTUD6B基因改变、突变和DNA甲基化，以及它们对预后的影响。TISIDB数据库结合R语言用于分析OTUD6B与免疫细胞浸润、免疫细胞生物标志物和免疫检查点的相关性。进行基因本体论(GO)分析和京都基因和基因组百科全书(KEGG)分析以研究OTUD6B基因的功能。此外，针对TNBC,首先来验证泛素化修饰是否参与TNBC细胞中FOXO3a的水平调控及OTU家族是否参与调控FOXO3a的水平。其次qPCR检测TNBC患者癌及癌旁组织OTUD6B 表达水平,CCK-8 及Transwell实验检测细胞增殖及迁移, 免疫共沉淀实验来验证FOXO3a与OTUD6B的相互作用。

结果：

在23种癌症中发现了OTUD6B的过表达。OTUD6B的过表达与乳腺癌中的分子分型，雌激素受体（ER）、孕激素受体（PR）及组织学类型显着相关,OTUD6B表达较高的患者预后较差。OTUD6B表达可以从正常组织中识别肿瘤（AUC=0.603）。1年、3年和5年生存ROC下的面积均高于0.5。OTUD6B基因改变率为13%, 体细胞突变频率为0.1%。在14个DNA甲基化CpG位点中，有4个与BRCA的预后相关。KEGG通路富集分析揭示了OTUD6B在RNA的转运、细胞周期通路富集。OTUD6B的表达与乳腺癌中的17种免疫细胞呈现正相关或者负相关，和免疫检查点细胞毒性T淋巴细胞抗原4（CTLA4）和程序性细胞死亡蛋白配体1（PD-L1）呈正相关。在TNBC细胞中，OTUD6B可能是 FOXO3a 的负调控因子，细胞功能实验结果表明：干扰OTUD6B显著抑制了TNBC细胞的增殖及迁移。其次Western blot、qPC及CoIP检测说明TNBC细胞中OTUD6B的过表达显著增强了FOXO3a的泛素化并下调了FOXO3a蛋白的水平，而对其mRNA水平无影响。鉴于MDM2在FOXO3a泛素化降解过程中具有重要作用，我们将MDM2作为候选研究对象。经Western blot实验发现OTUD6B 过表达明显增强了TNBC细胞中 MDM2的蛋白水 平，而对其mRNA水平无影响。通过免疫共沉淀实验发现OTUD6B 能够与 MDM2 进行结合，并且抑制MDM2的泛素化降解。

结论：

总体而言，OTUD6B在乳腺癌中具有良好预后价值，可能参与乳腺癌的免疫调节，有希望成为乳腺癌潜在的治疗靶点。此外，在细胞学功能水平上，OTUD6B可能通过结合MDM2并对其进行去泛素化，稳定TNBC细胞内MDM2水平， 进而增强MDM2对FOXO3a的泛素化降解，下调其稳定性和活性， 进而促进三阴性乳腺癌发生和侵袭转移。

第二部分

背景：

乳腺腺样囊性癌(adenoid cystic carcinoma, ACC)作为一种罕见类型的乳腺恶性肿瘤，肿瘤学行为尚不清楚。本研究回顾分析了乳腺ACC的临床表现、影像学特征及临床病理特征。探讨了乳腺ACC的预后相关因素且构建预后列线图，以便准确预测乳腺ACC患者的总生存期（overall survival，OS）。

方法：

基于监测、流行病学和结果（Surveillance, Epidemiology, and End Results，SEER）及中国医学科学院肿瘤医院的数据，我们筛选出病理诊断为乳腺腺样囊性癌420例患者的影像学特征、临床病理及治疗信息。将SEER数据库的患者按照7∶3分为训练集（285例）和验证集（121例）。基于R软件，对比SEER数据库与本院数据库中患者的临床病理特征，在训练集中采用单因素及多因素Cox回归模型、评价预后影响因素，并进一步构建列线图预后模型。最后，在训练集和验证集中使用C指数、ROC曲线(Receiver operating characteristic curve)和校准曲线、DCA曲线（Decision curve analysis）对列线图的效能及效益进行评价。

结果：

SEER数据库筛选出406例符合纳入标准的乳腺腺样囊性癌患者（训练集285例，验证集121例）。本院共14例患者纳入。本院数据提示乳腺腺样囊性癌最常见的临床表现是可触及的肿块（85.7%），所有本院患者的彩色多普勒超声和乳房X线摄影的影像学特征均无特异性。本院6例患者（42.9%）通过细针抽吸细胞学（FNAC）怀疑患为ACC，并通过术后组织学和免疫组织化学证实。SEER数据与中国医学科学院肿瘤医院数据对比，本医院的乳腺腺样囊性癌的保乳患者为4例（28.6%），它是低于SEER数据库乳腺腺样囊性癌保乳手术患者的例数273例(67.2%)。同时本院乳腺癌施行改良根治术比例4(28.6%)明显高于SEER数据库52(12.8%)。虽然Her-2状态及分子亚型在两个数据之间有统计学差异，可能是SEER数据库2010年前的占的比例较大造成。就Her-2而言，2010年前的患者的比例为227例 (55.9%)，明显高于Her-2阴性病例177例(43.6%)及阳性病例2例 (0.5%)。通过单因素和多因素分析显示年龄、组织学分级、Ｎ分期3个指标均是乳腺腺样囊性癌OS的独立预后因素（P均<0.05）。基于上述这些独立预后因素构建列线图模型，列线图在训练集中的C-指数为0.738，验证集的C-指数为0.803。训练集预测3、5、10年总生存率的ROC曲线下面积分别为0.749、0782、0.80；验证集预测3、5、10年总生存率的ROC曲线下面积分别为0.780、0.833、0.779，提示该列线图具有良好的预后预测能力。同时，校准曲线提示列线图在训练集和验证集中均有较好的校准度。决策曲线发现该模型在临床净效益方面较其他因素获益明显。

结论：

(1).乳腺ACC不能简单概括为三阴性乳腺癌，因为它还包括少数激素受体阳性乳腺癌。(2).临床表现及影像学检查很难确定术前诊断，FNAC可能是诊断的有用工具。(3).中国医学科学院肿瘤医院的乳腺腺样囊性癌患者行保留乳房的乳腺癌切除术的比例低于SEER数据库中乳腺腺样囊性癌行保乳手术的比例。(4).基于年龄、组织学分级、N分期的独立预后因素构建的列线图模型可有效预测乳腺腺样囊性癌患者的总生存率，对于指导治疗和评估患者预后具有一定参考价值。

第一部分

Background:

Breast cancer is the second leading cause of cancer death in the world, of which recurrence and metastasis are the main causes of death. Triple-negative breast cancer (TNBC) has high histological grade, advanced stage, high rate of distant metastasis, and high lymphocyte infiltration density. OTUD6B is a member of the ovarian tumor-associated protease (OTU) family, which is involved in almost all physiological activities including protein-specific degradation, signal transduction and protein localization, and is closely related to the occurrence and development of various tumors. In this study, we comprehensively analyzed the role of the deubiquitinating enzyme OTUD6B in breast cancer (BRCA) by using data from a set of public databases. Secondly, combined with the target substrate of OTUD6B Forkhead box 3a (FOXO3a), the mechanism of action of the two in triple-negative breast cancer was discussed.

Methods:

We firstly downloaded the transcriptome sequencing data of BRCA and BRCA subgroups from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) database, respectively. Then we analyzed the OTUD6B mRNA expression and further demonstrated the protein expression by Human Protein Atlas (HPA). Among them, R package was used to analyze the correlation between OTUD6B and breast cancer stage. ROC diagnostic curve, time dependent survival ROC curve and nomogram model were subsequently established. The cBioPortal and MethSurv were carried out to analyze the gene changes, mutation and DNA methylation of OTUD6B in breast cancer, and their effects on clinical prognosis. R software was then used to analyze data from TISIDB database in order to explore the correlation between OTUD6B and immune cell infiltration, immune cell biomarkers and immune checkpoints. And Gene Ontology (GO) analysis and Kyoto Encyclopedia of genes and genomes (KEGG) analysis were conducted to study the function of OTUD6B gene. What is more, for TNBC, First, we verified whether ubiquitination was involved in the regulation of FOXO3a levels in TNBC cells and whether the OTU family was involved in the regulation of FOXO3a levels. Secondly, qPCR was used to detect the expression levels of OTUD6B in TNBC patients' cancer and adjacent tissues, CCK-8 and Transwell assays were used to detect cell proliferation and migration, and co-immunoprecipitation experiments were used to verify the interaction between FOXO3a and OTUD6B.

Results:

We found the OTUD6B is highly expressed in 23 cancers and OTUD6B is significantly correlated with molecular typing, estrogen, progesterone, and histological types of breast cancer. Patients with high expression of OTUD6B have a poor prognosis. OTUD6B helps to distinguish cancer from normal tissue (AUC=0.603). The area of 1-year, 3-year and 5-year survival ROC was higher than 0.5. The change rate of OTUD6B gene was 13%, and the somatic mutation frequency was 0.1%. Of the 14 DNA methylated CpG sites, 4 were associated with the prognosis of BRCA. The expression of OTUD6B was positively correlated or negatively correlated with 17 kinds of immune cells in breast cancer, and positively correlated with immunophenotyping cytotoxic T lymphocyte antigen 4 (CTLA4) and programmed cell death protein ligand 1 (PD-L1). KEGG pathway enrichment analysis revealed that OTUD6B was enriched in RNA transport and cell cycle pathways. In TNBC cells, OTUD6B may be a negative regulator of FOXO3a. The results of cell function experiments showed that interfering with OTUD6B significantly inhibited the proliferation and migration of TNBC cells. Secondly, Western blot, qPC and CoIP detection showed that overexpression of OTUD6B in TNBC cells significantly enhanced the ubiquitination of FOXO3a and down-regulated the level of FOXO3a protein, but had no effect on its mRNA level. Given that MDM2 plays an important role in the ubiquitination and degradation of FOXO3a, we selected MDM2 as a candidate for study. Western blot showed that overexpression of OTUD6B significantly enhanced the protein level of MDM2 in TNBC cells, but had no effect on its mRNA level. Through co-immunoprecipitation experiments, it was found that OTUD6B can bind to MDM2 and inhibit the ubiquitination and degradation of MDM2.

Conclusions:

Overall, OTUD6B has good prognostic value and may be involved in the immune regulation of breast cancer, as well as a potential therapeutic target for patients with breast cancer. In addition, at the level of cytological function, OTUD6B may stabilize the level of MDM2 by binding and deubiquitinating MDM2 in TNBC cells, thereby enhancing the ubiquitination and degradation of FOXO3a by MDM2, downregulating its stability and activity, which results in occurrence, invasion and metastasis of triple-negative breast cancer.

第二部分

Background:

Adenoid cystic carcinoma (ACC) of the breast is a rare type of breast cancer with unknown oncological behavior. This study retrospectively analyzed the clinical manifestations, imaging features and clinicopathological features of breast ACC, explored its prognostic factors, and constructed a prognostic nomogram in order to accurately predict the overall survival (OS) of breast ACC patients.

Methods:

Based on the Surveillance, Epidemiology, and End Results (SEER) and the Cancer Hospital of the Chinese Academy of Medical Sciences^，data, we screened out 420 patients with pathologically diagnosed breast ACC for their imaging features, clinical features Information on pathology and treatment. The patients in SEER database were divided into training set (n=285) and validation set (n=121) according to 7:3. Based on R software, the clinicopathological characteristics of patients in SEER database and our hospital database were compared. At the same time, univariate and multivariate Cox regression model were used in the training set to evaluate the prognostic factors, and the nomogram prognostic model was further constructed. Finally, the performance and benefit of the nomogram were evaluated using the C index, ROC curve (Receiver operating characteristic curve), calibration curve, and DCA curve (Decision curve analysis) in the training set and validation set.

Results:

We screened 406 patients with adenoid cystic carcinoma of the breast who met the inclusion and exclusion criteria (285 in the training set and 121 in the validation set) from the SEER database. A total of 14 patients were included in our hospital. Our data suggest that the most common clinical manifestation of breast ACC is a palpable mass (85.7%), and the imaging features of Color Doppler ultrasonography and mammography in all our patients were nonspecific. ACC was suspected by fine needle aspiration cytology (FNAC) in 6 patients (42.9%) in our hospital and confirmed by postoperative histology and immunohistochemistry. Compared with the SEER and the data of our hospital, the proportion of breast-conserving patients in our hospital 4 (28.6%) was lower than that in the SEER database 273 (67.2%). At the same time, the proportion of modified radical mastectomy for breast cancer in our hospital 4 (28.6%) was significantly higher than that in SEER database 52 (12.8%). Although Her-2 status and molecular subtype were statistically different between the two data, it may be due to the larger proportion of SEER database before 2010. For Her-2 status, the proportion of patients before 2010 was 227 (55.9%), which was significantly higher than the 177 (43.6%) of Her-2-negative cases and 2 (0.5%) of positive cases. Significant factors affecting the prognosis, including age, histological grade and N stage were taken into the nomogram based on univariate and multivariate Cox regression analysis. The C-index of the nomogram in the training set was 0.738, while the areas under the ROC curve of the training set for predicting 3, 5, and 10-year overall survival were 0.749, 0.782, and 0.80, respectively. the c-index of the validation set was 0.803 and the areas under the ROC curve for the validation set to predict 3, 5, and 10-year overall survival were 0.780, 0.833, and 0.779, respectively. Such results suggest that the nomogram has a good prognostic predictive ability. At the same time, the calibration curve shows that the nomogram has a good degree of calibration in both the training set and the validation set. Decision curves found that the model was more beneficial than other factors in terms of net clinical benefit.