White Matter Microstructure Among Straight and Gay Cisgender Men, Sao Praphet Song, and Straight Cisgender Women in Thailand

Abstract

White matter (WM) microstructure is differentiated in relation to sex/gender, psychosexuality, and, among transgender people, gender-affirming hormone (GAH) use. Prior research focused on Western samples, which limits generalizability to other populations. Here, diffusion tensor imaging (DTI) was used to assess WM microstructure in a Thai sample (N = 128) of straight cisgender men, straight cisgender women, gay cisgender men, and sao praphet song (i.e., transfeminine individuals assigned male at birth and sexually attracted to cisgender men). Sao praphet song were further grouped by GAH use. Groups were compared on fractional anisotropy (FA), mean diffusivity (MD), axial diffusivity (AD), and radial diffusivity (RD) using whole-brain tract-based spatial statistics (TBSS). FA, AD, and RD were further examined via multivariate analysis to assess covariance across WM microstructural indices and participant groups. A significant multivariate pattern differentiated the feminine- from masculine-identifying groups irrespective of sex assigned at birth and suggested WM tissue organization was greater among the latter in the bilateral cingulum, anterior corona radiata, left corpus callosum, and right superior longitudinal fasciculus, forceps minor, and corticospinal tracts. TBSS analyses reinforced that WM differed by gender identity in various regions. Among sao praphet song, GAH use was associated with lower regional FA, suggesting less WM organization bilaterally in the corpus callosum, cingulum, and anterior corona radiata. The findings aligned with prior studies in Western samples, indicating cross-population generalizability of WM microstructural differentiation in relation to sex/gender, psychosexuality, and GAH use.

Keywords: Thailand; diffusion tensor imaging (DTI); gender identity; gender‐affirming hormones (GAH); sex/gender; sexual orientation; white matter microstructure.

FIGURE 1

Significant TBSS clusters indicating group differences in FA. Significant clusters (p _FWE < 0.05; k > 100 voxels) identified in FA (blue‐light blue) by TBSS analysis. Sample mean FA skeleton mask (green) overlaid on the sample mean FA map. Anatomical left is right.

FIGURE 2

Significant TBSS clusters indicating group differences in AD. Significant clusters (p _FWE < 0.05; k > 100 voxels) identified in AD (blue‐light blue). Sample mean FA skeleton mask (green) overlaid on FA map. Anatomical left is right.

FIGURE 3

Significant TBSS clusters indicating group differences in RD. Significant clusters (p _FWE < 0.05; k > 100 voxels) identified in RD (blue‐light blue). Sample mean FA skeleton mask (green) overlaid on FA map. Anatomical left is right.

FIGURE 4

One latent variable reveals a significant difference in WM microstructure across groups. (A) Brain pattern identified by the latent variable across groups. The direction of associations is interpreted in the inverse due to the blue clusters in panel B indicating a negative association with brain scores in panel A. Correlations with confidence intervals that do not overlap with zero indicate a stable contribution to the latent variable. AD = axial diffusivity; FA = fractional anisotropy; GAH+ SPS = sao praphet song using gender‐affirming hormones; GAH− SPS = sao praphet song not using gender‐affirming hormones; GCM = gay cisgender men; RD = radial diffusivity; SCM = straight cisgender men; SCW = straight cisgender women. (B) Stable regions of the latent variable. Bootstrap ratio is thresholded at ±2.5 as indicated by the gradient scale on the right. Anatomical left is left for coronal and axial view; the sagittal view shows the right hemisphere.