Brain Scan with AI Can Predict Whether the Antidepressant Sertraline Will Work

News
Article

A secondary analysis found a brain scan with an AI algorithm predicted within a week 1/3 of patients would respond to sertraline and 2/3 would not.

Brain Scan with AI Can Predict Whether the Antidepressant Sertraline Will Work

Maarten G. Poirot, MS

Credit: ResearchGate

Because of artificial intelligence (AI), patients may no longer have to wait weeks to see if an antidepressant will work to improve their depressive symptoms—a brain scan programmed with an AI algorithm can predict within a week whether the antidepressant, sertraline, will work, according to a new study.1

"This is important news for patients,” said investigator Liesbeth Reneman, professor of neuroradiology at Amsterdam UMC, in a press release.2 “Normally, it takes 6 to 8 weeks before it is known whether an antidepressant will work.”

Sertraline, one of the most frequently prescribed drugs in the United States and Europe, can lead to unpleasant adverse events, such as dry mouth, nausea, headaches, being unable to sleep, and diarrhea, among others.3 For some patients, sertraline may not improve depressive symptoms. Right now, prescribing antidepressants is a test-and-trial—a patient is prescribed sertraline and is checked up on after 6 – 8 weeks—or up to several months—to see if the medication works.1

If symptoms do not go away, they are prescribed another antidepressant. Trying out new antidepressants can take months and can prevent an individual from fully participating in life.

Investigators, led by Maarten G. Poirot, MS, from the University of Amsterdam, conducted a secondary analysis of the Establishing Moderators and Biosignatures of Antidepressant Response in Clinical Care (EMBARC) study, a multisite, double-blind, placebo-controlled randomized trial. The primary study assessed MRI scans and clinical data of 229 US patients with major depressive disorder before and after a week of treatment with sertraline or placebo. The sample had a mean age of 38 years, and 66% were female.

Investigators used balanced accuracy and area under the receiver operating characteristic curve scores to collect data on how well the brain scan predicted response and remission after 8 weeks of taking sertraline.

“The algorithm suggested that blood flow in the anterior cingulate cortex, the area of [the] brain involved in emotion regulation, would be predictive of the efficacy of the drug,” said Eric Ruhé, psychiatrist at Radboudumc, in the press release.2 “And at the second measurement, a week after the start, the severity of their symptoms turned out to be additionally predictive.”

Ultimately, the analysis revealed 1/3 of patients would respond to sertraline and 2/3 would not.

“With this method, we can already prevent 2/3 of the number of ‘erroneous’ prescriptions of sertraline and thus offer better quality of care for the patient,” Reneman said.

Therefore, instead of wasting time and money on taking sertraline, patients can try a new antidepressant that might work better for them regarding treatment response and fewer adverse events.

For predicting sertraline response, the brain scan algorithm had a balanced accuracy of 68% and an area under the receiver operating characteristic curve of 0.73, suggesting the algorithm predicted sertraline’s response and remission significantly better than chance.1 Patients taking a placebo had a balanced accuracy of 62% and an area under the receiver operating characteristic curve of 0.66. Moreover, patients who switched from placebo to sertraline had a balanced accuracy of 65% and an area under the receiver operating characteristic curve of 0.63, demonstrating the differences between sertraline treatment and placebo treatment.

After the follow-up examination, investigators found 1 in 3 patients with depression had no improvement in symptoms after trying many antidepressants. Thus, the team stressed the importance of predicting an antidepressant’s effectiveness even sooner than a week. Investigators will soon be strengthening the algorithm by including extra information.

“The study results confirm that early sertraline treatment response can be predicted; that the models are sertraline specific compared with placebo; that prediction benefits from integrating multimodal MRI data with clinical data; and that perfusion imaging contributes most to these predictions,” investigators wrote. “Using this approach, a lean and effective protocol could individualize sertraline treatment planning to improve psychiatric care.”

References

  1. Poirot MG, Ruhe HG, Mutsaerts HMM, et al. Treatment Response Prediction in Major Depressive Disorder Using Multimodal MRI and Clinical Data: Secondary Analysis of a Randomized Clinical Trial. Am J Psychiatry. Published online February 7, 2024. doi:10.1176/appi.ajp.20230206
  2. Artificial Intelligence Helps Predict Whether Antidepressants Will Work in Patients. EurekAlert! February 7, 2024. https://www.eurekalert.org/news-releases/1033410. Accessed February 14, 2024.
  3. Side Effects of Sertraline. National Health Service. https://www.nhs.uk/. Accessed February 14, 2024.
Recent Videos
Ashfaq Marghoob, MD: Artificial Intelligence, Smartphone Use for Pigmented Lesion Classification
Steve Nissen, MD | Credit: Cleveland Clinic
Major Diagnostic Challenges for Pigmented Lesions, with Ashfaq Marghoob, MD
Sherona Bau, NP | Credit: UCLA Health
Discussing Interim Findings on Nemolizumab for Atopic Dermatitis, with Diamant Thaçi, MD
Jessica Crimaldi, NP | Credit: Jessica Crimaldi on LinkedIn
Harpreet Bhatia, MD: Benefits of Universal Screening for Lp(a) Levels
© 2024 MJH Life Sciences

All rights reserved.