The AUGIS Survival Predictor: Prediction of Long-Term and Conditional Survival After Esophagectomy Using Random Survival Forests

Application of machine learning for high-throughput tumor marker screening

High-throughput sequencing and multiomics technologies have allowed increasing numbers of biomarkers to be mined and used for disease diagnosis, risk stratification, efficacy assessment, and prognosis prediction. However, the large number and complexity of tumor markers make screening them a substantial challenge. Machine learning (ML) offers new and effective ways to solve the screening problem. ML goes beyond mere data processing and is instrumental in recognizing intricate patterns within data. ML also has a crucial role in modeling dynamic changes associated with diseases. Used together, ML techniques have been included in automatic pipelines for tumor marker screening, thereby enhancing the efficiency and accuracy of the screening process. In this review, we discuss the general processes and common ML algorithms, and highlight recent applications of ML in tumor marker screening of genomic, transcriptomic, proteomic, and metabolomic data of patients with various types of cancers. Finally, the challenges and future prospects of the application of ML in tumor therapy are discussed.

Development and validation of a random survival forest model for predicting long-term survival of early-stage young breast cancer patients based on the SEER database and an external validation cohort

Young breast cancer (YBC) patients often face a poor prognosis, hence it's necessary to construct a model that can accurately predict their long-term survival in early stage. To realize this goal, we utilized data from the Surveillance, Epidemiology, and End Results (SEER) databases between January 2010 and December 2020, and meanwhile, enrolled an independent external cohort from Tianjin Medical University Cancer Institute and Hospital. The study aimed to develop and validate a prediction model constructed using the Random Survival Forest (RSF) machine learning algorithm. By applying the Least Absolute Shrinkage and Selection Operator (LASSO) regression analysis, we pinpointed key prognostic factors for YBC patients, which were used to create a prediction model capable of forecasting the 3-year, 5-year, 7-year, and 10-year survival rates of YBC patients. The RSF model constructed in the study demonstrated exceptional performance, achieving C-index values of 0.920 in the training set, 0.789 in the internal validation set, and 0.701 in the external validation set, outperforming the Cox regression model. The model's calibration was confirmed by Brier scores at various time points, showcasing its excellent accuracy in prediction. Decision curve analysis (DCA) underscored the model's importance in clinical application, and the Shapley Additive Explanations (SHAP) plots highlighted the importance of key variables. The RSF model also proved valuable in risk stratification, which has effectively categorized patients based on their survival risks. In summary, this study has constructed a well-performed prediction model for the evaluation of prognostic factors influencing the long-term survival of early-stage YBC patients, which is significant in risk stratification when physicians handle YBC patients in clinical settings.

Prediction of survival in patients with infected pancreatic necrosis: a prospective cohort study

BACKGROUND

Infected pancreatic necrosis (IPN) is a severe complication of acute pancreatitis, with mortality rates ranging from 15 to 35%. However, limited studies exist to predict the survival of IPN patients and nomogram has never been built. This study aimed to identify predictors of mortality, estimate conditional survival (CS), and develop a CS nomogram and logistic regression nomogram for real-time prediction of survival in IPN patients.

METHODS

A prospective cohort study was performed in 335 IPN patients consecutively enrolled at a large Chinese tertiary hospital from January 2011 to December 2022. The random survival forest method was first employed to identify the most significant predictors and capture clinically relevant nonlinear threshold effects. Instantaneous death risk and CS was first utilized to reveal the dynamic changes in the survival of IPN patients. A Cox model-based nomogram incorporating CS and a logistic regression-based nomogram were first developed and internally validated with a bootstrap method.

RESULTS

The random survival forest model identified seven foremost predictors of mortality, including the number of organ failures, duration of organ failure, age, time from onset to first intervention, hemorrhage, bloodstream infection, and severity classification. Duration of organ failure and time from onset to first intervention showed distinct thresholds and nonlinear relationships with mortality. Instantaneous death risk reduced progressively within the first 30 days, and CS analysis indicated gradual improvement in real-time survival since diagnosis, with 90-day survival rates gradually increasing from 0.778 to 0.838, 0.881, 0.974, and 0.992 after surviving 15, 30, 45, 60, and 75 days, respectively. After further variables selection using step regression, five predictors (age, number of organ failures, hemorrhage, time from onset to first intervention, and bloodstream infection) were utilized to construct both the CS nomogram and logistic regression nomogram, both of which demonstrated excellent performance with 1000 bootstrap.

CONCLUSION

Number of organ failures, duration of organ failure, age, time from onset to first intervention, hemorrhage, bloodstream infection, and severity classification were the most crucial predictors of mortality of IPN patients. The CS nomogram and logistic regression nomogram constructed by these predictors could help clinicians to predict real-time survival and optimize clinical decisions.

Total neoadjuvant therapy in oesophageal and gastro-oesophageal junctional adenocarcinoma

Adenocarcinoma of the oesophagus and gastro-oesophageal junction represent a large burden of cancer death in the Western World with an increasing incidence. In the past two decades, the overall survival of patients on a potentially curative treatment pathway has more than doubled due to the addition of perioperative oncological therapies to surgery. However, patients often fail to respond to oncological treatment or struggle to complete their treatment after surgery. In this review, we discuss the current evidence for total neoadjuvant therapy and options for assessment of treatment response.

An optimized postsurgery follow-up strategy for patients with esophageal cancer: a cohort study

BACKGROUND

After radical surgery, patients with esophageal cancer should undergo long-term surveillance of disease relapse. However, the optimal follow-up strategy remains to be explored.

METHOD

A total of 4688 patients were recruited. Recursive partition analysis was applied to develop recurrence risk stratification for patients. The follow-up strategies of each stratification were developed based on monthly recurrence probability and validated by bootstrap validation and an external dataset. A Markov decision-analytic model was constructed to evaluate the cost-effectiveness of the follow-up strategies.

RESULTS

Patients were stratified into four groups according to four pathological features. The authors applied a random survival forest to calculate the monthly recurrence probability of each group. Based on the temporal distribution of recurrences, the authors further established surveillance strategies for four groups. The strategies were validated as optimal protocols by bootstrap resampling and another dataset. Markov cost-effective analysis indicated that our recommended strategies outperformed the mainstream protocols from guidelines. Using less than 12 visits across the first 5 years on average, our follow-up strategies were more efficient than the NCCN recommended strategies (14 visits average). Our results also supported the computerized tomography from the neck to the upper abdomen as a routine examination and PETCT of distant metastasis for some groups with high risks.

CONCLUSION

Our study provided data-driven evidence of personalized and economic follow-up strategies for esophageal cancer patients and shed light on follow-up optimization for other cancer types.

The effect of surgical complications on long-term prognosis following oesophagectomy

INTRODUCTION

Complications are frequent after oesophagectomy, and there is evidence these adversely impact long-term prognosis. However, the effect of multiple complications, and the absolute magnitude of effect on survival is unclear. This study aimed to examine these effects in a single high-volume UK unit.

METHODS

Patients undergoing oesophagectomy for cancer and who survived to 90 days post-oesophagectomy were analysed. Complications were graded according to the Clavien-Dindo (CD) classification and the Comprehensive Complication Index (CCI). The effect and magnitude of effect of complications on survival were assessed using multivariable cox regression and the risk-adjusted population attributable fraction.

RESULTS

In total, 380 patients were included. Complications occurred in 251 (66.1%). Suffering ≥3 complications (HR 1.89, 95%CI 1.13-3.16, p = 0.015) or an unplanned escalation in care (HR 2.22, 95%CI 1.43-3.45, p < 0.001) significantly reduced survival whereas pulmonary complications and anastomotic leak did not. Patients with a CCI>30 had worse overall survival (HR 1.91, 95%CI 1.32-2.76, p < 0.001) and CCI>30 due to multiple minor complications gave a worse prognosis compared to CCI>30 due to major complications (HR 2.44, 95%CI 1.14-5.20, p = 0.022). An estimated 9.1% (95%CI 3.4-14.4%) of deaths at 5 years were attributable to a CCI>30.

CONCLUSION

Long-term survival following oesophagectomy for cancer is significantly affected by complications and the cumulative effect of multiple complications. Interestingly, multiple minor complications had a worse effect on survival than major complications. The absolute magnitude of effect is substantial: minimising all types of postoperative complications could have significant benefit to overall outcomes.

Development and Validation of a Machine-Learning Model to Predict Early Recurrence of Intrahepatic Cholangiocarcinoma

BACKGROUND

The high incidence of early recurrence after hepatectomy for intrahepatic cholangiocarcinoma (ICC) has a detrimental effect on overall survival (OS). Machine-learning models may improve the accuracy of outcome prediction for malignancies.

METHODS

Patients who underwent curative-intent hepatectomy for ICC were identified using an international database. Three machine-learning models were trained to predict early recurrence (< 12 months after hepatectomy) using 14 clinicopathologic characteristics. The area under the receiver operating curve (AUC) was used to assess their discrimination ability.

RESULTS

In this study, 536 patients were randomly assigned to training (n = 376, 70.1%) and testing (n = 160, 29.9%) cohorts. Overall, 270 (50.4%) patients experienced early recurrence (training: n = 150 [50.3%] vs testing: n = 81 [50.6%]), with a median tumor burden score (TBS) of 5.6 (training: 5.8 [interquartile range {IQR}, 4.1-8.1] vs testing: 5.5 [IQR, 3.7-7.9]) and metastatic/undetermined nodes (N1/NX) in the majority of the patients (training: n = 282 [75.0%] vs testing n = 118 [73.8%]). Among the three different machine-learning algorithms, random forest (RF) demonstrated the highest discrimination in the training/testing cohorts (RF [AUC, 0.904/0.779] vs support vector machine [AUC, 0.671/0.746] vs logistic regression [AUC, 0.668/0.745]). The five most influential variables in the final model were TBS, perineural invasion, microvascular invasion, CA 19-9 lower than 200 U/mL, and N1/NX disease. The RF model successfully stratified OS relative to the risk of early recurrence.

CONCLUSIONS

Machine-learning prediction of early recurrence after ICC resection may inform tailored counseling, treatment, and recommendations. An easy-to-use calculator based on the RF model was developed and made available online.

A novel staging system based on deep learning for overall survival in patients with esophageal squamous cell carcinoma

PURPOSE

We developed DeepSurv, a deep learning approach for predicting overall survival (OS) in patients with esophageal squamous cell carcinoma (ESCC). We validated and visualized the novel staging system based on DeepSurv using data from multiple cohorts.

METHODS

Totally 6020 ESCC patients diagnosed from January 2010 to December 2018 were included in the present study from the Surveillance, Epidemiology, and End Results database (SEER), randomly assigned to the training and test cohorts. We developed, validated and visualized a deep learning model that included 16 prognostic factors; then a novel staging system was further constructed based on the total risk score derived from the deep learning model. The classification performance at 3-year and 5-year OS was assessed by the receiver-operating characteristic (ROC) curve. Calibration curve and the Harrell's concordance index (C-index) were also used to comprehensively assess the predictive performance of the deep learning model. Decision curve analysis (DCA) was utilized to assess the clinical utility of the novel staging system.

RESULTS

A more applicable and accurate deep learning model was established, which outperformed the traditional nomogram in predicting OS in the test cohort (C-index: 0.732 [95% CI 0.714-0.750] versus 0.671 [95% CI 0.647-0.695]). The ROC curves at 3-year and 5-year OS for the model also showed good discrimination ability in the test cohort (Area Under the Curve [AUC] at 3-/5-year OS = 0.805/0.825). Moreover, using our novel staging system, we observed a clear survival difference among different risk groups (P < 0.001) and a significant positive net benefit in the DCA.

CONCLUSIONS

A novel deep learning-based staging system was constructed for patients with ESCC, which performed a significant discriminability for survival probability. Moreover, an easy-to-use web-based tool based on the deep learning model was also implemented, offering convenience for personalized survival prediction. We developed a deep learning-based system that stages patients with ESCC according to their survival probability. We also created a web-based tool that uses this system to predict individual survival outcomes.

Machine Learning-Based Prognostic Model for Patients After Lung Transplantation

Importance

Although numerous prognostic factors have been found for patients after lung transplantation (LTx) over the years, an accurate prognostic tool for LTx recipients remains unavailable.

Objective

To develop and validate a prognostic model for predicting overall survival in patients after LTx using random survival forests (RSF), a machine learning algorithm.

Design, Setting, and Participants

This retrospective prognostic study included patients who underwent LTx between January 2017 and December 2020. The LTx recipients were randomly assigned to training and test sets in accordance with a ratio of 7:3. Feature selection was performed using variable importance with bootstrapping resampling. The prognostic model was fitted using the RSF algorithm, and a Cox regression model was set as a benchmark. The integrated area under the curve (iAUC) and integrated Brier score (iBS) were applied to assess model performance in the test set. Data were analyzed from January 2017 to December 2019.

Main Outcomes And Measures

Overall survival in patients after LTx.

Results

A total of 504 patients were eligible for this study, consisting of 353 patients in the training set (mean [SD] age, 55.03 [12.78] years; 235 [66.6%] male patients) and 151 patients in the test set (mean [SD] age, 56.79 [10.95] years; 99 [65.6%] male patients). According to the variable importance of each factor, 16 were selected for the final RSF model, and postoperative extracorporeal membrane oxygenation time was identified as the most valuable factor. The RSF model had excellent performance with an iAUC of 0.879 (95% CI, 0.832-0.921) and an iBS of 0.130 (95% CI, 0.106-0.154). The Cox regression model fitted by the same modeling factors to the RSF model was significantly inferior to the RSF model with an iAUC of 0.658 (95% CI, 0.572-0.747; P < .001) and an iBS of 0.205 (95% CI, 0.176-0.233; P < .001). According to the RSF model predictions, the patients after LTx were stratified into 2 prognostic groups displaying significant difference, with mean overall survival of 52.91 months (95% CI, 48.51-57.32) and 14.83 months (95% CI, 9.44-20.22; log-rank P < .001), respectively.

Conclusions and relevance

In this prognostic study, the findings first demonstrated that RSF could provide more accurate overall survival prediction and remarkable prognostic stratification than the Cox regression model for patients after LTx.

Artificial intelligence to investigate predictors and prognostic impact of time to surgery in colon cancer

BACKGROUND AND OBJECTIVES

The role of time to surgery (TTS) for long-term outcomes in colon cancer (CC) remains ill-defined. We sought to utilize artificial intelligence (AI) to characterize the drivers of TTS and its prognostic impact.

METHODS

The National Cancer Database was utilized to identify patients diagnosed with non-metastatic CC between 2004 and 2018. AI models were employed to rank the importance of several sociodemographic, facility, and tumor characteristics in determining TTS, and postoperative survival.

RESULTS

Among 518 983 patients, 137 902 (26.6%) received intraoperative diagnosis of CC (TTS = 0), while 381 081 (74.4%) underwent elective surgery (TTS > 0) with median TTS of 19.0 days (interquartile range [IQR]: 7.0-33.0). An AI model, identified tumor stage, receipt of adequate lymphadenectomy, histologic grade, lymphovascular invasion, and insurance status as the most important variables associated with TTS = 0. Conversely, the type and location of treating facility and receipt of adjuvant therapy were among the most important variables for TTS > 0. Notably, TTS was among the most important variables associated with survival, and TTS > 3 weeks was associated with an incremental increase in mortality risk.

CONCLUSIONS

The identification of factors associated with TTS can help stratify patients most likely to suffer poor outcomes due to prolonged TTS, as well as guide quality improvement initiatives related to timely surgical care.

Machine learning-based radiomic computed tomography phenotyping of thymic epithelial tumors: Predicting pathological and survival outcomes

OBJECTIVE

For patients with thymic epithelial tumors, accurately predicting clinicopathological outcomes remains challenging. We aimed to investigate the performance of machine learning-based radiomic computed tomography phenotyping for predicting pathological (World Health Organization [WHO] type and TNM stage) and survival outcomes (overall and progression-free survival) in patients with thymic epithelial tumors.

METHODS

This retrospective study included patients with thymic epithelial tumors between January 2001 and January 2022. The radiomic features were extracted from preoperative unenhanced computed tomography images. After strict feature selection, random forest and random survival forest models were fitted to predict pathological and survival outcomes, respectively. The model performance was assessed by the area under the curve (AUC) and validated internally by the bootstrap method.

RESULTS

In total, 124 patients with a median age of 61 years were included. The radiomics random forest models of WHO type and TNM stage showed satisfactory performance with an AUC_WHO of 0.898 (95% CI, 0.753-1.000) and an AUC_TNM of 0.766 (95% CI, 0.642-0.886). For overall survival and progression-free survival prediction, the radiomics random survival forest models showed good performance (integrated AUCs, 0.923; 95% CI, 0.691-1.000 and 0.702; 95% CI, 0.513-0.875, respectively), and the integrated AUCs increased to 0.935 (95% CI, 0.705-1.000) and 0.811 (95% CI, 0.647-0.942), respectively, when combined with clinicopathological features.

CONCLUSIONS

Machine learning-based radiomic computed tomography phenotyping might allow for the satisfactory prediction of pathological and survival outcomes and further improve prognostic performance when integrated with clinicopathological features in patients with thymic epithelial tumors.

A prediction model for the 5-year, 10-year and 20-year mortality of medullary thyroid carcinoma patients based on lymph node ratio and other predictors

Aim

To explore the predictive value of lymph node ratio (LNR) for the prognosis of medullary thyroid carcinoma (MTC) patients, and constructed prediction models for the 5-year, 10-year and 20-year mortality of MTC patients based on LNR and other predictors.

Methods

This cohort study extracted the data of 2,093 MTC patients aged ≥18 years undergoing total thyroidectomy and neck lymph nodes dissection. Kaplan-Meier curves and log-rank tests were performed to compare survival curves between LNR < 15% group and LNR ≥ 15% group. All data was divided into the training set (n = 1,465) and the testing set (n = 628). The random survival forest model was constructed in the training set and validated in the testing set. The area under the curve (AUC) was employed for evaluating the predictive ability of the model.

Results

The 5-year, 10-year and 20-year overall survival (OS) and cause-specific survival (CSS) of MTC patients with LNR <15% were higher than those with LNR ≥15%. The OS was 46% and the CSS was 75% after 20 years' follow-up. The AUC of the model for the 5-year, 10-year, and 20-year OS in MTC patients was 0.878 (95%CI: 0.856-0.900), 0.859 (95%CI: 0.838-0.879) and 0.843 (95%CI: 0.823-0.862) in the training set and 0.845 (95%CI: 0.807-0.883), 0.841 (95%CI: 0.807-0.875) and 0.841 (95%CI: 0.811-0.872) in the testing set. In the training set, the AUCs were 0.869 (95%CI: 0.845-0.892), 0.843 (95%CI: 0.821-0.865), 0.819 (95%CI: 0.798-0.840) for the 5-year, 10-year and 20-year CCS in MTC patients, respectively. In the testing set, the AUCs were 0.857 (95%CI: 0.822-0.892), 0.839 (95%CI: 0.805-0.873) and 0.826 (95%CI: 0.794-0.857) for the 5-year CCS, 10-year CCS and 20-year CCS in MTC patients, respectively.

Conclusion

The models displayed good predictive performance, which might help identify MTC patients might have poor outcomes and appropriate interventions should be applied in these patients.

Endoscopic submucosal dissection versus esophagectomy for t1 esophageal squamous cell carcinoma: a propensity score-matched analysis

BACKGROUND

Endoscopic submucosal dissection (ESD) has been a preferred treatment option for superficial esophageal squamous cell carcinoma (SESCC).

OBJECTIVES

To compare the outcomes of ESD and esophagectomy in the treatment of SESCC, especially for lesions invading muscularis mucosa or submucosa (pT1a-MM/T1b).

DESIGN

We retrospectively analyzed data from patients with SESCC who underwent ESD or esophagectomy between 2015 and 2021.

METHODS

After propensity score matching, overall survival (OS), disease-specific survival (DSS), recurrence-free survival (RFS), and treatment-related events were compared between the ESD and esophagectomy groups. Furthermore, we performed a Cox regression analysis to identify factors associated with survival.

RESULTS

OS and DSS were significantly higher in the ESD group (n = 508) than that in the esophagectomy group (n = 466). After matching, 404 patients (202 per group) were included in the study. No significant differences were found between the ESD and esophagectomy groups in OS (p = 0.566), RFS (p = 0.586), and DSS (p = 0.912). The ESD group showed less blood loss, shorter procedure duration and hospital stay, lower hospital cost, and fewer adverse events. However, a lower R0 resection rate was observed in the ESD group compared to the esophagectomy group. Subgroup analysis showed comparable survival outcomes between the two groups. In Cox regression analysis, age was the independent factor associated with OS.

CONCLUSION

In the treatment of SESCC, ESD showed sufficient safety and advantages. Even for pT1a-MM/pT1b SESCC, ESD may be an alternative treatment to esophagectomy.

Development and validation of a deep learning model to predict survival of patients with esophageal cancer

Objective

To compare the performance of a deep learning survival network with the tumor, node, and metastasis (TNM) staging system in survival prediction and test the reliability of individual treatment recommendations provided by the network.

Methods

In this population-based cohort study, we developed and validated a deep learning survival model using consecutive cases of newly diagnosed stage I to IV esophageal cancer between January 2004 and December 2015 in a Surveillance, Epidemiology, and End Results (SEER) database. The model was externally validated in an independent cohort from Fujian Provincial Hospital. The C statistic was used to compare the performance of the deep learning survival model and TNM staging system. Two other deep learning risk prediction models were trained for treatment recommendations. A Kaplan–Meier survival curve was used to compare survival between the population that followed the recommended therapy and those who did not.

Results

A total of 9069 patients were included in this study. The deep learning network showed more promising results in predicting esophageal cancer-specific survival than the TNM stage in the internal test dataset (C-index=0.753 vs. 0.638) and external validation dataset (C-index=0.687 vs. 0.643). The population who received the recommended treatments had superior survival compared to those who did not, based on the internal test dataset (hazard ratio, 0.753; 95% CI, 0.556-0.987; P=0.042) and the external validation dataset (hazard ratio, 0.633; 95% CI, 0.459-0.834; P=0.0003).

Conclusion

Deep learning neural networks have potential advantages over traditional linear models in prognostic assessment and treatment recommendations. This novel analytical approach may provide reliable information on individual survival and treatment recommendations for patients with esophageal cancer.

The impact of endoscopic ultrasound on the management and outcome of patients with oesophageal cancer: an update of a systematic review

AIM

To provide an updated systematic review concerning the impact of endoscopic ultrasound (EUS) in the modern era of oesophageal cancer staging.

MATERIALS AND METHODS

To update the previous systematic review, databases including MEDLINE and EMBASE were searched and studies published from 2005 onwards were selected. Studies reporting primary data in patients with oesophageal or gastro-oesophageal junction cancer who underwent radiological staging and treatment, regardless of intent, were included. The primary outcome was the reported change in management after EUS. Secondary outcomes were recurrence rate and overall survival. Two reviewers extracted data from included articles. This study was registered with PROSPERO (CRD42021231852).

RESULTS

Eighteen studies with 11,836 patients were included comprising 2,805 patients (23.7%) who underwent EUS compared to 9,031 (76.3%) without EUS examination. Reported change of management varied widely from 0% to 56%. When used, EUS fine-needle aspiration precluded curative treatment in 37.5%-71.4%. Overall survival improvements ranged between 121 and 639 days following EUS intervention compared to patients without EUS. Smaller effect sizes were observed in a randomised controlled trial, compared to larger differences reported in observational studies.

CONCLUSION

Current evidence for the effectiveness of EUS in oesophageal cancer pathways is conflicting and of limited quality. In particular, the extent to which EUS adds value to contemporary cross-sectional imaging techniques is unclear and requires formal re-evaluation.

Impact of postoperative chemotherapy on survival for oesophagogastric adenocarcinoma after preoperative chemotherapy and surgery

BACKGROUND

Perioperative chemotherapy is widely used in the treatment of oesophagogastric adenocarcinoma (OGAC) with a substantial survival benefit over surgery alone. However, the postoperative part of these regimens is given in less than half of patients, reflecting uncertainty among clinicians about its benefit and poor postoperative patient fitness. This study estimated the effect of postoperative chemotherapy after surgery for OGAC using a large population-based data set.

METHODS

Patients with adenocarcinoma of the oesophagus, gastro-oesophageal junction or stomach diagnosed between 2012 and 2018, who underwent preoperative chemotherapy followed by surgery, were identified from a national-level audit in England and Wales. Postoperative therapy was defined as the receipt of systemic chemotherapy within 90 days of surgery. The effectiveness of postoperative chemotherapy compared with observation was estimated using inverse propensity treatment weighting.

RESULTS

Postoperative chemotherapy was given to 1593 of 4139 patients (38.5 per cent) included in the study. Almost all patients received platinum-based triplet regimens (4004 patients, 96.7 per cent), with FLOT used in 3.3 per cent. Patients who received postoperative chemotherapy were younger, with a lower ASA grade, and were less likely to have surgical complications, with similar tumour characteristics. After weighting, the median survival time after postoperative chemotherapy was 62.7 months compared with 50.4 months without chemotherapy (hazard ratio 0.84, 95 per cent c.i. 0.77 to 0.94; P = 0.001).

CONCLUSION

This study has shown that postoperative chemotherapy improves overall survival in patients with OGAC treated with preoperative chemotherapy and surgery.

Prediction of long-term survival after gastrectomy using random survival forests

BACKGROUND

No well validated and contemporaneous tools for personalized prognostication of gastric adenocarcinoma exist. This study aimed to derive and validate a prognostic model for overall survival after surgery for gastric adenocarcinoma using a large national dataset.

METHODS

National audit data from England and Wales were used to identify patients who underwent a potentially curative gastrectomy for adenocarcinoma of the stomach. A total of 2931 patients were included and 29 clinical and pathological variables were considered for their impact on survival. A non-linear random survival forest methodology was then trained and validated internally using bootstrapping with calibration and discrimination (time-dependent area under the receiver operator curve (tAUC)) assessed.

RESULTS

The median survival of the cohort was 69 months, with a 5-year survival of 53.2 per cent. Ten variables were found to influence survival significantly and were included in the final model, with the most important being lymph node positivity, pT stage and achieving an R0 resection. Patient characteristics including ASA grade and age were also influential. On validation the model achieved excellent performance with a 5-year tAUC of 0.80 (95 per cent c.i. 0.78 to 0.82) and good agreement between observed and predicted survival probabilities. A wide spread of predictions for 3-year (14.8-98.3 (i.q.r. 43.2-84.4) per cent) and 5-year (9.4-96.1 (i.q.r. 31.7-73.8) per cent) survival were seen.

CONCLUSIONS

A prognostic model for survival after a potentially curative resection for gastric adenocarcinoma was derived and exhibited excellent discrimination and calibration of predictions.

Development of a random forest model for hypotension prediction after anesthesia induction for cardiac surgery

BACKGROUND

Hypotension after the induction of anesthesia is known to be associated with various adverse events. The involvement of a series of factors makes the prediction of hypotension during anesthesia quite challenging.

AIM

To explore the ability and effectiveness of a random forest (RF) model in the prediction of post-induction hypotension (PIH) in patients undergoing cardiac surgery.

METHODS

Patient information was obtained from the electronic health records of the Second Affiliated Hospital of Hainan Medical University. The study included patients, ≥ 18 years of age, who underwent cardiac surgery from December 2007 to January 2018. An RF algorithm, which is a supervised machine learning technique, was employed to predict PIH. Model performance was assessed by the area under the curve (AUC) of the receiver operating characteristic. Mean decrease in the Gini index was used to rank various features based on their importance.

RESULTS

Of the 3030 patients included in the study, 1578 (52.1%) experienced hypotension after the induction of anesthesia. The RF model performed effectively, with an AUC of 0.843 (0.808-0.877) and identified mean blood pressure as the most important predictor of PIH after anesthesia. Age and body mass index also had a significant impact.

CONCLUSION

The generated RF model had high discrimination ability for the identification of individuals at high risk for a hypotensive event during cardiac surgery. The study results highlighted that machine learning tools confer unique advantages for the prediction of adverse post-anesthesia events.

Machine Learning for Future Subtyping of the Tumor Microenvironment of Gastro-Esophageal Adenocarcinomas

Tumor progression involves an intricate interplay between malignant cells and their surrounding tumor microenvironment (TME) at specific sites. The TME is dynamic and is composed of stromal, parenchymal, and immune cells, which mediate cancer progression and therapy resistance. Evidence from preclinical and clinical studies revealed that TME targeting and reprogramming can be a promising approach to achieve anti-tumor effects in several cancers, including in GEA. Thus, it is of great interest to use modern technology to understand the relevant components of programming the TME. Here, we discuss the approach of machine learning, which recently gained increasing interest recently because of its ability to measure tumor parameters at the cellular level, reveal global features of relevance, and generate prognostic models. In this review, we discuss the relevant stromal composition of the TME in GEAs and discuss how they could be integrated. We also review the current progress in the application of machine learning in different medical disciplines that are relevant for the management and study of GEA.

Machine Learning Models for the Prediction of Breast Cancer Prognostic: Application and Comparison Based on a Retrospective Cohort Study (Preprint)

Background

Over the recent years, machine learning methods have been increasingly explored in cancer prognosis because of the appearance of improved machine learning algorithms. These algorithms can use censored data for modeling, such as support vector machines for survival analysis and random survival forest (RSF). However, it is still debated whether traditional (Cox proportional hazard regression) or machine learning-based prognostic models have better predictive performance.

Objective

This study aimed to compare the performance of breast cancer prognostic prediction models based on machine learning and Cox regression.

Methods

This retrospective cohort study included all patients diagnosed with breast cancer and subsequently hospitalized in Fudan University Shanghai Cancer Center between January 1, 2008, and December 31, 2016. After all exclusions, a total of 22,176 cases with 21 features were eligible for model development. The data set was randomly split into a training set (15,523 cases, 70%) and a test set (6653 cases, 30%) for developing 4 models and predicting the overall survival of patients diagnosed with breast cancer. The discriminative ability of models was evaluated by the concordance index (C-index), the time-dependent area under the curve, and D-index; the calibration ability of models was evaluated by the Brier score.

Results

The RSF model revealed the best discriminative performance among the 4 models with 3-year, 5-year, and 10-year time-dependent area under the curve of 0.857, 0.838, and 0.781, a D-index of 7.643 (95% CI 6.542, 8.930) and a C-index of 0.827 (95% CI 0.809, 0.845). The statistical difference of the C-index was tested, and the RSF model significantly outperformed the Cox-EN (elastic net) model (C-index 0.816, 95% CI 0.796, 0.836; P=.01), the Cox model (C-index 0.814, 95% CI 0.794, 0.835; P=.003), and the support vector machine model (C-index 0.812, 95% CI 0.793, 0.832; P<.001). The 4 models’ 3-year, 5-year, and 10-year Brier scores were very close, ranging from 0.027 to 0.094 and less than 0.1, which meant all models had good calibration. In the context of feature importance, elastic net and RSF both indicated that TNM staging, neoadjuvant therapy, number of lymph node metastases, age, and tumor diameter were the top 5 important features for predicting the prognosis of breast cancer. A final online tool was developed to predict the overall survival of patients with breast cancer.

Conclusions

The RSF model slightly outperformed the other models on discriminative ability, revealing the potential of the RSF method as an effective approach to building prognostic prediction models in the context of survival analysis.

Consensus recommendations for the standardized histopathological evaluation and reporting after radical oesophago-gastrectomy (HERO consensus)

BACKGROUND

Variation in the approach, radicality, and quality of gastroesophageal surgery impacts patient outcomes. Pathological outcomes such as lymph node yield are routinely used as surrogate markers of surgical quality, but are subject to significant variations in histopathological evaluation and reporting. A multi-society consensus group was convened to develop evidence-based recommendations for the standardized assessment of gastroesophageal cancer specimens.

METHODS

A consensus group comprised of surgeons, pathologists, and oncologists was convened on behalf of the Association of Upper Gastrointestinal Surgery of Great Britain & Ireland. Literature was reviewed for 17 key questions. Draft recommendations were voted upon via an anonymous Delphi process. Consensus was considered achieved where >70% of participants were in agreement.

RESULTS

Consensus was achieved on 18 statements for all 17 questions. Twelve strong recommendations regarding preparation and assessment of lymph nodes, margins, and reporting methods were made. Importantly, there was 100% agreement that the all specimens should be reported using the Royal College of Pathologists Guidelines as the minimum acceptable dataset. In addition, two weak recommendations regarding method and duration of specimen fixation were made. Four topics lacked sufficient evidence and no recommendation was made.

CONCLUSIONS

These consensus recommendations provide explicit guidance for gastroesophageal cancer specimen preparation and assessment, to provide maximum benefit for patient care and standardize reporting to allow benchmarking and improvement of surgical quality.