Integrating precision medicine in the study and clinical treatment of a severely mentally ill person
Subject areasPsychiatry and Psychology, Genomics, Neurology, Translational Medicine, Ethical Issues
Deep brain stimulation (DBS) has emerged as a relatively safe and reversible neurosurgical technique that can be used in the clinical treatment of traditionally treatment resistant psychiatric disorders. DBS enables the adjustable and stable electrical stimulation of targeted brain structures. A recent paper by Höflich et al. (2013) notes variability in treatment outcomes for DBS patients, which is likely due to variable responses to differences in targeted stimulation regions and in post-operative stimulation parameters. Both sources of variation, the authors note, will effect the stimulation of different brain tissue fibers having different anatomical and functional connections. Furthermore, the authors suggest that not every target will be suitable for every person, as there exists a large degree of inter-individual variability of brain region activation during a reward task in healthy volunteers, and suggest that future work could (and should) focus on developing surgical plans based on individual-specific activations, functional connectivity and/or tractography. This work exemplifies the large degree of clinically relevant biological variability that exists in terms of individual clinical characteristics.
Ongoing clinical trials testing the “Effectiveness of Deep Brain Stimulation for Treating People With Treatment Resistant Obsessive-Compulsive Disorder” (Greenberg (2013)) detail the below exclusion criteria: •current or past psychotic disorder,•a clinical history of bipolar mood disorder, and/or•an inability to control suicide attempts, imminent risk of suicide in the investigator’s judgment, or a history of serious suicidal behavior, which is defined using the Columbia-Suicide Severity Rating Scale (C-SSRS) as either: one or more actual suicide attempts in the 3 years before study entry with the lethality rated at 3 or higher, or one or more interrupted suicide attempts with a potential lethality judged to result in serious injury or death.
These study criteria exclude the most severe cases of OCD, as many people with severe OCD also have severe depression, usually with passive (and sometimes active) suicidal ideation (Torres et al. (2011); Alonso et al. (2010); Balci & Sevincok (2010)). Obsessions and compulsions can be quite severe, with very poor insight, sometimes to a delusional or psychotic degree, and there can also be co-occurring psychoses in any individual. Each person is to some degree unique in his or her psychiatric presentation, and a tailored evaluation schema could prove more effective in clinical treatment. Due in part to these above hurdles, there are few detailed descriptions of the efficacy of DBS for OCD, with the number of published case studies on the efficacy of DBS for OCD covering upwards of ∼100 people (Roh et al. (2012); Goodman & Alterman (2012); Blomstedt et al. (2012); Burdick & Foote (2011); Mian et al. (2010); Haynes & Mallet (2010); Goodman et al. (2010); Denys et al. (2010); Komotar, Hanft & Connolly (2009); Jimenez-Ponce et al. (2009); Denys & Mantione (2009); Burdick, Goodman & Foote (2009); Shah et al. (2008); Figee et al. (2013); Lipsman et al. (2012); Lipsman, Neimat & Lozano (2007)).
An explosive growth in exome and whole genome sequencing (WGS) (Lyon & Wang (2012)) has occurred in parallel to the emergence of DBS for OCD, led in part by dramatic cost reductions. This in turn has given medical practitioners an efficient and comprehensive means to medically assess coding and non-coding regions of the genome, leading to much promise in terms of assessing and treating human disease. In our own efforts to push forward the field of precision medicine, we report here one effort to integrate the areas of clinical neuropsychiatry, brain machine interfaces and personal genomics in the individualized care of one person. We evaluate and treat an individual with DBS for treatment refractory OCD, gauge the feasibility and usefulness of the medical integration of genetic data stemming from whole genome sequencing, and search for rare variants that might alter the course of medical care for this person. As mentioned above, there have been relatively few reports on studies detailing the effective application of DBS for OCD; we report here one such study.
Research was carried out in compliance with the Helsinki Declaration. The corresponding author (GJL) conducted all clinical evaluations and he is an adult psychiatry and child/adolescent psychiatry diplomate of the American Board of Psychiatry and Neurology. GJL obtained IRB approval #00038522 at the University of Utah in 2009–2010 to evaluate candidates for surgical implantation of the Medtronic Reclaim® DBS Therapy for OCD, approved under a Humanitarian Device Exemption (HDE) for people with chronic, severe, treatment-resistant OCD (). The interdisciplinary treatment team consisted of one psychiatrist (GJL), one neurologist and one neurosurgeon. Implantation ultimately occurred on a clinical basis at another site. Written consent was obtained for phenotyping and whole genome sequencing through Protocol #100 at the Utah Foundation for Biomedical Research, approved by the Independent Investigational Review Board, Inc. Informed and written consent was also obtained using the Illumina Clinical Genome Sequencing test consent form, which is a clinical test ordered by the treating physician, GJL.
GJL received training regarding DBS for OCD at a meeting hosted by Medtronic in Minneapolis, Minnesota, in September 2009. The same author attended a Tourette Syndrome Association meeting on DBS for Tourette Syndrome, Miami, Florida, in December 2009. Approximately ten candidates were evaluated over a one-year period in 2010. The individual discussed herein received deep brain stimulation surgery at another site, and then returned for follow-up with GJL. Another psychiatrist, author RR, provided ongoing consultation throughout the course of this study. Although other candidates have since returned for follow-up (with GJL), no others have been surgically treated.
A 37-year old man and US veteran (here named with pseudonymous initials MA) was evaluated by GJL in 2010 for severe, treatment-refractory obsessive compulsive disorder (OCD), which is an illness that can be quite debilitating (Murphy, Zine & Jenike (2009)). MA had a lifelong history of severe obsessions and compulsions, including contamination fears, scrupulosity, and the fear of harming others, with much milder symptoms in childhood that got much worse in his early 20’s. His Yale-Brown Obsessive Compulsive Scale (YBOCS) (Goodman et al. (1989a); Goodman et al. (1989b)) ranged from 32 to 40, indicating extremely severe OCD. Perhaps the worst period of OCD included a 5-day, near continuous, period of tapping on his computer keyboard as a compulsion to prevent harm from occurring to his family members. MA had suffered throughout his life from significant periods of depression with suicidal ideation, and he had attempted suicide at least three times. His prior psychiatric history also includes episodes of paranoia relating to anxieties from his OCD, and he continues to be treated with biweekly injections of risperidone. His Global Assessment of Functioning (GAF) typically ranged from 5 to 15 on a 100 point scale.
His treatment history included over 15 years of multiple medication trials, including clomipramine and multiple SSRIs at high doses, including fluoxetine at 80 mg by mouth daily, along with several attempts with outpatient exposure and ritual prevention (ERP) therapy (Gillihan et al. (2012)). MA inquired and was evaluated by GJL at the University of Utah and then at two other centers independently offering deep brain stimulation for OCD. One of these centers required (as a condition for eligibility for an ongoing clinical trial) a two-week inpatient hospitalization with intensive ERP, which subsequently occurred and was documented as improving his YBOCS score to 24 at discharge. He maintains that he actually experienced no improvement during that hospitalization, but rather told the therapists what they wanted to hear, as they were “trying so hard”. See the File S11 for other clinical details.
The teams at the University of Utah and two other centers declined to perform surgery due to his prior history of severe depression, suicide attempts and possible psychoses with paranoia. Through substantial persistence of MA and his family members, a psychiatrist and neurosurgeon at a fourth center decided that he was an appropriate candidate for surgical implantation of the Medtronic Reclaim® DBS Therapy device for OCD, approved under a Humanitarian Device Exemption (HDE) for people with chronic, severe, treatment-resistant OCD (), and he was implanted in January of 2011 (Fig. 1). The device targets the nucleus accumbens/anterior limb of the internal capsule (ALIC). A detailed account of the surgical procedure can be found in the File S11.
After healing for one month, the implanted device (equipped with the Kinetra Model 7428 Neurostimulator) was activated on February 14, 2011, with extensive programming by an outpatient psychiatrist, with bilateral stimulation of the ALIC. Final settings were case positive, contact 1 negative on the left side at 2.0 V, frequency 130 Hz, and pulse width 210 µs, and case positive, contact 5 negative on the right side with identical settings.
Over the next few months, his voltage was increased monthly in increments of 0.2–0.5 V by an outpatient psychiatrist. He returned to one of the author’s (GJL) for psychiatric treatment in July 2011, at which time his voltage was set at 4.5 V bilaterally. His depression had immediately improved after the surgery, along with many of his most irrational obsessions, but his YBOCS score still remained in the 35–38 range. From July 2011–December 2011, his voltage was increased bilaterally on a monthly basis in increments of 0.2 V, with steady improvement with his OCD until his battery started to lose charge by December 2011. This caused him considerable anxiety, prompting him to turn off his battery in order to “save battery life”, which unfortunately led to a complete relapse to his baseline state in a 24 h period, which was reversed when he turned the battery back on. The battery was surgically replaced with a rechargeable Activa RC neurostimulator battery in January 2012, and the voltage has been increased monthly in 0.1–0.2 V increments until the present time (May 2013).
At every visit, MA has reported improvements, with reductions of his obsessions and compulsions, marked by an overall decline in his YBOCS score (Fig. 3). MA has started to participate in many activities that he had never previously been able to engage in. This includes: exercising (losing 50 pounds in two years) and volunteering at the church and other organizations, but not yet being able to work in any paid capacity. MA also started dating and recently got married, highlighting his improvement in daily functioning, with a GAF score ranging now from 40–50. New issues that MA reports are consistent tenesmus, occasional diarrhea (which he can now tolerate despite prior contamination obsessions) and improved vision (going from 20/135 to 20/40 vision, as documented by his optometrist), with him no longer needing to wear glasses. It is unknown whether the DBS implant has contributed to any of these issues. Attempts to add fluoxetine at 80 mg by mouth daily for two months to augment any efficacy from the DBS and ERP were unsuccessful, mainly due to no discernible benefit and prominent sexual side effects. MA still receives an injection of 37.5 mg risperidone every two weeks for his past history of psychoses; otherwise, he no longer takes any other medications. There has not been any exacerbation of psychoses in this individual during the two years of treatment with DBS.
Deep brain stimulation for MA’s treatment refractory OCD has provided a quantifiable and significant improvement in the management of his symptoms (Fig. 3). MA has regained a quality of life that he had previously not experienced in over 15 years, which is highlighted by his participating in regular exercise, working as a volunteer in his local church, dating, and eventually getting married, all of which illustrate a dramatic improvement in his daily functioning since receiving DBS treatment for his OCD.
One significant aspect of this study is the rechargeable, and hence depletable, nature of the Activa RC neurostimulator battery, which serves to illustrate the efficacy of DBS for OCD for this individual. On one such illustrative occasion, MA forgot to take the recharging device on a four-day weekend trip. Once his battery was depleted, all of his symptoms gradually returned to their full level over a ∼24 h period, including severe OCD, depression and suicidality. Since that episode, MA always takes his recharging device with him on extended trips, but there have been other such instances in which his battery has become depleted for several hours, with the noticeable and intense return of his OCD symptoms and the cessation of his tenesmus. The electrical stimulation is having a demonstrable effect on his OCD, and these data are complementary to other data-sets involving turning DBS devices off for one week at a time (Figee et al. (2013)).
There are many ethical and regulatory issues relating to deep brain stimulation that have been discussed elsewhere (Fins, Dorfman & Pancrazio (2012); Synofzik, Fins & Schlaepfer (2012); Fins et al. (2011a); Fins et al. (2011b); Fins & Schiff (2010); Fins (2010); Erickson-Davis (2012)), and we report here our one positive experience, made possible when the US Food and Drug Administration granted a Humanitarian Device Exemption (HDE) to allow clinicians to use this intervention. The rechargeable nature of the new battery has been reassuring to MA, as he is able to exert control over his battery life, whereas he previously had no control with the original “single-use” battery that must be replaced when the battery depletes (usually at least once annually). We assume that other persons treated with DBS for OCD will likely also start receiving rechargeable batteries. In this regard, it is worth noting that the recent development of an injectable class of cellular-scale optoelectronics paves the way for implanted wireless devices (Kim et al. (2013)), and we fully expect that there will be more brain-machine neural interfaces used in humans in the future (Alivisatos et al. (2013a); Alivisatos et al. (2013b); Pais-Vieira et al. (2013); Thomson, Carra & Nicolelis (2013); Nicolelis (2012)).
There are still many challenges in showing how any one mutation can contribute toward a clear phenotype, particularly in the context of genetic background and possible environmental influences (Moreno-De-Luca et al. (2013)). Bioinformatics confounders, such as poor data quality (Nielsen et al. (2011)), sequence inaccuracy, and variation introduced by different methodological approaches (O’Rawe et al. (2013)) can further complicate biological and genetic inferences. Although the variants discussed in the results section of our study have been previously associated with mental disease, we caution that the data presented are not sufficient to implicate any particular mutation as being necessary or sufficient to lead to the described phenotype, particularly given that mental illness results from a complex interaction of any human with their surrounding environment and social support structures. The genetic architecture of most neuropsychiatric illness is still largely undefined and controversial (Klei et al. (2012); Mitchell & Porteous (2011); Mitchell (2012); Visscher et al. (2011)). We provide our study as a cautionary one: WGS cannot act as a diagnostic and prognostic panacea for neuropsychiatric disorders, but instead could act to elucidate risk factors for psychiatric disease and pharmacogenetic variants that can inform future medication treatments.
During our study, we found that MA carries at least three alleles that have been associated with neuropsychiatric phenotypes, including variants in BDNF, MTHFR, and ChAT (Table 1). And, although we have discovered informative pharmacogenetic variants in this person, these discoveries have not led to the immediate alteration of this person’s medication schema. We have archived these discoveries, as described below, and expect that these variants will be useful over the course of his life-long medical care. We feel that this information is inherently valuable, as one can never predict with certainty what the future might hold, and a more complete medical profile on individual patients will facilitate more informed medical choices.
One can learn a substantial amount from detailed study of particular individuals (for just a small sampling, see Sacks (1995); Sacks (1998); Luria (1972); Luria (1976); Van Horn et al. (2012); Ratiu et al. (2004); Eichenbaum (2013); Worthey et al. (2011)), and we believe that we are entering an era of precision medicine in which we can learn from and collect substantial data on informative individual cases. Incorporating insights from a range of scientific and clinical disciplines into the study and treatment of any one person is therefore beginning to emerge as a tractable, and more holistic, approach, and we document here what we believe to be the first integration of deep brain stimulation and whole genome sequencing for precision medicine in the evaluation, treatment and preventive care for one severely mentally ill individual, MA. We have shown that DBS has been successful in aiding in the care and beneficial clinical outcome of his treatment refractory OCD, and we have also demonstrated that it is indeed feasible, given current technologies, to incorporate health information from WGS into the clinical care of one person with severe mental illness, including with the return of these health information to him directly. On a comparative level, deep brain stimulation has thus far been a more direct and effective intervention for his mental illness than anything discovered from his whole genome sequencing. Despite this, health information stemming from these WGS data was nevertheless immediately useful in the care of this person, as a variant associated with his ophthalmologic phenotype did indeed inform and enrich his care, and we expect that these data will continue to inform his care as our understanding of human biology and the genetic architecture of disease improves. Of course, the genomic data would have been more helpful if obtained much earlier in his medical course as it could have provided guidance on which medications to avoid or to provide in increased doses.