Understanding the challenge of comparative effectiveness research in focal epilepsy: A review of network meta-analyses and real-world evidence on antiepileptic drugs.

OBJECTIVE: Head-to-head randomized controlled trials (RCTs) are the gold standard for assessing comparative treatment effects. In the absence of direct comparisons between all possible antiepileptic drugs (AEDs), however, clinical decision-making in focal (partial onset) epilepsy relies on alternative evidence borne from indirect comparisons including network meta-analyses (NMAs) and from real-world evidence (RWE) studies. We review NMAs and observational RWE studies comparing AEDs in the adjunctive setting to compare the robustness of these methods and to formulate recommendations for future evidence development. METHODS: A literature review identified NMAs and RWE studies comparing AEDs for the adjunctive treatment of focal seizures published between January 2008 and October 2018. NMAs were evaluated for robustness using a framework based on guidelines from the National Institute for Health and Care Excellence Decision Support Unit and the International Society for Pharmacoeconomics and Outcomes Research. RWE studies were evaluated using the GRACE checklist. RESULTS: From a total of 1993 records, 11 NMAs and six RWE studies were eligible. Key limitations identified in the NMAs include nonsystematic selection of RCTs, unexplored heterogeneity between included RCTs in terms of study and patient characteristics, and selection of AEDs and AED doses or dosing strategies that are not reflective of clinical practice. The main limitations of RWE studies concern sample size, design, and analysis methods. Approximately 90% of comparisons between individual AEDs were nonsignificant in the NMAs. None of the RWE studies adjusted for baseline differences between comparator groups; therefore, they lack the validity to make comparative conclusions. SIGNIFICANCE: Current NMAs and RWE studies provide only nominal comparative evidence for AED treatments in focal epilepsy, and should be used with caution for decision-making due to their methodological limitations. To overcome these hurdles, adherence to methodological guidelines and concerted efforts to collect relevant outcome data in the real world are needed.

A randomized, open-label, multicenter comparative trial of levetiracetam and topiramate as adjunctive treatment for patients with focal epilepsy in Korea.

OBJECTIVE: The objective of this trial was to compare the effectiveness of levetiracetam (LEV) and topiramate (TPM) as adjunctive treatment for patients with focal seizures in Korea. METHODS: In this Phase IV, open-label, multicenter trial (NCT01229735), adults were randomized to treatment with LEV (1000-3000 mg/day) or TPM (200-400 mg/day). Only patients achieving LEV ≥1000 mg/day or TPM ≥100 mg/day after a 4-week up-titration entered the 20-week dose-finding and subsequent 28-week maintenance periods. The primary outcome was the 52-week retention rate; others included safety and exploratory efficacy outcomes. RESULTS: Of 343 randomized patients (LEV 177; TPM 166), 211 (61.5%) completed the trial. In the full analysis set (FAS), retention rate was 59.1% with LEV vs 56.6% with TPM (p = 0.7007), while in the prespecified sensitivity analysis, based on data from patients who received drug doses in the recommended range (LEV 176; TPM 113), it was 59.1% with LEV vs 42.5% with TPM (p = 0.0086). In the FAS, median percent reduction in seizure frequency from baseline was 74.47% with LEV and 67.86% with TPM (p = 0.0665); ≥50% responder rate was 69.0% vs 64.8% (p = 0.4205), and the 6-month seizure-freedom rate was 35.8% vs 22.3% (p = 0.0061). In the sensitivity analysis, differences between groups were greater, favoring LEV. Incidences of treatment-emergent adverse events (TEAEs) were 70.6% with LEV vs 77.1% with TPM; most frequently somnolence (20.3%), dizziness (18.1%), and nasopharyngitis (13.6%) with LEV; and decreased appetite (15.7%), dizziness (14.5%), and headache (14.5%) with TPM. Discontinuations due to TEAEs were 7.9% with LEV and 12.7% with TPM. CONCLUSIONS: In this open-label trial, the 52-week retention rate was not significantly different between LEV and TPM. However, LEV was associated with a substantially higher seizure freedom rate and a more favorable safety profile than TPM in this population of Korean patients with focal seizures.

The epidemiology of drug-resistant epilepsy: A systematic review and meta-analysis.

OBJECTIVE: The definition of drug-resistant epilepsy (DRE) affects case identification and treatment, and impacts prevalence or incidence estimates and health burden estimation in epidemiology. The objective of this systematic review is to evaluate the consistency between definitions of DRE in the literature and the official definition in the International League Against Epilepsy (ILAE) guidelines, and to estimate the incidence, prevalence, and risk factors for DRE. METHODS: MEDLINE and EMBASE were searched for observational studies of DRE published between January 1980 and July 2015. The definitions of DRE in these studies were compared with the definition in the ILAE guidelines. Random-effect model meta-analyses were used to generate pooled estimates of prevalence or incidence and pooled odds ratios of the association with risk factors. RESULTS: Thirty-five studies met inclusion criteria, including 13 080 epilepsy patients and 3941 patients with DRE. The definition of DRE varied widely across studies, with only 12% meeting the requirements of the ILAE definition. The pooled prevalence proportion of DRE among epilepsy patients was 0.30 (95% confidence interval [CI] 0.19-0.42), and the pooled incidence proportion was 0.15 (95% CI 0.11-0.19). Age at onset, symptomatic epilepsy, abnormal neuroimaging findings, abnormal electroencephalography results, history of mental retardation, neuropsychiatric disorders, febrile seizure, and status epilepticus increased risk for DRE. SIGNIFICANCE: There are limited high-quality data available on DRE. Lack of consistency in definitions limits the ability to obtain robust estimates on the burden of DRE. More data based on the ILAE definition from well-designed epidemiologic studies are needed to generate accurate and reliable results.

The left supramarginal gyrus contributes to finger positioning for object use: a neuronavigated transcranial magnetic stimulation study.

In everyday actions, we grasp dozens of different manipulable objects in ways that accommodate their functional use. Neuroimaging studies showed that grasping objects in a way that is appropriate for their use involves a left-lateralized network including the supramarginal gyrus (SMG), the anterior intraparietal area (AIP) and the ventral premotor cortex (PMv). However, because previous works premised their conclusions on tasks requiring action execution, it has remained difficult to discriminate between the areas involved in specifying the position of fingers onto the object from those implementing the motor programme required to perform the action. To address this issue, we asked healthy participants to make judgements about pictures of manipulable objects, while repetitive transcranial magnetic stimulation (rTMS) was applied over the left SMG, AIP, PMv or, as a control, the vertex. The participants were asked to name the part of the image where the thumb or the index finger was expected to contact the object during its normal utilization or where a given attribute of the same object was located. The two tasks were strictly identical in terms of visual display, working memory demands and response requirements. Results showed that rTMS over SMG slowed down judgements of finger positions but not judgements of object attributes. Both types of judgements remained unaffected when rTMS was applied over AIP or PMv. This finding demonstrates that, within the parieto-frontal network dedicated to object use, at least the left SMG is involved in specifying the appropriate position of the thumb and index onto the object.

Role of distinct parietal areas in arithmetic: an fMRI-guided TMS study.

Although several parietal areas are known to be involved in number processing, their possible role in arithmetic operations remains debated. It has been hypothesized that the horizontal segment of the intraparietal sulcus (hIPS) and the posterior superior parietal lobule (PSPL) contribute to operations solved by calculation procedures, such as subtraction, but whether these areas are also involved in operations solved by memory retrieval, such as multiplication, is controversial. In the present study, we first identified the parietal areas involved in subtraction and multiplication by means of functional magnetic resonance imaging (fMRI) and we found an increased activation, bilaterally, in the hIPS and PSPL during both arithmetic operations. In order to test whether these areas are causally involved in subtraction and multiplication, we used transcranial magnetic stimulation (TMS) to create, in each participant, a virtual lesion of either the hIPS or PSPL, over the sites corresponding to the peaks of activation gathered in fMRI. When compared to a control site, we found an increase in response latencies in both operations after a virtual lesion of either the left or right hIPS, but not of the PSPL. Moreover, TMS over the hIPS increased the error rate in the multiplication task. The present results indicate that even operations solved by memory retrieval, such as multiplication, rely on the hIPS. In contrast, the PSPL seems to underlie processes that are nonessential to solve basic subtraction and multiplication problems.

Contribution of the primary motor cortex to motor imagery: a subthreshold TMS study.

Motor imagery (MI) mostly activates the same brain regions as movement execution (ME) including the primary motor cortex (Brodmann area 4, BA4). However, whether BA4 is functionally relevant for MI remains controversial. The finding that MI tasks are impaired by BA4 virtual lesions induced by transcranial magnetic stimulation (TMS) supports this view, though previous studies do not permit to exclude that BA4 is also involved in other processes such as hand recognition. Additionally, previous works largely underestimated the possible negative consequences of TMS-induced muscle twitches on MI task performance. Here we investigated the role of BA4 in MI by interfering with the function of the left or right BA4 in healthy subjects performing a MI task in which they had to make laterality judgements on rotated hand drawings. We used a subthreshold repetitive TMS protocol and monitored electromyographic activity to exclude undesirable effects of hand muscle twitches. We found that BA4 virtual lesions selectively increased reaction times in laterality judgments on hand drawings, leaving unaffected a task of equal difficulty, involving judgments on letters. Interestingly, the effects of virtual lesions of left and right BA4 on MI task performance were the same irrespective of the laterality (left/right) of hand drawings. A second experiment allowed us to rule out the possibility that BA4 lesions affect visual or semantic processing of hand drawings. Altogether, these results indicate that BA4 contribution to MI tasks is specifically related to the mental simulation process and further emphasize the functional coupling between ME and MI.

Dissociation between manipulation and conceptual knowledge of object use in the supramarginalis gyrus.

Because tool naming activates motor-related areas in the posterior parietal cortex, it has been suggested that conceptual knowledge of tools relies on their unique manipulation patterns. However, this view is questioned by the finding that some patients impaired in retrieving manipulation knowledge of man-made objects are still able to perform conceptual judgments on them. To address this issue, we used repetitive transcranial magnetic stimulation (rTMS) to interfere with the functioning of the anterior part of the right or left supramarginalis gyrus (SMG), a region critically involved in object-directed actions. rTMS was delivered in healthy participants performing four judgment tasks designed to explore different aspects of manipulation and conceptual knowledge of man-made objects. The two manipulation judgment tasks consisted in determining whether (1) two objects displayed on a computer screen are normally used by adopting a comparable hand posture, or (2) a given hand posture is appropriate to use an object. In the two conceptual judgment tasks, subjects had to decide whether (1) two objects displayed on the computer screen are normally used in the same context or (2) they are functionally related. We found that virtual lesions of left SMG interfere only with the performance of the manipulation judgment task in which subjects had to decide whether two different objects are used by adopting the same hand posture, all the other tasks being unaltered. rTMS applied over the right SMG had no effect. These results challenge the assumption that conceptual knowledge of tools is grounded upon motor representations.

Double dissociation between motor and visual imagery in the posterior parietal cortex.

Because motor imagery (MI) and visual imagery (VI) are influenced differently by factors such as biomechanical constraints or stimulus size, it is conceivable that they rely on separate processes, possibly involving distinct cortical networks, a view corroborated by neuroimaging and neuropsychological studies. In the posterior parietal cortex, it has been suggested that the superior parietal lobule (SPL) underlies VI, whereas MI relies on the supramarginalis gyrus (SMG). However, because several brain imaging studies have also shown an overlap of activations in SPL and SMG during VI or MI, the question arises as to which extent these 2 subregions really contribute to distinct imagery processes. To address this issue, we used repetitive transcranial magnetic stimulation to induce virtual lesions of either SMG or SPL in subjects performing a MI (hand drawing rotation) or a VI (letter rotation) task. Whatever hemisphere was stimulated, SMG lesions selectively altered MI, whereas SPL lesions only affected VI, demonstrating a double dissociation between MI and VI. Because these deficits were not influenced by the angular distance of the stimuli, we suggest that SMG and SPL are involved in the reenactment of the motor and visual representations, respectively, and not in mental rotation processes per se.

Motor imagery while judging object-hand interactions.

Because corticospinal excitability, as assessed with transcranial magnetic stimulation, has been repeatedly shown to increase during motor imagery, we used this approach to determine whether appreciating object-hand interactions involves motor imagery. Corticospinal excitability was measured in nine healthy participants who were asked to decide whether a hand presented in a given posture was compatible with the use of an object. The control task consisted in deciding whether two hands were in the same posture; a dimming task was used to determine the baseline. We found a significant increase in corticospinal excitability while judging object-hand interactions in comparison with the two other tasks. This finding suggests that predicting the consequences of an action involves implicit motor imagery.

Distinct contribution of the parietal and temporal cortex to hand configuration and contextual judgements about tools.

Neuropsychological studies showed that manipulatory and semantic knowledge can be independently impaired in patients with upper-limb apraxia, leading to different tool use disorders. The present study aimed to dissociate the brain regions involved in judging the hand configuration or the context associated to tool use. We focussed on the left supramarginalis gyrus (SMG) and left middle temporal gyrus (MTG), whose activation, as evidenced by functional magnetic resonance imaging (fMRI) studies, suggests that they may play a critical role in tool use. The distinctive location of SMG in the dorsal visual stream led us to postulate that this parietal region could play a role in processing incoming information about tools to shape hand posture. In contrast, we hypothesized that MTG, because of its interconnections with several cortical areas involved in semantic memory, could contribute to retrieving semantic information necessary to create a contextual representation of tool use. To test these hypotheses, we used neuronavigated transcranial magnetic stimulation (TMS) to interfere transiently with the function of either left SMG or left MTG in healthy participants performing judgement tasks about either hand configuration or context of tool use. We found that SMG virtual lesions impaired hand configuration but not contextual judgements, whereas MTG lesions selectively interfered with judgements about the context of tool use while leaving hand configuration judgements unaffected. This double dissociation demonstrates that the ability to infer a context of use or a hand posture from tool perception relies on distinct processes, performed in the temporal and parietal regions. The present findings suggest that tool use disorders caused by SMG lesions will be characterized by difficulties in selecting the appropriate hand posture for tool use, whereas MTG lesions will yield difficulties in using tools in the appropriate context.