2025/08/05 by Danielle M. Andrade, Anne T. Berg, Arunan Selvarajah +35 · 1 voice
Biochemistry, Genetics and Molecular Biology · Medicine · #Autopsy Techniques and Outcomes #Ethics in Clinical Research #Genomics and Rare Diseases
paper · pdf · doi:10.1111/epi.18583
openalex publication_date 2025/08/05 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/27
We welcome critical engagement with our study, “Dravet Syndrome: From Neurodevelopmental to Neurodegenerative Disease?”1 and appreciate the opportunity to clarify our findings, strengthen our interpretations, and reassert their clinical importance.1 The critiques raised (Sisodiya et al., 2025) underscore the importance of epistemological rigor and conceptual precision, which we support.2 Yet, we must assert that our findings are robust, significant, and demand thoughtful reconsideration of current assumptions about the adult phenotype of Dravet syndrome (DS). Terminological accuracy must not come at the expense of timely scientific inquiry. DS was once called “severe myoclonic epilepsy of infancy,” a term that did not encompass the full clinical spectrum. Today, DS is classified as a developmental and epileptic encephalopathy (DEE), but this classification does not preclude the emergence of neurodegenerative processes later in life. Our article is clear in questioning neurodegeneration, rather than redefining disease classification. This hypothesis is grounded in converging evidence: early-onset parkinsonism, caregivers' perceptions of functional decline in activities of daily living, and positive dementia screening in a subset of patients. After our 2014 report of parkinsonism in adults with DS, interest in its biological basis grew.3 Subsequent investigations utilizing dopamine transporter single-photon emission computed tomography (DAT-SPECT) in adults with DS and parkinsonism did not show dopaminergic dysfunction.4, 5 Nevertheless, the absence of abnormalities on DAT-SPECT was not interpreted as evidence against the presence of parkinsonism, which remains fundamentally a clinical diagnosis. More sensitive methodologies identified a reduction in dopamine synthesis, as evidenced by decreased concentrations of neopterin, tetrahydrobiopterin, 5-hydroxyindoleacetic acid, and homovanillic acid in the cerebrospinal fluid of four patients with DS.5, 6 Our findings also challenge the “active epileptic encephalopathy” hypothesis raised by Sisodiya et al., which posits that progression in DS is due to poorly controlled seizures. In 2017, we compared DS to Lennox–Gastaut syndrome (LGS), another DEE with more frequent seizures and heavier antiseizure medication use in adults. Yet, parkinsonism was more prevalent in patients with DS, suggesting a distinct mechanism unrelated to seizure activity or medication alone.7 It is possible that some of the patients in Selvarajah et al. (2025) had hyperammonemia. In fact, when we evaluated the signs of hyperammonemia in patients with DS, we reported on their gait difficulties.8 However, that does not negate the previous neurologists’ observations of movement disorders in patients that did not have those metabolic abnormalities.3, 7, 9 The patients reported on in these three studies are part of the Adult Genetic Epilepsy (AGE) program in Toronto. At AGE, patients are routinely screened for hyperammonemia, so this can be confidently ruled out. Of note, our prospective study, which included physical examination and instrumented gait analysis, demonstrated a short-stepped shuffling gait, which further declined 5 years after the first evaluation. This supports the concept of DS as a progressive disorder.9 The Unified Parkinson's Disease Rating Scale (UPDRS) is not a diagnostic tool as parkinsonism is a clinical diagnosis. Nevertheless, the scale has long been validated and is a widely accepted assessment measure to evaluate the progression and severity of Parkinson's disease. We have discussed the limitations of our modified version, administered to poorly cooperative patients, yet the reported scores are far beyond what would be expected from normal aging alone. Finally, early onset Parkinsonism is increasingly recognized as a manifestation in adults with neurodevelopmental disorders, such as Dravet syndrome, Rett syndrome, and autism, including those where valproate is not used,10, 11 thus suggesting that valproate is not the only cause of parkinsonism in patients with developmental delay. The proposed mechanisms include mitochondrial dysfunction, abnormal neurotransmitter metabolism, endosomal trafficking, and autophagic and ubiquitin-proteasome system abnormalities,11-13 Andrade et al., submitted. Our findings of patients screening positive for dementia and caregivers reporting cognitive and functional decline are aligned with research showing dementia is underdiagnosed in patients with rare genetic neurodevelopmental disorders.14 Reduced cognitive reserve may explain earlier or atypical presentations. Dementia has been recognized in DS, tuberous sclerosis complex, Angelman syndrome, Phelan–McDermid syndrome, MBD5-developmental disorder, and more.15 Although status epilepticus may lead to a decline of previous neurological functioning, it is noted that dementia is seen in conditions not strongly associated with seizures, such as Prader–Willi.16 In 2011, Sisodyia and colleagues described one patient with DS diagnosed by them with dementia and parkinsonism. That patient had postmortem brain tissue analysis, which did not show abnormal phosphorylated tau (p-tau) or α-synuclein accumulation.17 Not all neurodegenerative diseases can be reduced to tauopathies and synucleinopathies. Emerging evidence suggests that alternative pathogenic mechanisms may drive neuronal degeneration with clinical decline. Multiple sclerosis, for example, involves neurodegeneration without proteinopathy accumulation.18 It is widely recognized that many assessment tools for patients with intellectual disability (ID), and especially DEEs, have limitations.19 We stated explicitly that those tools are screening—not diagnostic—measures. Their value as assessment measures is quantified by their sensitivity and specificity. Furthermore, Moss-PAS (Moss-psychiatric assessment schedule) is fully compliant with the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5); is validated as a screening tool for cognitive decline; including dementia in adults with ID; and is frequently used for patients with Down syndrome.20-22 The Moss-PAS, Vinelands and SCQ (Social Communication Questionnaire) are instruments used to quantify Observer Reported Outcomes (ObsROs). Concerns regarding potential selection bias and the use of ObsROs are acknowledged in our study: “Parental impression of functioning is a subjective evaluation, and random recall error may be present in the results of this part of the evaluation.” However, regulatory agencies such as the U.S. Food and Drug Administration (FDA) and European Medicines Agency (EMA) have formally recognized ObsROs measures as critical and necessary to capture how a patient feels and functions in his or her daily life, something a clinical exam or a performance measure is often ill-suited to do.23-28 The instruments we employed have documented psychometric properties including test–retest and inter-rater reliability.28 As we noted, this is acknowledged by the FDA and EMA,23-28 and in our study.1 However, bias could operate in both directions: families experiencing more decline may have been more motivated to participate in this study, whereas others may have been too overwhelmed or fatigued to do so. The potential for selection bias does not mean one exists. Our findings reflected similar levels of difficulties as those found in a Dravet Syndrome Foundation-wide study.29 Although a limitation, the consistent age-related trends across multiple domains—gait, cognitive, motor—supported by validated tools, are unlikely to reflect mere statistical noise. Sodium channel blockers (SCBs) and age were independently associated with clinical outcomes, as shown in “methods”1 and here in Table 1. The issue of valproate use is discussed above. This has been addressed for Vinelands and Moss-PAS. Regarding the motor component, caregivers were trained on how to create videos for centralized scoring of the modified Unified Parkinson's Disease Rating Scale (mUPDRS), which was done by two movement disorder specialists. Our methods of using caregiver-recorded videos were validated with in-person assessments on a portion of the sample.9 The pandemic disrupted care, but similar patterns of decline were seen in pre-pandemic data.29 Hence, COVID-related changes are unlikely to explain our findings. We agree that “association does not prove causation.” However, without association, there is no reason to investigate causation. Our data provide that association. Aligning research efforts and clinical practices with the true trajectory observed by families and clinicians offers tangible, immediate, and long-term benefits. Conversely, delaying this discussion risks perpetuating therapeutic stagnation, missing critical intervention windows, and failing to meet the evolving needs of individuals living with DS. We reaffirm our commitment to transparency, methodological rigor, and the clinical well-being of individuals with DS. Our findings, coupled with growing evidence in the published literature, provide a compelling basis for considering the possibility of a neurodegenerative process in adults with DS who demonstrate marked cognitive and motor decline. We have addressed other explanations for this association, and none can account for the findings. In science, as in any rigorous debate, arguments must be grounded in evidence—not assertion. We welcome an evidence-based dialogue on this important issue. To advance understanding, we invite collaboration on urgently needed longitudinal, multimodal studies. The signals are present, the evidence is growing, and the time to investigate is now. Arunan Selvarajah acquired data, analyzed the data, and drafted the manuscript for intellectual content. Andrea Sabo acquired data, analyzed the data, and reviewed the manuscript for intellectual content. Carolina Gorodetsky acquired data, analyzed the data, and reviewed the manuscript for intellectual content. Paula Marques acquired data and reviewed the manuscript for intellectual content. Ilakkiah Chandran acquired data, analyzed the data, and drafted the manuscript for intellectual content. Miles Thompson analyzed the data, and drafted the manuscript for intellectual content. Quratulain Zulfiqar Ali acquired data, analyzed the data, and drafted the manuscript for intellectual content. Mary Pat McAndrews acquired data and reviewed the manuscript for intellectual content. Maria Carmela Tartaglia acquired data and reviewed the manuscript for intellectual content. Victor S.T. Lira acquired data and reviewed the manuscript for intellectual content. Linda Huh acquired data and reviewed the manuscript for intellectual content. Mary Connolly acquired data and reviewed the manuscript for intellectual content. Arezoo Rezazadeh acquired data and reviewed the manuscript for intellectual content. Farah Qaiser acquired data and reviewed the manuscript for intellectual content. Tad Fantaneanu acquired data and reviewed the manuscript for intellectual content. Monica Duong acquired data and reviewed the manuscript for intellectual content. Karen Barboza acquired data and reviewed the manuscript for intellectual content. Lysa Boissé Lomax acquired data and reviewed the manuscript for intellectual content. Luciana Inuzuka Nakaharada acquired data and reviewed the manuscript for intellectual content. Kette Valente acquired data and reviewed the manuscript for intellectual content. Jack Arbinuch acquired data, analyzed data, and reviewed the manuscript for intellectual content. Mariana Espindola acquired data and reviewed the manuscript for intellectual content. Eliana Garzon acquired data and reviewed the manuscript for intellectual content. Gianluca Sorrento acquired data and reviewed the manuscript for intellectual content. Mary Anne Meskis acquired data and reviewed the manuscript for intellectual content. Nicole Villas acquired data and reviewed the manuscript for intellectual content. Veronica Hood acquired data and reviewed the manuscript for intellectual content. Marta Gonzalez acquired data and reviewed the manuscript for intellectual content. Elena Cardenal-Muñoz acquired data and reviewed the manuscript for intellectual content. Jose Angel Aiba acquired data and reviewed the manuscript for intellectual content. Lauraine McKenna acquired data and reviewed the manuscript for intellectual content. Christine Linehan acquired data and reviewed the manuscript for intellectual content. Orrin Devinsky acquired data and reviewed the manuscript for intellectual content. Ryan Yuen acquired data and reviewed the manuscript for intellectual content. Anne T. Berg acquired data and reviewed the manuscript for intellectual content. Babak Taati acquired data and reviewed the manuscript for intellectual content. Alfonso Fasano designed and conceptualized the study, acquired data, analyzed the data, and reviewed the manuscript for intellectual content. Danielle M. Andrade designed and conceptualized the study, analyzed the data, reviewed the manuscript for intellectual content, supervised, and acquired funding. This project was made possible by the Dravet Syndrome Foundation (USA). We thank all the families that participated in the study. This study was funded by the Dravet Syndrome Foundation (USA). D.M.A. received grant support from Dravet Syndrome Foundation; trial Support from Epygenix and Zogenix (UCB); And consultancy from Eisai, Biocodex, Stoke and Jazz; And serves on the medical advisory board for Dravet Syndrome Foundation. O.D. has equity and/or compensation from the following companies: Tilray, Tevard Biosciences, Regel Biosciences, Script Biosciences, Actio Biosciences, Empatica, Ajna Biosciences, Blackrock Neurotech, Emotiv, and Praxis Precision Therapeutics. He has received consulting fees or equity options from UCB Pharma and Ultragenyx. LIN UCB, PRAXIS, Abbott. J.A.A. is the President of the Dravet Syndrome Foundation Spain (DSF), and he and/or DSF have received grants and/or financial support from Jazz Pharma, UCB, Ovid Therapeutics, Encoded, Stoke, Takeda, Biocodex, Praxis, and StrideBio to help carry out some of DSF's founding activities or to provide consulting services. J.A.A. honoraria have always been donated directly or indirectly to the DSF. K.V. serves on São Paulo Research Foundation and receives support from Biocodex, Takeda, UCB, PRAXIS, Abbott, Prati-Donaduzzi, ACHE, and Eurofarma. The remaining authors have no conflicts of interest. 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 data that support the findings of this study are available from the corresponding author upon reasonable request.