Does transcranial magnetic stimulation (TMS) combined with cognitive training effectively treat mild to moderate Alzheimer's dementia?
That's a question the of the FDA Medical Devices Advisory Committee faces Thursday when it meets to decide whether to recommend the to the FDA.
If all goes the way device maker Neuronix hopes, neuroAD could be the first medical device approved by the FDA to treat Alzheimer's.
But that's a big "if." Because the device's pivotal clinical study failed to meet its primary endpoint -- and because of the agency's uncertainty about the study's post-hoc analysis and the device's "failure to consistently demonstrate clinically meaningful results" compared with sham -- "it is difficult to conclude that the available data demonstrates effectiveness or a clinical benefit of the neuroAD as an adjunctive treatment in the intended population," the FDA wrote in its .
The neuroAD system uses repetitive TMS to stimulate targeted areas of the brain. Tailored cognitive training with a computer touch screen occurs concurrently with TMS treatment. Although the mechanism is unknown, Neuronix hypothesizes that TMS may modulate synaptic activity and potentially enhance the effect of cognitive training. The therapy can be used in conjunction with Alzheimer's drugs.
Neuronix based its submission on a pivotal trial of mild-to-moderate Alzheimer's patients who had treatment or sham sessions over 6 weeks. The primary effectiveness endpoint was the change in Alzheimer's Disease Assessment Scale-Cognitive subscale () from baseline to week 7. (ADAS-Cog scores range from 0 to 70; higher scores indicate greater cognitive impairment and negative numbers signify improvement.) Secondary endpoints were changes in Alzheimer's Disease Cooperative Study-Clinical Global Impression of Change ( at week 7, and changes in ADAS-Cog and ADCS-CGIC at week 12.
Overall, 106 participants were included in the primary efficacy population: 57 in the active treatment group and 49 in the sham group. Sham performed better than neuroAD by 1.45 points on the ADAS-Cog primary endpoint at 7 weeks; the difference was caused by a numerical worsening in the neuroAD group from baseline (+0.07) and a numerical improvement in the sham group (-1.38).
At 12 weeks -- approximately 6 weeks after final treatment -- ADAS-Cog scores reversed direction, with the treatment group performing better than the sham by 0.42 points. The ADCS-CGIC also indicated a difference of 0.02 in favor of treatment at 7 weeks and 0.35 in favor of treatment at 12 weeks, but "it is not clear whether these changes lend any support to the clinical meaning of the changes found on the ADAS-Cog scale," the FDA wrote.
In , Neuronix stated it's possible the treatment had a delayed effect. The company also pointed out that a subset of patients with baseline ADAS-Cog ≤30 showed greater improvement after neuroAD treatment, citing two studies in Korea that showed similar findings. Based on a post-hoc analysis of the pivotal study, Neuronix proposed to limit the neuroAD patient population to people with a baseline ADAS-Cog ≤30, but the FDA seemed uncertain that ADAS-Cog ≤30 was a reliable subgroup with a higher probability of experiencing benefit than the overall cohort.
The FDA acknowledged the risks of using the neuroAD device are low, so questions on Thursday primarily will center on efficacy: whether the data demonstrate effectiveness, whether there is evidence of clinically meaningful benefit, and whether the overall risk-benefit risk assessment of neuroAD is favorable. In addition, the agency has asked the panel to discuss what minimum improvement in ADAS-Cog is clinically meaningful, as well as minimum improvement levels in ADCS-CGIC.
The neuroAD Therapy System submission is a premarket submission; other TMS devices that were first of a kind -- for major depressive disorder and obsessive compulsive disorder -- were cleared through the same de novo pathway. While the FDA will consider input from the panel after Thursday's meeting, it is not obliged to follow recommendations from its advisory committees.