Intrathecal Recombinant Human Hepatocyte Growth Factor for Acute Cervical Spinal Cord Injury with AIS Grade A at 72 h Post-Injury: Phase III Study of Safety and Efficacy

The acute management of traumatic spinal cord injury (SCI) continues to lack a universally accepted pharmacological intervention. We developed a novel therapeutic strategy, intrathecal administration of pharmaceutical recombinant human hepatocyte growth factor (KP-100), based on translational research using a non-human primate model of cervical SCI. A Phase I/II trial (multi-center, randomized, double-blind study including subjects with cervical SCI with modified Frankel grade A/B1/B2 at 72 h post-injury) demonstrated the safety and efficacy of intrathecal KP-100. A greater proportion of Frankel grade A subjects achieved ≥1 point improvement on their lower-extremity motor score (33.3% [KP-100] vs. 6.3% [placebo]). This study aimed to confirm these findings by including only subjects with the American Spinal Injury Association (ASIA) impairment scale (AIS) grade A. This open-label, non-randomized, single-group interventional Phase III study enrolled subjects with AIS grade A at 72 h post-injury. KP-100 was administered intrathecally immediately after enrollment. Subsequent doses were given once weekly, for a total of five administrations. Subjects were followed up for 168 days after the first administration. The primary end-point was the proportion of subjects who demonstrated improvement from AIS grade A at baseline to C or higher at day 168, compared with subjects with AIS grade A at 72 h post-injury in the General Spinal Cord Injury Center Data Bank in Japan (n = 81, the DB group). Additionally, we utilized data from modified Frankel grade A subjects in the PI/II study (n = 16, the PI/II Placebo group; n = 15, the PI/II KP-100 group) for post hoc analyses of therapeutic effects. Of the 31 pre-registered participants, 6 were excluded due to ineligibility, and 25 were included in the efficacy analysis (the PIII group). The primary end-point was not achieved (8.6% [DB, 95% confidence interval [CI]: 4.2–16.8] vs. 12.0% [PIII, 95% CI 4.2–30.0]). However, in post hoc exploratory comparisons, the PIII group tended to have greater proportion of subjects who improved to AIS grade B or higher (56.0% [37.1–73.3]) than the DB group (19.8% [12.5–29.7]). Whereas SCIs without fracture at the C3/4 level were more common in the PI/II KP-100 group, the PIII group tended to have more SCIs with fractures or dislocations and below C3/4. Accordingly, in exploratory comparisons, the proportion of subjects who improved to modified Frankel grade C1 or higher tended to be highest in the PI/II KP-100 group (6.3% in the PI/II Placebo, 26.7% in the PI/II KP-100, 12.0% in the PIII group). Conversely, ≥5 points of improvement on the upper-extremity motor score tended to be most frequent in the PIII group (31.3% in the PI/II Placebo, 20.0% in the PI/II KP-100, 56.0% in the PIII group). These results showed exploratory signals of neurological recovery in this highly impaired population, providing supportive evidence for potential biological activity of KP-100.