A phase 3, randomized, double-blind, sham-controlled trial of SI-6603 (condoliase) in patients with radicular leg pain associated with lumbar disc herniation.

Background Context: SI-6603 (condoliase) is a chemonucleolytic agent approved in Japan in 2018 for the treatment of lumbar disc herniation (LDH) associated with radicular leg pain. Condoliase, a mucopolysaccharidase with high substrate specificity for glycosaminoglycans (GAGs), offers a unique mechanism of action through the degradation of GAGs in the nucleus pulposus. As LDH management is currently limited to conservative approaches and surgical intervention, condoliase could offer a less invasive treatment option than surgery for patients with LDH.

Purpose: The Discovery 6603 study (NCT03607838) evaluated the efficacy and safety of a single-dose injection of SI-6603 (condoliase) vs sham for the treatment of radicular leg pain associated with LDH.

Study Design/setting: A randomized, double-blind, sham-controlled, phase 3 study conducted across 41 sites in the United States.

Patient Sample: Male and female participants (N=352; aged 30-70 years) with contained posterolateral LDH and unilateral radiculopathy/radicular leg pain for greater than 6 weeks.

Outcome Measures: The primary endpoint was the change from baseline (CFB) in average worst leg pain score at 13 weeks, assessed using the 100-mm visual analogue scale. Key secondary endpoints were CFB in average worst leg pain score at 52 weeks, herniation volume at 13 weeks, and Oswestry Disability Index (ODI) score at 13 weeks. Safety evaluations included adverse events (AEs) and imaging findings.

Methods: Participants were randomized 1:1 to receive a single intradiscal injection of condoliase (1.25 units) or sham injection followed by 52 weeks of observation. The primary and key secondary endpoints were assessed using a mixed model for repeated measures (MMRM) analysis and a protocol-specified multiple imputation (MI) sensitivity analysis on the modified intention-to-treat (mITT) population. A prespecified serial gatekeeping algorithm was used for multiple comparisons. Safety endpoints included AEs, laboratory tests, vital signs, imaging (by X-ray and magnetic resonance imaging [MRI]), and occurrence of posttreatment lumbar surgery.

Results: Of the 352 randomized participants, 341 constituted the mITT population (condoliase n=169; sham n=172) and the safety population (condoliase n=167; sham n=174). For the primary endpoint, the condoliase group showed significantly greater improvement in CFB in worst leg pain at Week 13 (least squares mean [LSM] CFB: -41.7) compared with sham injection (-34.2; LSM difference: -7.5; 95% confidence interval [CI]: -14.1, -0.9; p=.0263) based on the MMRM analysis. CFB in worst leg pain at Week 52 favored condoliase vs sham, but the difference was not statistically significant (p=.0558), which halted the serial gatekeeping testing algorithm and dictated that the CFB in herniation volume and ODI scores at Week 13 would be considered nonsignificant, regardless of their p-values. Treatment group differences in CFB in herniation volume and ODI score favored the condoliase group vs sham at all timepoints. The MI sensitivity analysis showed differences in CFB in worst leg pain at Week 13 (p=.0223) and Week 52 (p=.0433) in favor of the condoliase group. Treatment-emergent AEs (TEAEs) were more common in the condoliase group (≥1 TEAE: 71.9%; ≥1 treatment-related TEAE: 28.1%) compared with the sham group (≥1 TEAE: 60.3%; ≥1 treatment-related TEAE: 10.3%). Of the TEAEs, spinal MRI abnormalities and back pain occurred most frequently. No treatment-related serious AEs occurred.

Conclusions: Condoliase met its primary endpoint of significantly improving radicular leg pain at Week 13 and was generally well tolerated in patients with LDH. Chemonucleolysis with condoliase has the potential to provide a less invasive treatment option than surgery for those unresponsive to conservative treatment strategies.