_____20230922232700.thumb_head

Photo/Neuralink

Three years ago, Elon Musk's brain-computer interface company Neuralink livestreamed a demonstration of real-time neural activity from three pigs with implanted brain devices, sparking a new wave of "brain-computer craze." Musk was optimistic at the time that human trials of brain-computer interfaces would begin soon. However, the approval process was much more complex than anticipated.

This week, Neuralink finally received approval for human trials, advancing Musk's science fiction dream of "man-computer symbiosis" one step further.

Kristin Welle, a former FDA official and neuroscientist who oversees medical regulation, told NBD that the approval indicates that Neuralink has provided data that meets FDA safety requirements.

Musk's "man-computer symbiosis" dream takes an important step forward

On September 19, local time, Neuralink announced on its website that it would be recruiting participants for the first human clinical trial of its man-computer interface device. The trial will assess the safety and initial efficacy of the device, namely whether paralyzed patients can control external devices with their thoughts.

The trial, called PRIME (Precision Robotic Implantation of a Neural Interface), will use a robotic surgery system called R1 to implant a device called N1 in the part of the brain that controls movement. N1 is a thin, flexible thread that can record brain signals and wirelessly transmit them to an application that decodes movement intent.

Neuralink said it is looking for participants who have been paralyzed in all four limbs due to spinal cord injury or amyotrophic lateral sclerosis and have not shown any improvement in one year after their injury. The main trial will take about 18 months, and the entire clinical trial will take about six years if long-term follow-up appointments are included.

Neuralink's clinical trial has been four years in the making since its initial application, and it was rejected twice by the FDA during that time. In early 2022, the FDA made it clear that Neuralink needed to address dozens of issues before conducting human trials. The FDA's safety concerns included the safety of the lithium battery, whether the wires of the implant would move to other areas, and the safety of removal. In May 2023, the FDA finally granted Neuralink a conditional trial approval.

“The FDA does not disclose any information regarding their reviews of requests for clinical trials. So we don't know exactly what safety concerns were raised.” Kristin Welle, a former FDA official and current associate professor of neuroscience at the University of Colorado, told NBD. “However, because the FDA agreed to let the trial begin, it means that Neuralink provided data to satisfy the FDA's requirements for safety and for an appropriate clinical trial design. ”

The initial purpose of the trial is to allow paralyzed individuals to control a computer cursor or keyboard with their thoughts. According to Musk's statements in multiple occasions in recent years, Neuralink's short-term goal is to restore the vision of the visually impaired and allow the paralyzed to regain full body movement function. The ultimate goal is to achieve "human-machine symbiosis" in which the brain and the computer work together, which Musk believes can help to resist the artificial intelligence that he believes could threaten humanity.

Just like Musk's dream of colonizing Mars, brain-computer interfaces also have their science fiction origins. According to the Elon Musk Biography published in September, the inspiration for Neuralink came from the Culture series of space travel novels by Iain M. Banks, in which a device called "neural lace" is implanted in the human body to connect all human thought processes to a computer. Musk said, "When I first read Banks' work, I suddenly felt that this idea could become a shield for us in the face of artificial intelligence."

Brain-computer interface may need 5 to 10 years to commercialize

After a series of setbacks, Neuralink has fallen behind its competitors in the progress of human trials.

Another man-computer interface company in the United States, Synchron, received FDA approval for a trial in 2021 and announced the first human implantation in the United States in July 2022. In May 2023, a group of Swiss researchers published a paper in Nature, claiming that they implanted a brain device in a paralyzed man as early as July 2021, which allowed him to walk with the help of a walker.

Of course, Neuralink still has its own technological advantages. Traditional implantable brain-computer interfaces use a hard-wired electrode called the Utah array, while Neuralink uses a flexible electrode that can reduce brain rejection reactions and improve the quality of the neural information collected. In addition, Neuralink has developed a robotic surgery system for brain-computer interface surgery, which minimizes incisions during implantation and removal of the device.

To realize Musk's grand vision, this human trial is just a small step forward. "Neuralink is still in a very early stage of device development and commercialization, and is now in the early stage of human trials," Kristin Welle told NBD. "It still needs larger-scale trials, followed by pivotal trials, and then an application for premarket approval (PMA)."

Welle further explained to the reporter that Neuralink is still at the very beginning stages of device development and commercialization. They will begin with a small human study, and then likely use those results as the basis of a future request for one or more additional human trials. The device they are building will likely be classified as the highest risk level, Class 3, because it is implanted into the brain. This means that eventually, they will need to obtain a pre-market approval (PMA) from the FDA before they begin selling the device commercially. To obtain a PMA, they will need to conduct a pivotal clinical trial that is designed to demonstrate both the safety and the effectiveness of the device to treat a specific condition. A typical path for PMA would look like an early first-in-human trial (this is where Neuralink is right now), then a slightly larger feasibility trial, then a large pivotal trial, and then the submission of a PMA application to the FDA. 

This process requires a lot of money and time. Welle has previously told media that Neuralink may not be able to commercialize for at least 5 to 10 years.

On the other hand, Neuralink also faces multiple challenges in terms of technology and application prospects. Welle analyzed to NBD that Neuralink needs to prove that the device can be placed in the brain for a long time without the need for frequent replacement, that it can capture and interpret neural signals, and most importantly, that the technology is really helpful to patients.

"Even if Neuralink can demonstrate all of these things, the device will still be expensive and require brain surgery, so the number of patients who might be willing to be implanted might be small. " Welle said.

Editor: Alexander