Antidepressant drugs are used to reduce the symptoms of depression, but it is necessary to individualize the treatment depending on each patient and it takes several weeks to verify their effectiveness and determine the appropriate doses. During this time, people with this disorder continue to suffer and, in the most severe cases, there is a risk of suicide.
The results of a new study open a path of hope for these patients, since they have shown that with the use of artificial intelligence (AI) it could be predicted in just one week whether an antidepressant will work. The research has been carried out by scientists from Amsterdam UMC and Radboudumc who by entering a brain scan and an individual’s clinical history into an AI algorithm were able to see up to eight weeks faster whether the drug would work or not.
“This is important news for patients. Typically, it takes six to eight weeks before you know if an antidepressant will work,” said Liesbeth Reneman, Professor of Neuroradiology at UMC Amsterdam. The findings of the work have been published in the American Journal of Psychiatry.
Identify the most appropriate drug to treat severe depression
The researchers studied whether they could predict the effect of the antidepressant sertraline, the most prescribed antidepressant in the United States. In a previous study conducted in the United States, clinical data were analyzed and MRIs were performed on 229 patients with major depressive disorder before and after one week of treatment with sertraline or placebo. The Amsterdam researchers then developed and applied an algorithm to this data.
This analysis revealed that the drug would only work in a third of patients, allowing psychiatrists to adjust their treatment plan. “With this method we can already prevent 2/3 of the number of ‘wrong’ sertraline prescriptions and thus offer a better quality of patient care, because the drug also has side effects,” says Reneman.
“The algorithm suggested that the drug would help those who had a lot of blood flow in the anterior cingulate cortex, the area of the brain involved in regulating emotions. And in the second measurement, a week after the onset, it turned out to be the severity of their symptoms,” says Eric Ruhé, psychiatrist at Radboudumc.
“With this method we can prevent 2/3 of the number of ‘wrong’ sertraline prescriptions and thus offer a better quality of care to patients with depression”
In the future, this new method may help better tailor sertraline treatment to individual patients, since currently the drug is administered to the patient and within six to eight weeks – although this period is often It can last up to six months – check if it works. If the symptoms do not disappear, the patient is given another antidepressant and this process begins again and may be repeated several times.
This standard method usually takes weeks or months to work, with the consequences this has for the quality of life of all those affected, both patients and family members. It also entails an economic and social cost, because as long as the patient continues to experience severe depressive symptoms, he will not be able to fully participate in society.
In one in three depressed patients, the symptoms also do not improve after several phases of treatment and therefore there is an urgent need to find a solution that will allow the effectiveness of antidepressants in severe depression to be determined more quickly. The researchers’ next goal is to improve the algorithm by adding additional information.
