Halozyme Therapeutics, Inc. HALO

Key data

The business

Halozyme sells one thing and rents it out: an enzyme called rHuPH20. Injected alongside a large biologic drug, it briefly melts the mesh of hyaluronan under the skin so a dose that would otherwise need a long intravenous drip can be given as a shot in minutes, and the tissue knits back within a day (mechanism per Locke et al., Drug Delivery, 2019). Drugmakers license the enzyme, brand it ENHANZE, and pay Halozyme milestones plus a royalty on every subcutaneous dose sold. The customer list is the point: Johnson & Johnson's Darzalex, Roche's Phesgo, argenx's Vyvgart, and now Merck's subcutaneous Keytruda all ride on it. Because a royalty carries almost no cost, most of each new dollar drops to operating profit, which ran at 57% of sales over the past year. The moat is a thick wall of formulation patents and a decade of partner switching costs: once a $10 billion drug is reformulated, approved, and marketed as a shot, nobody re-opens that file. A small products arm (Hylenex, the Elektrofi auto-injector platform bought in late 2025) rounds it out but does not drive the model.

The numbers

The story is acceleration, with one ugly quarter to understand before trusting the rest.

Revenue reaccelerated in 2026: Q2 sales of $481.0M rose 48% from a year earlier, the biggest quarter ever, and net income of $229.9M rose 39%. The Q4 2025 loss was not the business breaking. It was a one-time write-off tied to the Elektrofi acquisition, and it is derived here (full year less nine months) because the vendor feed skipped that quarter. Strip it out and the line runs straight up. Q2 adjusted earnings of $2.28 landed above the $1.79 analysts modeled, the third upside print in four quarters.

Over five years revenue tripled, while earnings stayed bumpy because acquisitions and buybacks move the bottom line around. The compounding lever is simple: each partner drug that converts from IV to a shot adds a royalty stream that grows with that drug's own sales and lasts as long as the formulation patents hold. Merck's subcutaneous Keytruda is the giant in that pipeline. Merck is converting a roughly $30 billion franchise to the shot ahead of its 2028 patent loss on the IV version, and Halozyme earns on every subcutaneous vial. That single conversion is why analysts, though only three carry a 2028 line, model earnings more than tripling to about $11 to $12 a share, which is what turns 32 times this year's profit into single digits on their 2028 estimate. What the market has not settled is whether that conversion lands fast or slow. The print that decides it is Merck's subcutaneous Keytruda sales over the next four quarters.

Management

Helen Torley has run Halozyme since 2014 and built the royalty model deliberately. The capital record is aggressive on both ends. The company bought back stock every year it could, $342M in 2025 and over $1.5B across five years, shrinking the share count while earnings grew. To fund Elektrofi and keep buying, long-term debt jumped from $1.51B in September 2025 to $2.14B by year-end, pushing net debt to about 2.8 times EBITDA, still comfortable against 57% operating margins but no longer a fortress. Insiders only sell: zero purchases in the last year against $17.5M sold across 46 transactions, including three sales by Torley between April and August. Plan status is not disclosed in the data here, so read the selling as routine compensation rather than a signal until a Form 4 footnote says otherwise. Pay is tied to revenue and royalty growth, the right thing to reward for this model.

How it fails or surprises you

The patent clock (downside). Royalties live and die on how long formulation patents block copycat subcutaneous versions. The MDASE dispute and a May 2026 PTAB ruling have circled the core estate. It has not cracked results, Q2 was a record, so the overhang is still an overhang and not a wound. An adverse final ruling shortening protection on Darzalex or Keytruda subq would reprice the whole royalty tail. Watch the 10-K expiry table and any PTAB appeal.

Too few drugs (downside). A handful of partner products carry most of the royalty. If Darzalex growth slows or a partner's drug stumbles, there is no broad base to cushion it. The tell is the royalty line by partner in the 10-K and J&J's Darzalex sales each quarter.

Keytruda goes subcutaneous fast (right tail). Merck has every reason to convert its $30 billion franchise to the shot quickly, before its 2028 IV cliff, and a five-minute injection beats an hour-long infusion for doctors and patients. If conversion runs faster than the cautious pace priced in, Halozyme's royalty steps up years earlier than modeled. The first tell is Merck's subcutaneous Keytruda sales in its next two quarterly reports.

Closing thoughts

At 32 times trailing earnings and mid its range since 2017, Halozyme is priced as a good royalty compounder with a known patent overhang, roughly fair on what is visible today. The edge, if there is one, sits in the Keytruda subcutaneous ramp: Merck's own reported subq sales convert directly into Halozyme's royalty, so a fast ramp is upside the mid-range multiple does not yet pay for, and a stalled one is the disappointment. The right tail looks the fatter one here, because the patent fight would take years and appeals to actually shorten the tail, while the Keytruda conversion is already approved and only a question of speed. What is genuinely at risk on the downside is the durability of the royalty life, not this year's cash.

The bet is still that drugmakers keep moving their biggest medicines to a shot under the skin, Merck's Keytruda first, and that Halozyme keeps collecting on every dose. It breaks if the patents that protect those royalties are cut short before the new launches mature. The one pair of numbers that tells you first: Merck's subcutaneous Keytruda sales climbing while Halozyme's patent-expiry dates in the next 10-K hold their ground.