Medication development for alcohol use disorder (AUD) is a time-consuming and costly process. Unfortunately, no new medications for AUD have been approved in the past two decades, despite significant investments. A typical path to developing a new medication for AUD includes testing in animals, followed by safety testing in humans, followed by randomized clinical trials. Recently, it has been proposed that testing in humans using experimental psychopharmacology paradigms can detect the initial efficacy of a compound under development. As such, the “signal” of medication benefit over placebo is initially identified in animal models, followed by human laboratory testing, and ultimately tested in randomized clinical trials (RCT). In essence, at each phase in testing, scientists are tasked with making “go/no-go” decisions about candidate pharmacotherapies. In this context, approval by the FDA constitutes the final “go” decision and requires compelling efficacy demonstration in RCTs, which is the gold standard. While a host of factors are involved in making “go/no-go” decisions, the paradigms used in animal and human testing to detect an efficacy signal are crucial to the success of medication development. Further, how to evaluate the preclinical and human evidence for a compound in order to decide, is of paramount importance. To date, the question of which models should be used in preclinical studies and human laboratory studies and how the evidence they provide should be evaluated remains highly subjective. Scientists can argue for models they are most familiar with and preliminary data can be presented with a range of plausible interpretation, all of which is inherently subjective. The proposed R21 application seeks to conduct novel meta-analytic models to test the relationship between AUD medication effect sizes obtained in animal models, human laboratory models, and randomized clinical trials (RCTs). These analyses will test the degree to which models used at each stage of medication development for AUD are predictive of clinical outcomes in RCTs, the gold standard for improving healthcare.

---

---

Identifying treatment responders in medication trials for AUD using machine learning approaches                                                                                                Alcohol use disorder (AUD), as defined in DSM-5, represents a highly prevalent, costly, and often untreated condition in the United States. Pharmacotherapy offers a promising avenue for treating AUD and for improving clinical outcomes for this debilitating disorder. While developing novel medications to treat AUD remains a high priority research area, there remain major opportunities to further elucidate clinical response in completed medication trials. To that end, a key question in randomized clinical trials (RCTs) is which patients respond to a given pharmacotherapy. Identifying treatment responders provides major opportunities to advance clinical care for AUD by personalizing medication practices on the bases of variables/predictors of good clinical response. For example, while the effect size for medications such as naltrexone is deemed small-to-moderate, a host of studies over the past decade have shown that its effect size may be considerably larger for certain subgroups of patients. Towards advancing precision medicine for AUD and leveraging data from a host of carefully conducted RCTs for AUD, this R03 application seeks to conduct secondary data analysis. Specifically, we propose to analyze data from four RCTs conducted by the NIAAA Clinical Investigations Group (NCIG). These state-of-the-art RCTs for AUD have tested the following pharmacotherapies: (a) quetiapine, (b) Levetiracetam XR (Keppra XR®), (c) Varenicline (Chantix®), and (d) HORIZANT® (Gabapentin Enacarbil) Extended-Release. In this R03 application, we propose to use a machine learning approach to identify treatment responders in the NCIG RCTs. Machine learning represents a highly promising and underutilized data analytic strategy in the field of AUD treatment response. Machine learning models prioritize the ability to predict future outcomes over creating perfectly fitting models for the data at hand. This results in models which are more generalizable to future observations, which fits well with our goal of identifying responders in RCTs. Leveraging data from these pivotal RCTs through secondary data analysis and using novel analytic methods, namely machine learning, provides a cost-effective approach to identifying AUD pharmacotherapy responders.

---

Alcohol problems are highly prevalent among adults in the U.S. Specifically, twenty-five percent of adults in the U.S. report either currently having alcohol-related problems or drinking patterns that put them at risk for developing problems. Further, only 1 in 5 individuals with alcohol problems seek treatment and the available treatments are often not informed by science. In order to address the service needs of our community and to provide a training opportunity for advanced doctoral students in clinical psychology at UCLA, we propose to develop an outpatient alcoholism clinic providing evidence-based cognitive behavioral therapy for individuals with alcohol problems. In keeping with our service, training, and research missions, this clinical protocol will combine a research component with evidence-based practices. Specifically, participants will be asked to complete an alcohol cue-reactivity testing session at the end of their intake session and will discuss their reactions to alcohol cues with the therapist. Participants will be asked to repeat the alcohol cue-reactivity testing session after completing the 12-week CBT protocol. The CBT protocol will be derived from the Project MATCH Treatment Manual, which is an effective and well-disseminated evidence-based intervention for alcohol use disorder. Participants will complete a battery of assessments at the intake visit (30 minutes of assessments at intake) as well as weekly assessments immediately prior to their clinic visits (10-15 minutes of assessments each week), including a breathalyzer test, and random urine toxicology tests (one at intake and three random tests over the course of the 12-week treatment).