AK104

Cadonilimab (AK104), brand name Kaitanni in China, is Akeso Biopharma's bispecific antibody that hits PD-1 and CTLA-4 with a single molecule. It's already approved in mainland China for second-line recurrent/metastatic cervical cancer (2022) and first-line HER2-negative gastric/GEJ (gastroesophageal junction, where the esophagus meets the stomach) adenocarcinoma (2024) [2][3], but has no FDA or EMA approval. This Phase 2 trial (NCT06998394) [1] tests the drug alongside concurrent chemoradiotherapy in a narrow, hard-to-treat slice: oligometastatic stage IVB cervical cancer, where distant relapse defeats most standard regimens. The chemotherapy backbone is cisplatin (weekly or every 3 weeks) combined with pelvic external beam radiation and brachytherapy, with cadonilimab dosed at 10 mg/kg every 3 weeks [1]. The specific question here is whether adding dual checkpoint blockade to the existing chemoradiation backbone can extend progression-free survival in patients who otherwise almost universally progress. Beyond this single trial, the strategic question for Akeso is a US market path: cadonilimab is one of two Akeso bispecifics that could support an FDA-registration deal similar to Summit Therapeutics' ivonescimab licensing (up to $5B total, $500M upfront) [8]. Watching this trial in isolation misses the point. It matters mostly as a signal about how aggressively Akeso is expanding cadonilimab's use across cervical cancer stages.

Approved drug being pushed into a new setting. China's NMPA (National Medical Products Administration, roughly equivalent to the FDA) granted cadonilimab conditional approval in June 2022 for platinum-resistant recurrent/metastatic cervical cancer, based on the AK104-201 registrational study - a single-arm Phase 2 that showed a 33% objective response rate in 100 evaluable patients (43.8% in PD-L1 CPS≥1), median duration of response not reached, and median overall survival of 17.5 months [2][12]. The label was broadened in 2024 for first-line HER2-negative gastric/GEJ adenocarcinoma [3]. No FDA designations disclosed for this specific chemoradiation indication, and no publicly announced US IND filings, FDA meetings, or ex-China Phase 1 activity for cadonilimab as of mid-2026. Akeso runs a broad late-stage program: Phase 3 studies in hepatocellular carcinoma, gastric cancer, and locally advanced cervical cancer (NCT07400536, enrolling 378 patients [4]), alongside multiple Phase 2 combinations including gastric neoadjuvant and second-line settings [9]. NCT06998394 targets 30 participants with a primary endpoint of 1-3 year progression-free survival [1]; readout timelines have not been disclosed publicly. Akeso's FY2024 total commercial sales revenue was RMB 2,002.4 million (roughly $278M USD at 2024 rates), up 24.9% year-over-year, driven jointly by cadonilimab and ivonescimab; Akeso does not break out cadonilimab (Kaitanni) revenue separately in its public filings [5]. Cadonilimab holds NRDL coverage - inclusion in China's National Reimbursement Drug List, meaning government insurance pays for it, which drives volume but compresses per-patient economics. The commercially material question is whether Akeso secures a US or EU partner, since global registration would open a market roughly 10x larger than China at typical checkpoint inhibitor pricing. No such partner has been announced for cadonilimab as of this writing, and Akeso continues to run its late-stage program primarily in Chinese sites.

PD-1 and CTLA-4 are both molecular brakes on T cells, the immune cells whose job is to recognize and kill tumor cells. Tumors escape immune attack by activating these brakes. Ipilimumab plus nivolumab (Bristol Myers Squibb) hits both brakes with two separate antibodies and has driven durable responses in melanoma, kidney cancer, and MSI-high colorectal cancer (a tumor mutation pattern where cells lose mismatch repair function, generating high neoantigen loads that make cancers unusually responsive to checkpoint blockade) [6]. Cadonilimab does the same job with a single bispecific molecule instead of two drugs. Akeso's pitch is a toxicity advantage: because PD-1 and CTLA-4 are more likely to co-express on T cells inside tumors than in peripheral tissue, a bispecific format binds preferentially in the tumor microenvironment and reduces the systemic autoimmune toxicity that plagues ipi/nivo. Grade 3+ irAEs (immune-related adverse events - side effects caused by over-activation of the immune system attacking healthy tissue) run around 55-59% in the melanoma setting for ipi/nivo [6]. Cadonilimab China trials have shown lower Grade 3+ treatment-related adverse event rates (28.8% in AK104-201 [12]) than historical ipi/nivo benchmarks, though no head-to-head trial has been run. Cross-trial safety comparisons are unreliable, so the format advantage remains a plausible hypothesis, not confirmed fact. The mechanism itself is deeply validated by a decade of dual-checkpoint clinical data; what's new is the format, not the biology.

NCT06998394 [1] is a single-arm Phase 2 in stage IVB cervical cancer with oligometastatic disease, receiving concurrent chemoradiotherapy plus cadonilimab. The regimen: cisplatin weekly or every 3 weeks, pelvic external beam radiation, and brachytherapy, with cadonilimab 10 mg/kg every 3 weeks for up to 2 years or until progression. Enrollment target is 30 patients, primary endpoint is progression-free survival at 1-3 years, with secondary endpoints of objective response rate, disease control rate, and overall survival. The trial is running at a single site (First Affiliated Hospital of Zhengzhou University). Oligometastatic means limited distant spread, typically five or fewer lesions, where aggressive local therapy can still deliver long remissions, unlike widely metastatic disease where systemic control is the only realistic option. Standard concurrent chemoradiation drives many IVB cervical cancer patients to complete local response, but distant progression rates remain high, which is where a systemic immunotherapy backbone might extend benefit. Adding a checkpoint inhibitor during radiation follows a specific rationale: radiation releases tumor antigens and immune-priming signals, and the KEYNOTE-A18 trial (pembrolizumab plus chemoradiation) established this combination approach in locally advanced cervical cancer, gaining FDA approval in early 2024 [7]. Concerns: single-arm design limits interpretability against the pembrolizumab-plus-CRT standard, the patient population is niche (oligometastatic IVB is a small subset of cervical cancer - global cervical cancer incidence runs ~660,000 cases/year with ~15-20% presenting as stage IVB, and oligometastatic is a further subset of that), a 30-patient single-site study is more hypothesis-generating than confirmatory, and no comparator arm means efficacy claims will rest on historical controls. That's a low bar for hypothesis generation and a weak one for regulatory decision-making.

Our model estimates a 24% chance this drug is eventually approved. It starts from the historical base rate for Phase 3 drugs in this area (about 48%), then adjusts using ten facts about the trial and sponsor. What moves the number most: it is helped by a non-randomized design; it is held back by the sponsor's thin or weak approval record, smaller-than-typical enrollment for this phase, and weak or limited earlier-phase results. The other facts land near average for this stage, so they leave the estimate roughly where the base rate put it.

Efficacy risk is the largest exposure. Oligometastatic stage IVB cervical cancer is a small, heterogeneous population; even a positive result in 30 patients would need Phase 3 validation before regulatory traction outside China. Cross-trial comparison against pembrolizumab plus chemoradiation (KEYNOTE-A18) [7] will dominate interpretation, and cadonilimab has not been tested head-to-head against pembrolizumab in any cervical cancer setting. Safety risk is meaningful but not novel. Dual checkpoint blockade plus concurrent cisplatin-based chemoradiation stacks immune, myelosuppressive (cisplatin), and radiation toxicities - cisplatin brings nephrotoxicity and cytopenias, radiation brings GI and genitourinary damage, and dual PD-1/CTLA-4 blockade brings colitis, pneumonitis, and endocrinopathies. Cadonilimab's China safety profile has looked cleaner than historical ipi/nivo [12], but never in a chemoradiotherapy setting, where irAEs can be difficult to distinguish from radiation-driven bowel, bladder, and hematologic effects. Execution risk: Akeso is not an FDA-experienced sponsor, and any US registrational path likely requires a global partner. The precedent is Summit Therapeutics' up-to-$5B licensing deal for ivonescimab ($500M upfront plus up to $4.5B in regulatory and commercial milestones), a different Akeso bispecific targeting PD-1 and VEGF [8], but no equivalent partner exists for cadonilimab currently and no publicly disclosed US IND for cadonilimab. Commercial risk: pembrolizumab plus chemoradiation is already FDA-approved for locally advanced cervical cancer [7] and holds standard-of-care position in the US and EU. Cadonilimab would need superior efficacy or meaningfully lower toxicity to displace it. China's NRDL pricing gives Akeso volume but low per-patient economics; ex-China is where the real commercial upside sits.

Worth watching, but this specific trial is not the signal. The bigger commercial question for Akeso is whether Phase 3 data in first-line gastric cancer and follow-on cervical trials [9] support an FDA submission, and whether Akeso secures a US registration partner similar to what Summit did for ivonescimab [8]. This oligometastatic cervical Phase 2 is one of many exploratory combinations, and it will not move Akeso's stock or the drug's commercial trajectory alone. The datapoint that would matter: a Phase 3 readout showing cadonilimab beats or matches pembrolizumab in a directly comparable cervical or gastric cancer population, or a licensing deal announcement for ex-China rights. Check back mid-to-late 2026 for Phase 3 gastric OS updates and any pre-BLA activity. Akeso's China revenue trajectory - FY2024 total commercial sales of RMB 2.0B (~$278M USD), up 25% year-over-year [5] - is the leading commercial indicator to track quarterly, though Akeso does not break cadonilimab out separately from ivonescimab. The strategic angle is that Akeso is becoming the go-to Chinese bispecific manufacturer, and cadonilimab's fate will help determine whether that platform expands globally through licensing or stays regional.