Impact of Different Definitions of Vaso-Occlusion on Efficacy Assessments in Sickle Cell Disease Clinical Trials

Abstract

Introduction

Patients with sickle cell disease (SCD) experience recurrent, severe pain events due to vaso-occlusion. Eliminating these acute pain events is a key outcome in SCD clinical trials; however, the definition of a vaso-occlusive crisis (VOC) or a vaso-occlusive event (VOE) has not been consistently applied, hampering comparisons of treatment efficacy between different therapeutic approaches. We have examined the degree to which differing definitions of vaso-occlusion in clinical trial endpoints impact efficacy outcomes.

Methods

Descriptions of clinical endpoints related to vaso-occlusion and pain events were reviewed from trials of exagamglogene autotemcel (exa-cel), lovotibeglogene autotemcel (lovo-cel), renizgamglogene autogedtemcel (reni-cel), hydroxyurea, l-glutamine, voxelotor, and crizanlizumab. Patient-level data from the published exa-cel Phase 3 pivotal trial (CLIMB SCD-121; data cut 14 Jun 2023) was used to evaluate efficacy outcomes based on differing endpoint definitions of vaso-occlusion.

Results

In the seven clinical trials reviewed, definitions of vaso-occlusion and/or pain events varied by care setting, duration of care, treatments used, and associated complications, with the frequency and duration of medical facility visits for acute pain events being most dissimilar between trials. Definitions of severe VOCs (exa-cel), VOC (voxelotor), and sickle cell-related pain crises (SCPCs; crizanlizumab and l-glutamine) included pain events requiring a medical facility visit of any duration, whereas the definition of painful crises (hydroxyurea) required a medical facility visit of > 4 h and the definition of severe vaso-occlusive events (VOEs; lovo-cel and reni-cel) required a hospital or emergency room (ER) observation unit visit lasting ≥ 24 h or ≥ 2 visits to a day unit or ER over a 72-h period. Based on the definition of severe VOCs, 29/30 patients [96.7%; 95% confidence interval (CI): 82.8, 99.9] in the CLIMB SCD-121 trial were considered free from severe VOCs for ≥ 12 consecutive months, whereas when the severe VOEs definition was applied to the same data, all patients (30/30; 100.0%; 95% CI: 88.4, 100.0) were considered free from severe VOEs for ≥ 12 consecutive months.

Conclusion

Differences exist in definitions of vaso-occlusion and pain events used in SCD clinical trials. Severe VOCs (exa-cel), VOC (voxelotor), and SCPCs (crizanlizumab and l-glutamine) were more broadly inclusive than severe VOEs (lovo-cel and reni-cel) or painful crisis (hydroxyurea). Clinically, these differences resulted in differing numbers of patients being considered free from vaso-occlusion pain events, underscoring the challenge in comparing frequencies of pain events across SCD clinical trials.