It is still not well understood how neuroanatomical variability related to early substance involvement indicating future risk for developing substance use disorders may just reflect underlying predisposition and/or consequences of substance exposure. A question arises: What are the neuroanatomical features associated with early substance use initiation, and do they precede initiation? In quest of this question a study published in JAMA Network 1 for examining neuroanatomical features associated with early substance use initiation and the extent to which associations may reflect pre-existing vulnerability.
The ABCD study, also known as the Study of Adolescent Brain and Cognitive Development, constitutes a longitudinal study dealing with the multifaceted behavior and corporeal development from early childhood into young adulthood. A total of 11,875 children represented the baseline point of the study, aged 8.9 – 11 years, born between 2005 to 2009, and recruited from 22 US research sites. Parents or guardians gave written informed consent while children gave their assent to a research protocol approved by the institutional review board at each site. Race and ethnicity were reported by parent/guardian and were assessed to characterize the sociodemographic variability of our sample. Baseline neuroimaging data were obtained from ABCD data release 3.0, while all other data were sourced from release 5.0. Participants were reduced in number after excluding missing data in the final analytic samples of 6556 to 9804 subjects. The work complied with Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.
Data was analyzed from February to September 2024. Substance use initiation was tracked through a 3-year follow-up (i.e., under < age 15). Participants self-reported initiation of alcohol, nicotine, cannabis, and other substances, and baseline MRI-derived estimates of brain structure were also included (e.g., global and regional cortical volume, thickness, surface area, sulcal depth, and subcortical volume). Covariates included family (e.g., familial relationships), pregnancy (e.g., prenatal exposure to substances), child (e.g., sex and pubertal status), and MRI (e.g., scanner model) variables.
Of 9804 children (mean [SD] baseline age, 9.9 [0.6] years; 5160 boys [52.6%]; 213 Asian [2.2%], 1474 Black [15.0%], 514 Hispanic/Latino [5.2%], 29 American Indian [0.3%], 10 Pacific Islander [0.1%], 7463 White [76.1%], and 75 other [0.7%]) without missing baseline neuroimaging or covariate data, 3460 (35.3%) reported substance use initiation before age 15. Initiation of any substance or alcohol use was associated with thinner cortex in prefrontal regions (e.g. rostral middle frontal gyrus, β=-0.03; 95% CI, -0.02 to -0.05; P=6.99×10⁻⁶) but thicker cortex in all other lobes, larger globus pallidus, and hippocampal volumes, as well as greater global indices of brain structure (e.g., larger whole brain volume, β=0.05; 95% CI, 0.03 to 0.06; P=2.80×10⁻⁸), following Bonferroni or false discovery rate multiple testing correction. Cannabis initiation was associated with lower right caudate volume (β=-0.03; 95% CI, -0.01 to -0.05; P=.002). Post hoc analyses that restricted initiation to after baseline suggested that most associations, including a thinner prefrontal cortex and greater whole brain volume, preceded initiation.
Over the course of this cohort study involving a total of 9804 subjects, differential structural attributes in the brain exhibited significant associations with early initiation to substance use at ages less than 15 years in the forms of alcohol, nicotine, cannabis, or other substances. Aggregate increases in the size of whole, cortical, and subcortical brain volumes usually tend to be associated with a thinner prefrontal cortex, but those with a thicker cortex otherwise. Most features of brain structural measures noted against substance use initiation were observable in upfront and baseline naive children who later initiated substance use. This suggests that neurodevelopmental variability in brain structure may increase the risk of early substance involvement study has several limitations. First, although our study had sufficient power to detect small effects of any substance and alcohol use initiation, individual substances that were infrequently endorsed may lack power (i.e., 80% power for |β| > 0.04 at P < .005 for nicotine and cannabis). Such small effects are a way of saying that these findings aren’t clinically meaningful for people, but they inform and challenge current theoretical models of addiction.
There is limited geographical variability in alcohol and drug use involvement within the released data from the ABCD Study for this sample and thus limits the analysis to variance with respect to substance use initiation. However, later, problematic substance use phenotypes may produce greater effects with the use of multivariate methods, which may serve as omnibus tests to cut down the burden of multiple testing followed by post hoc univariate testing.
Second, even if there were more ABCD Study neuroimaging scans available, very few participants began using substances before the next scan (417 subjects), which limited the ability to create well-powered models to examine changes in neuroanatomy associated with substance initiation (i.e., possible exposure effects). At some future time, as the current sample matures and substance involvement becomes increasingly common and variable, it will be valuable to examine overlapping trajectories of substance use and neuroanatomy. Such studies cannot determine whether changes in neural phenotypes are due to substance exposure, developmental predisposition, or both, but genetically informed design strategies may allow tests of the plausibility of either explanation.
An invited commentary, “The Adolescent Brain Cognitive Development Study and How We Think About Addiction” published in JAMA Network by Felix Pichardo and Sylia Wilson 2. Both scholars documented the relevance of research to consideration of assumptions related to cause in models of brain disease due to addiction. The large sample size, longitudinal approach, and genetically informative components (family study design, nested twin subsamples, and DNA collection) of the ABCD study are advanced as important keys to causal inference concerning neural risk factors.
Felix Pichardo and Sylia Wilson also mentioned that some informants include several genetically informative components such as family studies design, nested twin subsamples, and DNA collection. This facilitates more causal inference and identification of mechanisms to complement temporal precedence. Genetically informed approaches, such as co-twin control, notwithstanding the longitudinal nature of ABCD data. They will infuse a further breadth in looking at core causal and mechanistic concerns other than correlational evidence.
The ABCD study is important to such work, like Miller et al 1, which will lead to critical insights into inform causal models of substance use and addiction and the developmental, maintenance, and harmful processes through which it evolves these findings contribute to the development of most targeted efforts for preventing early initiation, managing harmful use, and most effectively implementing interventions. It also has the potential to produce findings compelling us to rethink our causal assumptions and current theories about addiction through these key design features: large sample size, longitudinal assessments, and genetically informative components.
There were neuroanatomical features that predated substance exposure and were associated with substance use initiation in a cohort study of 9804 children. In terms of converging evidence from genetically informative (e.g., discordant twin or sibling) and other longitudinal studies, data increasingly render procrustean bed interpretations that variability with respect to substance comes only after substance use. Ultimately, a better understanding of the structural links between the brain and involvement with substances will uncover predispositional risk factors that may provide insight into early causes of substance use disorders. Furthermore, there will also be knowledge of clinically informative mechanisms through which many adverse health outcomes are associated with substance involvement.
Reference:
- Miller AP, Baranger DAA, Paul SE, et al. Neuroanatomical variability and substance use initiation in late childhood and early adolescence. JAMA Netw Open. 2024;7(12):e2452027. doi:10.1001/jamanetworkopen.2024.52027
- Pichardo F, Wilson S. The Adolescent Brain Cognitive Development Study and how we think about addiction. JAMA Netw Open. 2024;7(12):e2451997. doi:10.1001/jamanetworkopen.2024.51997
