Related Experiment Video
Updated: Dec 11, 2025

Competing-Risk Nomogram for Predicting Cancer-Specific Survival in Multiple Primary Colorectal Cancer Patients after Surgery
Published on: September 27, 2024
Risk of relapse and death from colorectal cancer and its related factors using non-Markovian Multi-State model
Saeideh Hajebi Khaniki1, Vahid Fakoor2, Soodabeh Shahid Sales3
1Student Research Committee, Department of Biostatistics, School of health, Mashhad University of Medical Sciences, Mashhad, Iran.
Aim:
This study aimed at modeling the risk of local relapse and death from colorectal cancer after the first treatment and its related factors using multi-state models.
Background:
In cancer studies modeling the course of disease regarding events which happen to patients is of great importance. By considering death as the final endpoint while incorporating the intermediate events, multi-state models have been developed.
Methods:
This was a historical cohort study in which 235 patients with colorectal cancer, who referred to Omid Hospital in Mashhad between 2006 and 2011, were studied and followed up until 2017. The transition probabilities to death due to metastasis with or without experiencing local relapse and variables related to them were determined using the non-Markovian multi-state model in three states of disease, local relapse and death.
Results:
The probability of not experiencing either of the events, just relapse and death in the first 5 years were 0.45, 0.09 and 0.46 respectively. If patients did not experience any event in the first year of treatment, the probability of relapse and death before the fifth year were 0.04 and 0.33 respectively and if they did experience relapse during this time, the probability of death by the fifth year was 0.62. The stage of cancer was associated with relapse and death, while ethnicity and history of addiction were related to death without relapse and BMI had a significant relationship with death after relapse (p<0.05).
Conclusion:
Risk of death in patients with colorectal cancer depends on local relapse and the time between them.
Related Concept Videos
Assumptions of Survival Analysis
Cancer Survival Analysis
Comparing the Survival Analysis of Two or More Groups
Kaplan-Meier Approach
Actuarial Approach
Consider the example of a high-risk surgical procedure with significant early-stage mortality. A two-year clinical study is conducted,...
Parametric Survival Analysis: Weibull and Exponential Methods
Weibull Distribution
The Weibull distribution is a flexible model used in parametric survival analysis. It can handle both increasing and decreasing hazard rates, depending on its shape parameter...

