On an ensemble algorithm for clustering cancer patient data

Pathological-Features-Modified TNM Staging System Improves Prognostic Accuracy for Rectal Cancer

BACKGROUND

Variations in survival outcomes are observed in the eighth edition of the American Joint Committee on Cancer TNM staging system.

OBJECTIVE

Machine learning ensemble methods were used to develop and evaluate the effectiveness of a pathological-features-modified TNM staging system in predicting survival for patients with rectal cancer by use of commonly reported pathological features, such as histological grade, tumor deposits, and perineural invasion, to improve the prognostic accuracy.

DESIGN

This was a retrospective population-based study.

SETTINGS

Data were assessed from the database of the Surveillance, Epidemiology, and End Results Program.

PATIENTS

The study cohort comprised 14,468 patients with rectal cancer diagnosed between 2010 and 2015. The development cohort included those who underwent surgery as the primary treatment, whereas patients who received neoadjuvant therapy were assigned to the validation cohort.

MAIN OUTCOME MEASURES

The primary outcome measures included cumulative rectal cancer survival, adjusted HRs, and both calibration and discrimination statistics to evaluate model performance and internal validation.

RESULTS

Multivariable Cox regression analysis identified all 3 pathological features as prognostic factors, after which patients were categorized into 4 pathological groups based on the number of pathological features (ie, 0, 1, 2, and 3). Distinct survival differences were observed among the groups, especially with patients with stage III rectal cancer. The proposed pathological-features-modified TNM staging outperformed the TNM staging in both the development and validation cohorts.

LIMITATIONS

Retrospective in design and lack of external validation.

CONCLUSIONS

The proposed pathological-features-modified TNM staging could complement the current TNM staging by improving the accuracy of survival estimation of patients with rectal cancer. See Video Abstract .

EL SISTEMA DE ESTADIFICACIN TNM CON CARACTERSTICAS PATOLGICAS MODIFICADO MEJORA LA PRECISIN DEL PRONSTICO DEL CNCER DE RECTO

ANTECEDENTES:Se observan variaciones en los resultados de supervivencia en el sistema de estadificación TNM del Comité Conjunto Americano del Cáncer 8º ediciónOBJETIVO:Se utilizaron métodos conjuntos de aprendizaje automático para desarrollar y evaluar la eficacia de un sistema de estadificación con características patológicas modificadas de tumores, ganglios y metástasis para predecir la supervivencia de pacientes con cáncer de recto, utilizando algunas características patológicas comúnmente informadas, como el grado histológico, depósitos tumorales e invasión perineural, para mejorar la precisión del pronóstico.DISEÑO:Este fue un estudio retrospectivo de base poblacional.ENTERNO CLINICO:Se recuperaron y evaluaron datos de la base de datos de Vigilancia, Epidemiología y Resultados Finales.PACIENTES:La cohorte del estudio estuvo compuesta por 14,468 pacientes con cáncer de recto diagnosticados entre 2010 y 2015. La cohorte de desarrollo incluyó a aquellos que se sometieron a cirugía como tratamiento primario, mientras que los pacientes que recibieron terapia neoadyuvante fueron asignados a la cohorte de validación.PRINCIPALES MEDIDAS DE RESULTADO:Las medidas de resultado primarias incluyeron supervivencia acumulada del cáncer de recto, índices de riesgo ajustados y estadísticas de calibración y discriminación para evaluar el rendimiento del modelo y la validación interna.RESULTADOS:El análisis de regresión multivariable de Cox identificó las tres características patológicas como factores pronósticos, después de lo cual los pacientes se clasificaron en cuatro grupos patológicos según el número de características patológicas (es decir, 0, 1, 2 y 3). Se observaron distintas diferencias en la supervivencia entre los grupos, especialmente en los pacientes en estadio III. La estadificación propuesta con características patológicas modificadas de tumores-ganglios-metástasis superó a la estadificación TNM tanto en las cohortes de desarrollo como en las de validación.LIMITACIONES:Diseño retrospectivo y falta de validación externa.CONCLUSIONES:La estadificación propuesta con características patológicas modificadas de tumores-ganglios-metástasis podría complementar la estadificación TNM actual al mejorar la precisión de la estimación de supervivencia de los pacientes con cáncer de recto. (Traducción- Dr. Francisco M. Abarca-Rendon ).

Machine Learning for Endometrial Cancer Prediction and Prognostication

Endometrial cancer (EC) is a prevalent uterine cancer that remains a major contributor to cancer-associated morbidity and mortality. EC diagnosed at advanced stages shows a poor therapeutic response. The clinically utilized EC diagnostic approaches are costly, time-consuming, and are not readily available to all patients. The rapid growth in computational biology has enticed substantial research attention from both data scientists and oncologists, leading to the development of rapid and cost-effective computer-aided cancer surveillance systems. Machine learning (ML), a subcategory of artificial intelligence, provides opportunities for drug discovery, early cancer diagnosis, effective treatment, and choice of treatment modalities. The application of ML approaches in EC diagnosis, therapies, and prognosis may be particularly relevant. Considering the significance of customized treatment and the growing trend of using ML approaches in cancer prediction and monitoring, a critical survey of ML utility in EC may provide impetus research in EC and assist oncologists, molecular biologists, biomedical engineers, and bioinformaticians to further collaborative research in EC. In this review, an overview of EC along with risk factors and diagnostic methods is discussed, followed by a comprehensive analysis of the potential ML modalities for prevention, screening, detection, and prognosis of EC patients.

Integrating additional factors into the TNM staging for cutaneous melanoma by machine learning

BACKGROUND

Integrating additional factors into the TNM staging system is needed for more accurate risk classification and survival prediction for patients with cutaneous melanoma. In the present study, we introduce machine learning as a novel tool that incorporates additional prognostic factors to improve the current TNM staging system.

METHODS AND FINDINGS

Cancer-specific survival data for cutaneous melanoma with at least a 5 years follow-up were extracted from the Surveillance, Epidemiology, and End Results Program of the National Cancer Institute and split into the training set (40,781 cases) and validation set (5,390 cases). Five factors were studied: the primary tumor (T), regional lymph nodes (N), distant metastasis (M), age (A), and sex (S). The Ensemble Algorithm for Clustering Cancer Data (EACCD) was applied to the training set to generate prognostic groups. Utilizing only T, N, and M, a basic prognostic system was built where patients were stratified into 10 prognostic groups with well-separated survival curves, similar to 10 AJCC stages. These 10 groups had a significantly higher accuracy in survival prediction than 10 stages (C-index = 0.7682 vs 0.7643; increase in C-index = 0.0039, 95% CI = (0.0032, 0.0047); p-value = 7.2×10-23). Nevertheless, a positive association remained between the EACCD grouping and the AJCC staging (Spearman's rank correlation coefficient = 0.8316; p-value = 4.5×10-13). With additional information from A and S, a more advanced prognostic system was established using the training data that stratified patients into 10 groups and further improved the prediction accuracy (C-index = 0.7865 vs 0.7643; increase in C-index = 0.0222, 95% CI = (0.0191, 0.0254); p-value = 8.8×10-43). Both internal validation using the training set and temporal validation using the validation set showed good stratification and a high predictive accuracy of the prognostic systems.

CONCLUSIONS

The EACCD allows additional factors to be integrated into the TNM to create a prognostic system that improves patient stratification and survival prediction for cutaneous melanoma. This integration separates favorable from unfavorable clinical outcomes for patients and improves both cohort selection for clinical trials and treatment management.

A prognostic system for epithelial ovarian carcinomas using machine learning

Introduction

Integrating additional factors into the International Federation of Gynecology and Obstetrics (FIGO) staging system is needed for accurate patient classification and survival prediction. In this study, we tested machine learning as a novel tool for incorporating additional prognostic parameters into the conventional FIGO staging system for stratifying patients with epithelial ovarian carcinomas and evaluating their survival.

Material and methods

Cancer‐specific survival data for epithelial ovarian carcinomas were extracted from the Surveillance, Epidemiology, and End Results (SEER) program. Two datasets were constructed based upon the year of diagnosis. Dataset 1 (39 514 cases) was limited to primary tumor (T), regional lymph nodes (N) and distant metastasis (M). Dataset 2 (25 291 cases) included additional parameters of age at diagnosis (A) and histologic type and grade (H). The Ensemble Algorithm for Clustering Cancer Data (EACCD) was applied to generate prognostic groups with depiction in dendrograms. C‐indices provided dendrogram cutoffs and comparisons of prediction accuracy.

Results

Dataset 1 was stratified into nine epithelial ovarian carcinoma prognostic groups, contrasting with 10 groups from FIGO methodology. The EACCD grouping had a slightly higher accuracy in survival prediction than FIGO staging (C‐index = 0.7391 vs 0.7371, increase in C‐index = 0.0020, 95% confidence interval [CI] 0.0012–0.0027, p = 1.8 × 10⁻⁷). Nevertheless, there remained a strong inter‐system association between EACCD and FIGO (rank correlation = 0.9480, p = 6.1 × 10⁻¹⁵). Analysis of Dataset 2 demonstrated that A and H could be smoothly integrated with the T, N and M criteria. Survival data were stratified into nine prognostic groups with an even higher prediction accuracy (C‐index = 0.7605) than when using only T, N and M.

Conclusions

EACCD was successfully applied to integrate A and H with T, N and M for stratification and survival prediction of epithelial ovarian carcinoma patients. Additional factors could be advantageously incorporated to test the prognostic impact of emerging diagnostic or therapeutic advances.

Expanding TNM for lung cancer through machine learning

Background

Expanding the tumor, lymph node, metastasis (TNM) staging system by accommodating new prognostic and predictive factors for cancer will improve patient stratification and survival prediction. Here, we introduce machine learning for incorporating additional prognostic factors into the conventional TNM for stratifying patients with lung cancer and evaluating survival.

Methods

Data were extracted from SEER. A total of 77 953 patients were analyzed using factors including primary tumor (T), regional lymph node (N), distant metastasis (M), age, and histology type. Ensemble algorithm for clustering cancer data (EACCD) and C‐index were applied to generate prognostic groups and expand the current staging system.

Results

With T, N, and M, EACCD stratified patients into 11 groups, resulting in a significantly higher accuracy in survival prediction than the 10 AJCC stages (C‐index = 0.7346 vs. 0.7247, increase in C‐index = 0.0099, 95% CI: 0.0091–0.0106, p‐value = 9.2 × 10⁻¹⁴⁷). There nevertheless remained a strong association between the EACCD grouping and AJCC staging (rank correlation = 0.9289; p‐value = 6.7 × 10⁻²²). A further analysis demonstrated that age and histological tumor could be integrated with the TNM. Data were stratified into 12 prognostic groups with an even higher prediction accuracy (C‐index = 0.7468 vs. 0.7247, increase in C‐index = 0.0221, 95% CI: 0.0212–0.0231, p‐value <5 × 10⁻³²⁴).

Conclusions

EACCD can be successfully applied to integrate additional factors with T, N, M for lung cancer patients.

Expanding the TNM for cancers of the colon and rectum using machine learning: a demonstration

Objective

The American Joint Committee on Cancer (AJCC) system for staging cancers of the colon and rectum includes depth of tumour penetration, number of positive lymph nodes and presence or absence of metastasis. Using machine learning, we demonstrate that these factors can be integrated with age, carcinoembryonic antigen (CEA) interpretation and tumour location, to form prognostic systems that expand the tumour, lymph node, metastasis (TNM) staging system.

Methods

Two datasets on colon and rectal cancers were extracted from the Surveillance, Epidemiology and End Results Programme of the National Cancer Institute. Dataset 1 included three factors (tumour, lymph nodes and metastasis). Dataset 2 contained six factors (tumour, lymph nodes, metastasis, age, CEA interpretation and tumour location). The Ensemble Algorithm for Clustering Cancer Data (EACCD) and the C-index were applied to generate prognostic groups.

Results

The EACCD prognostic system based on dataset 1 stratified patients into 10 risk groups, analogous to the 10 stages of the AJCC staging system. There was a strong inter-system association between EACCD grouping and AJCC staging (Spearman’s rank correlation=0.9046, p value=1.6×10⁻¹⁷). However, the EACCD system had a significantly higher survival prediction accuracy than the AJCC system (C-index=0.7802 and 0.7695, respectively for the EACCD system and AJCC system, p value=4.9×10⁻⁹¹). Adding age, or CEA interpretation, or location improved the prediction accuracy of the prognostic system-involving tumour, lymph nodes and metastasis. The EACCD prognostic system based on dataset 2 and all six factors stratified patients into 10 groups with the highest survival prediction accuracy (C-index=0.7914).

Conclusions

The EACCD can integrate multiple factors to stratify patients with colon or rectal cancer into risk groups that predict survival with a high accuracy.

Creating Prognostic Systems for Well-Differentiated Thyroid Cancer Using Machine Learning

Updates to staging models are needed to reflect a greater understanding of tumor behavior and clinical outcomes for well-differentiated thyroid carcinomas. We used a machine learning algorithm and disease-specific survival data of differentiated thyroid carcinoma from the Surveillance, Epidemiology, and End Results Program of the National Cancer Institute to integrate clinical factors to improve prognostic accuracy. The concordance statistic (C-index) was used to cut dendrograms resulting from the learning process to generate prognostic groups. We created one computational prognostic model (7 prognostic groups with C-index = 0.8583) based on tumor size (T), regional lymph nodes (N), status of distant metastasis (M), and age to mirror the contemporary American Joint Committee on Cancer (AJCC) staging system (C-index = 0.8387). We showed that adding histologic type (papillary and follicular) improved the survival prediction of the model. We also showed that 55 is the best cutoff of age in the model, consistent with the changes from the most recent 8th edition staging manual from AJCC. The demonstrated approach has the potential to create prognostic systems permitting data driven and real time analysis that can aid decision-making in patient management and prognostication.

Creating prognostic systems for cancer patients: A demonstration using breast cancer

Integrating additional prognostic factors into the tumor, lymph node, metastasis staging system improves the relative stratification of cancer patients and enhances the accuracy in planning their treatment options and predicting clinical outcomes. We describe a novel approach to build prognostic systems for cancer patients that can admit any number of prognostic factors. In the approach, an unsupervised learning algorithm was used to create dendrograms and the C‐index was used to cut dendrograms to generate prognostic groups. Breast cancer data from the Surveillance, Epidemiology, and End Results Program of the National Cancer Institute were used for demonstration. Two relative prognostic systems were created for breast cancer. One system (7 prognostic groups with C‐index = 0.7295) was based on tumor size, regional lymph nodes, and no distant metastasis. The other system (7 prognostic groups with C‐index = 0.7458) was based on tumor size, regional lymph nodes, no distant metastasis, grade, estrogen receptor, progesterone receptor, and age. The dendrograms showed a relationship between survival and prognostic factors. The proposed approach is able to create prognostic systems that have a good accuracy in survival prediction and provide a manageable number of prognostic groups. The prognostic systems have the potential to permit a thorough database analysis of all information relevant to decision‐making in patient management and prognosis.

Molecular Subtyping of Serous Ovarian Cancer Based on Multi-omics Data

Classification of ovarian cancer by morphologic features has a limited effect on serous ovarian cancer (SOC) treatment and prognosis. Here, we proposed a new system for SOC subtyping based on the molecular categories from the Cancer Genome Atlas project. We analyzed the DNA methylation, protein, microRNA, and gene expression of 1203 samples from 599 serous ovarian cancer patients. These samples were divided into nine subtypes based on RNA-seq data, and each subtype was found to be associated with the activation and/or suppression of the following four biological processes: immunoactivity, hormone metabolic, mesenchymal development and the MAPK signaling pathway. We also identified four DNA methylation, two protein expression, six microRNA sequencing and four pathway subtypes. By integrating the subtyping results across different omics platforms, we found that most RNA-seq subtypes overlapped with one or two subtypes from other omics data. Our study sheds light on the molecular mechanisms of SOC and provides a new perspective for the more accurate stratification of its subtypes.

An algorithm for expanding the TNM staging system

AIM

We describe a new method to expand the tumor, lymph node, metastasis (TNM) staging system using a clustering algorithm. Cases of breast cancer were used for demonstration.

MATERIALS & METHODS

An unsupervised ensemble-learning algorithm was used to create dendrograms. Cutting the dendrograms produced prognostic systems.

RESULTS

Prognostic systems contained groups of patients with similar outcomes. The prognostic systems based on tumor size and lymph node status recapitulated the general structure of the TNM for breast cancer. The prognostic systems based on tumor size, lymph node status, histologic grade and estrogen receptor status revealed a more detailed stratification of patients when grade and estrogen receptor status were added.

CONCLUSION

Prognostic systems from cutting the dendrogram have the potential to improve and expand the TNM.

A new cluster-based oversampling method for improving survival prediction of hepatocellular carcinoma patients

Liver cancer is the sixth most frequently diagnosed cancer and, particularly, Hepatocellular Carcinoma (HCC) represents more than 90% of primary liver cancers. Clinicians assess each patient's treatment on the basis of evidence-based medicine, which may not always apply to a specific patient, given the biological variability among individuals. Over the years, and for the particular case of Hepatocellular Carcinoma, some research studies have been developing strategies for assisting clinicians in decision making, using computational methods (e.g. machine learning techniques) to extract knowledge from the clinical data. However, these studies have some limitations that have not yet been addressed: some do not focus entirely on Hepatocellular Carcinoma patients, others have strict application boundaries, and none considers the heterogeneity between patients nor the presence of missing data, a common drawback in healthcare contexts. In this work, a real complex Hepatocellular Carcinoma database composed of heterogeneous clinical features is studied. We propose a new cluster-based oversampling approach robust to small and imbalanced datasets, which accounts for the heterogeneity of patients with Hepatocellular Carcinoma. The preprocessing procedures of this work are based on data imputation considering appropriate distance metrics for both heterogeneous and missing data (HEOM) and clustering studies to assess the underlying patient groups in the studied dataset (K-means). The final approach is applied in order to diminish the impact of underlying patient profiles with reduced sizes on survival prediction. It is based on K-means clustering and the SMOTE algorithm to build a representative dataset and use it as training example for different machine learning procedures (logistic regression and neural networks). The results are evaluated in terms of survival prediction and compared across baseline approaches that do not consider clustering and/or oversampling using the Friedman rank test. Our proposed methodology coupled with neural networks outperformed all others, suggesting an improvement over the classical approaches currently used in Hepatocellular Carcinoma prediction models.