Immunology & inflammation

Two Phase 2b misses in two months turns a mechanism question into a mechanism problem: MRGPRX2 antagonism was Evommune's platform bet across AD, CSU and migraine, and when the same target fails to move the needle in two distinct inflammatory indications, the more parsimonious read isn't "wrong molecule" — it's "wrong biology." The tell is the pivot itself: rather than pushing EVO756 into a third dermatology indication, Evommune is handing the lead AD slot to EVO301, an entirely different IL-18BP mechanism, while EVO756 gets demoted to a migraine hedge nobody's betting the company on. That leaves Evommune effectively a single-shot story riding on EVO301's 2027 Phase 2b — a long runway to prove out with a thinner cushion, and exactly the kind of back-to-back-failure narrative that compresses valuation and stiffens terms on whatever financing round comes next.

Schrödinger is monetizing its own discovery engine here, not just licensing it to others — a meaningful shift for a company whose story has long been 'platform, not pipeline.' Spinning SDGR-4594 and SDGR-8139 out to a VC-backed shell instead of advancing them in-house is capital-efficiency logic: Schrödinger keeps equity and royalty upside without funding preclinical burn off its own balance sheet, while NEA and RA Capital absorb the binary risk. The equity-plus-milestones structure, instead of a straight license fee, signals that early discovery-stage assets are hard to price for cash right now — everyone's betting on the come. Watch for more computational-discovery shops (Recursion, Insitro, Isomorphic) following this playbook: spin out rather than sell or self-fund, and let Series A investors underwrite the translational risk.

Shelving a Phase 3 asset with seven patients already dosed — not a preclinical program — tells you where the real cash crunch sits. This isn't routine reprioritization; it's a company admitting it can't fund its most advanced bet to the finish line, however the release frames it. The pivot to in vivo TCR-T for solid tumors is bet-the-company: trading a derisked clinical program for an earlier-stage platform nobody has yet proven works in vivo. High conviction, but also a tacit admission the allogeneic heme franchise found no partner willing to bankroll it. Allogeneic TCR-T for blood cancers has gotten capital-intensive with thin differentiation between competitors — expect more heme/autoimmune cell-therapy assets to hit the partnering block as smaller players run short of runway before readouts. Reaching only Q4 2027 after cutting 75% of headcount suggests the burn problem was structural, not confined to the heme program.

A miss with an asterisk, not a pivot. IFNGS-high response is biologically coherent — interferon signature is literally TYK2's mechanism — so this isn't pure spin like some post-hoc subgroup rescues. But shrinking Phase 3 to biomarker-enriched patients trades a bigger addressable market for a better shot at significance; that's the right call scientifically, wrong one commercially unless a companion diagnostic scales cleanly. Versus BMS's already-approved Sotyktu, Alumis needed a clean topline to differentiate TYK2 in autoimmune beyond psoriasis — it didn't get one. Expect the Street to stay skeptical until FDA blesses the enrichment strategy; a >50% drop doesn't reverse on a CEO's confidence alone. Psoriasis filing is now doing the heavy lifting for the story — good, since it's biomarker-agnostic — but lupus was supposed to be the platform proof point, and that thesis just got smaller.

Beating teriflunomide isn't the same as beating the field's real benchmark — ocrelizumab, natalizumab, the high-efficacy therapies most MS neurologists already reach for first. Still, the safety data are the real story here. Hepatotoxicity has dogged this drug class, including flags on Sanofi's rival tolebrutinib program, so a clean liver readout is a genuine differentiator, not just noise. If remibrutinib's selectivity truly spares the liver, Novartis gets real separation in the BTK race rather than a me-too label. Add this to the existing CSU approval and Novartis is building an actual platform molecule — one asset, multiple indications, shared manufacturing and commercial muscle. The number to watch next: whether that disability-progression trend holds up, since that's what actually shifts prescribing.

Autoantibody degradation goes after a gap FcRn inhibitors like Vyvgart can't close — those broadly dial down IgG, this targets the specific rogue antibodies. If MER511's Graves' data hold, it's a cleaner mechanism story than another B-cell depleter. The structure says Lilly isn't paying for hope: milestone-heavy pricing on preclinical/Phase 1 assets pushes the risk onto Merida's team hitting data, not onto Lilly's balance sheet. $2.88B is reasonable for a validated new modality, rich if MER511 stalls. It slots next to lebrikizumab and mirikizumab without cannibalizing either. The real signal is deal #13 this year — Lilly's recycling GLP-1 cash into next-gen immunology and allergy bets while it still has the war chest to shop early.

Autoimmune CAR-T was the industry's hottest oncology-to-new-indication pivot — Kyverna, Cabaletta, Autolus and multiple academic programs are all racing the same CD19 mechanism, so this becomes a class-level scrutiny event, not just a Novartis/BMS problem. The two signals aren't equivalent: three IEC-HS deaths is a materially different regulatory conversation than BMS's "reversible" events — expect FDA to treat rap-cel far more cautiously than zola-cel, even though headlines pair them. Multi-trial pauses cost both companies precious time on programs meant to diversify beyond their legacy oncology CAR-T franchises (Kymriah, Breyanzi) into a much larger chronic-disease market. If regulators respond with class-wide holds or REMS-style safeguards, smaller autoimmune CAR-T players inherit the delay and diligence burden without the balance sheets to absorb it.