Ross, J.R., Goller, K., Hardy, J., Riley, J., Broadley, K., A’hern, R., & Williams, J. (2005). Gabapentin is effective in the treatment of cancer-related neuropathic pain: A prospective, open-label study. Journal of Palliative Medicine, 8, 1118–1126.

To evaluate the effectiveness of gabapentin for cancer-related neuropathic pain

Gabapentin was administered at a dose of 300 mg/day up to 1.8 g/day over 15 days. Final doses were taken in three divided doses. Two parallel groups were recruited with either treatment-related (n = 25) or tumor-related (n = 37) neuropathic pain. Gabapentin was dose-escalated from 300 mg/day to 1.8 g/day. Median dose was 1,200 mg/day. Patient selection was nonrandomized.

• The study reported on 62 patients: 16 males and 46 females, 21 of whom had had breast cancer.
• A total of 41 patients completed to day 15.
• Patients were excluded from the study if they had started treatment that was likely to affect their neuropathic pain, such as anticonvulsants (< one month), antidepressants or steroids (< seven days), and anesthetic interventions, such as nerve blocks (< two weeks).

Patients were selected from inpatient and outpatient settings in a U.K. tertiary referral center.

Prospective, open-label studies of inpatients and outpatients

Patients completed pain and side-effect scores and were evaluated by a clinician on days 0, 8, and 15. In addition, pain scores were obtained on day 4. Pain was assessed using the modified Brief Pain Inventory (BPI) to measure pain on a 0–10 scale. Patient descriptors of pain and scores of activities of daily living (0–10 scale) were collated together with demographic data. Toxicity scores ranged from “not at all” to “a little,” “quite a bit,” or “very much.”

Baseline and final pain scores were not significantly different between the tumor-related and treatment-related groups. In addition, ANOVA did not show etiology to be a significant predictor of response to gabapentin. Data from both groups were pooled into the primary analysis. A total of 41 patients completed the study, with the median-range dose of gabapentin to be 1,200 mg/day. A total of 28 of 62 patients achieved at least a one-third reduction in their pain score (95% CI); the number needed to treat to obtain this benefit was 2.2. There was a significant reduction in all scores measuring the impact of pain on daily living (p < 0.003). ANOVA suggested that gender and cancer diagnosis, but not etiology of neuropathic pain or type of neuropathic pain, were independent predictors of change in average pain score. There was suggestion that beneficial effect was only in women, not men; men with bowel or lung cancer showed no benefit. This may be a chance subgroup finding. There was significant reduction in worst, average, and current pain scores (p < 0.002), but not in least pain scores. The tumor group described pain as shooting, pins and needles, and the treatment group described pain as hot, burning sensations. There was no significant difference in pain scores on day 8 compared to day 15. No further pain relief was reported from day 8 to day 15—if improvement is not seen in one week, this suggests that an alternative drug should be started. Six patients (10%) discontinued study because of probable drug-related side effects, 14% reported mild/moderate side effects, and 45% were already taking steroids or antidepressants, although doses did not change during the study.

Gabapentin is an effective treatment for cancer-related neuropathic pain from both tumor and treatment.

• The study time period was short and therefore may not have given enough time for patients to gain confidence and reduce the amount of opioid use. Further, patients were not randomized, so the placebo effect is not clear.
• Breakthrough medications were available, but their use was not documented.
• Women had a better response, which may be a chance subgroup finding or related to diagnosis (21 of 46 female patients had breast cancer).
• Gabapentin was not used prophylactically.
• Dose escalation was limited by side effects. Side effects were not clear; 10% left study due to side effects, and 14% had mild/moderate side effects.
• Of 62 patients, 41 completed study to day 15.
• Patients in the tumor group were more likely to be on steroids, and patients in the treatment group were more likely to be on antidepressants.
• This was a single-site study; there was no comparison between methods employed at different radiation centers.

Side effects included drowsiness, unsteadiness, dizziness, nausea, headache, and others. There was no change in opioids during study—one might question whether there could there be a synergistic effect between gabapentin and opioids. Studies need to be performed to determine if the escalation of a dose greater than 1,800 mg/day is beneficial between men and women and different cancer diagnoses.