Epithelial-mesenchymal transition biomarkers and support vector machine guided model in preoperatively predicting regional lymph node metastasis for rectal cancer

Progress in the diagnosis of lymph node metastasis in rectal cancer: a review

Historically, the chief focus of lymph node metastasis research has been molecular and clinical studies of a few essential pathways and genes. Recent years have seen a rapid accumulation of massive omics and imaging data catalyzed by the rapid development of advanced technologies. This rapid increase in data has driven improvements in the accuracy of diagnosis of lymph node metastasis, and its analysis further demands new methods and the opportunity to provide novel insights for basic research. In fact, the combination of omics data, imaging data, clinical medicine, and diagnostic methods has led to notable advances in our basic understanding and transformation of lymph node metastases in rectal cancer. Higher levels of integration will require a concerted effort among data scientists and clinicians. Herein, we review the current state and future challenges to advance the diagnosis of lymph node metastases in rectal cancer.

Predicting prognosis for patients with ESCC before surgery by SVMs ranking with nomogram analyses

OBJECTIVE

A SVM predictive model consisting of preoperative tumor markers and inflammatory factors was established to explore its significance in evaluating the prognosis of patients with ESCC.

METHODS

Clinical data of 311 patients with ESCC who underwent surgery were collected and followed up until October 2019. Statistical software SPSS version 22.0, and R (version 3.6.1) were used to analyze the data.

RESULTS

In the Test, Val1 and Val2 groups, the sensitivity of preoperative optimal combination (SVM5) to predict the prognosis of patients with ESCC was 88.89%, 76.92%, and 73.68%, respectively. The specificity was 92.00%, 74.42%, and 78.00%, respectively. The sensitivity and specificity were not statistically different from those of SVM9 (P > 0.05), while the sensitivity of SVM9+5 for predicting the prognosis of patients with ESCC was 91.84%, 82.26%, and 80.36%, respectively. The specificity was 97.44%, 75.93%, and 78.00%, respectively. Its sensitivity and specificity were higher than those of SVM9 (P < 0.001).

CONCLUSIONS

We used a nomogram to input the indicators in the SVM5 into the artificial intelligence program for patients with ESCC who have not yet developed an individualized plan. It can predict and evaluate the postoperative outcome of patients with ESCC with a sensitivity of 79.04%, specificity of 81.82%, PPV of 83.54%, NPV of 76.97%, and accuracy of 80.32%. For patients who have undergone surgery, we can enter the indicators in SVM9+5 into the artificial intelligence program.

Preoperative Prediction of Lymph Node Metastasis in Colorectal Cancer with Deep Learning

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.

The influencing factors of Harbin (China) residents' satisfaction with municipal solid waste treatment

China is experiencing an enormous increase in municipal household solid waste (MHSW) generation and is facing multiple problems associated with the treatment of MHSW. This paper analyses factors affecting residents' satisfaction with MHSW treatment performance. Six factors were identified by the Delphi method: (a) pick-up frequency by waste collection vehicles, (b) fund supply situation, (c) charging standard for waste treatment, (d) waste bin arrangement, (e) laws and regulations, (f) publicity and education. We examine the significance of these six influencing factors, estimating binary logistic regression models. Data for this study are derived from the survey responses of 469 households in Harbin, one of the largest cities in northeast China. The results indicate that 'pick-up frequency by waste collection vehicles' is ranked the first and most important determinant of Harbin residents' satisfaction with MHSW treatment; this is closely followed by 'publicity and education'. The third and fourth significant influencing factors, respectively, are 'fund supply situation' and 'charging standard for waste treatment'. The last two factors are 'laws and regulations' and 'waste bin arrangement'. By understanding the influence of various factors on residents' satisfaction, this study aims to help in designing an effective waste management system to reduce the cost of MHSW management, and to raise the residents' satisfaction with municipal solid waste treatment. Based on the research findings, we advocate that establishing a reasonable waste transport (pick-up) system as well as strengthening publicity and education of waste management are key to improving residents' satisfaction with the MHSW treatment performance.

Distant Metastasis Risk Definition by Tumor Biomarkers Integrated Nomogram Approach for Locally Advanced Nasopharyngeal Carcinoma

Identifying metastasis remains a challenge for death control and tailored therapy for nasopharyngeal carcinoma (NPC). Here, we addressed this by designing a nomogram-based Cox proportional regression model through integrating a panel of tumor biomarkers. A total of 147 locally patients with advanced NPC, derived from a randomized phase III clinical trial, were enrolled. We constructed the model by selecting the variables from 31 tumor biomarkers, including 6 pathological signaling pathway molecules and 3 Epstein-Barr virus-related serological variables. Through the least absolute shrinkage and selection operator (LASSO) Cox proportional regression analysis, a nomogram was designed to refine the metastasis risk of each NPC individuals. Using the LASSO Cox regression model, we constructed a 9 biomarkers-based prognostic nomogram: Beclin 1, Aurora-A, Cyclin D1, Ki-67, P27, Bcl-2, MMP-9, 14-3-3σ, and VCA-IgA. The time-dependence receiver operating characteristic analysis at 1, 3, and 5 years showed an appealing prognostic accuracy with the area under the curve of 0.830, 0.827, and 0.817, respectively. In the validation subset, the concordance index of this nomogram reached to 0.64 to identify the individual metastasis pattern. Supporting by this nomogram algorithm, the individual metastasis risk might be refined personally and potentially guiding the treatment decisions and target therapy against the related signaling pathways for patients with locally advanced NPC.

Molecular Decision Tree Algorithms Predict Individual Recurrence Pattern for Locally Advanced Nasopharyngeal Carcinoma

Background: Recurrence remains one of the key reasons of relapse after the radical radiation for locally advanced nasopharyngeal carcinoma (NPC). Here, the multiple molecular and clinical variables integrated decision tree algorithms were designed to predict individual recurrence patterns (with VS without recurrence) for locally advanced NPC.

Methods: A total of 136 locally advanced NPC patients retrieved from a randomized controlled phase III trial, were included. For each patient, the expression levels of 33 clinicopathological biomarkers in tumor specimen, 3 Epstein-Barr virus related serological antibody titer and 5 clinicopathological variables, were detected and collected to construct the decision tree algorithm. The expression level of 33 clinicopathological biomarkers in tumor specimen was evaluated by immunohistochemistry staining.

Results: Three algorithm classifiers, augmented by the adaptive boosting algorithm for variable selection and classification, were designed to predict individual recurrence pattern. The classifiers were trained in the training subset and further tested using a 10-fold cross-validation scheme in the validation subset. In total, 13 molecules expression level in tumor specimen, including AKT1, Aurora-A, Bax, Bcl-2, N-Cadherin, CENP-H, HIF-1α, LMP-1, C-Met, MMP-2, MMP-9, Pontin and Stathmin, and N stage were selected to construct three 10-fold cross-validation decision tree classifiers. These classifiers showed high predictive sensitivity (87.2-93.3%), specificity (69.0-100.0%), and overall accuracy (84.5-95.2%) to predict recurrence pattern individually. Multivariate analyses confirmed the decision tree classifier was an independent prognostic factor to predict individual recurrence (algorithm 1: hazard ration (HR) 0.07, 95% confidence interval (CI) 0.03-0.16, P < 0.01; algorithm 2: HR 0.13, 95% CI 0.04-0.44, P < 0.01; algorithm 3: HR 0.13, 95% CI 0.03-0.68, P = 0.02).

Conclusion: Multiple molecular and clinicopathological variables integrated decision tree algorithms may individually predict the recurrence pattern for locally advanced NPC. This decision tree algorism provides a potential tool to select patients with high recurrence risk for intensive follow-up, and to diagnose recurrence at an earlier stage for salvage treatment in the NPC endemic region.

Machine Learning Approach to Predicting Stem Cell Donor Availability

The success of unrelated donor stem cell transplants depends on not only finding genetically matched donors, but also donor availability. On average 50% of potential donors in the National Marrow Donor Program database are unavailable for a variety of reasons, after initially matching a patient, with significant variations in availability among subgroups (eg, by race or age). Several studies have established univariate donor characteristics associated with availability. Individual consideration of each applicable characteristic is laborious. Extrapolating group averages to the individual-donor level tends to be highly inaccurate. In the current environment with enhanced donor data collection, we can make better estimates of individual donor availability. We propose a machine learning based approach to predict availability of every registered donor, and evaluate the predictive power on a test cohort of 44,544 requests to be .77 based on the area under the receiver-operating characteristic curve. We propose that this predictor should be used during donor selection to reduce the time to transplant.

Construction of a 26‑feature gene support vector machine classifier for smoking and non‑smoking lung adenocarcinoma sample classification

The present study aimed to identify the feature genes associated with smoking in lung adenocarcinoma (LAC) samples and explore the underlying mechanism. Three gene expression datasets of LAC samples were downloaded from the Gene Expression Omnibus database through pre-set criteria and the expression data were processed using meta-analysis. Differentially expressed genes (DEGs) between LAC samples of smokers and non-smokers were identified using limma package in R. The classification accuracy of selected DEGs were visualized using hierarchical clustering analysis in R language. A protein-protein interaction (PPI) network was constructed using gene interaction data from the Human Protein Reference Database for the DEGs. Betweenness centrality was calculated for each node in the network and genes with the greatest BC values were utilized for the construction of the support vector machine (SVM) classifier. The dataset GSE43458 was used as the training dataset for the construction and the other datasets (GSE12667 and GSE10072) were used as the validation datasets. The classification accuracy of the classifier was tested using sensitivity, specificity, positive predictive value, negative predictive value and area under curve parameters with the pROC package in R language. The feature genes in the SVM classifier were subjected to pathway enrichment analysis using Fisher's exact test. A total of 347 genes were identified to be differentially expressed between samples of smokers and non-smokers. The PPI network of DEGs were comprised of 202 nodes and 300 edges. An SVM classifier comprised of 26 feature genes was constructed to distinguish between different LAC samples, with prediction accuracies for the GSE43458, GSE12667 and GSE10072 datasets of 100, 100 and 94.83%, respectively. Furthermore, the 26 feature genes that were significantly enriched in 9 overrepresented biological pathways, including extracellular matrix-receptor interaction, proteoglycans in cancer, cell adhesion molecules, p53 signaling pathway, microRNAs in cancer and apoptosis, were identified to be smoking-related genes in LAC. In conclusion, an SVM classifier with a high prediction accuracy for smoking and non-smoking samples was obtained. The genes in the classifier may likely be the potential feature genes associated with the development of patients with LAC who smoke.

Molecular Biology: Are We Getting Any Closer to Providing Clinically Useful Information?

Advances in molecular biology and biomarker research have significantly impacted our understanding and treatment of multiple solid malignancies. In rectal cancer, where neoadjuvant chemoradiation is widely used for locally advanced disease, most efforts have focused on the identification of predictors of response in an attempt to appropriately select patients for multimodality therapy. A variety of biomarkers have been studied, including genetic mutations, chromosomal copy number alterations, and single as well as multigene expression patterns. Also, as transanal resection of rectal tumors requires accurate preoperative detection of lymph node metastasis, the identification of biomarkers of regional nodal involvement has been another important field of active research. While preliminary results have been promising, lack of external validation means has a limited translation to clinical use. This review summarizes recent developments in rectal cancer biomarker research, highlighting the challenges associated with their adoption, and evaluating their potential for clinical use.

Diffusion kurtosis imaging evaluating epithelial-mesenchymal transition in colorectal carcinoma xenografts model: a preliminary study

Epithelial-mesenchymal transition (EMT) plays an important role in aggravating invasiveness and metastatic behavior of colorectal cancer (CRC). Identification of EMT is important for structuring treatment strategy, but has not yet been studied by using noninvasive imaging modality. Diffusion kurtosis imaging (DKI) is an advanced diffusion weighted model that could reflect tissue microstructural changes in vivo. In this study, EMT was induced in CRC cells (HCT116) by overexpressing Snail1 gene. We aimed to investigate the value of DKI in identifying EMT in CRC and decipher the correlations between DKI-derived parameters and EMT biomarker E-cadherin and cell proliferative index Ki-67 expression. Our results revealed that HCT116/Snail1 cells presented changes consistent with EMT resulting in significant increase in migration and invasion capacities. DKI could identify CRC with EMT, in which the DKI-derived parameter diffusivity was significantly lower, and kurtosis was significantly higher than those in the CRC/Control. Diffusivity was negatively and kurtosis was positively correlated with Ki-67 expression, whereas diffusivity was positively and kurtosis was negatively correlated with E-cadherin expression. Therefore, our study concluded that DKI can identify EMT in CRC xenograft tumors. EMT-contained CRC tumors with high Ki-67 and low E-cadherin expression were vulnerable to have lower diffusivity and higher kurtosis coefficients.

Prognostic value of autophagy related proteins ULK1, Beclin 1, ATG3, ATG5, ATG7, ATG9, ATG10, ATG12, LC3B and p62/SQSTM1 in gastric cancer

Autophagy-related (ATG) genes contributed to tumorigenesis and cancer progression. This study aims to investigate the expression of ATG proteins and their clinicopathological significance in gastric cancer. Nine well-known ATG proteins, (ULK1, Beclin 1, ATG3, ATG5, ATG7, ATG9, ATG10, ATG12 and LC3B) and p62/SQSTM1, which represented key regulators that participated in whole autophagosomes stepwise processes, were detected in a large cohort of 352 primary gastric cancer patients. Among these 352 patients, 117 cases were randomly assigned to the training set to detect the clinicopathological value of ATG proteins, and another 235 patients were used as the testing set for further validation. Except for Beclin 1, ATG9 and ATG10, another six ATG proteins and p62/SQSTM1 were closely correlated with histological types for gastric cancer. Moreover, low expression of ULK1, Beclin 1 and ATG10 were associated with lymph node metastasis. In addition, down-regulation of ULK1, Beclin 1, ATG7 and ATG10, up-regulation of ATG12 correlated with advanced TNM stage. Importantly, multivariate cox analysis identified ULK1, Beclin 1, ATG3 and ATG10 as favorable independent prognostic factors for overall survival. Combination analysis of ULK1, Beclin 1, ATG3, ATG10 revealed the improved prognostic accuracy for gastric cancer. Our study showed that ATG proteins might serve as novel prognostic biomarkers in gastric cancer, and supply a new valuable insight into cancer treatment targeting autophagy for patients.

Cancers Screening in an Asymptomatic Population by Using Multiple Tumour Markers

BACKGROUND

Analytic measurement of serum tumour markers is one of commonly used methods for cancer risk management in certain areas of the world (e.g. Taiwan). Recently, cancer screening based on multiple serum tumour markers has been frequently discussed. However, the risk-benefit outcomes appear to be unfavourable for patients because of the low sensitivity and specificity. In this study, cancer screening models based on multiple serum tumour markers were designed using machine learning methods, namely support vector machine (SVM), k-nearest neighbour (KNN), and logistic regression, to improve the screening performance for multiple cancers in a large asymptomatic population.

METHODS

AFP, CEA, CA19-9, CYFRA21-1, and SCC were determined for 20 696 eligible individuals. PSA was measured in men and CA15-3 and CA125 in women. A variable selection process was applied to select robust variables from these serum tumour markers to design cancer detection models. The sensitivity, specificity, positive predictive value (PPV), negative predictive value, area under the curve, and Youden index of the models based on single tumour markers, combined test, and machine learning methods were compared. Moreover, relative risk reduction, absolute risk reduction (ARR), and absolute risk increase (ARI) were evaluated.

RESULTS

To design cancer detection models using machine learning methods, CYFRA21-1 and SCC were selected for women, and all tumour markers were selected for men. SVM and KNN models significantly outperformed the single tumour markers and the combined test for men. All 3 studied machine learning methods outperformed single tumour markers and the combined test for women. For either men or women, the ARRs were between 0.003-0.008; the ARIs were between 0.119-0.306.

CONCLUSION

Machine learning methods outperformed the combined test in analysing multiple tumour markers for cancer detection. However, cancer screening based solely on the application of multiple tumour markers remains unfavourable because of the inadequate PPV, ARR, and ARI, even when machine learning methods were incorporated into the analysis.

Screening of feature genes in distinguishing different types of breast cancer using support vector machine

Objective

To screen the feature genes in estrogen receptor-positive (ER+) breast cancer in comparison with estrogen receptor-negative (ER−) breast cancer.

Methods

Nine microarray data of ER+ and ER− breast cancer samples were collected from Gene Expression Omnibus database. After preprocessing, data in five training sets were analyzed using significance analysis of microarrays to screen the differentially expressed genes (DEGs). The DEGs were further analyzed via support vector machine (SVM) function in e1071 package of R to construct a SVM classifier, the efficacy of which was verified by four testing sets and its combination with training sets using a leave-one-out cross-validation. Feature genes obtained by SVM classifier were subjected to function- and pathway-enrichment via the Database for Annotation, Visualization and Integrated Discovery and KEGG Orthology Based Annotation System, respectively.

Results

A total of 526 DEGs were screened between ER+ and ER− breast cancer. The SVM classifier demonstrated that these genes could distinguish different subtype samples with high accuracy of larger than 90%, and also showed good sensitivity, specificity, positive/negative predictive value, and area under receiver operating characteristic curve. The inflammatory and hormone biological processes were the common enriched results for two different function analyses, indicating that the inflammatory (ie, IL8) and hormone regulation (ie, CGA) genes may be the involved feature genes to distinguish ER+ and ER− types of breast cancer.

Conclusion

The gene-expression profile data can provide feature genes to distinguish ER+ and ER− samples, and the identified genes can be used for biomarkers for ER+ samples.

High expression of Beclin-1 predicts favorable prognosis for patients with colorectal cancer

PURPOSE

Beclin-1 is an autophagy gene. It promotes the formation of the autophagic vesicle as well as plays an essential role in guarding the cells against chromosomal instability. Overexpression of Beclin-1 has been reported to predict a favorable survival in various cancers. However, little is known about its prognostic significance in colorectal cancer.

METHODS AND MATERIALS

A total of three hundred and sixty-three (363) colorectal tissues from colorectal cancer (CRC) patients were collected. Tissue micro-arrays and immunohistochemistry were used to investigate the expression and prognostic significance of Beclin-1 in CRC. The associations among Beclin-1 expression, clinicopathological parameters and prognosis were evaluated.

RESULTS

Beclin-1 had a higher expression in CRC tissues than in normal tissues. A high expression of Beclin-1 was positively correlated with gender (P=0.027), histological grade (P=0.003), pM status (P=0.003) and clinical stage (P=0.024). Patients with a high Beclin-1 expression, when compared to those with a lower expression had both a better overall survival (OS, P=0.006) and disease-free survival (DFS, P=0.008). In the pT3 subgroup, Beclin-1 was also found to be a good prognostic indicator (P<0.05). Multivariate analysis showed a high expression of Beclin-1 was indeed a positive independent prognostic factor of OS and DFS for CRC patients (P<0.05).

CONCLUSION

Our results demonstrated that a high expression of Beclin-1 correlated with a better overall survival and disease-free survival, thus serving as a favorable independent prognostic marker in CRC.

Pre-operative predictive factors for intra-operative pathological lymph node metastasis in rectal cancers

BACKGROUND

A number of clinicopathologic factors have been found to be associated with pathological lymph node metastasis (pLNM) in rectal cancer; however, most of them can only be identified by expensive high resolution imaging or obtained after surgical treatment. Just like the Child-Turcotte-Pugh (CTP) and the model for end-stage liver disease (MELD) scores which have been widely used in clinical practice, our study was designed to assess the pre-operative factors which could be obtained easily to predict intra-operative pLNM in rectal cancer.

METHODS

A cohort of 469 patients who were treated at our hospital in the period from January 2003 to June 2011, and with a pathologically hospital discharge diagnosis of rectal cancer, were included. Clinical, laboratory and pathologic parameters were analyzed. A multivariate unconditional logistic regression model, areas under the curve (AUC), the Kaplan-Meier method (log-rank test) and the Cox regression model were used.

RESULTS

Of the 469 patients, 231 were diagnosed with pLNM (49.3%). Four variables were associated with pLNM by multivariate logistic analysis, age<60 yr (OR=1.819; 95% CI, 1.231-2.687; P=0.003), presence of abdominal pain or discomfort (OR=1.637; 95% CI, 1.052-2.547; P=0.029), absence of allergic history (OR=1.879; 95% CI, 1.041-3.392; P=0.036), and direct bilirubin ≥ 2.60 μmol/L (OR=1.540; 95% CI, 1.054-2.250; P=0.026). The combination of all 4 variables had the highest sensitivity (98.7%) for diagnostic performance. In addition, age<60 yr and direct bilirubin ≥ 2.60 μmol/L were found to be associated with prognosis.

CONCLUSION

Age, abdominal pain or discomfort, allergic history and direct bilirubin were associated with pLNM, which may be helpful for preoperative selection.

Levels of human replication factor C4, a clamp loader, correlate with tumor progression and predict the prognosis for colorectal cancer

BACKGROUND

Human replication factor C4 (RFC4) is involved in DNA replication as a clamp loader and is aberrantly regulated across a range of cancers. The current study aimed to investigate the function of RFC4 in colorectal cancer (CRC).

METHODS

The mRNA levels of RFC4 were assessed in 30 paired primary CRC tissues and matched normal colonic tissues by quantitative PCR. The protein expression levels of RFC4 were evaluated by western blotting (n = 16) and immunohistochemistry (IHC; n = 49), respectively. Clinicopathological features and survival data were correlated with the expression of RFC4 by IHC analysis in a tissue microarray comprising 331 surgically resected CRC. The impact of RFC4 on cell proliferation and the cell cycle was assessed using CRC cell lines.

RESULTS

RFC4 expression was significantly increased in CRC specimens as compared to adjacent normal colonic tissues (P <0.05). High levels of RFC4, determined on a tissue microarray, were significantly associated with differentiation, an advanced stage by the Tumor-Node-Metastasis (TNM) staging system, and a poor prognosis, as compared to low levels of expression (P <0.05). However, in multivariate analysis, RFC4 was not an independent predictor of poor survival for CRC. In vitro studies, the loss of RFC4 suppressed CRC cell proliferation and induced S-phase cell cycle arrest.

CONCLUSION

RFC4 is frequently overexpressed in CRC, and is associated with tumor progression and worse survival outcome. This might be attributed to the regulation of CRC cell proliferation and cell cycle arrest by RFC4.

Phosphorylated p38, a negative prognostic biomarker, complements TNM staging prognostication in colorectal cancer

Phosphorylated p38 (p-p38) played a pivotal role in the regulation of disease progression and correlated with tumor prognosis. Here, we characterized the prognostic effect of p-p38 in colorectal cancer (CRC). Three hundred and sixteen CRC patients in stages I-III were recruited in this study. P-p38 expression was semi-quantitatively evaluated using tissue microarrays and immunohistochemistry staining. Overall survival (OS), disease-free survival (DFS), local failure-free survival (LFFS), and distant metastasis-free survival (DMFS) of patient subgroups, segregated by p-p38 expression level and clinical stage, were compared using Kaplan-Meier analysis. We found that p-p38 was overexpressed in 48.1 % (152/316) CRC tissues, whereas low or deficiently expressed in normal adjacent epithelia. Overexpression of p-p38 predicted poor OS (P < 0.001), DFS (P = 0.002), LFFS (P = 0.016), and DMFS (P = 0.025) in CRC. Importantly, patient subgroups in the early stage (stages I + II) and with low p-p38 had similar OS, PFS, LFFS, and DMFS probabilities to that of stage I, whereas those with high p-p38 were similar to stage III disease. In addition, for stage III disease, the subgroup with low p-p38 had a similar survival probability to that of stage I, whereas the subgroup with high p-p38 had the worst survival. Multivariate Cox analysis confirmed that p-p38 was indeed a significantly independent factor for death, recurrence, and distant metastases in CRC. Our results demonstrated that p-p38 was a negative independent prognostic factor for CRC. Complementing TNM staging with p-p38 might refine the risk definition more accurately for a subset of patients.

Beclin 1 deficiency correlated with lymph node metastasis, predicts a distinct outcome in intrahepatic and extrahepatic cholangiocarcinoma

Autophagy can be tumor suppressive as well as promotive in regulation of tumorigenesis and disease progression. Accordingly, the prognostic significance of autophagy key regulator Beclin 1 was varied among different tumors. Here, we detected the clinicopathological and prognostic effect of Beclin 1 in the subtypes of intrahepatic cholangiocarcinoma (ICC) and extrahepatic cholangiocarcinoma (ECC). Beclin 1 expression level was detected by immunohistochemistry staining in 106 ICC and 74 ECC patients. We found that Beclin 1 was lowly expressed in 126 (70%) cholangiocarcinoma patients, consist of 72 ICC and 54 ECC. Moreover, the cholangiocarcinoma patients with lymph node metastasis (N1) had a lower Beclin 1 level than that of N0 subgroup (P=0.012). However, we did not detect any correlations between Beclin 1 and other clinicopathological features, including tumor subtypes, vascular invasion, HBV infection, liver cirrhosis, cholecystolithiasis and TNM stage. Survival analysis showed that, compared with the high expression subset, Beclin 1 low expression was correlated with a poorer 3-year progression-free survival (PFS, 69.1% VS 46.8%, P=041) for cholangiocarcinoma. Importantly, our stratified univariate and multivariate analysis confirmed that Beclin 1 lowly expressed ICC had an inferior PFS as well as overall survival than ECC, particularly than that of Beclin 1 highly expressed ECC patients. Thus, our study demonstrated that Beclin 1low expression, correlated with lymph node metastasis, and might be a negative prognostic biomarker for cholangiocarcinoma. Combined Beclin 1 level with the anatomical location might lead to refined prognosis for the subtypes of ICC and ECC.