Intracerebral delivery of antiseizure medications by microinvasive neural implants

Abstract

Focal epilepsy is a difficult disease to treat as two-thirds of patients will not respond to oral anti-seizure medications (ASMs) or have severe off-target effects that lead to drug discontinuation. Current non-pharmaceutical treatment methods (resection or ablation) are underutilized due to the associated morbidities, invasive nature and inaccessibility of seizure foci. Less invasive non-ablative modalities may potentially offer an alternative. Targeting the seizure focus in this way may avoid unassociated critical brain structures to preserve function and alleviate seizure burden. Here we report use of an implantable, miniaturized neural drug delivery system [microinvasive neural implant infusion platform (MINI)] to administer ASMs directly to the seizure focus in a mouse model of temporal lobe epilepsy. We examined the effect local delivery of phenobarbital and valproate had on focal seizures, as well as adverse effects, and compared this to systemic delivery. We show that local delivery of phenobarbital and valproate using our chronic implants significantly reduced focal seizures at all doses given. Furthermore, we show that local delivery of these compounds resulted in no adverse effects to motor function, whereas systemic delivery resulted in significant motor impairment. The results of this study demonstrate the potential of ASM micro dosing to the epileptic focus as a treatment option for people with drug resistant epilepsy. This technology could also be applied to a variety of disease states, enabling a deeper understanding of focal drug delivery in the treatment of neurological disorders.

Keywords: anti-seizure medication; brain; epilepsy; intracerebral delivery; neural implant.

Figure 1

Implantation of the MINI device and electrode in mice. (A) Schematic showing coordinates for the MINI device and electrode implantation in CA1 of the injected hippocampus. The electrode is implanted 0.7 mm posterior to the MINI device to allow for adequate space to connect the EEG tether for recording. Anatomic reference images courtesy of the Allen Mouse Brain Atlas. (B) Image of MINI device and recording electrode fixed in place with a dental cement headcap. (C) Representative mouse brain slice showing trypan blue infusion into CA1 through the MINI device. MINI = microinvasive neural implant infusion platform.

Figure 2

Seizure model validation in mice. Representative EEG recordings from epileptic mice from the kainate model, showing A(i) a monomorphic high-voltage sharp wave (HVSW) discharge with the boxed area in A(i) enlarged in a, and A(ii) ictal activity accompanying a generalized tonic-clonic seizure, followed by EEG suppression. All recordings are from an electrode located in the CA1 region of the injected hippocampus. [B(i and ii)] The (i) number and (ii) duration of seizure-like events (SLEs) recorded in the CA1 region of the hippocampus is shown over a period of 12 h following onset of the light phase. Data for number or duration of SLEs per hour are shown for six epileptic mice. Occurrence of generalized convulsive seizures (observed in two mice) is indicated by exclamation marks. Analysis of seizure frequencies and durations by one-way ANOVA did not indicate any significant differences over the 12-h time period shown (P = 0.110 and P = 0.312, respectively).

Figure 3

Effects on seizure-like events of local versus systemic delivery. Effects of phenobarbital (PB) and valproate (VPA) when given either intraperitoneally (i.p.) or intracerebrally on spontaneous seizure-like events (SLEs) in mice. [A(i and ii)] The number of SLEs recorded from the whole group of epileptic mice 45 min before drug treatment and 15 min–1 h after (i) intracerebral administration of 0.005 mg/kg (0.11 µg) or 0.017 mg/kg (0.4 µg) PB or (ii) intraperitoneal administration of 20 mg/kg or 60 mg/kg PB. [B(i and ii)] The same data are shown following (i) intracerebral administration of 0.029 mg/kg (0.7 µg) or 0.048 mg/kg (1.16 µg) VPA or (ii) intraperitoneal administration of 300 mg/kg or 400 mg/kg VPA, and [C(i and ii)] for (i) intracerebral or (ii) intraperitoneal administration of vehicle. To allow comparisons between different treatments, SLEs counted in the EEG in the 45-min block before drug treatment were normalized to 100% and compared with the SLEs counted in the EEG in the 45-min block after treatment. The number of SLEs before and after treatment for each individual mouse can be found in Supplementary Fig. 3. [D(i and ii)] The per cent change in SLE incidence 45 min before drug treatment and 15 min–1 h after both intraperitoneal and intracerebral delivery of (i) PB or (ii) VPA. All data are shown as means ± standard error of the mean; number of mice per experiment is indicated in the panels. Paired, two-tailed t-tests were performed to compare differences between baseline SLE activity and treatment. A series of unpaired, two-tailed t-tests were used to compare between different treatments, e.g. drug versus vehicle. For A–C, statistical differences between baseline control and treatment values within each group are indicated by asterisks (*P < 0.05, **P < 0.01, ****P < 0.0001). For D, statistical differences between equivalent intraperitoneal and intracerebral doses of a drug are indicated by asterisks (***P < 0.001), while statistical differences to vehicle are indicated by circles (^ooP < 0.01, ^oooP < 0.001, ^ooooP < 0.0001).

Figure 4

Evaluation of seizure-like event incidence over time following local and systemic delivery. The cumulative number of seizure like events (SLEs) over time is given for each mouse group following (A) systemic or (B) local administration of drug or vehicle. Because treatment groups all had different sample sizes, the total number of SLEs counted in the EEG in the 45-min baseline period before drug treatment were normalized to 100% and compared with the cumulative number of SLEs counted in 5 min intervals after treatment to allow for comparisons between groups. SLEs were recorded for at least 105 min following treatment and up to 165 min. The cumulative number of SLEs for each treatment group is given for the maximum amount of time EEG was recorded following treatment because EEG recordings were stopped or disturbed for some treatment groups past 105 min. The number of mice per experiment is indicated in the panels.

Figure 5

Behavioural testing for adverse effects in mice. Effects of either systemic or local administration of phenobarbital (PB) and valproate (VPA) on rotarod and open-field performance. The number of times a mouse fell off the rotarod was recorded during a 60 s trial after administration of either (A) PB or (B) VPA. (C) The total distance travelled in the open field test during a 15 min trial following systemic or local administration of PB. (D) The total number of revolutions (both clockwise and counterclockwise) in the same 15 min trial period. Testing began 15 min after administration of VPA and 30 min after administration of PB for both local and systemic administration. Systemic PB dose = 60 mg/kg, local PB dose = 0.017 mg/kg (0.4 µg), systemic VPA dose = 400 mg/kg and local VPA dose = 0.048 mg/kg (1.16 µg) for both tests. Data are shown as mean ± standard error of the mean. Group size for rotarod was n = 8 and group size for open field was n = 5. Statistical differences between groups are indicated by asterisks (*P < 0.05, **P < 0.01, ****P < 0.0001).