21b
Luis C. Farhat and Marc N. Potenza
The chapter by Links et al. (this volume) reviews the current state of understanding regarding medication management for personality disorders (PDs). The authors note that, compared to many other psychiatric disorders (especially those termed Axis I disorders in earlier editions of the Diagnostic and Statistical Manual; American Psychiatric Association, 2013), there are relatively few (if any) well validated pharmacotherapies for PDs, with most of the research focused on individuals with borderline personality disorder (BPD). The authors discuss limitations of many of the studies in this area, including the use of small sample sizes and uncontrolled designs, and note inconsistencies in recommendations regarding the extent to which medications are advised for administration in clinical settings. Whereas the United Kingdom National Institute for Healthcare and Excellence (NICE, 2015) does not recommend the use of psychopharmacologic agents in the treatment of individuals with BPD, the American Psychiatric Association supports the use of medications in a symptom-driven approach (APA, 2001). Further, medications may be used to treat other disorders that commonly co-occur with PDs. Indeed, the frequent co-occurrence of PDs with other psychiatric disorders may, in part, explain the observations in clinical practice settings that most people with PDs are prescribed pharmacotherapies.
In this commentary, we expand upon the important points raised by the authors regarding patient involvement in pharmacotherapy, consideration of adverse effects and quality of life, and the importance of therapeutic alliance in these processes. We also review some of the limitations of clinical trials for medications for PDs to date, and explore reasons why empirically supported pharmacotherapies for PDs may be lagging behind those for many other psychiatric disorders. Additionally, we discuss how some of these limitations may interfere with the aggregation and interpretation of study results, as well as the translation of findings to clinical settings. Finally, we hypothesize how different approaches, such as targeting symptom domains or focusing on co-occurring disorders (using gambling disorder as an example), could be potentially beneficial to the development of efficacious pharmacotherapies for PDs.
Historically, PDs were considered, at least by some, to be largely stable constructs. As such, PDs were considered Axis II disorders, differing from Axis I conditions that were thought to be more amenable to change, particularly with medications. This conceptualization may, in part, explain why there have been relatively fewer systematic large-scale pharmacotherapy randomized clinical trials for PDs than for most Axis I disorders.
Nonetheless, Links et al. (this volume) note that a broad range of classes of medications (mood stabilizers, antipsychotics, antidepressants, anxiolytics, and others) have been examined for PDs, especially borderline personality disorder (BPD), with many studies showing some positive findings, particularly within specific symptom domain categories. However, they also note multiple concerns with medication management of PDs (e.g., high drop-out and non-compliance rates; adverse effects that may impact quality of life). These represent important factors that must be taken into consideration when weighing the pros and cons of medications; for example, whereas antipsychotics like olanzapine may help specific symptoms in BPD, adverse effects relating to weight gain may lead to metabolic syndrome and poorer quality of life. As such, having a good therapeutic relationship that actively involves the patient in the pharmacotherapeutic process is extremely important in the treatment of individuals with PDs, as discussed in greater detail by Links et al.
As described by Links et al., most trials reviewed in the chapter show some promising results in the treatment of PDs, particularly with respect to specific symptom domains in BPD. However, existing trials of PDs often have several important methodological limitations that may hamper how their results are translated or transferred into clinical settings (Bateman, Gunderson, & Mulder, 2015).
Researchers may face challenges when determining the target characteristics of the samples to be studied in clinical trials. Like other psychiatric disorders, PDs are heterogeneous conditions. Wright et al. (2013) stated that heterogeneity in BPD is likely to influence how treatments may best be delivered and the outcomes of these treatments. Patients with PDs usually have more than one psychiatric diagnosis. For example, Zimmerman and Mattia (1999) found that 69.5 percent of individuals with BPD had three or more DSM-IV Axis I disorders, and 47.4 percent presented with four or more such diagnoses. Whereas researchers acknowledge heterogeneity related to comorbidities and try to control for this variability through statistical means, inclusion/exclusion criteria, and other approaches, the resulting approach may limit generalizability to clinical settings. Further, although some statistical approaches may be appropriate for disorders characterized by low rates of comorbidity and heterogeneity, they may not be as useful for disorders that are highly heterogeneous, such as PDs (Joyce, Kehagia, Tracy, Proctor, & Shergill, 2017).
Another strategy researchers may employ to reduce heterogeneity is to define inclusion criteria involving specific severity thresholds. For example, Black et al. (2014) only included BPD patients with total scores of 9 or more on the Zanarini Rating Scale for BPD (ZAN-BPD; Zanarini et al., 2003). Although this strategy has some benefits, it may only account for quantitative differences between patients. As many possible combinations of responses to a scale result in the same total score, PD trials may enroll participants who experience different aspects of PD psychopathology, and this difference may be overlooked. Therefore, such qualitative differences may not be properly addressed and may complicate findings and their interpretation. This effect has been termed weak aggregation (Joyce, Tracy, & Shergill, 2017).
Heterogeneity related to co-occurring disorders may also represent an opportunity to test and develop treatment algorithms. For example, consider gambling disorder, a condition for which there is currently no indicated medication. Based on existing data from pharmacotherapy trials that have selected patients based on patterns of co-occurring disorders, pharmacotherapy recommendations stemming from the presence or absence of specific disorders may be generated (Bullock & Potenza, 2012) and updated based on subsequent findings (e.g., de Brito et al., 2017; Grant et al., 2014; Grant, Potenza, Kraus, & Petrakis, 2017). This approach of employing pharmacological interventions based on co-occurring disorders is likely to resonate with prescribing physicians trained to evaluate patients systematically for the presence or absence of specific disorders.
Another important consideration involves the outcome measures employed in PD trials. Currently, there are no universally accepted instruments to report changes in PD psychopathology. For example, the ZAN-BPD was initially developed as an interview-based instrument, and, more recently, a self-report version was published (Zanarini, Weingeroff, Frankenburg, & Fitzmaurice, 2015). The ZAN-BPD evaluates changes in BPD psychopathology, and it has become well accepted as an important instrument to report improvement in BPD psychopathology in clinical trials (Black et al., 2014; Zanarini et al., 2011). Nevertheless, other recent trials have employed different outcome measures. For example, Rohde and colleagues (Rohde, Polcwiartek, Correll, & Nielsen, 2017) evaluated how their intervention influenced the number of psychiatric admissions, psychiatric bed-days, concomitant medications, serious adverse effects, and intentional self-harm or overdose. Meanwhile, a study by Bellino and colleagues (Bellino, Paradiso, Bozzatello, & Bogetto, 2010) used another measure, the BPD Severity Index (BPDSI; Arntz et al., 2003). The use of different outcome measures complicates comparison of results across trials. Consensus statements regarding how best to assess treatment outcome, as have been generated for studies of gambling disorder (Walker et al., 2006), may help in harmonizing measures across studies.
How improvements in outcome measures reported in trials relate to clinical well-being and quality of life is also complicated. One standard approach compares changes in endpoint scores between the active and control groups. If the difference is significant, efficacy is proposed. However, a statistical difference does not necessarily correspond to a significant clinical response. This is also a concern for other psychiatric conditions, and may explain why other disorders, such as trichotillomania (Houghton et al., 2015), do not have indicated pharmacologic treatment options.
Adverse effects also warrant consideration. One of the largest randomized clinical trials to date of a medication for BPD involved olanzapine, with some positive effects noted in specific symptom domains. However, the use of second-generation antipsychotics like olanzapine may result in significant weight gain, metabolic syndrome, and cardiovascular morbidity and mortality (Newcomer, 2007). Given that many individuals with BPD demonstrate weight gain over time, adverse effects such as these are important to monitor. Likewise, valproic acid and other anticonvulsants may have teratogenic effects (Ornoy, 2009); thus, given that most patients with BPD are women (American Psychiatric Association, 2013; Ten Have et al., 2016), clinicians should consider the risks and benefits of these medications, as well as possible alternatives. In light of these adverse effects, quality of life measures are important to administer and discuss with patients on an ongoing basis.
The issues described above may be addressed using different approaches. A current debate involves the relative utility of dimensional versus categorical approaches to psychopathology, with the research domain criteria (RDoC) approach representing an important example of the former (Insel et al., 2010). The Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV, American Psychiatric Association, 1994, p. 633) stated that “an alternative to the categorical approach is the dimensional perspective that Personality Disorders represent maladaptive variants of personality traits that merge imperceptibly into normality and into one another.” Consistent with this statement, several groups have proposed changes to PD assessment. Trull and colleagues (Trull, Tragesser, Solhan, & Schwartz-Mette, 2007) called attention to problems of the categorical diagnosis, such as poor stability of the PD diagnosis over time and high rates of heterogeneity among individuals with the same PD diagnosis. They proposed that a dimensional model would provide higher inter-rater reliability in assessing PD symptomatology and account for the aforementioned heterogeneity and deficits in diagnostic stability over time. Moreover, after evaluating several dimensional models for conceptualizing personality and PDs, they concluded that, despite their differences, these models overlapped in multiple ways. Widiger and Trull (2007) also discussed problems associated with categorical diagnoses of PDs and proposed that PDs be conceptualized within the dimensional Five-Factor Model. Others have supported this approach, with one survey finding that four out of five experts believe that PDs should be examined from a dimensional perspective (Bernstein, Iscan, & Masner, 2007).
PD clinical trials are increasingly integrating dimensional constructs, for example, by evaluating how risperidone may be safe and efficacious in the treatment of specific aspects of impulsivity in BPD. This approach should be expanded further – not only to relevant constructs within BPD, but also to dimensions specific to other PDs in respective trials of medications for these disorders. How researchers and clinicians in this field may best utilize both dimensional and categorical approaches to treatment development is an important and timely consideration that will likely have significant impact.
In conclusion, psychopharmacologic treatment of PDs is an emerging area that is presently difficult to evaluate, due to the limitations of the trials conducted to date. As additional data are generated using approaches that permit greater harmonization across studies, employ larger sample sizes, consider heterogeneities, and incorporate dimensional as well as categorical measures, it is hoped that there may be more safe and effective pharmacological treatments for people with PDs.
References
American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental Disorders (4th ed.). Washington, DC: American Psychiatric Association.
American Psychiatric Association. (2001). Practice guideline for the treatment of patients with borderline personality disorder. American Journal of Psychiatry, 158, 1–52.
American Psychiatric Association. (2013). Diagnostic and Statistical Manual of Mental Disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.
Arntz, A., Van den Hoorn, M., Cornelis, J., Verheul, R., van den Bosch, W. M., & de Bie, A. J. (2003). Reliability and validity of the borderline personality disorder severity index. Journal of Personality Disorders, 17, 45–59.
Bateman, A. W., Gunderson, J., & Mulder, R. (2015). Treatment of personality disorder. Lancet, 385, 735–743.
Bellino, S., Paradiso, E., Bozzatello, P., & Bogetto, F. (2010). Efficacy and tolerability of duloxetine in the treatment of patients with borderline personality disorder: A pilot study. Journal of Psychopharmacology, 24, 333–339.
Bernstein, D. P., Iscan, C., & Masner, J. (2007). Opinions of personality disorder experts regarding the DSM-IV personality disorders classifications system. Journal of Personality Disorders, 21(5), 536–551.
Black, D. W., Zanarini, M. C., Romine, A., Shaw, M., Allen, J., & Schulz, S. C. (2014). Comparison of low and moderate dosages of extended-release quetiapine in borderline personality disorder: A randomized, double-blind, placebo controlled trial. American Journal of Psychiatry, 171(11), 1174–1182.
Bullock, S. A., & Potenza, M. N. (2012). Pathological gambling: Neuropsychopharmacology and treatment. Current Psychopharmacology, 1(1), 67–85.
de Brito, A. M., de Almeida Pinto, M. G., Bronstein, G., Carneiro, E., Faertes, D., Fukugawa, V., … Tavares, H. (2017). Topiramate combined with cognitive restructuring for the treatment of gambling disorder: A two-center, randomized, double-blind clinical trial. Journal of Gambling Studies, 33(1), 249–263.
Grant, J. E., Odlaug, B. L., Chamberlain, S. R., Potenza, M. N., Schreiber, L. R., Donahue, C. B., & Kim, S. W. (2014). A randomized, placebo-controlled trial of N-acetylcysteine plus imaginal desensitization for nicotine-dependent pathological gamblers. Journal of Clinical Psychiatry, 75(1), 39–45.
Grant, J. E., Potenza, M. N., Kraus, S. W., & Petrakis, I. L. (2017). Naltrexone and disulfiram treatment response in veterans with alcohol dependence and co-occurring problem-gambling features. Journal of Clinical Psychiatry, 78(9), e1299–e1306.
Houghton, D. C., Capriotti, M. R., De Nadai, A. S., Compton, S. N., Twohig, M. P., Neal-Barnett, A. M., … Woods, D. W. (2015). Defining treatment response in trichotillomania: A signal detection analysis. Journal of Anxiety Disorder, 36, 44–51.
Insel, T., Cuthbert, B., Garvey, M., Heinssen, R., Pine, D.S., Quinn, K., … Wang, P. (2010). Research domain criteria (RDoC): Toward a new classification framework for research mental disorders. American Journal of Psychiatry, 167(7), 748–751.
Joyce, D. W., Kehagia, A. A., Tracy, D. K., Proctor, J., & Shergill, S. S. (2017). Realising stratified psychiatry using multidimensional signatures and trajectories. Journal of Translational Medicine, 15, 15. doi:10.1186/s12967-016-1116-1
Joyce, D. W., Tracy, D. K., & Shergill, S. S. (2017). Are we failing clinical trials? A case for strong aggregate outcomes. Psychological Medicine, 48(2), 177–186.
National Institute for Health and Care Excellence, UK [NICE]. (2015). Borderline Personality Disorder: Recognition and Management. www.nice.org.uk/guidance/cg78/resources/borderline-personality-dosrder-recognition-and-management-pdf-975635141317
Newcomer, J. W. (2007). Antipsychotic medications: Metabolic and cardiovascular risk. Journal of Clinical Psychiatry, 68(4), 8–13.
Ornoy, A. (2009). Valproic acid in pregnancy: How much are we endangering the embryo and fetus? Reproductive Toxicology, 28(1), 1–10.
Rohde, C., Polcwiartek, C., Correll, C. U., & Nielsen, J. (2017). Real-world effectiveness of clozapine for borderline personality disorder: Results from a 2-year mirror-image study. Journal of Personality Disorders, 31, 1–15.
Ten Have, M., Verheul, R., Kaasenbrood A., van Dorsselaer S., Tuithof, M., & Kleinjan, M. (2016). Prevalence rates of borderline personality disorder symptoms: A study based on the Netherlands Mental Health Survey and Incidence Study–2. BMC Psychiatry, 16, 249.
Trull, T. J., Tragesser, S. L., Solhan, M., & Schwartz-Mette, R. (2007). Dimensional models of personality disorder: Diagnostic and Statistical Manual of Mental Disorders Fifth Edition and beyond. Current Opinion in Psychiatry, 20(1), 52–56.
Walker, M., Toneatto, T., Potenza, M. N., Petry, N., Ladouceur, R., Hodgins, D. C., … Blaszczynski, A. (2006). A framework for reporting outcomes in problem gambling treatment research: The Banff, Alberta Consensus. Addiction, 101(4), 504–511.
Widiger, T. A., & Trull, T. J. (2007) Plate tectonics in the classification of personality disorder: Shifting to a dimensional model. American Psychologist, 62(2), 71–83.
Wright, A. G. C., Hallquist, M. N., Morse, J. Q., Scott, L. N., Stepp, S. D., Nolf, K. A., & Pilkonis, P. A. (2013). Clarifying interpersonal heterogeneity in borderline personality disorder using latent mixture modeling. Journal of Personality Disorder, 27(2), 125–143.
Zanarini, M. C., Schulz, S. C., Detke, H. C., Tanaka, Y., Zhao, F., Lin, D., … Corya, S. (2011). A dose comparison of olanzapine for the treatment of borderline personality disorder: A 12-week randomized, doubleblind, placebo-controlled study. Journal of Clinical Psychiatry, 72, 1353–1362.
Zanarini, M. C., Vujanovic, A. A., Parachini, E. A., Boulanger, J. L., Frankernburg, F. R., & Hennen, J. (2003). Zanarini Rating Scale for Borderline Personality Disorder (ZAN-BPD): A continuous measure of DSM-IV borderline psychopathology. Journal of Personality Disorders, 17(3), 233–242.
Zanarini, M. C., Weingeroff, J. L., Frankenburg, F. R., & Fitzmaurice, G. M. (2015). Development of the self-report version of the Zanarini Rating Scale for Borderline Personality Disorder. Personality and Mental Health, 9(4), 243–249.
Zimmerman, M., & Mattia, J.I. (1999). Axis I diagnostic comorbidity and borderline personality disorder. Comprehensive Psychiatry, 40(4), 245–252.