Evaluation of Prognostic Variables on Time to Relapse In Patients with Hodgkin's Lymphoma Using Classified Cox Model

Background and Objectives: Identifying the important influential factors of relapse in Hodgkin's Lymphoma is a great challenge in oncology studies. Cox regression model is a known method that could be used to evaluate the prognostic factors. This method identifies the main prognostic factors based on Proportional Hazard (PH) assumption. The aim of this study was to evaluate the stratified Cox model and assess the prognostic factors which affected the relapse time of Hodgkin's lymphoma in patients hospitalized in Ahvaz Shafa Hospital.

Material and Methods: The current study included all the cases of patients who had referred to the research center of oncology and hematology of Shafa hospital in Ahvaz Jundishapure University of medical sciences during 2002-2012. Survival time was considered from date of diagnosis to relapse (in months) after chemotherapy. Choosing the best model was based on Likelihood Ratio Test. Data analysis was performed using STATA13.

Results: The results of the stratified Cox regression model showed that the risk of relapse is 2 times higher for patients with hemoglobin lower than 10.5 (P=0.012). Furthermore, the patients that NLPHL was used as a biopsy method were 96 percent more likely to experience the relapse compared to CHL biopsy method (P = 0.155).

Conclusion: Using the stratified Cox regression model the hemoglobin less than 10.5 was identified as a main and significant prognostic factor.