Amgen makes complex biological medicines that patients take repeatedly, often for the rest of their lives. The company sells drugs for conditions like high cholesterol, bone disease, cancer, arthritis, and rare inflammatory diseases. Patients do not buy these medicines at a pharmacy counter the way they would a cold tablet. Instead, doctors prescribe them, hospitals and clinics administer many of them, and insurance companies or government health programs pay most of the bill. That chain, from laboratory to patient to payer, is how Amgen turns science into revenue. The diagram below traces where the money goes.

Five years of financial data tell a story of growth interrupted by a very large, very deliberate bet. From 2021 to 2022, Amgen was a steady machine: revenue held at roughly $26 billion, gross margin sat above 75%, and free cash flow ran between $8 billion and $9 billion. Then in October 2023, Amgen completed the acquisition of Horizon Therapeutics, adding TEPEZZA for thyroid eye disease and KRYSTEXXA for chronic gout to its medicine cabinet. The purchase price transformed the balance sheet overnight.

That debt load more than doubled from $25.4 billion in 2021 to $55.1 billion in 2023. At the same time, gross margin dropped from above 75% to 70% in 2023 and then to 61.5% in 2024, weighed down by the cost of absorbing Horizon's inventory at acquisition-date prices and by rising amortization charges. By 2025, gross margin had partially recovered to 67.2%, cost of sales had fallen as a share of revenue, and Amgen retired $6.0 billion of debt, bringing net debt to $50.1 billion. Revenue, meanwhile, climbed from $26.0 billion in 2021 to $36.8 billion in 2025, driven by volume growth of 13% in 2025 alone. The trajectory shows a business expanding its top line while slowly paying down the cost of expansion.

Inside that top-line growth, the picture is uneven. Repatha, the cholesterol drug, grew 36% in both 2024 and 2025, reaching $3.0 billion in sales. EVENITY for bone disease grew 34% in 2025. BLINCYTO for blood cancer grew 28%. TEZSPIRE for severe asthma grew 52%. These are the engines pulling the business forward. But ENBREL, once one of Amgen's biggest earners, fell 33% in 2025 to $2.2 billion, hammered by government pricing rules and discount programs. Prolia, the top-selling product at $4.4 billion, lost its US patent protection in February 2025, and multiple cheaper copies have already launched. XGEVA lost the same patent and faces the same pressure. The company is running two races at once: growing new products fast enough to cover the erosion of old ones.

Research spending rose sharply too. Amgen spent $7.3 billion on research and development in 2025, up from $4.8 billion in 2023. Most of that increase went into later-stage clinical programs, including six Phase 3 studies for MariTide, a monthly weight-loss injection that works differently from existing obesity drugs. MariTide is the most expensive unproven bet in the pipeline. If it clears Phase 3 and wins approval, it could enter one of the largest new drug markets in decades. If it does not, the research spending disappears without a commercial return.

Free cash flow tells its own story. It ran at $8.4 billion in 2021, dipped to $7.4 billion in 2023 as acquisition costs hit, recovered to $10.4 billion in 2024, and then fell back to $8.1 billion in 2025 as research spending accelerated. The business generates real cash, but that cash is being pulled in several directions at once: debt repayment, capital spending on manufacturing, dividends, and a growing research bill.

The documented risks are specific and serious. The US government has already set lower Medicare prices for ENBREL starting in 2026 and for Otezla starting in 2027. Up to 100 drugs could face government-set prices by 2031. Colorado has already declared ENBREL unaffordable under state pricing rules and set a price cap effective January 1, 2027, with Washington state reviewing it too. In 2025, a jury ordered Amgen to pay $406 million to Regeneron after finding that Amgen used unfair practices to boost Repatha sales at the expense of Regeneron's competing drug Praluent. The IRS is disputing how Amgen allocated profits between the US and Puerto Rico for 2010 through 2015, and a tax court decision is expected no earlier than mid-2026. Amgen's commercial manufacturing is heavily concentrated at one facility in Puerto Rico, meaning a single serious disruption there could limit drug supply across the business.

Prolia's patent expired in early 2025 and multiple biosimilar copies have already launched. XGEVA lost the same protection at the same time. ENBREL faces biosimilar competitors in Canada now, and US biosimilars have been approved even if not yet launched. Otezla's core composition patent expires in 2028. The Repatha antibody patent expires in the US in 2029. Several of the products that currently drive the most revenue will face intensifying competition well before the end of the decade.