A double-blind placebo-controlled clinical trial of recombinant human brain-derived neurotrophic factor (rhBDNF) in diabetic polyneuropathy

A randomized, double-blind, placebo-controlled study of brain-derived neurotrophic factor (rhBDNF) was conducted in 30 patients with insulin-treated diabetes mellitus, with obligatory abnormalities of sural nerve conduction studies and vibration perception threshold (VPT) at the great toe on recruitment. Nine patients received placebo, 11 rhBDNF (25 mug/kg) and 10 rhBDNF (100 mug/kg) s.c. daily for 3 months, and were assessed at days 0, 8, 15, 29, 43, 57 and 85 with nerve conduction and quantitative sensory and autonomic tests including VPT, thermal and light touch thresholds, and cutaneous axon-reflexes. No statistically significant differences were found among the 3 treatment groups between baseline and day 85 values. To examine possible reasons for lack of effect, post hoc analysis was performed. In the subset of patients with abnormal but detectable cool detection threshold (CDT) at baseline, there was improvement of CDT at day 85 when compared to baseline in the treated (p < 0.02) but not placebo group, Further, from days 43 to 85, in the treated group but not the placebo group, CDT was indistinguishable from a group of matched normal subjects (p > 0.05). Skin biopsies failed to show evidence of structural change; assessment of innervation of hair follicles, which is partly dependent on BDNF, was not possible because of the marked loss of this end-organ in diabetic neuropathic skin. The only side effects of rhBDNF were infrequent non-painful injection-site skin reactions and increased gut motility at the higher dose. We conclude that further preclinical studies are warranted before any future clinical trials to see if rhBDNF improves CDT and constipation in diabetics.