Relutrigine (PRAX-562)

Praxis Precision Medicines (PRAX) is under FDA Priority Review for relutrigine (PRAX-562), a first-in-class small-molecule inhibitor of the persistent sodium current (INaP), for pediatric SCN2A and SCN8A developmental and epileptic encephalopathies (DEEs). The FDA accepted the NDA in March 2026 based on the EMBOLD Phase 2 registrational cohort, and after a three-month extension the PDUFA date is now December 27, 2026 [3][7]. Relutrigine has Fast Track, Orphan Drug, Rare Pediatric Disease, and Breakthrough Therapy designations. The pitch is molecular selectivity: relutrigine shows a roughly 60-fold preference for the pathological persistent leak current over the normal firing (peak) current in patch-clamp assays, which older sodium channel blockers cannot match [5]. EMBOLD delivered a 46% placebo-adjusted median seizure reduction in 16 SCN2A/SCN8A patients, with open-label extension medians reaching ~75% [8]. With $1.5B in cash after a January 2026 follow-on, Praxis is funded into 2028 and does not need to raise to cross the PDUFA date [4]. A confirmatory Phase 3 (NCT07010471, n=160) is enrolling in parallel and is expected to read out around December 2026 [2]. This is no longer a Phase 3 binary. It is a near-term FDA approval decision with human-genetics-grade target validation [6].

Relutrigine is a first-in-class investigational INaP inhibitor with no approved products in this mechanistic class. Regulatory status as of Q2 2026: FDA accepted the NDA in March 2026 and granted Priority Review, with a PDUFA target action date originally September 27, 2026, then extended by three months to December 27, 2026 [3][7]. Designations in hand: Fast Track, Orphan Drug (SCN2A-DEE and SCN8A-DEE), Rare Pediatric Disease (which enables a Priority Review Voucher on approval, currently trading on the secondary market for roughly $150-200M based on recent Bavarian Nordic, Abeona, and Jazz transactions), and Breakthrough Therapy Designation granted July 2025 [9][10]. Two studies remain active. NCT05818553 is the EMBOLD Phase 2 program whose registrational cohort 1 (16 patients: 7 SCN2A, 9 SCN8A) generated the pivotal data supporting the NDA; registrational cohort 2 continues to enroll and is designed as supportive [1][8]. NCT07010471 is a confirmatory Phase 3 (n=160), active, not recruiting, with primary completion estimated December 2026 [2]. Financials: per the 10-K filed 2026-02-19, Praxis ended 2025 with $926.1M in cash and marketable securities; a January 2026 follow-on added $621.2M in net proceeds, extending runway into 2028 [4]. The company can fund through the PDUFA date and initial launch without dilution.

Neurons fire by briefly opening voltage-gated sodium channels. Sodium rushes in, the cell depolarizes, the channel inactivates within about a millisecond, and normal firing continues. In SCN2A (Nav1.2) and SCN8A (Nav1.6) gain-of-function mutations, a fraction of channels fail to fully inactivate. A small 'persistent' sodium current (INaP) keeps trickling in after the channel should have closed, and neurons stay chronically hyperexcitable. That leak drives the seizures and the cognitive plateau that define DEE, a group of severe childhood epilepsies where the seizures themselves damage brain development [6]. Relutrigine achieves selectivity for INaP through state-dependent binding. Channels that carry the pathological persistent current dwell longer in slow-inactivated conformations; relutrigine preferentially occupies channels in those states and stabilizes them there. In patch-clamp assays it inhibits mutation-induced persistent INa with IC50 values around 75-141 nM while blocking the normal peak (phasic) current only at roughly 60-fold higher concentrations, and it stabilizes Nav1.6 slow inactivation by left-shifting the steady-state inactivation curve [5]. Older sodium channel blockers like phenytoin, carbamazepine, and lacosamide all interact with sodium channels non-selectively enough to blunt normal firing at therapeutic doses. Lacosamide is the closest incumbent because it also enhances slow inactivation, but its selectivity for persistent over peak current is roughly 2-4-fold in comparable assays, an order of magnitude less than relutrigine's ~60-fold [5]. This is the quantitative basis of the mechanism thesis. The genetic case for the target is as clean as it gets in neurology: SCN2A and SCN8A gain-of-function variants cause DEE through this exact mechanism [6]. Few late-stage CNS drugs walk in with this grade of human-genetics validation.

The regulatory pivotal was EMBOLD (NCT05818553), a Phase 2 randomized, double-blind, placebo-controlled study whose registrational cohort 1 enrolled 16 patients across SCN2A-DEE and SCN8A-DEE. Topline: 46% median placebo-adjusted reduction in motor seizures over the 16-week double-blind period, with a safety profile in line with the sodium channel class [8]. Open-label extension data have shown even larger effects, with median reductions of ~75% and one patient exceeding 200 days seizure-free. FDA accepted this dataset as the basis for the NDA on the strength of the effect size, the genetic homogeneity of the target population, and Breakthrough Therapy Designation [10]. The confirmatory Phase 3 (NCT07010471) is a 160-patient global randomized double-blind placebo-controlled study in pediatric DEE with change in motor seizure frequency versus placebo as the primary endpoint; status is active, not recruiting, with estimated primary completion December 2026 [2]. Seizure frequency is the accepted regulatory endpoint for rare epilepsy programs, with cannabidiol (Epidiolex) and fenfluramine (Fintepla) approvals in Dravet and Lennox-Gastaut serving as the pathway precedents [1][2]. The residual design concern is variance. DEE seizure counts swing month to month, placebo response can drift high, and 160 patients spread across genotypes leaves the confirmatory trial exposed if the true effect proves smaller than EMBOLD suggested. Genotype pre-specification is the mitigant, and EMBOLD's placebo arm behaved cleanly. Enrollment closing on schedule is a positive execution signal for a small-cap running a global rare-disease Phase 3, where site activation and central seizure-diary adjudication are usually the choke points.

This drug is under FDA review (NDA/BLA), with a PDUFA decision date of 2026-09-27. Our estimate of 82% is the historical filing-approval rate for its area, adjusted for its rejection history (no prior Complete Response Letters). At this stage the early-trial design model no longer applies - what matters is that it reached the FDA and whether it has been rejected before.

Regulatory: the FDA extended review by three months in mid-2026, pushing the PDUFA date to December 27, 2026 [7]. Extensions are common and usually reflect submission of additional data at Agency request, not a signal of impending rejection, but they widen the uncertainty window and any label-negotiation friction (age range, genotype restrictions) will land here. Safety: the drug hits sodium channels centrally, and on-target CNS effects like somnolence, ataxia (loss of voluntary coordination), and cognitive blunting are the historical Achilles heel of this class. Cardiac conduction is the off-target concern because sodium channels also drive the cardiac action potential (SCN5A/Nav1.5 in the heart). Any QT interval (repolarization delay) or PR interval (atrioventricular conduction slowing) signal on ECG in a pediatric label would be devastating. Praxis has disclosed no cardiac issue to date, and relutrigine's in vitro Nav1.5 profile shows the same persistent-vs-peak preference seen at Nav1.2/1.6 [5], but regulators will scrutinize this class-wide risk closely. Efficacy: EMBOLD's 46% placebo-adjusted effect has to survive the confirmatory Phase 3 with more clinical heterogeneity. If real-world effect is smaller, the Phase 3 becomes a coin flip on statistical significance even if biology is right. Commercial: the addressable population is small. SCN2A variants account for roughly 1-2% of DEEs; combined SCN2A and SCN8A DEE US pediatric prevalence is estimated in the low thousands [6]. Orphan pricing (Epidiolex-like or higher) is required to justify the investment. Payer coverage for genotyped cases is straightforward; the harder question is off-label pressure from broader DEE and whether the label extends beyond the two channels tested. Competitive: Ionis and academic-industry collaborators have published individualized allele-selective antisense oligonucleotides (ASOs) for SCN2A that reduced seizure frequency by 26% and 90% in two treated patients, and Ionis has pipeline programs for both SCN2A and SCN8A [11]. These are earlier stage and more logistically complex than a small molecule, but if genotype-specific ASO platforms scale, they would eventually compete on the same patient populations.

Worth watching, and closely. Praxis is a small-cap whose valuation is materially tied to relutrigine, and the December 27, 2026 PDUFA date is the near-term binary that defines whether relutrigine becomes the first genotype-targeted DEE therapy. Because the NDA is already accepted with Priority Review, the risk profile has shifted from Phase 3 execution risk to regulatory/label risk. The three-month FDA extension in mid-2026 is worth flagging but is not, on its own, a bear signal [7]. Praxis's balance sheet is a strength: $926M at end-2025 plus $621M from the January 2026 follow-on funds operations into 2028, so no dilution is required to reach the PDUFA date or fund an initial commercial launch [4]. On approval, Praxis moves from clinical-stage to commercial-stage with a genotype-defined orphan asset in a payer-friendly rare-disease indication, plus a Priority Review Voucher (secondary market recently ~$150-200M based on Bavarian Nordic, Abeona, and Jazz transactions) [9] on the balance sheet. If the FDA issues a Complete Response Letter or a materially restricted label, the confirmatory Phase 3 readout in late 2026 becomes decisive, and the near-term Praxis pipeline (ulixacaltamide in essential tremor, PRAX-628 in focal epilepsy) has to carry the story. Signals to track: next Praxis 8-K around PDUFA, interim safety updates from EMBOLD open-label extension, and any FDA advisory committee announcement (none scheduled as of the last quarterly filing).