Abstract Preclinical and clinical studies suggest that neuronal nicotinic receptor (NNR) agonists may be a novel and effective therapy for numerous painful conditions. Analgesic efficacy and safety of the highly selective α4β2 NNR agonist ABT-894 was evaluated in 2 separate randomized, double-blind, multicenter, placebo-controlled clinical trials in patients with diabetic peripheral neuropathic pain (DPNP). Study 1 (280 patients randomized) tested 1, 2, and 4mg ABT-894 twice daily compared with placebo and 60mg duloxetine once per day over 8weeks of treatment. Study 2 (124 patients randomized) tested 6mg ABT-894 twice daily vs placebo for 8weeks. The primary efficacy outcome measure in both studies was the weekly mean of the 24-hour average pain score recorded in each patient’s diary. In both trials, none of the ABT-894 dose groups showed efficacy compared with placebo, whereas duloxetine achieved a statistically significant improvement over placebo in Study 1. All dose levels of ABT-894 were well tolerated, and no significant safety issues were identified. These results are in contrast to the outcome of a previously reported study of DPNP using the less selective α4β2 NNR agonist ABT-594, which demonstrated efficacy compared with placebo, albeit with significant tolerability limitations. The failure of the highly selective α4β2 NNR agonist ABT-894 indicates that it may not be possible to define a therapeutic index for this mechanism or that selectively targeting α4β2 NNRs may not be a viable approach to treating neuropathic pain.