A Case Study on Modeling the Time to Recurrence of Gastric Cancer Patients

Area of Science:

• Oncology
• Biostatistics
• Epidemiology

Background:

• Gastric cancer is a leading global cause of mortality.
• Understanding recurrence patterns is crucial for patient management.

Purpose of the Study:

• To model the time to first recurrence in gastric cancer patients.
• To identify significant prognostic factors influencing gastric cancer relapse.

Main Methods:

• Utilized data from gastric cancer patients at Tikur Anbesa Specialized Hospital (2013-2020).
• Employed Weibull, log-logistic, and lognormal baseline hazard functions with gamma and inverse Gaussian frailty distributions.
• Statistical analysis performed using R software.

Main Results:

• Median recurrence time was 23.96 months; 61.2% of patients experienced first recurrence.
• The clustering effect was significant in modeling recurrence time.
• The log-logistic inverse Gaussian frailty model identified sex, tumor size, smoking, treatment, vascular invasion, disease stage, H. pylori infection, and histological type as significant prognostic factors.

Conclusions:

• The Inverse Gaussian frailty model effectively describes gastric cancer recurrence.
• Key prognostic factors include gender, tumor size, treatment, vascular invasion, disease stage, H. pylori infection, and histological type.
• Findings necessitate targeted interventions to improve patient outcomes and prevent relapse based on identified risk factors.