"A Pill With the Efficacy of a Biologic"
That quote belongs to Dr. Melinda Gooderham, the dermatologist who presented Takeda's Phase 3 psoriasis data at the American Academy of Dermatology meeting in Denver this March.4 It is a sentence dermatologists have been waiting roughly twenty years to say out loud. Since the first TNF blockers arrived, the deal for moderate-to-severe plaque psoriasis has been simple and slightly humiliating: if you want clear skin, you learn to inject yourself. The pills, methotrexate, apremilast, and more recently deucravacitinib, have always been the compromise you accept when the injections are unavailable, unaffordable, or unwanted.
Zasocitinib, Takeda's once-daily oral TYK2 inhibitor, is the drug that is supposed to end that compromise. Takeda paid $4 billion up front for it, before a single Phase 3 patient had been dosed.9 It has now completed two 700-to-1,100-patient pivotal trials, a head-to-head against the only approved drug in its class, and produced week-16 complete-clearance rates that sit within shouting distance of injectable IL-23 blockers.5,6,7 Takeda is forecasting $3 to $6 billion in peak annual sales and calling it one of three launches that will replace the revenue it is losing on Entyvio and Vyvanse.5
Here is what makes this issue different from the usual "new drug, new hype" story. The numbers are, as far as I can tell, real and large. And yet, as of today, not one patient-level Phase 3 result has been peer-reviewed, the head-to-head exists only as a press release that withholds the comparator's actual number, and the safety database is roughly one-fifth the size and one-tenth the duration of what it took to learn the hard lessons of the last oral immunology class.4,7,10 So I want to do two things at once: give the efficacy data the credit it has earned, and be precise about how much of this story is still Takeda's word.
One Kinase Out of Four, Bound in the Wrong Place on Purpose
Psoriasis is, at its core, an IL-23/IL-17 disease. Dendritic cells in the skin release interleukin-23, which tells a population of T cells to pump out interleukin-17, which tells keratinocytes to proliferate, recruit neutrophils, and build the thick silvery plaques you see in clinic. The injectable biologics that dominate the market (risankizumab, guselkumab, bimekizumab, secukinumab) work by physically mopping up IL-23 or IL-17 in the bloodstream.14
The oral approach goes one step downstream. When IL-23 lands on its receptor, the signal is carried into the cell by a pair of enzymes from the Janus kinase family. There are four members: JAK1, JAK2, JAK3, and TYK2. IL-23 specifically depends on TYK2 paired with JAK2. So do IL-12 and the type I interferons. Almost everything else, the erythropoietin that makes your red blood cells, the IL-2 that runs your T-cell homeostasis, the cytokines that manage your lipids and your platelets, runs through JAK1, JAK2, and JAK3 without needing TYK2 at all.2
That distribution of labor is the whole thesis. The first-generation JAK inhibitors, tofacitinib, baricitinib, upadacitinib, bind the ATP pocket that all four kinases share, and inevitably hit several at once. In 2022, the ORAL Surveillance trial in 4,362 rheumatoid arthritis patients found tofacitinib carried a hazard ratio of 1.33 for major cardiovascular events and 1.48 for cancer versus TNF inhibitors, and the FDA put a boxed warning on the entire class.10 Every oral immunology drug since has lived in that shadow.
Deucravacitinib, approved as Sotyktu in September 2022, showed the way out. Instead of the shared catalytic site, it binds a regulatory "pseudokinase" domain, called JH2, that is unique enough to TYK2 that the drug leaves the other three JAKs largely alone. The reward was a label with no boxed warning, the first oral immunomodulator in the class to get one.12
Zasocitinib uses the same allosteric trick with a much tighter grip. Takeda's own pharmacology paper, published in the Journal of Investigative Dermatology this year, reports a binding constant for the TYK2 JH2 domain of 0.0087 nanomolar, selectivity over JAK1 that the authors describe as greater than one-million-fold, and no measurable inhibition of JAK1, JAK2, or JAK3 at all. Simulated 30 mg once-daily dosing keeps the drug above its TYK2 inhibitory threshold for a full 24 hours.2 Deucravacitinib's published selectivity, for comparison, is in the range of 100-fold over JAK1 and 2,000-fold over JAK2/3.12
I want to flag that these are test-tube numbers from the sponsor. A million-fold selectivity in a binding assay is a plausible reason to expect a clean drug, not evidence of one. But the strategic logic is sound: if you are more selective, you can inhibit TYK2 harder without picking up JAK-class toxicity, and if you inhibit TYK2 harder, you should clear more skin. Whether the second half of that sentence is true is what the LATITUDE program was built to answer.
Designed by Physics, Bought by Takeda, Proven in 259 Patients
Zasocitinib started life as NDI-034858 at Nimbus Therapeutics, a Boston biotech that designs molecules computationally using Schrödinger's physics-based modeling platform, which is why you will occasionally see it called the "AI-designed psoriasis pill." Nimbus ran the Phase 2b trial. Takeda announced it was buying the molecule in December 2022 for $4 billion up front plus up to $2 billion in sales milestones, closed the deal in February 2023, and renamed it TAK-279, then zasocitinib.9
Design. 55 sites in the US and Canada, adults with moderate-to-severe plaque psoriasis randomized 1:1:1:1:1 to zasocitinib 2, 5, 15, or 30 mg once daily, or placebo, for 12 weeks. Primary endpoint PASI 75 at week 12.1
Results: PASI 75 of 18% (2 mg), 44% (5 mg), 68% (15 mg), and 67% (30 mg) versus 6% on placebo. Complete clearance (PASI 100) at 30 mg: 33%. Adverse events were 44% on placebo and 53 to 62% across active arms with no dose relationship; no clinically meaningful laboratory changes.
Limitation: Small arms (about 50 patients each), 12 weeks, North America only, fewer than 8% Black participants. Nimbus and Takeda employees are co-authors. The 15 and 30 mg doses were indistinguishable on PASI 75, which raises the question of whether 30 mg was the right dose to take forward or simply the most aggressive one.
The Phase 2b result is the only zasocitinib psoriasis efficacy data that has been through peer review, and it is a good study. The dose-response is clean, the placebo rate is where psoriasis placebo rates live, and a 33% complete-clearance rate at 12 weeks from a pill was, in 2024, not a number anyone had seen before. A separate Phase 2b in psoriatic arthritis, published in Annals of the Rheumatic Diseases in 2025, found ACR20 responses of 54% versus 29% on placebo, a solid but not spectacular joint signal.3
1,801 Patients, Two Trials, One Very Low Bar
LATITUDE PsO-3001 and PsO-3002 were the pivotal trials. Both were randomized, double-blind, and ran across 21 countries. Both compared zasocitinib 30 mg once daily against placebo and against apremilast 30 mg twice daily, in roughly a 3:1:1 ratio. The co-primary endpoints, measured at week 16 against placebo, were the standard pair: a static Physician's Global Assessment score of clear or almost clear (sPGA 0/1) and a 75% improvement in the Psoriasis Area and Severity Index (PASI 75). Trial 3001 enrolled 693 patients and ran 52 weeks; trial 3002 enrolled 1,108 and ran 60 weeks with a randomized withdrawal at week 40. The population was typical: mean PASI around 20, about a quarter of body surface area involved, mean BMI 30, and about 30% had previously been on a biologic.4,5,6
Week 16, zasocitinib vs apremilast vs placebo (3001 / 3002). PASI 75: 75.7% / 71.4% vs 37.2% / 33.0% vs 12.1% / 12.3%. sPGA 0/1: 71.4% / 69.2% vs 32.1% / 29.7% vs 10.7% / 12.6%. PASI 90: 61.3% / 51.9% vs 16.8% / 15.9% vs 5.0% / 4.0%. PASI 100: 33.4% / 25.2% vs 2.9% / 4.3% vs 0.7% / 1.1%. All p<0.001 against both comparators.5,6
Week 24: PASI 100 rose to 42% in 3001 and 32% in 3002; PASI 90 to 69% and 63%. Onset was fast: PASI 75 at week 4 was 16.8% versus 4.3% on placebo in 3002.5
Limitation: Presented at a congress and in an SEC-filed investor deck; not peer-reviewed. Apremilast is a weak active comparator (PASI 75 around 35%), so the margin against it is expected. Takeda-sponsored, Takeda-analyzed, presented by investigators who disclose Takeda fees.
Let me put those numbers in context, because context is what turns a press release into a judgment. A PASI 75 of 71 to 76% at week 16 is roughly what the IL-23 biologics deliver at the same timepoint. A PASI 100 of 25 to 33%, climbing to 32 to 42% by week 24, is in the range that risankizumab reported at week 16 in its own pivotal trials, and it is roughly double what deucravacitinib achieved in POETYK.11,14 For a pill, that is not incremental. That is a different category.
Two pivotal trials, a head-to-head, and a Phase 2b in JAMA Dermatology. Only the last one has been through review.1,4,7
The durability data deserve a mention, with a caveat. In trial 3002, patients who had responded by week 40 were re-randomized to keep taking the drug or switch to placebo. Of those who stayed on zasocitinib, more than 90% held their sPGA 0/1, PASI 75, and PASI 90 responses through week 60. Of those switched to placebo, 59%, 69%, and 52% respectively still held those responses five months later.5 The first number is reassuring. The second is the more interesting one: it tells you that relapse after stopping is slow, which is a feature of the whole TYK2 class and matters for anyone who misses doses. But note that this is a responder-enriched analysis. It tells you how well the drug holds a response it has already produced, not how the whole starting population fared over a year.
Takeda also released data from the hard-to-treat sites in July. Among patients with scalp involvement, 74 to 77% reached a clear-or-almost-clear scalp score at week 16, versus 7 to 13% on placebo and 30 to 42% on apremilast. For palms and soles, 69 to 71% cleared versus 10 to 22% on placebo. Nail improvement was statistically significant against placebo, but no absolute numbers were released.8 The palmoplantar result was described as "numerically higher" against apremilast, which is trial-speak for "not formally tested."
The Trial That Matters Most Is the One We Know Least About
Beating apremilast proves you belong in the conversation. Beating deucravacitinib proves you belong at the front of it. That is why LATITUDE Atlas, the 606-patient head-to-head against Sotyktu, is the single most important trial in this program, and why it is frustrating that it currently rests on a June 11 press release.7
Design. Randomized, double-blind, 8 countries, zasocitinib 30 mg once daily versus deucravacitinib 6 mg once daily for 16 weeks. Primary endpoint: PASI 100 at week 16. Principal investigator Linda Stein Gold, Henry Ford Health.7
Results: "More than 35%" of zasocitinib patients reached PASI 100 versus a rate Takeda describes only as "more than 2.5 times" lower, which trade press has back-calculated to roughly 14%. Zasocitinib was also superior on PASI 90 and sPGA 0, with curves separating from week 8.
Limitation: No absolute comparator rate, no confidence intervals, no p-values, no safety data disclosed. Takeda "intends to present detailed data at upcoming medical congresses." The deucravacitinib dose is its approved dose, but it was chosen by Bristol Myers Squibb partly for safety margin, so "more TYK2 inhibition beats less TYK2 inhibition" is not surprising; the question it moves is whether the safety margin came with it.
I believe this result. The direction is consistent with the Phase 2b, consistent with both pivotal trials, and consistent with the mechanism. A 2.5-fold margin on complete clearance is not the kind of thing that evaporates in the full dataset. But "I believe it" is not the standard this newsletter runs on, and a company that has $4 billion sunk into a molecule and has already used the word "outperforms" in a headline has every incentive to release the flattering ratio and hold the inconvenient absolute numbers for later. When the full Atlas data are published, this section will get rewritten. Until then, it is a very persuasive rumor.
A 2.5-fold margin on complete clearance is not the kind of thing that evaporates in the full dataset. But "I believe it" is not the standard this newsletter runs on.
On the LATITUDE Atlas press releaseClean So Far, With One Ironic Exception
The pooled safety data from the two pivotal trials cover 970 zasocitinib patients, 412 on apremilast, and 417 on placebo through week 16, with a smaller set through week 24. Any adverse event: 62.1% on zasocitinib, 50.5% on apremilast, 46.9% on placebo. Discontinuation because of an adverse event: 3.2%, 2.6%, and under 1%. Serious adverse events: 3.0%, 1.5%, and under 1%. There was one death in the zasocitinib arm, the day after the first dose, judged unrelated.5
The individual events are what you would predict from switching off IL-23 and interferon signaling. Upper respiratory infections ran 10.1% versus 3.2% on placebo. Nasopharyngitis was 6.2%. And then there is the one that made me smile: acne, in 6.5% of zasocitinib patients through week 16 and 7.3% through week 24, versus under 1% on both comparators.5 Acne is a known effect of TYK2 inhibition; deucravacitinib produces it in 2 to 3% of patients in its own trials.12 Zasocitinib's rate is roughly double that, which is exactly what you would expect from hitting the same target harder. It is a mild, treatable side effect. It is also, in a drug whose entire purpose is to make your skin look better, a genuinely funny one.
What Takeda has said about the labs is reassuring but thin: lymphocytes, liver enzymes, and lipids showed "no clinically meaningful trends," and there were "no observed trends" toward the cholesterol elevations that dog the JAK class.5 The Phase 2b lab analysis found no dose-dependent movement in creatine kinase, blood counts, or hepatic or renal markers, which matters because deucravacitinib's label carries warnings for CPK elevation and rhabdomyolysis.1,12
The rare events are undisclosed
Herpes zoster, serious infections, cardiovascular events, blood clots, and malignancies have not been reported for Phase 3. Those are precisely the events that defined the JAK-class boxed warning, and ~970 patients over 16 to 24 weeks cannot detect them anyway.
Serious adverse events doubled
3.0% on zasocitinib versus 1.5% on apremilast and under 1% on placebo. Small absolute numbers, not characterized, possibly noise. But it is the one line in the safety table that goes the wrong direction and has not been explained.
No lipid or CPK signal so far
The metabolic fingerprints of broad JAK inhibition have not appeared. If that holds through the three-year extension study, it is the strongest argument that the selectivity story is real.
Acne in a skin drug
6.5 to 7.3% versus under 1% on comparators, about double deucravacitinib's rate. Manageable, but patients should be told before their psoriasis clears and their forehead breaks out.
The comparison that should worry a careful reader is with the deucravacitinib long-term data. The four-year POETYK extension covered 1,519 patients and 4,393 patient-years, and reported exposure-adjusted rates per 100 patient-years of 0.55 for zoster, 0.89 for malignancy, 0.32 for major cardiovascular events, and 0.07 for venous thromboembolism.12 Those numbers are what allowed Sotyktu to keep its clean label. Zasocitinib's longest disclosed exposure is 60 weeks in a subset of a few hundred patients. Its three-year extension study is enrolling.5 This is not a criticism of the drug; it is a statement about time. But anyone describing zasocitinib as "as safe as Sotyktu" is currently describing a hope.
The Oral Psoriasis Market Just Got Very Crowded, Very Fast
When Takeda bought this molecule in 2022, the oral landscape was apremilast and a freshly launched Sotyktu. It is not that landscape anymore. Johnson & Johnson's icotrokinra, an oral peptide that blocks the IL-23 receptor directly, was approved as Icotyde in March 2026 on the strength of a New England Journal of Medicine publication showing PASI 100 of about 27% at week 16 and 40% at week 24, and its own head-to-head wins against deucravacitinib.13 Alumis's envudeucitinib, another highly selective TYK2 inhibitor, reported PASI 100 rates of 28 to 29% at week 16 and around 40% at week 24 in trials of more than 1,700 patients, roughly six to twelve months behind zasocitinib.14
Cross-trial comparison is a sin I am about to commit with full disclosure. The LATITUDE population had about 30% biologic-experienced patients, which typically lowers response rates; ICONIC had a different mix; endpoints and imputation methods vary. With that said, zasocitinib and icotrokinra look remarkably similar on complete clearance, around 25 to 35% at week 16 and 32 to 42% at week 24. Both have beaten Sotyktu head-to-head. Neither has been compared with the other. And both still trail the IL-17 biologics, where bimekizumab produces PASI 100 in roughly 60% or more of patients at week 16.14
Then there is the commercial lesson Takeda says it has learned. Sotyktu was launched with a $4 billion peak-sales forecast. It sold $291 million in 2025, was ranked seventh or eighth of nine psoriasis therapies on efficacy in a Deutsche Bank survey of dermatologists, and Bristol Myers Squibb has said it will stop promoting the drug in dermatology in many markets.15 The reason was not that it failed; the reason was that a pill with 53 to 58% PASI 75 could not compete with injectables at 75 to 90% once payers set the rules. Zasocitinib's efficacy is a real answer to that problem. Its price will be the other half of the answer, and Takeda has said nothing. Sotyktu lists at about $74,000 a year and Icotyde launched at roughly $97,000 to $100,000.13,15 Expect parity with Icotyde.
What Would Change My Rating
I keep a mental list of the things a sponsor could release tomorrow that would move a drug from Promising to Strong Evidence, and for zasocitinib the list is unusually specific. First, a peer-reviewed publication of LATITUDE 3001 and 3002 with the full tables, including the zoster and serious-infection counts that were left out of the investor deck. Second, the Atlas comparator number, with confidence intervals. Third, an FDA label, which would mean regulators with access to patient-level data have signed off on the safety story. Fourth, the three-year extension data. Takeda has said the NDA will be filed "starting in fiscal 2026" and is guiding toward a launch in the first half of 2027, which implies either a very fast filing or a priority review that has not been announced.5 As of the July quarterly report, the application had not been submitted.
The other item on my list is a voice without a Takeda relationship. The principal investigators, the congress presenters, the authors of the pharmacology paper, and the dermatologist who called the drug "game-changing" in a trade-press interview all disclose fees from Takeda or Nimbus.2,4 The closest thing to independent commentary is an August review from Steven Feldman's group at Wake Forest, which concluded that zasocitinib "may have a role but will likely face competition from highly effective, established therapies."14 That is not a dissent. It is a shrug. It is also, at the moment, the most skeptical thing anyone credentialed has put in print.
Anyone describing zasocitinib as "as safe as Sotyktu" is currently describing a hope.
Dr. Maren ColeThe Best Oral Psoriasis Data Ever Presented. Emphasis on Presented.
I want to be clear about what zasocitinib has actually demonstrated, because it is a lot. Two large, well-designed, placebo- and active-controlled Phase 3 trials produced a replicated, very large effect: roughly seven in ten patients clear or almost clear at week 16, and a quarter to a third with no plaques at all, rising to a third to 42% by week 24. A head-to-head against the only approved drug in the class reportedly more than doubled its complete-clearance rate. The Phase 2b is published in a top journal with a clean dose-response. The mechanism is coherent, and the early safety data show none of the metabolic fingerprints that sank the broader JAK class. If you have moderate-to-severe psoriasis and have been waiting for a pill that is not a consolation prize, this is the most credible candidate that has ever existed.
And yet the Phase 3 program lives entirely in press releases, congress slides, and an SEC filing. The head-to-head withholds the comparator's number. The safety table omits the rare events that matter most. The exposure is too short and too small to say anything about cardiovascular events, clots, or cancer. The drug is not approved anywhere. And every expert quoted on the record is paid by the sponsor. None of that means the data are wrong. It means the data have not yet been checked by anyone who does not stand to gain from them.
That is the textbook definition of Promising: encouraging, plausible, large, and not yet independently verified. I expect this to earn Strong Evidence within eighteen months, and I will say so when the papers and the label arrive. But this newsletter rates what has been shown, not what has been promised, and what has been shown is a series of very good slides.
Zasocitinib looks like the first pill that clears psoriasis like a biologic. It also looks that way entirely on Takeda's say-so. Believe the direction, wait for the paper.
- 1. Armstrong AW, Gooderham M, Lynde C, et al. Tyrosine Kinase 2 Inhibition With Zasocitinib (TAK-279) in Psoriasis: A Randomized Clinical Trial. JAMA Dermatology. 2024;160(10):1066–1074. Phase 2b, n=259 dosed, 12 weeks; PASI 75 at week 12: 67% (30 mg) vs 6% placebo; PASI 100 33%.
- 2. Mehrotra S, Sano Y, Halkowycz P, et al. Pharmacological Characterization of Zasocitinib (TAK-279): An Oral, Highly Selective, and Potent Allosteric TYK2 Inhibitor. Journal of Investigative Dermatology. 2026;146(1):214–222. Ki 0.0087 nM; >1-million-fold selectivity over JAK1. Takeda-authored.
- 3. Kivitz A, Baraliakos X, Muensterman ET, et al. Highly selective TYK2 inhibition with zasocitinib improves outcomes in active psoriatic arthritis: a randomised phase 2b study. Annals of the Rheumatic Diseases. 2025;84(10):1660–1674. n=290; ACR20 54.2% vs 29.2% placebo.
- 4. Gooderham M, Laquer V, Zhang J, et al. Once-daily oral zasocitinib demonstrates rapid and reproducible skin clearance: Results from two randomized Phase 3 trials (LATITUDE PsO-3001 and PsO-3002). Late-breaking abstract #79730, American Academy of Dermatology Annual Meeting, Denver, March 27–31, 2026. Unpublished. n=693 + 1,108.
- 5. Takeda Pharmaceutical Company. Investor presentation: Zasocitinib Phase 3 Psoriasis Results. March 28, 2026. Filed with the US SEC as Form 6-K. Contains full baseline, week 16/24/60 efficacy, and pooled TEAE tables (zasocitinib n=970; apremilast n=412; placebo n=417).
- 6. Takeda Pharmaceutical Company. Takeda's Zasocitinib Delivered Rapid and Durable Skin Clearance in Phase 3 Trials. Press release, March 28, 2026. sPGA 0/1 71.4% / 69.2%; PASI 100 33.4% / 25.2%; any TEAE 62.1%.
- 7. Takeda Pharmaceutical Company. Takeda's Zasocitinib Significantly Outperforms Deucravacitinib in Head-to-Head Phase 3 Psoriasis Study (LATITUDE Atlas, TAK-279-PsO-3004, NCT06973291). Press release, June 11, 2026. n=606; PASI 100 at week 16 ">35%" vs ">2.5×" the deucravacitinib rate; comparator rate not disclosed.
- 8. Takeda Pharmaceutical Company. Zasocitinib Shows Consistent Skin Clearance in Phase 3 Psoriasis Studies (scalp, nail, palmoplantar data presented at AAD Innovation Academy). Press release, July 16, 2026. ssPGA 0/1 77% / 74%; hfPGA 0/1 71% / 69%.
- 9. Takeda Pharmaceutical Company. Takeda to Acquire Late-Stage, Potential Best-in-Class Oral Allosteric TYK2 Inhibitor NDI-034858 From Nimbus Therapeutics. Press release, December 13, 2022. $4 billion upfront plus up to $2 billion in milestones; closed February 2023.
- 10. Ytterberg SR, Bhatt DL, Mikuls TR, et al. Cardiovascular and Cancer Risk with Tofacitinib in Rheumatoid Arthritis (ORAL Surveillance). New England Journal of Medicine. 2022;386(4):316–326. n=4,362; MACE HR 1.33 (95% CI 0.91–1.94); malignancy HR 1.48 (1.04–2.09).
- 11. Armstrong AW, Gooderham M, Warren RB, et al. Deucravacitinib versus placebo and apremilast in moderate to severe plaque psoriasis: Efficacy and safety results from the 52-week, randomized, double-blinded, placebo-controlled phase 3 POETYK PSO-1 trial. Journal of the American Academy of Dermatology. 2023;88(1):29–39. n=666; PASI 75 at week 16: 58.4% vs 12.7% placebo vs 35.1% apremilast; PASI 100 ~14%.
- 12. Armstrong AW, et al. Deucravacitinib in plaque psoriasis: Four-year safety and efficacy results from the POETYK PSO-1, PSO-2 and long-term extension trials. Journal of the European Academy of Dermatology and Venereology. 2025. 1,519 patients, 4,392.8 patient-years; exposure-adjusted incidence per 100 PY: herpes zoster 0.55, malignancy 0.89, MACE 0.32, VTE 0.07. Sotyktu prescribing information: no boxed warning; warnings for infections, TB, malignancy, CPK/rhabdomyolysis, triglycerides, liver enzymes.
- 13. Bissonnette R, Soung J, Hebert AA, et al. Oral Icotrokinra for Plaque Psoriasis in Adults and Adolescents (ICONIC-LEAD). New England Journal of Medicine. 2025. n=684; PASI 90 at week 16: 50% vs 4%; PASI 100 at week 24: 40%. Johnson & Johnson press release: FDA approval of ICOTYDE (icotrokinra), March 18, 2026; launch price ~$8,100–8,360 per 30 tablets.
- 14. McGuirt V, Razler D, Moore SG, Feldman SR. Evaluating zasocitinib as an oral therapy for moderate to severe plaque psoriasis. Expert Opinion on Pharmacotherapy. Published online August 16, 2026. Independent review including cross-trial context for envudeucitinib (ONWARD1/2), risankizumab (UltIMMa-1/2), and bimekizumab (BE VIVID / BE READY).
- 15. Recon Strategy. Winning on route of administration isn't enough: lessons from Sotyktu. December 2025. Sotyktu 2025 net sales $291 million vs $4 billion original peak forecast; WAC $6,164.78 per 30 days; Deutsche Bank dermatologist survey ranking; BMS decision to reduce dermatology promotion (Q1 2026).