(356) Cumulative Plasma Exposure, Not Singular Concentrations, Yields a Best Correlation for Brivaracetam, Levetiracetam to Photoparoxysmal Abolition in Photosensitive Epilepsy: Results of Randomized, Double-blind, Crossover Study

Rationale: Brivaracetam-BRV has different physiochemical properties: higher potency, lipophilicity, 10-15x-greater SV2A affinity than Levetiracetam-LEV (Klitgaard, Epilepsia 2016; 57:538; Nicolas, Epilepsia 2016; 57:201), suggesting a potential for more rapid CNS effect.  BRV i.v. had a faster time (61% [95% CI= 9-84% faster, p= 0.039], median 5.5 min) to CNS peak effect (abolition of EEG photoparoxysmal response [PPR]) than an equi-potent dose of i.v. LEV in a previously conducted randomized, double-blind, 2-period crossover study in 9 patients with epilepsy (PwE) + photosensitivity  (Reed etal; AES 2019 Abst 1.302). Herein, we have retrospectively explored the relationship of AED plasma concentrations to time to PPR abolition.

Methods: Our protocol was IRB-approved and registered on ClinTrials.gov (NCT03580707). Study design details, including IPS parameters, have been stated previously (Reed etal); Part 1= 1500 mg LEV, 100 mg BRV post-15-min i.v.; Part 2= same AED doses at-5-min i.v. infusion. All patients had Hx GTCS, myoclonic seizures, or both. All serial blood samples for LEV, BRV were collected from an indwelling heparin-lock from each patient’s arm vein contralateral to i.v. BRV or LEV infusion; plasma was stored at > -20 degrees C.  Plasma BRV analysis was performed by NMS Labs, Horsham, PA; plasma LEV by ClinLab, St. Louis, MO; both used a proprietary LC-MS/MS method. The lower limit of quantification for BRV= 0.1 mcg/mL (10 ng/ml), and LEV= 1.0 mcg/mL; within- & between-day coefficient of variation (%CV) for a low concentration quality control BRV sample= 3.1 & 2.9%, respectively; LEV= 8% for both. All natural log (Ln) Area-Under-the-plasma-Concentration-time-curves (AUC, or exposure) were calculated using a linear trapezoidal method. Plasma correlation plots were constructed via least-squares multiple linear regression; associated p values were obtained via Excel.

Results: Nine PwE (6F; age= 27.8 [18-42]yr) participated in/completed the study; seven (4F) did both Parts 1&2. PWE showed PPR abolition (31/32 instances). A greater BRV:LEV difference in PPR abolition was observed for 15-min infusion vs. 5-min.  Greater variability in time to PPR abolition was seen in Part 2. At PPR abolition (n=16), median plasma [BRV]= 250 (range= 30-4,100) ng/ml; [LEV]= 28.35 [range 1-86.7] mcg/mL. For Part 1&2 separately, then combined, a very strong Pearson correlation coefficient (r) was observed when plotting time to PPR abolition versus Ln AUC for both LEV and BRV, whereas singular Ln plasma concentrations for either BRV or LEV did not correlate well with PPR abolition time, except for [BRV] in Part 1 (see Table).

Conclusions: Cumulative Ln plasma [BRV] and [LEV] concentration-exposure best correlates to abolition of the pharmacodynamic EEG biomarker of photosensitivity (PPR) in the majority of PWE. Singular Ln plasma concentrations do not correlate well.  Frequent plasma samples are needed when evaluating novel AED compounds in the ‘Photosensitivity Model of Epilepsy’ to calculate cumulative [AED] exposure.

Funding: Please list any funding that was received in support of this abstract.: UCB funded our investigator-initiated study (IIS).
Click here to view image/table