Early cannabis use and later opioid use disorder - the role of adverse childhood experiences, genetic liability and comorbid stimulant use

Abstract
Opioid use disorder (OUD) is rare (2%) but associated with extraordinarily high morbidity and mortality. Opioids
are the primary contributors to the soaring numbers of U.S. drug overdose deaths, with increasing co-use of
psychostimulants exacerbating the burden. Alongside this opioid crisis, are increasing rates of cannabis use, a
drug that is now legal for recreational use in 24 states. Early cannabis use onset (eCan) has been documented
in cross-national, cross-sectional, and longitudinal studies to increase the likelihood of using and misusing both
cocaine/psychostimulants and opioids. Shared genetic vulnerability has been suggested as a mechanism
undergirding both eCan and OUD, although it has not been estimated. However, even within discordant twin
pairs, twins who endorse eCan more commonly report stimulant misuse and OUD, when compared to their co-
twins (each at 4 times increased odds), suggesting that genetic factors alone do not explain this association.
Also, rodents pretreated with delta-9 tetrahydrocannabinol during adolescence show variable behavioral and
epigenetic responses to opioids and cocaine, suggesting that eCan may modify sensitization to these drugs. Yet,
due to the rarity of OUD in samples that typically assess eCan and the absence of eCan measurement in large
biobanks from which OUD data are drawn, a detailed examination of the association between eCan and opioid
involvement is pending. In this R21 data analysis proposal in response to NOT-DA-22-003, using data from
multiple deeply phenotyped samples, predominantly those ascertained for substance use disorders
(N~138,000), we examine the association between eCan and opioid involvement. In aim 1, we estimate the
association between early cannabis use (eCan) and proximal (e.g., initial subjective responses, indexing
sensitization) and distal aspects of opioid involvement (e.g., OUD criterion count), and whether associations are
modified by co-occurring stimulant (non-prescription, including cocaine) misuse. In aim 2, we test whether eCan
and OUD (including overdose in the presence of psychostimulants) share an architecture of common genetic
influences and parse evidence for non-genetic, putatively causal influences of eCan on opioid involvement using
both genomic causality and data from pairs of twins and siblings discordant for eCan. In aim 3, we estimate the
independent role of Adverse Childhood Experiences (ACEs) on eCan and opioid involvement and its interplay
with genetic susceptibility in worsening the likelihood of OUD in those who start using cannabis at an earlier age.
Understanding the relationship between eCan and specific aspects of opioid involvement provides insights into
whether future changes in cannabis use might exacerbate OUD, and establishes targets for experimental and
mechanistic studies. If eCan modifies opioid sensitization and increases likelihood of OUD, or is correlated with
opioid overdose death, then heightened vigilance is urgently needed to delay onsets amidst efforts to legalize
the drug.

Grant Number: 5R21DA061592-02
NIH Institute/Center: NIH
Principal Investigator: ARPANA AGRAWAL