Cox regression models have been used for biomarker prognostic prediction for years. But due to model over-fitting and the improper way of model development and various other factors, very few published prognostic signature has found its clinical success in applications. This study is aimed to minimize these issues by applying an iterated four-step K-fold cross-validated (CV) Cox model building and testing procedure for triple negative breast cancer (TNBC) patients based on their microRNA profiling. The four steps include: feature selection, model selection, addition of significant clinical covariates, and model performance assessment. At each step, various methods were compared. Model performance was assessed on the test data by cross-validated AUC of 3-year or 5-year recurrence and ROC calculated using the time-dependent ROC method. The optimal procedure suggested the best-performing predictive model based on the final biomarker signature.