Study design challenges and strategies in clinical trials for rare diseases: Lessons learned from pantothenate kinase-associated neurodegeneration

**Background:** Clinical trials for rare neurodegenerative diseases face significant design and methodology challenges, particularly when the disease involves central nervous system (CNS) degeneration with heterogeneous clinical manifestations and natural history. Traditional trials focus on symptom-based endpoints, which require established symptomatic disease—often indicating advanced neurodegeneration with exhausted brain reserve. The authors argue for a paradigm shift toward pre-symptomatic treatment, using a Primary Disease Activity Biomarker (PDAB) that directly indexes the core disease mechanism, enabling accelerated approval followed by long-term post-marketing studies. This review uses pantothenate kinase-associated neurodegeneration (PKAN) as a model to illustrate strategies applicable to other inborn errors of metabolism (IEMs) with movement disorders.

**Methods:** The paper is a narrative review that synthesizes lessons from prior PKAN clinical trials (e.g., the FORT trial of fosmetpantotenate and the deferiprone trial), as well as trials in other rare neurodegenerative genetic disorders such as Niemann-Pick disease type C (NP-C) and neuronal ceroid lipofuscinosis type 2 (CLN2). The authors discuss strategies derived from natural history reviews, engagement with physicians, patient advocacy groups, regulatory agencies (FDA, EMA), and biostatisticians. They also consider the potential of wearable technology, speech assessments, and neuroimaging as endpoints.

**Key Results:** The paper does not present original data but summarizes critical findings from prior studies. In the FORT trial (N=84, 24 weeks), the primary endpoint (PKAN-ADL) showed a modest 1–2 point improvement in both placebo and active arms, with no statistically significant difference between groups. The deferiprone trial (N=88, 18 months) did not meet its co-primary endpoints (BAD scale and PGI-I), though it effectively lowered brain iron levels as measured by MRI-R2*. In NP-C, miglustat improved horizontal saccadic eye movement velocity (HSEM) and swallowing, while arimoclomol significantly reduced annual disease progression on the NPCCSS. For CLN2, cerliponase alfa showed less decline in motor and language function compared to historical controls. The review emphasizes that study duration, endpoint sensitivity, and patient heterogeneity are critical factors; for example, the saltatory progression of classic PKAN (stepwise decline over days/weeks followed by plateaus) and the very slow progression of atypical PKAN (plateaus lasting months to years) necessitate different trial designs.

**Clinical Implications:** The authors recommend several strategies: (1) separate patient cohorts for classic vs. atypical PKAN, with different endpoints (dystonia scales for classic, speech/neuropsychiatric assessments for atypical) and study durations; (2) use of natural history controls or within-patient pre-/post-treatment comparisons when placebo controls are unethical or impractical; (3) incorporation of wearable technology for objective, home-based movement assessment; (4) combining alpha-adjusted p-values from multiple prespecified analysis methods to increase statistical power in small samples; and (5) early and continuous regulatory feedback to leverage flexibility in trial design. The ultimate goal is to enable pre-symptomatic intervention based on PDABs, which could prevent or substantially delay disease manifestation, rather than merely slowing progression in already-disabled patients. The authors call for regulatory agencies to apply the broadest possible flexibility, as articulated by FDA's Dr. Billy Dunn, to tolerate greater uncertainty in approving treatments for severe, life-threatening rare diseases with substantial unmet need.