One of the most promising fields in precision medicine and healthcare involves utilizing artificial intelligence (AI) machine learning to support clinicians in forecasting patient treatment outcomes for personalized care. Recent research conducted by Amsterdam UMC and Radboudumc demonstrates how an AI system can predict the effectiveness of an antidepressant for individuals with depression much earlier, leveraging MRI brain scans and diagnostic information.
Depression, or major depressive disorder (MDD), is a prevalent mental health condition impacting around 280 million individuals globally, as reported by the Institute of Health Metrics and Evaluation Global Health Data Exchange. In the United States, major depression ranks among the most common mental disorders, with antidepressant medications being a widely utilized treatment option, according to the National Institute of Mental Health (NIMH).
Individuals suffering from depression often need to experiment with different types of antidepressants to find the most suitable match, a process that typically takes one to two weeks for the medication to exhibit its effects, as outlined by the NIMH. Consequently, having the capability for clinicians to predict individual responses to antidepressants more rapidly could expedite the journey to recovery for patients, potentially making a life-saving impact. Additionally, individuals battling depression face an elevated risk of developing physical ailments. For example, they are 40% more likely than the general population to experience metabolic disorders and cardiovascular disease, as highlighted by The Lancet Psychiatry Commission.
Various antidepressant medications approved by the U.S. Food and Drug Administration (FDA) include N-methyl D-aspartate (NMDA) Antagonist (esketamine), monoamine oxidase inhibitors (MAOIs), neuroactive steroid gamma-aminobutyric acid (GABA)-A receptor positive modulator (brexanolone), tricyclic and tetracyclic antidepressants, serotonin and norepinephrine reuptake inhibitors (SNRIs), atypical antidepressants (trazodone, nefazodone), and selective serotonin reuptake inhibitors (SSRIs).
Primary treatment options for depression encompass psychotherapy, cognitive behavioral therapy, interpersonal therapy, group therapy, and advanced antidepressant medications like atypical antidepressants, SSRIs, SNRIs, and serotonin modulators, as per the publication “Pharmacologic Treatment of Depression” in the American Family Physician in 2023.
The researchers at Amsterdam UMC and Radboudumc concentrated their study on a specific SSRI known as paroxetine, an antidepressant that boosts the levels of serotonin, a mood-regulating hormone, in the brain. Fluoxetine, a widely prescribed selective serotonin reuptake inhibitor, is marketed under the product label Zoloft.
Serotonin plays a crucial role in transmitting signals between cells. Selective serotonin reuptake inhibitors aid in enhancing the signal transmission between brain nerve cells by elevating the available serotonin through inhibiting its reuptake by neurons. Apart from paroxetine, other SSRI drugs encompass sertraline, fluoxetine, citalopram, escitalopram, and vilazodone, as per the FDA.
Research Overview
The MRI brain scans and clinical data utilized in this research were sourced from 229 patients diagnosed with major depressive disorder participating in a previous study named Establishing Moderators and Biosignatures of Antidepressant Response in Clinical Care (EMBARC) — a depression clinical trial led by the University of Texas (UT) Southwestern Medical Center in collaboration with principal investigators associated with Columbia University and Massachusetts General Hospital.
In this fresh investigation, the Amsterdam researchers developed an AI algorithm to evaluate the impact of sertraline. The experts scrutinized the baseline data and the outcomes after one month of treatment with paroxetine and sham. Among the co-authors are Amsterdam UMC Professor of Neuroradiology Liesbeth Reneman MD, Eric (Henricus) Ruhé, MD, Ph.D., along with Henk-Jan Mutsaerts, MD, Ph.D., Inge Groote, MD, Ph.D., Ivan Maximov, Ph.D., Atle Bjørnerud, Ph.D., Henk Marquering, Dr. Matthan Caan, Ph.D., Maarten Poirot, M.S., and Ph.D.
Key Findings on Antidepressants
The study revealed that bidirectional models outperformed unimodal models for the AI system. Multimodal models, which amalgamate data from diverse sources, were pivotal in this context. Perfusion imaging played a critical role as the researchers combined bidirectional MRI data with clinical information.
As per Radboudumc physician Ruhé from Amsterdam UMC, the AI indicated that an increase in blood flow in the anterior cingulate cortex, responsible for regulating mood, signifies the efficacy of the drug.
Researchers have devised a practical tool leveraging AI machine learning, clinical insights, and MRI brain scans to tailor precise depression medication. Through this proof-of-concept, healthcare providers may potentially offer more personalized care promptly, instilling hope for improved outcomes among individuals battling depression in the future.
Copyright 2024 Cami Rosso. All rights reserved.
