Wang, JieJieWangLiu, XiaodongXiaodongLiuXu, FanyongFanyongXuLu, WeicongWeicongLuShao, RobinRobinShaoLi, XiaoyueXiaoyueLiWang, RunhuaRunhuaWangDr. LAM Yin-Hung, BessBessDr. LAM Yin-HungChen, JinyongJinyongChenCai, AolingAolingCaiGoldstein, BenjaminBenjaminGoldsteinDu, FeiFeiDuMcIntyre, Roger S.Roger S.McIntyreWu, JinfengJinfengWuLin, KangguangKangguangLin2026-02-022026-02-022026BMJ Mental Health, 2026, vol. 29(1), pp. 1-9.2755-9734http://hdl.handle.net/20.500.11861/26658Open access<jats:sec> <jats:title>Objectives</jats:title> <jats:p>Bipolar disorder (BD) often begins in adolescence, a period marked by dynamic neurodevelopment. However, the neurobiological basis from genetic risk and subthreshold symptoms to diagnosed BD remains unclear.</jats:p> </jats:sec> <jats:sec> <jats:title>Methods</jats:title> <jats:p> We conducted a cross-sectional analysis using data from the Recognition and Early Intervention of Prodromal Bipolar Disorders cohort ( <jats:ext-link xmlns:xlink="http://www.w3.org/1999/xlink" ext-link-type="clintrialgov" xlink:href="NCT01863628">NCT01863628</jats:ext-link> ), including 392 participants aged 12–25 years with a balanced sexual distribution, stratified into five groups: offspring of patients with BD with (OBDs, n=48) or without (OBDns, n=62) subthreshold symptoms, individuals without BD family history but with subthreshold symptoms (nOBDs, n=63), patients diagnosed with BD (n=133) and healthy controls (HCs, n=86). Cortical thickness relative to HC was assessed using high-resolution T1-weighted images and FreeSurfer V.7.3.2. Gene expression patterns were derived from the Allen Human Brain Atlas, and partial least squares regression, along with gene enrichment analyses, were applied to link cortical alterations with underlying transcriptomic profiles. </jats:p> </jats:sec> <jats:sec> <jats:title>Findings</jats:title> <jats:p>Cross-sectional analyses revealed graded cortical thickness differences across the BD risk spectrum, with patients with BD showing the most pronounced deviations and high-risk individuals with subthreshold symptoms displaying intermediate features relative to HCs. Cortical changes were significantly associated with spatial gene expression patterns, particularly in genes involved in mitochondrial ATP production, oxidative phosphorylation and synaptic signalling. Gene set enrichment revealed that BD-specific cortical thinning correlated with downregulation of excitatory synaptic pathways and excitatory neuron-related gene expression. Conversely, high-risk individuals exhibited upregulation of both excitatory and inhibitory neuronal markers. Developmental transcriptomic enrichment further linked significant genes to mid-childhood and adolescence.</jats:p> </jats:sec> <jats:sec> <jats:title>Discussion</jats:title> <jats:p>By identifying distinct transcriptomic signatures associated with cortical thinning at different stages, our findings underscore the potential of transcriptomic markers for early detection and intervention in BD.</jats:p> </jats:sec> <jats:sec> <jats:title>Clinical implications</jats:title> <jats:p>The findings highlight the potential for using transcriptomic markers for early detection and intervention, suggesting that identifying these markers could lead to improved outcomes for at-risk adolescents. This research has the potential to inform clinical practices and policies aimed at early screening and preventive strategies for BD.</jats:p> </jats:sec>enPeer Reviewed Journal Article10.1136/bmjment-2025-302006