OPADE: Ending the trial-and-error era in depression treatment

For all the advances in modern medicine, the treatment of depression remains, in many ways, lacking the precision seen in other fields. Diagnosis is largely based on conversation, while treatment is often a process of trial and error, particularly when drugs are prescribed. For millions of patients, the path to recovery can be long, uncertain and deeply challenging, with treatment programmes that are often difficult to endure.

“Depression is a really big problem of our time,” says Dr. Giulio Corrivetti, Psychiatrist and Principal Investigator of the OPADE project, and Vice President of the European Biomedical Research Institute of Salerno (EBRIS), as well as Director of the Department of Mental Health at ASL Salerno. “It is already one of the most common disorders in the world, and it may become the leading one in the years ahead. But for psychiatrists, it is still very difficult to characterise.”

A systematic updated analysis on the prevalence and global burden of mental disorders from 1990 to 2023, within the Global Burden of Disease (GBD) 2023 study, published in The Lancet this year, demonstrated that in 2023, approximately 1.17 billion people worldwide are living with a mental disorder, nearly twice as many as in 1990. The increase in burden has mainly been seen in anxiety disorders and major depressive disorder.

At the heart of this mental disorder challenge is a fundamental gap: unlike many other areas of medicine, psychiatry still lacks objective biological markers to guide diagnosis and treatment. Where other specialities can draw on measurable data to inform clinical decisions, depression is still largely assessed through behavioural observation, patient narratives and clinician interpretation.

This reliance on subjective assessment inevitably introduces uncertainty into both diagnosis and treatment, shaping not only how the condition is understood, but also how it is treated and managed. “We have some bias from the subjective style of observation,” Corrivetti explains. “We organise information from observation and conversation with the patient, rather than from biological data that can guide us. This makes it much more difficult to define the condition and to choose the most effective treatment.”

The consequences are profound. Across Europe, only a small proportion of patients receive what could be considered optimal treatment. Many never reach specialist care at all. And even for those who do, the process of finding an effective pharmacological antidepressant treatment can take weeks or months. “If they don’t see a response within a month, it’s very hard to keep a patient on a drug,” adds Professor Riccardo Panella, Head of Research at EBRIS. “So, if we can predict in advance which option will be most effective, we can improve both outcomes and adherence.”

Biological insight
It is this ambition to move from reactive, trial-and-error prescribing to informed, predictive decision-making that sits at the core of OPADE. The project represents a shift in how depression is understood and treated, moving from a predominantly behavioural model toward one that integrates behavioural parameters with biological markers, paving the way for a more detailed characterisation of the disease, integrating data and measurable indicators. It brings together clinical centres across Europe and beyond to build one of the most comprehensive datasets yet assembled in this field. Patients diagnosed with major depressive disorder are followed over time, with repeated assessments capturing both clinical progression and underlying biological changes.

“We recruited 350 patients, which is already a very strong cohort for this kind of study,” says Dr. Alessandra Marenna, who leads the Microbiome Core at EBRIS and is part of OPADE coordinating team. “From every patient, we take behavioural data, biological samples like blood, stools, saliva, as well as other indicators.”

The project is designed to uncover relationships that cannot be captured through any one type of observation alone. By combining biological, behavioural and clinical data, OPADE is building a more complete picture of the disease that reflects its underlying complexity rather than reducing it to a single set of symptoms. Genetics, microbiome composition, immune response and metabolic activity are analysed alongside behavioural assessments and patient-reported experience, creating a multi-dimensional view of the condition.

“There is a very large variability in depression,” Corrivetti explains. “Genetic variability, behavioural variability and it is very difficult to connect these. We are looking to address this variability by bringing together the different layers of data to better understand how the biological and behavioural aspects of the disorder interact.”

Crucially, this approach reflects a recognition that depression is not just a disorder of the mind, but one that is deeply rooted in the body. “The depression is in the mind, but also in the brain and in the body,” Corrivetti adds. “So, from this type of analysis, we can gain the knowledge we need to change how we approach treatment.”

The real breakthrough, however, lies not just in collecting this data, but in making sense of it. The biological, behavioural and clinical information gathered forms a complex multi-omics dataset, which captures the genetic, microbiome, immune and metabolic signals alongside patient experience. This is then integrated through artificial intelligence and machine learning, identifying patterns and correlations that would be impossible to detect manually.

“All these data together are fed to our AI partner,” explains Riccardo Panella. “They find patterns and help us understand which biomarkers are more predictive than others — particularly in terms of how patients are likely to respond to different treatments.”

Rather than searching for a single defining marker, the AI is identifying combinations – clusters of biological signals that, together, can predict how a patient will respond to treatment. “It might be one metabolite, three microbiome markers, two microRNAs,” Panella further explains. “But all together, they become very predictive for response to therapy.”

This iterative process allows the model to refine itself over time, narrowing down from a broad, exploratory dataset to a focused set of clinically useful indicators. “We started with an unbiased approach,” adds Marenna. “Now we are filtering down to the biomarkers that can actually be used in practice.”

Clinical decision-making
The ultimate goal for OPADE is to translate this analytical capability into a tool that can support real-world clinical decision-making. Today, antidepressant prescribing often begins without a clear indication of how a patient will respond, requiring weeks of observation before effectiveness can be assessed. In the OPADE vision, that uncertainty is replaced with early, data-informed guidance.

“Patients will take behavioural tests and provide biological samples, and then in a matter of days you can know which treatment will be most effective,” says Panella. “This will allow clinicians to move more quickly towards the most appropriate therapeutic approach, rather than working through a sequence of options over time.”

Crucially, this is not about reducing depression to a set of rigid categories. Instead, it reflects a deeper understanding of the condition’s variability, not just between patients, but in how individuals respond to treatment. “We don’t have just responders and non-responders,” notes Corrivetti. “We have different levels of efficacy. Treatment response exists along a spectrum, shaped by the complex interaction of biological and behavioural factors.”

“By identifying patterns within this variability, the system can begin to group patients into more meaningful subtypes,” he continues. “Each one of these can then be associated with a different likelihood of responding to specific therapies.”

The result is not personalised medicine in the sense of developing entirely new treatments for each individual, but a new knowledge that will allow for more precise patient stratification: matching patients to the therapies most likely to work for them based on their underlying profile. It is a shift away from one-size-fits-all prescribing towards a model that reflects the complexity of depression itself.

Faster response
For patients, this shift could be transformative. One of the most persistent challenges in depression treatment is the delay before antidepressants take effect – typically four to six weeks. During this time, many patients experience little improvement and may abandon treatment altogether.

“A lot of patients drop out after the first month,” says Corrivetti. “But when they feel an effect in a few days, with energy and well-being, adherence to treatment plans improves dramatically.”

By predicting effective treatments from the outset, OPADE aims to shorten the timeline between prescription and a patient beginning to feel the benefits of therapy. In turn, this has the potential to reduce both the duration and severity of depressive episodes, limiting the prolonged uncertainty that often characterises treatment.

“If we can show an effect earlier in the right group of patients,” adds Marenna, “it becomes much easier to keep them engaged with the treatment. That improves adherence and ultimately leads to better clinical outcomes.”

The implications extend beyond clinical practice. The absence of objective biological markers has not only shaped how depression is diagnosed and treated, but also how it is perceived, contributing to a persistent stigma distinguishing it from many other medical conditions. Without measurable indicators, the disorder is often framed in subjective or personal terms, rather than as a clearly defined clinical condition.

By grounding depression in measurable biological processes, OPADE has the potential to shift that perception. “With a clear set of tests, we can approach the disease in a more objective way,” says Panella. “This can give the disease the dignity of being defined with measurable parameters.”

For Corrivetti, this reflects a broader need to reconnect the psychological and biological dimensions of the disorder. “Depression is in the mind, but also in the brain and in the body,” he explains. “From this kind of analysis, we can gain the knowledge we need to change how we approach treatment. In a field still shaped by stigma, this shift could be significant and will help us reframe depression not as a personal failing, but as a complex, biologically grounded condition requiring targeted and informed intervention, just like other diseases.”

Future discovery
While the immediate focus is on treatment response, the project’s long-term value may lie in the data itself. The extensive multi-omics dataset being built will form a resource for future research, enabling new hypotheses to be tested and new therapeutic targets to be identified.

“We are collecting all the information we can,” says Panella. “Even things we are not analysing today could become important in the future. This allows us to build a bank of samples, information and data that will be an added value for the entire scientific community and have the potential to support future analysis and research. This includes factors such as diet and the gut–brain axis, areas of growing interest in mental health research.”

Ultimately, OPADE is about more than improving individual prescriptions. It is about aligning psychiatry with the broader shift toward precision medicine seen across healthcare. “The future is more personalised and closer to the patient,” says Corrivetti. “We want to bring psychiatry in line with other medical disciplines, but this will take time. We have to consider regulatory frameworks, clinical guidelines and healthcare systems, and all will need to be adapted.

“But the direction of travel is clear,” he continues. “Paving the way for new clinical practices based on biomarkers and multiomic approach.”

OPADE is not just attempting to simplify the biological, psychological and social dimensions of depression, but to provide a framework for understanding them more clearly.

And, by combining clinical data, behavioural test, response to treatment biological data and AI, it offers a path beyond trial and error towards a future where treatment is faster, more effective and grounded in the individual realities of each patient. “This may mark the beginning of a new era in mental health care,” Corrivetti concludes, “one where psychiatry finally joins the precision medicine revolution already transforming the rest of healthcare.”