Methodological sources of neuroimaging outcomes in acute cortisol challenge studies

Introduction

Cortisol modulates limbic and prefrontal circuits via temporally distinct non-genomic mechanisms with onset within minutes and genomic mechanisms emerging from approximately 60 minutes post-exposure [1]. Dysregulation of the hypothalamic–pituitary–adrenal axis is implicated in major depressive disorder, post-traumatic stress disorder, and anxiety disorders, and glucocorticoid-based interventions remain under investigation [2]. Pharmacological challenge combined with neuroimaging enables the investigation of such drug–brain relationships in vivo [3]. Findings from acute hydrocortisone challenge studies are heterogeneous, with opposing effects reported within comparable regions and paradigms [4]. It was hypothesised that such variability reflects methodological factors, including dose, route of administration, scan timing relative to pharmacodynamic windows, and circadian phase.

Aims

To derive design recommendations for future cortisol challenge studies by mapping how methodological parameters relate to neuroimaging outcomes of acute hydrocortisone administration.

Methods

A systematic review was conducted in accordance with PRISMA 2020 [5]. PubMed was searched; extension to additional databases (Embase, PsycINFO) is planned. Included studies administered hydrocortisone as an acute pharmacological challenge in humans and reported neuroimaging outcomes from MRI-based modalities (BOLD-fMRI, resting-state fMRI, arterial spin labelling), PET-based modalities ([¹⁸F]FDG, H₂¹⁵O, receptor PET), or other modalities (EEG, NIRS, TMS). Studies were excluded if cortisol was used therapeutically or for replacement, if pharmacological effects could not be isolated, or if the source constituted a review, meta-analysis, or protocol. Variables extracted included dose, administration route, dosing time, baseline cortisol verification, dose-to-scan interval, modality, paradigm, sample, analysis type (whole-brain vs region-of-interest), coordinate type, and number of foci. Design parameters were cross-tabulated against effect direction in two of the most frequently reported regions, the amygdala and hippocampus. Coordinate-based meta-analysis (e.g. Activation Likelihood Estimation, ALE) remains under consideration.

Results

The included literature comprised 2,186 participants (1,863 healthy, 323 clinical); clinical populations included post-traumatic stress disorder, depression, phobia, schizophrenia, and Alzheimer’s disease. Females comprised approximately one-third of healthy participants (594 / 1,863). Hydrocortisone was administered orally (43 / 56 studies, 10–100 mg) or intravenously as bolus or infusion (13 / 56 studies). Dose-to-scan intervals clustered between 31–90 minutes, and only 9 / 56 studies sampled the early window (≤30 minutes) corresponding to non-genomic effects. Circadian dosing time ranged from 08:00 to 23:00 and was inconsistently reported. Task-based fMRI was the most frequently used modality (31 / 56), followed by PET-based modalities (10 / 56) and EEG (8 / 56). The amygdala and hippocampus were reported in 27 and 29 of 56 studies respectively; effect directions in both regions diverged across studies.

Conclusions

Neuroimaging findings following acute hydrocortisone administration are heterogeneous, co-occurring with substantial variation in scan timing, dose, route of administration, and circadian phase across studies. Consistent reporting of circadian dosing time and dose-to-scan interval should become standard. Sampling of the early post-administration window, corresponding to non-genomic effects, remains limited. Findings reported here represent interim results; database extension and quantitative synthesis are in progress.

Modality	Studies (k)	Sample (N)	Intravenous administration (k)
Task-based fMRI	31	1,598	1
PET-based	10	153	7
EEG	8	180	1
Resting-state fMRI	4	183	1
Other*	3	72	3
Total	56	2,186	13

*ASL (k = 2), TMS (k = 1)

References

1. De Kloet, E. R., Joëls, M., & Holsboer, F. (2005). Stress and the brain: From adaptation to disease. Nature Reviews Neuroscience, 6(6), 463–475.
2. Lanzenberger, R., Wadsak, W., Spindelegger, C., Mitterhauser, M., Akimova, E., Mien, L.K., Fink, M., Moser, U., Savli, M., Kranz, G.S., Hahn, A., Kletter, K., Kasper, S., 2010. Cortisol plasma levels in social anxiety disorder patients correlate with serotonin-1A receptor binding in limbic brain regions. Int. J. Neuropsychopharmacol. 13 (9), 1129–1143.
3. Gryglewski, G., Klöbl, M., Berroterán-Infante, N., Rischka, L., Balber, T., Vanicek, T., Pichler, V., Kautzky, A., Klebermass, E.-M., Reed, M.B., Vraka, C., Hienert, M., James, G.M., Silberbauer, L., Godbersen, G.M., Unterholzner, J., Michenthaler, P., Hartenbach, M., Winkler-Pjrek, E., Wadsak, W., Mitterhauser, M., Hahn, A., Hacker, M., Kasper, S., Lanzenberger, R., 2019. Modeling the acute pharmacological response to selective serotonin reuptake inhibitors in human brain using simultaneous PET/MR imaging. Eur. Neuropsychopharmacol. 29 (6), 711–719.
4. Henckens, M.J.A.G., Pu, Z., Hermans, E.J., van Wingen, G.A., Joëls, M., Fernández, G., 2012. Dynamically changing effects of corticosteroids on human hippocampal and prefrontal processing. Hum. Brain Mapp. 33 (12), 2885–2897.
5. Page, M.J., McKenzie, J.E., Bossuyt, P.M., Boutron, I., Hoffmann, T.C., Mulrow, C.D., Shamseer, L., Tetzlaff, J.M., Akl, E.A., Brennan, S.E., Chou, R., Glanville, J., Grimshaw, J.M., Hróbjartsson, A., Lalu, M.M., Li, T., Loder, E.W., Mayo-Wilson, E., McDonald, S., McGuinness, L.A., Stewart, L.A., Thomas, J., Tricco, A.C., Welch, V.A., Whiting, P., Moher, D., 2021. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 372, n71.