What is KAI-7535?

What KAI-7535 is and who makes it

KAI-7535 is an oral small-molecule GLP-1 receptor agonist in development for obesity, overweight, and type 2 diabetes. Hengrui Pharma, one of China's largest drugmakers, discovered the molecule and runs its Chinese trials under the name HRS-7535. In 2024, Hengrui licensed the rights outside Greater China to Kailera Therapeutics, a US biotech formed specifically to develop Chinese-originated obesity and metabolic drugs for Western markets. Outside China, the same molecule goes by KAI-7535.

That two-name setup matters for the timeline. Hengrui is running its own Chinese development program and plans to file for approval there on the strength of its own trials. Kailera has to generate its own data for the US and other markets before it can file anywhere else, which puts KAI-7535 much closer to a Chinese approval than a US one.

Why "small molecule" is the detail that matters

The distinction between a peptide and a small-molecule GLP-1 drug determines how the pill has to be taken. Oral semaglutide is a peptide, and peptides get broken down by stomach acid, so Novo Nordisk built a special absorption-enhancing formulation around it. Even with that engineering, oral semaglutide requires a strict fasting window: take it with a small amount of water, then wait 30 minutes before eating, drinking, or taking other medication.

Small molecules like orforglipron, CX11, and now KAI-7535 survive digestion without that workaround. No fasting window, no timing rules, take it like any other daily pill. For a fuller rundown of how that split shaped the oral GLP-1 category, see the oral GLP-1 pill guide.

What the obesity trial found

HARBOR-1 (NCT06904105) enrolled 556 adults in China with obesity or overweight, mean BMI 34.0, randomized 2:2:1 to 120 mg, 180 mg, or placebo once daily. At week 44, the 120 mg arm averaged 9.5% weight loss and the 180 mg arm averaged 10.9%, against 2.5% for placebo. A follow-up, ad hoc analysis at week 50 put the 180 mg group at 11.1%, still trending down with no sign of a plateau.

Responder rates at the 180 mg dose: 68.2% of participants lost at least 5% of body weight, 46.6% lost at least 10%, and 26.0% lost at least 15%. Those numbers put KAI-7535 in the same general range as orforglipron and CX11, the other oral small-molecule GLP-1s furthest along in development.

What the diabetes trial found

OUTSTAND-2 (NCT06589765) enrolled 810 adults in China with type 2 diabetes, testing 30 mg, 60 mg, and 90 mg of HRS-7535 against dapagliflozin, an SGLT2 inhibitor already approved for diabetes. At week 32, HbA1c dropped 1.58% at 30 mg, 1.50% at 60 mg, and 1.68% at 90 mg, versus 1.28% for dapagliflozin. All three doses beat the active comparator, which is the bar Hengrui needs to clear for a Chinese diabetes filing.

The tolerability problem

This is where KAI-7535's story gets more complicated. Every GLP-1 drug causes some nausea and vomiting, concentrated early in treatment and during dose increases. HRS-7535's rates are the highest reported for any oral GLP-1 pill so far: nausea hit 70.0-70.3% of participants in HARBOR-1, versus 16.2% on placebo. Vomiting hit 66.7-68.6%, versus 4.5%. Diarrhea affected 35.9-36.9%, versus 15.3% on placebo.

For comparison, CX11 reported vomiting in just 12-16% of its Phase 2 participants. Those are different trials with different doses and durations, so the comparison is directional rather than exact, but the gap is large enough to be a real signal, not noise.

The strange part: despite those GI numbers, treatment discontinuation stayed low, 3.1-4.1% on HRS-7535 versus 2.7% on placebo. Most people who felt nauseated or threw up stayed on the drug rather than quitting. Kailera reported no liver safety signal, consistent with earlier HRS-7535 trials. Whether the tolerability gap closes with a gentler dosing schedule, or turns out to be an inherent feature of this molecule, is the open question the next round of trials has to answer.

How KAI-7535 compares to the other oral GLP-1s

Three data points, with the caveat that applies to every early-stage comparison in this category: different trials, different doses, different durations, different populations. None of this is a head-to-head.

KAI-7535's efficacy holds up against its two closest oral rivals. Its tolerability profile does not, at least not yet. That gap is exactly what Kailera's global Phase 2 trial, which started in April 2026 across the US and Australia, is designed to test: a slower titration schedule starting at 15 mg and stepping up every four weeks to a maximum of 180 mg, instead of the faster ramp used in the Chinese trials.

What is next for KAI-7535

Hengrui plans to submit new drug applications in China for both the obesity and type 2 diabetes indications, using the HARBOR-1 and OUTSTAND-2 data directly. That path could move relatively fast, since the Phase 3 work is already done.

Outside China, the timeline is much longer. Kailera's global Phase 2 trial, roughly 320 participants across the US and Australia, is still enrolling, with data expected in 2027. If that trial shows the slower titration meaningfully improves tolerability without giving up efficacy, Kailera would still need to run its own Phase 3 program before seeking FDA approval. Realistically, that puts a US approval, if it happens at all, several years out. For context on how the broader next-generation obesity drug pipeline is shaping up, see what is retatrutide, which is much further along with a substantially higher efficacy ceiling.

For the two oral GLP-1 pills you can actually get a prescription for today, see the oral semaglutide vs orforglipron comparison.