von Delius, S., Eckel, F., Wagenpfeil, S., Mayr, M., Stock, K., Kullmann, F., . . . Lersch, C. (2007). Carbamazepine for prevention of oxaliplatin-related neurotoxicity in patients with advanced colorectal cancer: Final results of a randomised, controlled, multicenter phase II study. Investigational New Drugs, 25, 173–180.

This study evaluated the efficacy and safety of carbamazepine for the prevention of oxaliplatin-associated neuropathy in patients treated for advanced-stage colorectal cancer.

Patients were randomly assigned to receive carbamazepine 200 mg orally daily, beginning six days before the first oxaliplatin via IV, with dose increases to yield a plasma level of 4-6 mg/dl. All patients received oxaliplatin 85 mg/m² via IV every two weeks, plus folinic acid 500 mg/m² via IV followed by 5-FU 2,000 mg/m² infused over 24 hours every week. A cycle consisted of six weeks of treatment and a two week rest. Groups were compared in relation to worst neurotoxicity and neurotoxicity at each cycle.

• The study consisted of 36 total patients, with 19 assigned to the carbamazepine group and 17 to the control group.
• To be included, patients had confirmed advanced-stage colorectal cancer, were 18 years of age or older, had a World Health Organization performance status of 0 or 1, and had an anticipated life expectancy of at least three months. Laboratory criteria also needed to be met.
• Patients were excluded if they had preexisting neurologic diseases, had been treated with a drug that could interact with carbamazepine, had prior treatment with oxaliplatin, and were currently being treated with gabapentin, vitamin B, magnesium, or calcium.

The study had a randomized, controlled, multicenter phase II design.

Data were collected at baseline and following each treatment cycle using Levi’s Neurotoxicity Rating Scale (0–4), and  peripheral neuropathy score based on sensory symptoms and examination (vibrational sense, strength, and deep tendon reflexes), each scored 0–3. 

No group differences were noted in grade of neurotoxicity or in grade 3 or 4 neurotoxicity using the Levi Neurotoxicity Rating Scale for either the worst toxicity across all cycles or cycle-based comparisons. The groups did not differ in scores on the individual components of the peripheral neuropathy score or the overall peripheral neuropathy score based on worst toxicity or cycle-specific comparisons between groups. Of note, only two participants discontinued  carbamazepine for CNS side effects.

Because the current study was underpowered, no definitive conclusions can be drawn regarding efficacy and safety.

• Regarding limitations, the final sample size was not adequate enough to achieve the power needed to detect differences between groups.
• No data on reliability or validity for the Levi Neurotoxicity Rating Scale were provided.
• Reliability and validity of the two approaches used to grade or score severity of neuropathy was not discussed.
• Adherence to carbamazepine was not assessed but the plasma levels suggest that adherence was acceptable during the early phases of the study.
• The results are inconsistent with a nonrandomized pilot study conducted by the same group.