The IRA's discount gets the headlines. The clock is what rewrites your launch

Melonie WarfelChief Commercial Officer, SwishX·10 min read
Contents
  1. Introduction
  2. Three cycles in, the program is no longer theoretical
  3. How the clock and the ceiling actually work
  4. The discount is a cost. The clock is a deadline on value
  5. The first indication has become a pricing decision
  6. The case that this is overblown, and why I only half agree
  7. Europe has priced value at launch for years
  8. What launch teams should change now

Medicare's negotiated prices are real, but they land late in a drug's life and finance teams can model them. What the Inflation Reduction Act really changes is when a medicine has to prove its value, and that moves the work to launch day.

Most of what gets written about Medicare drug price negotiation is about the discount. I think that is the wrong number to stare at. The discount arrives nine or thirteen years after approval, it is bounded by a formula, and any decent revenue management team can model it. The bigger change is the clock: a fixed date by which a medicine has to have earned its keep in Medicare, and a clock that starts on the day of the first approval, whatever that first approval happens to be for.

A fixed clock pulls the hard work of a brand's life forward. Which indication goes first, whether a molecule is a pill or a biologic, how fast the evidence program runs and how ready the commercial team is on launch day all now sit inside a window that does not move. My career has been on the vendor side of this industry, including time at Model N on pharma pricing and revenue management, and these are my views on what follows. They are not legal advice.

Three cycles in, the program is no longer theoretical

The first ten negotiated prices took effect on January 1, 2026, all for Part D drugs, including Eliquis, Jardiance, Xarelto and Stelara. CMS estimated that if those prices had applied in 2023, Medicare would have spent about $6 billion less on them, a 22% net reduction, and that people with Medicare would save about $1.5 billion out of pocket this year.

Last November the second cycle closed. CMS reached agreement on all 15 drugs, Ozempic, Rybelsus and Wegovy among them, with prices that take effect on January 1, 2027. CMS put the savings at about $8.5 billion, roughly 36% lower net spending, had those prices applied in 2024.

The third cycle is the one that changes the shape of the program. In January, CMS selected 15 more drugs and, for the first time, some are paid under Part B, the benefit that covers drugs physicians administer in the clinic. Botox, Cimzia, Entyvio and Orencia are on the list. The 15 drugs accounted for about $27 billion in Medicare spending over twelve months, and CMS also picked Tradjenta for the program's first renegotiation. Under the published timeline, CMS's final offers were due by the end of September, companies have until October 31 to respond, and CMS must publish the prices by November 30. They apply from 2028.

From 2027, up to 20 drugs a year can be selected across both benefits, and CMS proposed the rule for that fourth cycle in June. KFF counts the 40 drugs selected so far as $125 billion of Medicare Part B and Part D spending in 2024, about 36% of the total before rebates. Whatever you think of the policy, it already reaches more than a third of what Medicare spends on drugs.

How the clock and the ceiling actually work

A small molecule becomes eligible for selection once it is at least seven years past FDA approval, and the negotiated price takes effect about two years later, so call it nine years of Medicare pricing set by the market. A biologic gets eleven years to selection and roughly thirteen to the new price. If a drug has several approvals, CMS counts from the earliest. A marketed generic or biosimilar takes a drug out of the program, and drugs with low Medicare spending are excluded.

The part I find most revealing is the ceiling. The negotiated price, the maximum fair price, cannot exceed a percentage of the drug's non-federal average manufacturer price, and the percentage falls as the drug ages: 75% for a drug between nine and twelve years past approval, 65% between twelve and sixteen, and 40% after that. CMS then negotiates down from the ceiling, weighing the company's own data alongside the evidence on therapeutic alternatives, comparative effectiveness and unmet need. In other words, the law prices age. The longer a drug has been on the market, the less room the formula leaves.

There is also a plumbing change that gets less attention than it deserves. Since January, pharmacies buy the selected drugs at their usual price, Part D plans pay them at the negotiated price, and the manufacturer makes up the difference, with claims data routed through CMS's Medicare Transaction Facilitator. The manufacturer has 14 days from receiving the dispensing data to work out what it owes. Anyone who has run gross-to-net knows what a new payment stream with a two-week clock does to forecasting, disputes and cash. Part B drugs will add a different version of that problem in 2028.

The discount is a cost. The clock is a deadline on value

Before the IRA, a slow launch was expensive but forgiving. If uptake took four years instead of two, you lost revenue early, but the top of the curve was still there until the patent cliff. A negotiation date changes that arithmetic for any drug big enough to be selected. Every month a brand spends ramping is a month taken off the most profitable stretch of its Medicare life, because the end date does not move when the launch is late.

I find it useful to think of this as a value window: the years between first approval and the date a negotiated price would apply. There are three ways to lose that window, and only one of them is about price. You can start slowly, with a field force, payer story and content that are not ready on approval day. You can prove value late, with the evidence that would change a prescriber's or a payer's mind arriving in year five. Or you can spend the clock on a first indication that is too small to matter.

The third one is the least intuitive, and oncology shows why. A Health Affairs study of 184 cancer drug approvals from 2000 to 2024 found that 41.8% of them went on to win at least one follow-on approval, and 59.7% of those follow-on indications were in earlier stages of disease than the original. The classic pattern was to launch in late-line patients and walk the drug back toward earlier, usually larger, populations over many years. Under a clock that starts at first approval, that pattern puts the biggest population at the end of the window, closest to the negotiated price.

The first indication has become a pricing decision

Companies are already adjusting. In interviews Tapestry Networks published in December 2024, industry leaders said they expected to move away from the old sequencing playbook and prioritize larger populations for the first indication, and some described larger companies developing parallel molecules against the same target rather than adding indications to one. Pfizer has said it will favor biologics in its oncology pipeline, a choice that BioSpace and others have tied to the law. I would treat interview data and single examples with care, but the direction is consistent with the incentives.

Congress has already changed the clock once. The One Big Beautiful Bill Act, signed on July 4, 2025, widened the orphan drug exclusion to drugs designated for one or more rare diseases, and for an orphan drug that later wins a non-orphan approval, it starts the clock at that later approval. The change applies from the 2028 cycle. CBO put the cost at $8.8 billion over ten years, and KFF expects it to delay the selection of Keytruda and Opdivo by at least a year. For companies with a rare disease asset, orphan-first sequencing is protected again. For a small molecule aimed at a common disease, nothing changed.

The small molecule gap is the other open question. The EPIC Act, introduced in the House in February 2025 as H.R. 1492, would give small molecules the same thirteen years as biologics. As of October 2026 it has not become law, so the nine-versus-thirteen gap, the so-called pill penalty, is what boards are planning against. That gap makes the modality choice partly a pricing choice, made years before anyone in commercial is asked for an opinion.

Then there is the patient mix. Selection is driven by Medicare spending, so a drug used mostly by people over 65 feels the full force of the clock, while one used mostly by commercially insured patients may never be selected at all. Two indications with the same clinical promise can now carry very different economic lives depending on who the patients are.

The case that this is overblown, and why I only half agree

The strongest counterargument goes like this. Most drugs will never be selected. Only the highest-spending products, 15 or 20 a year, get picked, and drugs below a spending threshold are excluded. Nine years is a long time, and many brands face generic entry within that window anyway. CMS has already set dates for some negotiated drugs to leave the program, NovoLog and Fiasp among them, the step the law takes once a generic or biosimilar is on the market. The rules also keep moving, through the orphan change, the pending EPIC Act and the administration's separate voluntary pricing deals, such as the $245 a month Medicare price for GLP-1 drugs announced in November 2025. On that view, rebuilding launch strategy around one statute is overreaction.

For the median drug, I think that is right. But the decisions that matter most are made for the drugs a company hopes will be big, and those are exactly the drugs that end up on the list. Being selected is the price of winning. A launch plan that assumes success has to assume the clock.

Where I am less sure is how much of the shift shows up in R&D rather than in commercial behavior. Much of the data on falling small molecule investment comes from advocates on one side of the EPIC Act fight, and I would not build a strategy on it. The commercial consequences are easier to see, because they follow directly from a fixed end date.

Europe has priced value at launch for years

None of this would surprise a European market access team. Germany's AMNOG process assesses added benefit at launch, and the 2022 GKV Financial Stabilization Act shortened the window of free pricing from twelve months to six, so the negotiated reimbursement amount now applies from the seventh month. Since January 12, 2025, new cancer medicines and advanced therapies filed with the EMA also go through an EU joint clinical assessment, with orphan medicines following in 2028 and all new medicines in 2030.

The US got to a similar place by a different route. Europe asks for the value case at the door. The IRA lets you in without one and then sets a date by which the value had better be visible. The practical result is the same on both sides of the Atlantic: the evidence, the value story and the commercial capability need to be ready at launch, not assembled in year two.

For global brands that is an opportunity. One evidence plan built for launch can serve a German benefit assessment, an EU joint clinical assessment, a US payer review and, eventually, a negotiation with CMS. The companies that still run these as four separate projects will pay for it four times.

What launch teams should change now

Put the negotiation date in the launch forecast. Every asset that could plausibly reach the spending threshold should have a modeled value window by indication sequence, with the eligibility date, the ceiling that would apply and the Medicare share of the likely patient population. If the first indication is chosen before pricing, access and brand have seen that model, the most expensive decision in the asset's life has been made without them.

Build the evidence the negotiation will ask for, starting now. CMS weighs comparative effectiveness, unmet need and the effect on specific populations. The data a company submits in year seven is the data it chose to generate in years zero to five. The same evidence feeds the payer conversation at launch, so plan the AMCP dossier, the real-world evidence program and the head-to-head questions as one program with one timeline.

Compress the ramp. That means the field force trained and certified before approval, payer and HCP materials through review before the PDUFA date, and the capacity to update every asset quickly when a new indication or a label change lands, because each supplemental approval now has fewer years to pay back. Content is usually the slowest part of that chain, which is why it is worth building an MLR-ready content engine before launch rather than after the first delays. A claims library that is already approved makes the second and third indication weeks of work instead of months.

Bring revenue management in early. Effectuating the maximum fair price, the 14-day payment cycle and the arrival of Part B prices in 2028 all belong in gross-to-net models, contract strategy and channel planning now, not in a scramble the quarter before they bite. And watch three dates: November 30, when the third-cycle prices are due; the final rule for the 2029 cycle, expected this fall; and any movement on the EPIC Act.

Under the IRA, the most expensive thing a brand can waste is not a discount point. It is the first two years.

Filed underPerspectives

Keep reading

All articles

Want to see Magic Studio on your brand?

Video, visual aids, ads, approved email and web, made from your approved claims and ready for MLR review.