Acknowledgement:

This project was supported by the Ministry of University and Research within the Complementary National Plan PNC-I.1 “Research initiatives for innovative technologies and pathways in the health and welfare sector, D.D. 931 of 06/06/2022, PNC0000002 DARE – Digital Lifelong Prevention CUP: B53C22006440001.

Abstract:

The Choroid Plexus (ChP) is a vascular brain structure involved in vital brain functions [1]. Most clinical studies focused on analyzing ChP Volume (ChPV) in diseased populations, hypothesizing a link between increased ChPV and neuroinflammatory processes [2,3]. However, the influence of biological factors (e.g., age, gender) and brain anthropometric measures on ChPV variability remains underexplored due to the limited size of healthy control (HC) cohorts [4-6]. Therefore, we develop and validate a normative modeling (NM) approach to characterize the natural variation of ChPV across the adult lifespan aiming to develop an imaging biomarker [7,8]. The ChP of 1,036 HC (36-88 years) from publicly available datasets [9,10] was automatically segmented with ASCHOPLEX [11]. Lateral ventricles volume and total intracranial volume, extracted with FreeSurfer [12], with age and gender were used as independent regressors for a Bayesian hierarchical NM. The NM was tested on two independent clinical cohorts of patients affected by depression or multiple sclerosis (MS) [11,13]. The individual deviation from the normal HC distribution was evaluated through the z-score. The proposed model explains up to 40% of ChPV normal variability in HC (Fig.1). The Area Under the Curve (AUC) for the compared z-score distributions (p<0.001) demonstrated the ability of the NM in distinguishing a HC from a patient (AUC: HC-depression=0.78; HC-MS=0.69), as confirmed by ANOVA test (p<0.001). NM can detect ChPV alterations at the level of individual patients, replicating existing findings of altered ChPV in pathological conditions and might help in accelerating its translation to its clinical use.

Figure 1: The hierarchical Bayesian normative modelling (NM) of the choroid plexus volume (ChPV)[mm³] as function of aging: ChPV ~ Age + Gender + LVV + TIV + (1 | Dataset). The blue lines represent the percentiles between 2.5% and 97.5%. The ChPV individual values are shown in grey for healthy controls (HC) used for training the NM, in red for multiple sclerosis (MS) subjects, and in green for depression (DEP) subjects.

References:

[1] Spector et al., Exp Neurol, 2015, 267:78–86.

[2] Monaco et al., Int J Mol Sci, 2020, 21(21).

[3] Kolahi et al., Neuroradiology, 2024, 66(11):1869-1886.

[4] Alisch et al., Front Aging Neurosci, 2021, 13(734992).

[5] Eisma et. al., J Cereb Blood Flow Metabol, 2022, 43(2):269–280.

[6] Eisma et al., Fluids Barriers CNS, 2024, 21(21).

[7] Marquand et al., Mol Psychiatry, 2019, 24(10):1415–1424.

[8] Rutherford et al., eLife, 2022, 11(e72904).

[9] Shafto et al., BMC Neurology, 2014, 14(204).

[10] Bookheimer et al., NeuroImage, 2019, 185:335-348.

[11] Visani et al, Comput Biol Med, 2024, 182(109164).

[12] Fischl, NeuroImage, 2012, 62(2):774-781.

[13] BIODEP, NIMA consortium, https://www.neuroimmunology.org.uk/