Developmental variation in basal ganglia tissue iron, neurocognitive functioning, and impulsivity is associated with substance use trajectories in youth

Article https://doi.org/10.1038/s41467-026-73611-1 Developmental variation in basal ganglia tissue iron, neurocognitive functioning, and impulsivity is associated with substance use trajectories in youth Received: 22 August 2025 1234567890():,; 1234567890():,; Accepted: 15 May 2026 Ashley C. Parr 1,2 , Amar Ojha 2,3, Daniel J. Petrie1, Finnegan J. Calabro1,4, Brenden Tervo-Clemmens 5, Will Foran 1, Douglas Fitzgerald1, Susan F. Tapert 6, Kate Nooner 7, Wesley Thompson8, David B. Goldston9, Duncan Clark1 & Beatriz Luna1 Check for updates Neurodevelopmental models implicate dopaminergic and neurocognitive maturation in adolescent risk-taking, yet their joint contribution to substance use trajectories in humans remains unclear. We examined basal ganglia tissue iron, a marker of dopamine-related neurobiology, alongside impulsivity and inhibitory control in relation to longitudinal substance use patterns in the NCANDA-A cohort (N = 802; ages 12–30; 6,078 visits). Growth Mixture Models identified four trajectories: no/low use (30% of participants), youth peak (26%), adolescent increasing (17%), and adult increasing (26%). Substance use, inhibitory control, and tissue iron increased with age, while impulsivity declined. Greater substance use was associated with heightened impulsivity, low inhibitory control, and low tissue iron, prominently in early adolescence among youth peak patterns. Trajectories were further distinguished by divergent maturation of impulsivity and tissue iron. Results suggest that developmental variation in tissue iron and neurocognition contribute to youth substance use, highlighting adolescence as a sensitive window for risk stratification and prevention. Illicit substance use increases dramatically during the teenage years. In 2024, only 15% of 8th graders in the United States reported using substances, compared to ~37% of 12th graders1. Substance use is a leading contributor to preventable morbidity and mortality in youth, driving a range of behavioral and adverse health outcomes including suicidality and self-harm2,3, interpersonal violence2, and accidental death (i.e., road injuries2 and drug-related toxicity4,5). Accordingly, youth substance use poses a critical public health concern, necessitating evidence-based policies and developmentally informed prevention and intervention strategies. Neurodevelopmental theories conceptualize youth substance use within dual systems frameworks, including the Driven Dual Systems Model6,7. This model proposes that a developmental peak in subcortical, dopaminergically-mediated (DA) reward systems interacts with still-maturing, yet largely accessible, 1 Department of Psychiatry, University of Pittsburgh, Pittsburgh, PA 15213, US. 2Center for the Neural Basis of Cognition (CNBC), Carnegie Mellon University, Pittsburgh, PA 15213, US. 3Center for Neuroscience, University of Pittsburgh, Pittsburgh, PA 15213, US. 4Department of Bioengineering, University of Pittsburgh, Pittsburgh, PA 15213, US. 5Department of Psychiatry and Behavioral Sciences, University of Minnesota, Minneapolis, MN 55455, US. 6Department of Psychiatry, University of California, San Diego, La Jolla, CA 92093, US. 7Department of Psychology, University of North Carolina Wilmington, Wilmington, NC 28403, US. 8 Oxley College of Health and Natural Sciences, University of Tulsa, Tulsa, OK 74136, US. 9Department of Psychiatry and Behavioral Sciences, Duke University, e-mail: Durham, NC 27705, US. Nature Communications | (2026)17:4861 1 Article prefrontal cognitive control systems. As such, heightened reward reactivity can bias decision-making processes and contribute to normative increases in sensation seeking and risk-taking, creating a window of vulnerability for elevated substance use. Specifically, ‘dual-risk’ theories of adolescent substance use propose that hyper-active striatal reward reactivity coupled with hypofunction of neurocognitive systems confers risk for problematic use8–11, potentially via an increase of an already-enhanced reward drive12 coupled with ongoing maturation of prefrontal cognitive control systems8. Empirical work supports this framework, with striatal reward hyper-activation identified as a reliable functional brain marker of adolescent substance use vulnerability8,13, as well as studies showing that cognitive performance and cortical activation are associated with substance use risk-factors (i.e., impulsivity and externalizing14,15) and substance use onset16. Normative adolescent increases in risk-taking have been linked to enhanced mesocorticolimbic functional connectivity, including ventromedial PFC (vmPFC) - ventral striatum (VS) coupling17, and VS connectivity predicts youth substance use onset and frequency18,19. In adults, substance use is associated with impaired inhibitory control and distributed system-level alterations across cortical control regions (i.e., vmPFC, cingulate, insula, and dorsolateral PFC20,21), potentially reflecting long-term dysregulation of top-down control over striatal DA22–25. Together, these findings implicate mesocorticolimbic circuits underlying neurocognition and reward processing as a core feature of substance use vulnerability; however, how developmental interactions among DA neurobiology, cognitive control, and impulsivity shape adolescent substance use – and function as neurodevelopmental markers that predict its time course across adolescence into adulthood – remains unclear. Animal studies demonstrate dynamic adolescent changes in DA systems, including increased striatal DA availability26,27 and signaling28, supporting heightened reward drive, risk-taking27,29–31, and drugseeking behavior28. Human fMRI studies similarly show amplified adolescent VS blood oxygen level dependent (BOLD) activation during reward processing12,32–34 that relate to risk-taking32,35, and predict increases in substance use13. However, striatal BOLD is an indirect proxy for DA system function, and the lack of non-invasive in vivo markers in pediatric populations precludes understanding DA’s role in risk-taking and substance use behavior in youth. Molecular imaging studies in adults reveal distinct DAergic processes involved in substance use progression36–42, but the few molecular imaging studies in youth have generally included small samples, primarily targeting young adults, and have yielded mixed results – with studies showing either no changes43 or blunted DA function in high-risk youth44. Based on these results, changes in DA neurobiology could reflect a long-term consequence of increased midbrain DA biosynthesis in response to substance use (i.e., downregulation of striatal DA signaling)20, but findings in adults cannot disentangle preexisting vulnerability markers from long-term consequences of exposure. Animal work suggests that lower baseline levels of striatal D2 receptor availability and DA release promotes increased drug-seeking behavior in animals with no prior history of exposure25,45, rais (...truncated)