Search

BackgroundTo construct prognostic model of colorectal cancer (CRC) recurrence and metastasis (R&M) with traditional Chinese medicine (TCM) factors based on different machine learning (ML) methods. Aiming to offset the defects in the existing model lacking TCM factors.MethodsPatients with stage I-III CRC after radical resection were included as the model data set. The training set and the internal verification set were randomly divided at a ratio of 7: 3 by the “set aside method”. The average performance index and 95% confidence interval of the model were calculated by repeating 100 tests. Eight factors were used as predictors of Western medicine. Two types of models were constructed by taking “whether to accept TCM intervention” and “different TCM syndrome types” as TCM predictors. The model was constructed by four ML methods: logistic regression, random forest, Extreme Gradient Boosting (XGBoost) and support vector machine (SVM). The predicted target was whether R&M would occur within 3 years and 5 years after radical surgery. The area under curve (AUC) value and decision curve analysis (DCA) curve were used to evaluate accuracy and utility of the model.ResultsThe model data set consisted of 558 patients, of which 317 received TCM intervention after radical resection. The model based on the four ML methods with the TCM factor of “whether to accept TCM intervention” showed good ability in predicting R&M within 3 years and 5 years (AUC value > 0.75), and XGBoost was the best method. The DCA indicated that when the R&M probability in patients was at a certain threshold, the models provided additional clinical benefits. When predicting the R&M probability within 3 years and 5 years in the model with TCM factors of “different TCM syndrome types”, the four methods all showed certain predictive ability (AUC value > 0.70). With the exception of the model constructed by SVM, the other methods provided additional clinical benefits within a certain probability threshold.ConclusionThe prognostic model based on ML methods shows good accuracy and clinical utility. It can quantify the influence degree of TCM factors on R&M, and provide certain values for clinical decision-making.

Positive circumferential resection margin (CRM) was associated with a higher recurrence rate and worse survival in rectal cancer. Predictors of CRM in rectal cancer have widely been investigated. Our study aims to determine the incidence, predictors and prognostic implications of positive CRM following colon cancer (CC) surgery in a Chinese high-volume cancer center. The clinicopathological features and oncological outcomes of CC patients undergoing surgery between January 2008 and December 2018 were identified from Fudan University Shanghai Cancer Center database. Positive CRM was defined as resection margin ≤1 mm. A total of 5268 stage I-IV CC patients were identified in our study, 108 (2.05%) of whom had positive CRM. Multivariate logistic analysis found that advanced N stage, distant metastases and poorly differentiated tumor had increased risk of positive CRM. After propensity score matching, the 5-year overall survival rates of the patients with positive and negative CRM were 33.2% and 39.8% (P=0.005), respectively. Multivariable COX regression model showed that positive CRM was an independent prognostic factor for OS in CC patients. The overall rate of positive CRM in our center is lower than that in western population. Several adverse pathological parameters deserve more attention to identify CC patients at a high risk of positive CRM. Adoption of appropriate surgical techniques and multidisciplinary treatment planning are expected to improve oncological outcomes for high selected CC patients with “high-risk” CRM involvement.

The aim of this study was to explore the prognostic factors in stage I-III colorectal cancer (CRC) patients who had survived for over five years. A total of 9754 stage I-III CRC patients who received curative surgery in the Department of Colorectal Surgery, Fudan University Shanghai Cancer Center were enrolled in this study. Of them, 3640 patients had survived for over five years after surgery. Univariate and multivariate Cox regression analyses were performed in the entire cohort and those who had survived for over five years. Compared with patients in the entire cohort, patients who had survived for over five years were more likely to be younger, have less disease of signet ring cell histology, perineural invasion and vascular invasion, more well differentiated tumors and stage I disease. In the entire cohort, increased age, signet ring cell, poor differentiation, more advanced pathological stage, perineural invasion and vascular invasion were inversely associated with disease-free survival (DFS) and overall survival (OS) using multivariable Cox regression analyses. Only age, pathological stage and perineural invasion remained significant in patients who had survived for over five years. Moreover, tumor location was an independent factor for OS in this subgroup. Predictors for prognosis of CRC change over time. Age, pathological stage and perineural invasion deserve more attention among patients who have survived for over five years.

ObjectiveTo elucidate the prognostic significance of mean corpuscular volume (MCV), with implications of habitual alcohol intake in stage II-III colorectal cancer (CRC).BackgroundMCV had the potential to become an ideal prognostic biomarker and be put into clinical application. Few studies, however, have explored whether habitual alcohol intake which greatly increased the value of MCV would affect the prognostic role of MCV.MethodsEligible patients were identified from the CRC database of Fudan University Shanghai Cancer Center (FUSCC) between January 2012 and December 2013. Survival analyses were constructed using the Kaplan–Meier method to evaluate the survival time distribution, and the log-rank test was used to determine the survival differences. Univariate and multivariate Cox proportional hazard models were built to calculate the hazard ratios of different prognostic factors.ResultsA total of 694 patients diagnosed with stage II-III CRC between January 2012 and December 2013 were identified from FUSCC. Low pretreatment MCV was independently associated with 72.0% increased risk of overall mortality compared with normal MCV (HR = 1.720, 95%CI =1.028-2.876, P =0.039, using normal MCV as the reference). In patients with habitual alcohol intake, however, pretreatment MCV positively correlated with the mortality (P = 0.02) and tumor recurrence (P = 0.002) after adjusting for other known prognostic factors.ConclusionsIn CRC patients without habitual alcohol intake, low (

BackgroundOsteosarcoma (OTS) is aggressive bone malignancy without well-recognized prognosis biomarker. Tumor mutational burden (TMB) has been proved as effective biomarker in predicting clinical outcomes in several cancer types. However, its prognostic value in OTS remains unknown. In this study, we aim to evaluate the implication of TMB in OTS patients.MethodsTo depict the landscape of somatic mutations in OTS, we performed Whole-Exome Sequencing (WES) on 31 OTS tissue samples and corresponding White Blood Cells (WBCs) as matched control. TMB was calculated as the total number of somatic alterations in coding regions normalized to the per sequenced genomic megabase (~30.4Mb in WES). The prognostic values of TMB were evaluated by Kaplan-Meier methods and Cox regression models.ResultsThe median age was 16.0 years at diagnosis, and 54.8% of patients were male. The most common genetic alterations were mainly involved in cell cycle and DNA damage response and repair, including H3F3A, TP53, MYC, and CDKN2A/B. The median progression-free survival (PFS) was 775.5 days in TMB-High (defined as third quartile of TMB value, 400 days), while 36.4% of all patients with TMB-Low were PFS-Long (P=0.003). TMB were significantly greater in PFS-Long than in PFS-Short (