Background: The present study aimed at identifying the risk factors associated with overall survival (OS) of breast cancer (BC) patients and projecting the most suitable accelerated failure time (AFT) model and semi-marko multistate models.

Method: This retrospective cohort study was conducted on 318 women with BC who were followed up for 10 years. Kaplan-Meier (KM) and log-rank statistics, Cox proportionality hazard (PH), Cox extended models (semi-parametric methods), and parametric PH and AFT models, such as exponential, Weibull, Generalized Gamma (GG), and log-logistic, were used to analyze the data, and the best AFT model was selected according to minimum Akaike information criteria (AIC). Stata software version 14.2 and SPSS 22 were used for data analysis (P <0.05). 

Results: The median age of the patients was 55 years and the median survival time was 29 months. Five- and 10-year OS were estimated at 67% and 32%, respectively. The results of multivariate analysis using the Cox extended model showed that lymph node involvement (HR = 2.21, P=0.038) and tumor size (T4) (HR=10.38, P=0.029) had significant effects on OS. Based on AIC, likelihood ratio (LR), and Cox-Snell residuals, the Weibull model was considered the AFT model showing the best fit into the BC data. 

Discussion: According to the Weibull AFT model, the estimated time ratio (TR) for prognostic factors indicated that women with lymph node involvement (LN+), positivity of human epidermal growth factor receptor 2 (HER2+), positivity of estrogen receptor (ER+), negative expression of progesterone receptor (PR-), advanced disease grade, and large tumor sizes were more likely to have a shorter survival compared to other women, suggesting that these factors can bring forward death. 

Professor Kourosh Sayehmiri is a highly regarded biostatistician who has made significant contributions to the field of medical research. He currently serves as the head of the Department of Biostatistics at Ilam University of Medical Sciences in Iran.

Sayehmiri received his Bachelor's degree in Statistics from Shahid Chamran University in 2004, followed by a Master's degree in Biostatistics from Tehran University of Medical Sciences, and a PhD in Biostatistics from the same university. He has also completed a three-month sabbatical research program at Nantes University in France.

Throughout his career, Sayehmiri has focused on developing and applying statistical methods in medical research, with an emphasis on meta-analysis, systematic review, survival analysis, clinical trials, and statistical modeling. He has published over 370 research papers, book chapters, and conference proceedings, and his work has been cited over 8772 times with an h-index of 45 on Google Scholar.

Sayehmiri is also an experienced teacher and mentor. He has taught courses in biostatistics, meta-analysis, and research methods at several universities in Iran and has supervised many graduate students in their research projects. He is also a reviewer and editorial board member for several scientific journals.

Sayehmiri's contributions to the field of biostatistics have been recognized through numerous awards and honors, including the Outstanding Researcher Award and the Top Researcher Award from Ilam University of Medical Sciences and the Distinguished Researcher Award from the Iranian Ministry of Health and Medical Education.

Today, Sayehmiri continues to conduct research, teach, and mentor students in the field of biostatistics. His work has had a significant impact on the field, and he is widely regarded as one of the leading experts in his area of expertise.