According to an NIH-&-others-funded study published online February 10 in Nature Biotechnology, and discussed at medscape (free to all, just need to register):
Personalized treatment for depression may soon become a reality, thanks to an artificial intelligence (AI) algorithm that accurately predicts antidepressant efficacy in specific patients.
A landmark study of more than 300 patients with major depressive disorder (MDD) showed that a latent-space machine-learning algorithm tailored for resting-state electroencephalography (EEG) robustly predicted patient response to sertraline [e.g. Zoloft] The findings were generalizable across different study sites and EEG equipment.
"We found [this algorithm identifying the EEG signature] for patients who do well on sertraline," study investigator Madhukar Trivedi, MD, professor of psychiatry at the University of Texas Southwestern Medical Center in Dallas, told Medscape Medical News. [and on looking further] "...that patients with that same EEG signature do not do well on placebo..."
How widely this discovery may apply seems iffy, even if promising. For one thing, there’s growing acknowledgement of depression as an adverse side-effect of many commonly-prescribed meds themselves. For another, the published research is not yet full-free-access —only the abstract is available— so the data about ages, genders, ethnicity, socioeconomic status, or other factors in the study subjects and groups is unknown — gender, and ethnicity are among known factors in genetic capacity to metabolize drugs, and age and socioeconomic status among known epigenetic factors.
Another question is accuracy of diagnosis.
Currently, major depression is defined using a range of clinical criteria. As such, it encompasses a heterogeneous mix of neurobiological phenotypes. Such heterogeneity may account for the modest superiority of antidepressant medication relative to placebo.
While recent research suggests resting-state EEG may help identify treatment-predictive heterogeneity in depression, these studies have also been hindered by a lack of cross-validation and small sample sizes.
What's more, these studies have either identified nonspecific predictors or failed to yield generalizable neural signatures that are predictive at the individual patient level.
For these reasons, there is currently no robust neurobiological signature for an antidepressant-responsive phenotype that may help identify which patients will benefit from antidepressant medication. Nevertheless, said [study investigator Madhukar Trivedi, MD, professor of psychiatry at the University of Texas Southwestern Medical Center in Dallas], detailing such a signature would promote a neurobiological understanding of treatment response, with the potential for notable clinical implications.
"The idea behind this NIH-funded study was to develop biomarkers that can distinguish treatment outcomes between drug and placebo," he said. "To do so, we needed a randomized, placebo-controlled trial that has significant breadth in terms of biomarker evaluation and validation, and this study was designed specifically with this end in mind.
"There has not been a drug-placebo study that has looked at this in patients with depression. So in that sense this was really a pivotal study," he explained.
The medscape article is about three fairly readable pages long, ending with includes Dr. Trivedi’s across-lifetime financial relationships with a somewhat startling array of commercial entities in the pharmaceutical industry together with past grant/funding relationships including govt and nonprofit.
A consideration to keep in mind is that research into personalized prescribing of psychotropics alone, including use of computerized methods, dates back to at least 1986, and has centered on both efficacy and avoidance/prevention of harm. But it’s been a thin and patchy stream of research until somewhat recently, apparently due to absence of funding.
Research into the usefulness of antidepressants beyond psychotropic treatment alone, runs a parallel track, if still extensively involving off-label prescribing in practice.
One may speculate that the ruling-out of patients as candidates for this or that prescription drug did not become financially interesting to the medical industries until quite huge lawsuits, damage awards, and major fines and other disciplinary actions from governmental agencies across a broad range of adverse drug events —and adverse drug event patterns in countries/societies with extensive available healthcare— mounted high enough to begin cast a serious pall upon the reputations of doctors and drugs generally.
And abridging profitability.
In other words, the basic research, and worthwhile directions to pursue for the benefit of patients have been in place several decades. Now that there’s a pecuniary incentive for the industries, this neglected area of research may become, as the researchers say, “robust.”
Meanwhile, we can be cautiously hopeful.