Abstract
Dear Editor:
Neuropsychiatric problems are some of the most devastating consequences of traumatic brain injury (TBI) because of their impact on a person’s sense of identity, interactions with others, and engagement with their wider community. Yet, our understanding of how TBI leads to these problems and effective interventions is limited. One serious challenge facing researchers and clinicians working in this area is the categorical diagnostic approach to classification of post-TBI psychopathology that heavily relies on ICD or DSM criteria. These criteria assign diagnostic labels based on the presence of mutually exclusive fixed sets of symptoms. This means that patient and disease heterogeneity, comorbidity, and the full range of symptomology are poorly considered. Furthermore, the diagnostic labels do not provide a framework for investigating underlying mechanisms. It is particularly likely that in a heterogeneous condition, such as TBI, multiple disparate and overlapping neurological and psychological mechanisms may contribute to similar clinical presentations. Therefore, diagnosing a TBI survivor with “depression,” for example, identifies a syndrome without necessarily revealing anything about contributing mechanisms or a reasoned basis for treatment.
Carmichael et al. 1 propose the Hierarchical Taxonomy of Psychopathology following TBI (HiTOP-TBI), a novel dimensional classification of neuropsychiatric problems after TBI that builds on the transdiagnostic approach in the HiTOP framework. 2 In the HiTOP framework, a psychopathological presentation is considered to be composed of different components and traits, which are hierarchically grouped into scales, spectra, and syndromes based on their similarity, covariation, and co-occurrence. Applying the HiTOP assessment battery and approach to a large cohort of moderate-to-severe TBI survivors, the authors show that this cohort experienced significant problems in the internalizing and detachment spectra (comprising, for example, affective lability, cognitive problems, and social withdrawal), while having lower scores for thought disorder and antagonistic externalizing spectra (comprising, for example, fantasy proneness and grandiosity). The authors then develop a hierarchical dimensional model for TBI (the HiTOP-TBI model) comprising seven factors that cover the range of post-TBI psychopathology. This model of classification allows for heterogeneity of presentation and accounts for all presenting signs and symptoms. Crucially, this approach paves the way for rational investigation of mechanisms underlying post-TBI neuropsychiatric presentations.
It is likely that clinicians treating post-TBI psychopathology already apply a dimensional lens to patient management. For example, sleep disturbance, fatigue, or affective lability are the treatment targets, rather than “depression.” However, the field has yet to codify this approach to psychopathology assessment for research. A transdiagnostic approach is already being used to gain important insights into the development of post-trauma symptoms in the AURORA study, 3 and using HiTOP-TBI as a tool to facilitate this would bring many benefits. Current recruiting practices, based on DSM/ICD diagnosis, likely lead to a covertly heterogenous population, while using single assessment scales risks similar problems by ignoring comorbidity. In addition, Hi-TOP factors, being closer to neurobiological mechanisms, may also show value as outcome measures, though further external validation work needs to be carried out, for example, to examine how closely they relate to broader functional outcome and quality of life measures.
A body of work has already been done that enables the HiTOP to be used to describe individual patients, 4 including tools that enable “conversion” to ICD and DSM codes for insurance billing or clinical audit purposes. The same work needs to be done for HiTOP-TBI to enable its use in the clinic and research studies. In addition, this study should motivate the development of lab-based tasks that interrogate distinct HiTOP-TBI components to facilitate the study of the underlying neurobiological mechanisms. A key mechanistic question is how different HiTOP-TBI factors map onto different TBI biomarker profiles, brain lesion locations, white matter, and brain network disruption. One specific interesting question is how mechanisms underlying psychopathology after TBI overlap with mechanisms underlying cognitive deficits. For example, if salience network dysfunction 5 is key to post-TBI attentional and memory deficits, how might it contribute to psychopathology components given the importance of the salience network in emotional processing.
Another important future direction could include leveraging the HiTOP-TBI framework for exploratory clinical trials of candidate therapeutics. Instead of asking whether a therapeutic improves a specific DSM- or ICD-based diagnostic entity (e.g., “depression”) in the context of TBI, we could explore the effects on a hierarchical spectrum of TBI-related psychopathologies. While the HiTOP-TBI battery is relatively long (average 1.5 h to complete), the time required is not out of the range of other commonly used clinical trial assessment batteries, and it can be self-administered. Importantly, the HiTOP-TBI provides substantial dimension reduction, which in turn would improve the statistical power of such therapeutic trials by mitigating the challenges posed by issues of multiple comparisons relative to other exploratory approaches. Of course, the results of any such exploratory clinical trials would have to be replicated in more focused later-stage trials. Given the relatively poor success rate in the field, a new and potentially more efficient approach to matching up candidate therapeutics with domains of impairment is warranted. As a specific hypothetical example, we could envision asking the question of which portions of the TBI-related spectrum of psychopathologies are affected by stimulants like methylphenidate. There may be some useful surprises; anecdotally, symptoms of anhedonia, workplace performance anxiety, and apathy may be improved as much or more than attention deficit symptoms. Another example could involve an exploratory trial of a brain-penetrant Bruton’s tyrosine kinase inhibitor such as tolebrutinib. Many have hypothesized that autoantibodies to brain antigens may play a deleterious role in chronic post-TBI sequelae, but as yet there is not enough knowledge to design a focused clinical trial using a specific DSM- or ICD-based diagnostic entity as a primary outcome measure. The HiTOP-TBI approach could be used in an initial exploratory trial as a place to start. As correctly noted by Carmichael et al., a great deal of additional work remains to be performed in the field, but the potential for benefit is tremendous.