Psychiatric drug development has one of the lowest success rates of any therapeutic area. One analysis found that just 7.9 percent of psychiatric drug candidates entering clinical development ultimately received regulatory approval, compared with an average of 9.6 percent across all therapeutic areas.
One challenge is that psychiatric diagnoses can encompass patients with very different underlying biology. Two people with major depressive disorder (MDD), for example, may experience similar symptoms but have different biological processes contributing to their illness and respond differently to the same treatment.
Despite this, clinical trials typically group patients by diagnosis, relying heavily on symptom scales to assess whether an experimental drug works. However, if a treatment only benefits a subset of patients, its effects may be harder to detect across the whole trial population, potentially obscuring a benefit in those most likely to respond.
There is now a growing movement to incorporate biological markers to stratify patients, monitor treatment response, and improve clinical trial design. Alto Neuroscience is among the companies pursuing this approach, combining biological and cognitive measurements with machine learning to guide psychiatric drug development.
But can this approach improve the odds of success? In 2024, Alto's investigational drug ALTO-100 failed to meet the primary endpoint in a Phase 2b trial for MDD. The result raised questions about the promise of biomarker-guided drug development — and what evidence is needed to show that selecting patients based on their biology can improve clinical trial outcomes.
When a trial misses its target
ALTO-100 is one of seven investigational drugs in Alto Neuroscience’s pipeline. It is designed to influence pathways involving brain-derived neurotrophic factor (BDNF), a protein involved in neuronal health and plasticity. So, Alto’s biomarker strategy was aimed at finding patients with a memory-based cognitive biomarker, who were deemed more likely to respond to the drug.
ALTO-100 initially showed encouraging results. In January 2023, Alto reported that its Phase 2a study had found greater improvements in depression symptoms among patients with MDD who matched the biomarker profile associated with the drug’s proposed mechanism. After six weeks of treatment, these patients had an average 15.5-point reduction on the Montgomery–Åsberg Depression Rating Scale (MADRS), compared with a 10.6-point reduction among patients who did not match the profile. The biomarker-defined group also had a higher response rate, with 61 percent experiencing at least a 50 percent reduction in symptoms, compared with 33 percent of the other patients.
The findings supported advancing ALTO-100 into a larger Phase 2b trial. But the study had limitations: It was open-label, meaning participants and researchers knew the treatment was being given, and the comparison was between patients with and without the biomarker profile rather than between the drug and a placebo within a biomarker-selected population.
In 2024, that Phase 2b trial failed. ALTO-100 did not significantly outperform placebo on the main measure of depression symptom improvement, challenging the earlier findings and raising questions about the reliability of the biomarker-guided approach.
Alto Neuroscience has since analyzed the full dataset and CEO Amit Etkin now believes that the results exposed a problem with trial execution rather than the drug or its biomarker.
The trial included two prespecified treatment groups: patients receiving ALTO-100 alone and those receiving it alongside an existing antidepressant. According to Etkin, medication adherence in the monotherapy group was much lower than expected. In contrast, the biomarker-positive subgroup receiving adjunctive treatment showed what he described as a promising treatment effect, although the analysis was not powered to demonstrate statistical significance.
“The people who were not taking the medication obviously did not [show the effect],” Etkin told DDN.
Etkin argued that inconsistent medication use and problems verifying participants' eligibility can make psychiatric trials difficult to interpret. He also raised concerns about potential professional patients — people who enter clinical trials primarily for financial compensation rather than to seek treatment — suggesting that their participation could undermine the reliability of trial results.
Making clinical trial recruitment more rigorous
Alto has responded by tightening its requirements for both trial sites and individual participants. Rather than relying solely on screening questionnaires or a patient's reported medical history, the company is now verifying diagnoses and existing treatments through medical records, prescriptions, and refill histories. It can also check for the presence of medication in blood or urine.
The company has incorporated these measures into its ongoing Phase 2b trials of ALTO-207 for treatment resistant depression, ALTO-100 for bipolar depression, and ALTO-300 for MDD. According to Alto’s 2026 update, its eligibility review process screened out 11 percent of patients considered for the ALTO-100 trial and 13 percent of those considered for ALTO-300 who might otherwise have been enrolled.
The company also reported that 97 percent of samples in periodic pharmacokinetic testing showed expected drug concentrations. In the ALTO-300 trial, 100 percent of patients had demonstrated ongoing adherence to their existing antidepressant treatment, as assessed through urinalysis during the study.
However, more stringent screening comes with trade-offs. Alto now expects topline results from the ALTO-300 trial in the first half of 2027 and from the ALTO-100 bipolar depression trial in mid-2027, later than previously anticipated. Etkin acknowledged that stricter screening can slow recruitment, reduce the number of sites a company can use, and limit how many studies it can conduct simultaneously.
Whether these changes will improve the reliability of efficacy results remains to be seen. Better adherence and more rigorous eligibility checks can help ensure that a trial tests a drug under the intended conditions, but they cannot guarantee that the drug will outperform placebo or that a biomarker will reliably identify those most likely to benefit.
A more targeted approach to psychiatric drug development
The experience with ALTO-100 highlights the difficulty of translating promising early findings into successful clinical trials. But Alto’s work on ALTO-300 offers another perspective on how biomarker-guided drug development might progress.
ALTO-300, also known as agomelatine, acts on melatonin receptors and blocks the serotonin receptor 5-HT2C, influencing dopamine and norepinephrine signaling. In a Phase 2a trial, patients with MDD who matched an EEG-based biomarker profile experienced greater improvements in depression symptoms when ALTO-300 was added to their existing antidepressant treatment.
Like ALTO-100’s Phase 2a study, this was an early, exploratory trial rather than a placebo-controlled test of the drug’s efficacy in biomarker-positive patients. Crucially, however, researchers have thoroughly investigated the biology underlying ALTO-300’s EEG biomarker. The signal is characterized by greater irregularity in brain activity and has been linked to reduced connectivity in the medial prefrontal cortex, a brain region implicated in depression. In mice, activating the 5-HT2C receptor increased the EEG signal, providing a potential link between the biomarker and the drug’s mechanism of action.
“It seems like it’s a biomarker of the drug response, rather than a biomarker of the disease process itself,” Etkin said. This provides a biological rationale for using the biomarker to identify patients more likely to benefit from ALTO-300.
The approach is also designed to be practical. In the Phase 2a trial, around half of the EEG recordings were collected in patients’ homes, Etkin said. The biomarker relies on a signal from a single electrode, which he believes could make it easier to implement beyond specialist research settings.
Alto is now evaluating ALTO-300 in a randomized, double-blind, placebo-controlled Phase 2b trial in patients with MDD. The study includes patients with and without the EEG biomarker, allowing researchers to examine whether the signal helps distinguish those who benefit from treatment.
The results could show whether biomarker-guided trial design can reliably identify patients most likely to benefit from a treatment — and ultimately help more patients receive treatments that work for them.
