Comparing Biomarker-Guided DBS Programming With Standard Clinical Monopolar Programming

Deep brain stimulation (DBS) is an established therapy for patients with Parkinson's disease (PD) with motor fluctuations inadequately controlled by medication. However, DBS programming is currently based on monopolar review, a time-consuming process that relies heavily on clinician experience and subjective patient feedback. This approach does not systematically incorporate objective anatomical or neurophysiological information regarding the optimal stimulation contact and may require several months to achieve satisfactory clinical benefit.

Recent advances in DBS technology allow the integration of objective markers into the programming process. These include postoperative anatomical reconstruction of electrode location using neuroimaging and the recording of local field potentials (LFPs), particularly beta-band activity, from implanted DBS systems. In parallel, wearable inertial sensors enable continuous, objective quantification of motor states such as OFF time, dyskinesias, motor fluctuations, and gait disturbances in daily life.

This study is a pilot, randomized, controlled, parallel-group clinical trial with single-blind assessment, designed to evaluate whether DBS programming guided by objective markers (electrode localization on neuroimaging combined with beta LFP sensing) improves clinical outcomes compared with conventional programming based on monopolar clinical review. The study will be conducted in patients with Parkinson's disease undergoing DBS surgery at the Movement Disorders Unit of the Hospital de la Santa Creu i Sant Pau.

A total of 20 participants with idiopathic Parkinson's disease will be enrolled and randomized in a 1:1 ratio to one of two programming strategies: (1) conventional DBS programming based on clinical monopolar review, or (2) DBS programming guided by anatomical electrode reconstruction using Brainlab Elements™ and neurophysiological beta biomarkers obtained via BrainSense™. All participants will be implanted with a clinically approved DBS system (Percept™ PC or RC with SenSight™ directional leads, Medtronic) and will wear PDMonitor® inertial sensors for objective motor assessment. The only difference between groups is the DBS programming strategy.

All participants will undergo a 7-day PDMonitor® recording and a full clinical evaluation during the preoperative visit. DBS surgery will be followed by an approximately one-month postoperative period without stimulation. Initial DBS programming will begin around week 5 after surgery. A mid-programming clinical evaluation and PDMonitor® recording will be performed at approximately week 9. At 3 months after initiation of programming, participants will complete another 7-day PDMonitor® recording and a full clinical assessment, which will serve as the primary endpoint evaluation. Adaptive DBS (aDBS) will then be activated in both groups. A final follow-up visit at 6 months will include clinical assessments and sensor-based measurements; these post-aDBS data will be analyzed descriptively and exploratorily only.

The primary outcome is the percentage change in daily time spent in the OFF state between baseline (preoperative PDMonitor® recording) and 3 months after the start of DBS programming. Secondary outcomes include changes in quality of life measured by the PDQ-39, motor symptoms and motor complications assessed by the MDS-UPDRS parts III and IV, and non-motor symptoms assessed by the MDS-NMS, all between baseline and 3 months. Exploratory outcomes include descriptive changes in motor and functional measures between months 3 and 6 after activation of adaptive DBS.

PDMonitor® is used exclusively as a research assessment tool and does not constitute a therapeutic intervention. All implanted DBS devices, imaging reconstruction tools, and adaptive stimulation algorithms are CE-marked and used within their approved clinical indications.

The statistical analysis will follow a Bayesian framework appropriate for a pilot randomized study. Descriptive analyses will be performed for all variables. Bayesian linear mixed-effects models will be used to compare changes from baseline to 3 months between groups for the primary and secondary outcomes. Results will be expressed using posterior distributions, 95% credible intervals, Bayes factors, and comparison with a region of practical equivalence (ROPE). Analyses will be conducted primarily on an intention-to-treat basis, with per-protocol and sensitivity analyses to assess robustness. Missing data will be handled using Bayesian or multiple imputation methods as appropriate, except for the primary outcome, which will not be imputed in the absence of valid PDMonitor® recordings.

The study is exploratory in nature and is not designed as a confirmatory efficacy trial. It will be conducted in accordance with Good Clinical Practice (ICH-E6), the Declaration of Helsinki, and applicable European and Spanish data protection regulations. All participants will provide written informed consent prior to inclusion.