Innovent gives another cytokine a shot

After Innovent delivered a massive licensing deal for its lead cytokine-based asset, IBI363, the company is trying to repeat the trick with a different cytokine. IBI3026, which is based around not IL-2 but IL-12, is about to enter its first human study, according to recent listings on the clinicaltrials.gov registry.

Both molecules are fusion proteins whose structure includes an anti-PD-1 antibody, the aim being to deliver the chosen cytokine selectively to PD-1-positive cells. This could be a way to reduce the toxicity that has held back the development of earlier cytokine therapeutics, and one company that’s convinced is Takeda, which paid Innovent $1.2bn for IBI363 last October.

IBI363 is an anti-PD-1 x IL-2α-biased fusion protein, while IBI3026 fuses anti-PD-1 with attenuated IL-12. Both IL-2 and IL-12 stimulate the growth and survival of T cells, and by designing IBI363 to focus on IL-2Rα bias (with IL-2βγ attenuation), Innovent aimed to reduce the toxicity of broad IL-2 stimulation.

Meanwhile, IL-12 has more typically been used as “armouring” in the design of some Car-T therapies. In the case of IBI3026 the IL-12 element is designed with potency attenuation, according to preclinical presentations, the aim again being to avoid excessive toxicity. Preclinical data suggest “selective” immune system activation, and a favourable therapeutic index.

IBI3026 and IBI363 both use activity at PD-1 to “tether” the molecule to PD-1-positive cells, including T cells, as well as to block PD-1. However, the antibody each project uses appears not to be the same: preclinical disclosures suggest that IBI363 uses a monovalent anti-PD-1 MAb, and IBI3026 a bivalent one. China cleared IBI3026’s phase 1 trial last August, and the study begins this month.

Recently disclosed first-in-human studies*

Note: *these projects were first listed on the clinicaltrials.gov database between 29 Dec 2025 and 8 Jan 2026.

Among other first-in-human trial entrants Pfizer’s PF-08032562 stands out as especially interesting.

This is disclosed only as being an oral small molecule, but a LinkedIn post by Dr Manesh Sharma, one of the investigators, reveals it to be an inhibitor of KAT6A, KAT6B and KAT7. Accordingly, the molecule seems to be related to prifetrastat, the KAT6 inhibitor Pfizer took into phase 3 last year; additional activity at KAT7 is also being pursued by Ideaya, for instance.

Also of note is Cartography’s anti-LY6G6D T-cell engager CBI-1214. This is the private company’s first clinical project, and comes after the group raised $67m last year. The target is pretty unique: the only other clinical-stage project to hit LY6G6D is QLSF Biotherapeutics' QL335, according to OncologyPipeline, and Roche in 2024 discontinued a similarly acting T-cell engager, linclatamig, which disappointed in phase 1.

Finally, the mechanism of Evopoint’s XNW34017 will be of note to Nurix investors. The molecule is described as an oral protein degrader targeting Aurka while simultaneously degrading Myc.

This appears to mirror a preclinical molecule Nurix revealed at last year’s AACR conference, coded NRX-4972 and said to be an orally bioavailable and brain penetrant Aurka degrader. OncologyPipeline reveals no other industry projects with degrader activity on the Aurka (aurora A kinase) oncogene.