Array of Markers Collectively Predicts MDD Treatment Response

Nancy A. Melville

April 11, 2018

WASHINGTON — Predictors of response to mainstream therapies for major depressive disorder (MDD) range from baseline comorbid anxiety, brain connectivity patterns on neuroimaging, genomic patterns, and novel findings of inflammatory system abnormalities, new research suggests.

Further analyses from the Prediction of Remission to Individual and Combined Treatments (PREDICT) randomized controlled trial, which included 344 treatment-naive patients with MDD, as well as findings from the 1000 patient‒strong international Study to Predict Optimized Treatment in Depression (iSPOT-D), were presented here at the Anxiety and Depression Association of America (ADAA) 2018 Conference.

"What I believe is happening in our field is we have collectively uncovered a number of factors, including clinical, imaging, and genetic, that together have an effect on predicting treatment response to psycho- or pharmacotherapy," PREDICT coinvestigator Charles B. Nemeroff, MD, PhD, clinical director of the Center on Aging, Leonard L. Miller School of Medicine, University of Miami, Florida, told meeting attendees.

"What we will end up with is a sophisticated modeling system to plug in all of these variables and then have a reasonable degree of certainty of what patients should be treated with," Nemeroff added.

Predictors of Response, Recurrence

In PREDICT, patients with MDD were randomly assigned to receive either 12 weeks of cognitive-behavioral therapy (CBT) or 12 weeks of therapy with antidepressant therapy with the selective serotonin reuptake inhibitor (SSRI) escitalopram (Lexapro, Forest Labs) or the serotonin and norepinephrine reuptake inhibitor (SNRI) duloxetine (Cymbalta, Lilly). This was followed by a 21-month maintenance phase consisting of continued medication or up to three CBT booster sessions per year.

After the initial 12 weeks, patients who did not respond to therapy were switched to the opposing therapy for another 12 weeks. In other words, patients who failed to respond to antidepressant therapy would be switched to CBT, and vice versa.

Recent analysis of recurrence rates in 94 patients who experienced remission after receiving 12 weeks of either CBT or pharmacotherapy showed that 15.5% experienced a recurrence during the maintenance period, study coauthor  W. Edward Craighead, PhD, Department of Psychiatry and Behavioral Sciences at Emory University, Atlanta, Georgia, reported.

There were no significant differences in recurrence according to treatment groups.

Key factors that were found to significantly predict recurrence included having residual depressive symptoms after the initial monotherapy treatment, measured by the Hamilton Depression Rating Scale (hazard ratio [HR], 1.31; P = .03) and having comorbid anxiety at study baseline (HR, .31; P = .03). No other clinical measures significantly predicted recurrence.

These findings were published in March in the Journal of Consulting and Clinical Psychology.

Newer findings were presented by Craighead with respect to 112 participants whose depression failed to go into remission during the initial 12 weeks of treatment. These patients were switched to the alternative therapy. Results showed that after switching treatment, 48.2% of the patients experienced remission.

Interestingly, there were no significant differences between the groups in terms of type of therapy or in the sequence of therapies in terms of remission.

fMRI Imaging

In addition to the clinical predictors, findings from pretreatment resting-state functional MRI (fMRI) for 112 patients showed significant differences between those who did and those who did not achieve remission with CBT compared with antidepressant therapy, reported coauthor Boadie Dunlop, MD, of Emory University.

The differences were seen in three regions in the subcallosal cingulate cortex (SCC) known to be implicated in depression ― the left frontal operculum, the left ventromedial prefrontal cortex, and the periaqueductal gray/dorsal raphe.

Greater connectivity in those regions, as demonstrated on fMRI, was significantly associated with remission with CBT and treatment failure with medication, whereas lower connectivity in the regions was predictive of remission with medication and treatment failure with CBT.

These findings were recently published in the American Journal of Psychiatry.

Even patient preference for treatment modality was found to be associated with treatment completion, as reported in findings published in 2017.

Specifically, whereas the receipt of a patient's preferred modality (CBT or antidepressant medication) did not predict remission at 12 weeks (P = .31), those patients who did not receive their preferred treatment modality were significantly more likely to drop out of the trial (P = .01), usually within the first few weeks of treatment, said Dunlop.

"The findings indicate that patients who do not receive their preferred form of treatment are more likely to drop out of treatment, but not less likely to remit," he added.

The study's neuroimaging investigations were led by senior author Helen S. Mayberg, MD, professor of psychiatry, neurology, and radiology and the Dorothy Fuqua Chair in Psychiatry Imaging and Therapeutics at Emory University School of Medicine.

Inflammatory Markers

In his presentation, Nemeroff further discussed novel findings from the PREDICT trial. Never-treated patients with MDD showed marked abnormalities in the inflammatory system and cascade, compared to normal volunteers.

These untreated patients with MDD "are already in what I would call a cytokine storm," he said.

The untreated patients were found to have elevations in levels of Interleukin-6 (IL-6), IL-7A, and tumor necrosis factors, as well as elevations in a number of the inflammatory cytokines. There were also elevations in anti-inflammatory cytokines in these patients, Nemeroff noted.

Not only were levels of individual cytokines elevated, but abnormalities were found across the entire inflammasome (a complex of immune system receptors and sensors that regulate the inflammatory response).

"What we think is happening is the depressed patients show a profound proinflammatory response. It's then compensated to some extent by the anti-inflammatory cytokines, and the net effect is a muted T-cell response in terms of disease," he explained.

"This may be one of the reasons why depressed patients are vulnerable to medical disorders," he added.

During the 12-week initial treatment period, there were unique differences between patients who responded to treatment and those who did not. Among responders, inflammation marker levels stabilized but did not decrease. Among nonresponders, levels continued to increase.

There were no differences in the patterns with respect to treatment modality. The patterns were the same regardless of whether patients were treated with pharmacotherapy or psychotherapy.

"Novel" Findings

"Remission, regardless of treatment, was associated with inflammatory marker changes," Nemeroff told meeting attendees.

"This is one of the first times we see the exact same biological effect of treatment response to antidepressants and psychotherapy, and we looked at this every which way with every cytokine, and there was no difference," he said.

Although the association between inflammatory markers and depression is well established, the patterns observed are notable, Nemeroff later told Medscape Medical News.

"The findings are very novel," he said. "It is likely that these patients exhibit a very different treatment response to available treatments, and that is the challenge for the next step in research."

The findings could have important clinical implications, Nemeroff added.

"The biggest question is whether potent anti-inflammatory interventions would produce an antidepressant response, either alone or in combination with conventional treatment," he said. "Stay tuned."

iSPOT Predictors

The neuroimaging findings from the PREDICT trial mirror data also reported in the session from the iSPOT-D study. That study included 1008 unmedicated patients with MDD and 336 patients who acted as healthy controls. They were enrolled at 17 sites in five countries.

All received 8 weeks of treatment with one of three antidepressants: extended-release venlafaxine (Effexor XR, Wyeth Pharmceuticals), escitalopram, or sertraline (Zoloft, Pfizer).

The strongest predictors of remission were a history of abuse in early life, cognitive impairment, and two measures observed on neuroimaging: a loss of white matter integrity in the brain, and disruptions in resting intrinsic functional brain connectivity. These findings were similar to findings in the PREDICT study.

Key genetic factors also emerged as predictors of response to each of the drugs. ABCB1 variants were found to be differential predictors of remission with fewer side effects. Among them, G/G homozygotes predicted remission with escitalopram, and T/T homozygotes predicted remission with venlafaxine-XR.

Presenting author Leanne Williams, PhD, professor of psychiatry and behavioral sciences at Stanford University, California, said the findings underscore the potential usefulness of brain imaging in psychiatry at the diagnostic level.

"The imaging is routine in so many other health specialties but not yet in mental health," she told Medscape Medical News.

"Certainly the infrastructure is there for these other complex conditions, but we haven't had the data on usefulness in psychiatry. But I think more standard use of brain scans is not far away. The cost is less than an annual exam, and if someone isn't responding to treatment, it really could be worth the cost to determine what might work best for them," Williams said.

The iSPOT-D study is supported by Brain Resource, Ltd and by the Office of the Director of the National Institutes of Health, administered by the National Heart, Lung and Blood Institute. Dr Nemeroff has received research grants from the National Institutes of Health (NIH) and the Stanley Medical Research Institute; has consulted for Xhale, Takeda, Taisho Pharmaceuticals, PrismicPharmaceuticals, Fortress Biotech, Sunovion Pharmaceuticals, Sumitomo Dainippon Pharma, Janssen Research and Development, Magstim, Navitor Pharmaceuticals, TCMDS, and Intra-Cellular Therapies; is a stockholder of Xhale, Celgene, Seattle Genetics, AbbVie, OPKO Health, Antares, BI Gen Holdings, and Corcept Therapeutics Pharmaceuticals Company; is on the scientific advisory boards of the American Foundation for Suicide Prevention (AFSP), the Brain and Behavior Research Foundation, Xhale, ADAA, Skyland Trail, Bracket, and the Laureate Institute for Brain Research; is on the board of directors for the AFSP, Gratitude America, and the ADAA; has income sources or equity of $10,000 or more from American Psychiatric Publishing, Xhale, Bracket, CME Outfitters, Takeda, Intra-Cellular Therapies, and Magstim; and holds patents on a method of and devices for transdermal delivery of lithium and a method of assessing antidepressant drug delivery via an ex vivo assay. Dr Dunlop has received industry funds to Emory for his research from Acadia, Assurex, Axsome, Janssen, Otsuka, and Takeda. Dr Craighead receives support from the NIH, the Mary and John Brock Foundation, and the Fuqua family foundations, and book royalties from John Wiley & Sons; is a consultant to Skyland Trail; and is on the scientific advisory board of the ADAA and the AIM Foundation. Dr Craighead is an officer of Hugarheill, an Icelandic company for the prevention of depression. Dr Williams is a scientific advisor for Psyberguide, a division of the nonprofit One Mind Institute, and is a consultant for Blackthorn Therapeutics.

Anxiety and Depression Association of America (ADAA) Conference 2018. Session 308R, presented April 6, 2018.

For more Medscape Psychiatry news, join us on Facebook and Twitter.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.