**1. Introduction**

Deep brain stimulation (DBS), which involves the insertion of electrical leads into important cortical or subcortical structures to treat various neurological diseases, is often performed in a single surgery where the leads are subsequently connected to an implanted stimulator. Alternatively, DBS can be performed in a two-staged surgery where the leads (or electrodes) are first implanted, externalized (i.e., the non-neural side of the electrode exits the body), and connected to an external stimulator for a trial period to assess therapeutic benefit over 3–10 days. DBS trials with lead externalization have been used extensively over the last 50 years specifically for treating chronic pain [1–3]. Lead externalization permits inpatient testing of stimulation e ffects prior to permanent implantation, which is especially important for novel or unapproved DBS indications to ensure successful therapeutic response. This is especially important for diseases that encompass broad domains of symptoms spanning from neuropsychiatric to somatic, such as chronic pain syndromes. Because optimal brain targets for chronic pain are still unknown, such trials o ffer a key opportunity for the exploration and validation of both targets and pulse parameters that may modulate activity in multiple, relevant brain networks underlying various symptoms. Further, such a trial period allows the recording of neurophysiological activity from critical brain regions to further research and understand the mechanisms of action of stimulation. Modern recordings have aimed to uncover key biomarkers of chronic pain states toward the development of adaptive (or "closed-loop") control algorithms where

stimulation delivery is adjusted in response to biomarkers to increase e fficacy or avert the development of long-term tolerance.

Such trial periods that allow patients to "test drive" the neuromodulation therapy are used even when the general target regions are known, such as in spinal cord stimulation (SCS) [4]. A trial period is often necessary because of the large heterogeneity in patient response, to assess safety, and for the fine tuning of electrode position relative to the neural target so that patient benefit can be maximized before committing to an expensive therapy. It is worth pointing out that a short trial period to determine chronic brain stimulation targets assumes that an acute response to stimulation can be found (within minutes or hours of stimulation) and that these effects will translate to long-term efficacy. It remains to be determined whether this is the case for chronic pain.

The global burden of chronic pain is significant and growing alongside that of non-communicable diseases, which account for 78.6% of years lived with disability worldwide [5]. The economic impact of chronic neurological conditions such as chronic pain and epilepsy is greater than that of many other health conditions due to absenteeism, reduced levels of productivity, and increased risk of leaving labor markets, indicating that there is a significant indirect economic cost associated with these conditions [6]. Chronic pain also poses a significant cost to the healthcare system; analysis of a Canadian database indicates that the costs attributed to the healthcare of patients with painful neuropathic disorders were considerably higher than those in patients without chronic pain in the same age and sex demographics [6]. Perhaps most importantly, chronic neurological conditions have a significant debilitating e ffect on the overall quality of life of patients and are associated with some of the poorest quality-of-life indices, with the potential to impact social relationships, economic participation, and mental health. DBS for pain is still performed with a very limited trial period (o ff-label in the U.S.) resulting in highly variable success rates across patients.

In this review, we outline important considerations for DBS trial periods for chronic pain including possible brain targets for stimulating and recording. We then propose an argumen<sup>t</sup> for the use of stereoelectroencephalography (sEEG) in the trial and discuss possible trial related risks. An sEEG based approach is elaborated in the context of characterizing pain biomarkers and neurophysiological effects of stimulation. Finally, we o ffer practical advice for conducting a trial, including mitigating the placebo e ffect and rigorously ensuring therapeutic e fficacy before permanent implantation.
