Exploring factors related to metastasis free survival in breast cancer patients using Bayesian cure models

Factors Affecting Long-Survival of Patients with Breast Cancer by Non-Mixture and Mixture Cure Models Using the Weibull, Log-logistic and Dagum Distributions: A Bayesian Approach

BACKGROUND

Breast cancer is a top biomedical research priority, and it is a major health problem. Therefore, the present study aimed to determine the prognostic factors of breast cancer survival using cure models.

METHODS

In this retrospective cohort analytic study, data of 140 breast cancer patients were collected from Ali Ibn Abi Taleb hospital, Rafsanjan, Southeastern Iran. Since in this study, a part of the population had long-term survival, cure models were used and evaluated using DIC index. The data were analyzed using Openbugs Software.

RESULTS

In this study, of 140 breast cancer patients, 23 (16.4%) cases died of breast cancer. Based on the findings, the Bayesian nonmixture cure model, with type I Dagum distribution, was the best fitted model. The variables of BMI, number of children, number of natural deliveries, tumor size, metastasis, consumption of canned food, tobacco use, and breastfeeding affected patients' survival based on type I Dagum distribution.

CONCLUSION

The results of the present study demonstrated that the Bayesian nonmixture cure model, with type I Dagum distribution, can be a good model to determine factors affecting the survival of patients when there is the possibility of a fraction of cure. In this study, it was found that adapting a healthy lifestyle (eg, avoiding canned foods and smoking) can improve the survival of breast cancer patients.

Application of a non-parametric non-mixture cure rate model for analyzing the survival of patients with colorectal cancer in Iran

OBJECTIVES

Colorectal cancer (CRC) patients are considered to have been cured when the mortality rate of individuals with the disease returns to the same level as expected in the general population. This study aimed to assess the impact of various risk factors on the cure fraction of CRC patients using a real dataset of Iranian CRC patients with a non-mixture non-parametric cure model.

METHODS

This study was conducted on the medical records of 512 patients who were definitively diagnosed with CRC at Taleghani Hospital, Tehran, Iran from 2001 to 2007. A non-mixture non-parametric cure rate model was applied to the data after using stepwise selection to identify the risk factors of CRC.

RESULTS

For non-cured cases, the mean survival time was 1,243.83 days (95% confidence interval [CI], 1,174.65 to 1,313.00) and the median survival time was 1,493.00 days (95% CI, 1,398.67 to 1,587.33). The 1- and 3-year survival rates were 92.9% (95% CI, 91.0 to 95.0) and 73.4% (95% CI, 68.0 to 79.0), respectively. Pathologic stage T1 of the primary tumor (estimate=0.58; p=0.013), a poorly differentiated tumor (estimate=1.17; p<0.001), a body mass index (BMI) between 18.6 and 24.9 kg/m² (estimate=−0.60; p=0.04), and a BMI between 25.0 and 29.9 kg/m² (estimate=−1.43; p<0.001) had significant impacts on the cure fraction of CRC in the multivariate analysis. The proportion of cured patients was 64.1% (95% CI, 56.7 to 72.4).

CONCLUSIONS

This study found that the pathologic stage of the primary tumor, tumor grade, and BMI were potential risk factors that had an impact on the cure fraction. A non-mixture non-parametric cure rate model provides a flexible framework for accurately determining the impact of risk factors on the long-term survival of patients with CRC.

Long-Term Disease-Free Survival of Non-Metastatic Breast Cancer Patients in Iran: A Survival Model with Competing Risks Taking Cure Fraction and Frailty into Account

Introduction: Survival modeling is a very important tool to detect risk factors and provide a basis for health care planning. However, cancer data may have properties leading to distorted results with routine methods. Therefore, this study aimed to cover specific factors (competing risk, cure fraction and heterogeneity) with a real dataset of Iranian breast cancer patients using a competing risk-cure-frailty model. Materials and methods: For this historical cohort study, information for 550 Iranian breast cancer patients who underwent surgery for tumor removal from 2001 to 2007 and were followed up to March 2017, was analyzed using R 3.2 software. Results: In contrast to T-stage and N-stage, hormone receptor status did not have any significant effect on the cure fraction (long-term disease-free survival). However, T-stage, N-stage and hormone receptor status all had a significant effect on short-term disease-free survival so that the hazard of loco-regional relapse or distant metastasis in cases positive for a hormone receptor was only 0.3 times that for their negative hormone receptor counterparts. The likelihood of locoregional relapse in the first quartile of follow up was nearly twice that of other quartiles. The least cumulative incidence of time to locoregional relapse was for cases with a positive hormone receptor, low N stage and low T stage. The effect of frailty term was significant in this study and a model with frailty appeared more appropriate than a model without, based on the Akaike information criterion (AIC); values for the frailty model and one without the frailty parameter were 1370.39 and 1381.46, respectively. Conclusions: The data from this study indicate ae necessity to consider competing risk, cure fraction and heterogeneity in survival modeling. The competing risk-cure-frailty model can cover complex situations with survival data.

A Hybrid Computer-aided-diagnosis System for Prediction of Breast Cancer Recurrence (HPBCR) Using Optimized Ensemble Learning

Cancer is a collection of diseases that involves growing abnormal cells with the potential to invade or spread to the body. Breast cancer is the second leading cause of cancer death among women. A method for 5-year breast cancer recurrence prediction is presented in this manuscript. Clinicopathologic characteristics of 579 breast cancer patients (recurrence prevalence of 19.3%) were analyzed and discriminative features were selected using statistical feature selection methods. They were further refined by Particle Swarm Optimization (PSO) as the inputs of the classification system with ensemble learning (Bagged Decision Tree: BDT). The proper combination of selected categorical features and also the weight (importance) of the selected interval-measurement-scale features were identified by the PSO algorithm. The performance of HPBCR (hybrid predictor of breast cancer recurrence) was assessed using the holdout and 4-fold cross-validation. Three other classifiers namely as supported vector machines, DT, and multilayer perceptron neural network were used for comparison. The selected features were diagnosis age, tumor size, lymph node involvement ratio, number of involved axillary lymph nodes, progesterone receptor expression, having hormone therapy and type of surgery. The minimum sensitivity, specificity, precision and accuracy of HPBCR were 77%, 93%, 95% and 85%, respectively in the entire cross-validation folds and the hold-out test fold. HPBCR outperformed the other tested classifiers. It showed excellent agreement with the gold standard (i.e. the oncologist opinion after blood tumor marker and imaging tests, and tissue biopsy). This algorithm is thus a promising online tool for the prediction of breast cancer recurrence.

A Comparison between Cure Model and Recursive Partitioning: A Retrospective Cohort Study of Iranian Female with Breast Cancer

Background. Breast cancer which is the most common cause of women cancer death has an increasing incidence and mortality rates in Iran. A proper modeling would correctly detect the factors' effect on breast cancer, which may be the basis of health care planning. Therefore, this study aimed to practically develop two recently introduced statistical models in order to compare them as the survival prediction tools for breast cancer patients. Materials and Methods. For this retrospective cohort study, the 18-year follow-up information of 539 breast cancer patients was analyzed by “Parametric Mixture Cure Model” and “Model-Based Recursive Partitioning.” Furthermore, a simulation study was carried out to compare the performance of mentioned models for different situations. Results. “Model-Based Recursive Partitioning” was able to present a better description of dataset and provided a fine separation of individuals with different risk levels. Additionally the results of simulation study confirmed the superiority of this recursive partitioning for nonlinear model structures. Conclusion. “Model-Based Recursive Partitioning” seems to be a potential instrument for processing complex mixture cure models. Therefore, applying this model is recommended for long-term survival patients.

Application of a Non-Mixture Cure Rate Model for Analyzing Survival of Patients with Breast Cancer

BACKGROUND

As a result of significant progress made in treatment of many types of cancers during the last few decades, there have been an increased number of patients who do not experience mortality. We refer to these observations as cure or immune and models for survival data which include cure fraction are known as cure rate models or long-term survival models.

MATERIALS AND METHODS

In this study we used the data collected from 438 female patients with breast cancer registered in the Cancer Research Center in Shahid Beheshti University of Medical Sciences, Tehran, Iran. The patients had been diagnosed from 1992 to 2012 and were followed up until October 2014. We had to exclude some because of incomplete information. Phone calls were made to confirm whether the patients were still alive or not. Deaths due to breast cancer were regarded as failure. To identify clinical, pathological, and biological characteristics of patients that might have had an effect on survival of the patients we used a non-mixture cure rate model; in addition, a Weibull distribution was proposed for the survival time. Analyses were performed using STATA version 14. The significance level was set at P ≤ 0.05.

RESULTS

A total of 75 patients (17.1%) died due to breast cancer during the study, up to the last follow-up. Numbers of metastatic lymph nodes and histologic grade were significant factors. The cure fraction was estimated to be 58%.

CONCLUSIONS

When a cure fraction is not available, the analysis will be changed to standard approaches of survival analysis; however when the data indicate that the cure fraction is available, we suggest analysis of survival data via cure models.