The Promise of Precision Medicine for Motor Recovery after Stroke

By: Emma Savelon

“Will everyone soon receive personal medication advice?” is the article headlining the news¹. This article published in 2023 highlights a personalized approach to medicine and states that using a patient’s DNA profile for tailor-made medication leads to a 30% reduction in side effects. Tailoring treatment to a subset of patients, termed precision medicine, revolutionized treatment strategies since first being introduced in 1999, but why is this relevant for stroke recovery today?

Stroke recovery

Stroke causes damage to brain tissue due to a disruption of blood supply, this damaged area is referred to as a lesion. While recent advancements in treatments have increased post-stroke survival, the prevalence of stroke-related disability has also increased. Of the patients suffering from motor deficits in the early days after stroke, 65% report the persistence of motor impairment 6 months post-stroke, severely impacting their quality of life². Following initial motor impairment after a stroke, some patients spontaneously recover their motor function over time (Fig. 1, green line) while others may only recover some (Fig 1. blue line) or almost none of their motor function (Fig. 1 red line), even after treatment. The complex and heterogeneous nature of stroke makes a “one-suits-all” treatment approach unsuitable. Moreover, stroke is regarded as a network disease; areas located far away from the lesion can still be damaged due to their functional or structural connection with the lesioned area. Since a stroke manifests differently among patients, different areas of the brain are affected, leading to different motor function outcomes. This high inter-patient variability and the significant number of non-responders to treatment have pushed the stroke recovery field toward precision medicine. 

Precision medicine

Precision medicine can be seen as tailoring a suit. Just as a tailor takes individual measurements from an individual to ensure a good fit, precision medicine incorporates the patient’s genetic, environmental, and lifestyle factors to develop targeted treatments. Precision medicine aims to develop targeted interventions based on a patient’s characteristics, improve treatment efficacy, increase the number of responders, and reduce side effects. While precision medicine is well established in the field of oncology, where the molecular genetics of tumors can shape treatment protocols for patients³, other fields have not been able to reproduce the same success. A disease like stroke has complex genetics and stroke-specific variables such as lesion location and size. These factors contribute to risk and outcome, making it hard to personalize treatment based on genotyping alone. However, the introduction of large-scale clinical data allows us to gather large amounts of information about the stroke’s impact on the body (lesion site and location) and behavior (motor impairment, attention deficits and others). This has led to new developments in precision medicine for stroke recovery⁴. Just like a suit that fits better when a tailor has more measurements, additional clinical information can provide more targeted treatment.

Biomarkers for stroke recovery: predicting motor outcomes

As mentioned before, it is crucial to have patient-related information to be able to provide targeted treatment. This is where biomarkers come in. Biomarkers are objective measures of physiological processes happening in the body and can be used for either prognostic or predictive purposes. In precision medicine, biomarkers assess inter-patient variability and stratify patients into groups for targeted therapy and rehabilitation programs⁵. Imaging techniques can be used to assess the integrity of motor pathways, which are responsible for transmitting signals from the brain to the muscles. The integrity of these pathways acts as a reliable biomarker to predict the extent of motor recovery. The less damaged the pathway, the more likely the patient will make a full recovery. However, no single biomarker can accurately predict motor outcomes. A combination of stroke-specific biomarkers and clinical outcomes of initial impairment (as depicted in Fig 2.) gives us the most reliable predicted motor outcome. 

The Predict Recovery Potential 2 (PREP2) algorithm is an example of a prediction tool used in the clinic⁶. Using biomarkers and clinical measures taken within the first 72 hours from stroke symptom onset, a patient’s functional outcome of the affected limb at 3 months post-stroke can be predicted. Knowing whether a patient will regain full or limited limb function allows for realistic patient goal-setting and tailored rehabilitation protocols. For instance, patients who are predicted to achieve excellent recovery will benefit from physiotherapy that focuses on recovering normal use of their affected limb. Meanwhile, patients who are predicted to not regain useful movement will instead undergo physiotherapy that focuses on reducing disability, such as using the non-affected limb for daily activities and preventing complications like spasticity and pain.

The challenges of precision medicine

Before precision medicine can be implemented in healthcare for routine use, certain elements must be taken into consideration. Due to the high variability amongst patients, extensive clinical data is required to tailor rehabilitation protocols to the patient’s needs. Currently, there is a lack of infrastructure and technology to accommodate and analyze the vast amount of data that precision medicine brings in. Artificial intelligence and machine learning-based ‘big data’ platforms have been developed but are in need of further advancement before being implemented on a larger scale⁷. Concerns around privacy have been raised, with the question if patient anonymity can be preserved. Strict policies will have to be implemented to ensure the protection of patient data. Additionally, identifying effective population-based biomarkers will require the inclusivity of underrepresented populations in genome-wide association studies (GWAS), in particular the African population. This was highlighted by a study that showed that this population was completely absent in 31 stroke-related GWAS identifying potential stroke subtype biomarkers⁸. Just as different fabrics require different sewing techniques, different populations have different individual characteristics that require different interventions, highlighting the need for precision medicine.  Despite these challenges, the concept of implementing precision medicine for stroke recovery in the clinic in the future is hopeful, with prediction models already being used worldwide.   

Motor recovery: future perspectives

So, what could precision medicine for stroke recovery look like in the future aside from prediction models? In recent years, non-invasive brain stimulation (NIBS) techniques have become a staple of stroke rehabilitation research for diagnostic and therapeutic purposes. NIBS modulates the excitability of the brain to promote motor recovery as well as provide information on the integrity of motor pathways⁹. Recently, NIBS research has shifted focus to personalized rehabilitation protocols to address the high heterogeneity and number of non-responders following treatment. Rehabilitation protocols are tailored to individuals’ specific characteristics such as age, level of impairment and location of stroke to address the wide variability in treatment response. Aside from NIBS, novel technologies have opened new opportunities in the field of stroke rehabilitation, with robot and virtual reality-assisted therapy being the latest developments for patient-specific treatment¹⁰. Robot-assisted therapy often makes use of a brain-computer interface, a system that can read brain signals and translate them into executable commands for a robotic arm or glove to support motor function¹¹. Despite promising advancements being made, more research is needed to improve the validity and efficacy of precision medicine before thinking of ways precision medicine can be implemented for individualized rehabilitation protocols in the clinic. 

Conclusion

Precision medicine in stroke recovery is a promising field that aims to improve efficacy and increase the number of responders to treatment. By having large datasets to find relevant biomarkers and stratify patients into the most suited treatment group, precision medicine has started to become implemented in the clinic, directly impacting a patient’s outcome. However, the challenges regarding the implementation of precision medicine for routine use must be addressed. The recent advances in technology keep the field moving forward and allow for the validation of more accurate and predictive biomarkers, giving more individualized rehabilitation protocols, or a more well-fitting suit.

About the author

Further reading

1. Van Straten, B. (2023, 3^rd, February). Krijgt iedereen binnenkort een persoonlijk medicatie advies?. NOS, 1-1. https://nos.nl/artikel/2462380-krijgt-iedereen-binnenkort-een-persoonlijk-medicatie-advies
   
   ↩︎
2. Dobkin, B. H. (2005). Rehabilitation after Stroke. New England Journal of Medicine, 352(16), 1677-1684. https://doi.org/10.1056/NEJMcp043511
   
   ↩︎
3.  Rostanski SK, Marshall RS. Precision Medicine for Ischemic Stroke. JAMA Neurol. 2016;73(7):773–774. https://doi.org/10.1001/jamaneurol.2016.0087
   
   ↩︎
4. Olaiya, M. T., Sodhi-Berry, N., Dalli, L. L., Bam, K., Thrift, A. G., Katzenellenbogen, J. M., Nedkoff, L., Kim, J., & Kilkenny, M. F. (2022). The Allure of Big Data to Improve Stroke Outcomes: Review of Current Literature. Current neurology and neuroscience reports, 22(3), 151–160. https://doi.org/10.1007/s11910-022-01180-z
   
   ↩︎
5.  Simpkins, A. N., Janowski, M., Oz, H. S., Roberts, J., Bix, G., Doré, S., & Stowe, A. M. (2020). Biomarker Application for Precision Medicine in Stroke. Translational stroke research, 11(4), 615–627. https://doi.org/10.1007/s12975-019-00762-3
   
   ↩︎
6. Stinear, C. M., Byblow, W. D., Ackerley, S. J., Smith, M. C., Borges, V. M., & Barber, P. A. (2017). PREP2: A biomarker-based algorithm for predicting upper limb function after stroke. Annals of clinical and translational neurology, 4(11), 811–820. https://doi.org/10.1002/acn3.488
   
   ↩︎
7. Blatter, T. U., Witte, H., Nakas, C. T., & Leichtle, A. B. (2022). Big Data in Laboratory Medicine-FAIR Quality for AI?. Diagnostics (Basel, Switzerland), 12(8), 1923. https://doi.org/10.3390/diagnostics12081923
   
   ↩︎
8. Olaiya, M. T., Sodhi-Berry, N., Dalli, L. L., Bam, K., Thrift, A. G., Katzenellenbogen, J. M., Nedkoff, L., Kim, J., & Kilkenny, M. F. (2022). The Allure of Big Data to Improve Stroke Outcomes: Review of Current Literature. Current neurology and neuroscience reports, 22(3), 151–160. https://doi.org/10.1007/s11910-022-01180-z
   
   ↩︎
9. Nicolo P, Ptak R, Guggisberg AG. Variability of behavioural responses to transcranial magnetic stimulation: Origins and predictors. Neuropsychologia. (2015) 74:137–44. 10.1016/j.neuropsychologia.2015.01.033
   
   ↩︎
10. Xiong, F., Liao, X., Xiao, J., Bai, X., Huang, J., Zhang, B., Li, F., & Li, P. (2022). Emerging Limb Rehabilitation Therapy After Post-stroke Motor Recovery. Frontiers in aging neuroscience, 14, 863379. https://doi.org/10.3389/fnagi.2022.863379
    
    ↩︎
11. Colamarino, E., Pichiorri, F., Toppi, J., Mattia, D., and Cincotti, F. (2022). Automatic Selection of Control Features for Electroencephalography-Based Brain-Computer Interface Assisted Motor Rehabilitation: The GUIDER Algorithm. Brain Topogr. 35, 182–190. doi: 10.1007/s10548-021-00883-9
    
    ↩︎