The unintended consequences of the inflation reduction act on biosimilar market incentives and Medicare savings

Abstract

Introduction

The Inflation Reduction Act (IRA) authorizes Medicare price negotiation but includes a “special rule” deferring negotiation for biologics with “imminent” biosimilar competition. This study examined the potential impact of this provision on Medicare spending.

Methods

We modeled Medicare savings under three hypothetical scenarios. In Scenario 1, we applied historical price reductions following biosimilar entry (2017-2024) for 10 reference products and 30 biosimilars to forecast savings for ustekinumab (Stelara) after biosimilar entry. Scenario 2 estimated savings from the IRA's negotiated maximum fair price for ustekinumab. Scenario 3 modeled a modified IRA implementation policy in which ustekinumab was excluded from negotiation and replaced by palbociclib (Ibrance), an eligible high spend drug that was not selected for the first round of Medicare price negotiation.

Results

Across all biologics in our sample, historic market-weighted prices declined to 40.3% of pre-entry levels within five years of biosimilar launch. Negotiating ustekinumab under the IRA yielded greater first-year savings, but cumulative savings were highest in the modified scenario-combining negotiated discounts for palbociclib with biosimilar-driven price declines for ustekinumab.

Conclusion

Selecting biologics with near-term biosimilar competition for IRA negotiation may produce short-term savings but forgo greater long-term savings achievable through competition.

Introduction

Biosimilars play a critical role in promoting competition and lowering spending on biologic therapies. Prior research shows that within three years of biosimilar entry, net prices in the combined biosimilar-reference product market decline to roughly 63% of pre-biosimilar levels.¹

The Inflation Reduction Act (IRA) of 2022 authorized the Department of Health and Human Services (HHS) to negotiate drug prices for select high-expenditure Medicare drugs, with the first negotiated prices taking effect in January 2026.² In August 2024, CMS announced the negotiated prices for first round of drugs, reporting list-price discounts ranging from 38% to 79%³ and net price discounts from 8% to 42%.⁴

To preserve market-based incentives, the IRA includes a “special rule” allowing HHS to defer negotiation for biologics expected to face biosimilar competition within two years. Eligibility requires evidence of a licensed or pending biosimilar and a credible expectation of market entry within the designated timeframe (Appendix 1 summarizes key IRA provisions relevant to biosimilars).

Despite this safeguard, it remains unclear how this special rule will affect Medicare savings. Negotiations may lead to low immediate prices before biosimilar entry but simultaneously weaken biosimilar manufacturers’ incentives to launch (ie, ability to recoup development and launch costs) and thus dampen future biosimilar investment.

This study addresses that policy tension by evaluating the IRA's implications for Medicare savings and biosimilar incentives. In a set of hypothetical scenarios using real-world pricing trends, we compare historical price reductions following biosimilar launch with the negotiated price levels for the first IRA drug list. Ustekinumab (Stelara) was included because, at the time of selection, it met all eligibility criteria and had no biosimilar yet marketed in the United States, although multiple ustekinumab biosimilars did indeed launch in early 2025. We modeled an alternative scenario in which ustekinumab was removed from the negotiation list and replaced with palbociclib, (Ibrance), which was predicted to be among eligible drugs for the first round of Medicare price negotiation but not selected.⁵

Data and methods

Hypothetical scenarios

We modeled Medicare savings under three hypothetical policy scenarios to assess how IRA implementation may interact with biosimilar competition.

Scenario 1: biosimilar competition

We estimated historic price declines following biosimilar entry from 2017-2024 and applied these trends to forecast hypothetical future savings for biologics included in the first IRA negotiation round that have existing or imminent biosimilar competition, focusing on ustekinumab.

Scenario 2: current IRA implementation

We compared these estimated savings to the expected discount from the IRA's maximum fair price (MFP) for ustekinumab, assuming no biosimilar introduction.

Scenario 3: modified IRA implementation

We modeled an alternative scenario in which ustekinumab was excluded from negotiation because of near-term biosimilar entry and replaced with palbociclib, a high-expenditure, single-sourced small molecule estimated to be among eligible drugs for the first negotiation round and ultimately not selected,⁵ to evaluate the implications of selection policy for total Medicare savings.

Data sources

We obtained Average Sales Price (ASP) data for July 2017-June 2025 from the Centers for Medicare and Medicaid Services (CMS) quarterly payment files.⁶ ASP represents the average net transaction price to payers after rebates and discounts. Because ASP data capture only Part B drugs, we supplemented with net price estimates from SSR Health's Net Pricing Tool for self-administered (Part D) drugs. SSR derives net price by dividing manufacturer-reported net revenue by estimated prescription volume. Total Medicare spending came from the CMS Part D Drug Spending Dashboard.⁷

Market share data (2017-2024) came from the IQVIA Longitudinal and Adjudicated Data Set (LAAD), covering approximately 300 million patients (about 92% of retail prescriptions and 70% of mail order prescriptions). For therapeutic areas not included in LAAD, we supplemented with the IQVIA PharMetrics® Plus database, which contains deidentified, fully adjudicated claims from commercial health plans representing over 210 million enrollees.

Sample

We analyzed 10 reference biologics and 30 associated biosimilars approved and marketed as of June 30^th 2024. We excluded tbo-filgrastim (approved via a full Biologics License Application before the biosimilar pathway was established) and insulin products as we do not consider them to be specialty drugs. Appendix 2 lists the included products and marketing dates.

Methods

In Scenario 1, for each originator-biosimilar family (a biologic and all its marketed biosimilars), we calculated market-share–weighted average prices per quarter by multiplying each product's ASP by its market share within the family. We then measured the percentage change in the weighted price relative to the quarter preceding the first biosimilar launch and averaged these percentage changes across all families to estimate mean market price reductions. For self-administered drugs (adalimumab, tocilizumab), we repeated this analysis adding SSR Health net price data, extrapolating quarter-over-quarter price changes to a five-year horizon following biosimilar entry. Prices were not adjusted for inflation.

We estimated annual Medicare savings by applying these percentage reductions to gross spending on ustekinumab.

In Scenario 2 we used published estimates of IRA MFP discounts from net prices for ustekinumab (40%) and applied these to projected annual Medicare spending through 2029, assuming continuation of the 2019-2023 spending growth trend.^4,7

In Scenario 3, we modeled the potential savings if ustekinumab were replaced on the negotiation list with palbociclib, assuming a 29.6% average MFP discount (the mean among the first ten negotiated drugs).⁴ Total savings under this scenario were defined as the sum of biosimilar-driven savings for ustekinumab) and negotiated savings for palbociclib.

All analyses were performed using Microsoft Excel 2021. Because negotiated price effects and biosimilar uptake under the IRA are not yet observable, we used a comparative static approach to enable transparent, policy-relevant estimates based on historical trends and published price projections.

Results

Scenario 1: biosimilar competition

Overall, market prices declined substantially following biosimilar entry (Figure 1). Across all originator-biosimilar families, the mean market-weighted ASP fell to 64.9% of its pre-entry level three years after the first biosimilar and to 45.9% after five years. When incorporating net pricing data for part D drugs (adalimumab and tocilizumab), price declines were similar but slightly greater, reaching 58.9% at three years and 40.3% at five years post launch). Applying these historical market trends to ustekinumab, we estimated that biosimilar competition would yield gross Medicare savings of $251-$416 million in the first year after biosimilar entry, increasing to approximately $1.4 billion in year three and $2.3 billion in year five (Figure 2).

Figure 1.

Market-weighted mean originator-biosimilar net prices since first biosimilar launch, part B & part D drugs. Source: Author's analysis of market share data of 10 originator and 30 biosimilar products from IQVIA Longitudinal Access and Adjudicated Dataset (LAAD) and PharMetrics Plus database and Average Sales Price (ASP) information from the Center for Medicare and Medicaid Services (CMS). For Part D drugs, we supplemented the sample with net pricing information from SSR Health.

Figure 2.

Estimated annual Medicare savings, by scenario. Notes: In Scenario 1: Biosimilar Competition, we applied historic price declines following biosimilar entry to forecast hypothetical future savings on ustekinumab. In Scenario 2: Current IRA Implementation, we estimated expected discounts from the IRA's maximum fair price (MFP) for ustekinumab. In Scenario 3: Modified IRA Implementation, we modeled savings from excluding ustekinumab from IRA negotiation and replacing with palbociclib.

Scenario 2: current IRA implementation

Published estimates indicate that the IRA will produce an average 40% discount from net price for ustekinumab.⁴ Applying these discounts to projected Medicare spending, we estimated potential CMS savings of approximately $1.4 billion in the first year of implementation, rising to $1.5 billion in year three and $1.7 billion in year five (Figure 2).

Scenario 3: modified IRA implementation

In a hypothetical scenario where palbociclib replaces ustekinumab on the negotiation list and ustekinumab is instead subject to biosimilar competition, total estimated CMS savings would range from $976 million to $1.1 billion in the first year, increasing to $2.2-2.4 billion in year 3, and $3.2-$3.5 billion in year five (Figure 2).

Discussion

This analysis compared historical biosimilar price reductions due to market competition with the initial negotiated prices established under the IRA for the first price applicability year (IPAY) 2026. We found that both mechanisms can generate substantial savings for Medicare, though they do so at different rates and over different time horizons. While negotiated prices under the IRA may achieve greater near-term fiscal savings, our modeling suggests that the cumulative price reductions achieved through biosimilar competition may surpass those of the IRA by the third year following market entry. If biologics such as ustekinumab were sufficiently deferred from negotiation under the IRA's “special rule” to allow biosimilars to enter the market, overall Medicare savings could be even greater-combining near-term negotiation gains on other drugs with longer-term price erosion driven by competition. Indeed, in 2025 nine ustekinumab biosimilars launched with list price discounts of up to 90% lower than Stelara.⁸

These results illustrate a policy trade-off between short-term savings through brand-name drug price negotiations and savings achievable through long-term market competition via biosimilars following a drug's loss of exclusivity (LOE). Including high-expenditure biologics such as ustekinumab in the first negotiation round generates immediate budgetary relief but may weaken incentives for biosimilar development and launch if expected post-entry returns decline. In contrast, replacing such biologics with high-expenditure therapies not facing imminent LOE could preserve short-term savings while allowing market forces to drive deeper more durable reductions in biologic spending. The modeling projections suggest that negotiation and competition should be viewed as complementary mechanisms-best calibrated to allow biosimilar competition to lower prices while reserving negotiation for products lacking near-term competition.

Failing to balance short-term savings with long-term market competition carries risks beyond the lost savings identified in this study. Chief among them is a potential decline U.S. biosimilar development-a trend already seen in Europe, where despite a record number of upcoming loss-of-exclusivity events by 2030, fewer than one third of molecules (29%) have a biosimilar in development.⁹ Sustaining a robust biosimilar pipeline is essential for preserving competition, maintaining multiple supple sources, and mitigating drug shortages.¹⁰ For CMS, calibrating IRA negotiation to avoid discouraging biosimilar entry will be critical to securing durable savings and a resilient biologics market.

Several policy refinements could help improve the situation. CMS could strengthen drug-selection criteria by incorporating a clear “biosimilar readiness” review before adding biologics to the negotiation list. Such a review should explicitly consider whether a biosimilar has been approved, tentatively approved, or is credibly expected to enter the market within 1-2 years of the date of negotiation. The absence of this consideration contributed to ustekinumab's inclusion in the first negotiation despite imminent biosimilar entry. Incorporating this assessment would help ensure that drug selection appropriately reflects expected post-LOE price competition.

Greater transparency regarding negotiation timing and eligibility criteria would also improve predictability for all stakeholders. Developers have expressed uncertainty about how “imminent biosimilar entry” is defined and applied, and clearer guidance from CMS and HHS would stabilize expectations and reduce perceived regulatory risk.¹¹ Finally, CMS and the Office of the Assistant Secretary for Planning and Evaluation should monitor how negotiation affects biosimilar launch and uptake, adjusting implementation rules as evidence emerges.

Our study has limitations. Analyses were based on historical ASP and net price data that may not fully capture evolving contracting strategies, rebate structures, or payer responses following IRA implementation. Assumptions about biosimilar uptake, spending growth and prices were extrapolated from pre-IRA trends and may differ under the new policy environment. For example, our analysis did not account for the observed price declines for ustekinumab that may have resulted from the launch of its biosimilars in 2025 and may underestimate the savings with biosimilar competition. In addition, results are limited to Medicare and do not account for possible spillover effects in commercial markets or Medicaid.

Conclusion

Our results suggest that selecting biologics with imminent biosimilar competition for IRA negotiation may yield near-term savings but forego larger savings achievable through competition. As future negotiation rounds are implemented, policymakers should consider biosimilar readiness, improve transparency in deferral decisions, and support reimbursement incentives to ensure that negotiation and competition operate in alignment to sustain both short-term and long-term savings. Doing so would strengthen the long-term sustainability of the U.S. biosimilar market while preserving the IRA's goal of reducing drug costs for Medicare beneficiaries.

Supplementary material

Supplementary material is available at Health Affairs Scholar online.

Funding

This work was funded by Genentech Inc.

Conflicts of interest

James Chambers received funding for this study from Genentech, Inc. William Wong and Stella Ko are employees of Genentech, Inc and own stock in Roche Holding AG. Please see ICMJE form(s) for author conflicts of interest. These have been provided as material.

Please see ICMJE form(s) for author conflicts of interest. These have been provided as supplementary materials.