Ivonescimab (AK112) is a humanized tetravalent bispecific antibody simultaneously targeting PD-1 and VEGF-A [6]; a phase 1a dose-escalation study established the MTD at 20 mg/kg every 2 weeks [6]. Its evidence base sits in lung cancer: in advanced non-small cell lung cancer with EGFR variant after progression on TKI therapy, a placebo-controlled phase 3 trial enrolled 322 patients (ivonescimab n=161, placebo n=161), with IRRC-assessed PFS as the primary endpoint [1]; HARMONi-2 compared ivonescimab head-to-head with pembrolizumab as first-line therapy for PD-L1-positive advanced NSCLC across 55 centers in China [4]. The drug was approved in China in 2024 for non-small cell lung cancer [5] and is covered in the annual Antibodies to Watch review [3].
At ESMO 2026, two new lines were reported. First, biliary tract cancer: live posts state that in the first-line phase 3 trial of ivonescimab plus gemcitabine/cisplatin versus durvalumab plus chemotherapy, ORR, PFS and OS were all better, with a 25% lower death risk; per one poster, this is the drug's 4th positive phase 3 randomized trial and the first outside its lung cancer indications already approved in China. Second, renal cell carcinoma: in a phase Ib/II monotherapy study in first-line IMDC favorable-risk advanced RCC, one post reported 21 evaluable patients, ORR 57.1%, DCR 95.2%, median PFS not reached; another post reported 23 evaluable patients, ORR 65.2%, DCR 95.7%, 12-month PFS 85.5%, and no deaths, with both posts citing grade ≥3 TRAEs of 26.9% (proteinuria 15.4% listed alongside), 13 months of follow-up; the latter referenced CheckMate 9ER's favorable-risk ORR of 66.2% as a cross-trial comparator and framed the finding as hypothesis-generating.
The site news did not independently cover these two readouts, but added circumstantial evidence of pipeline expansion: Ruijin Hospital newly registered two trials in advanced malignant pleural mesothelioma—one phase 1 with a 3+3 dose-escalation design, planning 9 patients across 3, 10, and 20 mg/kg dose levels [N2], and one phase 1b/2 single-arm multicenter trial planning 78 patients, first determining the RP2D before proceeding to phase 2 efficacy evaluation, registered on October 21 with recruitment not yet started [N3]; another industry report mentioned NSCLC and thymic carcinoma oral presentations involving ivonescimab at this meeting but gave no drug-specific numbers [N1]. These align with the official press release describing data across multiple tumor types and phase II data supporting clinical development expansion. Separately, a FiercePharma post noted that, against the backdrop of AstraZeneca's $2B investment in Summit, the Akeso CEO intends to explore more combinations.
Relative to prior knowledge, the step forward is extending the evidence base from lung cancer into biliary tract cancer—a phase 3 win over an active comparator—and a single-agent signal in favorable-risk RCC, alongside apparent expansion into new tumor types such as mesothelioma. Limitations are evident: the RCC data come from a small single-arm study with only 13 months of follow-up, the two posts disagree on evaluable patient counts and ORR with no explanation in the materials; sample size and median OS/PFS for the BTC phase 3 are undisclosed; cross-trial comparisons are hypothesis-generating only, and proteinuria warrants follow-up on the safety side.
