Compliance With Once-Daily Divalproex Extended-Release Tablets (Depakote-ER) Versus Multiple-Daily Dose Valproic Acid Capsules (Depakene) in Epilepsy:

x_ Prospective x__ Single-center __ Multi-center x__ Open-label __
Double-blind __ Single-blind x_ Randomized (please provide randomization ratio): 1:1
This population includes both patients whose seizures are relatively well- controlled on
their present conventional, enteric-coated, twice-daily or three-times daily Divalproex
sodium (Depakote®, [DR]) regimen. Patients on DR monotherapry are preferred, but not
required. Partial onset seizures and primarily generalized seizures will both be
represented in this study. Patients will be randomized in 1:1 fashion.

Group #1: 10 patients will be randomized to IR-VPA, to be taken 3 to 4 times a day, in a
total daily dose equivalent to DR. The dosing regimen will likely be taken at meal times
for those receiving IR-VPA tid (not q 8 h), and additionally at bedtime for those on a QID
(not q 6 h) regimen. The choice of TID vs QID regimen will be dictated by the patients’
total daily dose requirements, with the attempt to keep the number of 250 mg capsules
identical for each dose throughout the day.

Group #2: 10 patients will be randomized to ER, to be taken once-daily in the AM or PM at
the investigator’s discretion and patient’s choice, since dosing ER in the AM or PM does not
substantially differ as to plasma VPA concentrations [17]. Once AM or PM once-daily ER
dosing is chosen for a patient, it will not be changed. ER once-daily dosing will
incorporate the recommended dose-proportional increase of 8-20% in total daily dose over
that of DR for patients with epilepsy (18). However, only one tablet strength (500 mg ER)
will be utilized, so total daily dose will be rounded to the nearest 500 mg increment.

A total of 20 patients should provide sufficient power (80%) to adequately detect a
statistically (p < 0.05) and clinically meaningful change in compliance rate between the
groups, if one exists.

Compliance and precision will both be measure in this short-term study. Compliance, defined
as the number of times the multiple-daily IR-VPA dose or ER dose was actually taken vs that
prescribed daily, will be tallied with a computer-chip recording device (MEMS unit [13]) and
via the patient’s daily diary / calendar. Precision, defined as the actual time the daily
dose for ER was actually taken (or multiple-daily doses for IR-VPA were actually taken)
compared to the prescribed times ( + 15 min), as documented by the MEMS unit.

------------------------------------------------------------------------------

It will take 3 months to recruit #20 patients according to our protocol. The total protocol
can last from 49 to 70 days. There will be up to a 21-day screening period per patient,
followed by a 7-day period for the patient to practice the MEMS unit and gain familiarity
and demonstrate competency with the device. After randomization at Day 0, there will be 42
consecutive days of prospective evaluation.

From the end of study, the time to first abstract / manuscript submitted: 2 months
x__Abstract/manuscript/ poster assistance is anticipated

Washout period: N/A Screening Period: 0 to 21 days prior to baseline observation. Baseline
observation period: 7 days prior to randomization. This period allows the patient to
practice with the MEMS unit, gaining familiarity with it and demonstrating competency with
the device.

Treatment period: The treatment period is forty-two (# 42) days after Randomization.

A rapid, immediate conversion from multiple-daily dosed DR to once-daily divalproex-ER has
been recommended for adults with epilepsy since it is a simple maneuver to understand with a
low predicted chance for side effects (19). Likewise, patients randomized to the IR-VPA
group will be immediately converted from DR to multiple-daily dose IR-VPA without tapering
the DR or gradual titration upward of IR-VPA.

Taper period: N/A

Age: adults (16 and older) Gender: male/female Disease under study: Patients with a
chronic, yet relatively stable partial and /or generalized seizure disorder will be included
in this study.

Twenty patients currently currently undergoing treatment with enteric-coated divalproex [DR]
for whom the treating physician elects to change to once-daily divalproex-ER will be
candidates for this study. Ten patients will be assigned to each group randomly, which
should yield sufficient power to make a definite statement regarding the best conversion
strategy to use in this patient population.

XX__ Depakote® Tablets (for Lead-in) XX__ Depakote® ER

- No Drug Required __ Depakote® Sprinkle

- Depacon® XX_Depakene® Capsules Lead-in: Patients with epilepsy who are already being
treated with conventional, enteric-coated, delayed-release Depakote ([DR]). Patients
on DR monotherpay are preferred, but patients may be taking additional AEDs.

ACTIVE: Depakote-ER. For those randomized to this group, an 8-20% mg dose increase will be
utilized in DR to ER conversion, as recommended by the Divalproex-ER package insert (Abbott)
and based upon published evidence for the need to compensate for the lesser bioavailability
of ER. Only 500 mg ER tablets will be used, such that total daily dose will be rounded to
the nearest 500 mg.

ACTIVE COMPARATOR: Depakene Capsules. Depakene is a liquid-filled capsule containing 250 mg
IR-VPA. The same total daily dose of IR-VPA as DR will be utilized. IR-VPA will be given
TID (not q 8 h) or QID (not q 6 h), depending upon which regimen allows for an identical
number of capsules to be given for each dose.

Note: Compliance and precision will only be recorded for either ER or IR-VPA, not for other
AEDS, in those patients who happen to be taking concomitant AEDs.

1) Patients, age 16 and above, currently taking Divalproex-DR for any seizure disorder; 2)
Other AEDs are permitted concurrently, although compliance with these will not be recorded.
Other medications for co-morbid disease are permitted, provided no plans for changes in
medications used for the treatment of the concomitant disorder are expected.

3) Patients must demonstrate a 75% or greater compliance rate with DR via calendar during
the week of familiarity with the MEMs unit. The threshold value of 75% has been chosen
since research shows that people take approximately 75% of their AED(s) as prescribed
(13,14), and the same numerical value is frequently used in determining whether or not to
retain a patient in clinical Phase 2a-3b industry-sponsored study.

1. patients with a recent history of status epilepticus;

2. patients who have refractory or unstable epilepsy;

3. patients with acute illnesses requiring changes in concurrent drugs;

4. patients unwilling to change from their present DR regimen to divalproex-ER or IR-VPA.

5. Patients unwilling or unable to utilize the MEMs monitoring unit;

6. Pregnant or lactating women.

IRB approval will be sought. After the appropriate IRB approval, written, informed consent
will be obtained for all patients participating.

Patients in either Group will be instructed to take their ER or IR-VPA according to the PI’s
usual practice, without different or special coaching or prompting techniques to enhance
compliance throughout the study.

Patients will be observed closely by physicians and allied health personnel on a weekly
basis, by actual visit to the study center (Days 0, +14, +28 & +42 and via phone on Days +7,
+21 & +35. Blood work, including CBC, platelet counts, LFTs, serum chemistry panel and
total valproate will be measured prior to and at the end of the study on Day +42
post-randomization.

Criteria for premature discontinuation for safety reasons:

1. Patients who pass screening but who do not demonstrate proficiency or competence with
the MEMs monitoring unit for the 7-day period prior to randomization;

2. Patients with a 50% increase in seizure frequency compared to their historic baseline,
or the development of seizures requiring hospitalization at any time during the study;

3. Patients with a clinically significant increase in the number or severity of AEs, as
determined by the investigator;

4. Patients with a MEMs (if data is immediately accessible and available) or
calendar-documented compliance rate of < 35% at any time during the study.

5. Patients with a MEMs (if data is immediately accessible and available) or
calendar-documented compliance rate between > 35% and < 75% on three separate
occasions or two consecutive occasions during the study.

Primary Measures:

The primary efficacy parameter will be the difference in mean compliance rate (defined as
the # does taken divided by the # doses prescribed) for the ER vs IR-VPA group at Day + 42.

Secondary Measures:

1) The change in compliance rate from Day 0 to Day +42 will be compared for both groups; 2)
The precision of dose administration, determined by the number of doses administered within
+ 15 min of their assigned times will be compared for all dose times over the course of the
study between the two groups; 3) Drop out rates for both groups, due to lack of seizure
control, adverse effects or noncompliance, will be compared between groups.

x_Statistical analysis assistance may be required

None foreseen. Patients selected for this study will only be those who, upon
recommendations or orders from their physician, will be switched from multiple-daily dose of
divalproex-DR to the divalproex extended-release formulation, administered once-daily or to
IR-VPA.

This project is considered “research” by virtue of the randomization of epilepsy patients to
two different groups, utilizing dosage formulations and regimens commonly encountered in
clinical practice.