ALTO-100

ALTO-100 is an oral small molecule from Alto Neuroscience (NYSE: ANRO) enrolling a 200-patient Phase 2 trial in bipolar depression [1]. It is a pivot. On October 22, 2024, Alto reported that ALTO-100 missed its primary endpoint in a Phase 2b major depressive disorder trial that had used the company's memory-based cognitive biomarker to pre-select likely responders [2][3][8]. ANRO shares fell about 70 percent the next trading day, from roughly $14.50 to $4.36 [9]. Alto's platform pitch, use EEG signatures and cognitive tests to enrich for patients whose brains look like they will respond, was supposed to be the differentiator that made small psychiatric trials worth running. When the MDD signal did not hold up in the biomarker-selected subgroup, the whole thesis took damage. The bipolar depression trial (NCT06656416) is now the main way to test whether the compound and the biomarker approach have a shot at all. Bipolar depression is a smaller commercial market than MDD but a genuine unmet need, since only a handful of drugs (quetiapine, olanzapine/fluoxetine, lurasidone, cariprazine, lumateperone) are FDA-approved for it. A clean win here would rebuild both Alto and precision-psychiatry as a category. A miss makes the platform very hard to defend.

ALTO-100 is a novel investigational compound, not previously approved anywhere. The active trial is NCT06656416, a Phase 2, randomized, double-blind, placebo-controlled study in 200 adults with bipolar I or II depression, six-week primary endpoint on MADRS [1]. No FDA designations (breakthrough, fast track, RMAT) have been disclosed for ALTO-100 in bipolar depression, and given the prior MDD miss none would be expected at this stage. Completed studies include the Phase 2b MDD trial (NCT05712187, n=301) that read out on October 22, 2024 as a primary endpoint miss in the biomarker-selected subgroup [2][3][8], an earlier Phase 2 in MDD and PTSD (NCT05117632, n=245) [4], and a decentralized pilot (NCT05419869, n=21) [5]. The bipolar Phase 2 initiated in 2024 and is currently recruiting. Alto has not issued firm topline guidance for the bipolar readout. With a six-week treatment window and a 200-patient enrollment, topline data most likely arrive in late 2026 or the first half of 2027 if enrollment stays on pace. Alto's Q1 2026 financial results frame the coming year as when the biomarker-based platform gets tested against real efficacy data, with ALTO-100 in bipolar depression and ALTO-203 in schizophrenia as the two most consequential readouts [6][7].

The molecular target of ALTO-100 is not publicly confirmed by Alto. The compound is described as pro-neuroplasticity/pro-neurogenesis, meaning it is supposed to help brain cells rewire and, in the hippocampus, generate new neurons. Multiple public sources (Wikipedia, patent-mining databases) identify ALTO-100 as the former NSI-189 program, originally developed by Neuralstem Inc. (later Seneca Biopharma, then routed through Palisade Bio) and acquired by Alto around 2021 [10]. NSI-189 was discovered by phenotypic screen against human fetal hippocampal neural stem cells and is proposed to act indirectly through BDNF (brain-derived neurotrophic factor) signaling, though no clean receptor or enzyme target has ever been published. In depression, the mainstream biological theory is that chronic stress and depressive states damage the fine wiring of the prefrontal cortex and hippocampus, and that fast-acting antidepressants (ketamine is the reference case) work by regrowing those connections. Drugs designed around this theory usually engage glutamate signaling, BDNF, or downstream plasticity pathways. Readers should treat the mechanism as pharmacologically ambiguous. The stronger story is not the molecule but the selection strategy. Alto's platform records EEG (a wire-on-the-scalp brain activity readout), administers computerized cognitive tests, and pulls in wearable data, then uses that combined signature to pick patients whose baseline biology looks like it should respond [11]. Importantly, the Phase 2b MDD trial used a memory-based cognitive biomarker rather than an EEG signature as its enrichment cut [2][8]. Alto has not publicly disclosed whether the bipolar trial uses the same cognitive cut, an EEG signature, or a modified composite. The mechanism case for ALTO-100 therefore rests almost entirely on the biomarker being predictive, and the Phase 2b MDD miss in the biomarker-selected subgroup [2][3][8] is the exact result you would not want to see for that thesis.

NCT06656416 is a Phase 2, randomized, double-blind, placebo-controlled trial in adults with bipolar I or II disorder currently in a depressive episode, target enrollment 200 [1]. Primary endpoint is mean change on the Montgomery-Asberg Depression Rating Scale (MADRS) from baseline (Day 1) to Week 6, evaluated in a pre-defined biomarker-selected subgroup rather than the full intent-to-treat population (intent-to-treat meaning the full enrolled cohort analyzed regardless of biomarker status). That structural choice is a yellow flag given the MDD readout: a biomarker subgroup as primary endpoint buys statistical power in a smaller trial, but it means the whole result hinges on the biomarker actually stratifying responders, and the MDD study using the same design principle did not deliver on that biomarker [2][3][8]. Enrollment status is listed as recruiting. Six weeks of active treatment is standard for psychiatric studies of this type. Placebo control is appropriate here since placebo effects in bipolar depression trials can move MADRS scores by 8 to 10 points on their own, so an active-only arm would tell you nothing. Comparator is placebo added to background mood stabilizers, which mirrors how the drug would be positioned commercially as adjunctive therapy (taken on top of existing medication, not replacing it) if approved. There is no active comparator arm, so any efficacy signal will be relative to placebo rather than head-to-head against lumateperone (Caplyta) or cariprazine, which limits the commercial and pricing case this single trial can support.

Our model estimates a 4% 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 more secondary endpoints than usual; 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.

Efficacy risk is dominant. The MDD readout showed Alto's cognitive biomarker did not pick out responders the way the platform predicted [2][3][8]. Alto did not publicly disclose granular numeric MADRS delta or p-values for the biomarker-selected subgroup beyond stating that the primary endpoint was not met and that key secondary analyses did not show benefit over placebo [8]. Without those numbers, external readers cannot distinguish a near-miss (a real but underpowered signal) from a full null, and that ambiguity itself argues for caution when inferring what the bipolar trial will show. Bipolar depression uses a biomarker cut that Alto has not fully specified in public materials, and there is no independent published validation that the platform generalizes across indications. If the biomarker again fails to enrich responders, ALTO-100 misses. Endpoint risk is elevated because the primary is defined in a biomarker-selected subgroup within a 200-patient trial. A modestly underpowered subgroup analysis has real odds of coming in numerically positive but statistically ambiguous, which is the hardest kind of readout to work with commercially. Safety risk is not the main concern. Prior ALTO-100 studies dosed hundreds of patients across MDD and PTSD cohorts without disclosed dose-limiting toxicities [4][5][8], so a late safety signal would be surprising. Commercial risk assumes approval. Bipolar depression already has FDA-approved options (quetiapine, olanzapine/fluoxetine, lurasidone, cariprazine, lumateperone) and generic quetiapine is essentially free. A new branded adjunct needs a clear efficacy or tolerability edge over lumateperone (Caplyta), which is the newest and most commercially aggressive entrant, to win formulary access (placement on insurer-approved drug lists, which determines whether the drug gets prescribed at all). Payers will resist any launch price above roughly $500 to $800 per month without differentiated data. Execution risk includes cash runway, but here the balance sheet is a positive rather than a concern. As of March 31, 2026, Alto reported $264.2 million in cash, cash equivalents, and restricted cash (up from $177.0 million at year-end 2025 after a $120 million PIPE closed in March 2026), against a Q1 2026 net loss of $26.2 million [6]. Management guides that this position, combined with the $100 million registered direct offering priced July 13, 2026 [12], funds operations through 2029, which is well past both the ALTO-100 bipolar readout and the ALTO-203 schizophrenia readout. Cash runway is therefore not the binary here; the binary is efficacy.

Worth watching, calibrated as a small-weight, high-optionality position. ALTO-100 is not a base-case win. It is a low-probability asset attached to a company (Alto Neuroscience, ANRO) whose entire equity story is the biomarker platform. Some investor context: ANRO priced its IPO at $16.00 in February 2024, was trading around $14.50 heading into the October 22, 2024 topline, and closed at $4.36 the day after the miss, a roughly 70 percent single-day drop that wiped out over $200 million in market cap [9]. The stock has since traded well below IPO, with the company funding continued development through two 2026 equity raises rather than through revenue or partnership [6][12]. The single signal worth waiting for is the NCT06656416 primary endpoint: mean MADRS change from Day 1 to Week 6 in the biomarker-selected subgroup, versus placebo [1]. A clinically meaningful separation (at least a 3-point MADRS delta with a tight p-value) would partly rehabilitate the platform thesis and reset ANRO. A miss, or a modest numeric edge that does not clear statistical significance in the pre-specified subgroup, effectively closes the door on ALTO-100 as a commercial asset and forces the company to lean harder on ALTO-203 and the rest of its pipeline. Check-back timing is set by enrollment. Six-week treatment plus follow-up plus data lock puts a realistic recruit-to-readout window at 12 to 18 months, meaning late 2026 to mid 2027 for topline if the trial holds pace. Watch 8-K filings for enrollment-completion announcements, which usually precede readout by six to nine months. The most instructive read on Alto right now is not ALTO-100 alone but the shape of its 2026 SEC filings and investor presentations: how the biomarker platform is being described to investors after the MDD miss, whether the bipolar biomarker cut is explicitly defined, and how R&D is allocated across the pipeline now that funding pressure has been relieved through 2029.