RESUMO
BACKGROUND: Sjögren's Syndrome (SS) is a rare chronic autoimmune disorder primarily affecting adult females, characterized by chronic inflammation and salivary and lacrimal gland dysfunction. It is often associated with systemic lupus erythematosus, rheumatoid arthritis and kidney disease, which can lead to increased mortality. Early diagnosis is critical, but traditional methods for diagnosing SS, mainly through histopathological evaluation of salivary gland tissue, have limitations. METHODS: The study used 100 labial gland biopsy, creating whole-slide images (WSIs) for analysis. The proposed model, named Cell-tissue-graph-based pathological image analysis model (CTG-PAM) and based on graph theory, characterizes single-cell feature, cell-cell feature, and cell-tissue feature. Building upon these features, CTG-PAM achieves cellular-level classification, enabling lymphocyte recognition. Furthermore, it leverages connected component analysis techniques in the cell graph structure to perform SS diagnosis based on lymphocyte counts. FINDINGS: CTG-PAM outperforms traditional deep learning methods in diagnosing SS. Its area under the receiver operating characteristic curve (AUC) is 1.0 for the internal validation dataset and 0.8035 for the external test dataset. This indicates high accuracy. The sensitivity of CTG-PAM for the external dataset is 98.21%, while the accuracy is 93.75%. In comparison, the sensitivity and accuracy for traditional deep learning methods (ResNet-50) are lower. The study also shows that CTG-PAM's diagnostic accuracy is closer to skilled pathologists compared to beginners. INTERPRETATION: Our findings indicate that CTG-PAM is a reliable method for diagnosing SS. Additionally, CTG-PAM shows promise in enhancing the prognosis of SS patients and holds significant potential for the differential diagnosis of both non-neoplastic and neoplastic diseases. The AI model potentially extends its application to diagnosing immune cells in tumor microenvironments.
Assuntos
Síndrome de Sjogren , Síndrome de Sjogren/diagnóstico , Síndrome de Sjogren/patologia , Humanos , Feminino , Estudos de Coortes , Curva ROC , Processamento de Imagem Assistida por Computador/métodos , Pessoa de Meia-Idade , Aprendizado Profundo , Área Sob a Curva , Adulto , AutomaçãoRESUMO
BACKGROUND: Thyroid cancer is a common thyroid malignancy. The majority of thyroid lesion needs intraoperative frozen pathology diagnosis, which provides important information for precision operation. As digital whole slide images (WSIs) develop, deep learning methods for histopathological classification of the thyroid gland (paraffin sections) have achieved outstanding results. Our current study is to clarify whether deep learning assists pathology diagnosis for intraoperative frozen thyroid lesions or not. METHODS: We propose an artificial intelligence-assisted diagnostic system for frozen thyroid lesions that applies prior knowledge in tandem with a dichotomous judgment of whether the lesion is cancerous or not and a quadratic judgment of the type of cancerous lesion to categorize the frozen thyroid lesions into five categories: papillary thyroid carcinoma, medullary thyroid carcinoma, anaplastic thyroid carcinoma, follicular thyroid tumor, and non-cancerous lesion. We obtained 4409 frozen digital pathology sections (WSI) of thyroid from the First Affiliated Hospital of Sun Yat-sen University (SYSUFH) to train and test the model, and the performance was validated by a six-fold cross validation, 101 papillary microcarcinoma sections of thyroid were used to validate the system's sensitivity, and 1388 WSIs of thyroid were used for the evaluation of the external dataset. The deep learning models were compared in terms of several metrics such as accuracy, F1 score, recall, precision and AUC (Area Under Curve). RESULTS: We developed the first deep learning-based frozen thyroid diagnostic classifier for histopathological WSI classification of papillary carcinoma, medullary carcinoma, follicular tumor, anaplastic carcinoma, and non-carcinoma lesion. On test slides, the system had an accuracy of 0.9459, a precision of 0.9475, and an AUC of 0.9955. In the papillary carcinoma test slides, the system was able to accurately predict even lesions as small as 2 mm in diameter. Tested with the acceleration component, the cut processing can be performed in 346.12 s and the visual inference prediction results can be obtained in 98.61 s, thus meeting the time requirements for intraoperative diagnosis. Our study employs a deep learning approach for high-precision classification of intraoperative frozen thyroid lesion distribution in the clinical setting, which has potential clinical implications for assisting pathologists and precision surgery of thyroid lesions.
Assuntos
Aprendizado Profundo , Secções Congeladas , Câncer Papilífero da Tireoide , Neoplasias da Glândula Tireoide , Humanos , Neoplasias da Glândula Tireoide/patologia , Neoplasias da Glândula Tireoide/diagnóstico , Neoplasias da Glândula Tireoide/cirurgia , Câncer Papilífero da Tireoide/patologia , Câncer Papilífero da Tireoide/diagnóstico , Câncer Papilífero da Tireoide/cirurgia , Carcinoma Papilar/patologia , Carcinoma Papilar/cirurgia , Carcinoma Papilar/diagnóstico , Adenocarcinoma Folicular/patologia , Adenocarcinoma Folicular/diagnóstico , Adenocarcinoma Folicular/cirurgia , Glândula Tireoide/patologia , Glândula Tireoide/cirurgia , Carcinoma Neuroendócrino/patologia , Carcinoma Neuroendócrino/diagnóstico , Carcinoma Neuroendócrino/cirurgia , Feminino , Masculino , Pessoa de Meia-Idade , Adulto , Período Intraoperatório , Carcinoma Anaplásico da Tireoide/patologia , Carcinoma Anaplásico da Tireoide/diagnóstico , Carcinoma Anaplásico da Tireoide/cirurgiaRESUMO
AIMS: To clarify claudin18.2 expression and its clinicopathological features in various cancers, especially in lung adenocarcinoma. METHODS: Immunohistochemistry staining and fluorescence in situ hybridisation (FISH) were performed to detect claudin18.2 expression and CLDN18 gene rearrangement in adenocarcinoma from different organs. RESULTS: The results showed that claudin18.2 expression was found in 68% (27 of 40) of lung mucinous adenocarcinoma, 52% (16 of 31) of cholangiocarcinoma, 2% (10 of 423) of colorectal adenocarcinoma tissue microarray, 27% (6 of 22) of colorectal mucinous adenocarcinoma and 30% (3 of 10) of cervical adenocarcinoma, but not in all 39 cases of invasive breast adenocarcinoma by immunohistochemistry staining. There was significantly positive correlation between ratio of claudin18.2-positive carcinoma cells and staining intensity in lung mucinous adenocarcinoma and cholangiocarcinoma. Claudin18.2 expression was much more in female patients than male patients with lung mucinous adenocarcinoma. In addition, cholangiocarcinoma with claudin18.2 expression was more aggressive and had perineural invasion. Intraductal papillary neoplasm of the bile duct and epithelial dysplasia of the adjacent bile in cholangiocarcinoma also showed claudin18.2 expression. All three cases of cervical adenocarcinoma with claudin18.2 expression were moderately differentiated adenocarcinoma including one human papillomavirus (HPV)-associated carcinoma, two non-HPV-associated and gastric-type carcinoma. CLDN18 gene rearrangement was not found in all 22 cases with high claudin18.2 expression by FISH. CONCLUSIONS: Our results suggest claudin18.2 might be a potential biomarker for targeted therapy on lung mucinous adenocarcinoma, cholangiocarcinoma, colorectal mucinous adenocarcinoma and gastric-type cervical adenocarcinoma.
RESUMO
Digitization of pathological slides has promoted the research of computer-aided diagnosis, in which artificial intelligence analysis of pathological images deserves attention. Appropriate deep learning techniques in natural images have been extended to computational pathology. Still, they seldom take into account prior knowledge in pathology, especially the analysis process of lesion morphology by pathologists. Inspired by the diagnosis decision of pathologists, we design a novel deep learning architecture based on tree-like strategies called DeepTree. It imitates pathological diagnosis methods, designed as a binary tree structure, to conditionally learn the correlation between tissue morphology, and optimizes branches to finetune the performance further. To validate and benchmark DeepTree, we build a dataset of frozen lung cancer tissues and design experiments on a public dataset of breast tumor subtypes and our dataset. Results show that the deep learning architecture based on tree-like strategies makes the pathological image classification more accurate, transparent, and convincing. Simultaneously, prior knowledge based on diagnostic strategies yields superior representation ability compared to alternative methods. Our proposed methodology helps improve the trust of pathologists in artificial intelligence analysis and promotes the practical clinical application of pathology-assisted diagnosis.
Assuntos
Inteligência Artificial , Patologistas , Humanos , Diagnóstico por Computador/métodosRESUMO
RNA binding proteins (RBPs) contributes to cancer progression, but the underlying mechanism reminds unclear. Here, we find that DDX21, a representative RBP, is highly expressed in colorectal cancer (CRC), which leads to CRC cell migration and invasion in vitro, and CRC to liver metastasis and lung metastasis in vivo. This effect of DDX21 on CRC metastasis is correlated to the activation of Epithelial-mesenchymal transition (EMT) pathway. Moreover, we reveal that DDX21 protein is phase separated in vitro and in CRC cells, which controls CRC metastasis. Phase-separated DDX21 highly binds on MCM5 gene locus, which is markedly reduced when phase separation is disrupted by mutations on its intrinsically disordered region (IDR). The impaired metastatic ability of CRC upon DDX21 loss is restored by ectopic expression of MCM5, indicating MCM5 is a key downstream target of DDX21 for CRC metastasis. Furthermore, co-higher expressions of DDX21 and MCM5 is significantly correlated with poor survival outcomes of stage III and IV CRC patients, indicating the importance of this mechanism in CRC late and metastatic stage. Altogether, our results elucidate a new model of DDX21 in regulating CRC metastasis via phase separation.
Assuntos
Neoplasias Colorretais , Transição Epitelial-Mesenquimal , Humanos , Linhagem Celular Tumoral , Transição Epitelial-Mesenquimal/genética , Proteínas/genética , Neoplasias Colorretais/patologia , Movimento Celular/genética , Regulação Neoplásica da Expressão Gênica , Metástase Neoplásica , Proliferação de Células , Proteínas de Ciclo Celular/genética , Proteínas de Ciclo Celular/metabolismo , RNA Helicases DEAD-box/genética , RNA Helicases DEAD-box/metabolismoRESUMO
BACKGROUND: Determining the origin of bone metastatic cancer (OBMC) is of great significance to clinical therapeutics. It is challenging for pathologists to determine the OBMC with limited clinical information and bone biopsy. METHODS: We designed a regional multiple-instance learning algorithm to predict the OBMC based on hematoxylin-eosin (H&E) staining slides alone. We collected 1041 cases from eight different hospitals and labeled 26,431 regions of interest to train the model. The performance of the model was assessed by ten-fold cross validation and external validation. Under the guidance of top3 predictions, we conducted an IHC test on 175 cases of unknown origins to compare the consistency of the results predicted by the model and indicated by the IHC markers. We also applied the model to identify whether there was tumor or not in a region, as well as distinguishing squamous cell carcinoma, adenocarcinoma, and neuroendocrine tumor. FINDINGS: In the within-cohort, our model achieved a top1-accuracy of 91.35% and a top3-accuracy of 97.75%. In the external cohort, our model displayed a good generalizability with a top3-accuracy of 97.44%. The top1 consistency between the results of the model and the immunohistochemistry markers was 83.90% and the top3 consistency was 94.33%. The model obtained an accuracy of 98.98% to identify whether there was tumor or not and an accuracy of 93.85% to differentiate three types of cancers. INTERPRETATION: Our model demonstrated good performance to predict the OBMC from routine histology and had great potential for assisting pathologists with determining the OBMC accurately. FUNDING: National Science Foundation of China (61875102 and 61975089), Natural Science Foundation of Guangdong province (2021A15-15012379 and 2022A1515 012550), Science and Technology Research Program of Shenzhen City (JCYJ20200109110606054 and WDZC20200821141349001), and Tsinghua University Spring Breeze Fund (2020Z99CFZ023).
Assuntos
Adenocarcinoma , Neoplasias Ósseas , Carcinoma de Células Escamosas , Aprendizado Profundo , Humanos , Algoritmos , Neoplasias Ósseas/diagnósticoRESUMO
AIMS: The automation of quantitative evaluation for breast immunohistochemistry (IHC) plays a crucial role in reducing the workload of pathologists and enhancing the objectivity of diagnoses. However, current methods face challenges in achieving fully automated immunohistochemistry quantification due to the complexity of segmenting the tumor area into distinct ductal carcinoma in situ (DCIS) and invasive carcinoma (IC) regions. Moreover, the quantitative analysis of immunohistochemistry requires a specific focus on invasive carcinoma regions. METHODS AND RESULTS: In this study, we propose an innovative approach to automatically identify invasive carcinoma regions in breast cancer immunohistochemistry whole-slide images (WSIs). Our method leverages a neural network that combines multi-scale morphological features with boundary features, enabling precise segmentation of invasive carcinoma regions without the need for additional H&E and P63 staining slides. In addition, we introduced an advanced semi-supervised learning algorithm, allowing efficient training of the model using unlabeled data. To evaluate the effectiveness of our approach, we constructed a dataset consisting of 618 IHC-stained WSIs from 170 cases, including four types of staining (ER, PR, HER2, and Ki-67). Notably, the model demonstrated an impressive intersection over union (IoU) score exceeding 80% on the test set. Furthermore, to ascertain the practical utility of our model in IHC quantitative evaluation, we constructed a fully automated Ki-67 scoring system based on the model's predictions. Comparative experiments convincingly demonstrated that our system exhibited high consistency with the scores given by experienced pathologists. CONCLUSIONS: Our developed model excels in accurately distinguishing between DCIS and invasive carcinoma regions in breast cancer immunohistochemistry WSIs. This method paves the way for a clinically available, fully automated immunohistochemistry quantitative scoring system.
RESUMO
Primary lymphoma in soft tissue is very rare. In order to understand the clinicopathological features of primary lymphoma in soft tissue, we found 13 cases (0.3%) of primary lymphoma in soft tissue by reviewing 4303 lymphomas diagnosed in our institution from 2010 to 2019. Tumors were found in the following sites: 8 in lower extremity (2 in leg, 1 in calf, 1 in knee and 4 in buttock), 1 in upper extremity (left shoulder) and 4 in the trunk (3 in waist and 1 in thoracolumbar). The most common histologic type was DLBCL (7/13, 54.8%). 6 cases of which had follow-up information. 25 patients were also selected by screening the English literature search (from Jan 2010 to December 2019) including 1102 studies. Compared to the results of literature review, our results are similar with them. The tumor sites were as follows: 10 in lower extremity, 4 in upper extremity, 9 in the trunk and 2 in masticatory muscle. The most common histological type was also DLBCL (n=11/25, 44%). Overall survival analysis of all 31 patients including our 6 cases with primary lymphoma in soft tissue showed no significant difference between different histological type (Log Rank P=0.120, Breslow P=0.157). The differential diagnosis includes malignant melanoma, rhabdomyosarcoma and metastatic carcinoma in soft tissue.
RESUMO
Tumor segmentation is a fundamental task in histopathological image analysis. Creating accurate pixel-wise annotations for such segmentation tasks in a fully-supervised training framework requires significant effort. To reduce the burden of manual annotation, we propose a novel weakly supervised segmentation framework based on sparse patch annotation, i.e., only small portions of patches in an image are labeled as 'tumor' or 'normal'. The framework consists of a patch-wise segmentation model called PSeger, and an innovative semi-supervised algorithm. PSeger has two branches for patch classification and image classification, respectively. This two-branch structure enables the model to learn more general features and thus reduce the risk of overfitting when learning sparsely annotated data. We incorporate the idea of consistency learning and self-training into the semi-supervised training strategy to take advantage of the unlabeled images. Trained on the BCSS dataset with only 25% of the images labeled (five patches for each labeled image), our proposed method achieved competitive performance compared to the fully supervised pixel-wise segmentation models. Experiments demonstrate that the proposed solution has the potential to reduce the burden of labeling histopathological images.
Assuntos
Neoplasias , Aprendizado de Máquina Supervisionado , Algoritmos , Humanos , Processamento de Imagem Assistida por Computador/métodos , Neoplasias/diagnóstico por imagemRESUMO
Compared with the traditional single therapy, nanomedicine has promoted a multimodal combination treatment for various carcinomas, especially the development of corresponding intelligent multifunctional biomaterials based on advanced DNA nanotechnology has great potential in cancer combination therapy. Herein, we describe a strategy to "backpack" aptamer PL1, which specifically binds to PD-L1 and Pcsk9 siRNA on well-defined DNA tetrahedral nanoparticles (TDNs) via DNA hybridization, which collectively contributes to the effective therapy for colorectal cancer (CRC). In addition, we designed a targeted TDN upon folic acid (FA) recognition, limiting its release to the sites of tumors where folic acid receptor (FAR) is encountered. Our results demonstrated that the TDN-FA/PL1/Pcsk9-siRNA could free immune cells to target CRC cells and attenuate 83.48% tumor growth in mouse models of CT26 CRC. Mechanically, the cancer-targeting FA guided TDN-FA/PL1/Pcsk9-siRNA into tumor cells, thereby ensuring that the aptamer PL1 could choke the mutual effects between PD-1 and PD-L1, followed by a 1.69-fold increase in T cell number and a 1.9-fold suppression of T cell activity by the PD-1/PD-L1 pathway, while Pcsk9 siRNA decreased Pcsk9 expression averagely to the extent of 65.13% and then facilitated intratumoral infiltration of cytotoxic T cells robustly with IFN-γ and Granzyme B expression. Our results reveal that the multifunctional TND-FA/PL1/Pcsk9-siRNA is effective and safe for CRC therapy, thereby expanding the application of DNA nanotechnology for innovative therapies of various cancers.
Assuntos
Antígeno B7-H1 , Neoplasias Colorretais , Animais , Antígeno B7-H1/metabolismo , Linhagem Celular Tumoral , Neoplasias Colorretais/tratamento farmacológico , Neoplasias Colorretais/genética , DNA , Ácido Fólico , Camundongos , Oligonucleotídeos , Receptor de Morte Celular Programada 1 , Pró-Proteína Convertase 9/genética , RNA Interferente Pequeno/genéticaRESUMO
We collected 61 craniopharyngioma (CP) specimens to investigate the expression of TrkA, ß-catenin, BRAF gene mutation, and NTRK1 fusion in CP. There were 37 male and 24 female individuals with a median age of 34 years (range, 4-75 years). Histologically, there were 46 cases of adamantinomatous craniopharyngioma (ACP), 14 cases of papillary craniopharyngioma (PCP), and 1 case with a mixed adamantinomatous and papillary pattern. By immunohistochemistry, we found that moderate/high TrkA expression was detected in 47% (28/60) CP and was significantly higher in adult patients (p = 0.018). Interestingly, TrkA is more expressed in "whorled epithelium" cells in ACP, similar to the localization of abnormal ß-catenin. The abnormal expression rate of ß-catenin was 70% (43/61), and the medium/high cyclin D1 expression rate was 73% (44/60), both of which were significantly higher in ACP than in PCP. Of the CP, 41% (21/51) had a moderate/strong P16-positive signal; 58% (34/59) showed a high Ki-67 expression, and there was a significant correlation between high Ki-67 L.I. and high tumor recurrence (p = 0.021). NTRK1 fusion was not found in CP by fluorescence in situ hybridization (FISH). By PCR, 26% (15/58) CP showed BRAF V600E gene mutation, which mainly occurred in PCP (100%, 14/14) except one case of mixed CP. Moreover, TrkA expression was negatively correlated with Ki-67 index and positively correlated with P16 expression. There was a significantly negative correlation between BRAF V600E mutation and abnormal ß-catenin expression. Our results demonstrate for the first time that TrkA expression might occur in CP, especially in adult CP patients, and suggest that cyclin D1 could be used for ACP histological classification in addition to ß-catenin and BRAF V600E mutation, while Ki-67 could be used as a marker to predict CP recurrence.
Assuntos
Craniofaringioma , Neoplasias Hipofisárias , Proteínas Proto-Oncogênicas B-raf , Receptor trkA , beta Catenina , Adolescente , Adulto , Idoso , Ciclo Celular , Criança , Pré-Escolar , Craniofaringioma/genética , Ciclina D1/genética , Feminino , Humanos , Hibridização in Situ Fluorescente , Antígeno Ki-67/genética , Masculino , Pessoa de Meia-Idade , Mutação , Neoplasias Hipofisárias/diagnóstico , Neoplasias Hipofisárias/genética , Neoplasias Hipofisárias/metabolismo , Prognóstico , Proteínas Proto-Oncogênicas B-raf/genética , Receptor trkA/genética , Adulto Jovem , beta Catenina/genéticaRESUMO
AIMS: Vascular tumors are composed of benign, intermediate, and malignant lesions. The diagnosis is challenging because some entities demonstrate overlapping morphologies and harbor the same genetic alterations. We describe herein a cohort of vascular tumors with clinicopathologic, immunohistochemical, and molecular features. METHODS AND RESULTS: 118 vascular tumors including 56 angiosarcomas, 18 epithelioid haemangioendotheliomas (EHE), 25 epithelioid haemangiomas (EH), 8 pseudomyogenic haemangioendotheliomas (PHE), 1 papillary intralymphatic angioendothelioma (PILA), 2 kaposiform haemangioendotheliomas (KHE), 3 Kaposi sarcomas, 2 retiform haemangioendotheliomas (RHE), and 3 anastomosing haemangiomas were assessed. FOSB, c-Fos, CAMTA1, and TFE3 expression and gene rearrangements were analyzed by immunohistochemical staining and FISH, respectively. Our results showed that FOSB expression was diffusely positive in all 8 PHEs, focally or sparsely in 12 EHs, and in 2 angiosarcomas. C-FOS expression was sparsely to diffusely positive in 15 EHs, focally or sparsely in 17 angiosarcomas, 1 EHE, 1 Kaposi sarcoma, and 1 PHE. CAMTA1 expression was positive in only 12 EHEs. TFE3 expression was focally or sparsely positive in all 8 PHEs, 22 angiosarcomas, 6 EHEs, 3 EHs, 2 Kaposi sarcomas, and 2 AHs. FOSB rearrangement was found in 5 PHEs, FOS rearrangement only in 1 EH, CAMTA1 rearrangement in 4 EHEs. CONCLUSIONS: FOSB and CAMTA1 are useful diagnostic markers for PHE and EHE, respectively. FOSB and FOS fusion represent a subset of epithelioid haemangioma. TFE3 is not a diagnostically meaningful marker in a majority of vascular tumors. The combined utility of these markers will facilitate the differential diagnosis in vascular tumors with morphologic overlap.
RESUMO
BACKGROUND: The expression of thymidylate synthase (TYMS) is significantly up-regulated in various cancers and associated with the poor prognosis of patients. However, the role of TYMS in the progression of colorectal cancer (CRC) is unclear. METHODS: Cell function assay, biology information analysis, and RNA sequencing were used to investigate the role of TYMS in the progression of CRC and underlining molecular mechanism. SPSS22.0 statistical software and GraphPad Prism 5 (Graphpad software) were used for statistical analysis. RESULTS: Our results showed that TYMS expression was higher in CRC tissues than that in non-tumor colorectal mucosa tissues. TYMS knockdown inhibited the proliferation, migration and invasion of HCT116 and HT29 cells, and the spheroid formation of HCT116 cells. The underling mechanism demonstrated that TYMS promoted the progression of CRC by regulating EMT-related proteins including E-cadherin, Vimentin, MMP-9 and stem cell biomarkers including CD133 and CD44. Furthermore, DEG sequencing showed that TYMS knockdown enriched the pathways of metastasis and metabolism by GO and KEGG analysis. We identified TM4SF4 was the downstream target of TYMS in CRC cells. TM4SF4 overexpression increased migration and invasion of CRC cells by regulating EMT and CD133 expression. CONCLUSIONS: Our findings suggest that TYMS-TM4SF4 axis may promote the progression of CRC by EMT and upregulating stem cell markers.
RESUMO
BACKGROUND: Programmed Cell Death 2 Like (PDCD2L) correlates with cell proliferation, apoptosis and mouse embryonic development. However, the role of PDCD2L in human cancers is unclear. METHODS: Multiple bioinformatic methods, in vitro function experiments and validation were performed to clarify the oncogenic role of PDCD2L in human cancers. RESULTS: Our study found that PDCD2L was aberrantly expressed in multiple types of human cancers, and associated with clinical stage and molecular subtype. Furthermore, overexpression of PDCD2L predicted poor overall survival in adrenocortical carcinoma(ACC), kidney chromophobe(KICH), acute myeloid leukemia(LAML), brain lower grade glioma(LGG),liver hepatocellular carcinoma(LIHC), mesothelioma(MESO), uveal melanoma(UVM) and poor diseases free survival in ACC, bladder urothelial carcinoma(BLCA), cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC), kidney renal clear cell carcinoma(KIRC), kidney renal papillary cell carcinoma(KIRP), LGG, LIHC, and UVM. PDCD2L expression was negatively associated with cancer associated fibroblast in breast invasive carcinoma (BRCA), lung squamous cell carcinoma (LUSC), sarcoma (SARC), stomach adenocarcinoma (STAD) and testicular germ cell tumors (TGCT). Mechanically, we found that PDCD2L expression was associated with apoptosis, invasion and cell cycle by investigating single cell sequencing data. For further validation, PDCD2Lwas highly expressed in colorectal cancer (CRC) cell lines and tissue samples compared with the normal colon cell line and non-tumor adjacent colorectal mucosa tissues. PDCD2L knockdown induced the apoptosis and proliferation of CRC cells. CONCLUSIONS: Our study shows that the oncogenic role of PDCD2L in various cancers and PDCD2L could be served as a biomarker of CRC.
RESUMO
Objective. To develop an artificial intelligence method predicting lymph node metastasis (LNM) for patients with colorectal cancer (CRC). Impact Statement. A novel interpretable multimodal AI-based method to predict LNM for CRC patients by integrating information of pathological images and serum tumor-specific biomarkers. Introduction. Preoperative diagnosis of LNM is essential in treatment planning for CRC patients. Existing radiology imaging and genomic tests approaches are either unreliable or too costly. Methods. A total of 1338 patients were recruited, where 1128 patients from one centre were included as the discovery cohort and 210 patients from other two centres were involved as the external validation cohort. We developed a Multimodal Multiple Instance Learning (MMIL) model to learn latent features from pathological images and then jointly integrated the clinical biomarker features for predicting LNM status. The heatmaps of the obtained MMIL model were generated for model interpretation. Results. The MMIL model outperformed preoperative radiology-imaging diagnosis and yielded high area under the curve (AUCs) of 0.926, 0.878, 0.809, and 0.857 for patients with stage T1, T2, T3, and T4 CRC, on the discovery cohort. On the external cohort, it obtained AUCs of 0.855, 0.832, 0.691, and 0.792, respectively (T1-T4), which indicates its prediction accuracy and potential adaptability among multiple centres. Conclusion. The MMIL model showed the potential in the early diagnosis of LNM by referring to pathological images and tumor-specific biomarkers, which is easily accessed in different institutes. We revealed the histomorphologic features determining the LNM prediction indicating the model ability to learn informative latent features.
RESUMO
Spred1 is highly expressed in normal hematopoietic stem cells (HSCs). Lack of Spred1 function has been associated with aberrant hematopoiesis and acute leukemias. In chronic myelogenous leukemia (CML), Spred1 is reduced in patients with accelerated phase (AP) or blast crisis (BC) CML, thereby suggesting that deficit of this protein may contribute to disease transformation. In fact, Spred1 knockout (KO) in SCLtTA/BCR-ABL CML mice either globally, or restricted to hematopoietic cells (i.e., HSCs) or to endothelial cells (ECs), led to transformation of chronic phase (CP) CML into AP/BC CML. Upon BCR-ABL induction, all three Spred1 KO CML models showed AP/BC features. However, compared with global Spred1 KO, the AP/BC phenotypes of HSC-Spred1 KO and EC-Spred1 KO CML models were attenuated, suggesting a concurrent contribution of Spred1 deficit in multiple compartments of the leukemic bone marrow niche to the CML transformation. Spred1 KO, regardless if occurred in HSCs or in ECs, increased miR-126 in LSKs (Lin-Sca-1+c-Kit+), a population enriched in leukemic stem cells (LSCs), resulting in expansion of LSCs, likely through hyperactivation of the MAPK/ERK pathway that augmented Bcl-2 expression and stability. This ultimately led to enhancement of Bcl-2-dependent oxidative phosphorylation that supported homeostasis, survival and activity of LSCs and drove AP/BC transformation.
Assuntos
Proteínas Adaptadoras de Transdução de Sinal/metabolismo , Proteínas Adaptadoras de Transdução de Sinal/fisiologia , Transformação Celular Neoplásica/patologia , Resistencia a Medicamentos Antineoplásicos , Células-Tronco Hematopoéticas/patologia , Leucemia Mielogênica Crônica BCR-ABL Positiva/patologia , Células-Tronco Neoplásicas/patologia , Proteínas Adaptadoras de Transdução de Sinal/genética , Animais , Transformação Celular Neoplásica/efeitos dos fármacos , Transformação Celular Neoplásica/metabolismo , Regulação Leucêmica da Expressão Gênica , Células-Tronco Hematopoéticas/efeitos dos fármacos , Células-Tronco Hematopoéticas/metabolismo , Humanos , Leucemia Mielogênica Crônica BCR-ABL Positiva/tratamento farmacológico , Leucemia Mielogênica Crônica BCR-ABL Positiva/genética , Leucemia Mielogênica Crônica BCR-ABL Positiva/metabolismo , Camundongos , Camundongos Endogâmicos C57BL , Camundongos Knockout , Células-Tronco Neoplásicas/efeitos dos fármacos , Células-Tronco Neoplásicas/metabolismo , Inibidores de Proteínas Quinases/farmacologiaRESUMO
Esophageal squamous cell carcinoma (ESCC) is a common malignant tumor with high mortality and poor prognosis. Ferroptosis is a newly discovered form of cell death induced by iron-catalyzed excessive peroxidation of polyunsaturated fatty acids (PUFAs). However, the prognostic value of ferroptosis-related genes (FRGs) for ESCC remains unclear. Based on the ESCC dataset from the Gene Expression Omnibus (GEO) database, we identified 39 prognostic FRGs through univariate Cox regression analysis. After LASSO regression and multivariate Cox regression analyses, a multigene signature based on 10 prognostic FRGs was constructed and successfully divided ESCC patients into two risk groups. Patients in the low-risk group showed a significantly better prognosis than patients in the high-risk group. In addition, we combined the risk score with clinical predictors to construct a nomogram for ESCC. The predictive ability of the nomogram was further verified by ROC curves and calibration plots in both the training and validation sets. The predictive power of the nomogram was demonstrated to be better than that of either the risk score or clinical variable alone. Furthermore, functional analysis revealed that the 10-FRG signature was mainly associated with ferroptosis, differentiation and immune response. Connectivity map analysis identified potential compounds capable of targeting FRGs in ESCC. Finally, we demonstrated the prognostic value of SRC gene in ESCC using the clinical samples and found that SRC inhibition sensitized ESCC cells to ferroptosis inducers by in vitro experiments. In conclusion, we identified and verified a 10-FRG prognostic signature and a nomogram, which provide individualized prognosis prediction and provide insight into potential therapeutic targets for ESCC.
RESUMO
BACKGROUND: During acute myeloid leukemia (AML) growth, the bone marrow (BM) niche acquires significant vascular changes that can be offset by therapeutic blast cytoreduction. The molecular mechanisms of this vascular plasticity remain to be fully elucidated. Herein, we report on the changes that occur in the vascular compartment of the FLT3-ITD+ AML BM niche pre and post treatment and their impact on leukemic stem cells (LSCs). METHODS: BM vasculature was evaluated in FLT3-ITD+ AML models (MllPTD/WT/Flt3ITD/ITD mouse and patient-derived xenograft) by 3D confocal imaging of long bones, calvarium vascular permeability assays, and flow cytometry analysis. Cytokine levels were measured by Luminex assay and miR-126 levels evaluated by Q-RT-PCR and miRNA staining. Wild-type (wt) and MllPTD/WT/Flt3ITD/ITD mice with endothelial cell (EC) miR-126 knockout or overexpression served as controls. The impact of treatment-induced BM vascular changes on LSC activity was evaluated by secondary transplantation of BM cells after administration of tyrosine kinase inhibitors (TKIs) to MllPTD/WT/Flt3ITD/ITD mice with/without either EC miR-126 KO or co-treatment with tumor necrosis factor alpha (TNFα) or anti-miR-126 miRisten. RESULTS: In the normal BM niche, CD31+Sca-1high ECs lining arterioles have miR-126 levels higher than CD31+Sca-1low ECs lining sinusoids. We noted that during FLT3-ITD+ AML growth, the BM niche lost arterioles and gained sinusoids. These changes were mediated by TNFα, a cytokine produced by AML blasts, which induced EC miR-126 downregulation and caused depletion of CD31+Sca-1high ECs and gain in CD31+Sca-1low ECs. Loss of miR-126high ECs led to a decreased EC miR-126 supply to LSCs, which then entered the cell cycle and promoted leukemia growth. Accordingly, antileukemic treatment with TKI decreased the BM blast-produced TNFα and increased miR-126high ECs and the EC miR-126 supply to LSCs. High miR-126 levels safeguarded LSCs, as shown by more severe disease in secondary transplanted mice. Conversely, EC miR-126 deprivation via genetic or pharmacological EC miR-126 knock-down prevented treatment-induced BM miR-126high EC expansion and in turn LSC protection. CONCLUSIONS: Treatment-induced CD31+Sca-1high EC re-vascularization of the leukemic BM niche may represent a LSC extrinsic mechanism of treatment resistance that can be overcome with therapeutic EC miR-126 deprivation.
Assuntos
Medula Óssea/patologia , Regulação Leucêmica da Expressão Gênica , Leucemia Mieloide Aguda/genética , MicroRNAs/genética , Células-Tronco Neoplásicas/patologia , Animais , Medula Óssea/irrigação sanguínea , Humanos , Leucemia Mieloide Aguda/patologia , Leucemia Mieloide Aguda/terapia , Camundongos , Camundongos Endogâmicos C57BL , Regulação para Cima , Tirosina Quinase 3 Semelhante a fms/genéticaRESUMO
SLC2A3 is a membrane transporter that belongs to the solute carrier family, whose function includes transmembrane transport and glucose transmembrane transport activity. To clarify the expression and role of SLC2A3 in colorectal cancer (CRC), we analyzed the TCGA and GEO databases and found that SLC2A3 mRNA levels were significantly higher in CRC tissues than that in adjacent non-tumor tissues. Furthermore, high expression of SLC2A3 predicted poor overall survival and disease free survival for CRC patients. For validation, we collected 174 CRC samples and found that SLC2A3 expression was higher in CRC tissues than that in adjacent non-tumor colorectal mucosa tissues by immunohistochemistry staining. Further study showed that high expression of SLC2A3 was enriched in epithelial-mesenchymal transition (EMT) classical pathway, interferon-γ pathway by GSEA analysis enrichment, indicating that SLC2A3 may play a key role in the progression of CRC through EMT and immune response, which also has been validated by the global gene expression profiling of human CRC cell lines. The expression of SLC2A3 was positively correlated with CD4 and CD8+T cells by using TIMER and EPIC algorithm, respectively. SLC2A3 knockdown suppressed migration and inhibited the expression of Vimentin and MMP9 in CRC cell line SW480 and RKO. Meanwhile, PD-L1 expression was also significantly attenuated in SW480 and RKO cells transfected with SLC2A3 siRNA. The result suggests that SLC2A3 may be involved in the immune response of CRC by regulating PD-L1 immune checkpoint. In our series, SLC2A3 and PD-L1 positive expression was 74% (128/174) and 22% (39/174) of CRC, respectively. SLC2A3 expression was significantly associated with perineural invasion in CRC patients. In conclusion, SLC2A3 may play an important role in progression of CRC by regulating EMT and PD-L1 mediated immune responses.
RESUMO
BACKGROUND: Gastrointestinal stromal tumor (GIST) can arise anyplace along the gastrointestinal (GI) tract. The uncommon tumor location in groin area is rarely reported. CASE PRESENTATION: We herein reported a metastasized case presented as GI hemorrhage complicated with indirect hernia, and underwent tumor cytoreduction, herniorrhaphy and chemotherapy for jejunal GIST. The case was described consecutively based on the process of surgical management, with a good follow-up result. A literature review by searching similar case reports from two national medical databases was performed to summarize clinical features of such unusual presentation of GIST, which included hernia characteristics, short- and long-term outcomes of this disease. It showed GIST presenting as groin hernia was rarely reported and all available 11 cases suggested a primary tumor and required both tumor resection and hernia repair. The long-term results indicated 64.3% overall survival at 5 years after the incidental diagnosis. CONCLUSIONS: Inguinal hernia is an extremely rare presentation of GIST, with limited case reports available in the literature. A radical involving tumor resection plus hernia repair is an optimal surgical approach for such uncommon condition. An adjuvant medication mounting on mutated KIT gene should be strictly followed for high risk cases.