Effect of Surveillance Imaging in Relapse Detection on Survival of Patients with Follicular Lymphoma Open Access

Wang, Zhuowei (Spring 2020)

Permanent URL: https://etd.library.emory.edu/concern/etds/1n79h527b?locale=en


Background: Serving as an approach for early detection of tumor relapse, the routine implementation of surveillance imaging (SI) is still highly controversial due to its disadvantages like low specificity, huge costs and so on. Most importantly, previous studies did not show improvement in overall survival (OS) by SI. This study focused on patients with recrudescent follicular lymphoma (FL) and divided them into clinical detection group and SI detection group to examine whether SI results in better survival outcomes compared with traditional clinical detection methods.

Methods: We first conducted descriptive analysis on patients’ demographic and disease characteristics. Then survival analyses were performed starting by comparing survival probabilities and survival curves between two groups of patients using Kaplan-Meier method. After that, simple and multiple Cox proportional hazard models were built to investigate the association between detection methods and three survival outcomes. Supremum test and standardized score process plots were finally applied to check the proportional hazard assumption.

Results: In LEAD cohort, the hazard ratios of SI detection versus clinical detection were 1.71 (95% CI: 0.77, 3.80, p=0.187), 0.14 (95% CI: 0.10-3.36, p=0.222) and 0.28 (95% CI: 0.03-2.51, p=0.255) for PFS, OS from diagnosis and OS from relapse, respectively, in each multiple Cox model. In MER cohort, the values were 1.05 (95% CI: 0.71-1.55, p=0.806), 1.17 (95% CI: 0.53-2.59, p=0.691) and 1.10 (95% CI: 0.50-2.47, p=0.799), respectively. Although Cox models and hazard ratios are different in these two cohorts, similar results of no statistically significant association between detection method and survival outcomes were generated. 

Conclusion: The early detection of relapse by SI did not bring in much improvement in survival outcome for patients with FL in our study so the necessity of routine utilization of SI should be reconsidered. Further studies in prospective cohorts with more balanced patient and disease characteristics would be needed to validate these findings in FL and in other indolent lymphomas. 

Table of Contents

1. Introduction 1

2. Method 4

2.1 Study Population 4

2.2 Statistical Analysis Method 5

2.2.1 Descriptive analysis 5

2.2.2 Survival analysis 5

3. Result 7

3.1 Descriptive analysis 7

3.2 Survival analysis 8

3.2.1 LEAD cohort 8

3.2.2 MER cohort 9

4. Conclusion and Discussion 10

Tables and Figures 13

Reference 22

About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
Subfield / Discipline
  • English
Research Field
Committee Chair / Thesis Advisor
Committee Members
Last modified

Primary PDF

Supplemental Files