Summary/Abstract
Anabolic-androgenic steroids (AAS) use is a significant public health problem, with nearly 4 million Americans
having used AAS. Roughly 30% of AAS users develop AAS dependence, among the highest dependence
rates of all abused drugs. Polydrug abuse is highly prevalent among AAS users. AAS use causes acute
psychiatric effects such as aggression and violence, and as we reported in past, long-term AAS use is
associated with visuospatial memory dysfunction on tests predictive of early dementia. To date, human brain
correlates of long-term AAS use are largely unexplored. Our pilot imaging studies in long-term AAS users
produced 3 compelling findings. First, the amygdala is enlarged by AAS, consistent with controlled preclinical
studies. Second, AAS reduced functional connectivity between the amygdala and dorsal anterior cingulate
cortex (dACC), a cognitive control region. Third, in magnetic resonance spectroscopy studies, AAS users had
elevated glutamine/glutamate ratios and lower scyllo-inositol levels in dACC, suggestive of ongoing dysfunction
and possible neurotoxicity. As the amygdala participates in emotion regulation, visuospatial processing,
sensory integration/processing of appetitive/aversive stimuli, cost/benefit decision-making, and drug reward,
seeking, and cue reactivity, and together with the dACC modulates approach/avoidance learning and monitors
emotional conflicts, amygdala and dACC abnormalities could impair all of these processes. Because
commonly used AAS increase ß-amyloid levels and scyllo-inositol prevents ß-amyloid clumping, our scyllo-
inositol finding may be particularly important. In this R01 application, we aim to build upon initial findings by
directly probing amygdala and hippocampal function with task-based BOLD fMRI paradigms, including a
source memory paradigm and the Hariri emotional face paradigm. We also will acquire MRS spectra from
dACC and parieto-occipital cortex, the latter of which is a target for early ß-amyloid accumulation, to determine
whether glutamine/glutamate and scyllo-inositol/creatine metabolite ratio abnormalities also occur in posterior
cingulate cortex, a region normally exhibiting early ß-amyloid accumulation. Proposed studies involve large
sample sizes that are adequately powered to test a priori hypotheses. Resulting data will help to identify the
neural bases for psychiatric and cognitive abnormalities in AAS users, to gauge the severity of brain effects
from long-term AAS use, and to inform the design of future studies to examine the progression of such effects
with continued AAS use and/or aging, including interventional studies with agents such as scyllo-inositol or
NMDA receptor antagonists. We have access to well-characterized AAS users and controls, recruited for our
nearly-completed NIDA-funded cardiovascular studies, and the ability to recruit new subjects meeting
inclusion/exclusion criteria. Our team is highly experienced with these populations and with proposed
assessment methods, and thus is uniquely poised to conduct this research. Accordingly, this program is
feasible and, if successful, likely will exert a sustained and powerful influence on the field.