An analysis of real-world administrative claims data in patients with chronic lymphocytic leukemia (CLL) has highlighted the need for treatments that result in fewer adverse events, according to data presented at ASH 2018.
The substantial incidence of adverse events among patients with CLL who received systemic anticancer therapy was associated with significant costs, mostly related to inpatient stays.
The most common regimen in the first-line setting was bendamustine (Bendeka) plus rituximab (Rituxan), in this patient population, and ibrutinib (Imbruvica) was most often used in the second and third lines of therapy.
“Adverse events are frequent and costly among patients with CLL, and healthcare costs increase with the number of adverse events observed during treatment episodes,” said Stacey Dacosta Byfield, PhD, MPH, Vice President, Research, Health Economics and Outcomes Research, Optum, Eden Prairie, MN. “The economic burden identified in this study underscores the need for treatments with fewer adverse events,” she added.
Approximately 21,000 patients will be diagnosed with CLL this year, according to Dr Byfield, and the disease will result in approximately 5000 deaths. Although CLL is the most common type of leukemia in adults, few studies have provided real-world current data on treatment patterns, treatment-related adverse events, and the economic burden associated with the various treatment regimens.
Dr Byfield and colleagues used medical and pharmacy claims data from the Optum Research Database, a large national database, to identify adults with CLL who received systemic therapy for CLL between July 2012 and May 2017. Patients who had a stem-cell transplant during that period were excluded from the analysis.
The researchers used timing and receipt of systemic CLL therapy to capture up to 3 lines of therapy and then created cohorts based on the most common treatment regimens. Potential treatment-related adverse events were identified by International Classification of Diseases (ICD), Ninth Revision and ICD-10 diagnosis codes. All-cause costs and adverse events–related resource utilization and costs per-patient per-month were analyzed.
The Cost of Adverse Events
A total of 3292 patients with CLL met the study criteria, including 65% Medicare Advantage enrollees and 35% with commercial insurance. Of these 3292 patients, 31% received ≥2 lines of treatment, and 10% received ≥3 lines of treatment. Bendamustine plus rituximab was the most common (23%) first-line regimen; ibrutinib was the most common second-line (22%) and third-line (17%) regimens.
Of the 4509 treatment episodes observed, 3177 involved the 5 most common treatment regimens: rituximab, excluding maintenance therapy (30%); bendamustine plus rituximab (28%); ibrutinib (20%); obinutuzumab (Gazyva), with or without chlorambucil (14%); and cyclophosphamide plus fludarabine and rituximab (8%).
The most common nonhematologic adverse events associated with these regimens were anemia (34%), hypertension (33%), infection (32%), thrombocytopenia (12%), neutropenia (12%), atrial fibrillation (11%), arthralgia or myalgia (10%), hemorrhage or bleeding (9%), and renal failure (7%).
The overall cost estimates for patients with CLL per-patient per-month were approximately $14,000 for treatment periods that had no adverse events. These costs trended higher when patients experienced adverse events.
Adverse events associated with the highest mean per-patient per-month costs included renal failure, $14,237 (standard deviation [SD], $60,043); stroke, $8037 (SD, $23,198); myocardial infarction, $6976 (SD, $10,206); and neutropenia, $6787 (SD, $28,627).
“Per-patient per-month costs associated with adverse events were substantial, and the majority of costs were associated with inpatient stays,” Dr Byfield emphasized.
The investigators noted the lack of clinical or prognostic variables that may influence treatment and outcomes as a study limitation.
“With claims data, there’s no causality with treatment and adverse events like we can abstract from medical records,” said Dr Byfield, adding that adverse events and associated costs might have been underestimated because of varying coding practices.
“Claims databases are great when it comes to utilization and cost but do not provide the clinical information that we need to determine what is really an adverse event,” Dr Byfield said.