Major depressive disorder (MDD) is one of the most heterogeneous disorders in psychiatry and first line treatments are inadequate for the majority of patients, likely because they do not target an individual?s subtype. Improving our understanding of MDD subtypes will allow us to (1) identify treatments that target subtype- specific pathophysiology and (2) determine which subgroup of MDD patients will best respond to these treatments, thus improving antidepressant outcomes. In a neuroinflammatory subtype, MDD may manifest via chronic neuroinflammation. The translocator protein (TSPO), located on the outer mitochondrial membrane of microglia and astrocytes, is regarded as a marker of this neuroinflammation and can be measured in vivo by positron emission tomography (PET). In support of a neuroinflammatory subtype of MDD, TSPO, measured by PET, was found to be elevated by 30% on average in the prefrontal cortex (PFC) in MDD relative to healthy individuals. Further, we have preliminary data from a repeated social defeat stress mouse model that shows elevated PFC TSPO and other elevated neuroinflammation markers in a subset of ?depressed?, non-resilient mice, and that this phenotype is reversed by elimination of TSPO expressing glial cells! To parallel this effect in humans, brain-penetrant anti-inflammatory medications such as celecoxib can be used. Celecoxib has antidepressant effects in MDD; however, the observed effect sizes are highly variable, likely reflecting the biological heterogeneity of MDD. We hypothesize that anti-inflammatory treatments such as celecoxib will be most effective in those with the neuroinflammatory subtype of MDD and that the mechanism of antidepressant action is through a reduction of neuroinflammation. We will test these hypotheses in parallel studies in humans in rodents. For the human aim, we will take advantage of an ongoing study of celecoxib efficacy currently being performed at Stony Brook Medicine (PI: Parsey) by recruiting participants who are already being treated with celecoxib (8 weeks, 400mg/day). 53 MDD participants will be enrolled, with 42 expected to complete the study involving TSPO PET imaging before and after treatment. We hypothesize that higher PFC TSPO (as measured by PET) prior to treatment will be correlated with better response to celecoxib and further, that reductions in PFC TSPO will be correlated to depression improvement after adjusting for covariates. In a parallel study in rodents, we hypothesize that PFC TSPO and other CNS inflammation markers in our repeated social defeat stress mouse model of depression will be elevated as measured by microPET, quantitative protein/mRNA level analysis, and reactive microglial morphology, when compared to wild type mice. Further, we hypothesize that elevated PFC TSPO and other neuroinflammatory markers will be reduced after celecoxib treatment. If mouse and human studies do not agree, this suggests that TSPO PET provides a clinically- relevant proxy of neuroinflammation. However, if our hypotheses are confirmed, this would validate that TSPO is a marker of the neuroinflammatory MDD subtype and a target of treatment.
The results of this translational study will provide the first evidence that a targeted intervention with an anti- inflammatory medication can reverse both the underlying pathology and depressive state in ?inflammatory depression? in rodents and humans. As such, our findings will be of direct benefit to MDD patients by providing alternative, targeted treatment options, so as to improve poor remission rates currently associated with use of monoaminergic antidepressants.