By Rachael Levy
(Reuters) - Neuralink's disclosure last week that tiny wires inside the brain of its first patient had pulled out of position is an issue the Elon Musk company has known about for years, according to five people familiar with the matter.
The company knew from animal testing it had conducted ahead of its U.S. approval last year that the wires might retract, removing with them the sensitive electrodes that decode brain signals, three of the sources said. Neuralink deemed the risk low enough for a redesign not to be merited, the sources added.
Neuralink is testing its implant to give paralyzed patients the ability to use digital devices by thinking alone, a prospect that could help people with spinal cord injuries.
The company said last week that the implant's tiny wires, which are thinner than a human hair, retracted from a patient's brain in its first human trial, resulting in fewer electrodes that could measure brain signals.
The signals get translated into actions, such as moving a mouse cursor on a computer screen. The company said it managed to restore the implant's ability to monitor its patient’s brain signals by making changes that included modifying its algorithm to be more sensitive.
The sources declined to be identified, citing confidentiality agreements they had signed with the company. Neuralink and its executives did not respond to calls and emails seeking comment.
The U.S. Food and Drug Administration was aware of the potential issue with the wires because the company shared the animal testing results as part of its application to begin human trials, one of the people said.
The FDA declined to comment on whether it was aware of the issue or its possible significance. The agency told Reuters it would continue to monitor the safety of patients enrolled in Neuralink's study.
Were Neuralink to continue the trials without a redesign, it could face challenges should more wires pull out and its tweak to the algorithm proves insufficient, one of the sources said.
But redesigning the threads comes with its own risks. Anchoring them in the brain, for example, could result in brain tissue damage if the threads dislodge or if the company needs to remove the device, two of the sources said.
The company has sought to design the threads in a way that makes their removal seamless, so that the implant can be updated over time as the technology improves, current and former employees say.
In January, Neuralink implanted the device in the brain of its first patient, Noland Arbaugh, who is paralyzed from the shoulders down due to a 2016 diving accident.
In the weeks after the surgery, “a number of threads retracted from the brain,” Neuralink said in a blog update last week. The post made no mention of adverse health effects to Arbaugh and did not disclose how many of the device’s 64 threads pulled out or stopped collecting brain data.
So far, the device has allowed Arbaugh to play video games, browse the internet and move a computer cursor on his laptop by thinking alone, according to company blog posts and videos. Neuralink says that soon after the surgery, Arbaugh surpassed the world record for the speed at which he can control a cursor with thoughts alone.
It is common for medical device companies to troubleshoot different designs during animal trials and for issues to arise during animal and clinical testing, according to outside researchers and sources who have worked at Neuralink and other medical device companies.
Specialists who have studied brain implants say the issue of threads moving can be hard to solve, partly due to the mechanics of how the brain moves inside the skull.
Robert Gaunt, a neural engineer at the University of Pittsburgh, described the movement of the wires so soon after the surgery as disappointing but said that is not unforeseen. “In the immediate days, weeks, months after an implant like this, it’s probably the most vulnerable time,” he said.
PIG HEAVING
In 2022, the FDA initially rejected Neuralink’s application to begin human trials, and raised safety concerns about the threads, Reuters exclusively reported last year.
Neuralink conducted additional animal testing to address those concerns, and the FDA last year granted the company approval to begin human testing.
The company found that a subset of pigs implanted with its device developed a type of inflammation in the brain called granulomas, raising concerns among Neuralink’s researchers that the threads could be the cause, according to three sources familiar with the matter and records seen by Reuters.
Granulomas are an inflammatory tissue response that can form around a foreign object or an infection.
In at least one case, a pig developed a severe case of the condition. Company records reviewed by Reuters show that the pig developed a fever and was heaving after surgery. Neuralink’s researchers did not recognize the extent of the problem until examining the pig’s brain post-mortem.
Inside Neuralink, researchers debated how to rectify the issue and commenced a months-long investigation, said the sources familiar with the events.
Ultimately, the company could not determine the cause of the granulomas, but concluded that the device and the attached threads were not to blame, one of the sources said.