Abstracts

RATIONALE: VPA is often regarded as being relatively benign in terms of adverse cognitive effects, whereas the high incidence of cognitive complaints in early clinical studies with TPM suggested a potentially greater negative impact on cognitive function. A randomized, double-blind, placebo-controlled trial compared the effects of TPM and VPA as add-on therapy to CBZ in adults with partial-onset seizures, using neuropsychometric tests to objectively evaluate effects on cognitive function. Initial analyses showed a modest but statistically significant negative effect of TPM vs. VPA in 2/30 tests after 20 wks of treatment. Additional analyses examined the distribution of change scores in patient subsets to further delineate the pattern of sensitivity to drug-induced cognitive effects.
METHODS: Patients with [gte]3 seizures during a 4-week baseline were randomized to TPM 400 mg/day, VPA 2250 mg/day, or placebo added to CBZ. 23 neuropsychometric tests and 7 mood tests were conducted at baseline, end of titration (8 wks), and study end (20 wks).
RESULTS: 76 adults (17-66 yrs; 57% women) received TPM (N=34; mean dose, 395 mg/day), VPA (N=29; 1928 mg/day), or placebo (N=13) added to CBZ (1020-1200 mg/day). Cognitive complaints with TPM and VPA were similar, although slightly more TPM-treated patients reported psychomotor slowing (TPM, n=3; VPA, n=1), mood problems (TPM, 3; VPA, 0), speech (eg, word finding) difficulty (TPM, 4; VPA, 2), and confusion (TPM, 2; VPA, 1). Of the 30 tests, 6 (7%) showed significantly (P[lte]0.05) greater negative baseline-to-titration changes for TPM vs. VPA. However, after 20 wks, only symbol digit modalities (SDM) and controlled oral word association (COWAT) were significantly different for TPM vs. VPA. These differences could be accounted for by a small subset of TPM-treated patients in whom scores deteriorated [gt]1 standard deviation (SD) from baseline. In a post-hoc analysis, mean Z-score changes were calculated for the remaining 28 tests in TPM- and VPA-treated patients with [gt]1 SD on the SDM or COWAT (SDM/COWAT Sensitive) and [lt]1 SD change (SDM/COWAT Insensitive). With TPM, mean Z-score change was -0.01 for SDM/COWAT Sensitive subset and +0.10 for SDM/COWAT Insensitive; with VPA, mean Z-score change was -0.34 vs. +0.19, respectively (positive Z-score = [dsquote]improvement[dsquote]).
CONCLUSIONS: In this study, a subset of patients were more sensitive to drug effects from both TPM and VPA. VPA-treated SDM/COWAT Sensitive patients tended to have negative effects in other tests. In TPM-treated SDM/COWAT Sensitive patients, negative effects were limited to SDM and COWAT. With TPM or VPA, SDM/COWAT Insensitive patients tended to improve in other tests.
Support: Ortho-McNeil Pharmaceutical
Disclosure: Salary - Hulihan, Kamin, Karim- Ortho-McNeil Pharmaceutical; Grant - Meador- clinical investigations- Ortho-McNeil Pharmaceutical; Stock - Hulihan, Kamin, Karim- Ortho-McNeil Pharmaceutical; Honoraria - Meador- Ortho-McNeil Pharmaceutical