IBD 2026: Merck TL1A Phase 3 Win, Spyre at $106

Inflammatory bowel disease is the densest lane in immunology, and the trade is not what most people think. Five mechanisms fight over one $20 billion-plus market: TL1A, IL-23, JAK, S1P, and anti-integrin. Everyone already owns the IL-23 incumbents. The money over the next three years is made on the TL1A swing, and the only way to actually buy that swing without a rounding-error position in a mega-cap is a pure play trading at $106. The reason this lane is so hard to trade is structural. Three of the five mechanisms are owned by three pharma giants, AbbVie, J&J, and BMS, each a $129 billion to $652 billion company. A drug that captures 5% of a $20 billion market is a $1 billion franchise, which moves none of them. The share-shift math, not the single-binary trade, is the actual edge here.

The TL1A war is the headline, and Merck just took the lead. Tulisokibart, the anti-TL1A antibody Merck bought through its $10.8 billion Prometheus acquisition, met its primary endpoint of clinical remission per the Modified Mayo Score at week 12, plus its key secondary endpoints, in the Phase 3 ATLAS-UC induction study on June 22, 2026, the first anti-TL1A biologic to hit Phase 3 remission in ulcerative colitis. Merck did not disclose the specific remission rate or p-value, saving the numbers for a scientific congress. The Phase 2 signal is the published reference point: 26% clinical remission versus 1% for placebo, a 25-point difference (p<0.001), in a refractory population, per the NEJM Phase 2 publication. Merck is running tulisokibart across seven indications, including the Phase 3 ARES-CD trial in Crohn’s disease with initial results toward the end of 2028. The catch is the company behind it. Merck trades at $150.66, a $376 billion market cap, and is fighting a KEYTRUDA patent cliff. A first-in-class IBD drug is a real pipeline signal, but it is a rounding error to a $376 billion enterprise until it proves it can clear $1 billion in annual sales.

Roche is the best-in-class challenger, and here is where the briefs get it wrong. The second TL1A asset is not “Roche/Teva RVT-3101.” It is Roche’s afimkibart (RVT-3101), bought through the $7.1 billion Telavant acquisition from Roivant in 2023. Phase 2b data showed 36% clinical remission and 50% endoscopic improvement at week 56, a maintenance result that edges Merck’s induction numbers on durability. Roche is running afimkibart in Phase 3. The separate TL1A program that involves Teva is duvakitug, partnered with Sanofi. Three TL1A entrants, not two, all targeting the same immuno-fibrosis mechanism.

The pure play is Spyre Therapeutics. Spyre’s SPY002, a half-life-extended anti-TL1A antibody, cleared a Phase 2 ulcerative colitis readout that roughly doubled the stock since March, and Spyre now trades at $106.65. This is the only way to isolate the TL1A thesis. Every other TL1A program is buried inside a company where the drug cannot move the share price, which is exactly why the immunology shelf ranking treats mechanism wars as share-shift math rather than single-binary trades.

The IL-23 incumbents are the revenue reality, and they are not losing yet. AbbVie’s Skyrizi (risankizumab) did $5.505 billion in Q2 2026, up 24.4%, overtaking Humira for the first time, and AbbVie has guided the drug to a $20 billion peak, now raised toward $21.7 billion. Rinvoq (upadacitinib), the JAK option, did $2.525 billion, up 24.5%. Together that is an $8 billion quarterly immunology engine, detailed in the AbbVie Q2 analysis where Skyrizi overtook Humira. Johnson & Johnson’s Tremfya (guselkumab) grew 72.5% in Q2 as Stelara collapsed 55% to $740 million on biosimilar erosion, which means the IL-23 lane is consolidating share toward Skyrizi and Tremfya even before TL1A arrives. Skyrizi is running a head-to-head ulcerative colitis trial against Takeda’s Entyvio, the anti-integrin standard, and an SC induction filing for Crohn’s is pending at the FDA. The IL-23 moat is deep, and it buys AbbVie and J&J years of runway.

The JAK lane is the oral convenience trade with a safety asterisk. Rinvoq is the dominant oral option across UC and Crohn’s, but its class-wide boxed warning on serious infections, malignancy, and cardiovascular events keeps it behind the biologics in the treatment algorithm. That safety label is precisely why J&J’s oral IL-23 pill icotrokinra, a psoriasis CHMP story already on this site, matters: it promises IL-23 efficacy in a pill without the JAK black box. If icotrokinra lands in IBD, it competes directly with Rinvoq for the oral lane.

The S1P lane is the quiet fourth mechanism. Bristol Myers Squibb’s Zeposia (ozanimod) is the approved oral S1P modulator for UC, trading convenience for a first-dose cardiac monitoring requirement. BMS sits at $67.28, a $128.9 billion market cap, and Zeposia is a small piece of a portfolio trying to replace Revlimid and Eliquis. This lane does not win share; it harvests patients who want an oral and cannot take a JAK. The anti-integrin lane is similarly defensive: Takeda’s Entyvio remains the entrenched subcutaneous standard, which is exactly why AbbVie is running Skyrizi head-to-head against it rather than against the S1P or JAK options. That head-to-head is the IL-23 lane’s attempt to pull the anti-integrin franchise onto its own mechanism, and its result decides whether Skyrizi can displace Entyvio.

The cautionary read is Sanofi. The Sanofi Q2 earnings analysis showed a company cutting its immunology pipeline (amlitelimab, itpekimab) while leaning on Dupixent, and its amlitelimab discontinuation is the reminder that a mechanism is not a drug. Sanofi is now a duvakitug partner rather than an IBD leader, which means its IBD exposure is a single TL1A bet buried in a $102.8 billion company. The argenx’s Forte acquisition and its FcRn franchise context shows the same pattern: the value migrates to whoever can prove a mechanism with a registrational win, not whoever licenses it first.

The risks are specific. Merck’s tulisokibart still has to show maintenance and endoscopic differentiation against the IL-23 incumbents; Citi analysts noted after the ATLAS-UC readout that tulisokibart takes a back seat to IL-23s and JAKs unless it shows maintenance, endoscopic, histologic-endoscopic, or biomarker differentiation. Spyre is a Phase 2 asset with a market cap that has already doubled on the readout, so the next data print is a binary that can halve it. Roche’s afimkibart has to convert Phase 2b durability into a Phase 3 win, and Roche is not a pure play. The TL1A class could also stumble on fibrosis or infection signals that none of the Phase 2 data has yet revealed. And the incumbents are not standing still: AbbVie’s SC induction filing for Skyrizi in Crohn’s and the Entyvio head-to-head are the near-term label-expansion events that could widen the IL-23 moat before TL1A ever reaches the market.

Here is the call. The IL-23 incumbents keep winning share for the next two years, because tulisokibart’s Crohn’s data does not land until late 2028. But the mechanism that takes the most share by 2030 is TL1A, and Merck is ahead. The problem for investors is that none of Merck, Roche, AbbVie, J&J, or BMS can be owned for this thesis: a single IBD drug cannot move a $376 billion to $652 billion stock. That leaves Spyre as the only tradable TL1A swing, and it is a binary, not a core holding. Own the IL-23 incumbents for revenue, and treat Spyre as a 1% to 2% option on the mechanism war. The real money in IBD over three years is made by whoever proves TL1A induces remission and keeps it, and Merck just took the first step.