Targeting Inflammation May Help Bipolar Depression Subtype

Batya Swift Yasgur, MA, LSW

May 20, 2019

An anti-inflammatory agent does not improve depressive symptoms in patients with bipolar depression but shows promise in a subpopulation of this group who have a history of trauma, new research suggests.

Investigators compared adjunctive infliximab (multiple brands), a tumor necrosis factor (TNF) antagonist, to placebo in 60 adults with bipolar depression over a 12-week period and found that the reduction in symptom severity at week 12 did not differ significantly between the two groups.

However, a secondary analysis showed that infliximab-treated individuals with a history of childhood maltreatment, particularly physical abuse, exhibited significantly greater reductions in depressive symptoms and higher response rates, compared with those who received placebo.

"The punchline is that adults having a bipolar depressive episode who reported being exposed to childhood diversity, which is so common — either physical or sexual abuse, for example — exhibited much greater anti-inflammatory response than those who had received placebo," lead author Roger McIntyre, MD, professor of psychiatry and pharmacology, University of Toronto and head of the Mood Disorders Psychopharmacology Unit, University Health Network, Canada, told Medscape Medical News.

"I think this provides reason to believe that if we can target inflammation, people with bipolar depression who exhibit inflammation — either by history or biochemical evidence — can benefit from an anti-inflammatory approach, especially if they have a history of childhood maltreatment," said McIntyre, who is also director of the Depression and Bipolar Support Alliance.

The study was published online May 8 in JAMA Psychiatry.

Inflammatory Subtype

The variable response to treatment for bipolar depression "implicates distinct biotypes within heterogeneous populations with bipolar disorder," the authors write.

A body of evidence suggests a role for immunoinflammatory disturbances during the onset, phenomenology, comorbidity, and treatment response in bipolar disorder.

Previous research has suggested that some anti-inflammatory agents may have "variable antidepressant effects" in adults with both unipolar and bipolar depression.

"We know that current treatments help some people a lot, but not enough, and we also know that one of the reasons and variables that decreases the efficacy of treatment is a history of trauma — not necessarily current PTSD [posttraumatic stress disorder] but a trauma history, which seems to attenuate the likelihood that medications will be helpful for bipolar disorder," McIntyre said.

"One conceptual framework proffered to explain this is that when a person is exposed to trauma, this activates the immune system, and when the body is in an inflammatory state — and there are many ways to be in that state —treatments don't seem to cooperate as well," he continued.

Previous research has "well-documented the inflammatory state created by trauma and, along with growing interest in the role of inflammation in subserving symptoms of bipolar disorder, depression, and cognition, provided interest in conducting a study looking at people with bipolar illness, especially those with elevated inflammation," he added.

A previous study found no significant differences between infliximab and placebo treatment in adults with major depressive disorder (MDD) and bipolar disorder, but a post hoc analysis revealed an antidepressant effect in favor of infliximab in those who exhibited elevated pretreatment C-reactive protein (CRP) levels, suggesting "the notion of a possible inflammatory biotype that is more likely to respond to an on-target anti-inflammatory treatment," the authors state.

The current study investigated "whether adults with bipolar disorder I/II depression who have biochemical and/or phenotypic evidence of immuno-inflammatory activation before randomization would be more likely to show an antidepressant response to infliximab, compared with placebo."

Biochemical or Phenotypic Criteria

The researchers randomly assigned 60 participants to receive either infliximab (n = 29 [48%]; mean [SD] age, 45.0 [11.7] years; 71% female) or placebo (n = 31 [52%]; mean age, 46.8 [10.2] years; 87% female) for a 12-week period.

Participants were required to meet either one biochemical (eg, CRP ≥5 mg/L) or 1 phenotypic inflammatory criterion at baseline, consisting of:

  • Obesity (eg,, ethnicity-specific waist circumference or body mass index [BMI] ≥30 (phenotypic) plus one or more of the following:

    • Increased triglyceride levels

    • Decreased high-density lipoprotein (HDL) cholesterol level

    • Elevated blood pressure

  • Type 1 or 2 diabetes

  • Inflammatory bowel disorder (IBD)

  • Rheumatologic disorder

  • Daily cigarette smoking

  • Migraine headaches

Individuals with concurrent psychiatric disorders, active psychotic symptoms, substance abuse/dependence during the past 6 months, previous exposure to anti-TNF agents, and other medical conditions were excluded.

The primary outcome was baseline-to-endpoint change in the Montgomery-Asberg Depression Rating Scale (MADRS) total score, with exploratory analyses to evaluate the "moderating effects" of baseline CRP level, illness severity and course, and self-reported childhood maltreatment (based on the 28-item Childhood Trauma Questionnaire [CTQ]).

Anti-Inflammatory Modalities

Significant baseline-to-endpoint change in MADRS total score was found at all weeks across treatment × time interaction (X2 = 10.33; P = .04).

The treatment × time interaction, however, was significant only at week 2  (RR, 0.86; 95% confidence interval [CI],0.75 - 0.98; df = 1; P = .024) and was no longer significant at week 12 (relative risk [RR], 1.09; 95% CI, 0.80 - 1.50; df = 1; P = .60).

No time × treatment × CRP interaction or time × treatment × clinical severity interaction were observed.

The secondary analysis revealed a significant treatment × time × childhood maltreatment interaction, in which infliximab-treated individuals with childhood history of physical abuse exhibited greater reductions in MADRS total score (X2 = 12.20; P = .02) and higher response rates (X2 = 4.05; P = .04) compared with those who were treated with placebo.

The authors state that it is "of note" that baseline CTQ physical abuse subdomain scores were associated with clinical severity (r = 0.32; 95% CI, 0.04 - 0.56; P = .02) and CRP (r = 0.31; 95% CI, 0.01 - 0.55; P = .03).

Moreover, the time × treatment × physical abuse interaction effect remained significant, even after adjustment for baseline clinical severity and CRP.

"People with a history of trauma are more likely to have inflammatory activation, and may stratify themselves as being more likely to benefit from an anti-inflammatory approach, which is interesting conceptually and academically and may have clinical relevance to how we treat people in the future," McIntyre commented.

He suggested that future research might "seek to determine whether any anti-inflammatory modality could confer benefit," including "more naturalistic treatments," such as improved sleep, physical exercise, mindfulness, L-Carnitine, curcumin, and folate.

"We're ultimately trying to help people who have brain-based disease that to some extent is caused by inflammation, and if we engage that inflammatory process through any modality, we could have a viable treatment on our hands," McIntyre said.


Commenting on the study for Medscape Medical News, Antonio Lucio Teixeira, MD, PhD, professor and director, of the Neuropsychiatry Program at the University of Texas Health in Houston, noted that a "major limitation of the study, partially acknowledged by the authors, is the way they defined their 'immune-inflammatory activation' group of patients" because they "used only one biochemical criterion — CRP levels — while other inflammatory markers, notably TNF and related molecules such as TNFRs, could have been more meaningful."

Moreover, their "phenotypic criteria" included "very heterogeneous conditions, some, such as diabetes, with much stronger inflammatory component than others, such as migraine," said Teixeira, who was not involved with the current study.

One take-home message is that "the way we define 'clinical phenotypes' matters, [and] this statement is valid for clinical research and practice," Teixeira emphasized.

Additionally, "regarding inflammation [and] immune role in major psychiatric disorders, despite all the hype around the issue but unfulfilled promises, growing evidence suggests that part — maybe around 20% to 30% — of patients might benefit from some related strategy," he suggested.

"In this context, optimizing the treatment of medical comorbidities associated with inflammation, such as diabetes and IBD, among others, could play an adjunctive role in the management of psychiatric symptoms," Teixeira stated.

Also commenting on the study for Medscape Medical News, Yann Quidé, PhD, of the School of Psychiatry, University of New South Wales, Australia and Neuroscience Research Australia, said that dysregulations of the immune system have been reported in both major psychiatric disorders and individuals exposed to childhood trauma, independent of their clinical status.

"This proof-of-concept study may have critical implications for the therapeutic practices in the field," said Quidé, who was not associated with the study.

"Identifying subtypes of patients can help reduce the social and economic burden associated with treatment of these severe disorders," he continued.

"However, it also remains unclear if all patients exposed to childhood trauma present a high inflammatory profile [which is] unlikely, and further research is needed to identify finer subtypes of patients for whom personalized treatments can be offered," Quidé said.

"Hot" Pathways

Also commenting on the study for Medscape Medical News, Charles Raison, MD, of the University of Wisconsin-Madison, described it as a "really rigorous piece of work."

Raison, who was not involved with this study but was the first author on the original study of infliximab in depression, said that a take-home message of the current study is not to "assume that all depressed patients are inflamed, or that all depressed patients will benefit from anti-inflammatories."

"Rather," he continued, "the situation is more complex."

The study, Raison said, "should be a spur to continue examining anti-inflammatory agents in depressed patients with high inflammation measured by biomarkers."

"I think the study's post hoc finding that childhood abuse predicted response to infliximab is very interesting and consistent with evidence that early adversity may sensitize inflammatory and other stress pathways to run 'hot' in adulthood," Raison added.

McIntyre noted that the "time is now ready by far where we need a large study to show whether the systematic and routine evaluation of a patient's inflammatory markers would be cost effective and result in improved health outcomes for individuals and for the system."

The study was supported by a grant from the Stanley Medical Research Institute. McIntyre reported receiving grants from Stanley Medical Research Institute during the conduct of the study; receiving grants from the Canadian Institutes of Health Research/Global Alliance for Chronic Diseases/Chinese National Natural Research Foundation outside the submitted work; and receiving speaking/consultation fees from Lundbeck, Janssen, Shire, Purdue, Pfizer, Otsuka, Allergan, Takeda, Neurocrine, Sunovion, and Minerva outside the submitted work. The other authors' disclosures are listed on the original article. Teixeira, Raison, and Quidé have disclosed no relevant financial relationships.

JAMA Psychiatry. Published online May 8, 2019. Abstract, Editorial

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.