RAP-219

RAP-219 is Rapport Therapeutics' lead drug, a once-daily oral agent for focal epilepsy designed to block the brain's main excitatory signal selectively in the forebrain where seizures start, while sparing the cerebellum and brainstem [1]. The pitch is simple: do what perampanel does for seizure control without the sedation and psychiatric side effects that limit perampanel's use [2]. A Phase 2a proof-of-concept in refractory focal epilepsy (NCT06377930, n=30) read out positive in 2025, the Phase 3 registrational trial FOCUS-1 (NCT07563881, n=333) launched in Q2 2026 following an FDA-accelerated path, and Rapport has expanded into bipolar I disorder as a second indication (NCT07046494, n=250) [3][4][9]. Rapport ended Q1 2026 with $476.8M cash, guiding runway into H2 2029 - so FOCUS-1 readout is fully funded without a forced dilution [11]. This is the asset Rapport's June 2024 IPO ($174M total raise including private placement) was priced on, and FOCUS-1 is the company-defining catalyst [12].

Novel compound, never approved anywhere. The key study is FOCUS-1 (NCT07563881), a Phase 3 trial of 333 adults with focal seizures using median percent change in seizure frequency as the primary endpoint - the standard FDA-acceptable endpoint for antiepileptic approvals [3]. FOCUS-1 began enrolling in Q2 2026 following FDA agreement to accelerate into registrational trials; company guidance points to topline in 2027, though with a Q2 2026 start a 2028 readout is also plausible depending on enrollment pace. A parallel Phase 2 study in bipolar I disorder (NCT07046494, n=250) uses the Young Mania Rating Scale, with topline expected in H1 2027 - opening a second commercial lane [4]. An open-label long-term safety extension (NCT07219407) is underway for Phase 2a completers, with initial data expected H2 2026 [5]. No FDA breakthrough, fast track, or orphan drug designations have been publicly disclosed. The drug carries an RxNorm CUI (2728018), reflecting NLM recognition as a distinct investigational entity [6]. The Eilat XVII conference review of new antiepileptic drugs in development includes RAP-219 among the assets to watch in this generation [1].

AMPA receptors are the protein channels on neurons that handle most fast excitatory signaling - when glutamate hits them, sodium rushes in and the neuron fires. Seizures are runaway firing. Block AMPA receptors and you damp the storm. Perampanel proved this works clinically and is approved for focal seizures, but because AMPA receptors sit on neurons everywhere - cerebellum, brainstem, forebrain - blocking them broadly causes sedation, dizziness, and a black-box psychiatric warning [2]. RAP-219's trick is targeting TARP γ-8 (transmembrane AMPA receptor regulatory protein γ-8, gene name CACNG8), a helper protein that escorts AMPA receptors to the cell surface and tunes their gating. TARP γ-8 is concentrated in the forebrain - hippocampus and cortex, where focal seizures originate - and largely absent from cerebellum and brainstem [7]. By binding TARP γ-8 rather than the AMPA receptor itself, RAP-219 dials down excitatory signaling only where the seizures live. The genetic case is moderate: CACNG8 shows up in Open Targets with epilepsy and seizure association scores around 0.60, and TARP γ-8 knockout mice are resistant to chemically induced seizures [7]. Validation comes from pharmacology, not human genetics. The mechanism is biologically reasonable and the target-selectivity logic is clean - whether forebrain-selectivity actually translates to a cleaner CNS side-effect profile in humans is what Phase 2 began to answer and Phase 3 has to confirm.

The Phase 2a study (NCT06377930, n=30) was a clever design that used patients with implanted NeuroPace RNS (Responsive Neurostimulation) devices - intracranial electrodes that continuously record seizure-related electrical activity. Primary endpoint was change in 'long episode' frequency on intracranial EEG, a pharmacodynamic biomarker that tracks days-to-weeks ahead of clinical seizures, letting Rapport read out signal on 30 patients instead of waiting on a 200-patient clinical-seizure trial [9]. Smart capital efficiency, but two caveats: the long-episode endpoint is not FDA-accepted for approval, and the study was open-label single-arm - the headline numbers (see PoS) are baseline-subtracted, not placebo-controlled. There's also a representativeness gap: the Phase 2a population was device-implanted and highly refractory (RNS patients have already failed multiple medications and qualified for surgery), while FOCUS-1 enrolls a broader refractory focal epilepsy population. Pharmacodynamic responsiveness in the sickest subset doesn't guarantee the same magnitude of clinical effect in less-refractory patients - an under-appreciated source of false-positive Phase 2-to-Phase 3 risk. The Phase 3 FOCUS-1 trial (NCT07563881) reverts to the standard regulatory paradigm: 333 adults with focal seizures, randomized, adjunctive (add-on to existing AED therapy, not replacement), median percent change in seizure frequency over the maintenance period as the primary endpoint versus placebo [3]. That's the same endpoint structure used to approve perampanel, brivaracetam, and cenobamate. The long-term safety extension (NCT07219407) feeds into the chronic-exposure data package FDA requires for new CNS drugs [5]. Trial design is conventional and FDA-aligned. Enrollment risk exists - focal epilepsy trials compete for the same refractory patients across programs - but Rapport has a head start and the Tenacia Greater China partnership adds Phase 3 sites in China that should help [12].

Our model gives this drug a 12% chance of eventually being approved. That starts from a 24% historical approval rate for Phase 2 drugs in this area, then adjusts based on ten facts about the trial and sponsor. The estimate gets a small boost from a non-randomized trial design and more secondary endpoints than usual, but is pulled down by the sponsor's thin approval record and weak earlier-phase results. The remaining factors are close to average for this stage, so they don't move the number much.

Three live failure modes. First, efficacy: the Phase 2a used 30 patients, no placebo arm, and an intracranial-EEG biomarker the FDA hasn't blessed for approval. If the pharmacodynamic signal on long episodes doesn't translate into a placebo-adjusted ~20-25% median seizure-frequency reduction in FOCUS-1, the drug doesn't separate from placebo and the program dies. Approved adjunctive AEDs cluster in that range. Second, safety and tolerability: the entire commercial case rests on RAP-219 being cleaner than perampanel on CNS side effects. If patients in Phase 3 show somnolence, dizziness, or psychiatric events at rates similar to perampanel, the differentiation collapses and the drug becomes a me-too in a market with multiple generic AEDs and cenobamate as the recent branded entrant [2]. The Phase 2a population was small and pre-selected for tolerability in the device-implanted cohort - tolerability in a broader Phase 3 population is not yet established. Third, commercial: even with approval, refractory focal epilepsy is crowded. Generic levetiracetam, lamotrigine, and topiramate dominate first and second line. Branded options - perampanel, brivaracetam, cenobamate - fight for the adjunctive market. Payers will demand head-to-head differentiation or steep discounting. Cenobamate's launch (SK Biopharmaceuticals/Xcopri) showed branded AED uptake is possible but slow. Cash runway extends into H2 2029, so the financial risk of a forced raise pre-readout is low - but Rapport remains a single-asset company and FOCUS-1 failure likely takes the equity to cash value [11].

Worth watching. The mechanism logic is the cleanest version of the 'selective glutamatergic modulation' thesis anyone has run into Phase 3, and the precedent drug (perampanel) is approved - so this is a tolerability bet, not a target bet, which is a meaningfully easier scientific question. The specific data point that would turn this from interesting to signal: FOCUS-1 topline showing ≥20% placebo-adjusted median seizure-frequency reduction AND a CNS adverse-event profile materially cleaner than perampanel's label. Either alone is not enough - efficacy without differentiation is a me-too in a generic-dominated market; tolerability without efficacy is a failed drug. Near-term checkpoints: long-term OLE data (NCT07219407) in H2 2026 for chronic safety, bipolar I Phase 2 topline (NCT07046494) in H1 2027 for indication expansion, and FOCUS-1 topline in the 2027-2028 window. Rapport Therapeutics ($RAPP, ~$1.3B market cap) is a single-asset story until the bipolar I Phase 2 reads out, so the equity reflects the binary - small position sizing if you're playing it, with the silver lining that cash runway to H2 2029 ($476.8M as of Q1 2026) means dilution before readout is unlikely [11]. The Tenacia Greater China deal ($20M upfront, up to $308M milestones, March 2026) is modest external validation of the asset and adds Phase 3 sites in China [12]. The Lilly heritage (Rapport spun out of Lilly's neuroscience group) gives mechanistic credibility but doesn't guarantee execution. On IP: composition-of-matter coverage on a small-molecule chemical entity filed mid-2010s through early 2020s, plus any patent term extension, should provide exclusivity into the late 2030s - comfortably beyond an expected late-2020s NDA filing - but specific patent expiry dates were not located in this pass and should be confirmed against the S-1 IP section before any sizing decision.