structural-biology.com

Casebook · From Marc's papers

2026 – Discovery of the Orally Bioavailable Isoform Selective Janus Kinase 1 (JAK1) Compound Povorcitinib (INCB054707) for the Treatment of Inflammatory and Autoimmune Diseases

2 Jun 2026 · updated 17 Sep 2026 · 3 min read

10PI · JAK1 kinase (JH1 domain) in complex with povorcitinib · X-ray · 1.54 Å · Open the four lenses →

Discovery of the Orally Bioavailable Isoform Selective Janus Kinase 1 (JAK1) Compound Povorcitinib (INCB054707) for the Treatment of Inflammatory and Autoimmune Diseases

  • Selectivity by geometry: Povorcitinib exploits a roughly 2 Å P-loop shift that opens a lipophilic groove only in JAK1.
  • Sparing JAK2: Leaving JAK2 intact avoids the anemia and cytopenias tied to erythropoietin and thrombopoietin blockade.
  • Residues that decide: JAK1 His885 faces solvent while JAK2 Asn859 and Phe860 close the pocket against the inhibitor’s tail.
  • Oral exposure by design: An intramolecular hydrogen bond masks the amide’s polarity, lifting permeability and oral bioavailability across species.
  • From structure to clinic: The 46-fold-selective design now shows durable activity in hidradenitis suppurativa, validating the JAK1-only strategy.
  1. Discovery of the Orally Bioavailable Isoform Selective JAK1 Compound Povorcitinib (INCB054707): Zhuo et al., J. Med. Chem., 2026
  2. A Comprehensive Overview of Globally Approved JAK Inhibitors: Shawky et al., Pharmaceutics, 2022
  3. Functional and Structural Characterization of Clinical-Stage JAK2 Inhibitors Identifies Determinants for Drug Selectivity: Miao et al., J. Med. Chem., 2024
  4. Intramolecular Hydrogen Bonding: An Opportunity for Improved Design in Medicinal Chemistry: Caron et al., Med. Res. Rev., 2019
  5. Efficacy and Safety of the Oral JAK1 Inhibitor Povorcitinib in Hidradenitis Suppurativa: Kirby et al., J. Am. Acad. Dermatol., 2024

How a retracted loop and a fluorinated tail let povorcitinib tell two near-identical kinase pockets apart

The four Janus kinases share an ATP pocket so conserved that inhibiting one without the others is a real medicinal-chemistry problem, which is why most early JAK drugs hit two or more isoforms at once, as cataloged across the approved JAK inhibitor class. That promiscuity has a cost. JAK2 carries erythropoietin and thrombopoietin signaling, so blocking it invites anemia, thrombocytopenia, and other cytopenias, a liability the structural profiling of JAK2 inhibitors ties directly to its role in hematopoiesis. JAK1 tells a different story. It sits at the center of the inflammatory cytokine network, pairing with JAK2, JAK3, and TYK2 to relay interleukin and interferon signals. A compound that shuts down JAK1 while leaving JAK2 untouched should calm autoimmune disease without starving the bone marrow. The hard part is finding something in the binding site that actually distinguishes the two.

A recent Incyte paper of mine reconstructs how povorcitinib (INCB054707) was engineered to do exactly that. Co-crystal structures of the inhibitor bound to the isolated JAK1 and JAK2 kinase domains, deposited as PDB 10PI and 10PJ, show that its 3,5-dimethylpyrazole hinge binder forms the expected hydrogen bonds with the backbone of Leu932 and Glu930 in both isoforms. Selectivity is not written at the hinge, where the two kinases look almost identical. It is written further out along the elongated arm of the molecule, which extends past the gatekeeper toward the glycine-rich P-loop.

“JAK1 selectivity was conferred by the amide’s alkyl group accessing a small lipophilic groove”

There, the two isoforms diverge. In JAK1, the P-loop rides about 2 Å higher, opening a shallow lipophilic groove that neatly accepts the terminal trifluoro-isopropyl amide. In JAK2, the equivalent loop closes: JAK1 presents His885 pointing into the solvent, JAK2 swings Asn859 inward to hydrogen-bond with Glu1015, and a nearby Phe860 rotates the pocket shut. The same fluorinated tail that fits JAK1 now clashes, and affinity drops.

Two design moves made the difference. Trading the metabolically vulnerable pyrrolopyrimidine for the smaller dimethylpyrazole pushed that tail deeper into the JAK1-specific space, roughly doubling isoform selectivity to about 46-fold. An intramolecular hydrogen bond then masked the amide’s polarity, rescuing permeability and oral exposure across species. The payoff is visible beyond the crystallography: povorcitinib has produced durable responses in hidradenitis suppurativa, with 2026 long-term readouts reinforcing the safety case for selective JAK1 inhibition. A two-residue difference, read correctly, becomes a drug.