Effectiveness and tolerability of fenfluramine in pediatric and adult patients with developmental and epileptic encephalopathies: A multicenter, retrospective, real‐world clinical‐practice study Authors: Vicente Villanueva, Elena González‐Alguacil, Patricia Smeyers, Sara Hernández, Victor Soto‐Insuga, Alejandro Fernández‐Cabrera, Laura Olivié, Teresa de Santos, Beatriz Parejo, Lucas Iacampo, Belén Ortuño, Javier Aparicio, Montserrat Fuentes, Juan Rodriguez‐Uranga, Paula Martínez‐Agredano, José Carlos Estévez, Angel Aledo‐Serrano, Rocio Martin‐Alvarez, David Conejo, Belen Baena, Blanca Mercedes‐Álvarez, Yolanda López, Alba Sierra‐Marcos, Rocio Trincado‐Lamuño, Mouna Ennazeh, Jesús González de la Aleja, Julia Renau, Maria Dolores Castro‐Vilanova, Fátima Romero‐Aguilera, José M Serratosa Published: 2026 ABSTRACT AbstractObjectiveDevelopmental and epileptic encephalopathies (DEEs) are characterized by drug‐resistant seizures and developmental slowing/regression. We examined the efficacy and tolerability of fenfluramine (FFA) in pediatric and adult patients with Lennox–Gastaut syndrome (LGS), Dravet syndrome (DS), and other DEEs.MethodsIn this multicenter, retrospective, real‐world study, FFA treatment was initiated ≥12 months before database closure. Data from patients' charts at 3, 6 and 12 months were recorded in the Spanish Epilepsy Society's DEEs registry. Effectiveness endpoints included seizure reductions of 100%, ≥75%, and ≥50% (responder), median percent seizure reduction, and seizure worsening. Safety endpoints included adverse event (AE) rates.ResultsThe study included 166 patients (LGS, n = 84; DS, n = 42; other DEE, n = 40; mean age 16.6 ± 12.1 years [111 pediatric, median 10.1 years; 55 adult, median 26.5 years]). Patients had a median of 3 prior failed anti‐seizure medications (ASMs). At 12 months, the median FFA dose was 0.49 mg/kg/day, with 77.1% treatment retention. There was a 68.8% median reduction in seizure frequency between baseline and 12 months (p < .001), with significant reductions across all diagnostic groups (LGS, p < .001; DS, p < .001; other DEE, p = .004). The 12‐month ≥50% responder rate was 61%, with no statistically significant differences according to age. At 12 months, there were significant reductions in the mean number of concomitant ASMs (p = .004) and the proportion of patients requiring rescue medication (p < .001). Meaningful improvements in Clinical Global Impression scores for cognition (59.3% of patients), behavior (40.9%), sleep (24.3%), and caregiver (44.8%) domains were observed. At 12 months, AEs were reported in 68.1% of patients (mostly mild‐to‐moderate) and AEs leading to discontinuation in 12%, with no significant differences according to age.SignificanceFFA was effective, with predictable tolerability, in patients with different DEEs and seizure types, supporting its use as an effective treatment option in pediatric and adult patients with DEEs.Keywords: anticonvulsants, drug‐resistant, epilepsy, seizures Key points Fenfluramine was effective in reducing seizures in adult and pediatric patients with developmental and epileptic encephalopathies. Fenfluramine was effective in reducing seizures in adult and pediatric patients with developmental and epileptic encephalopathies. Responder rates were similar across age categories, including in pediatric and adult patients. Responder rates were similar across age categories, including in pediatric and adult patients. Fenfluramine treatment reduced concomitant anti‐seizure medication (ASM) use, the need for rescue medication, and the frequency of scheduled healthcare visits. Fenfluramine treatment reduced concomitant anti‐seizure medication (ASM) use, the need for rescue medication, and the frequency of scheduled healthcare visits. Improvements in Clinical Global Impression (CGI) scores for cognition, behavior, sleep, and caregiver domains were observed. Improvements in Clinical Global Impression (CGI) scores for cognition, behavior, sleep, and caregiver domains were observed. The most frequent adverse events with fenfluramine were somnolence and loss of appetite. The most frequent adverse events with fenfluramine were somnolence and loss of appetite. INTRODUCTION Developmental and epileptic encephalopathies (DEEs) are characterized by seizures and/or epileptiform activity, which contribute to developmental slowing/regression. 1 The umbrella of DEEs includes those with a clear genetic, structural, or other etiology; those with established diagnostic criteria (e.g., Dravet syndrome [DS], Lennox–Gastaut syndrome [LGS]); and those without clearly defined syndromes or with unknown causes. 2 Unmet needs in DEEs include inadequate seizure control, a high burden of adverse events (AEs) related to polytherapy, increased sudden unexpected death in epilepsy (SUDEP) risk, cognitive and behavioral impairments, and significant caregiver burden. 3 , 4 , 5 , 6 In addition, there are challenges in establishing multidisciplinary care teams, difficulties during transition from pediatric to adult care, and increased use of emergency resources. 7 Recently, new anti‐seizure medications (ASMs), including rufinamide, stiripentol, cannabidiol, and ganaxolone, have been developed as orphan drugs for a subset of patients within the DEE group. In clinical trials, DEEs are mostly represented by individuals with DS or LGS, and many individuals with ultra‐rare DEEs or undefined syndromes are excluded. Fenfluramine (FFA) is the most recently approved therapy for seizures associated with DS and LGS. It is licensed in the United States for use in patients 2 years of age and older, and in Europe for a comparable indication as adjunctive therapy alongside other ASMs. 8 , 9 FFA has a dual mechanism of action, modulating both serotonergic transmission, via modulation of distinct 5‐Hydroxytryptamine receptors, and the σ1 receptor. 10 Data from four randomized controlled trials (RCTs) have demonstrated the efficacy and tolerability of FFA in patients with DS and LGS. 11 , 12 , 13 , 14 Long‐term effectiveness and tolerability have been confirmed in open‐label extension (OLE) studies. 15 , 16 In addition, early access programs (EAPs) and compassionate use have provided evidence supporting the effectiveness and tolerability of FFA in patients with DS, LGS, and other DEEs. 17 , 18 However, reports on clinical practice experience post‐marketing remain limited, particularly among adult patients. 19 , 20 , 21 , 22 , 23 , 24 , 25 The aim of this study was to describe the 1‐year effectiveness and tolerability of FFA in the largest post‐marketing cohort of patients with DEEs, including DS and LGS, and to provide insights into its use in clinical practice. METHODS Study design This multicenter, retrospective, observational study was conducted in 18 Spanish hospitals (Table S1) using data from the DEEs Registry of the Pharmacological Therapy Working Group of the Spanish Epilepsy Society. The study protocol was approved by the ethics committee of Hospital Universitario y Politécnico La Fe (Valencia) and followed the Declaration of Helsinki and Strengthening the Reporting of OBservational studies in Epidemiology (STROBE) guidelines. All patients received FFA post‐marketing (after July 2023). Inclusion criteria were signed informed consent by patients/legal representatives and initiation of FFA in a clinical practice setting ≥12 months before database closure (October 2025). Exclusion criteria included unreliable clinical records according to participating physicians, and participation in clinical trials or the EAP. Although FFA is approved for DS and LGS, some patients with other DEEs were treated within a compassionate‐use program when deemed suitable by participating physicians. DS and LGS were defined according to International League Against Epilepsy (ILAE) criteria, 26 and DEEs according to the published operational definition. 1 Study outcomes The effectiveness population comprised patients who fulfilled the inclusion criteria and had at least one effectiveness measurement. The safety population comprised patients who fulfilled the inclusion criteria and received ≥1 dose of FFA. Effectiveness was assessed based on monthly seizure frequency and seizure type (according to updated ILAE Seizure Classification). 27 Primary effectiveness endpoints included seizure freedom, ≥75% and ≥50% seizure reduction, median percent seizure reduction from baseline, and seizure worsening, analyzed overall and by diagnostic category. Effectiveness outcomes were evaluated for total and most disabling seizures (convulsive and drop seizures). Convulsive seizures included clonic, tonic–clonic, and focal to bilateral tonic–clonic seizures; drop seizures included tonic or atonic seizures associated with falls. Secondary effectiveness endpoints included outcomes in other seizure types (e.g., absences, myoclonic, focal without bilateral progression), retention rate, change in seizure severity, ≥50% seizure reduction at 12 months (by age), changes in status epilepticus, healthcare resource utilization at 12 months vs baseline, concomitant and rescue medication use vs baseline, and non‐seizure outcomes evaluation (i.e., Clinical Global Impression [CGI] for cognition, behavior, sleep, and caregiver domains) at 12 months. Primary safety endpoints included the proportion of patients with AEs (by age), AE severity (mild, moderate, or severe according to participating physicians), and the proportion of patients with AEs who discontinued at 3, 6, and 12 months (by age). Only AEs considered related to FFA were included in the analysis. Effectiveness and safety in DEEs by etiology and genetic diagnosis were also assessed. Data collection Patients' clinical charts were reviewed at each center and data were collected by participating physicians according to usual clinical practice. Patients were usually seen at baseline and at 3‐, 6‐, and 12‐month visits following FFA initiation. Blood samples were taken at least once during follow‐up. Echocardiography was performed at baseline and every 6 months by a cardiologist during the 1 year of follow‐up. Cardiac valvulopathy and pulmonary hypertension were evaluated. Data were collected through the Spanish Epilepsy Society's DEEs registry. Study data were collected and managed using Research Electronic Data Capture (REDCap). At baseline, patient demographic information, seizure frequency and type (including status epilepticus episodes in the last 3 months), prior and concomitant ASM use (including rescue medication), and healthcare resource use in the last 3 months were collected. At each visit, FFA dosage, seizure frequency, change in seizure severity from baseline (based on 3‐item Caregiver Global Impression scale, scoring as improvement, similar, or worsening), status epilepticus frequency, concomitant and rescue ASM use, and healthcare resource use (number of prearranged visits, emergency department visits, and hospital admissions) were collected. Change in non‐seizure outcomes previously defined from baseline were evaluated using the 7‐point CGI scale, ranging from very much improved to very much worse. Statistical analysis The log‐rank test was used to compare time‐to‐event variables between groups. Treatment response (i.e., 100%, ≥75%, ≥50% reduction and worsening in seizure frequency) was compared between groups using the chi‐square test. A binary logistic regression was used to relate worsening with patient information. Continuous and dichotomous‐related variables were analyzed using Wilcoxon's and McNemar tests, respectively. Comparisons of continuous and discrete variables between groups were analyzed using the Kruskal–Wallis and chi‐square tests, respectively. AEs were recorded as frequency and percentages. IBM SPSS Statistics 28.0 (IBM Corporation, Armonk, NY, USA) was used for all statistical analyses; the threshold for statistical significance was 5% (p < .05). RESULTS Patient baseline characteristics Of 181 patients included in the study database, 166 (56% female) fulfilled the inclusion/exclusion criteria and were eligible for final analysis. All excluded patients started FFA <12 months before database closure. There were 111 pediatric patients (<18 years; 66.9%) and 55 adult patients (≥18 years; 33.1%). Eighty‐four patients (50.6%) had LGS, 42 (25.3%) had DS, and 40 (24.1%) had other DEEs. The most common etiologies were genetic (83 patients; 50%) and structural (23; 13.9%) (Table 1). TABLE 1Patient demographics and disease characteristics at baseline.CharacteristicLGS (n=84)DS (n=42)Other DEEs (n=40)All DEEs (N=166)Female, n (%)29 (34.5)20 (47.6)24 (60)93 (56) Mean age, years (range), n (%) 20.2 (2–66)12.5 (1–53)11 (3–46)16.6 (1–66)≤6 year old7 (8.3)16 (38.1)7 (17.5)30 (18.1)7–12 year old16 (19)10 (23.8)17 (42.5)43 (25.9)13–17 year old22 (26.2)8 (19)8 (20)38 (22.9)≥18 year old38 (46.4)8 (19)8 (20)55 (33.1)Median duration of epilepsy, years (IQR)4.9 (1.9–13.4)5 (2.1–9.8)5.8 (4.8–11)5.6 (2.5–11.6)Etiology, n (%)Genetic a 25 (17.8)42 (100)16 (40)83 (50)Structural15 (17.9)0 (0)8 (20)23 (13.9)Genetic and structural6 (7.1)0 (0)7 (17.5)13 (7.8)Genetic and metabolic0 (0)0 (0)1 (2.5)1 (0.6)Infectious4 (4.8)0 (0)0 (0)4 (2.4)Metabolic1 (1.2)0 (0)1 (2.5)2 (1.2)Other34 (40.5)0 (0)7 (17.5)41 (24.7)Prior ASMs, n (%)Median (IQR)4 (3–5)3 (2–4)3 (2.5–4)3 (3–4)Mean (SD)4.1 (2.3)3 (0.8)3.4 (1.4)3.6 (1.9)1–334 (40.5)31 (73.8)23 (57.5)88 (53)4–648 (57.1)11 (26.2)16 (40)75 (45.2)>65 (6)01 (2.5)3 (1.8)Concomitant ASMs, n (%)Median (IQR)3 (3–4)3 (2–3)3 (2–4)3 (3–4)Mean (SD)3.4 (0.9)2.9 (0.7)3.1 (1.1)3.2 (1)11 (1.2)1 (2.4)3 (7.5)5 (3)211 (13.1)10 (23.8)8 (20)29 (17.5)334 (40.5)23 (54.8)17 (42.5)74 (44.6)≥438 (45.2)8 (19)12 (30)58 (34.9)Monthly seizure frequencyAll seizuresMean (SD)79 (83)5.7 (8.9)96.2 (203)61 (111)Median (IQR)58 (26.7–100)1.7 (1–4.7)20.7 (1–4.7)27 (3.3–70)Convulsive seizuresMean (SD)24.6 (38.3)3.8 (6)31 (53.8)17.8 (53.8)Median (IQR)6.7 (2–30)1.8 (1–4)4 (1–20)4 (1–20)Drop seizuresMean (SD)34 (70.2)1.8 (1.6)54.3 (62.6)35.1 (67.9)Median (IQR)19.2 (2.33)1.8 (6.7–75)38.3 (6.7–75)14.5 (2–38.3)Other seizures b Mean (SD)49.3 (65.7)6.6 (9.7)27.1 (51.3)36.6 (51.3)Median (IQR)30 (10–60)1 (0.7–8)6.3 (3.5–28.3)20.7 (3.3–33.3)Abbreviations: ASMs, anti‐seizure medications; DEE, developmental and epileptic encephalopathy; DS, Dravet syndrome; IQR, interquartile range; LGS, Lennox–Gastaut syndrome; SD, standard deviation. a All patients with DS had pathogenic variants in SCN1A. Among patients with LGS, two had pathogenic variants in CDKL5, whereas single cases had pathogenic variants in ADGRG1, ALDH7A1, IQSEC2, CHD2, DYNC1H1, KMT2, DCX, SPTAN1, STXBP1, TCF4, TSC2, and GNA genes. Chromosomal abnormalities included 9p deletion with partial 13q trisomy, 13q14q31 deletion, t (13q14 17q11.2), deletion of the 15q11.2‐q13 region (n = 3), microduplication 15q11‐q13 (n = 1), 15q duplication, 15q tetrasomy (n = 2), and 22q13.1. In seven LGS patients, genetic status was not recorded. Among other DEEs, three had PAFAH1B1 pathogenic variants, two had pathogenic variants in CDKL5, two in DCX, two in NEXMIF (Case 1: p.Asn1153Lysfs*8 [crX:73960933], DEE with a predominance of myoclonic absences and eyelid myoclonia; Case 2: c.1882C>T, p.Arg628* [crX:73962510], DEE with a history of myoclonic absences and predominant generalized tonic–clonic seizures in recent years), two in SCN1A outside the Dravet phenotype, two in TSC2, and one each in SCN1B and SCN8A (c.2952C>A, p.Asn984Lys, gain‐of‐function variant), WWOX, PCDH19, TSC1, and BCL11B. Chromosome alterations included an Xq28 duplication (n = 2), a 2q24 deletion, a 15q duplication, and an inversion duplication 15q. b Includes absences, myoclonic, focal seizures without bilateral progression and other seizures. Patient demographics and disease characteristics at baseline. Mean age, years (range), n (%) Abbreviations: ASMs, anti‐seizure medications; DEE, developmental and epileptic encephalopathy; DS, Dravet syndrome; IQR, interquartile range; LGS, Lennox–Gastaut syndrome; SD, standard deviation. All patients with DS had pathogenic variants in SCN1A. Among patients with LGS, two had pathogenic variants in CDKL5, whereas single cases had pathogenic variants in ADGRG1, ALDH7A1, IQSEC2, CHD2, DYNC1H1, KMT2, DCX, SPTAN1, STXBP1, TCF4, TSC2, and GNA genes. Chromosomal abnormalities included 9p deletion with partial 13q trisomy, 13q14q31 deletion, t (13q14 17q11.2), deletion of the 15q11.2‐q13 region (n = 3), microduplication 15q11‐q13 (n = 1), 15q duplication, 15q tetrasomy (n = 2), and 22q13.1. In seven LGS patients, genetic status was not recorded. Among other DEEs, three had PAFAH1B1 pathogenic variants, two had pathogenic variants in CDKL5, two in DCX, two in NEXMIF (Case 1: p.Asn1153Lysfs*8 [crX:73960933], DEE with a predominance of myoclonic absences and eyelid myoclonia; Case 2: c.1882C>T, p.Arg628* [crX:73962510], DEE with a history of myoclonic absences and predominant generalized tonic–clonic seizures in recent years), two in SCN1A outside the Dravet phenotype, two in TSC2, and one each in SCN1B and SCN8A (c.2952C>A, p.Asn984Lys, gain‐of‐function variant), WWOX, PCDH19, TSC1, and BCL11B. Chromosome alterations included an Xq28 duplication (n = 2), a 2q24 deletion, a 15q duplication, and an inversion duplication 15q. Includes absences, myoclonic, focal seizures without bilateral progression and other seizures. The median number of prior failed ASMs (including concomitant medications) was 3 (interquartile range [IQR] 3–4); the most frequent were valproic acid (VPA; 111 patients; 66.9%), clobazam (105; 63.3%), and cannabidiol (78; 47%). The median number of concomitant ASMs at baseline was 3 (IQR 3–4); overall, the most frequent were VPA (110; 66.3%), clobazam (104; 62.7%), and cannabidiol (51; 30.7%). At baseline, 9 patients (5.5%) had undergone epilepsy surgery, 22 (13.3%) had vagus nerve stimulation implanted, and 18 (10.9%) were receiving a ketogenic diet. Rescue medication was used by 36.1% of patients at baseline, including midazolam (22.3%), diazepam (10.2%), clonazepam (1.8%), lorazepam (1.2%), and clobazam (0.6%). Dosage and titration The median FFA daily dose was 0.15 mg/kg at 3 months (range 0.03–0.88), 0.45 mg/kg/day at 6 months (0.09–0.94), and 0.49 mg/kg/day at 12 months (0.39–0.63). FFA Daily doses by age group are shown in Table 2. A dose cap was applied in all patients (Appendix S1). TABLE 2Fenfluramine dosages according to age.Age groupDose unit3 months, mean (median), range6 months, mean (median), range12 months, mean (median), rangeAllmg/kg/day a 0.42 (0.15), 0.03–0.880.46 (0.45), 0.09–0.940.51 (0.49), 0.39–0.630–6 yearsmg/kg/day0.45 (0.4), 0.1–0.880.45 (0.4), 0.1–0.930.54 (0.5), 0.12–0.947–11 yearsmg/kg/day0.45 (0.43), 0.09–0.990.47 (0.49), 0.09–0.80.52 (0.49), 0.09–0.9412–17 yearsmg/day14.1 (13.2), 1.1–2016.8 (15.4), 5.75–3618.6 (17.6), 6.9–3612–17 yearsmg/kg/day0.33 (0.32), 0.09–0.620.38 (0.36), 0.15–0.620.42 (0.42), 0.16–0.70≥18 yearsmg/day20.4 (23.0), 4.4–26.021.1 (26.0), 4.4–26.021.7 (26.0), 6.6–26.0≥18 yearsmg/kg/day0.32 (0.35), 0.07–0.450.32 (0.33), 0.07–0.640.33 (0.35), 0.11–0.64 a Dose expressed as mg/kg/day for all age groups and as mg/day for patients aged ≥12 years. Fenfluramine dosages according to age. Dose expressed as mg/kg/day for all age groups and as mg/day for patients aged ≥12 years. Overall, 36.2% of patients reached the target FFA dose at 2 weeks from baseline, 11% at 3 weeks, 22.7% at 4 weeks, and 30.1% at >4 weeks. There were statistically significant differences in titration rates between diagnostic groups (Appendix S1). Retention rates FFA treatment retention rates were 96.4% (160/166), 86.1% (143/166), and 77.1% (128/166) at 3, 6, and 12 months, respectively (Figure S1). The mean treatment retention time was 11.5 months (95% confidence interval [CI] 9.4–11.3; Figure S2) and was significantly higher in the DS group compared with the other groups (p = .015; log rank test). Seizure frequency At 12 months, there was a median percentage reduction in seizure frequency of 68.8% for total seizures (p < .001; Wilcoxon test), 78.9% for convulsive seizures (p = .001), 80% for drop seizures (p < .001), and 80.6% for other seizures (p < .001; including 83.3% for absence seizures [p = .047]). Of 32 patients with absence seizures at baseline, the mean seizure frequency per month fell from 31 ± 48 at baseline to 18 ± 46 at 12 months. The proportions of patients with 100%, ≥75%, and ≥50% reductions in frequency of total, convulsive, drop, and other seizures, and worsening of seizure frequency, are shown in Figure 1. Proportion of patients with a reduction or worsening in seizure frequency following fenfluramine initiation according to diagnosis. Bars show reductions (100%, ≥75%, and ≥50%) or worsening in seizure frequency at 3, 6, and 12 months after fenfluramine initiation. (A) Total seizures; (B) Convulsive seizures; (C) Drop seizures; (D) Other seizures. DEE, developmental and epileptic encephalopathy; DS, Dravet syndrome; LGS, Lennox–Gastaut syndrome. Significant reductions in total seizures were observed in the LGS (68.1%; p < .001; Wilcoxon test), DS (88.9%; p < .001), and other DEE (66.7%; p = .004) subgroups, with numerically (but not significantly) greater reductions in the DS group. Worsening of seizure frequency was associated with an absence of concomitant VPA (p = .02). Reductions in the frequency of individual seizure types were comparable between diagnostic groups. In the LGS, DS, and other DEE groups, convulsive seizures decreased by 81.7%, 76.4%, and 58.3%; drop seizures by 80%, 88.9%, and 90%; and other seizure types by 62.2%, 100% and 77.4%, respectively. Responders by age and DEE diagnosis Within each diagnosis category, there was no significant difference in the proportion of responders to FFA (i.e., ≥50% reduction in seizure frequency between baseline and 12 months) according to age (Table 3). For patients with LGS, a higher responder rate was observed in adults vs children 0–6 or >6–12 years of age, possibly related to disease evolution, although the number of patients with LGS in the younger age groups was low. There were no significant differences in rates of seizure freedom, ≥50% responders, and seizure worsening between pediatric and adult patients (Table S2). Among adult patients, 5 of 54 (9.3%) achieved seizure freedom; in adults with drop or convulsive seizures, 7of 39 (17.9%) became seizure‐free and 26 of 39 (66.7%) had a ≥50 reduction in drop/convulsive seizure frequency (Table S3). TABLE 3Responder rate (≥50%) at 12 months according to age and epilepsy diagnosis.AgeLGSDSOther DEEsAll DEEsAll ages, n/N (%)48/83 (57.8)32/42 (76.2)21/40 (52.5)101/165 (61.2)0–6 years, n/N (%)3/7 (42.9)12/16 (75)5/7 (71.4)20/30 (66.7)>6–12 years, n/N (%)6/16 (37.5)8/10 (80)10/17 (58.8)24/43 (55.8)>12–17 years, n/N (%)16/22 (72.7)6/8 (75)4/8 (50)26/38 (68.4)≥18 years, n/N (%)23/38 (60.5)6/8 (75)2/8 (25)31/54 (57.4) p‐value.14.99.29.56Abbreviations: DEEs, developmental and epileptic encephalopathies; DS, Dravet syndrome; LGS, Lennox–Gastaut syndrome. Responder rate (≥50%) at 12 months according to age and epilepsy diagnosis. Abbreviations: DEEs, developmental and epileptic encephalopathies; DS, Dravet syndrome; LGS, Lennox–Gastaut syndrome. Seizure severity change At 12 months, an improvement in seizure severity was observed in 81 of 156 patients (52%); seizure severity was classified as similar in 62 (39.7%) and worse in 13 (8.3%) (Table S4). Status epilepticus At baseline, 11 patients (6.6%) had status epilepticus within the preceding 3 months, 5 with LGS (6%), 4 with DS (9.5%), and 2 with other DEEs (5%). At 12 months, this fell to 5 patients (3%): 1 with LGS (1.2%), 3 with DS (7.1%), and 1 with other DEEs (2.5%). Despite this numerical decrease, the low event count precluded meaningful statistical analysis (McNemar test, p = .146). Rescue medication The proportion of patients requiring the usual rescue medication decreased between baseline and 12 months (36.1 → 15.1%; p < .001; McNemar test). Significant reductions were also observed in patients with LGS (34.5 → 13.1%; p < .001) and DS (50 → 19%; p < .001), but not in those with other DEEs. Healthcare resources The median number of scheduled visits per month decreased by 24.1% between baseline and 12 months (0.67 → 0.33/month; p < .001; Wilcoxon test). Reductions were observed across all diagnostic groups (LGS 50%, p < .001; DS 50%, p = .001; other DEEs 50%, p = .020). There were no significant differences in the number of visits to the emergency department or in the number of hospital admissions between baseline and 12 months overall, or within diagnostic subgroups. Concomitant ASMs The mean number of concomitant ASMs decreased between baseline and 12 months in the overall population (3.2 → 3.1; p = .004, Wilcoxon test) and in the DS group (2.9 → 2.5; p = .002). No significant difference in the mean number of concomitant ASMs was observed in the LGS (p = .110) and other DEEs (p = 1.0) groups. At baseline, 79.5% of patients were receiving >2 concomitant ASMs, decreasing to 69.2% at 12 months. The proportion of patients receiving >2 ASMs decreased across diagnoses (LGS: 85.7 → 79.7%; DS: 73.8 → 52.4%; other DEEs: 72.5 → 65%). At 12 months, 42 patients (25.3%) had reduced ≥1 concomitant ASM; the most frequently reduced were cannabidiol (30.7 → 23.7%), topiramate (16.9 → 12.7%) and stiripentol (12.7 → 7.2%). Changes and discontinuations in concomitant ASMs by diagnostic group are shown in Appendix S1 and Table 4, and the most frequently discontinued ASMs are shown in Table S5. TABLE 4Adverse events.LGS, n = 84DS, n = 42Other DEEs, n = 40All DEEs (N = 166)Cumulative AEs, n (%)3 months41 (48.8)31 (73.8)20 (50)92 (55.4)6 months45 (53.6)34 (81)27 (67.5)106 (63.9)12 months52 (61.9)34 (81)27 (67.5)113 (68.1)AEs at ≤12 months (≥2% of patients), n (%)Somnolence23 (27.4)14 (33.3)13 (32.5)50 (30.1)Lack of appetite11 (13.1)14 (33.3)9 (22.5)34 (20.5)Loss of weight8 (9.5)7 (16.7)3 (7.5)18 (10.8)Diarrhea10 (11.9)6 (14.3)0 (0)16 (9.6)Agitation6 (7.1)2 (4.8)3 (7.5)11 (6.6)Aggressivity5 (6)2 (4.8)1 (2.5)8 (4.8)Insomnia3 (3.6)1 (2.4)2 (5)6 (3.6)Seizure worsening1 (1.2)1 (2.4)3 (7.5)5 (3)Tremor0 (0)4 (9.5)1 (2.5)5 (3)Ataxia/gait disturbances3 (3.6)1 (2.4)1 (2.5)5 (3)Sialorrea4 (4.8)0 (0)0 (0)4 (2.4)Behavioral disturbances2 (2.4)2 (4.8)0 (0)4 (2.4)Mood changes1 (1.2)1 (2.4)2 (5)4 (2.4)AE severity at 12 months, n (%)Mild21 (25)19 (45.2)10 (25)50 (30.1)Moderate25 (29.8)10 (23.8)11 (27.7)46 (27.7)Severe6 (7.1)5 (11.9)6 (15)17 (10.2)Discontinuation due to AEs, n (%)10 (11.9)1 (2.4)9 (22.5) a 20 (12) Note: AEs occurring in <2% of patients included: three cases each of constipation, lethargy, and fatigue; two cases each of gut disturbance, asthenia, autolytic ideation, pyrexia, and irritability; one case each of hallucinations, self‐lesions, headache, dehydration, leg edema, urinary incontinence, dizziness, vomiting, aphasia, bradypsychia, fall, neurological impairment, skin mouth lesion, increase of mucus, pneumonia, and status epilepticus.Abbreviations: AE, adverse event; DEEs, developmental and epileptic encephalopathies; DS, Dravet syndrome; LGS, Lennox–Gastaut syndrome. a p = .02 for overall group comparison (DS < LGS < other DEEs). Adverse events. Note: AEs occurring in <2% of patients included: three cases each of constipation, lethargy, and fatigue; two cases each of gut disturbance, asthenia, autolytic ideation, pyrexia, and irritability; one case each of hallucinations, self‐lesions, headache, dehydration, leg edema, urinary incontinence, dizziness, vomiting, aphasia, bradypsychia, fall, neurological impairment, skin mouth lesion, increase of mucus, pneumonia, and status epilepticus. Abbreviations: AE, adverse event; DEEs, developmental and epileptic encephalopathies; DS, Dravet syndrome; LGS, Lennox–Gastaut syndrome. p = .02 for overall group comparison (DS < LGS < other DEEs). No association was found between the number of ASM adjustments and patient response. Non‐seizures outcomes Between baseline and the 12‐month visit, 51 of 86 participants (59.3%), 36 of 88 (40.9%), 18 of 74 (24.3%), and 30 of 67 participants (44.8%) with CGI data showed improvements in cognition, behavior, sleep and caregiver domains, respectively; 3 of 86 (3.5%), 13 of 88 (14.8%), 11 of 74 (14.9%), and 4 of 67 (6%) reported worsening (Figure 2). Behavioral outcomes were worst in the other DEEs group (Kruskal–Wallis test, p = .012). Change in Clinical Global Impression scores for cognition, behavior, sleep, and caregiver domains between baseline and 12 months according to diagnosis. DEE, developmental and epileptic encephalopathy; DS, Dravet syndrome; LGS, Lennox–Gastaut syndrome. Safety The cumulative percentage of AEs was 55.4% at 3 months, 63.9% at 6 months, and 68.1% at 12 months. At 12 months, 30.1% of AEs were mild, 27.7% were moderate, and 10.2% were severe. The most frequent AEs were somnolence (30.1%), loss of appetite (20.5%), and weight loss (10.8%). AEs leading to discontinuation were reported in 12% of participants; the most frequent were somnolence (3.6%), seizure worsening (2.4%), lack of appetite (1.8%), aggressivity (1.2%), and agitation (1.2%) (Table 4). There were no significant differences in the proportion of patients with AEs between diagnostic groups at 12 months, although AEs were numerically highest in patients with DS (p = .073; chi‐square test; Table 4). Discontinuations due to AEs were significantly higher in patients with other DEEs (22.5%; p = .02) vs LGS (11.9%) or DS (2.4%). The most common AEs leading to discontinuation in the LGS group were somnolence (4.8%), aggressivity (2.4%), and loss of weight (2.4%); in DS, these were agitation (2.4%) and status epilepticus (2.4%); and in other DEEs, seizure worsening (7.5%; including two cases with structural etiology [Sturge–Weber syndrome and left frontal dysplasia] and one with genetic etiology [Xq28 duplication]), somnolence (5%), and loss of appetite (5%). There were no significant differences in the proportion of patients with AEs or AEs leading to discontinuation between pediatric and adult patients, although discontinuations were numerically higher in the pediatric population (Table S2). No relevant laboratory disturbances were observed, and no cases of cardiac valvulopathy or pulmonary hypertension were reported. Subgroup analysis by overall and genetic etiology Effectiveness and safety endpoints according to overall etiology and genetic etiology are shown in Tables S6 and S7 for the “other DEE” subgroup, and Table S8 for all DEEs with tuberous sclerosis complex (TSC) and Cyclin‐Dependent Kinase‐Like (CDKL5) pathogenic variants. Owing to small sample sizes, we cannot draw definitive conclusions on the impact of etiology on effectiveness and safety. DISCUSSION In this large, real‐world clinical‐practice study, FFA demonstrated sustained effectiveness across multiple DEE types and ages groups, with numerically higher responder rates in patients with DS. FFA treatment was associated with a reduction in some healthcare resource utilization; no new safety signals were revealed. Our efficacy data are consistent with outcomes from RCTs, 11 , 12 , 13 , 14 OLE studies, 15 , 16 and real‐world studies. 19 , 20 , 21 , 22 , 23 , 24 , 25 We observed significant reductions in seizures across diagnostic subgroups and across seizure types. This is particularly important in regard to convulsive and drop seizures due to their association with SUDEP 28 and traumatic injuries. The 79% reduction in convulsive seizure frequency (76% in DS; 82% in LGS) is slightly higher than reported in DS (54%–65% reduction) 11 , 12 , 13 and LGS (46%–58% reduction) clinical trials. 29 Moreover, those outcomes are in line with a review of FFA use in generalized tonic–clonic seizures across different studies in patients with DEEs (47%–100% median percent reduction in tonic–clonic seizures). 20 In patients with DS, the percentage of ≥50% responders at 12 months was comparable with that reported in a German cohort of 78 patients (63% vs 76% in our study); responder rates for convulsive seizures were 71% in our study compared with 54% in the German series. 18 In another real‐world study, the ≥50% responder rate at 3 months was slightly higher (94%) than in our study. 22 In patients with LGS, an OLE study reported a ≥50% responder rate of 33.3% for drop seizures, compared with 73% in our study. 16 These differences could be explained by the restricted inclusion criteria of the OLE and the long duration of follow‐up. Outcomes in other DEEs are underreported due to the restricted indication of FFA; in a small series of 20 children with other DEEs, 40% achieved long‐term ≥75% seizure reduction, 23 slightly higher than the 28% achieving a ≥75% response at 1 year in our other DEE cohort. Heterogeneity in the other DEE group may have influenced lower response in our study. Another small case series reported a 50% responder rate with FFA in patients with Rett syndrome and drug‐resistant epilepsy. 30 Similarities in patients' responses to FFA across DEE phenotypes may sometimes be related to pathogenic variants in the same gene. A subset of patients showed worsening of seizures, partly reflecting the drug resistance of this polymedicated group over time, which can be aggravated when ASMs are reduced after adding FFA. The only significant association with worsening in our study was the absence of concomitant VPA, although this analysis is limited by the small sample size. Seizure severity generally mirrored ≥50% responder rates, with best outcomes in DS, followed by LGS and other DEEs. In other DEEs, modest severity improvement likely reflects better control of less disabling seizures rather than drop seizures. Considering the underrepresentation of adult patients in RCTs and most clinical practice studies, an important finding of our study is the apparent lack of significant age‐related differences in responder rates among patients with DEEs, with particularly consistent rates in those with DS and LGS. This is similar to findings from a European EAP for FFA, which reported >75% seizure reduction in over half of patients with DS, and comparable magnitude, consistency, and durability of responses across age groups. 17 Sustained seizure reduction was also reported across age groups in an OLE study of patients with LGS (27.6% in children; 40% in adults). 16 Our observations support the effectiveness of FFA in both pediatric and adult patients with DEE, although outcomes in the DS and other DEE groups should be considered with caution due to the relatively low proportion of adult patients in these groups (19% and 20%, respectively). Moreover, there is a major limitation as not all genetic variants were included in the study. As specific variants can impact the choice of ASM, 31 , 32 , 33 it is of utmost importance that these are considered in future studies. The dosing and titration patterns observed in our study reflect real‐world prescribing behavior and highlight the importance of individualized dose optimization. The 12‐month mean daily dose (0.49 mg/kg) was similar to that reported by Guerrini et al. (0.39 mg/kg at last follow‐up), 17 and in the LGS OLE study (0.4 mg/kg). 16 Although previously unreported, our data in mg/day, using the usual practical approach in adults, showed mean doses of 20.4 mg/day at 3 months and 21.1 mg/day at 1 year. Titration was slower than that recommended in the prescribing information but is in line with previously published practical recommendations. 34 , 35 FFA dosage adjustments may be required when FFA is co‐administered with strong enzyme inhibitors or inducers, particularly stiripentol 34 ; also, a reduction of the dosage of bromides when used concomitantly with FFA has recently been reported. 25 Treatment retention in our study was high (96%, 86%, and 77% at 3, 6, and 12 months, respectively), in line with a German DS cohort (92% in adults, 83% in children) 18 and an Italian EAP (86%). Previously reported retention outcomes in LGS were slightly lower (64% in the OLE study). 16 In our study, retention was highest in the DS group; however, outcomes in the other groups were also robust, especially considering the highly drug‐resistant population and extensive prior ASM exposure. We observed a significant reduction in concomitant ASMs following FFA treatment, particularly in patients with DS. Similar reductions in concomitant ASMs have been reported in two DS series (45%–71% of population simplified baseline treatment after FFA) and one study in patients with DEEs. 18 , 19 , 22 Reducing polypharmacy is especially relevant in DEEs, where cumulative ASM exposure contributes to adverse cognitive, behavioral, and systemic effects. 36 We also observed reductions in healthcare resource utilization following FFA treatment. Scheduled visits decreased by 24% and the proportion of patients requiring rescue medication declined from 36% to 15% over 12 months. For patients with DS treated with FFA, a recent Danish study reported a 52% reduction in hospital contacts, 37 and an Italian series demonstrated improved status epilepticus episodes and reduced rescue medication. 38 Together, these findings, across different healthcare systems, highlight the wider societal impact of achieving seizure control for these patients. FFA treatment was associated with improvements in multiple CGI domains, including cognition, behavior, sleep, and caregiver domains. Although assessed using CGI rather than formal testing, these changes are clinically meaningful and reflect domains of high relevance to patients and caregivers. The findings are consistent with results from 65 caregivers of patients with DS who had participated in an FFA long‐term extension Phase 3 clinical trial or U.S. EAP. 39 Caregivers reported quality‐of‐life (QoL) benefits for patients, including improvements in cognitive function, alertness, and academic performance, alongside improvements in sleep quality and mental health. 39 Similarly, a long‐term extension study in patients with DS reported clinically meaningful improvements in emotion and cognitive regulation following FFA based on the Behavior Rating Inventory of Executive Function (BRIEF). 40 Finally, benefits in everyday executive functions were reported in adults with LGS enrolled in RCTs following BRIEF assessments. 41 Together, these findings highlight potential benefits of FFA beyond seizure control, warranting further investigation. Some patients (2.4%) experienced behavioral deterioration following treatment with FFA, which was associated with treatment discontinuation. Behavioral deterioration is related to QoL 4 , 5 , 6 and should therefore be considered when adding a new ASM. The safety profile observed here is consistent with those reported in RCTs and long‐term extension studies. 11 , 12 , 13 , 15 AEs were common (68% at 1 year) but predominantly mild to moderate. Discontinuation due to AEs was most frequent among patients with other DEEs, consistent with a recent real‐world study, 22 and there was no significant difference between discontinuation rates in pediatric vs adult patients; the rates within diagnostic groups were similar or slightly higher than those observed in other FFA studies (DS, 0%–5.1%; LGS, 5%–5.7%; and other DEEs, 13%). 14 , 15 , 18 , 22 Somnolence was the most frequent AE across groups (30% overall; 33% in the DS cohort). This was higher than reported by Guerrini et al. (16%) 17 but lower than reported by Strzelczyk et al. (36%) 18 in patients with DS. These differences could be explained by different populations and drug loads. The second most frequent AE, lack of appetite, occurred more often in DS patients (33.3%) than in LGS patients (13.1%), as reported in the extension studies (26.7% in DS; 16.1% in LGS). No new safety signals were identified and no clinically relevant cardiac abnormalities were observed, in line with previous findings, despite the range of age groups. Nevertheless, longer follow‐up is needed to confirm these outcomes. 15 This study has several limitations. Based on its retrospective, observational design and no control group, there is potential for selection and reporting biases. Seizure frequency and non‐seizure outcomes were derived from routine clinical records and caregiver reports, which may be affected by recall bias, documentation variability between centers, and missing information. The heterogeneous cohort, comprising different DEE etiologies and epilepsy syndromes, may have introduced variability in treatment response and tolerability, and subgroup analyses (especially for other DEEs) were limited by heterogeneity and small sample sizes. Moreover, the uneven distribution of diagnoses across age groups may confound the relationship between age and syndrome type. Treatment decisions, titration schedules, and adjustments of concomitant ASMs were not standardized and were based on clinician judgment, reflecting real‐world practice; this could potentially confound treatment effects. Finally, a major limitation is that not all genetic variants were included in the study; as specific variants may impact the choice of ASM, these should be explored in future studies. Nevertheless, the study included a large and clinically representative cohort with multiple DEE syndromes and a broad range of patient ages, including a large group of adult patients (particularly in the LGS group). CONCLUSIONS This real‐world study extends evidence supporting FFA as an effective and durable treatment for patients with DEEs across diagnostic categories and age groups. The inclusion of a substantial adult population (one third of the cohort) and different phenotypes, both underrepresented in most studies, and analysis of healthcare utilization and non‐seizure outcomes, allow a comprehensive assessment of FFA in routine care, complementing evidence from randomized trials and other real‐world studies. Compared with pediatric patients, adult responder rates appeared higher in LGS but lower in other DEEs, although these differences were not statistically significant and could be impacted by confounding between age and diagnostic group. Our data reinforce FFA as a valuable therapeutic option in the management of DEEs. Additional data are needed on safety (including cardiac evaluation), different phenotypes, and potential non‐seizure effects of treatment. AUTHOR CONTRIBUTIONS All authors contributed to the conception and design of the study and to acquisition of data. Vicente Villanueva organized the database, analyzed the data, wrote the first draft, and created tables and figures. All authors discussed the results, revised the first draft, and approved the final manuscript. FUNDING INFORMATION This article includes data obtained through the Spanish Epilepsy Society (SEEP) Registry owned and financed by the SEEP. UCB is one of the contributors that supports the registry through an unrestricted educational grant. In addition, UCB provided financial support for this study with an IIS‐2023‐600442. UCB was not involved in the study design, collection, analysis or interpretation of the data, the writing of the report, or the decision to submit the article for publication. CONFLICT OF INTEREST STATEMENT VV has received honoraria and/or research funds from Adium, Angelini Pharma, Bial, Eisai, Jazz Pharmaceuticals, Neuraxpharm, Novartis, Rapport, Takeda, UCB Pharma, and Xenon. EG has participated in advisory boards, research and educational activities supported or organized by Eisai, Jazz Pharmaceuticals, Nutricia, and UCB Pharma. PS has received fees for conferences and other educational activities from Eisai, GW Pharmaceutical Company, Humana, Neuraxpharm, and UCB Pharma. VS has participated in advisory boards, research and educational activities supported or organized by Biocodex, Eisai, Jazz Pharmaceuticals, Nutricia, Takeda, and UCB Pharma. AF has received honoraria and/or research funds from Angelini Pharma, Eisai, Jazz Pharmaceuticals, and UCB Pharma. LI has received honoraria from Angelini Pharma, Bial, Eisai, Jazz Pharmaceuticals, and UCB Pharma. JRU has received honoraria and/or research funds from Angelini Pharma, Bial, Eisai, Europharma, GW Pharma, Jazz Pharmaceuticals, Lundbeck, Neuraxpharm, Novartis, Orión, and UCB Pharma. PM has received honoraria and/or research funds from Angelini Pharma, Eisai, Jazz Pharmaceuticals, Bial, and UCB Pharma. AA has received funding for research and educational activities from Jazz Pharmaceuticals, Angelini Pharma, Biocodex, and UCB Pharma. YL has received funding educational activities from Angelini Pharma, Bial, Eisai, Adamed, and UCB Pharma. AS has received fees for conferences and other educational activities from Angelini, UCB Pharma, GW Pharmaceutical Company, Eisai, Bial, and Neuraxpharm. MDC has received honoraria from Angelini Pharma, Bial, Eisai, Jazz Pharmaceuticals, Neuraxpharm, and UCB Pharma. JS has received honoraria and/or research funds from Arvelle, Bial, Eisai, Esteve, GW Pharmaceuticals, Sanofi, and UCB Pharma. SH, LO, TS, BP, BO, JA, MF, JCE, RM, DC, BB, BM, RT, ME, JG, JR, and FR have no conflicts of interest to disclose. ETHICAL PUBLICATION STATEMENT We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. The study protocol was approved by the hospital's ethics committee and was in accordance with the code of ethics set out in the 1964 Declaration of Helsinki and its later amendments. CONSENT For all patients participating in the study, written informed consent was obtained from the patient or their legal representative according to the study protocol. Supporting information Table S1. Participants and centers. Table S2. Responder rates and adverse event rates in pediatric versus adult patients at 12 months. Table S3. Effectiveness outcomes in adult patients (≥18 years old) according to diagnosis. Table S4. Change in seizure severity at 12 months according to diagnosis. Table S5. Most frequently discontinued ASMs during FFA treatment. Table S6. Effectiveness and safety endpoints according to overall etiology at 12 months. Table S7. Responder (≥50% reduction in seizure frequency) and adverse event rates according to genetic etiology in other DEEs at 12 months. Table S8. Effectiveness and safety endpoints in all patients with tuberous sclerosis complex or CDKL5 pathogenic variant at 12 months. Figure S1. Patient disposition. Figure S2. Treatment retention time (Kaplan‐Meier analysis) for (A) All DEEs and (B) by DEE diagnosis. Appendix S1. Dose capping and titration. ACKNOWLEDGMENTS The authors acknowledge Patricia Santagueda for assistance with statistical analysis. Medical writing support was provided by Ian Marshall (WriteMedical Ltd). DATA AVAILABILITY STATEMENT The data that support the findings of this study are available from the corresponding author upon reasonable request. REFERENCES Associated Data Supplementary Materials Table S1. Participants and centers. Table S2. Responder rates and adverse event rates in pediatric versus adult patients at 12 months. Table S3. Effectiveness outcomes in adult patients (≥18 years old) according to diagnosis. Table S4. Change in seizure severity at 12 months according to diagnosis. Table S5. Most frequently discontinued ASMs during FFA treatment. Table S6. Effectiveness and safety endpoints according to overall etiology at 12 months. Table S7. Responder (≥50% reduction in seizure frequency) and adverse event rates according to genetic etiology in other DEEs at 12 months. Table S8. Effectiveness and safety endpoints in all patients with tuberous sclerosis complex or CDKL5 pathogenic variant at 12 months. Figure S1. Patient disposition. Figure S2. Treatment retention time (Kaplan‐Meier analysis) for (A) All DEEs and (B) by DEE diagnosis. Appendix S1. Dose capping and titration. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request.