RESUMO
BACKGROUND: This article is aimed to make predictions in terms of prognostic factors and compare prediction methods by using Cox proportional hazards regression analysis (CPH), some machine learning techniques and Accelerated Failure Time (AFT) model for post-treatment survival probabilities according to clinical presentations and pathological information of early-stage breast cancer patients. MATERIAL AND METHODS: The study was carried out in three stages. In the first stage, the CPH method was applied. In the second stage, the AFT model and in the last stage, machine learning methods were applied. The data set consists of 697 breast cancer patients who applied to Marmara University Hospital oncology clinic between 01.01.1994 and 31.12.2009. The models obtained by using various parameters of the patients were compared according to the C index, 5-year survival rate and 10-year survival rate. RESULTS AND CONCLUSION: According to the models obtained as a result of the analyses applied, MetLN and age were obtained as a significant risk factor as a result of CPH method and AFT methods, while MetLN, age, tumor size, LV1 and extracapsular involvement were obtained as risk factors in machine learning methods. In addition, when the c-index values of the handheld models are examined, it is obtained as 69.8 for the CPH model, 70.36 for the AFT model, 72.1 for the random survival forest and 72.8 for the gradient boosting machine. In conclusion, the study highlights the potential of comparing conventional statistical methods and machine-learning algorithms to improve the precision of risk factor determination in early-stage breast cancer prognosis. Additionally, efforts should be made to enhance the interpretability of machine-learning models, ensuring that the results obtained can be effectively communicated and utilized by clinical practitioners. This would enable more informed decision-making and personalized care in the treatment and follow-up processes for early-stage breast cancer patients.
RESUMO
Optimal scoring system for clinical prognostic factors in patients with unresectable hepatocellular carcinoma (HCC) is currently uncertain. We aimed to develop and externally validate an easy to use tool, particularly for this population, and named it the "unresectable hepatocellular carcinoma prognostic index" (UHPI). We evaluated the data of patients with treatment-naive unresectable HCC who were diagnosed in the training center from 2010 to 2019 (n = 209). A simple prognostic model was developed by assigning points for each covariate in proportion to the beta coefficients in the Cox multivariable model. Predictive performance and distinction ability of the UHPI were further evaluated in an independent European validation cohort (n = 147) and compared with 11 other available models. A simple scoring system was derived, assigning 0.5/1/2 scores for six independent covariates including, the Child-Pugh score, Eastern Cooperative Oncology Group performance status, maximum tumor size, vascular invasion or extrahepatic metastasis, lymph node involvement, and alpha-fetoprotein. The UHPI score, ranging from 0 to 6, showed superior performance in prognosis prediction and outperformed 11 other staging or prognostic models, giving the highest homogeneity (c-index, 6-month and 1-year area under the receiver operator characteristic curves), lowest Akaike information criterion, and -2 log-likelihood ratio values. The UHPI score allocated well the risk of patients with unresectable HCC for mortality within the first year, using two cut-off values (low-risk, <0.5; intermediate-risk, 0.5-2; high-risk, >2). Conclusion: The UHPI score can predict prognosis better than other systems in subjects with unresectable HCC and can be used in clinical practice or trials to estimate the 6-month and 1-year survival probabilities for this group.