ALTO-300

Alto Neuroscience is testing ALTO-300, a repurposed version of agomelatine, as an add-on to standard antidepressants in major depressive disorder (MDD). The Phase 2b study (NCT05922878) enrolls 321 patients and uses a resting-state EEG biomarker to pre-select the subgroup predicted to respond [1][11]. Agomelatine is already sold in Europe by Servier as Valdoxan, but the FDA never approved it after Novartis withdrew the US application in 2011 citing hepatotoxicity signals and marginal efficacy over SSRIs [7][8]. Alto's thesis: the molecule works, but only for a specific brain-signature subpopulation that previous all-comers trials never isolated. The commercial payoff hinges on whether EEG screening can define a real responder group and whether that group shows enough separation from placebo on MADRS at week 6 to justify a US regulatory filing in a market where generic SSRIs cost pennies and payers will push back hard on any biomarker-gated branded antidepressant. The company announced a favorable interim analysis in 2025 and now guides topline to 1H 2027 [12][13].

Not a novel compound. Agomelatine was approved by the EMA in November 2009 and sold by Servier as Valdoxan and Thymanax [7]. Novartis held US rights and voluntarily withdrew the FDA submission in 2011 after regulators raised concerns about transaminase elevations (liver enzyme spikes that can signal liver injury) and effect sizes that did not clearly beat SSRIs [8]. ALTO-300 is Alto Neuroscience's repackaging of the molecule paired with the company's EEG-based biomarker protocol, which screens patients whose resting-state EEG signatures match a machine-learning-derived responder profile [11]. Because the molecule is already approved in another jurisdiction with an established safety database, the expected US regulatory route is a 505(b)(2) NDA, which allows Alto to reference Servier's prior nonclinical and clinical safety data and reduces the required data package versus a full NDA. No publicly disclosed FDA breakthrough, fast-track, or RMAT designation for ALTO-300. Alto Neuroscience went public on Nasdaq in February 2024 under ticker ANRO [10]. The company's lead compound ALTO-100 failed its Phase 2b trial in October 2024, which raised the stakes on ALTO-300 to deliver positive Phase 2b data [9]. The ALTO-300 Phase 2b (NCT05922878) is recruiting per ClinicalTrials.gov [1]. In 2025 Alto reported a favorable outcome from a pre-specified interim analysis of this trial, and current company guidance places topline readout in 1H 2027 [12][13].

Agomelatine hits three receptors. It activates melatonin receptors MT1 and MT2, which normally sense the hormone melatonin and run the body's internal 24-hour clock, the one that tells you when to sleep. Blocking the 5-HT2C serotonin receptor removes a brake on dopamine and norepinephrine release in the frontal cortex, which many researchers believe contributes to the antidepressant effect [11]. The combination is unusual among antidepressants because it does not touch the serotonin transporter, meaning it works through a completely different pathway than Prozac, Lexapro, or Zoloft. An EEG (electroencephalogram) measures electrical brain activity through scalp sensors, the same test used to detect seizures. Alto's biomarker is derived from resting-state EEG and reflects what the company describes as greater neural variability, or 'noisy cortices,' in responders [11]. The mechanistic rationale: increasing 5-HT2C activity or depleting dopamine, both the opposite of ALTO-300's action, produces greater EEG irregularity in preclinical models, so patients whose baseline EEG shows this signature may represent a neurobiological state that agomelatine can normalize. Alto estimates roughly 50% of MDD patients are biomarker-positive [11]. The validation case is mixed. On the plus side, the drug is approved and sold in Europe for MDD, so there is enough efficacy to clear a European regulator [7]. On the minus side, meta-analyses of agomelatine's original Phase 3 program have been inconsistent, with some showing effect sizes barely larger than placebo and others showing modest but real benefit [4][8]. The 5-HT2C/MT1/MT2 combination is pharmacologically distinct and has not been copied by any approved competitor, so if the drug works it is doing something SSRIs cannot. Alto's bet: previous trials enrolled every MDD patient and diluted the signal, but an EEG-selected subgroup will show a cleaner effect [1][11].

NCT05922878 is a randomized, double-blind, placebo-controlled Phase 2b study enrolling 321 adults with MDD who are already on a stable dose of an antidepressant with inadequate response [1]. Participants receive either ALTO-300 or placebo added to their existing SSRI or SNRI for six weeks. The dosing mirrors the European Valdoxan label (25 mg once daily, with a possible titration to 50 mg), which keeps the hepatic safety profile comparable to the EU post-marketing database. The primary endpoint is change in the Montgomery-Asberg Depression Rating Scale (MADRS), a clinician-rated depression severity score, measured over six weeks in a pre-specified EEG-biomarker-positive subgroup [1]. Design choices worth flagging. Pre-specification of the biomarker subgroup is methodologically clean. Alto locked in the subgroup definition and statistical analysis plan before unblinding, so a positive result in that subgroup is defensible rather than p-hacked. The six-week MADRS endpoint is standard for MDD and matches FDA expectations under the 2018 draft guidance on MDD endpoint development [14]. The adjunctive-to-SSRI framing targets the inadequate-responder segment, which per STAR*D represents the majority of MDD patients (only about one-third of patients achieve remission with their first SSRI) [15]. The concern: adjunctive trials in MDD fail at a high rate because placebo-arm response is inflated by background antidepressant activity, making separation hard. Two prior ALTO-300 Phase 2 studies (NCT05118750, n=91; NCT05157945, n=148) generated the biomarker signal this Phase 2b is designed to confirm [2][3]. That makes NCT05922878 a signal-replication study rather than a discovery study, which is methodologically the right posture. A favorable pre-specified interim analysis was reported in 2025, which does not confirm success but does increase the probability the full readout clears the primary endpoint [12].

Our model estimates a 5% chance this drug is eventually approved. It starts from the historical base rate for Phase 2 drugs in this area (about 24%), then adjusts using ten facts about the trial and sponsor. What moves the number most: it is helped by larger-than-typical enrollment for this phase; it is held back by heavier-than-usual blinding, the sponsor's thin or weak approval record, 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.

Hepatotoxicity. Agomelatine carries liver injury warnings in its European label after post-marketing cases of transaminase elevations and rare acute liver failure [7][8]. European prescribers are required to test liver enzymes at baseline and at weeks 3, 6, 12, and 24 of treatment, and whenever dose increases. This monitoring burden is both a commercial friction point (prescribers have to order and track labs) and a trial interpretation risk: if the week-6 primary endpoint coincides with an LFT draw and the biomarker-positive subgroup shows elevated transaminases, the readout is a confounded safety/efficacy picture rather than a clean efficacy win. Any US regulatory review will revisit this. Biomarker non-replication. Pre-specified subgroup trials in psychiatry have a long history of failing to replicate Phase 2 signals in Phase 3. The EEG signature Alto defined off two small Phase 2 trials (n=91 and n=148) may hold here but still has to survive a Phase 3 [2][3]. The favorable 2025 interim analysis reduces but does not eliminate this risk [12]. Effect size inadequate. Even if the biomarker-positive arm separates from placebo, the magnitude of MADRS improvement needs to be clinically meaningful. The FDA's 2018 draft guidance on MDD endpoint development discusses MCID in the range of roughly 1.6 to 1.9 points on MADRS versus placebo as a reasonable floor for clinical meaningfulness [14]. Agomelatine's European Phase 3 trials often produced separations near or below that threshold [8]. Commercial pricing and competition. If approved in the US as a branded product, ALTO-300 competes against generic SSRIs costing a few dollars a month and sits in a market where zuranolone (Zurzuvae) launched at roughly $20,000 per course and is still struggling for payer coverage, and esketamine (Spravato) is REMS-restricted with in-clinic administration burden. A biomarker-gated branded add-on will face hard payer step-therapy barriers. Execution risk. Alto's Q2 2025 10-Q reported cash, cash equivalents, and restricted cash of approximately $148.1 million as of June 30, 2025, down from $168.7 million at year-end 2024, with management guiding runway into 2029 [16]. Runway is adequate through the readout, but a mixed or negative result would compress the strategic options to a dilutive raise or a wind-down.

Worth watching for one specific data point: the pre-specified biomarker-positive subgroup MADRS separation at week 6 in NCT05922878 [1]. Everything else is noise around that number. If the biomarker-selected arm shows roughly a 2-point or greater MADRS improvement over placebo (above the FDA's 2018 draft-guidance MCID range of about 1.6 to 1.9 points) [14] with a p-value under 0.05 and no new hepatotoxicity signals, Alto has a story that could support a 505(b)(2) NDA or, more likely, a partnership with a large pharma company that wants an MDD asset. The relevant commercial comps are zuranolone (approved, $20K/year, limited payer uptake) and esketamine (approved for TRD, REMS and in-clinic burden), both of which define the ceiling for a biomarker-gated branded MDD add-on and show how hard it is to get payer traction even with approval. If the subgroup separates by less than the MCID floor or shows ambiguous liver signals, this is a dead story in the US market regardless of what the all-comers data shows. Readout window has slipped from earlier 2026 guidance to 1H 2027 per Alto's Q1 2026 update [13]. The 2025 favorable interim analysis is a real but partial de-risking event [12]. Any 8-K from Alto Neuroscience with a trial-related subject line is worth reading in full given the thin pipeline. ALTO-100's Phase 2b failure in October 2024 made ALTO-300 the asset that determines Alto Neuroscience's near-term commercial trajectory [9]. A positive readout likely drives a sizable stock move or a buyout conversation. A negative readout likely drives an existential financing or an orderly wind-down. Binary outcome, binary stakes. Not investment advice.