Perampanel in real-world epilepsy setting: results from the PERSICILIA study

In this multicentre, retrospective, real-world study, we evaluated 254 people with epilepsy initiating adjunctive perampanel in routine clinical practice. Treatment retention was high at 6 and 12 months (90.9% and 81.9%). In observed cases, ≥50% seizure reduction was achieved by 80.5% and 78.8%, while seizure freedom was reached by 28.1% and 25.5%, respectively. Conservative baseline-denominator estimates yielded responder rates of 73.2% and 64.6%, with seizure freedom in 25.6% and 20.9%; sustained seizure freedom occurred in 14.6% (37/254). Adverse events were reported in 10.4% at 6 months and 4.3% at 12 months, mostly irritability-related, leading to discontinuation in 14 patients. Median total drug load increased from 1.20 at baseline to 2.00 at 6 months and 1.92 at 12 months (p < 0.001), whereas the load of concomitant antiseizure medications remained largely stable, with modest reductions observed particularly among sodium channel blockers and SV2A ligands. Pittsburgh Sleep Quality Index and Epworth Sleepiness Scale scores improved significantly at both follow-ups, whereas Quality of Life in Epilepsy Inventory-31 total score remained stable, despite some domain-level improvements. Adjusted models indicated higher odds of response in temporal lobe epilepsy (OR 3.27; p = 0.006), and lower odds in patients with neuropsychiatric comorbidities, longer epilepsy duration, or female sex. Neuropsychiatric comorbidity was also strongly associated with adverse events (OR 6.01, p < 0.001). These results support sustained real-world effectiveness and manageable tolerability of perampanel across heterogeneous centres and patient subgroups, highlighting the importance of monitoring comorbidities and concomitant therapy adjustments, and suggesting potential benefits beyond seizure control.