A new approach to personalized treatments in addiction

Published in Healthcare & Nursing
Like

Share this post

Choose a social network to share with, or copy the shortened URL to share elsewhere

This is a representation of how your post may appear on social media. The actual post will vary between social networks

Substance use disorders affect millions of people globally and over 40 million people in the United States. Rates of drug-related overdose deaths are climbing to levels never before reported. Importantly, individuals experience addiction, and vulnerability to addiction onset, in very different ways. We and others have shown that addiction is a complex disorder involving multiple mechanisms, such as abnormal processing of rewards, altered cognitive function, and heightened negative emotions - and that all of these mechanisms are linked to changes in brain function as well. At the same time, we know that the benefit seen from currently available addiction treatments remains limited. Up to 85% of people who attempt treatment will return to use in the long-term. Adding to the problem, only a fraction of people with addiction seek out treatment options. Using Alcohol Use Disorder as an example, the number of people with chronic alcohol misuse who seek treatment is as low as 8%. In part, this is due to people having previous negative experiences or low confidence in how successful treatment will be for them. Since we know that the mechanisms underlying addiction vary widely between people, we believe that this ‘treatment crisis’ is caused by a lack of personalized treatment options. If we could identify which mechanism is driving an individual’s drug use, we could provide them with a treatment that is focused on that mechanism specifically. For example, if someone is using drugs to alleviate heavy feelings of anxiety, a combination of medication and talk therapy to reduce anxiety may be the best option for them, whereas someone who primarily uses drugs to experience pleasure may not benefit from those treatments.   

Defining subgroups based on the mechanisms underlying addiction  

Our and our colleagues’ research has demonstrated that regardless of the specific drug used, there are at least three important mechanisms underlying addiction. The first of these is altered approach-related behavior, which includes behaviors such as sensation seeking and social risk-taking. The second is lower executive function, which includes cognitive flexibility, attention shifting, and planning. The third is increased negative emotionality, which includes internalizing behaviors, psychiatric symptoms, and aggression. To develop personalized treatments for addiction will require a more detailed understanding of the individual differences on these three mechanisms. In the addiction literature, these three mechanisms are often described as consistently, cyclically contributing to one another. However, psychological research covering many decades has reported that they are at least partially independent. Neuroscience research has confirmed that separable brain systems underlie function for each of them. Therefore, it seems necessary to assess individual differences across all three mechanisms.

 In the past, many research groups have looked into individual differences in addiction and attempted to define meaningful subtypes (or subgroups) within addiction. However, they have usually not used data that directly assesses function on the underlying mechanisms, but rather have used data that assesses severity of clinical symptoms. For example, they defined subtypes as simply low, moderate, or heavy users or as users experiencing early versus late-onset of their addiction. Our goal in this study was to identify subtypes defined by individual differences in function on the three main addiction-related mechanisms, hence using a “mechanism-based” subtyping approach. We further expected that “mechanism-based”  subtypes would also have distinct profiles in brain function, as assessed by functional magnetic resonance imaging (fMRI). 

What did we find? 

To address this, we used the Nathan Kline Institute Enhanced Rockland Sample (NKI-RS) data set. This is a large, community-representative sample of hundreds of individuals, 28% of whom had a history of addiction across different substances of abuse. We analyzed data that comprehensively assessed function on the three mechanisms described above in a total of 593 individuals. We included 74 different assessments per individual. To reduce this large amount of data to a set of underlying factors, we used a data-reduction method, known as factor analysis. This analysis summarized the data into 12 underlying ‘factors’: internalizing, psychiatric symptoms, effortful control, executive function, sensation seeking, unethical behavior, urgency, openness/sensitivity, extraversion/sociability, risk perception, negative affect, and social risk-taking. We then analyzed how these factors correlated to one another and determined that the factors described different aspects of the three functional mechanisms discussed above. 

Once we identified that these factors represented different aspects of the three mechanisms, we aimed to determine if subtypes existed. Subtypes were identified using a clustering method (specifically, latent profile analysis) on 173 individuals, all with a history of addiction. The goal was to identify if there were different subgroups with similar profiles in function based on their factor scores. This analysis revealed three subtypes that were each uniquely impaired on only one of the three mechanisms (See Figure). Specifically, we found 1) a “reward type” with altered reward-related behavior, including increased sensation-seeking, social risk-taking, and unethical behavior, 2) a “cognitive type” with executive function challenges including decreased openness/sensitivity and lower executive function, and 3) a “relief type” with greater negative emotionality including high internalizing, psychiatric symptoms, and negative affect. Demographic and clinical data aligned with the subtypes’ behavioral profiles. The “reward type” had higher current levels of drug use, the “cognitive type” had lower levels of education, and the “relief type” had a higher rate of internalizing disorders (e.g., anxiety, depression).

Next, we wanted to characterize each subtypes’ brain function. We investigated how their brain function was linked to their levels of current drug use. Each subtype displayed different, unique changes in brain function. In the “reward type”, brain function was altered in brain regions that have previously been implicated in abnormal reward-processing. In the “cognitive type”, brain function was different in brain regions that have been previously linked to cognitive functioning. In the “relief type” the altered brain regions were those that have been related to higher vigilance, a key behavior associated with negative emotionality, especially anxiety. In summary, each subtype showed unique profiles of brain function related to their current levels of drug use and aligned with their unique behavioral profiles.

What is the significance of our findings?

Our results add evidence for the relevance of a three-mechanism model for addiction across specific substance use disorders. Most importantly, however, these results stress the  importance of studying individual differences across these three mechanisms. We provide evidence of the existence of subtypes (or subgroups) within a sample of individuals with a history of addiction with very distinct behavioral profiles. Notably, these subtypes existed  independent of the primary drug of choice. Moreover, this research also demonstrates the value of defining subtypes using data that assesses underlying mechanisms, and establishes a new approach that may be of interest beyond the study of addiction. Finally, this work is a critical first step towards the development of personalized addiction treatments that are focused on the specific and different challenges an individual may be experiencing, with the goal of increasing treatment success rates in the future.

Figure: Phenotypic Profiles of the 3 Subtypes.

Subtypes were identified using a clustering method (specifically, latent profile analysis) on 173 individuals with a history of addiction. We found 1) a “reward type” with altered reward-related behavior, including increased sensation-seeking, social risk-taking, and unethical behavior, 2) a “cognitive type” with executive function challenges including decreased openness/sensitivity and lower executive function, and 3) a “relief type” with greater negative emotionality including high internalizing, psychiatric symptoms, and negative affect. Independent samples t-test were done to compare the subtypes statistically (p<.05*, p<.01**). Error bars are standard error.

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in