Neuralink adds chronic stroke to a new US implant trial

By Inside BCI · Edited by Velco Dar · November 13, 2026

A stroke can leave the hand that once typed, swiped and wrote barely working. Neuralink is now recruiting some of those people. Its new US study, PRIME2, adds chronic ischaemic stroke to the spinal cord injury and ALS covered by PRIME, its first US trial, and will test whether participants can control a computer cursor or virtual keyboard with its brain implant.

What Neuralink registered

PRIME2 was first posted on ClinicalTrials.gov on 10 November under protocol N1-EFS-300. It is an early feasibility study of the N1 Implant, placed in the motor cortex by Neuralink’s R1 surgical robot, and the record says it “builds on knowledge gained in a previous clinical investigation of this system.”

Up to 10 participants will be enrolled at up to five sites in the United States. The University of Miami, a PRIME site since 2025, is the only one listed so far and is recruiting, with neurosurgeon Jonathan Jagid as principal investigator. Participants will be followed for about two years after implantation, or about three if they choose extended follow-up. There is no comparison group; each person’s results are compared with their own baseline.

The primary outcome is serious adverse device effects from the day of surgery through a 90-day follow-up visit. The study will also assess digital daily activities, computer-control performance, independent use of the system, and outcomes reported by participants, clinicians and caregivers. The record gives an estimated start of November 2026 and primary completion in November 2027.

Who can join

The first version of the record, which cleared the registry’s checks on 5 November and was posted on 10 November, set out the stroke criterion in detail. It required a chronic ischaemic stroke at least 12 months before screening, “resulting in residual hemiparesis of the pre-morbid dominant upper extremity, with a completion time on the Nine-Hole Peg Test (9-HPT), per standardized protocol, that exceeds 3 standard deviations above the respective established age- and hand-specific normative mean completion times for both hands.” Hemiparesis is weakness on one side of the body; the Nine-Hole Peg Test is a standard timed test of hand dexterity.

That version also excluded anyone with damage to the primary motor cortex region on MRI and, for stroke patients, anyone with imaging evidence of stroke in the planned implant target area.

A version submitted on 11 November and posted on 12 November shortened the eligibility section. It now opens with a one-sentence summary, “Adults with a diagnosis of ALS, spinal cord injury or chronic ischemic stroke who have severe weakness in the arms and hands that limits independent use of a computer or phone”, followed by short lists of inclusion and exclusion criteria. The motor cortex and implant-target exclusions no longer appear in the public listing. Both versions exclude people taking blood thinners that cannot be paused for surgery.

How it differs from PRIME

PRIME required “severe quadriplegia” from spinal cord injury or ALS for at least a year, limited participants to ages 22 to 75, and measured device and procedure adverse events at 12 months. PRIME2 accepts adults from 18 with no upper age limit and adds stroke.

Neuralink’s other registry entries have included stroke. Its US speech study, VOICE, lists stroke among the conditions that qualify. Its device-control studies in Britain and the United Arab Emirates admit stroke where it leaves weakness in both arms, and the British criteria name brainstem stroke. PRIME2 is the first Neuralink study to name chronic ischaemic stroke.

Signal: Why it matters

Stroke affects far more people than the conditions PRIME covered. More than 795,000 people in the United States have a stroke each year, and about 87% of strokes are ischaemic, caused by a blocked blood supply to the brain. By comparison, there are about 18,500 new traumatic spinal cord injuries a year. In a 2005 review in the New England Journal of Medicine, neurologist Bruce Dobkin wrote that six months after a stroke, about 65% of patients cannot incorporate the affected hand into their usual activities.

The trial itself is small and its criteria are strict, excluding people with cognitive impairment and, in the first version of the record, damage to the motor cortex, so only a fraction of stroke survivors would qualify. It is the first step in testing whether the population Neuralink can serve extends beyond paralysis from spinal cord injury and ALS.

Other implant makers have registered studies that list stroke, including Synchron’s COMMAND and Paradromics’ Connect-One. Devices for chronic stroke are also authorised for US marketing. The FDA approved Vivistim, an implanted vagus nerve stimulator made by MicroTransponder, now Mobia Medical, for chronic ischaemic stroke patients with moderate to severe arm impairment in August 2021. It granted De Novo authorisation to Neurolutions’ IpsiHand, a non-invasive EEG brain-computer interface used in rehabilitation, in April 2021. Both are used to improve arm and hand function during rehabilitation. PRIME2 measures how well participants control a computer through the implant.

Not disclosed

Neuralink has not said why it added stroke, why the eligibility criteria were shortened, which side of the brain it will implant in stroke patients, or when it expects its first stroke implant. It has not named the other sites. The most recent participant count on Neuralink’s website is 21 people enrolled in its trials worldwide, from January 2026.

Forward Radar: What to watch

Whether the first stroke participant is implanted at the University of Miami, and when other sites are added.

How cursor and keyboard control in stroke patients compares with Neuralink’s results in spinal cord injury and ALS.

Whether the stroke criteria change again before enrolment.

Sources

Primary:

Market and regulatory context:

Cross-reference to prior InsideBCI coverage:

news Neuralink PRIME2 N1-Implant R1-Robot stroke chronic-ischemic-stroke clinical-trials University-of-Miami Jonathan-Jagid Vivistim IpsiHand