Radiotherapy for Patients With Resected Tumor Deposit-Positive Colorectal Cancer: A Surveillance, Epidemiology, and End Results-Based Population Study

MiR-378a-5p exerts a radiosensitizing effect on CRC through LRP8/β-catenin axis

BACKGROUND

MiRNAs are closely related to tumor radiosensitivity. MiR-378a-5p level is down-regulated in colorectal cancer (CRC). Therefore, this study intends to explore the role of miR-378a-5p in CRC, especially radiosensitivity.

METHODS

The expression of miR-378a-5p was analyzed in CRC samples. CRC cell lines were treated with different doses of X-rays. Bioinformatics analysis, dual-luciferase reporter assay and RT-qPCR were used to detect the expressions and binding relationship of miR-378a-5p and low-density lipoprotein receptor-related protein 8 (LRP8). MiR-378a-5p inhibitor or/and siLRP8 were transfected into CRC cells with or without irradiation. Subsequently, clonogenic assay, flow cytometry and in vivo experiments including tumorigenesis assay, immunohistochemistry, RT-qPCR and Western blot were performed to clarify the role of miR-378a-5p/LRP8 axis in the radiosensitivity of CRC.

RESULTS

The down-regulated expression of miR-378a-5p in CRC is related to histological differentiation and tumor-node-metastasis (TNM) stage. After irradiation, the survival fraction of CRC cells was decreased, while the apoptotic rate and the level of miR-378a-5p were increased. Restrained miR-378a-5p repressed apoptosis and apoptosis-related protein expressions, yet promoted the proliferation and the radioresistance of cells by regulating β-catenin in CRC cells. LRP8 was highly expressed in CRC, and targeted by miR-378a-5p. SiLRP8 improved radiosensitivity and reversed the effect of miR-378a-5p down-regulation on CRC cells. Overexpressed miR-378a-5p and irradiation enhanced the level of miR-378a-5p, yet suppressed the expressions of Ki67 and LRP8 as well as tumorigenesis.

CONCLUSION

MiR-378a-5p may exert a radiosensitizing effect on CRC through the LRP8/β-catenin axis, which may be a new therapeutic target for CRC radioresistance.

Colorectal cancer: a comprehensive review of carcinogenesis, diagnosis, and novel strategies for classified treatments

Colorectal cancer is the third most common and the second deadliest cancer worldwide. To date, colorectal cancer becomes one of the most important challenges of the health system in many countries. Since the clinical symptoms of this cancer appear in the final stages of the disease and there is a significant golden time between the formation of polyps and the onset of cancer, early diagnosis can play a significant role in reducing mortality. Today, in addition to colonoscopy, minimally invasive methods such as liquid biopsy have received much attention. The treatment of this complex disease has been mostly based on traditional treatments including surgery, radiotherapy, and chemotherapy; the high mortality rate indicates a lack of success for current treatment methods. Moreover, disease recurrence is another problem of traditional treatments. Recently, new approaches such as targeted therapy, immunotherapy, and nanomedicine have opened new doors for cancer treatment, some of which have already entered the market, and many methods have shown promising results in clinical trials. The success of immunotherapy in the treatment of refractory disease, the introduction of these methods into neoadjuvant therapy, and the successful results in tumor shrinkage without surgery have made immunotherapy a tough competitor for conventional treatments. It seems that the combination of those methods with such targeted therapies will go through promising changes in the future of colorectal cancer treatment.

Activation of M2 macrophage autophagy by rapamycin increases the radiosensitivity of colorectal cancer xenografts

BACKGROUND

Tumor-associated macrophages (TAMs) are intimately involved in cancer radiochemotherapy resistance. However, the mechanism by which macrophages affect radiosensitivity through autophagy remains unclear. The purpose of our study was to investigate how activating autophagy in type-II macrophages (M2) by using rapamycin (RAP) would affect the radiosensitivity of colorectal cancer (CRC) xenografts.

MATERIALS AND METHODS

A nude mouse CRC model was established by injecting LoVo CRC cells. After tumor formation, supernatant from M2 cells (autophagy-unactivated), autophagy-activated M2 cells, or autophagy-downregulated M2 cells was injected peritumorally. All tumor-bearing mice were irradiated with 8-Gy X-rays twice, and the radiosensitivity of CRC xenografts was analyzed in each group.

RESULTS

The mass, volume, and microvessel density (MVD) of tumors in the autophagy-unactivated M2 group significantly increased; however, supernatant from M2 cells that were autophagy-activated by rapamycin significantly decreased tumor weight, volume, and MVD compared with negative control. Combining bafilomycin A1 (BAF-A1) with RAP treatment restored the ability of the M2 supernatant to increase tumor mass, volume, and MVD. Immunohistochemical and Western blot results showed that compared with the negative control group, supernatant from M2 cells that were not activated by autophagy downregulated the expression of Livin and Survivin in tumor tissues; activation of M2 autophagy further downregulated the protein levels.

CONCLUSIONS

Therefore, autophagy-activated M2 supernatant can downregulate the expression of the antiapoptotic genes Livin and Survivin in CRC xenografts, improving the radiosensitivity of CRC by inducing apoptosis in combination with radiotherapy and inhibiting the growth of transplanted tumors.

Union With Recursive Feature Elimination: A Feature Selection Framework to Improve the Classification Performance of Multicategory Causes of Death in Colorectal Cancer

Despite the use of machine learning tools, it is challenging to properly model cause-specific deaths in colorectal cancer (CRC) patients and choose appropriate treatments. Here, we propose an interesting feature selection framework, namely union with recursive feature elimination (U-RFE), to select the union feature sets that are crucial in CRC progression-specific mortality using The Cancer Genome Atlas (TCGA) dataset. Based on the union feature sets, we compared the performance of 5 classification algorithms, including logistic regression (LR), support vector machines (SVM), random forest (RF), eXtreme gradient boosting (XGBoost), and Stacking, to identify the best model for classifying 4-category deaths. In the first stage of U-RFE, LR, SVM, and RF were used as base estimators to obtain subsets containing the same number of features but not exactly the same specific features. Union analysis of the subsets was then performed to determine the final union feature set, effectively combining the advantages of different algorithms. We found that the U-RFE framework could improve various models' performance. Stacking outperformed LR, SVM, RF, and XGBoost in most scenarios. When the target feature number of the RFE was set to 50 and the union feature set contained 298 deterministic features, the Stacking model achieved F1_weighted, Recall_weighted, Precision_weighted, Accuracy, and Matthews correlation coefficient of 0.851, 0.864, 0.854, 0.864, and 0.717, respectively. The performance of the minority categories was also significantly improved. Therefore, this recursive feature elimination-based approach of feature selection improves performances of classifying CRC deaths using clinical and omics data or those using other data with high feature redundancy and imbalance.

Identifying the long-term survival beneficiary of chemotherapy for stage N1c sigmoid colon cancer

Sigmoid colon cancer often has an unsatisfactory prognosis. This study explored the effect of tumor deposits (TDs) on survival, and whether their presence/absence influence individualized treatment. Data of postoperative patients with sigmoid colon cancer were extracted from the Surveillance, Epidemiology, and End Results database. Overall survival (OS) was calculated using the Kaplan-Meier method and prognostic factors were identified using Cox regression analysis and random forest (RF). The nomogram's discrimination performance was evaluated using a concordance index (C-index), integrated discrimination improvement (IDI), calibration curves, and decision-curve analysis. The N1c group showed a worse prognosis than the N0 group. For N1c patients, a combination of surgery and chemotherapy prolonged survival, compared to surgery alone; however, the chemotherapy-surgery combination did not affect the OS of patients younger than 70 years, in stage T1-2, and/or of black race. Multivariable analysis and RF presented Age, T stage, and N stage were the most important predictors for OS. The novel nomogram had superiority to the TNM staging system with improved C-index and IDI, as well as good consistency and higher clinical benefit. TDs are associated with poor survival from sigmoid colon cancer, and considering TDs can inform the formulation of individual treatment regimens. The nomogram shows satisfactory prediction ability for OS.

Multimetric feature selection for analyzing multicategory outcomes of colorectal cancer: random forest and multinomial logistic regression models

Colorectal cancer (CRC) is one of the most common cancers worldwide, and a leading cause of cancer deaths. Better classifying multicategory outcomes of CRC with clinical and omic data may help adjust treatment regimens based on individual's risk. Here, we selected the features that were useful for classifying four-category survival outcome of CRC using the clinical and transcriptomic data, or clinical, transcriptomic, microsatellite instability and selected oncogenic-driver data (all data) of TCGA. We also optimized multimetric feature selection to develop the best multinomial logistic regression (MLR) and random forest (RF) models that had the highest accuracy, precision, recall and F1 score, respectively. We identified 2073 differentially expressed genes of the TCGA RNASeq dataset. MLR overall outperformed RF in the multimetric feature selection. In both RF and MLR models, precision, recall and F1 score increased as the feature number increased and peaked at the feature number of 600-1000, while the models' accuracy remained stable. The best model was the MLR one with 825 features based on sum of squared coefficients using all data, and attained the best accuracy of 0.855, F1 of 0.738 and precision of 0.832, which were higher than those using clinical and transcriptomic data. The top-ranked features in the MLR model of the best performance using clinical and transcriptomic data were different from those using all data. However, pathologic staging, HBS1L, TSPYL4, and TP53TG3B were the overlapping top-20 ranked features in the best models using clinical and transcriptomic, or all data. Thus, we developed a multimetric feature-selection based MLR model that outperformed RF models in classifying four-category outcome of CRC patients. Interestingly, adding microsatellite instability and oncogenic-driver data to clinical and transcriptomic data improved models' performances. Precision and recall of tuned algorithms may change significantly as the feature number changes, but accuracy appears not sensitive to these changes.

Prognostic Nomogram for Rectal Cancer Patients With Tumor Deposits

Aim

Tumor deposits (TDs) are an aggressive hallmark of rectal cancer, but their prognostic value has not been addressed in current staging systems. This study aimed to construct and validate a prognostic nomogram for rectal cancer patients with TDs.

Methods

A total of 1,388 stage III–IV rectal cancer patients who underwent radical surgical resection from the Surveillance, Epidemiology, and End Results (SEER) database were retrospectively analyzed to identify the clinical value of TDs. TD-positive rectal cancer patients in the SEER database were used as the training set to construct a prognostic model, which was validated by Fujian Cancer Hospital. Three models were constructed to predict the prognosis of rectal cancer patients with TDs, including the least absolute shrinkage and selection operator regression (LASSO, model 1), backward stepwise regression (BSR, model 2), and LASSO followed by BSR (model 3). A nomogram was established among the three models.

Results

In the entire cohort, TD was also identified as an independent risk factor for overall survival (OS), even after adjusting for baseline factors, stage, other risk factors, treatments, and all the included variables in this study (all P < 0.05). Among patients with TDs, model 3 exhibited a higher C-index and area under the curves (AUCs) at 3, 4, and 5 years compared with the American Joint Committee on Cancer staging system both in the training and validation sets (all P < 0.05). The nomogram obtained from model 3 showed good consistency based on the calibration curves and excellent clinical applicability by the decision curve analysis curves. In addition, patients were divided into two subgroups with apparently different OS according to the current nomogram (both P < 0.05), and only patients in the high-risk subgroup were found to benefit from postoperative radiotherapy (P < 0.05).

Conclusion

We identified a novel nomogram that could not only predict the prognosis of rectal cancer patients with TDs but also provide reliable evidence for clinical decision-making.

A Six-microRNA Signature Nomogram for Preoperative Prediction of Tumor Deposits in Colorectal Cancer

Purpose

Tumor deposits (TDs) are acknowledged negative prognostic factors in colorectal cancer (CRC), and their pathogenesis remains a puzzle. This study aimed to construct and validate a nomogram available for preoperative TDs prediction in CRC patients.

Patients and Methods

Patients from the Surveillance, Epidemiology, and End Results (SEER) and the cancer genome atlas (TCGA) databases were randomly divided into training and validation sets according to the sample size ratio of 7:3. Univariate logistic regression was performed for identifying differentially expressed microRNAs between TDs and non-TDs. Nomograms for TDs prediction were developed from the multivariate logistic regression model with least absolute shrinkage and selection operator and were validated internally in terms of accuracy, calibration, and clinical utility. Based on the target genes, pathways tightly associated with TDs were selected using enrichment analysis.

Results

Six clinicopathologic factors and expressions of six microRNAs (miR-614, miR-1197, miR-4770, miR-3136, miR-3173, and miR-4636) differed significantly between TDs and non-TDs CRC patients from the SEER and TCGA training sets. We compared potential prediction discrimination between two nomograms: a clinicopathologic nomogram and a six-microRNA signature nomogram. The six-microRNA signature nomogram revealed better accuracy than the clinicopathologic one for TDs prediction (AUC values of 0.96 and 0.93 in the validation cohort). The calibration plots and decision curve analysis demonstrated that the six-microRNA signature nomogram had better validity and a greater prognostic benefit versus the clinicopathologic one for TDs prediction. Calcium signaling pathways were closely associated with roles of the six microRNAs in TDs of CRC patients.

Conclusion

The six-microRNA signature nomogram can be used as an efficient tool for preoperative TDs prediction in CRC patients.

Performance and efficiency of machine learning algorithms for analyzing rectangular biomedical data

Most biomedical datasets, including those of 'omics, population studies, and surveys, are rectangular in shape and have few missing data. Recently, their sample sizes have grown significantly. Rigorous analyses on these large datasets demand considerably more efficient and more accurate algorithms. Machine learning (ML) algorithms have been used to classify outcomes in biomedical datasets, including random forests (RF), decision tree (DT), artificial neural networks (ANN), and support vector machine (SVM). However, their performance and efficiency in classifying multi-category outcomes of rectangular data are poorly understood. Therefore, we compared these metrics among the 4 ML algorithms. As an example, we created a large rectangular dataset using the female breast cancers in the surveillance, epidemiology, and end results-18 database, which were diagnosed in 2004 and followed up until December 2016. The outcome was the five-category cause of death, namely alive, non-breast cancer, breast cancer, cardiovascular disease, and other cause. We analyzed the 54 dichotomized features from ~45,000 patients using MatLab (version 2018a) and the tenfold cross-validation approach. The accuracy in classifying five-category cause of death with DT, RF, ANN, and SVM was 69.21%, 70.23%, 70.16%, and 69.06%, respectively, which was higher than the accuracy of 68.12% with multinomial logistic regression. Based on the features' information entropy, we optimized dimension reduction (i.e., reduce the number of features in models). We found 32 or more features were required to maintain similar accuracy, while the running time decreased from 55.57 s for 54 features to 25.99 s for 32 features in RF, from 12.92 s to 10.48 s in ANN, and from 175.50 s to 67.81 s in SVM. In summary, we here show that RF, DT, ANN, and SVM had similar accuracy for classifying multi-category outcomes in this large rectangular dataset. Dimension reduction based on information gain will increase the model's efficiency while maintaining classification accuracy.

Prognostic value of tumor deposits in rectal cancer: A monocentric series of 505 patients

BACKGROUND AND OBJECTIVES

It has been suggested that tumor deposits (TDs) may have a worse prognosis in rectal cancer compared with colonic cancer. The aim of this study was to assess TDs prognosis in rectal cancer.

METHODS

Patients who underwent total mesorectum excision for rectal adenocarcinoma (2011-2016) were included. A case-matched analysis was performed to assess the accurate impact of TDs for each pN category after exclusion of synchronous metastasis.

RESULTS

A total of 505 patients were included. TDs were observed in 99 (19.6%) patients, (pN1c = 37 [7.3%]). TDs were associated with pT3-T4 stage (P = .037), synchronous metastasis (P = .003), lymph node (LN) invasion (P = .041), vascular invasion (P = .001), and perineural invasion (P < .001). TD was associated with a worse 3-year disease-free survival (DFS) among pN0 (51.2% vs 79.8%; P < .001); pN1 patients (35.2% vs 70.1%; P = .004) but not among pN2 patients (37.5% vs 44.7%; P = .499). After matching, pN1c patients had a worse 3-year DFS compared with pN0 patients (58.6% vs 82.4%; P = .035) and a tendency toward a worse DFS among N1 patients (40.1% vs 64.2%; P = .153). DFS was worse when one TD was compared with one invaded LN (40.8% vs 81.3%; P < .001).

CONCLUSION

In rectal cancer, TDs have a metastatic risk comparable to a pN2 stage which may lead to changes in adjuvant treatment.

Association of KRAS mutation with tumor deposit status and overall survival of colorectal cancer

PURPOSE

To examine associations of KRAS mutation with tumor deposit status and overall survival in colorectal cancer (CRC) patients.

METHODS

This retrospective cohort study included patients with incidental CRC diagnosed during 2010-2014 and recorded statuses of KRAS and tumor deposit in the National Cancer Database of the USA. Multivariable logistic regression and time-varying Cox regression analyses were used.

RESULTS

We included 45,761 CRC patients with KRAS status (24,027 [52.5%] men, 24,240 [53.0%] < 65 years old, 17,338 [37.9%] with KRAS mutation). Adjusted for microsatellite instability, age, pathologic stage and tumor grade, KRAS mutation (versus wild type) was associated with tumor deposit presence (odds ratio = 1.11, 95% CI 1.02-1.20). KRAS mutation was also linked to worse overall survival of CRC patients regardless of tumor deposit status (adjusted Hazard ratio [HR] = 1.20, 95% CI 1.07-1.33 for CRC with tumor deposits, and adjusted HR = 1.24, 95% CI 1.14-1.35 or CRC without) or tumor stage (adjusted HR = 1.32, 95% CI 1.14-1.54 for early-stage and adjusted HR = 1.18, 95% CI 1.10-1.27 for late-stage). Microsatellite instability was associated with better overall survival in CRC without tumor deposit (adjusted HR = 0.89, 95% CI 0.79-0.99), but not in CRC with tumor deposit (adjusted HR = 1.12, 95% CI 0.97-1.30).

CONCLUSION

KRAS mutation is independently associated with tumor deposit presence and a worse overall survival in CRC patients.

Prognostic value of N1c in colorectal cancer: a large population-based study using propensity score matching

PURPOSE

We conducted this large population-based study to investigate the prognostic significance of N1c.

METHODS

Patients diagnosed with colorectal cancer from the surveillance, epidemiology, and end results (SEER) database between January 1, 2010, and December 31, 2010, were included in the sample. The primary outcome of interest used in our study was cause-specific survival (CSS). Cox proportional hazards models and Kaplan-Meier methods were used to evaluate the prognostic value of N1c. Propensity score matching (PSM) was implemented to reduce the possibility of selection bias using a logistic regression model.

RESULTS

A total of 19,991 patients diagnosed with colorectal cancer were identified from the SEER database. The median follow-up time of the whole cohort was 60 months (0-71 months). Multivariate Cox analysis showed that N1c was associated with significantly higher risk of colorectal cancer-specific mortality compared with N0 (HR = 1.962, 95%CI = 1.642 to 2.343, P < 0.001) and N1a (HR = 0.818, 95%CI = 0.678 to 0.987, P = 0.036); N1c was associated with significantly lower risk of colorectal cancer-specific mortality compared with N2a (HR = 1.296, 95%CI = 1.081 to 1.554, P = 0.005) and N2b (HR = 1.663, 95%CI = 1.391 to 1.989, P < 0.001). Yet the CSS difference between N1b and N1c did not achieve statistical difference (HR = 1.089, 95%CI = 0.909 to 1.304, P = 0.354).

CONCLUSIONS

The large population-based and propensity score-matched study with long follow-up time provides the first evidence that CSS difference between N1b and N1c does not achieve a statistical difference.