PURPOSE: To determine the analgesic effect of the addition of gabapentin to opioids in the management of neuropathic cancer pain. PATIENTS AND METHODS: One hundred twenty-one consecutive patients with neuropathic pain due to cancer, partially controlled with systemic opioids, participated in a multicenter, randomized, double-blind, placebo-controlled, parallel-design, 10-day trial from August 1999 to May 2002. Gabapentin was titrated from 600 mg/d to 1,800 mg/d in addition to stable opioid dose. Extra opioid doses were available as needed. Zero to 10 numerical scale was used to rate average daily pain. The average pain score over the whole follow-up period was used as main outcome measure. Secondary outcome measures were: intensity of burning pain, shooting/lancinating pain, dysesthesias (also scored on 0 to 10 numerical scale), number of daily episodes of lancinating pain, presence of allodynia, and daily extra doses of opioid analgesics. RESULTS: Overall, 79 patients received gabapentin and 58 (73%) completed the study; 41 patients received placebo and 31 (76%) completed the study. Analysis of covariance (ANCOVA) on the intent-to-treat population showed a significant difference of average pain intensity between gabapentin (pain score, 4.6) and placebo group (pain score, 5.4; P =.0250). Among secondary outcome measures, dysesthesia score showed a statistically significant difference (P =.0077; ANCOVA on modified intent-to-treat population = 115 patients with at least 3 days of pain assessments). Reasons for withdrawing patients from the trial were adverse events in six patients (7.6%) receiving gabapentin and in three patients receiving placebo (7.3%). CONCLUSION: Gabapentin is effective in improving analgesia in patients with neuropathic cancer pain already treated with opioids.