Economic Burden of Neurologic Toxicities Associated with Treating Relapsed Refractory Diffuse Large B-Cell Lymphoma in the United States

Introduction: Chimeric antigen receptor T (CAR-T) cell therapies targeting CD19 antigen can yield durable remissions in relapsed or refractory diffuse large B-cell lymphoma (r/r DLBCL). Yet the use of CAR-T can be limited by potentially severe toxicities, principally cytokine release syndrome (CRS) and neurologic toxicities. Management of neurologic toxicities requires vigilant monitoring, supportive treatment and resource allocation. We found no studies reporting healthcare costs associated with treatment-related neurologic adverse events (NEAEs) in r/r DLBCL patients. The objective of this study was to develop an evidence-based list of r/r DLBCL treatment-related NEAEs and to estimate the healthcare costs associated with these NEAEs in a real-world setting.

Methods:

Grade 3 or higher NEAEs that occurred in ≥2% of patients were identified by reviewing U.S. drug prescribing information (PI), European Medicines Agency summaries of product characteristics, and published clinical trials for treatments of r/r DLBCL. Then, adult patients ≥18 years old with r/r DLBCL were identified from a U.S. administrative claims database containing de-identified claims for over 150 million people across over 11 years. Patients were included if they had: 1) evidence of treatment beyond first-line (2L+) during the identification period (07/01/14 - 12/31/18); and 2) ≥1 inpatient or ≥2 outpatient claims for DLBCL (ICD-9-CM codes: 200.7X; ICD-10-CM codes: C83.3X) during the study period (01/01/14 - 12/31/18) with ≥1 having occurred prior to or on the date when the 2L+ treatment was received. 2L+ treatments were selected based on National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines and clinical expert input. To maximize the identification of patients treated with CAR-T, treatments were categorized hierarchically into 4 groups: CAR-T therapy (axicabtagene ciloleucel, or tisagenlecleucel), high-intensity cytotoxic therapy (carboplatin, cisplatin, cyclophosphamide, or ifosfamide), low-intensity cytotoxic therapy (bendamustine, lenalidomide, or gemcitabine), and targeted monotherapy (rituximab, ibrutinib, or brentuximab vedotin). The index date was defined as the start date of the 2L+ treatment. Patients in the cytotoxic and targeted therapy groups were also required to have evidence of an earlier line of cytotoxic or targeted therapy. All patients were required to have ≥6-months continuous enrollment before the index date. The outcomes of interest were the rates of NEAEs and total healthcare costs for patients with and without NEAEs during the 30-day post index period. Costs were inflated to 2018 US dollars. Descriptive statistics were reported.

Results: Twenty-three NEAEs were identified based on the review: 17 for CAR-T, 10 for conventional immunochemotherapy regimens, and 4 for both. In the claims database, a total of 349 adult patients with r/r DLBCL were identified, including 27 CAR-T therapy, 262 high-intensity cytotoxic therapy, 50 low-intensity cytotoxic therapy, and 10 targeted therapy users. Mean (median, SD) patient age was 72.3 (73; 10.3) years, with 47.9% being female and 83.4% having Medicare insurance. Patients were mainly from the South (44.1%) and the Midwest (31.5%). The mean (SD) Charlson comorbidity index was 4.4 (3.6) and mean (SD) number of chronic conditions was 7.3 (2.2). Forty-five (12.9%) patients had ≥1 NEAE at some point during the 30-day post-index period. Of these, 14 (31.1%) were CAR-T users. Eleven (40.7%) of the CAR-T users had encephalopathy. Mean total healthcare costs were $99,611 higher for patients with NEAEs [mean (SD): $153,435 (227,771)] than those without any NEAEs [$53,824 (96,170)]. Among patients with NEAEs, 72% of the healthcare costs were accrued in the inpatient setting. Among patients without NEAEs, 63% of the healthcare costs were for outpatient medical services. The trend of higher costs in patients with NEAEs was consistent across treatment groups.

Conclusion: This is the first study of the economic burden of NEAEs associated with treating r/r DLBCL in a real-world setting with data that reflects the current range of treatment options. Patients with r/r DLBCL who have NEAEs incur substantially higher costs than those without such events. In this analysis, CAR-T is overrepresented for NEAEs, although the sample size is small. We intend to repeat the analysis when more claims data becomes available in the near future.