"We know that there's excessive high-frequency gamma band activity that really relates to degree of cognitive impairment and severity of anxiety, and with a single dose of the Spinogenix drug in Fragile X, that completely normalized. That's the strongest effect we've ever seen."
Dr. Craig Erickson didn't start his career in synaptic science. He started at summer camps for kids with autism.
Before medical school, Craig worked directly with youth and young adults living with severe developmental disabilities. What struck him wasn't the diagnosis itself — it was the degree to which communication and cognitive deficits shaped every aspect of their day-to-day lives, requiring lifelong adult supervision. That experience set him on a path into neurodevelopmental research that has now spanned more than two decades.
Along the way, the science kept pointing him in the same direction: the synapse. Research emerging through the 90s and early 2000s — the formative years of his training — highlighted how aberrant signaling via neurotransmitters and disrupted protein synthesis at synaptic connections were central to a wide range of brain disorders. That insight became the foundation of his work.
Today, Craig is a Research Director and Research Endowed Professor at Cincinnati Children's, where he leads the Cincinnati Fragile X Research and Treatment Center. He's also the Chief Medical Advisor at Spinogenix, where he's helping develop drugs that aim to restore synapses — the fundamental connections between brain cells — across conditions as diverse as Fragile X syndrome, Alzheimer's, ALS, and schizophrenia.
Why Synapses Are The Common Thread
The brain contains millions of neurons that need to connect. Those connections — synapses — are the links in the chain of cell-to-cell communication, each with a pre-synaptic and post-synaptic component. When those links break down, the result is pathology. And they break down in a remarkable number of ways across a remarkable number of conditions.
Neurodegenerative disorders like ALS and Alzheimer's. Severe mental illness like schizophrenia. Neurodevelopmental disorders like Fragile X and autism. The underlying causes are different — and the timing varies from fetal development to late adulthood — but the common endpoint is a brain that can't functionally connect.
What makes the Spinogenix approach distinctive is that it targets that shared endpoint. If you can restore synaptic connectivity, you have the potential to impact a broad set of disorders in ways that don't typically come from a single therapeutic platform.
The TAGS Platform: Changing Brain Morphology, Not Just Signaling
Spinogenix operates two main programs under its therapeutic platform. SPG-302, used in ALS, Alzheimer's, and schizophrenia, works through what Craig described as a novel small-molecule approach that enables neurons to generate or regenerate mature synapses where they haven't existed. It's not modulating neurotransmitter levels the way an SSRI or dopaminergic drug would — it's changing the actual morphology of synaptic connections. Craig was candid that he never thought that was possible.
SPG-601, the Fragile X program, takes a different but related approach through modulation of the large conductance BK potassium channel. It's more of a traditional neurotransmission modulator, but it also restores synaptic signaling and connectivity. Both programs share the synapse as their target, but they arrive there through different mechanisms.
A New Trial Model: Small Studies, Objective Brain Data
Craig's involvement with Spinogenix wasn't just about synaptic science. It was also about how trials get run.
Over the past decade-plus, his NIH-funded research has focused on using EEG and other quantitative measures to detect brain target engagement in early-phase studies. The logic is straightforward: if you're running a first-in-disease study with a small number of patients, you need tools that can objectively show whether brain function is normalizing — not just subjective clinical scales that are vulnerable to placebo effects.
This is where Craig's career arc becomes especially relevant. He described being involved in essentially every negative double-blind, placebo-controlled, multi-site, large-scale trial in autism over the past two decades. Treatments that appeared to help two-thirds of patients in open-label clinical use would consistently fail to beat placebo in formal studies of 100 or 200 participants. The field kept chasing post-hoc subgroup analyses, trying to find signal in noise.
That track record of failure forced a rethinking. Instead of enrolling hundreds of patients and relying on subjective measures, the Spinogenix approach was to run a small, highly quantitative trial — 10 men with Fragile X — using computer-based cognitive testing and high-density EEG to measure whether the drug was actually changing brain function. There's no placebo response when you're measuring whether brain oscillations are normalizing at high or low frequencies. It either is or isn't happening.
The result: statistically significant improvements in computer-based cognitive performance and normalization of EEG signals after a single dose. In Craig's words, the strongest effect they've ever seen in an early-phase Fragile X study.
Why EEG Isn't An Approvable Endpoint — Yet
A natural question: if EEG is so effective at detecting signal, why isn't everyone using it? And why isn't it an FDA-approvable endpoint?
Craig was direct about the challenge. The FDA's biomarker qualification process for an approvable endpoint is onerous — there's no other way to describe it. It requires years of validation studies linking quantitative measures to clinical outcomes, both cross-sectionally and over time. Even getting a new pen-and-paper measure through FDA qualification takes hundreds of patients and years of work. A biologic electrophysiologic marker faces an even higher bar.
Where Craig sees EEG fitting today is in three roles: as an outcome measure in earliest-phase studies to justify further development, as a tool for making go or no-go decisions based on quantitative data, and as inclusion or stratification criteria for selecting the right patients.
In the Fragile X work with Spinogenix, for example, the EEG signal effect was specifically linked to males who make zero Fragile X protein — the most severely affected subgroup. That biological link between molecular phenotype and EEG phenotype can then inform who gets enrolled in larger studies.
There has been some progress. The FDA has become open to using the NIH Toolbox, a quantitative computer-based test, as a primary outcome in Fragile X. The Spinogenix trial actually showed statistically significant improvement on part of that test — a result that could be moved forward as an approvable endpoint. But that opening came after years of NIH-funded development work. The bar remains high.
Fragile X Heterogeneity And Precision Patient Selection
One of the most striking aspects of Craig's approach is how carefully he thinks about patient selection — even within a single-gene disorder.
Fragile X syndrome results from a triplet repeat expansion in the FMR1 gene, leading to reduced or silenced production of Fragile X messenger ribonucleoprotein. But the deficit isn't absolute — it's continuous. Females with Fragile X are phenotypically variable due to random X inactivation: they may or may not have intellectual disability, may or may not have autism, though the majority experience anxiety. Males are more uniformly affected since they only have one X chromosome, but even among males, roughly a third make zero protein, a third make trace amounts (about 5% of normal), and the remaining third fall in a 5–20% range.
That's meaningful heterogeneity in what's often described as a straightforward genetic condition. For the first Spinogenix trial, Craig enrolled only males making zero or trace FMRP — because he knew they had a consistent EEG phenotype and could set up EEG as a reliable marker of target engagement.
The lesson extends beyond Fragile X. Craig described parallel success with a company called Stelicla, which used machine learning to identify a subphenotype within the broader autism population. A small 12-patient trial in that subphenotype, using EEG and cognitive outcomes, produced strong results. The principle is the same: winnow down heterogeneous populations, select for biological consistency, use quantitative measures, and you have a real chance of detecting signal — or reliably concluding that there isn't one.
Chronic Dosing: Can The Gains Be Sustained?
The single-dose data is remarkable, but the critical question is whether the effects hold up over time. Craig acknowledged the concern directly — tachyphylaxis (the wearing-out of a drug's effect with repeated use) has been a real problem in Fragile X research, particularly with mGluR5 antagonists that showed promising early results but couldn't sustain them.
The SPG-601 molecule appears to be different on this front. It was in-licensed by Spinogenix with existing phase 2 safety data from work in multiple sclerosis, including repeated dosing data out to 28 days with no evidence of tolerance. Preclinical validation funded by the FRAXA Research Foundation specifically tested for tachyphylaxis using the same assays that flagged the problem with mGluR5 antagonists — and found none.
Craig is cautious but optimistic. A brain that immediately showed improved cognitive performance and normalized EEG signals is, in his view, a brain that is now able to process information differently — able to learn and communicate in ways that the excessive high-frequency interference of Fragile X had been preventing. The chronic dosing studies will need to confirm that, but the preclinical and early clinical evidence is encouraging.
What Meaningful Improvement Looks Like For Patients And Families
When Craig talks about the potential impact of chronic dosing, he's specific about what improvement means in the lives of the patients he sees every day.
The males in the zero-protein subgroup are severely affected. They have significant intellectual disability, high anxiety, pronounced ADHD symptoms. The vast majority require lifelong adult support and supervised housing. A very small percentage live even semi-independently. Almost all have at least one parent or caregiver who has had to forgo normal employment to provide full-time care — and who worries constantly about who will take over as they age.
The kinds of functional gains Craig envisions aren't abstract. An individual who needed a van ride every day might be able to take public transportation. Someone who could only sit in a day program watching television might move to supported employment. Someone in supported employment might reach independent employment. Those are the outcomes that would transform quality of life not just for the individual, but for entire families — and they carry enormous financial implications at a societal level as well.
Looking Ahead: Synaptic Repair And Beyond
Spinogenix's near-term roadmap centers on moving SPG-601 into chronic dosing studies in Fragile X, starting with the well-defined male subpopulation and then expanding to females and mosaic males. For SPG-302, long-term treatment data in ALS is expected soon, with decisions about prioritization across ALS, Alzheimer's, and schizophrenia to follow based on results.
Craig also pointed to the potential for rational combination therapy. In Fragile X, for example, Shionogi is currently running phase 3 studies of a phosphodiesterase type 4 inhibitor focused on cognitive enhancement. That molecule has a different EEG profile from SPG-601 — it doesn't show the same bold normalization of high-frequency activity. Craig sees potential synergy: two compounds with different brain response profiles that could be complementary. EEG could serve as the tool for evaluating those combinations with more sophistication than simply pairing two safe drugs and hoping for the best.
Beyond synaptic restoration, Craig expressed excitement about gene and cellular therapy — AAV-mediated CRISPR gene editing, cell-penetrating protein work, and the promise that spinal muscular atrophy's treatment with AAV-based gene therapy has opened for other rare genetic conditions. But he sees small molecules and gene therapies as complementary, not competitive. Gene therapy may eventually reach neonates, but there are far more adults living with Fragile X than children — and they need effective strategies to improve their quality of life now.
As Craig put it: he has a rolodex in his head of every patient he's ever seen. That's what drives the work. You understand the challenging outcomes, and you pound the door down every day.
