Using machine learning to predict treatment outcomes for neuropathic pain

Machine-Learning Based EEG Biomarkers for Personalized Interventions

This study is testing if a new computer program can help predict how well a brain treatment will work for people with nerve pain from spinal cord injuries.

Quick facts

• Study type: Observational
• Enrollment: 58 (estimated)
• Ages: 18 Years to 99 Years
• Sex: All
• Sponsor: Institut Guttmann (other)
• Locations: 1 site (Badalona, Barcelona)
• Trial ID: NCT06531317 on ClinicalTrials.gov

What this trial studies

This observational study aims to develop a machine learning model that predicts the effectiveness of transcranial direct current stimulation (tDCS) in treating neuropathic pain resulting from spinal cord injuries. Participants will undergo EEG recordings to capture brain activity and complete clinical assessments before and after the treatment. The study seeks to determine if EEG data combined with clinical evaluations can accurately forecast treatment success, thereby personalizing interventions for patients suffering from neuropathic pain.

Who should consider this trial

Good fit: Ideal candidates are adults over 18 years old with subacute neuropathic pain following a spinal cord injury, experiencing pain intensity of at least 4 out of 10.

Not a fit: Patients whose neuropathic pain is primarily related to spasms or those not meeting the specified pain intensity criteria may not benefit from this study.

Why it matters

Potential benefit: If successful, this approach could lead to more effective and personalized treatments for patients with neuropathic pain.

How similar studies have performed: While the use of machine learning in predicting treatment outcomes is an emerging field, similar studies have shown promise in other areas of neuromodulation and pain management.

Eligibility criteria

Inclusion Criteria:

* Age: Over 18 years old.
* Neuropathic Pain (NP): Subacute NP at or below the lesion level for at least 1 month following spinal cord injury or disease. Persistent NP is defined as pain in an area of sensory abnormality corresponding to the spinal cord injury according to international criteria (Bryce et al. 2012). The pain should not be primarily related to spasms or any other movement.
* Pain Intensity: At least 4 out of 10 on the Numerical Rating Scale (NRS) at the time of screening (rated during the previous 24 hours).
* Pharmacological Treatment: Stable treatment including antiepileptic, antidepressant, or antispastic drugs (Gabapentin (GBP) with a minimum dose of 900 mg/day, Pregabalin (PGB) with a minimum dose of 150 mg/day, Amitriptyline with a minimum dose of 25 mg/day). No dose changes for at least 2 weeks prior to treatment and no additional antiepileptic medication. The pharmacological regimen must be maintained without changes during the 10-day stimulation period and until the electrophysiological measurement. It is recommended to keep the regimen stable until the completion of the following two evaluations (4 and 12 weeks after the end of treatment). Only paracetamol or anti-inflammatory drugs are allowed as rescue treatment.

Exclusion Criteria:

* Patients with severe pain (NRS \> 7) from other sources, such as musculoskeletal pain, inflammatory pain, or cancer-related pain.
* Subjects with traumatic brain injury.
* Subjects with alcohol abuse.
* Subjects with neurological diseases other than the specified spinal cord injury.
* Subjects with substance abuse.
* Subjects with any other chronic medical condition where transcranial tDCS is relatively contraindicated, such as pregnancy or epilepsy.

Where this trial is running

Badalona, Barcelona

• Institut Guttmann — Badalona, Barcelona, Spain (RECRUITING)

Study contacts

• Principal investigator: Dolors Soler, PhD — Institut Guttmann
• Study coordinator: Dolors Soler Fernandez, PhD

Conditions: Neuropathic Pain, Spinal Cord Injuries, Central Neuropathic Pain, EEG, Machine Learning, Spinal Cord Injury, tDCS