As previously reported, the initial Phase 3 clinical trial (DPN 3-1, N=500 subjects, 9 months) did not meet its primary efficacy endpoint, but its double-blind placebo-controlled 3-month extension study (DPN 3-1B, a subset of N=101 subjects) met its primary endpoint (12 months long-term safety) and key secondary endpoint (analgesic efficacy at Day 365).
Last September, Helixmith reported PK anomalies observed in its VM202 DPN Phase 3 trial, indicating the possibility of mix-up of materials between placebo and active groups. The investigation has now been completed and root causes have been identified. The investigation revealed that there was no placebo-drug mix-up among the study subjects. Because there was no evidence of drug misdosing or randomization errors, the company was now able to validate the previous data and derive following conclusions:
(This Phase 3 trial was composed of two programs, DPN 3-1 (9 months, N = 500 subjects) and DPN 3-1B (12 months, a subset of N = 101 subjects). The latter was the extended safety trial conducted in a double-blind, placebo-controlled, manner, that was aimed to gather both 12-month safety and efficacy data)
1. Safety of Engensis was confirmed in this large-scale Phase 3 trial, in both the 9 month (DPN 3-1) and 12 month (DPN 3-1B) studies. The remarkable safety of Engensis (VM202) from this trial is consistent with data from Phase 1 and Phase 2 studies done for painful DPN and also from other clinical studies for critical limb ischemia (CLI), coronary artery disease (CAD), and amyotrophic lateral sclerosis (ALS).
2. The 9-month DPN 3-1 study did not meet its primary endpoint (pain reduction at 3 months), showing no statistically significant difference between placebo and active groups.
3. However, the extended trial, DPN 3-1B (12 months, N = 101), clearly demonstrated both VM202’s safety and efficacy.
(1) Safety: The occurrence of adverse events (AEs) was no different between the VM202 and placebo groups; the Treatment-Emergent AE occurrence rate was lower in the VM202 group than in the placebo group.
(2) Efficacy (Pain reduction): In the intent-to-treat (ITT) population (N=101), VM202 showed clinically meaningful and statistically significant pain reductions vs. placebo at months 6, 9, and 12. The pain reduction differences (delta) between the two arms at months 6, 9, and 12 were –1.1, –0.9, and –0.9, respectively with the p-values being <0.01 or <0.05.
(3) Improvement in clinical status: The Patient Global Impression of Change (PGIC) scale at 12 months was significantly better for VM202 group as many more subjects responded with “very much improved” efficacy vs. placebo. The p-value was <0.01.
(4) Greater efficacy in subjects not taking DPN medication gabapentin and pregabalin: The difference in pain reduction between the VM202 and placebo groups was greater in the population not on gabapentinoid medication (N=53 total). At 6, 9, and 12 months, the pain reduction differences vs. placebo were –1.3, –1.2, and –1.5. The p-values were <0.05.
(5) Regeneration potential: The last follow-up visit of this study occurred more than 8 months from the last injection of VM2020. The fact that the VM202 group sustains pain reductions strongly suggests that VM202 may have nerve regeneration properties.
Understanding the difference in efficacy between DPN 3-1 and DPN 3-1B:
Subjects in the DPN 3-1 and DPN 3-1B trials shared almost identical demographics and baseline characteristics, but there were clear differences in efficacy between the N=500 subjects enrolled in the full DPN 3-1 study and the subset of 101 subjects enrolled in the DPN 3-1B study. Results from additional post-hoc statistical analyses indicated that there was a noticeable difference between subjects enrolled in the first half and the second half of the trial. The company has identified possible causes, and the improvements will be implemented in the next Phase 3 trial.
The company plans two or three independent Phase 3 trials for painful diabetic peripheral neuropathy (DPN). One of these trials will be aimed to demonstrate the neuronal regenerative capacity of VM202.