RAP-219

RAP-219 is Rapport Therapeutics' lead drug, an oral small molecule that blocks a specific subset of AMPA glutamate receptors found only in the forebrain, aimed at focal-onset seizures in adults [1]. The compound is now in two Phase 3 trials, FOCUS-1 (NCT07563881, n=333) and a parallel study (NCT07594119, n=312), both recruiting adults with focal seizures inadequately controlled by existing antiepileptic drugs [2][3]. The commercial pitch is straightforward: match the efficacy of perampanel (Eisai's Fycompa, the only approved AMPA receptor blocker for epilepsy) while avoiding the psychiatric side effects that come from blocking AMPA receptors everywhere in the brain. Perampanel carries a boxed warning for serious psychiatric and behavioral reactions [4]. The competitive bar is cenobamate (SK Biopharm's Xcopri), which delivered ≥50% seizure-reduction responder rates of 57.8% at 200 mg and 60.4% at 400 mg versus 21.7% placebo in its pivotal Study C017, the strongest efficacy seen in adjunctive focal epilepsy [10]. If RAP-219 clears that efficacy bar and reads out clean on behavioral safety, it slots into a real commercial gap.

RAP-219 is a novel small molecule, not approved anywhere for any indication. The program advanced to Phase 3 in 2026 based on a Phase 2a proof-of-concept study (NCT06377930, n=30) in patients with refractory focal epilepsy who already had responsive neurostimulation devices implanted. Responsive neurostimulation (RNS) is a closed-loop brain implant that continuously monitors electrical activity in the seizure focus and delivers a brief stimulation pulse when it detects an abnormal pattern, aiming to abort the seizure before it spreads. The trial used the device's brain-recording output to measure long ictal episodes (prolonged seizure-like electrical bursts detected by the implant) as a pharmacodynamic readout (a direct measurement showing the drug is engaging its biological target in the brain) [5]. Phase 2a topline was positive: 85.2% of patients achieved ≥30% reduction in intracranial long episodes and 72.0% achieved ≥50% reduction in clinical seizures over eight weeks, with the follow-up period showing a 90% median reduction in clinical seizures in weeks 9-12 and a 59% median reduction in weeks 13-16 versus baseline; the drug was generally well tolerated with a 10% discontinuation rate [11]. Two Phase 3 trials are now recruiting: FOCUS-1 (NCT07563881, n=333) and a companion study (NCT07594119, n=312), both measuring median percent change in seizure frequency as the primary endpoint [2][3]. Rapport is running the program without any publicly disclosed FDA expedited designations such as breakthrough therapy or fast track. A separate Phase 2 study in adults with bipolar I disorder (NCT07046494, n=253) is active and not recruiting, testing the compound against the Young Mania Rating Scale [6]. An open-label long-term safety extension (NCT07219407) is enrolling patients from the earlier focal epilepsy studies. Given both Phase 3 trials are still recruiting, primary readouts are most likely 2027 to 2028, with an NDA submission plausible in 2028 to 2029 if both hit their primary endpoints. Cash position supports that timeline: Rapport ended Q2 2026 with $436.1M in cash and short-term investments and quarterly operating burn of $41.5M, with company guidance that current cash funds operations into the second half of 2029, past the expected Phase 3 readout window without a required dilutive raise [12].

AMPA receptors are the main gates through which the neurotransmitter glutamate excites neurons across the brain. Too much AMPA signaling drives the runaway electrical activity of a seizure, and blocking these receptors reduces seizures. Perampanel proved this works clinically: it is approved for focal and generalized epilepsy and cuts seizure frequency meaningfully as adjunctive therapy [4]. The problem is that AMPA receptors sit on almost every neuron, so blocking them everywhere produces cognitive and psychiatric side effects severe enough to warrant a boxed warning [4]. RAP-219 tries to solve this by targeting only the AMPA receptors that carry an auxiliary protein called TARP gamma-8, encoded by the gene CACNG8. TARP gamma-8 is expressed almost exclusively in the forebrain (cortex and hippocampus), which is where most focal seizures originate. Cerebellum and brainstem use different TARP subtypes, so a TARP gamma-8-selective blocker should leave motor control and autonomic regions largely untouched. Open Targets records CACNG8 with genetic and functional evidence linking it to focal epilepsy phenotypes, and TARP gamma-8 controls AMPA receptor trafficking and gating in exactly the circuits where focal seizures start [7]. Multiple companies have chased this target; Eli Lilly disclosed LY3130481 years ago but did not advance it. Whether RAP-219 achieves clean forebrain selectivity at doses high enough to stop seizures is the entire clinical question.

FOCUS-1 (NCT07563881) is the anchor Phase 3, enrolling 333 adults with focal-onset seizures inadequately controlled on one to three background antiepileptic drugs, with median percent change in seizure frequency during the maintenance period as the primary endpoint [2]. The parallel Phase 3 (NCT07594119) enrolls 312 patients on the same primary endpoint [3]. The two-trial structure is standard for antiepileptic drug NDAs: FDA has approved every recent adjunctive focal AED (perampanel, cenobamate, brivaracetam) on the basis of two placebo-controlled add-on trials using this exact endpoint framework. Both trials are recruiting. Rapport has not publicly disclosed the specific tested doses or the placebo-to-active randomization ratio in the primary registry entries, which is a modest transparency gap for evaluating dose selection versus Phase 2a exposures. The Phase 2a study (NCT06377930) was small (n=30) but methodologically interesting: it enrolled patients with implanted responsive neurostimulation devices and used long episode counts recorded by the device as a pharmacodynamic surrogate. Long episodes showed greater than 90% median concordance with electrographic seizures in this cohort, which is what allowed Rapport to use the surrogate as a translational bridge into Phase 3 [11]. The design provided a mechanism-of-action readout at low cost but was open-label and not powered to establish clinical efficacy against placebo. The bipolar I Phase 2 (NCT07046494, n=253) is a separate program using the Young Mania Rating Scale and is now closed to enrollment; a positive readout would open a second indication for the same molecule [6].

Our model estimates a 28% chance this drug is eventually approved. It starts from the historical base rate for Phase 3 drugs in this area (about 51%), then adjusts using ten facts about the trial and sponsor. What moves the number most: it is helped by a non-randomized design; it is held back by the sponsor's thin or weak approval record, smaller-than-typical enrollment for this phase, and weak or limited earlier-phase results. The other facts land near average for this stage, so they leave the estimate roughly where the base rate put it.

Efficacy risk. TARP gamma-8 selectivity in animal models does not guarantee that human focal seizure networks respond to partial forebrain AMPA modulation at tolerable doses. Phase 2a was open-label, n=30, without placebo control, so even the strong directional response (72% ≥50% clinical seizure responders) carries residual placebo-effect and selection uncertainty going into a placebo-controlled Phase 3 [11]. Safety risk splits two ways. On the psychiatric front, the entire clinical case for RAP-219 rests on avoiding perampanel-style behavioral toxicity, and any signal of aggression, irritability, or suicidal ideation in Phase 3 collapses the differentiation story [4]. On general CNS tolerability, AEDs in this space routinely show dose-limiting somnolence, dizziness, and gait disturbance; the Phase 2a discontinuation rate of 10% was reassuring but the safety database at Phase 3 scale is still open [11]. Execution risk. Rapport Therapeutics IPO'd in June 2024 and is running two Phase 3 trials plus a bipolar Phase 2 on a single-molecule pipeline [9]; however, the Q2 2026 balance sheet of $436.1M with runway guided into 2H 2029 means the company can reach Phase 3 readout without a required equity raise, materially lowering the near-term dilution and cash-runway concern [12]. Commercial risk. Two forces compress branded pricing power in this class simultaneously. Cenobamate (Xcopri) is the efficacy benchmark, with 57.8% (200 mg) and 60.4% (400 mg) ≥50% responder rates versus 21.7% placebo in Study C017 [10] - RAP-219 must approach these numbers to command premium pricing. Perampanel generic entry: Fycompa's estimated generic launch date is on or around July 1, 2026 based on Eisai's patent estate and ANDA filings, meaning by the time RAP-219 could launch (2028-2029) it would be entering a market with a fully genericized branded predecessor as the low-cost anchor [13]. Positioning becomes 'clean-behavioral-safety differentiation at a premium to generic perampanel and at parity with branded cenobamate' - a narrower commercial wedge than the writeup would otherwise suggest.

Worth watching. The mechanistic story is one of the more coherent ones in the current epilepsy pipeline: perampanel already proved the target works, and forebrain selectivity via TARP gamma-8 is the obvious next move to strip out the psychiatric toxicity. The Phase 2a follow-up data is more directionally supportive than the surrogate-only framing suggested, and the Q2 2026 cash position removes near-term financing overhang. Specific data points to watch: first, any interim update on FOCUS-1 (NCT07563881) enrollment pace, because two concurrent Phase 3 trials on a single-asset company cannot afford recruitment delays even with cash to spare; second, the Phase 2 bipolar I readout (NCT07046494), which is active-not-recruiting and could deliver in 2026 or 2027, opening a second indication and validating the CNS-selectivity story outside epilepsy [6]; third, any Rapport disclosure about psychiatric adverse event rates in the ongoing open-label safety extension (NCT07219407), because that is the differentiation pillar; fourth, cenobamate real-world responder data as the moving efficacy benchmark. The 35.6% model score is a fair anchor; a clean bipolar signal and confirmed on-track Phase 3 enrollment would move it toward 45%, while any psychiatric adverse event signal or enrollment slowdown would drop it below 25%.