A Data-Driven Approach to Define Parsimonious Eligibility Criteria in First-Line Clinical Trials for Diffuse Large B-Cell Lymphoma 公开

Abstract

Background: Diffuse large B-cell lymphoma (DLBCL) is clinically and genetically heterogeneous. Forty percent of patients relapse or are refractory to first-line therapy and have inferior outcomes, indicating unmet treatment needs for high-risk disease. Randomized controlled trials (RCTs) designed to improve outcomes for these high-risk groups have been largely unsuccessful, potentially due to restrictive eligibility criteria that in fact limit enrollment of high-risk patients. We define evidence-based methods to streamline enrollment of patients with high-risk disease for first-line DLBCL clinical trials incorporating novel therapeutics. Methods: We enumerated enrollment criteria from 19 first-line DLBCL RCTs. We proposed eligibility criteria for four eligibility criterion categories: International Prognostic Index (IPI) score, age at diagnosis, Eastern Cooperative Oncology Group (ECOG) performance status (PS), and Ann Arbor stage. Using study-specific eligibility criteria and proposed criteria, we identified eligible patients in eight DLBCL data sets representing institutional, regional, and national populations. We performed survival analysis according to eligibility status to determine whether prior RCTs and proposed criteria targeted high-risk groups. We calculated sensitivity and specificity of combinations of proposed criteria to identify patients meeting eligibility criteria for prior studies and developed receiver operating characteristic plots to identify optimal combinations. We characterized the mutational profile of the eligible patient population. Results: We identified 52 eligibility criterion categories across 19 trials. We proposed the inclusion criteria IPI score ≥ 2, age at diagnosis ≥ 18, ECOG PS 0–2, and stage II–IV. Proposed criteria risk-stratified patients with hazard ratios for eligible versus ineligible patients of 1.37–3.58 for overall survival across data sets and defined eligibility for high-risk subgroups. Subsets of the proposed criteria lacking full IPI factors identified patients who were eligible for prior RCTs with sensitivity ≥ 0.75 for at least 14 of 19 RCTs when using data sets containing data for all data types included in analysis. We described patterns of DLBCL mutations for high-risk, eligible patients. Conclusion: Subsets of modernized eligibility criteria for first-line DLBCL RCTs identified high-risk, eligible patient groups with high sensitivity. We identified relationships between eligibility status and cohort genomics that facilitate precision medicine RCT design for DLBCL.

Table of Contents

Introduction 1

Background 2

Methods 9

Results 18

Discussion 28

References 36

Tables 42

Table 1. Patient demographics and baseline disease characteristics from all data sets 42

Table 2. Randomized controlled trials included in analysis (n = 19) 43

Table 3. Criterion categories in 19 diffuse large B-cell lymphoma randomized controlled trials 44

Figures 45

Figure 1. Eligibility criteria trends for IPI score, age at diagnosis, ECOG PS, and Ann Arbor stage in DLBCL RCTs spanning the R-CHOP era 45

Figure 2. Study eligibility in the Emory University and Schmitz et al. DLBCL cohorts using study-specific and proposed criteria 46

Figure 3. Study specific HRs for OS and PFS comparing eligible and ineligible groups across 19 DLBCL RCTs and using proposed criteria 47

Figure 4. Cox proportional hazards results comparing eligible groups from 19 DLBCL RCTs with eligible groups defined using proposed criteria 48

Figure 5. ROC plot illustrating the capacity of combinations of proposed criteria to identify patients eligible for the PHOENIX trial in the Schmitz et al. cohort 49

Figure 6. Impact of eligibility criteria on genetic alteration prevalence in the Reddy et al. data set 50

Figure 7. Impact of eligibility criteria on genetic alteration prevalence in the Chapuy et al. data set 51

Figure 8. Impact of eligibility criteria on genetic alteration prevalence for patients in genetic Cluster 1 in the Chapuy et al. data set 52

Figure 9. Impact of eligibility criteria on genetic alteration prevalence for patients in genetic Cluster 2 in the Chapuy et al. data set 53

Figure 10. Impact of eligibility criteria on genetic alteration prevalence for patients in genetic Cluster 3 in the Chapuy et al. data set 54

Figure 11. Impact of eligibility criteria on genetic alteration prevalence for patients in genetic Cluster 4 in the Chapuy et al. data set 55

Figure 12. Impact of eligibility criteria on genetic alteration prevalence for patients in genetic Cluster 5 in the Chapuy et al. data set 56

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.

School	Laney Graduate School
Department	Clinical Research
Degree	M.S.
Submission	Master's Thesis
Language	English
Research Field	Health Sciences, Oncology
关键词	Diffuse large B-cell lymphoma Eligibility criteria Clinical trials
Committee Chair / Thesis Advisor	Christopher Flowers, MD, MS, FASCO, MD Anderson Cancer Center
Committee Members	Amita Manatunga, PhD, Emory University Jean Koff, MD, MS, Emory University

最新修改

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	A Data-Driven Approach to Define Parsimonious Eligibility Criteria in First-Line Clinical Trials for Diffuse Large B-Cell Lymphoma ()	2020-04-25 20:24:59 -0400	Press to Select an action Download

Supplemental Files

Thumbnail	Title	Date Uploaded	Actions