The bipolar maze: a roadmap through translational psychopathology

This exciting issue of Acta Psychiatrica Scandinavica focusses on bipolar disorder from different perspectives, including epidemiology on age at onset 1, 2 and comorbidity 3, biomarkers 4, clinical trial methodology 5, treatment 6, and medication adherence 7, thus covering some of the hottest topics in current clinical research on this condition. Indeed, bipolar disorder research has blossomed over the past two decades 8, 9, and bipolar disorder units and programs are more widespread than ever 10. This condition has been now unprecedentedly renamed as 'the heartland of Psychiatry' 11, 12. However, there is still a huge disparity between the burden of this particular disease and the (little) amount of funding devoted to research in this area of biomedical research 13. Hence, because of its psychopathologic complexity and limited investment, the translational road from bench to bed and vice versa can be defined as a maze. We are at the point of starting to work on a reasonable roadmap to walk it through. The first challenge in the maze is the phenotype. DSM-5 has upgraded bipolar disorders at the same level as psychotic disorders and depressive disorders, which may seem reasonable. However, the DSM system seems to lead nowhere over the foreseeing future. There may be small incremental improvements, such as the dimensional assessment of mixed states 14, 15, but the questionable validity of the DSM constructs, along with the limited improvement as regards to their reliability 16, suggests that a shift in paradigm is necessary. This fact prompted a debate in United States between the American Psychiatric Association and the National Institute of Mental Health (NIMH) on the validity of DSM-5 and what were the alternatives, and the NIMH announced their proposal to fund research based on research domains criteria (RDoC), rather than DSM-5. The RDoC initiative can be seen as an attempt to build a neuroscience-based classification of mental disorders 17. The positive side of this project is the will to anchor the new classification to well established circuits and biological mechanisms: the downside, the actual risk of ending nowhere, at least from the clinical use perspective. There is no question that some of the findings that RDoC-related research may bring up over the coming years will enrich our current knowledge on what we call 'bipolar disorder' and beyond. But it is hard to foresee an alternative to the traditional phenotype of manic-depressive illness that clinicians may be able to recognize and treat. It is much more likely, thus, that future translational research looking into concepts such as 'positive valence' or 'systems for social processes' may assist in better defining the bipolar phenotype allowing for a real 'personalized or stratified medicine' 18, rather than replacing it. The ROAMER project (Roadmap for Mental Health Research in Europe) may provide guidance on how to implement the foundation for an alternative to the US radical shift in mental health research priorities represented by RDoC, at the European level. The World Health Organization may want to learn from RDoC and ROAMER to make their own proposal of diagnostic research criteria (which should not necessarily be the same as those used for clinical practice) for the International Classification of Diseases, 11th edition. After decades of research on the heritability of bipolar illness, still the best proof of it are the epidemiological studies (twin, adoption, and population studies), but we certainly made some progress in molecular genetics. The second challenge of the bipolar maze is the genotype. To date, very little of the heritability of bipolar disorder is explained by common polymorphisms 19. Two genes appear to be the strongest candidates from genome-wide association studies, the ANK3, a gene involved in the function of voltage gated sodium channels), and the CACNA1C, a gene encoding a protein that is part of a voltage dependent calcium channel 20, 21. Many other candidate genes have been identified as correlates of certain traits and behaviors, such as impulsivity or suicidality 22, 23, giving further support to the need of a new nosological model based on translational or molecular psychopathology. Hopefully the next generation of genetic studies using whole exome or whole genome screening may shed more light into the genetics of the syndrome and particularly into the molecular genetics of its psychopathological components. The studies that have looked into the structural and functional neuroimaging of bipolar disorder have also increased our understanding of the condition. Again, it would be unfair to say that we have not made substantial progress in our understanding of the brain structures involved in the bipolar phenotype, but it is also true that we lack a compelling model connecting genes, cells, circuits, and structures. Indeed, the third challenge of the bipolar maze is the functional neuroanatomy of the bipolar brain. Brain structural magnetic resonance imaging (MRI) studies suggest cortical and subcortical abnormalities within networks subserving emotional regulation. There is evidence of neuroprogressive loss of gray matter volume in prefrontal and anterior cingulate cortex and the subgenual region 24 and further abnormalities in subcortical structures, including the hippocampus, the thalamus, and the caudate nucleus 25. Moreover, adolescents at risk of bipolar disorder may have some abnormal amygdala neurodevelopment 26. Unfortunately, none of these findings is specific enough to assist in the diagnostic process. Functional neuroimaging techniques have revealed consistent findings with the structural neuroimaging studies. Thus, abnormal amygdala activation during face processing appears to be more pervasive in children and adolescents than in adults with bipolar disorder 27, and neurobiological changes in activation patterns involving fronto-limbic circuitry have been reported in relation to different illness phase and mood states. The difficulties to deactivate the default mode network is another consistent finding across mood states 28, 29, but might be very unspecific. Resting state studies have reported changes in the connectivity of the medial prefrontal cortex and the anterior cingulated cortex with limbic-striatal structures, and in spatial extent in regional homogeneity when studying the default mode network 30. Neurocognitive dysfunctions may not be necessarily present before illness onset 31, 32, but they have been extensively described in first-episode patients 33 and may persist beyond acute episodes 34, 35. Cognitive impairment may be considered a measure of allostatic load 36 and may correlate with some neuroinflammatory and oxidative stress markers 37, 38. Neurocognitive impairment may reflect a poor course and outcome of the disease, especially concerning the number of manic episodes 39, 40, a history of psychosis 41, subthreshold symptoms 42, and comorbidity 43. Although all studies control for the effects of medication, there are very few studies with medication-free patients 44, and the influence of drugs on cognitive performance may be bimodal, with both positive and negative effects 45, 46. The relevance of assessing cognition in patients with bipolar disorder is its high correlation with functional outcome 47-51. This model, which was initially tested in schizophrenia and subsequently in bipolar disorder, is now being analyzed in patients with unipolar depression 52. As cognition is a crucial mediator of functional outcome, cognitive remediation techniques, and particularly functional remediation, have been successfully applied to patients with bipolar disorder 53. Understanding the underlying mechanisms of cognitive deficits, their role as an endophenotype, and their impact in psychosocial outcome is the fourth challenge in the bipolar maze. It is customary to say that the progress in the treatment of mental disorders has been slow and limited over the past 30 years and that nothing has matched the success of chlorpromazine, imipramine, or lithium. However, the armamentarium for the treatment of bipolar disorder has increased in quantity and quality, and patients are treated at a much earlier stage, resulting in the practical disappearance of catatonic and severely bizarre psychotic manic and depressive syndromes in the Western world. We have learned to use the available therapies as monotherapies or, most commonly, in combination 54, 55. Guidelines have been developed to reduce the variability of clinical practice 56, 57, and clinicians have become aware of the importance of medication adherence and psychoeducation 58-60. We learned to measure the relative efficacy of drugs for bipolar disorder by means of the number needed to treat 61 and their profile according to the polarity index 62, which is a useful measure that can be easily misunderstood 63 because it provides advice on what not to use, rather than what to use, depending on the predominant polarity of manic and depressive episodes in a given patient 64. The concept is now being used to choose among psychosocial interventions as well 65. Over the past decade, we learnt to better use the second generation of antipsychotics and a good bunch of evidence has been delivered on the use of antipsychotics in combination with lithium or valproate and the combination of the latter two in the long-term treatment of bipolar disorder 66. Unfortunately, on the negative side, the use of lithium is progressively declining despite the strong evidence base in favor of its use 67, and we still lack good data to support the use of antidepressants in this condition 68. The fifth challenge of the maze is to overcome the current drug development paradigm and enter a new era in neuropsychopharmacology, which is easy to say, difficult to do. It is now quite obvious that the drug pipelines for psychiatric disorders in general and bipolar disorder in particular are drying up 69. This is one of the reasons for the NIMH shift toward a neuroscience-based classification of mental disorders. We need methodological upgrades 70, technological sophistication, and much better outcomes. We may also need more incentives to CNS research. Only a few, if any, of the forthcoming new drugs may actually represent a significant advance in the treatment of manic-depressive illness; there is some chance with NMDA receptor antagonists and partial agonists for bipolar and unipolar treatment-resistant depression, and little more. Further effort is needed to explore the full potential of brain stimulation techniques and newer psychotherapies. Staging models should be developed to adjust the benefit risk of any intervention to the particular morbidity of every patient, moving toward personalized psychiatry. Clearly, some treatments are more suited for early intervention than others 71, and medicalization is only necessary in patients in whom the diagnosis is solidly confirmed. The bipolar maze is full of challenges and dead ends. Current research is providing slow but incremental progress in our knowledge of the pathophysiology of manic-depressive illness. But if we want to make significant steps forward, a paradigm shift is necessary in defining the phenotype and in drug development. This is the envisaged roadmap to the bipolar maze. Needless to say, the maze has no exit, the real success is to enjoy the journey. Dr. Vieta has received grants and served as consultant, advisor or CME speaker for the following entities: Adamed, Alexza, Almirall, AstraZeneca, Bial, Bristol-Myers Squibb, Elan, Eli Lilly, Ferrer, Forest Research Institute, Gedeon Richter, Glaxo-Smith-Kline, Janssen-Cilag, Jazz, Johnson & Johnson, Lundbeck, Merck, Novartis, Organon, Otsuka, Pfizer, Pierre-Fabre, Qualigen, Roche, Sanofi-Aventis, Servier, Shering-Plough, Shire, Solvay, Sunovion, Takeda, Teva, the Spanish Ministry of Science and Innovation (CIBERSAM), the Seventh European Framework Programme (ENBREC), the Stanley Medical Research Institute, United Biosource Corporation, and Wyeth.