Background: Preclinical studies have demonstrated that nerve growth factor may prevent or reverse peripheral neuropathy. We have therefore tested the effects of recombinant human nerve growth factor in patients with diabetic polyneuropathy.

Methods: A total of 250 patients with symptomatic diabetic polyneuropathy randomly received either placebo or one of two doses of recombinant human nerve growth factor for 6 months. Patients were assessed for symptoms and signs of polyneuropathy before and after treatment.

Results: Compared with placebo, recombinant human nerve growth factor led to significant improvement after 6 months of treatment, as measured by the sensory component of the neurologic examination, two quantitative sensory tests, and the impression of most subjects that their neuropathy had improved. Three prospectively identified multiple endpoint analyses indicated improvements in the nerve growth factor treatment groups over the placebo group in all three analyses (p = 0.032; p = 0.008;p = 0.005). Recombinant human nerve growth factor was well tolerated, with injection site discomfort reported as the most frequent adverse event.

Conclusions: Recombinant human nerve growth factor appears to be safe and shows preliminary evidence of efficacy in patients with symptomatic diabetic polyneuropathy.