The aim of this project is to evaluate the efficacy of the influenza vaccine in individuals
with major depressive disorder (MDD) as well as to elucidate the nature of the immunological
abnormalities in MDD using a quasi-experimental design. Specifically, the investigators plan
to induce transient, mild inflammation in medically-healthy study participants using the
influenza vaccine. Initially the investigators will conduct a pilot project with up to 20
individuals in order to evaluate the time-point at which the peak inflammatory response to
the vaccine occurs. Subjects will receive the seasonal influenza vaccine and provide blood
samples 4 hours, 2 days, and 30 days post vaccination. Subsequent to the pilot study, both
depressed and psychiatrically-healthy participants will be randomized in a parallel group,
double-blind design so that they receive either influenza vaccine (seasonal vaccine) or
saline (i.m). At baseline, subjects will provide a blood sample, complete a number of rating
scales to measure mood and fatigue, and may complete approximately one hour of MRI scanning
with or without simultaneous EEG recording. Two-days post vaccination, they will provide a
second blood sample, complete more clinical ratings and may complete another identical MRI
session with or without simultaneous EEG. Four weeks later, participants will be asked to
return to provide a third blood sample and complete additional clinical ratings. The blood
samples will be used to measure both innate and adaptive immune function and may be used to
correlate the vaccine-induced immunological changes to neurophysiological changes in the
brain measured by MRI and/or EEG.
Psychological stress has been shown to impair the efficacy of the influenza vaccine,
particularly in older individuals (> 60 years). However, the effect of depression (as opposed
to stress per se) on vaccine response has not been measured, especially in younger people.
One possible cause of this putative impaired vaccine response is suppression of the immune
system in a subgroup of individuals with depression. Major depressive disorder (MDD) has been
associated with a decline in total numbers of circulating B-cells, T-cells, and natural
killer cells (NKC), a reduction in the proliferative responses of NKC and T-cells in vitro
that is ameliorated by administration of selective serotonin reuptake inhibitors (SSRIs),
increased susceptibility to various viral infections, and a decline in the efficacy of the
herpes zoster vaccine in the elderly (1, 2). On the other hand, MDD appears also to be
characterized by immunological abnormalities that manifest both as low-grade inflammation,
including over-activity of the innate immune system (3). Whether inflammation and immune
suppression are two different manifestations of the same underlying abnormality (unitary
model) or whether inflammation and immune suppression are distinct phenomena occurring in
different MDD subtypes (independent model) is not known.
MDD is also characterized by neurophysiological abnormalities of striatal, cortical midline,
and limbic brain structures that regulate hormonal, autonomic, and emotional responses to
reward and stress (4-6). Yet, it remains unclear how immune function relates to these
depression-associated functional neuroimaging abnormalities. Studies of healthy volunteers
(HCs) administered endotoxin or the typhoid vaccine have reported inflammation-associated
increases in negative affect along with neurophysiological changes in brain regions
comprising the affective and reward circuitry (7-8). However, because these studies were
limited to HCs, it is possible that the observed inflammation-induced neurophysiological
changes were adaptive and therefore not applicable to MDD populations. That is, it remains
unclear whether any experimentally-induced immune abnormalities in MDD are correlated and
therefore potentially causally-related to the neuroimaging abnormalities observed in MDD.
In order to address these questions, we use a parallel group, placebo-controlled influenza
vaccine challenge to induce transient inflammation that peaks 2-days post vaccination in MDD
subjects and HCs. Subjects may complete a pre-and-post vaccination MRI scan, the latter
during the peak inflammatory response.
Aim 1: To examine whether influenza vaccine effectiveness is compromised in a
young-to-middle-aged sample (18-55 years of age) of individuals with MDD.
Hypothesis 1.1. Compared with HCs (n=40), individuals with MDD (n=40) will display reduced
vaccine titers and reduced influenza virus-induced CD4+ memory T-cell proliferation four
weeks post vaccination (T2).
Hypothesis 1.2. A greater number of MDD subjects than HCs will fail to show a four-fold
increase in antibody titer, the conventional standard for determining a clinically
significant response to a viral vaccine.
Aim 2: To test the competing models of immune dysfunction in MDD.
Hypothesis 2.1. At T0 and T1, stimulated and unstimulated monocytes from MDD subjects will
show increased interleukin 6 (IL-6) and tumor necrosis factor (TNF) production compared with
HCs. The effect will be most salient at T1 in individuals receiving vaccine.
Hypothesis 2.2. Consistent with the unitary model, monocyte IL-6 and/or TNF production at T0
and T1 will be inversely correlated with antibody titers at T2 in the MDD group.
Aim 3: To test for neurophysiological differences between MDD patients and HCs in resting
state cerebral blood flow (with arterial spin labeling), resting EEG activity, and BOLD
response to monetary rewards and interoceptive stimuli (with fMRI) at T0 and T1, and to
correlate depression-associated perfusion, EEG activity, reward response, and interoceptive
function with immune function at T0 and T1.
Hypothesis 3.1: Compared with HCs, the MDD subjects will display reduced perfusion of the
subgenual anterior cingulate cortex (sgACC) and/or pregenual ACC (pgACC), hemispheric
asymmetry in the alpha (8-13 Hz) power band (left prefrontal cortex hypoactivation), as well
as reduced hemodynamic response to anticipatory reward in the ventral striatum. These
differences will be most salient at T1 in the MDD group receiving vaccine versus the HCs
Hypothesis 3.2: The functional imaging abnormalities in the MDD group at T0 and/or T1 will be
correlated with the abnormalities in monocyte activation in the MDD group at T0 and/or T1.
Major Depressive Disorder Patient Group - Currently Depressed: Subjects will have met
Diagnostic and Statistical Manual of the American Psychiatric Association (DSM-V) criteria
for primary MDD in a current major depressive episode and current Hamilton Depression
Rating Scale (HAM-D) or Montgomery Asberg Depression Rating Scale (MADRS) score in the
mild-to-severely depressed range and will have been deemed to be medically stable by a
physician listed on this protocol. Subjects who do not wish to receive treatment with
psychotropic medication in the future and have not taken psychotropic medication for at
least 3 weeks will be included in the study.
Healthy Comparison Group: Subjects will be selected who have not met criteria for any Axis
I psychiatric disorder, have no known first-degree relatives with mood or anxiety
disorders, and have a current score on the HAM-D or MADRS in the non-depressed range.
Inability to provide informed consent, pregnant or nursing women, known hypersensitivity to
vaccines, age of onset of MDD > 40 years, metal implants or other factors that would
preclude MRI scanning, serious risk of suicide, delusions or hallucinations, medical or
neurological illnesses (such as diabetes, autoimmune disorders or inflammatory bowel
disease) that affect brain structure, function or immune measurements, previous head injury
with loss of consciousness, abuse of drugs or alcohol within the previous year or a
lifetime history of substance dependence, treatment with medications that impact immune
function (e.g. prednisone), HIV or other chronic infection, a recent acute illness (e.g.
influenza), receipt of a vaccine within 3 months of commencing the study. Subjects whose
first major depressive episodes arose temporally after other major medical or psychiatric
conditions will also be excluded, since their functional imaging results generally differ
from those reported in primary MDD.
Subjects will be asked not to take any over-the-counter non-steroidal anti-inflammatory
drugs or drink alcohol for 48 hr. before each study day. The use of oral contraceptives and
menstrual phase or status are not exclusion criteria but this information will be recorded.
The same exclusion criteria apply to healthy controls with the added criterion that
first-degree relatives should be free of major psychiatric illness as assessed with the
Family Interview for Genetic Studies (FIGS).