The nation's grim opioid crisis surges on, with the fentanyls (high potency synthetic opioids) driving unprecedented mortality rates. Drug overdose deaths are now the leading cause of death in those under age 50, with more than 47,000 Americans dying of opioid overdose in 2017. As of December 2018, Philadelphia had the third highest rate of opioid overdose deaths in the country (out-ranked only by Pittsburgh and Baltimore). Fentanyl is present in 84% of the fatal opioid overdoses in Philadelphia. Medication-assisted treatment (MAT) for opioid use disorders – whether full opioid agonist (methadone), partial opioid agonist (buprenorphine), or a full antagonist (naltrexone) – is critical for reducing opioid use, and for preventing overdose deaths. Unfortunately, compliance with these life- saving medications is often poor, with ancillary use of non-opioid drugs (especially cocaine) as a common culprit. Cocaine is found in almost half of the opioid overdose deaths in Philadelphia. Identifying promising adjunctive medications that reduce cocaine and other illicit drug use during MAT could improve adherence and save thousands of lives each year. Further, measuring how these medications “engage” the intended brain targets will speed rational medication development. Toward both these goals, we have established a Clinical Laboratory with Integrated Neuroscience (CLIN) for Evaluation of Medications for Substance Use Disorders at the University of Pennsylvania Center for Studies of Addiction. In the initial demonstration project for the UG1, we will test the promise of a dopamine D3- preferring agent, cariprazine (Vraylar, 1.5 mg daily), vs. placebo both for (Aim 1) preliminary clinical efficacy (reduced illicit drug use, and improved adherence to life-saving buprenorphine), and (Aim 2) for target engagement (e.g., limbic activation by opioid and cocaine cues), in patients with both opioid use disorder (OUD) and cocaine use disorder (CocUD), in an 8 week outpatient trial. Recently, the UG1 Laboratory grants at UPenn and VCU have, in collaboration with NIDA, proposed to screen a medication identified by data-mining, olanzapine, for clinical efficacy (Primary Aim: reduction in illicit opioid use) in a two-site (n=24 per site), 8-week observational trial for buprenorphine-stabilized OUD patients with symptoms of Serious Mental Illness (SMI). Secondary Aims feature reduction in other illicit drug use and improvement in sleep; Exploratory Aims include improvements in MAT adherence, in study retention, and in thought and mood disorder symptoms. Both the initial demonstration project (cariprazine) and the olanzapine participants will also receive a battery of phenotypic probes (e.g., for reward/inhibition; neurocognition) to capture relapse vulnerability, and likely heterogeneity in medication response.