(C-243) Real-World Evidence of the Impact of Voxelotor on Care Utilization for Patients With Sickle Cell Disease

Background: Voxelotor is an oral therapy for patients with sickle cell disease (SCD) that inhibits sickle hemoglobin polymerization. In September 2024, Pfizer withdrew voxelotor for SCD treatment based on clinical data indicating the benefit of the drug no longer outweighed the risk. Pfizer specifically cited a potentially different risk in vaso-occulsive crises (VOC). Prior published real-world data analyses suggested reduced VOC utilization among voxelotor patients.

Objectives: To determine the impact of voxelotor on SCD-related care utilization for SCD patients using complete, longitudinal health service utilization data.

Methods: This observational study was conducted on 15,266 de-identified SCD patients with utilization history from 2019 to 2023 using the Komodo Research Dataset, which contains medical and pharmacy benefit data for about 88% of the US patient population. The study includes a new-user design estimating the observational analog of the intention-to-treat effect. Patients were eligible if they were aged ≥12 years, had ≥1 VOC event in the year prior to their index date, and were continuously enrolled. Initiators were new users of voxelotor and non-users were patients who never initiated voxelotor.

Propensity score matching was used to balance the treatment and comparison groups based on demographic information and number of VOC events in the year prior to the index date. We estimated incidence rate ratios (IRR) for care utilization metrics using a negative binomial model with an offset to account for differences in individual observed follow-up time.

Results: After matching, each group had 1,307 patients. For almost all outcomes, patients initiating voxelotor had an increase in utilization compared to non-initiators.

For inpatient measures, the incidence rate for the treatment group was 32% (IRR = 1.32 [95%CI: 1.17,1.50]) higher for inpatient VOCs, 93% (1.93 [1.60,2.33]) higher for inpatient visits for acute chest syndrome (ACS), and 22% (1.22 [1.08,1.38]) higher for all-cause hospitalizations.

For ED measures, no statistically significant difference was found for ED VOC visits (1.09 [0.96,1.24]), however the treatment group had a 53% (1.53 [1.20,1.95]) higher incidence rate for ED ACS visits. Lastly, the incidence rate of SCD-related outpatient visits was 47% (1.47 [1.35,1.60]) higher among voxelotor initiators.

Conclusions: This study found higher care utilization rates among voxelotor patients, particularly for inpatient VOC events, prior to the market withdrawal of voxelotor. The findings demonstrate that leveraging comprehensive and complete real-world data is important when evaluating post-market therapy impact.