Anti-IL-23
The dual-benefit claims analysis
$47.3B in 2025 gross claims. One molecule split three ways. The first oral IL-23 agent. And a class that moved from dermatology to gastroenterology without anyone calling it.
Press arrow keys, space, or Enter to advance
IL-23 is a signalling protein that drives inflammation in psoriasis, psoriatic arthritis and inflammatory bowel disease; blocking it is how this class works. It is built from two subunits, and which one a drug targets defines the generation. p40 agents bind a subunit IL-23 shares with IL-12, so they switch off both pathways — that is Stelara and its biosimilars. p19 agents bind the subunit unique to IL-23, hitting one pathway instead of two — Skyrizi, Tremfya, Ilumya, Omvoh.
2026 is excluded from year-over-year comparisons throughout: the panel is capped at 2026-06-30, so it is a half year.
Stelara now competes against its own authorized generic and eight ustekinumab biosimilars. The p40 line on this chart is all three combined.
Skyrizi, Tremfya, Ilumya and Omvoh are four distinct launches with different indications and different routes.
What began as a dermatology class is now majority inflammatory-bowel-disease by diagnosed patients — see the indication slide.
p19 agents accounted for 74.2% of all class events in 2025, across both benefits.
Stelara, Janssen's own authorized generic and 8 FDA-approved biosimilars all deliver the same molecule — ustekinumab. Shown apart, the brand still holds 22.6% of class claims while biosimilars have taken 3.5% in their first full year. Added back together the molecule still trails Skyrizi by 405K pharmacy claims. Entry is more concentrated than the count suggests: two products carry 78.4% of biosimilar volume.
The signal: Biosimilars crossed the Stelara brand on monthly pharmacy claims in April 2026 — sixteen months after the first US biosimilar fill.
Median gap between consecutive fills for the same patient. No dosing schedule was supplied to the query — the regimens below were recovered from refill timing alone, and 5 of 7 agents land their most common gap exactly on the label interval.
The signal: the median runs slightly under label because refill gaps measure acquisition — patients collect before running out. The mode is the sharper read, and it lands on 28, 56 and 84 days: exact 4-, 8- and 12-week multiples, recovered from claim timing alone. Icotyde is excluded — a daily oral has no comparable interval, and with 890 intervals across four months its modal gap reads as launch noise.
2025, full year. Counted per diagnosis, not per patient — one patient can carry more than one, so shares are of recorded diagnoses.
Stelara crossed in 2019: by 2025, 79.9% of its recorded diagnoses were Crohn’s or colitis rather than psoriasis. Skyrizi and Tremfya are climbing but have not crossed — 47.9% and 32.1%. Skyrizi’s Crohn’s population went from 7 patients in 2019 to 33,708. Biosimilars inherit the skew at 79.7%.
These counts cover patients with at least one medical claim carrying a relevant diagnosis — roughly 40% of the treated base. Read the shares as shares of the diagnosed subset.
Anti-VEGF is a retina story and SGLT-2 a primary-care one. IL-23 splits across three specialties treating three different diseases with the same molecules — and gastroenterology now administers 48.9% of medical-benefit volume against dermatology's 6.1%, the clearest single measure of where this class has moved.
The signal: a class with three prescriber populations has three formulary fights, three sales forces and three competitive sets. Read it as one dermatology market and the gastroenterology volume — the larger and faster-growing half — is invisible. Medical benefit only: pharmacy claims do not carry a provider specialty, so they are excluded here rather than shown as a spurious category.
Where the class’s intravenous administrations actually happen, 2025 full year. Covers the 5 agents with intravenous volume: Omvoh, Skyrizi, Stelara, Tremfya, Ustekinumab biosimilar.
A meaningful share of rows resolve to Other — the place-of-service field is a proprietary numeric scheme and is sparsely populated. Treat the office share as a floor.
Pharmacy claims by payment type, 2025 full year. All 7 agents in the class: Ilumya, Omvoh, Skyrizi, Stelara, Tremfya, Ustekinumab authorized generic, Ustekinumab biosimilar.
75.2% of 2025 pharmacy claims are commercial — a higher concentration than most specialty classes, and the reason formulary decisions move this market so quickly.
The signal: Share is computed on claim volume, not spend.
65.6% of Ilumya patients are 65 or older — by far the sharpest age skew in the class, and consistent with its Medicare-dominant medical-benefit payer mix.
The 35–64 bands carry most of the class — psoriatic disease and inflammatory bowel disease both present in mid-life, and the treated population sits well below Medicare age outside the Ilumya exception.
Every patient in the panel, not a single year — so the newest agents are represented and the counts are lifetime rather than annual. They will therefore run higher than the calendar-year slides. Patients under 18 are excluded from the age chart, and records with no recorded sex are excluded from the split.
Masked and null state codes are excluded from the map rather than shown as a state.
Every point below pairs one year of IMX pharmacy claims with the US net sales the manufacturer filed with the SEC for that same year. Nothing is fitted to make them agree — the claims come from the panel, the dollars from the 10-K.
Plotted: Skyrizi, the class's fastest-growing franchise. Reported sales are an external benchmark from public filings, not an IMX figure. Stelara is scored to 2024 because its biosimilars reset net price in 2025 — that shift is the next slide.
R² is the share of the year-to-year variation in reported sales that claim volume alone explains. 1.000 would be a perfect straight line.
The signal: across three franchises and seven years, claim volume tracks the filed top line almost exactly. The panel is not a proxy for the market — it is a monthly read on the same quantity the annual report confirms twelve months later.
The through-line: growth is patients, not intensity — patients up ~27%/yr while events per patient held flat at ~4.0. At that rate the class reaches ~6.4M events by 2030 (dashed). Growth has decelerated from ~40%/yr since 2016, so read it as a path, not a floor.
IMX Data delivers the claim-level depth behind the p40-to-p19 handover.
Every agent · Every benefit · Every patient.