5
Looking Ahead
This final chapter highlights overall messages and directions for potential future work on traumatic brain injury (TBI) biomarkers reported by workshop participants, as well as ongoing challenges to integrating biomarkers into clinical practice.
WHERE TBI BIOMARKERS CAN CURRENTLY HAVE IMPACT
Leslie Prichep, BrainScope Company, introduced the final discussion session. With many potential directions to pursue and limited resources, she said, identifying areas with the greatest potential for return on investment and added value to patients and families can help chart a course for the future. Referring to Figure 1-1 (see Chapter 1), she also reflected that although the dimensions of type, time after injury, and purpose of biomarkers are listed separately, in reality these areas overlap. Researchers and clinicians in TBI cannot look at any one area without considering them in the context of the other two. This fact reinforces the potential advances that will arise from combining information from multiple biomarkers and from integrating biomarker data with other clinical information to more optimally inform decision-making.
Based on the presentations and discussions, Prichep conducted an informal poll of workshop participants (in person and online) on which time points, contributions to care and research, and considerations for integration can have the most evidence for immediate impact on the TBI field. She provided three questions for the audience to consider:
- Rank the time points at which biomarkers can currently play a significant role in TBI care and research: acute (≤ 72 hours); post-acute (> 72 hours and < 6 months); long term (> 6 months); and research.
- Rank the contributions that could currently be made by biomarkers to TBI care and research: diagnosis/classification; optimizing/targeting treatment; monitoring recovery; and research to fill gaps.
- What are the main considerations in integration of biomarkers into the current workflow? Food and Drug Administration (FDA) clearance/approval, ease of use, actionable result, or more research needed?
Responses from in-person and remote attendees were collected using Survey Gizmo. After a short pause, during which approximately 80 responses were received, Prichep briefly reviewed the results. For question 1, respondents ranked the time points during which TBI biomarkers could currently have the greatest effect as acute, research, post-acute, and long term. Most of the biomarker data presented during the workshop addressed the acute period, she noted, and it is generally agreed that there is currently the least amount of evidence for the performance of TBI biomarkers over the longer term. This was reflected in the responses with the highest ranking for the acute period and the lowest ranking for the long-term period. The understanding and analysis of biomarker performance over the longer term represents an ongoing direction for the field.
Similarly, for the second question, respondents ranked the contributions that TBI biomarkers could currently play as being greatest for diagnosis/classification, optimizing/targeting treatment, research to fill gaps, and monitoring recovery.
The third informal poll question asked about key considerations for integrating biomarkers into clinical workflows. The relative rankings in response to this question were close to one another, Prichep said, with actionable results and ease of use being the highest ranked, and FDA clearance and more research needed ranked a bit lower.
POTENTIAL NEXT STEPS AND OPPORTUNITIES IDENTIFIED BY PARTICIPANTS
The briefing poll exercise captured a snapshot of participants’ views about the state of development of TBI biomarkers and their roles in TBI care and research. During the final sessions of the workshop, moderated by Prichep, participants expanded on a number of key points about the state of the science, roles that biomarkers currently play in the TBI landscape, their potential futures uses, and ideas to move the field forward. Speaker
and participant comments touched on future directions such as progress in techniques for analyzing information provided by multiple types of TBI biomarkers, incorporating feedback from expanded clinical use of biomarkers, evaluating the utility of biomarkers in expanded care settings and in additional types of patient populations, and addressing implementation issues, including incorporation into clinical practice guidance, clinician and patient buy-in, and effects on workflow. A number of participants also noted the importance of enhanced partnerships among researchers, clinicians, industry developers, government agencies, health information systems, and insurance providers. Comments made are summarized below.
Biomarkers Provide Added Information
TBI care is already informed by biomarkers, including those obtained through neuroimaging, while FDA-cleared tests for blood and electrophysiological biomarkers are beginning to reach clinical practice. Several participants noted that it is becoming increasingly clear that multiple TBI biomarkers are at the cusp of additional clinical applications for diagnosis and classification, prognostication, and monitoring of recovery. TBI is a complex multisystem phenomenon, a participant emphasized. There will likely never be a single biomarker that captures all the important dysfunction that occurs after injury; however, a multidimensional profile of TBI is emerging and may be very useful, not only for diagnosis and classification, but ultimately in informing treatment. The profiles of these TBI biomarkers have potential to help researchers understand laboratory and clinical TBI endophenotypes, and to aid in the management of TBI throughout the continuum of care and across the spectrum of injury. However, the lessons from incorporating troponin into cardiac care demonstrate the importance of conducting biological variation studies and understanding reference ranges, standardizing assays, and establishing laboratory turnaround time needs and expectations as part of the translation and implementation of biomarkers into practice.
There are a lot of opportunities in the biomarker tool kit, another participant commented, and there is a need for different types of tools. It will be important for clinicians and researchers to identify the problem that needs to be solved, assess the capabilities of the available biomarker tools, and identify the right ones to use, based on the indication and context of use. In addition to their applicability for clinical diagnosis and the continuum of care after a TBI, biomarkers such as blood-based glial fibrillary acidic protein (GFAP) can serve as potential pharmacodynamic response biomarkers for preclinical drug screening. Such efforts may aid in the enhanced assessment and development of TBI therapeutics.
A number of participants emphasized the value of incorporating the spectrum of biomarker measures as well as the potential arising from combining different types of biomarker measures in multimodal analyses. For example, a participant noted that combining biomarkers based on electroencephalograms (EEGs) and blood biomarkers would provide two perspectives that have different timelines and that characterize different aspects of brain function and injury. Challenges and strategies raised during the workshop included the further development of analytic pipelines and standards, and the importance of further developing analytic techniques to integrate such diverse, multimodal information, as well as obtaining FDA clearance.
Levels of Evidence and FDA Clearance
Several participants emphasized the consideration of which biomarkers are ready for use now, in which settings and contexts, including which biomarkers can be measured using FDA-cleared assays and devices. As the informal poll reflected, more evidence is currently available on the use of biomarkers in acute TBI care. But others noted potential limitations in the generalizability of studies and that more research is likely needed in certain populations (including in children and in older adults), as well as more fully understanding baseline biomarker values and the role of intra- and interindividual variation. For example, a participant highlighted how the current state of care for geriatric TBI reflects a crisis of underrepresentation in research. Without studies that represent this population, she said, older adults will continue to lack evidence-based TBI treatment and management guidelines. Another participant agreed with the need to study biomarker profiles in the full range of patients who experience TBI, noting that it will be crucial to ensure gender and racial diversity in the ongoing research and care agenda.
The continued collection and analysis of biomarker profiles over time and how these profiles inform TBI prognostication over the longer term is an area that needs to be further explored and represents a potential future evolution for the field. Several participants also reflected on issues of clarity and transparency when discussing TBI biomarker assays and results. Using the example of GFAP, discussions during the workshop raised the issue that not all assays measure the same GFAP structure (e.g., the same isoform or breakdown product), in the same ways or at the same antibody concentration levels, or use the same cutoff values. To compare results and understand next steps, it is important to know the specific details of the targeted protein and assay parameters being used.
One participant commented that when using and interpreting biomarker tests, clinicians will want systems that are easy to use but will likely also want access to the data, not only to automated interpretations that
an assay or device generates (e.g., that a test value is in the normal range, elevated, etc. based on specified cutoff ranges). Providers need to be able to use judgment in integrating the biomarker into the full assessment, the participant argued. Using the example of the BrainScope Concussion Index, a participant commented that algorithms were initially able to provide only categorical binary or ternary results. However, she continued, the score and the threshold for this measure are now provided, enabling a provider to review and decide the next care steps and to evaluate change over time. Feedback has shown that clinicians often use their judgment on how close the result score is to the threshold to inform their decision, and this offers more nuance than a binary yes or no outcome. Looking ahead, several participants agreed that feedback from the clinical use of biomarkers will be important in helping to create and refine the next generation of measures.
Adoption and Implementation of TBI Biomarkers into Clinical Practice
A number of participants noted issues associated with the effective incorporation of biomarkers into clinical workflows, including economic and workforce considerations. Such implementation elements cannot be overlooked in the translation and adoption of biomarkers into clinical practice. For example, a participant noted that incorporating biomarkers into practice in emergency department settings will require addressing the speed of results and effect on emergency department throughput, cost-effectiveness, and patient acceptance, as well as how results can be incorporated into patient electronic medical records. Widespread implementation of TBI biomarkers in emergency settings also has the potential to greatly increase the number of patients diagnosed with mild TBI who need post-acute care, a participant emphasized. While the ability to better diagnose these patients is important, participants cautioned that health care systems will need to work now to develop plans to meet the anticipated increase in patients seeking post-acute TBI care.
In the setting of longer-term TBI, a participant asked about combating challenges with insurance reimbursement and the ability to provide care to patients who depend on rehabilitation interventions to change their long-term recovery trajectories. To address this, a participant suggested that clinicians will need to be armed with objective evidence and informational aids that make connections among biomarker results, rehabilitation interventions, and improved outcomes to be accepted by those who make such cost and approval determinations.
In addition to documenting anticipated effects and returns on investment from incorporating further types of TBI biomarkers in various clinical settings, other participants highlighted the important role of obtaining clinician buy-in to use new biomarker tests. Establishing clinical practice
guidelines and policies covering the use of TBI biomarker tests will be important aspects in this adoption and buy-in process.
TBI Classification
As described in prior chapters, several participants noted that current TBI classification into the categories of mild, moderate, and severe does a disservice to many patients. During the course of the workshop, a number of participants emphasized the opportunity for the TBI field to incorporate additional biomarker information in ongoing efforts to develop more precise classification systems for TBI.
Considering Future Questions and Directions
The workshop highlighted a variety of areas in which further study is needed—including further research in various patient populations, for different mechanisms and severities of injury, in different clinical settings and contexts of use, and in further biomarker discovery, validation, and multimodal analyses.
For patients and families experiencing TBI, a participant reiterated that much of the lived experience of TBI takes place outside of a hospital or clinic, often in the home, workplace, or community. While biomarkers may be measured and incorporated into formal health records, other types of personnel, such as athletic trainers, emergency medical providers, and caregivers should have access to tools and information as well. Building on the discussions of the potential use of biomarkers in prehospital settings, several participants commented that research and development of TBI biomarkers in prehospital settings is an important area for further study, noting also that this area could draw on the ability of EMS to take blood samples or record an EEG in the minutes following an injury.
Over the longer term, the prognostic value of biomarker information to inform loved ones caring for TBI patients is the next frontier, a participant emphasized. The ability to better provide patients and families with information about what can be done to support them in optimal recovery would be transformative. The question was raised whether biomarkers can help establish a benchmark for the progress of TBI recovery, while not losing the importance of the human factor when assessing patients. As an example, Geoffrey Manley, University of California, San Francisco, referred to a South American study that found the same outcomes from monitoring intracranial pressure and from physicians at the bedside conducting frequent neurological exams and imaging.1 This result reinforces
___________________
1 See Chesnut et al., 2012.
the importance of being a clinician and speaking with patients and families, first and foremost, he said. After the acute phase, an important role of TBI care is to have hands on the patient, reassure them, and educate them about their history and anticipated recovery trajectory. The continued longitudinal study of biomarker information to better understand injury progression, risk of late complications, and predictive value in TBI remains important in this consultation process.
With such a heterogeneous population of TBI patients, adequately studying the full course of injury and recovery and measuring the right markers effectively remain important areas of development. During the course of the meeting, a number of participants noted patient populations in which biomarker performance after TBI injury and during recovery needs to be better understood, including geriatric TBI patients, people who have experienced multiple head injuries, and those with coexisting medical conditions, including seizures, epilepsy, or dementia. The ability to account for the potential confounding effect on biomarker results from situations such as preexisting head injury and coexisting medical conditions is important.
Several participants commented on the intersection of biomarker assays with the data infrastructure world. One participant suggested the need to further build out data use agreements with academic and private partners and the need for additional efforts to design information systems to gather and synthesize relevant TBI biomarker information. Efforts that collate and synthesize the available information could also make this information more accessible to providers outside of major trauma centers. A patient should not have to go to a top hospital’s intensive care unit to survive a TBI, the participant noted.
Finally, a number of additional research directions arose during the discussions. Future areas noted by participants included the continued expansion of biomarker discovery platforms within TBI consortia, as well as studying and validating potential “good” biomarkers, whose levels increase in positive states of recovery, compared to what most of the discussion covered, which are “bad” biomarkers whose levels are elevated after injury. One participant highlighted the use of non-hypothesis-based methods as a promising avenue to identify which fluid-based biomarkers are most informative for TBI diagnosis and prognosis and to determine pathological mechanisms to inform future therapeutics. In the field of electrophysiological biomarkers, tools such as EEG are well established, give immediate results, and are sensitive to connectivity between brain regions and networks, another participant commented, although they are sensitive to factors that are difficult to control. Suggestions for future development included advances in analytical technologies for furthering the understanding of the flow of information between regions of the brain, the use of techniques such as network analysis, and the continued incorpora-
tion of measures that complement EEG, such as heart rate variability, eye tracking, and clinical history.
WORKSHOP WRAP-UP
Presentations and discussions over the course of the 1-day workshop brought together clinicians and TBI care providers, patient and family advocacy organizations, agency policy makers, industry partners, and others. Corinne Peek-Asa, University of California, San Diego, closed out the workshop, thanking speakers and participants and noting that the sessions highlighted multiple ways in which biomarkers can play expanding roles in the assessment, care, and monitoring of traumatic brain injury in preclinical TBI research, and in informing clinical trials for new treatments and interventions. This is an active area for the TBI field, although a number of challenges will need to be overcome to collectively push the frontiers of the field forward and fully realize the promise of biomarkers in improving care for TBI patients and families.
