Personalizing the Treatment of Substance Use Disorders

To achieve a dimensional approach to treatment requires thinking anew about how we develop new treatments and what we expect in a treatment.

The existing pharmacopoeia for substance use disorders is severely limited. The FDA has approved medications only for alcohol, nicotine, and opioid use disorders (Table 1), and currently there are no approved medications for cannabis, cocaine, methamphetamine, or inhalant use disorders. The absence of medications to treat most substance use disorders and the limited number of existing medications for alcohol, nicotine, and opioid use disorders make development of new therapeutics a high priority. Yet drug development for substance use disorders faces great hurdles.

To obtain FDA approval for most substance use disorders, medications until recently had to demonstrate that they produce abstinence in a significant subset of patients, as measured by negative urine tests. However, the abstinence endpoint is a high bar to achieve, equivalent to requiring remission of pain from an analgesic or remission of depression from an antidepressant. Yet, the FDA granted approval of analgesics and antidepressants on the basis of reduction of symptom severity, not remission (2). The high bar for addiction medications has discouraged investment by the pharmaceutical industry, and significant public sector help was required to bring many of the currently available medications for substance use disorders to market, including buprenorphine, extended-release naltrexone, lofexidine, and naloxone nasal spray.

Treatment programs for substance use disorders inherited a dichotomous working definition of recovery from the 12-step world of past generations, where being completely “drug free” was not merely the gold standard but the only standard, short of which an addicted individual was regarded as having failed or would not be considered to be “recovering.” Yet evidence indicates that abstinence is not the only clinically relevant outcome for every individual and that alternative endpoints can contribute to recovery even when abstinence is not completely achieved.

Reduced alcohol use (measured as percentage of heavy drinking days) is now being used as an endpoint in clinical trials for treatments for alcohol use disorder. The FDA has also recently expressed its openness to considering endpoints other than abstinence as targets in medication development for other substance use disorders (3, 4). Given the illegality of many addictive drugs, it has been argued that any reduction in use should be considered a benefit to the individual’s health and safety (5). Every time a person addicted to heroin must obtain the drug, he or she faces the risks associated with the drug trade as well as with exposure to fentanyl or a contaminant that could lead to overdose or poisoning.

Recently, researchers found in a pooled sample of study participants with cocaine use disorder that those who had high-frequency use at the start of the study and had reduced to low-frequency use by the end of the study showed outcomes at 1-year follow-up similar to those of participants who had quit altogether (6).

Endpoints other than abstinence may lead not only to treatments that are helpful in reducing drug use but also to the use of compounds that target specific neurobiological processes and symptoms relevant to addiction and the risk for relapse.

In April 2018, the FDA, in partnership with the Addiction Policy Forum and the National Institute on Drug Abuse (NIDA), convened a meeting to solicit input from patients with opioid use disorder as part of its Patient-Focused Drug Development initiative (7). Among other things, participants emphasized their desire for a more holistic and individualized approach to treatment, as well as their wish for medications that would address specific symptoms of withdrawal, such as cravings, depression, cognitive impairments, pain, and sleep problems. The same year, the FDA approved lofexidine for treating physical symptoms of opioid withdrawal during detoxification—the first approved drug for treating symptoms associated with opioid use disorder with a restricted purpose and not expected to lead, by itself, to continued abstinence. After detoxification, the individual would ideally be treated with naltrexone or buprenorphine as a longer-term treatment to help prevent relapse and achieve recovery. Other potential targets for medications are those that, while not addressing addiction directly, target major risk factors for relapse.

One such factor is insomnia, for it is frequently interrelated with substance use disorders, with each exacerbating the risk of the other. Findings of shared targets and circuits between disrupted sleep and addiction offer unique opportunities for treatment development. For example, while studying the role of orexin in narcolepsy, researchers serendipitously discovered an unusually high number of orexin-producing neurons in the postmortem brain of a heroin-addicted individual (8). They subsequently established in preclinical models and postmortem brain studies that long-term use of heroin was associated with an increase in orexin-producing neurons. Since orexin is already targeted by suvorexant, an FDA-approved drug for insomnia, NIDA is funding research to test its efficacy, along with that of other novel orexin receptor antagonists, as therapeutic agents in opioid use disorder.

Similarly, dysphoria and depression, which are frequently associated with protracted withdrawal, are another relevant area where our growing understanding of underlying neurocircuitry could guide selection of promising new targets. For example, the habenular complex is intricately involved in dysphoria and negative emotional states and is associated with depression (9) and addiction (10). Both alpha-5 nicotinic acetylcholine receptors and mu-opioid receptors are highly expressed in the habenula, where they modulate its activity, contributing to the adverse symptoms of withdrawal that follow nicotine and heroin discontinuation, respectively, and to the relief that follows during intoxication. Targeting the habenula has already been shown to be beneficial in animal models of addiction treatment (11), and it has been a target for deep brain stimulation for the treatment of depression (12).

Because of the high comorbidity of substance use disorder with depression, psychiatrists have used antidepressants off-label to treat their addicted patients, even though randomized clinical trials of antidepressants have failed to achieve the desired outcome of abstinence. Recognizing that improving depression could still be beneficial for patients with substance use disorders, studies should revisit the possible efficacy of antidepressants as an element of addiction treatment, using endpoints other than abstinence. Bupropion, which blocks the dopamine and norepinephrine transporters and is an approved antidepressant medication, is also approved for the treatment of nicotine addiction. Given the involvement of the mu-opioid receptor system in mood, it would be expected that targeting depression might have particular value in treating opioid use disorder; an interesting feature of the opioid partial agonist buprenorphine is that it has antidepressant properties (13), and opioid-addicted patients who have depression respond particularly well to this medication (14).

Another important therapeutic target is that of addressing social isolation, and while this might be optimally achieved with behavioral interventions, including group treatment, medications could still hold promise. Addicted individuals report reduced pleasure from social contact, as well as fear of the stigma attached to their drug use, and thus they tend to isolate themselves. Isolation in turn drives drug taking (15). Here again, we could take advantage of our increased understanding of the neurobiology of social attachment to bolster social connections. For example, oxytocin, a neurochemical involved in social bonding that also modulates key processes associated with addiction, including reward and stress responses, is being evaluated as a possible addiction treatment and may enhance the efficacy of psychosocial addiction treatments (16, 17).

A dimensional approach to the treatment of substance use disorder is also relevant to neuromodulation. Early research has shown that transcranial magnetic stimulation (TMS) and transcranial direct current stimulation (tDCS) may be useful in reducing drug cravings, and TMS is already an approved therapy for treatment-resistant depression. Research is needed to study how TMS, tDCS, or peripheral nerve stimulation could be used to improve symptoms associated with addiction, from acute symptoms of withdrawal to the more protracted symptoms of dysphoria and sleep problems. As we understand better how to use neuromodulation technologies to modify brain circuits, it may create opportunities to strengthen specific circuits that can buffer or compensate for others that have been impaired by drug use or constitute a predisposing vulnerability.

Behavioral therapies are also suited to dimensional approaches to substance use disorder treatment. Considerable research already shows the benefits of cognitive-behavioral treatments in improving self-regulation and of contingency management in strengthening the degraded motivation to engage in non-drug-related activities, so clearly these modalities are effective for addressing specific dimensions of the addiction process. Similarly, behavioral treatments to improve executive function could help build resilience against relapse, as shown by methylphenidate’s reported ability to reduce impulsivity in individuals with cocaine use disorder (18).

Trajectories of use vary among people who use drugs, ranging from persistent use or declining use to cessation and relapse or sustained cessation. Studies of people who inject opioids, for example, have identified factors that, to some extent, are predictive of these trajectories (19). Being in a stable relationship, for instance, has been associated with early cessation (highlighting the importance of social support).

Addiction is an evolving disorder that changes through time and across the lifespan of the individual and one that has an unpredictable element that springs from the unique experiences an individual is exposed to. Some widely used behavioral treatments already accommodate and address this changeability of substance use disorder. Cognitive-behavioral therapy teaches the individual to identify external triggers and respond more appropriately to internal states (e.g., mood, craving) that place them at risk for relapse. New technologies are developing algorithms to identify indicators of relapse risk and incorporating them into wearable devices and smartphones with the goal of delivering an intervention in a timely, targeted manner. In the future, as big-data analytics and machine-learning algorithms yield more insight into behavioral and biological markers of relapse risk, tools or devices to avert relapse farther in advance may be developed.

Neuroscience has revealed that addiction involves a set of interconnected processes that can be targeted strategically, rather than being a disorder defined principally by a single behavior (uncontrollable excessive drug use). Addiction medicine is also increasingly recognizing that factors traditionally associated with recovery are components of treatment. For example, for any meaningful recovery to occur, the individual must be able to integrate him- or herself into a socially meaningful environment. People with substance use disorders who are professionally active or engage in meaningful activity and have a caring family face less of a challenge than those who have no social supports and whose isolation places them at high risk for relapse. The integration of peer mentors, recovery coaching, and supportive housing into addiction treatment is an example of this shift, but more research is needed to determine the most effective ways to sustain social inclusion and to achieve recovery (20).

Addiction is a complex disorder that involves brain circuits necessary for survival and one that is strongly influenced by genes, development, and social factors. We now understand the underlying mechanisms well enough that we can turn this complexity into an opportunity to include these dimensions as targets for substance use disorder treatment, as well as to personalize interventions to accommodate the unique neurobiological characteristics and social contexts of individual patients.

The author thanks Eric M. Wargo and Emily B. Einstein for their valuable help in the preparation of this article.