Back to search

Boston University

Role of allopregnanolone and pregnanolone in cognition in individuals with posttraumatic stress symptoms

Abstract

dc:description.abstract

Background: Changes in blood-based neuroactive steroid and inflammatory biomarker levels have been associated with posttraumatic stress disorder (PTSD) and cognitive dysfunction. How these biomarker systems relate to neuropsychological performance among individuals with PTSD symptoms, however, remains unclear. This dissertation first reviews literature on the role of the neuroactive steroids allopregnanolone (Allo) and pregnanolone (PA)—equipotent positive allosteric modulators of the effects of γ-aminobutyric acid (GABA) at GABAA receptors— in memory and cognition in relation to PTSD. It then examines whether resting Allo+PA levels (i.e., assessed 45-minutes after sitting), indices of their synthesis capacity, or inflammatory biomarkers moderate the association between a) PTSD severity and neuropsychological performance cross-sectionally, and b) PTSD severity at time point 1 (T1) and cognitive change longitudinally. Additionally, pilot pharmacokinetic (PK) studies evaluated the effects of two doses of intravenous (IV) Allo administered ~2−3 hours after learning on a non-aversive episodic source memory task. The low dose infusion (5-minute 1.7 mcg/kg loading dose followed by a 5-hour 2.6 mcg/kg/hr continuous infusion) was intended to raise and sustain resting plasma Allo levels by ~ 1500 pg/mL during the 5−6 hour window of memory consolidation. The effects of the low dose were compared to a higher dose 30-minute infusion (28 mcg/kg) that targeted third-trimester pregnancy plasma Allo levels before returning to baseline over the same period. Methods: PTSD diagnosis and severity were assessed using the Clinician-Administered PTSD Scale (CAPS). Cross-sectional analyses of the relationships between PTSD severity and cognition were conducted in 712 post-9/11 veterans (mean age = 34.7 ± 9.2 years; 88.8% male; 73.5% White) enrolled in the Translational Research Center for TBI and Stress Disorders (TRACTS) longitudinal study at the United States Department of Veteran Affairs (VA) Healthcare Systems in Boston and Houston. Longitudinal analyses included 410 participants who returned for follow-up an average of 1.9 years later. Confirmatory factor analyses modeled four latent cognitive domains: Working Memory, Attention/Executive Function, Episodic Memory, and Processing Speed. Linear regression tested cross-sectional moderation, and cross-lagged panel models evaluated longitudinal associations. Biomarkers included plasma concentrations of the sum of the equipotent GABAergic neuroactive steroids Allo and PA (i.e., Allo+PA), as well as ratios of neuroactive steroid levels that index the function of two key enzymes involved in their synthesis: a) Allo+PA/5α-dihydroprogesterone (5α-DHP), which indexes 3α-hydroxysteroid dehydrogenase (3α-HSD) function, and b) 5α-DHP/Progesterone (PROG), which indexes 5α-reductase function. Inflammatory biomarkers included plasma levels of C-reactive protein (CRP), tumor necrosis factor α (TNF-α), interleukin 6 (IL-6), and IL-10. In pilot PK studies, a two-day episodic source memory test was administered to 10 participants (mean age = 29.7 ± 13.4 years; 20% male) with chronic PTSD (i.e., PTSD ≥ 3 months in duration), and linear mixed-effects models assessed IV Allo effects of the low dose (N = 5) versus high dose (N = 5) IV on 24-hour free recall, recognition, and source memory. Results: Cross-sectionally, Allo+PA moderated the association between PTSD severity and Working Memory (interaction: unstandardized B = 0.0002, p = .010), such that PTSD severity was negatively associated with Working Memory at lower Allo+PA levels (B = -0.009, p = .011) but not at higher Allo+PA levels (B = 0.002, p = .617). Higher CRP and IL-6 levels were associated with poorer Attention/Executive Function (CRP: B = -0.231, p = .010; IL-6: B = -0.085, p = .041) and slower Processing Speed (CRP: B = -0.239, p = .005; IL-6: B = -0.089, p = .026). CRP also moderated the associations between PTSD severity and both Attention/Executive Function (interaction: B = 0.008, p = .020) and Processing Speed (interaction: B = 0.008, p = .008), such that PTSD severity was negatively associated with these domains at lower CRP levels (Attention/Executive Function: B = -0.010, p < .001; Processing Speed: B = -0.010, p < .001), but not at higher CRP levels (Attention/Executive Function: B = -0.004, p = .126; Processing Speed: B = -0.004, p = .115). In longitudinal cross-lagged panel models, T1 Allo+PA levels were negatively associated with Working Memory (standardized β = -0.168, p = .003) and positively associated with Processing Speed (β = 0.133, p = .010) at time point 2 (T2). T1 TNF-α levels were positively associated with Processing Speed at T2 (β = 0.070, p = .022), while T1 CRP levels were positively associated with PTSD severity at T2 (βs = 0.054 – 0.056, ps = .008 – .013). Elevated T1 CRP levels further strengthened the negative association between T1 PTSD severity and Processing Speed at T2 (interaction: β = -0.118, p = .013). Although CRP × PTSD severity and IL-6 × PTSD severity interactions at T1 reached statistical significance for T2 Episodic Memory (CRP interaction: β = -0.057, p = .012) and Processing Speed (IL-6 interaction: β = -0.062, p = .036), respectively, follow-up simple slope analyses indicated that the PTSD severity-cognition associations evaluated at lower and higher biomarker levels were not statistically significant, suggesting weak modification effects or insufficient power. No main effects of Allo+PA synthesis capacity or IL-10 on cognitive outcomes, nor interaction effects with PTSD severity, were observed. In the pilot PK study, the model including the Allo dose x recall day interaction provided better fit than the model without the interaction (∆χ2(1) = 3.91, p = .048) and demonstrated greater Day 2 free recall in the higher-dose group (53.8% ± 26.5) relative to the lower-dose group (29.4% ± 11.0). No group differences in free recall were observed on Day 1. No effects of IV Allo dose on recognition or source memory were observed. Conclusions: These findings demonstrate that GABAergic neuroactive steroids and inflammatory biomarkers are associated with domain-specific cognitive performance in a trauma-exposed veteran cohort. This work also provides evidence that Allo+PA and markers of inflammation moderate the effects of PTSD severity on cognition. Notably, while higher CRP levels were associated with poorer Attention/Executive Function and Processing Speed in cross-sectional analyses, the negative association between PTSD severity and these domains was evident only at lower CRP levels. Longitudinal results further indicate that these biomarkers influence cognition and PTSD symptoms over time. The pilot PK findings provide preliminary evidence for a possible dose-dependent effect of Allo on episodic memory consolidation, indexed by recall. However, replication in a larger, placebo-controlled trial is needed to clarify these potential dose-dependent IV Allo effects and establish clinical significance. Collectively, the data suggests that alterations in Allo+PA and inflammatory marker levels may signal risk for cognitive decline and PTSD chronicity. This work thus may inform future investigation of precision-based interventions targeting inflammatory processes and/or the strategic timing of neuroactive steroid administration to improve cognitive outcomes. Future research should clarify mechanisms underlying the observed relationships of Allo+ PA and the inflammatory factors of interest to the cognitive domains queried in this study. Research should also evaluate whether modulation of these biomarkers changes the longitudinal trajectory of cognitive performance among individuals with PTSD and in turn facilitates recovery.

Degree

thesis:*
Name thesis:degree_name
Doctor of Philosophy
Level thesis:degree_level
doctoral
Discipline thesis:degree_discipline
Behavioral Neurosciences
Grantor
Boston University
Year dc:date.issued
2026

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Brown, Kayla
Advisor dc:contributor.advisor
  • Ratner, Marcia
ORCID dc:identifier.orcid

Subjects

dc:subject × 9

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/2144/53453

Chain of custody

source
Harvested from
Boston University
Base URL
open.bu.edu/oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Brown, Kayla. Role of allopregnanolone and pregnanolone in cognition in individuals with posttraumatic stress symptoms. doctoral thesis, Boston University, 2026. https://hdl.handle.net/2144/53453