Artificial intelligence helps identify ketamine as potential cocaine addiction treatment

Researchers identified ketamine as a possible treatment for cocaine use disorder using artificial intelligence and professional expertise, according to findings published in the journal Addiction.

“This study is a great example of addressing an intractable problem by the creative use of AI using different sources of data,” Pamela B. Davis, MD, PhD, the Arline H. and Curtis F. Garvin Research Professor at the Case Western Reserve School of Medicine in Cleveland, said in a press release. “It is our hope that this approach will suggest therapeutic approaches for other difficult problems.”

Davis and colleagues implemented their AI-based KG-Predict system to identify the 35 FDA-approved drugs that were most likely to be candidates for cocaine use disorder (CUD) treatments. The advisory committee of the National Drug Abuse Treatment Clinical Trials Network evaluated the AI system’s rankings and elected to retrospectively assess ketamine.

The researchers reviewed electronic health records in the global TriNetX database to evaluate CUD remission among patients prescribed ketamine or non-ketamine anesthesia or who were prescribed ketamine or antidepressants/midazolam for depression.

In total, there were 379,409 patients with CUD diagnosed from January 2007 to June 2022, 16,754 of whom were prescribed ketamine as an anesthetic or for depression.

Compared with people prescribed a non-ketamine anesthetic, those who received ketamine for anesthesia had greater odds of CUD remission (HR = 1.98; 95% CI, 1.42-2.78). Subgroup analyses by race and sex revealed no disparities, with ketamine anesthesia increasing the odds of remission among white patients (HR = 1.71; 95% CI, 1.09-2.68), Black patients (HR = 2.12; 95% CI, 1.24-3.63), women (HR = 2.32; 95% CI, 1.47-3.65) and men (HR = 2.35; 95% CI, 1.33-4.16).

Analyses of people with CUD prescribed ketamine for depression demonstrated more than four times greater odds of remission compared with those prescribed antidepressants or midazolam (HR = 4.39; 95% CI, 2.89-6.68). Again, no significant disparities in remission between men and women or between Black and white patients were identified.

People with CUD who received anesthesia but had no diagnosis of depression were found to be more than twice as likely to enter remission if they received ketamine vs. non-ketamine anesthesia (HR = 2.23; 95% CI, 1.02-4.91). Similarly, in analyses of people with CUD who received depression medication but not any type of anesthesia, those prescribed ketamine had more than three times the odds of remission compared with those who received antidepressants or midazolam (HR = 3.37; 95% CI, 1.45-7.83).

Further analyses indicated that ketamine targets 10 CUD-associated genes and several enriched pathways associated with neuroactive ligand-receptor interaction, alcohol and cocaine addiction and cAMP (cyclic adenosine 3,5-monophosphate) signaling, according to the study.

“These findings, taken together with findings from the small clinical trials that have tested ketamine for CUD, suggest additional clinical trials are warranted,” Davis and colleagues wrote.

References:

• Gao Z, et al. Addiction. 2023;doi:10.1111/add.16168.
• New study reveals ketamine could be effective treatment for cocaine-use disorders. https://thedaily.case.edu/new-study-reveals-ketamine-could-be-effective-treatment-for-cocaine-use-disorders/. Published Feb. 27, 2023. Accessed Feb. 28, 2023.