SCOTTSDALE, Arizona — The combination of a key neuroimaging biomarker and behavioral factors is a strong multimodal predictor of successful response to treatment for cocaine dependence. However, behavior offers accuracy at a lower cost, new research shows.
The results from a study conducted by investigators at Columbia University College of Physicians and Surgeons and the New York State Psychiatric Institute in New York City showed that patients who had an early response to treatment and were motivated, as reflected by their cumulative clinic attendance, had a predictive performance of 83% that was seen by the 3rd week of the 24-week study.
Similarly, positron-emission tomography (PET) imaging results revealed that the ventral striatal PET signal alone had strong accuracy in predicting individual cocaine addiction treatment response, with a cross-validated correct rate of 82%.
"What this shows is that in using PET, you can tell about 80% of the time who will be a responder or nonresponder," first author Sean X. Luo, MD, explained.
"But it also shows that you don't need PET ― you can just use the average clinic attendance in the first 3 weeks to tell who will be a responder and nonresponder, and these are equivalent measures in terms of predictive performance," Dr. Luo added.
The findings were presented here at the American Academy of Addiction Psychiatry (AAAP) 24th Annual Meeting & Symposium.
Clinical Marker Just as Good
Neuroimaging studies have previously shown the [11C]raclopride signal (a dopamine D2 receptor radioligand) in the ventral striatum on PET images to be associated with a substantially greater chance of success in a cocaine dependency treatment program that focuses on positive reinforcement.
In the new study, the evaluation of the PET images of 24 cocaine-dependent patients and 24-week follow-up supported those findings, but they also showed a correlation between the PET signal and patients who had regular adherence to treatment sessions, which predicted treatment success.
To compare the predictive rate with a clinical marker, the researchers chose treatment attendance, a sign of motivation.
"We wondered if you could substitute PET with a clinical marker that would be just as good. Since patient motivation has been long postulated as a key predictive variable in whether patients respond to treatment, we used early clinical attendance, which is easily measurable," Dr. Luo told conference delegates.
PET Scans Not Practical
The investigators also evaluated predictive models of PET signals from 5 regions of the striatum and behavioral factors among study participants.
They found the predictive performance in the PET imaging was not improved by the inclusion of information from other regions of interest in the striatum, with the exception of small improvement with adding the posterior caudate. Inclusion of baseline demographic variables in the model also did not improve predictive performance.
The PET signal identified in patients less likely to respond to treatment is consistent with previous research showing that chronic addiction and acute activation of the brain reward system result in a depletion and underactivity of the dopamine system that, in turn, can partly explain poor treatment response, Dr. Nunes explained.
"The depleted dopamine system, in theory, leads to an inability to respond to normal, healthy reinforcers ― the kinds of social rewards inherent in participating in treatment, or doing the kinds of things treatment encourages, like spending time on friends, family, recreation, or work, rather than taking drugs," he said.
Whereas PET scans are still far too expensive to have practical use in routine clinical practice, the study's finding that the PET signal correlated with patient behaviors leading to a greater chance of successful treatment adds important validation to the value of the behavioral predictors on their own, said Dr. Luo.
"We spend so much money and time trying to figure out if there is a biomarker for response, yet, this shows that according to our own intuition as physicians, we can do just as well with a careful physical and history," he said.
A Guide for Future Studies
Commenting on the findings for Medscape Medical News, Byron Adinoff, MD, Veterans Affairs North Texas Health Care System and distinguished professor in drug and alcohol abuse research at the University of Texas Southwestern Medical Center, in Dallas, said that the study contributes important insights into the understanding of the inner workings of the cocaine-addicted brain, behavioral characteristics, and the response to treatment.
"This is very interesting and promising study," he told Medscape Medical News.
"It offers important information on the relationship between striatal dopamine binding potential, ie, the number of dopamine receptors in the brain coupled with how strongly dopamine binds to the receptor, and relapse."
"This offers a guide for future studies as to how clinicians could target specific biological disruptions in the treatment of cocaine addiction," he said.
He noted, however, some important considerations include the utility of the PET signal as a marker for clinical treatment, including the fact that to make [11C]raclopride, an on-site cyclotron is needed, which is extremely expensive to buy and maintain.
Dr. Nunes, first author Sean Luo, MD, and coauthor Diana Martinez, MD, reported no relevant financial relationships. Coauthor Kenneth M. Carpenter, MD, is a part-time employee for Center for Motivation and Change. Coauthor Mark Slifstein, PhD, is a consultant for Amgen and receives research support from Pierre Fabre. Dr. Adinoff reported that he has received funding from the National Institute on Drug Abuse.
American Academy of Addiction Psychiatry (AAAP) 24th Annual Meeting & Symposium: Presented December 7, 2013.
Medscape Medical News © 2013 WebMD, LLC
Send comments and news tips to news@medscape.net.
Cite this: Cocaine Addiction: Patient Behavior Predicts Treatment Success - Medscape - Dec 12, 2013.
Comments