Racial and Socioeconomic Disparities Among Patients with Chronic Lymphocytic Leukemia (CLL) Treated in the US Community Oncology Setting

Background: Since the first targeted therapy for CLL was approved by the Food and Drug Administration in 2014, new therapies and clinical practices have improved the care for patients with CLL. There is increased attention to racial disparities in healthcare.

Objectives: To describe characteristics and outcomes of patients with CLL and assess whether there is equitable benefit from contemporary CLL therapies across racial and socioeconomic groups by analyzing real-world data collected in a large community oncology setting in the US.

Methods: Patients with CLL treated in the US Oncology Network (The Network) from January 2015 through June 2023 were included. Data were sourced from structured data fields in iKnowMed™ (iKM), an oncology-specific electronic health record (EHR) system that captures outpatient practice encounters, including data on demographic and relevant clinical information for nearly 40% of US community oncology practices. iKM records are updated daily, and deterministically linked to supplementary Datavant mortality data and social determinants of health (SDoH) EHR data. Kaplan-Meier methods, univariate and multivariable cox regression were used to analyze overall survival (OS) and to determine the adjusted hazard ratio (HR) of OS. Considering that estimating the OS from initial diagnosis date is subject to immortal time bias, we performed the following two analyses: (1) time from the first visit date at a Network practice on or after the diagnosis date to OS, and (2) time from the initial diagnosis date among those newly diagnosed patients. In the primary analysis of OS, patients without a death date were censored at their last visit at The Network. In a sensitivity analysis, all follow-up data were censored at July 1, 2023.

Results: The study included 12,253 CLL patients across all 4 census regions in the US, 60% were male. Of the 10,521 (86%) patients with race reported, 91.0% were white, 5.1% Black, and 3.9% were Other race. The median (Q1, Q3) age was 69 (61, 76) with Black patients 1.2 years younger than White patients (median: 69.0 versus 67.8, P=0.004). The initial CLL diagnosis was available for 11,576 (94%) patients and the majority (6,874; 59%) either had their initial diagnosis in The Network or had their first visit in The Network within 30 days after their initial diagnosis elsewhere. The median time from initial diagnosis to the first visit in The Network was slightly shorter for Black patients (2.4 weeks) as compared to White (3.9 weeks, p=0.001). At the initial diagnosis, more Black patients had Rai Stage III or IV than White patients (54% versus 46%, P=0.002). Death dates were available for 2,231 (18%) of patients. The analysis of time from first visit to death showed that the 8-year death rate was significantly higher for Black patients (61%) than White patients (43%, P=0.002) with statistically significant HR (95% confidence interval [CI]) of 1.34 (1.11, 1.61), P=0.002 in the univariate analysis. The median (95% CI) OS was 7.5 (6.4, not estimable) years for Black patients. The median OS was not estimable for White patients for this study because the study was designed to use data in the last 8 years and the median survival for White patients was longer than 8 years. After adjusting for education level and income, the race effect was no longer significant 1.25 (0.76, 2.05), P=0.37. The wider CI was due to SDoH data being only available for 2,928 (24%) of patients. Even with reduced statistical power, income of <$30k was significantly associated with higher risk of a shorter time to death with HR (95% CI) of 1.78 (1.32, 2.41), P<0.001 as compared to >$75k in the adjusted model. The analysis of time from the initial diagnosis date to OS among those newly diagnosed patients and the sensitivity analyses showed similar trends.

Conclusion: This study identifies racial disparities in this modern era of CLL therapy using real-world data from a large community setting in the US. Black race and low income are prognostic of less favorable OS in CLL. Further research is needed to determine whether racial and income disparities in CLL are due to differences in access to therapy, quality of care, disease biology, comorbidities or other factors. This research may contribute to developing interventions to ensure that progress in CLL therapy benefits all patients.