Safiglipron

updated

Safiglipron (HRS-7535 at Hengrui, KAI-7535 at Kailera) is an unapproved oral small-molecule GLP-1 receptor agonist. In a peer-reviewed phase 2 trial in 235 Chinese adults with obesity, placebo-adjusted weight change at 26 weeks reached -6.87% at 180 mg. A completed 556-patient phase 3 trial in obesity (HARBOR-1) has results only as a company statement dated 7 July 2026 of 8.0% and 9.8% versus 2.4% at week 44, and two published phase 3 trials in type 2 diabetes lowered HbA1c.

Oral small-molecule GLP-1 receptor agonist

  • Safiglipron is PubChem CID 166094939, C32H31ClFN3O5, 592.1 g/mol, CAS 2757506-39-9, with HRS-7535 and KAI-7535 among its synonyms. Kailera says Hengrui Pharma developed it and licensed it to Kailera outside of Greater China in 2024.
  • In the phase 2 trial (235 Chinese adults with obesity, 26 weeks), placebo-adjusted weight change was -0.49%, -4.60%, -3.67% and -6.87% at 30, 60, 120 and 180 mg. The 30 mg dose did not differ from placebo.
  • The company reported on 7 July 2026 that HARBOR-1 (556 adults, week 44) gave 9.5% and 10.9% weight loss at 120 and 180 mg versus 2.5% on placebo (efficacy estimand), and 8.0% and 9.8% versus 2.4% (treatment policy estimand). That result is not peer reviewed.
  • OUTSTAND-1 (284 adults) and OUTSTAND-2 (810 adults), phase 3 trials in type 2 diabetes in Nature Medicine, showed placebo-adjusted HbA1c differences of -1.20% to -1.45% and non-inferiority to dapagliflozin.
  • Nausea occurred in about 70% and vomiting in about 67 to 69% of treated participants in HARBOR-1, per the company's 7 July 2026 statement, versus 16.2% and 4.5% on placebo.
  • No approval appears in any source reviewed. A brokerage repost dated 2 September 2026 reports an NMPA acceptance notice for the type 2 diabetes application, and no US application or label was found on 10 October 2026.

What is safiglipron, and is it approved anywhere?

Safiglipron (INN) is the name for the oral small-molecule GLP-1 receptor agonist known as HRS-7535 at Hengrui and KAI-7535 at Kailera. It is not approved anywhere in the sources reviewed. A brokerage repost dated 2 September 2026 reports an NMPA acceptance notice for the type 2 diabetes application, and no US application or label was found.

PubChem CID 166094939 is safiglipron, listed as an INN, C32H31ClFN3O5, 592.1 g/mol, CAS 2757506-39-9. HRS-7535, HRS7535, KAI-7535 and KAI7535 are among its synonyms, so Hengrui's code is HRS-7535. Kailera's release of 1 October 2026 calls it the oral small molecule GLP-1 receptor agonist safiglipron, also known as KAI-7535 (HRS-7535), and its release of 7 July 2026 says HRS-7535 was developed by Hengrui Pharma and licensed to Kailera outside of Greater China in 2024. The Nature Medicine report of OUTSTAND-2 uses the name safiglipron for the trial registered under HRS-7535.

The registry sponsor of the China trials is Shandong Suncadia Medicine Co., Ltd. The registry itself does not tie that company to Hengrui, and Kailera's releases say the trials were conducted by Hengrui.

In a statement dated 7 July 2026, Hengrui said it plans to submit NDAs for HRS-7535 in type 2 diabetes and obesity in China, without stating that an application had been accepted. A brokerage repost dated 2 September 2026 of a Hengrui announcement says its subsidiary Shandong Shengdi Pharmaceutical Co., Ltd. received an NMPA Acceptance Notice for the application for HRS-7535 tablets in adult type 2 diabetes, based on OUTSTAND-1 and OUTSTAND-2. The repost does not mention an obesity application, and one aggregator's report of an obesity acceptance was not confirmed. Acceptance of a filing is not approval.

On 10 October 2026 an openFDA Drugs@FDA search by generic name returned no records and a DailyMed search returned 0 labels, so there is no US application or label.

PubChem: safiglipron identity, with HRS-7535 and KAI-7535 codes

Chemical identity record

National Library of Medicine. PubChem compound record CID 166094939 (safiglipron).

PubChem CID 166094939 is safiglipron, listed as an INN, C32H31ClFN3O5, 592.1 g/mol, CAS 2757506-39-9, with HRS-7535, HRS7535, KAI-7535 and KAI7535 among its synonyms, so Hengrui's code is HRS-7535. Kailera's release of 1 October 2026 calls the molecule oral small molecule GLP-1 receptor agonist safiglipron, also known as KAI-7535 (HRS-7535). Its release of 7 July 2026 says HRS-7535 was developed by Hengrui Pharma and licensed to Kailera outside of Greater China in 2024. The 2026 Nature Medicine report of OUTSTAND-2 uses the name safiglipron for the trial registered as NCT06589765 under HRS-7535. The registry sponsor of the China trials is Shandong Suncadia Medicine Co., Ltd.

Participants / model
Chemical identity, not a clinical population
Treatment
Defined molecule, oral tablet
Follow-up
Not applicable
Study design
NLM chemical record, with company statements and registry sponsor information on the code mapping
Funding
not stated in the source

PubChem and Kailera's release use all three names. The registry itself does not tie the sponsor Shandong Suncadia Medicine Co., Ltd. to Hengrui, and Kailera's releases say the trials were conducted by Hengrui.

Read the original source
Company statement, 7 July 2026: NDA plans in China; no US application or label found

Regulatory record

Jiangsu Hengrui Pharmaceuticals announcement, 7 July 2026, on hengrui.com; Kailera release of 26 May 2026; openFDA Drugs@FDA and DailyMed searches run on 10 October 2026.

Hengrui says it plans to submit NDAs for HRS-7535 for the treatment of type 2 diabetes and obesity in China. The release does not state that an application has been accepted. Kailera's release of 26 May 2026 likewise says it plans to submit an NDA for type 2 diabetes. On 10 October 2026 an openFDA Drugs@FDA search by generic name safiglipron returned no records and a DailyMed search returned 0 labels, so there is no US application or label.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Statements dated 26 May and 7 July 2026; searches run on 10 October 2026
Study design
Company statement and dated regulatory search
Funding
not stated in the source

A plan to submit is not a submission, and a submission is not an approval. The US search shows what the databases held on that date.

Read the original source
Repost, 2 September 2026: NMPA acceptance notice for the type 2 diabetes application

Regulatory record

Brokerage repost dated 2 September 2026 of a Hengrui announcement, read on iTiger.

A brokerage repost dated 2 September 2026 of a Hengrui announcement says its subsidiary, Shandong Shengdi Pharmaceutical Co., Ltd., received an NMPA Acceptance Notice for the marketing authorization application for HRS-7535 tablets in adult type 2 diabetes, based on OUTSTAND-1 and OUTSTAND-2. The repost read does not mention an obesity application; one aggregator reported an obesity acceptance that was not confirmed. Acceptance of a filing is not approval, and no approval exists in any source reviewed.

Participants / model
Adults with type 2 diabetes, planned label
Treatment
HRS-7535 tablets
Follow-up
Statement dated 2 September 2026
Study design
News repost of a company announcement; not a regulator record
Funding
not stated in the source

The announcing subsidiary Shandong Shengdi Pharmaceutical Co., Ltd. is not tied to Hengrui by the registry itself. Acceptance for obesity is unconfirmed.

Read the original source

What did the peer-reviewed studies in obesity find?

In a 26-week phase 2 trial in 235 Chinese adults with obesity and no diabetes, placebo-adjusted weight change was -0.49%, -4.60%, -3.67% and -6.87% at 30, 60, 120 and 180 mg. The 30 mg dose did not differ from placebo. A phase 1 study in healthy participants reported a 6.63% weight fall on day 29 versus 1.18% on placebo.

The phase 2 trial (NCT06250946) was run at 29 centers in China. It randomized 235 adults with obesity (BMI 28.0 to 40.0) and no diabetes 1:1:1:1:1 to once-daily oral HRS-7535 at target doses of 30, 60, 120 or 180 mg, or placebo. At week 26 the least-squares mean body weight changes were -2.99%, -7.09%, -6.17% and -9.36% versus -2.50% on placebo. The placebo-adjusted differences were -0.49%, -4.60%, -3.67% and -6.87% (P = 0.7104, 0.0006, 0.0062 and <0.0001). Gastrointestinal events were most common, mostly mild to moderate. The 30 mg dose did not differ from placebo.

The phase 1 trial in healthy participants was randomized and placebo-controlled, with single doses of 15, 60 and 120 mg, a food-effect part and 28 days of multiple doses. On day 29, body weight fell 4.38 kg (6.63%) on HRS-7535 versus 0.8 kg (1.18%) on placebo. That is a healthy-participant safety study, not an obesity efficacy result.

Phase 2: 235 Chinese adults with obesity, 26 weeks, placebo-adjusted -0.49% to -6.87%

Randomized controlled trial

Gu W et al. Nat Commun. 2026;17(1). PMID 42331800.

At week 26 the least-squares mean body weight changes were -2.99%, -7.09%, -6.17% and -9.36% at 30, 60, 120 and 180 mg versus -2.50% on placebo, with placebo-adjusted differences of -0.49%, -4.60%, -3.67% and -6.87% (P = 0.7104, 0.0006, 0.0062 and <0.0001). Gastrointestinal events were most common, mostly mild to moderate. The 30 mg dose did not differ from placebo.

Participants / model
235 Chinese adults with obesity (BMI 28.0 to 40.0) and no diabetes, at 29 centers
Treatment
HRS-7535 at target doses of 30, 60, 120 or 180 mg orally once daily, or placebo, randomized 1:1:1:1:1
Follow-up
26 weeks
Study design
Randomized phase 2 trial (NCT06250946)
Funding
not stated in the source

The changes from baseline and the placebo-adjusted differences are different numbers. The 120 mg arm lost less than the 60 mg arm.

Read the original source
Phase 1 in healthy participants: median Tmax 5.98 to 10.98 hours, half-life 5.28 to 9.08 hours after single doses

Randomized controlled trial

Wu J et al. Diabetes Obes Metab. 2024;26(3):901-910. PMID 38100147.

Nausea and vomiting were the most frequent adverse events. The median Tmax was 5.98 to 5.99 hours after single doses and 5.98 to 10.98 hours on day 28; geometric mean half-life was 5.28 to 9.08 hours after single doses and 6.48 to 8.42 hours on day 28; exposure was approximately dose-proportional. On day 29, body weight fell 4.38 kg (6.63%) on HRS-7535 versus 0.8 kg (1.18%) on placebo.

Participants / model
Healthy participants
Treatment
HRS-7535 oral, once daily: single doses of 15, 60 and 120 mg; a 90 mg food-effect part; multiple doses reaching 120 mg over 28 days
Follow-up
Single doses; 28 days of multiple doses, with weight reported on day 29
Study design
Randomized placebo-controlled phase 1 trial
Funding
not stated in the source

Ranges are reported for Tmax and half-life, not single values. Weight change in healthy participants over 28 days is not an obesity efficacy result.

Read the original source

What did the two phase 3 studies in type 2 diabetes find?

Both were reported in Nature Medicine. In OUTSTAND-1 (284 adults) placebo-adjusted HbA1c fell by 1.22%, 1.20% and 1.45% at 30, 60 and 90 mg. In OUTSTAND-2 (810 adults on metformin) all three doses were non-inferior to dapagliflozin, and the 90 mg dose met superiority.

OUTSTAND-1 (NCT06672172) was a phase 3 placebo-controlled trial at 46 sites in China. It randomized 284 adults with type 2 diabetes managed with diet and exercise (mean HbA1c 7.95%) to safiglipron 30, 60 or 90 mg or placebo for 32 weeks. Placebo-adjusted HbA1c differences were -1.22% (95% CI -1.53 to -0.91), -1.20% (-1.52 to -0.89) and -1.45% (-1.75 to -1.14), all P<0.0001 under the treatment policy estimand. Body weight differences were modest (-0.65%, -2.23% and -3.56% versus placebo).

OUTSTAND-2 (NCT06589765) was a phase 3 double-dummy active-comparator trial at 98 sites in China. It randomized 810 adults with type 2 diabetes on metformin (mean HbA1c 8.60%) to safiglipron 30, 60 or 90 mg or dapagliflozin 10 mg for 32 weeks plus a 20-week extension. HbA1c differences versus dapagliflozin at week 32 were -0.30% (95% CI -0.51 to -0.09), -0.22% (-0.44 to -0.01) and -0.40% (-0.62 to -0.19), establishing non-inferiority for all three doses. Under the treatment policy estimand the 90 mg dose met superiority (difference -0.25%; 95% CI -0.45 to -0.05; one-sided P=0.0067) and the 60 mg dose did not (P=0.0789). Mean body weight changes were -2.35%, -3.55% and -4.17% versus -3.60% on dapagliflozin.

OUTSTAND-1: 284 Chinese adults with type 2 diabetes, placebo-adjusted HbA1c -1.20% to -1.45%

Randomized controlled trial

Yu M et al. Nat Med. 2026. PMID 42811187.

Placebo-adjusted HbA1c differences were -1.22% (95% CI -1.53 to -0.91), -1.20% (-1.52 to -0.89) and -1.45% (-1.75 to -1.14) at 30, 60 and 90 mg, all P<0.0001 (treatment policy estimand). Body weight differences were modest (-0.65%, -2.23% and -3.56% versus placebo). Adverse events led to discontinuation in 1.4%, 2.9% and 6.9% on the three doses and 0% on placebo.

Participants / model
284 Chinese adults with type 2 diabetes managed with diet and exercise (mean HbA1c 7.95%), at 46 sites
Treatment
Safiglipron 30, 60 or 90 mg orally once daily, or placebo
Follow-up
32 weeks
Study design
Phase 3 placebo-controlled trial (NCT06672172), reported in Nature Medicine
Funding
not stated in the source

HbA1c and weight figures are placebo-adjusted differences under the treatment policy estimand.

Read the original source
OUTSTAND-2: 810 Chinese adults with type 2 diabetes on metformin, HbA1c versus dapagliflozin

Randomized controlled trial

Guo L et al. Nat Med. 2026. PMID 42834229.

Differences versus dapagliflozin in HbA1c at week 32 were -0.30% (95% CI -0.51 to -0.09), -0.22% (-0.44 to -0.01) and -0.40% (-0.62 to -0.19) at 30, 60 and 90 mg, establishing non-inferiority for all three doses. Under the treatment policy estimand the 90 mg dose met superiority (difference -0.25%; 95% CI -0.45 to -0.05; one-sided P=0.0067) and the 60 mg dose did not (P=0.0789). Mean body weight changes were -2.35%, -3.55% and -4.17% versus -3.60% on dapagliflozin. Adverse events led to discontinuation in 4.0%, 4.0% and 3.9% versus 1.5%.

Participants / model
810 Chinese adults with type 2 diabetes on metformin (mean HbA1c 8.60%), at 98 sites
Treatment
Safiglipron 30, 60 or 90 mg orally once daily, or dapagliflozin 10 mg
Follow-up
32 weeks plus a 20-week extension
Study design
Phase 3 double-dummy active-comparator trial (NCT06589765), reported in Nature Medicine
Funding
not stated in the source

The comparator was dapagliflozin, not placebo.

Read the original source

What has the company reported from the phase 3 trial in obesity?

On 7 July 2026 the company reported HARBOR-1 in 556 Chinese adults with obesity or overweight: week 44 weight loss of 9.5% and 10.9% at 120 and 180 mg versus 2.5% on placebo (efficacy estimand), and 8.0% and 9.8% versus 2.4% (treatment policy estimand). It is not peer reviewed.

The company reported on 7 July 2026 (Kailera, matching a Hengrui release of the same date) that the phase 3 HARBOR-1 trial (NCT06904105, completed, with no results posted in the registry) randomized 556 adults with obesity or overweight in China 2:2:1 to once-daily HRS-7535 120 mg, 180 mg or placebo. Mean weight was 94.1 kg, mean BMI 34.0, and 62% were female.

At week 44, mean weight loss was 9.5% (120 mg) and 10.9% (180 mg) versus 2.5% on placebo under the efficacy estimand, and 8.0% and 9.8% versus 2.4% under the treatment policy estimand. Under the treatment policy estimand, at least 5% weight loss was reached by 58.6% and 68.2%, at least 10% by 39.6% and 46.6%, and at least 15% by 18.5% and 26.0%. A week 50 ad hoc analysis (efficacy estimand) gave 9.5% and 11.1% versus 2.6%.

These figures are company-reported and not peer reviewed. The releases label two estimands without saying which is primary in the text read. A PubMed search on 10 October 2026 returned 6 records for the compound and none reports HARBOR-1.

Company statement, 7 July 2026: HARBOR-1 in 556 adults, not peer reviewed

Regulatory record

Kailera Therapeutics press release, 7 July 2026, matching a Hengrui release of the same date, read on Yahoo Finance.

The phase 3 HARBOR-1 trial (NCT06904105, completed, no results posted in the registry) randomized 556 adults with obesity or overweight in China (mean weight 94.1 kg, BMI 34.0, 62% female) 2:2:1 to once-daily HRS-7535 120 mg, 180 mg or placebo. At week 44, mean weight loss was 9.5% (120 mg) and 10.9% (180 mg) versus 2.5% on placebo under the efficacy estimand, and 8.0% and 9.8% versus 2.4% under the treatment policy estimand. Under the treatment policy estimand, at least 5% weight loss was reached by 58.6% and 68.2%, at least 10% by 39.6% and 46.6%, and at least 15% by 18.5% and 26.0%. A week 50 ad hoc analysis (efficacy estimand) gave 9.5% and 11.1% versus 2.6%. Nausea occurred in 70.3% (120 mg) and 70.0% (180 mg) versus 16.2% on placebo, vomiting in 66.7% and 68.6% versus 4.5%, and diarrhea in 36.9% and 35.9% versus 15.3%. Discontinuation due to treatment-emergent adverse events was 4.1% (120 mg) and 3.1% (180 mg) versus 2.7% on placebo. The company states no liver safety signal was observed, consistent with prior HRS-7535 trials. Most events were mild to moderate and gastrointestinal.

Participants / model
556 Chinese adults with obesity or overweight (mean weight 94.1 kg, BMI 34.0, 62% female)
Treatment
HRS-7535 120 or 180 mg orally once daily, or placebo, randomized 2:2:1
Follow-up
Week 44 primary time point; week 50 ad hoc analysis
Study design
Company press release on a completed phase 3 trial; not peer reviewed
Funding
not stated in the source

The Kailera and Hengrui releases label two estimands without saying which is primary in the text read.

Read the original source
PubMed search, 10 October 2026: 6 records, none reports HARBOR-1

PubMed search record

PubMed search for HRS-7535 OR safiglipron OR HRS7535, run on 10 October 2026.

A PubMed search for HRS-7535 OR safiglipron OR HRS7535 returned 6 records, namely the phase 1 report, the phase 2 obesity report, a phase 2 diabetic kidney disease trial, a randomized trial in type 2 diabetes inadequately controlled on metformin, and the two Nature Medicine phase 3 type 2 diabetes reports (OUTSTAND-1 and OUTSTAND-2). None reports HARBOR-1, the obesity phase 3 trial, so the 9.8% and 10.9% weight-loss figures remain company statements.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Search run on 10 October 2026
Study design
Dated literature search
Funding
not stated in the source

A search result shows what PubMed returned on that date. It is not proof that no unindexed report exists.

Read the original source

Are there studies outside China?

One: Kailera's global phase 2 obesity trial of KAI-7535 (NCT07497880) in the United States and Australia, active and not recruiting with 306 enrolled and data expected in 2027. It has no results.

Kailera's global phase 2 obesity trial of KAI-7535 (NCT07497880, sponsor Kailera, United States and Australia) is registered as Active, not recruiting, with 306 participants enrolled, registry start on 6 April 2026 and primary completion on 7 July 2027. The primary outcome is percent change in body weight at week 44 in participants without diabetes. Kailera's 7 July 2026 release says the trial tests doses up to 180 mg, and up to 360 mg in a higher-dose cohort, with data expected in 2027.

A registration is not an outcome, and this trial has no results. The peer-reviewed phase 2 and phase 3 trials were all run in China.

Registry: Kailera phase 2 of KAI-7535 outside China, 306 enrolled, data expected 2027

Trial registry record

ClinicalTrials.gov record NCT07497880, read on 10 October 2026; Kailera release of 7 July 2026.

Kailera's global phase 2 obesity trial of KAI-7535 (NCT07497880, sponsor Kailera, United States and Australia) is registered as Active, not recruiting with 306 participants enrolled (actual), registry start 2026-04-06, primary completion 2027-07-07, and primary outcome percent change in body weight at week 44 in participants without diabetes. Kailera's 7 July 2026 release says the trial tests doses up to 180 mg (up to 360 mg in a higher-dose cohort), with data expected in 2027. A registration is not an outcome; there are no results.

Participants / model
Adults with obesity or overweight without diabetes, planned and enrolled
Treatment
KAI-7535 oral tablets once daily
Follow-up
Primary outcome at week 44; primary completion 7 July 2027
Study design
Trial registry record; no results
Funding
not stated in the source

This is the only study outside China in the sources reviewed, and it has no results.

Read the original source

Is safiglipron licensed by Kailera from Hengrui and in phase 3 in China, and what is the Hengrui code?

Yes on all three. Hengrui's code is HRS-7535 and the INN is safiglipron. Three phase 3 trials in China are complete: HARBOR-1 in obesity (company-reported) and OUTSTAND-1 and OUTSTAND-2 in type 2 diabetes (published). Kailera says HRS-7535 was licensed to it outside Greater China in 2024.

Sellers describe the compound as an oral small-molecule GLP-1 agonist licensed by Kailera from Hengrui, in phase 3 in China, with the Hengrui code left open.

PubChem lists HRS-7535 as a synonym of safiglipron alongside KAI-7535, so Hengrui's code is HRS-7535. Kailera says HRS-7535 was developed by Hengrui Pharma and licensed to Kailera outside of Greater China in 2024. The phase 3 claim holds by the registry, company statements and publications: HARBOR-1 is completed, and OUTSTAND-1 and OUTSTAND-2 are published phase 3 trials in type 2 diabetes. The obesity phase 3 results are company statements and are not peer reviewed.

PubChem: safiglipron identity, with HRS-7535 and KAI-7535 codes

Chemical identity record

National Library of Medicine. PubChem compound record CID 166094939 (safiglipron).

PubChem CID 166094939 is safiglipron, listed as an INN, C32H31ClFN3O5, 592.1 g/mol, CAS 2757506-39-9, with HRS-7535, HRS7535, KAI-7535 and KAI7535 among its synonyms, so Hengrui's code is HRS-7535. Kailera's release of 1 October 2026 calls the molecule oral small molecule GLP-1 receptor agonist safiglipron, also known as KAI-7535 (HRS-7535). Its release of 7 July 2026 says HRS-7535 was developed by Hengrui Pharma and licensed to Kailera outside of Greater China in 2024. The 2026 Nature Medicine report of OUTSTAND-2 uses the name safiglipron for the trial registered as NCT06589765 under HRS-7535. The registry sponsor of the China trials is Shandong Suncadia Medicine Co., Ltd.

Participants / model
Chemical identity, not a clinical population
Treatment
Defined molecule, oral tablet
Follow-up
Not applicable
Study design
NLM chemical record, with company statements and registry sponsor information on the code mapping
Funding
not stated in the source

PubChem and Kailera's release use all three names. The registry itself does not tie the sponsor Shandong Suncadia Medicine Co., Ltd. to Hengrui, and Kailera's releases say the trials were conducted by Hengrui.

Read the original source
Company statement, 7 July 2026: HARBOR-1 in 556 adults, not peer reviewed

Regulatory record

Kailera Therapeutics press release, 7 July 2026, matching a Hengrui release of the same date, read on Yahoo Finance.

The phase 3 HARBOR-1 trial (NCT06904105, completed, no results posted in the registry) randomized 556 adults with obesity or overweight in China (mean weight 94.1 kg, BMI 34.0, 62% female) 2:2:1 to once-daily HRS-7535 120 mg, 180 mg or placebo. At week 44, mean weight loss was 9.5% (120 mg) and 10.9% (180 mg) versus 2.5% on placebo under the efficacy estimand, and 8.0% and 9.8% versus 2.4% under the treatment policy estimand. Under the treatment policy estimand, at least 5% weight loss was reached by 58.6% and 68.2%, at least 10% by 39.6% and 46.6%, and at least 15% by 18.5% and 26.0%. A week 50 ad hoc analysis (efficacy estimand) gave 9.5% and 11.1% versus 2.6%. Nausea occurred in 70.3% (120 mg) and 70.0% (180 mg) versus 16.2% on placebo, vomiting in 66.7% and 68.6% versus 4.5%, and diarrhea in 36.9% and 35.9% versus 15.3%. Discontinuation due to treatment-emergent adverse events was 4.1% (120 mg) and 3.1% (180 mg) versus 2.7% on placebo. The company states no liver safety signal was observed, consistent with prior HRS-7535 trials. Most events were mild to moderate and gastrointestinal.

Participants / model
556 Chinese adults with obesity or overweight (mean weight 94.1 kg, BMI 34.0, 62% female)
Treatment
HRS-7535 120 or 180 mg orally once daily, or placebo, randomized 2:2:1
Follow-up
Week 44 primary time point; week 50 ad hoc analysis
Study design
Company press release on a completed phase 3 trial; not peer reviewed
Funding
not stated in the source

The Kailera and Hengrui releases label two estimands without saying which is primary in the text read.

Read the original source
OUTSTAND-1: 284 Chinese adults with type 2 diabetes, placebo-adjusted HbA1c -1.20% to -1.45%

Randomized controlled trial

Yu M et al. Nat Med. 2026. PMID 42811187.

Placebo-adjusted HbA1c differences were -1.22% (95% CI -1.53 to -0.91), -1.20% (-1.52 to -0.89) and -1.45% (-1.75 to -1.14) at 30, 60 and 90 mg, all P<0.0001 (treatment policy estimand). Body weight differences were modest (-0.65%, -2.23% and -3.56% versus placebo). Adverse events led to discontinuation in 1.4%, 2.9% and 6.9% on the three doses and 0% on placebo.

Participants / model
284 Chinese adults with type 2 diabetes managed with diet and exercise (mean HbA1c 7.95%), at 46 sites
Treatment
Safiglipron 30, 60 or 90 mg orally once daily, or placebo
Follow-up
32 weeks
Study design
Phase 3 placebo-controlled trial (NCT06672172), reported in Nature Medicine
Funding
not stated in the source

HbA1c and weight figures are placebo-adjusted differences under the treatment policy estimand.

Read the original source
OUTSTAND-2: 810 Chinese adults with type 2 diabetes on metformin, HbA1c versus dapagliflozin

Randomized controlled trial

Guo L et al. Nat Med. 2026. PMID 42834229.

Differences versus dapagliflozin in HbA1c at week 32 were -0.30% (95% CI -0.51 to -0.09), -0.22% (-0.44 to -0.01) and -0.40% (-0.62 to -0.19) at 30, 60 and 90 mg, establishing non-inferiority for all three doses. Under the treatment policy estimand the 90 mg dose met superiority (difference -0.25%; 95% CI -0.45 to -0.05; one-sided P=0.0067) and the 60 mg dose did not (P=0.0789). Mean body weight changes were -2.35%, -3.55% and -4.17% versus -3.60% on dapagliflozin. Adverse events led to discontinuation in 4.0%, 4.0% and 3.9% versus 1.5%.

Participants / model
810 Chinese adults with type 2 diabetes on metformin (mean HbA1c 8.60%), at 98 sites
Treatment
Safiglipron 30, 60 or 90 mg orally once daily, or dapagliflozin 10 mg
Follow-up
32 weeks plus a 20-week extension
Study design
Phase 3 double-dummy active-comparator trial (NCT06589765), reported in Nature Medicine
Funding
not stated in the source

The comparator was dapagliflozin, not placebo.

Read the original source

How does safiglipron work, and how long does it stay in the body?

Kailera describes it as an oral small-molecule GLP-1 receptor agonist. In healthy participants, median Tmax was 5.98 to 5.99 hours after single doses, and geometric mean half-life was 5.28 to 9.08 hours after single doses and 6.48 to 8.42 hours on day 28.

Kailera's release of 1 October 2026 calls the molecule an oral small molecule GLP-1 receptor agonist. The sources reviewed give no receptor pharmacology beyond that description. The weight and HbA1c results in the trials are the evidence for its effect in people.

In the phase 1 trial in healthy participants, the median Tmax was 5.98 to 5.99 hours after single doses and 5.98 to 10.98 hours on day 28. Geometric mean half-life was 5.28 to 9.08 hours after single doses and 6.48 to 8.42 hours on day 28, and exposure was approximately dose-proportional.

PubChem: safiglipron identity, with HRS-7535 and KAI-7535 codes

Chemical identity record

National Library of Medicine. PubChem compound record CID 166094939 (safiglipron).

PubChem CID 166094939 is safiglipron, listed as an INN, C32H31ClFN3O5, 592.1 g/mol, CAS 2757506-39-9, with HRS-7535, HRS7535, KAI-7535 and KAI7535 among its synonyms, so Hengrui's code is HRS-7535. Kailera's release of 1 October 2026 calls the molecule oral small molecule GLP-1 receptor agonist safiglipron, also known as KAI-7535 (HRS-7535). Its release of 7 July 2026 says HRS-7535 was developed by Hengrui Pharma and licensed to Kailera outside of Greater China in 2024. The 2026 Nature Medicine report of OUTSTAND-2 uses the name safiglipron for the trial registered as NCT06589765 under HRS-7535. The registry sponsor of the China trials is Shandong Suncadia Medicine Co., Ltd.

Participants / model
Chemical identity, not a clinical population
Treatment
Defined molecule, oral tablet
Follow-up
Not applicable
Study design
NLM chemical record, with company statements and registry sponsor information on the code mapping
Funding
not stated in the source

PubChem and Kailera's release use all three names. The registry itself does not tie the sponsor Shandong Suncadia Medicine Co., Ltd. to Hengrui, and Kailera's releases say the trials were conducted by Hengrui.

Read the original source
Phase 1 in healthy participants: median Tmax 5.98 to 10.98 hours, half-life 5.28 to 9.08 hours after single doses

Randomized controlled trial

Wu J et al. Diabetes Obes Metab. 2024;26(3):901-910. PMID 38100147.

Nausea and vomiting were the most frequent adverse events. The median Tmax was 5.98 to 5.99 hours after single doses and 5.98 to 10.98 hours on day 28; geometric mean half-life was 5.28 to 9.08 hours after single doses and 6.48 to 8.42 hours on day 28; exposure was approximately dose-proportional. On day 29, body weight fell 4.38 kg (6.63%) on HRS-7535 versus 0.8 kg (1.18%) on placebo.

Participants / model
Healthy participants
Treatment
HRS-7535 oral, once daily: single doses of 15, 60 and 120 mg; a 90 mg food-effect part; multiple doses reaching 120 mg over 28 days
Follow-up
Single doses; 28 days of multiple doses, with weight reported on day 29
Study design
Randomized placebo-controlled phase 1 trial
Funding
not stated in the source

Ranges are reported for Tmax and half-life, not single values. Weight change in healthy participants over 28 days is not an obesity efficacy result.

Read the original source

What adverse effects have been reported?

Gastrointestinal events dominate. In HARBOR-1, the company reported on 7 July 2026 that nausea occurred in about 70% and vomiting in about 67 to 69% of treated participants versus 16.2% and 4.5% on placebo. Discontinuation for adverse events was 3.1% to 4.1% in HARBOR-1 and up to 6.9% in the type 2 diabetes trials.

In HARBOR-1, the company reported on 7 July 2026 nausea in 70.3% (120 mg) and 70.0% (180 mg) versus 16.2% on placebo, vomiting in 66.7% and 68.6% versus 4.5%, and diarrhea in 36.9% and 35.9% versus 15.3%. Discontinuation due to treatment-emergent adverse events was 4.1% (120 mg) and 3.1% (180 mg) versus 2.7% on placebo. Most events were mild to moderate and gastrointestinal. The company states no liver safety signal was observed, consistent with prior HRS-7535 trials. These figures are company-reported and not peer reviewed.

In the peer-reviewed trials, nausea and vomiting were the most frequent adverse events in the phase 1 trial, and gastrointestinal events were most common in the phase 2 trial, mostly mild to moderate. In OUTSTAND-1, adverse events led to discontinuation in 1.4%, 2.9% and 6.9% of participants on 30, 60 and 90 mg and 0% on placebo. In OUTSTAND-2, discontinuation was 4.0%, 4.0% and 3.9% on 30, 60 and 90 mg versus 1.5% on dapagliflozin.

The Kailera phase 2 trial outside China has no results, and interactions are not described in the sources reviewed.

Company statement, 7 July 2026: HARBOR-1 in 556 adults, not peer reviewed

Regulatory record

Kailera Therapeutics press release, 7 July 2026, matching a Hengrui release of the same date, read on Yahoo Finance.

The phase 3 HARBOR-1 trial (NCT06904105, completed, no results posted in the registry) randomized 556 adults with obesity or overweight in China (mean weight 94.1 kg, BMI 34.0, 62% female) 2:2:1 to once-daily HRS-7535 120 mg, 180 mg or placebo. At week 44, mean weight loss was 9.5% (120 mg) and 10.9% (180 mg) versus 2.5% on placebo under the efficacy estimand, and 8.0% and 9.8% versus 2.4% under the treatment policy estimand. Under the treatment policy estimand, at least 5% weight loss was reached by 58.6% and 68.2%, at least 10% by 39.6% and 46.6%, and at least 15% by 18.5% and 26.0%. A week 50 ad hoc analysis (efficacy estimand) gave 9.5% and 11.1% versus 2.6%. Nausea occurred in 70.3% (120 mg) and 70.0% (180 mg) versus 16.2% on placebo, vomiting in 66.7% and 68.6% versus 4.5%, and diarrhea in 36.9% and 35.9% versus 15.3%. Discontinuation due to treatment-emergent adverse events was 4.1% (120 mg) and 3.1% (180 mg) versus 2.7% on placebo. The company states no liver safety signal was observed, consistent with prior HRS-7535 trials. Most events were mild to moderate and gastrointestinal.

Participants / model
556 Chinese adults with obesity or overweight (mean weight 94.1 kg, BMI 34.0, 62% female)
Treatment
HRS-7535 120 or 180 mg orally once daily, or placebo, randomized 2:2:1
Follow-up
Week 44 primary time point; week 50 ad hoc analysis
Study design
Company press release on a completed phase 3 trial; not peer reviewed
Funding
not stated in the source

The Kailera and Hengrui releases label two estimands without saying which is primary in the text read.

Read the original source
OUTSTAND-1: 284 Chinese adults with type 2 diabetes, placebo-adjusted HbA1c -1.20% to -1.45%

Randomized controlled trial

Yu M et al. Nat Med. 2026. PMID 42811187.

Placebo-adjusted HbA1c differences were -1.22% (95% CI -1.53 to -0.91), -1.20% (-1.52 to -0.89) and -1.45% (-1.75 to -1.14) at 30, 60 and 90 mg, all P<0.0001 (treatment policy estimand). Body weight differences were modest (-0.65%, -2.23% and -3.56% versus placebo). Adverse events led to discontinuation in 1.4%, 2.9% and 6.9% on the three doses and 0% on placebo.

Participants / model
284 Chinese adults with type 2 diabetes managed with diet and exercise (mean HbA1c 7.95%), at 46 sites
Treatment
Safiglipron 30, 60 or 90 mg orally once daily, or placebo
Follow-up
32 weeks
Study design
Phase 3 placebo-controlled trial (NCT06672172), reported in Nature Medicine
Funding
not stated in the source

HbA1c and weight figures are placebo-adjusted differences under the treatment policy estimand.

Read the original source
OUTSTAND-2: 810 Chinese adults with type 2 diabetes on metformin, HbA1c versus dapagliflozin

Randomized controlled trial

Guo L et al. Nat Med. 2026. PMID 42834229.

Differences versus dapagliflozin in HbA1c at week 32 were -0.30% (95% CI -0.51 to -0.09), -0.22% (-0.44 to -0.01) and -0.40% (-0.62 to -0.19) at 30, 60 and 90 mg, establishing non-inferiority for all three doses. Under the treatment policy estimand the 90 mg dose met superiority (difference -0.25%; 95% CI -0.45 to -0.05; one-sided P=0.0067) and the 60 mg dose did not (P=0.0789). Mean body weight changes were -2.35%, -3.55% and -4.17% versus -3.60% on dapagliflozin. Adverse events led to discontinuation in 4.0%, 4.0% and 3.9% versus 1.5%.

Participants / model
810 Chinese adults with type 2 diabetes on metformin (mean HbA1c 8.60%), at 98 sites
Treatment
Safiglipron 30, 60 or 90 mg orally once daily, or dapagliflozin 10 mg
Follow-up
32 weeks plus a 20-week extension
Study design
Phase 3 double-dummy active-comparator trial (NCT06589765), reported in Nature Medicine
Funding
not stated in the source

The comparator was dapagliflozin, not placebo.

Read the original source
Phase 1 in healthy participants: median Tmax 5.98 to 10.98 hours, half-life 5.28 to 9.08 hours after single doses

Randomized controlled trial

Wu J et al. Diabetes Obes Metab. 2024;26(3):901-910. PMID 38100147.

Nausea and vomiting were the most frequent adverse events. The median Tmax was 5.98 to 5.99 hours after single doses and 5.98 to 10.98 hours on day 28; geometric mean half-life was 5.28 to 9.08 hours after single doses and 6.48 to 8.42 hours on day 28; exposure was approximately dose-proportional. On day 29, body weight fell 4.38 kg (6.63%) on HRS-7535 versus 0.8 kg (1.18%) on placebo.

Participants / model
Healthy participants
Treatment
HRS-7535 oral, once daily: single doses of 15, 60 and 120 mg; a 90 mg food-effect part; multiple doses reaching 120 mg over 28 days
Follow-up
Single doses; 28 days of multiple doses, with weight reported on day 29
Study design
Randomized placebo-controlled phase 1 trial
Funding
not stated in the source

Ranges are reported for Tmax and half-life, not single values. Weight change in healthy participants over 28 days is not an obesity efficacy result.

Read the original source
Phase 2: 235 Chinese adults with obesity, 26 weeks, placebo-adjusted -0.49% to -6.87%

Randomized controlled trial

Gu W et al. Nat Commun. 2026;17(1). PMID 42331800.

At week 26 the least-squares mean body weight changes were -2.99%, -7.09%, -6.17% and -9.36% at 30, 60, 120 and 180 mg versus -2.50% on placebo, with placebo-adjusted differences of -0.49%, -4.60%, -3.67% and -6.87% (P = 0.7104, 0.0006, 0.0062 and <0.0001). Gastrointestinal events were most common, mostly mild to moderate. The 30 mg dose did not differ from placebo.

Participants / model
235 Chinese adults with obesity (BMI 28.0 to 40.0) and no diabetes, at 29 centers
Treatment
HRS-7535 at target doses of 30, 60, 120 or 180 mg orally once daily, or placebo, randomized 1:1:1:1:1
Follow-up
26 weeks
Study design
Randomized phase 2 trial (NCT06250946)
Funding
not stated in the source

The changes from baseline and the placebo-adjusted differences are different numbers. The 120 mg arm lost less than the 60 mg arm.

Read the original source

Why does safiglipron sit in B tier?

B fits a published phase 1, a published phase 2 in obesity and two published phase 3 trials in type 2 diabetes. It is not A because the obesity phase 3 is reported only by the company, trials outside China are limited to an ongoing phase 2, weight loss is below the injectable class, nausea and vomiting were frequent in HARBOR-1, and nothing is approved.

The compound has a published phase 1, a published phase 2 in obesity (235 patients, -9.36% at 180 mg versus -2.50% on placebo at 26 weeks), two published phase 3 trials in type 2 diabetes (284 and 810 patients, in Nature Medicine), and a completed 556-patient phase 3 in obesity reported only by a company statement dated 7 July 2026 (8.0% and 9.8% versus 2.4% at 44 weeks, treatment policy estimand).

It stays at B because the obesity phase 3 is unpublished, trials outside China are limited to an ongoing phase 2, weight loss is below the injectable class, nausea occurred in about 70% and vomiting in about 67 to 69% of treated participants in the obesity trial, and nothing is approved.

Phase 1 in healthy participants: median Tmax 5.98 to 10.98 hours, half-life 5.28 to 9.08 hours after single doses

Randomized controlled trial

Wu J et al. Diabetes Obes Metab. 2024;26(3):901-910. PMID 38100147.

Nausea and vomiting were the most frequent adverse events. The median Tmax was 5.98 to 5.99 hours after single doses and 5.98 to 10.98 hours on day 28; geometric mean half-life was 5.28 to 9.08 hours after single doses and 6.48 to 8.42 hours on day 28; exposure was approximately dose-proportional. On day 29, body weight fell 4.38 kg (6.63%) on HRS-7535 versus 0.8 kg (1.18%) on placebo.

Participants / model
Healthy participants
Treatment
HRS-7535 oral, once daily: single doses of 15, 60 and 120 mg; a 90 mg food-effect part; multiple doses reaching 120 mg over 28 days
Follow-up
Single doses; 28 days of multiple doses, with weight reported on day 29
Study design
Randomized placebo-controlled phase 1 trial
Funding
not stated in the source

Ranges are reported for Tmax and half-life, not single values. Weight change in healthy participants over 28 days is not an obesity efficacy result.

Read the original source
Phase 2: 235 Chinese adults with obesity, 26 weeks, placebo-adjusted -0.49% to -6.87%

Randomized controlled trial

Gu W et al. Nat Commun. 2026;17(1). PMID 42331800.

At week 26 the least-squares mean body weight changes were -2.99%, -7.09%, -6.17% and -9.36% at 30, 60, 120 and 180 mg versus -2.50% on placebo, with placebo-adjusted differences of -0.49%, -4.60%, -3.67% and -6.87% (P = 0.7104, 0.0006, 0.0062 and <0.0001). Gastrointestinal events were most common, mostly mild to moderate. The 30 mg dose did not differ from placebo.

Participants / model
235 Chinese adults with obesity (BMI 28.0 to 40.0) and no diabetes, at 29 centers
Treatment
HRS-7535 at target doses of 30, 60, 120 or 180 mg orally once daily, or placebo, randomized 1:1:1:1:1
Follow-up
26 weeks
Study design
Randomized phase 2 trial (NCT06250946)
Funding
not stated in the source

The changes from baseline and the placebo-adjusted differences are different numbers. The 120 mg arm lost less than the 60 mg arm.

Read the original source
OUTSTAND-1: 284 Chinese adults with type 2 diabetes, placebo-adjusted HbA1c -1.20% to -1.45%

Randomized controlled trial

Yu M et al. Nat Med. 2026. PMID 42811187.

Placebo-adjusted HbA1c differences were -1.22% (95% CI -1.53 to -0.91), -1.20% (-1.52 to -0.89) and -1.45% (-1.75 to -1.14) at 30, 60 and 90 mg, all P<0.0001 (treatment policy estimand). Body weight differences were modest (-0.65%, -2.23% and -3.56% versus placebo). Adverse events led to discontinuation in 1.4%, 2.9% and 6.9% on the three doses and 0% on placebo.

Participants / model
284 Chinese adults with type 2 diabetes managed with diet and exercise (mean HbA1c 7.95%), at 46 sites
Treatment
Safiglipron 30, 60 or 90 mg orally once daily, or placebo
Follow-up
32 weeks
Study design
Phase 3 placebo-controlled trial (NCT06672172), reported in Nature Medicine
Funding
not stated in the source

HbA1c and weight figures are placebo-adjusted differences under the treatment policy estimand.

Read the original source
OUTSTAND-2: 810 Chinese adults with type 2 diabetes on metformin, HbA1c versus dapagliflozin

Randomized controlled trial

Guo L et al. Nat Med. 2026. PMID 42834229.

Differences versus dapagliflozin in HbA1c at week 32 were -0.30% (95% CI -0.51 to -0.09), -0.22% (-0.44 to -0.01) and -0.40% (-0.62 to -0.19) at 30, 60 and 90 mg, establishing non-inferiority for all three doses. Under the treatment policy estimand the 90 mg dose met superiority (difference -0.25%; 95% CI -0.45 to -0.05; one-sided P=0.0067) and the 60 mg dose did not (P=0.0789). Mean body weight changes were -2.35%, -3.55% and -4.17% versus -3.60% on dapagliflozin. Adverse events led to discontinuation in 4.0%, 4.0% and 3.9% versus 1.5%.

Participants / model
810 Chinese adults with type 2 diabetes on metformin (mean HbA1c 8.60%), at 98 sites
Treatment
Safiglipron 30, 60 or 90 mg orally once daily, or dapagliflozin 10 mg
Follow-up
32 weeks plus a 20-week extension
Study design
Phase 3 double-dummy active-comparator trial (NCT06589765), reported in Nature Medicine
Funding
not stated in the source

The comparator was dapagliflozin, not placebo.

Read the original source
Company statement, 7 July 2026: HARBOR-1 in 556 adults, not peer reviewed

Regulatory record

Kailera Therapeutics press release, 7 July 2026, matching a Hengrui release of the same date, read on Yahoo Finance.

The phase 3 HARBOR-1 trial (NCT06904105, completed, no results posted in the registry) randomized 556 adults with obesity or overweight in China (mean weight 94.1 kg, BMI 34.0, 62% female) 2:2:1 to once-daily HRS-7535 120 mg, 180 mg or placebo. At week 44, mean weight loss was 9.5% (120 mg) and 10.9% (180 mg) versus 2.5% on placebo under the efficacy estimand, and 8.0% and 9.8% versus 2.4% under the treatment policy estimand. Under the treatment policy estimand, at least 5% weight loss was reached by 58.6% and 68.2%, at least 10% by 39.6% and 46.6%, and at least 15% by 18.5% and 26.0%. A week 50 ad hoc analysis (efficacy estimand) gave 9.5% and 11.1% versus 2.6%. Nausea occurred in 70.3% (120 mg) and 70.0% (180 mg) versus 16.2% on placebo, vomiting in 66.7% and 68.6% versus 4.5%, and diarrhea in 36.9% and 35.9% versus 15.3%. Discontinuation due to treatment-emergent adverse events was 4.1% (120 mg) and 3.1% (180 mg) versus 2.7% on placebo. The company states no liver safety signal was observed, consistent with prior HRS-7535 trials. Most events were mild to moderate and gastrointestinal.

Participants / model
556 Chinese adults with obesity or overweight (mean weight 94.1 kg, BMI 34.0, 62% female)
Treatment
HRS-7535 120 or 180 mg orally once daily, or placebo, randomized 2:2:1
Follow-up
Week 44 primary time point; week 50 ad hoc analysis
Study design
Company press release on a completed phase 3 trial; not peer reviewed
Funding
not stated in the source

The Kailera and Hengrui releases label two estimands without saying which is primary in the text read.

Read the original source
Registry: Kailera phase 2 of KAI-7535 outside China, 306 enrolled, data expected 2027

Trial registry record

ClinicalTrials.gov record NCT07497880, read on 10 October 2026; Kailera release of 7 July 2026.

Kailera's global phase 2 obesity trial of KAI-7535 (NCT07497880, sponsor Kailera, United States and Australia) is registered as Active, not recruiting with 306 participants enrolled (actual), registry start 2026-04-06, primary completion 2027-07-07, and primary outcome percent change in body weight at week 44 in participants without diabetes. Kailera's 7 July 2026 release says the trial tests doses up to 180 mg (up to 360 mg in a higher-dose cohort), with data expected in 2027. A registration is not an outcome; there are no results.

Participants / model
Adults with obesity or overweight without diabetes, planned and enrolled
Treatment
KAI-7535 oral tablets once daily
Follow-up
Primary outcome at week 44; primary completion 7 July 2027
Study design
Trial registry record; no results
Funding
not stated in the source

This is the only study outside China in the sources reviewed, and it has no results.

Read the original source

Studies and sources

PubChem: safiglipron identity, with HRS-7535 and KAI-7535 codes

Chemical identity record

National Library of Medicine. PubChem compound record CID 166094939 (safiglipron).

PubChem CID 166094939 is safiglipron, listed as an INN, C32H31ClFN3O5, 592.1 g/mol, CAS 2757506-39-9, with HRS-7535, HRS7535, KAI-7535 and KAI7535 among its synonyms, so Hengrui's code is HRS-7535. Kailera's release of 1 October 2026 calls the molecule oral small molecule GLP-1 receptor agonist safiglipron, also known as KAI-7535 (HRS-7535). Its release of 7 July 2026 says HRS-7535 was developed by Hengrui Pharma and licensed to Kailera outside of Greater China in 2024. The 2026 Nature Medicine report of OUTSTAND-2 uses the name safiglipron for the trial registered as NCT06589765 under HRS-7535. The registry sponsor of the China trials is Shandong Suncadia Medicine Co., Ltd.

Participants / model
Chemical identity, not a clinical population
Treatment
Defined molecule, oral tablet
Follow-up
Not applicable
Study design
NLM chemical record, with company statements and registry sponsor information on the code mapping
Funding
not stated in the source

PubChem and Kailera's release use all three names. The registry itself does not tie the sponsor Shandong Suncadia Medicine Co., Ltd. to Hengrui, and Kailera's releases say the trials were conducted by Hengrui.

Read the original source
Phase 1 in healthy participants: median Tmax 5.98 to 10.98 hours, half-life 5.28 to 9.08 hours after single doses

Randomized controlled trial

Wu J et al. Diabetes Obes Metab. 2024;26(3):901-910. PMID 38100147.

Nausea and vomiting were the most frequent adverse events. The median Tmax was 5.98 to 5.99 hours after single doses and 5.98 to 10.98 hours on day 28; geometric mean half-life was 5.28 to 9.08 hours after single doses and 6.48 to 8.42 hours on day 28; exposure was approximately dose-proportional. On day 29, body weight fell 4.38 kg (6.63%) on HRS-7535 versus 0.8 kg (1.18%) on placebo.

Participants / model
Healthy participants
Treatment
HRS-7535 oral, once daily: single doses of 15, 60 and 120 mg; a 90 mg food-effect part; multiple doses reaching 120 mg over 28 days
Follow-up
Single doses; 28 days of multiple doses, with weight reported on day 29
Study design
Randomized placebo-controlled phase 1 trial
Funding
not stated in the source

Ranges are reported for Tmax and half-life, not single values. Weight change in healthy participants over 28 days is not an obesity efficacy result.

Read the original source
Phase 2: 235 Chinese adults with obesity, 26 weeks, placebo-adjusted -0.49% to -6.87%

Randomized controlled trial

Gu W et al. Nat Commun. 2026;17(1). PMID 42331800.

At week 26 the least-squares mean body weight changes were -2.99%, -7.09%, -6.17% and -9.36% at 30, 60, 120 and 180 mg versus -2.50% on placebo, with placebo-adjusted differences of -0.49%, -4.60%, -3.67% and -6.87% (P = 0.7104, 0.0006, 0.0062 and <0.0001). Gastrointestinal events were most common, mostly mild to moderate. The 30 mg dose did not differ from placebo.

Participants / model
235 Chinese adults with obesity (BMI 28.0 to 40.0) and no diabetes, at 29 centers
Treatment
HRS-7535 at target doses of 30, 60, 120 or 180 mg orally once daily, or placebo, randomized 1:1:1:1:1
Follow-up
26 weeks
Study design
Randomized phase 2 trial (NCT06250946)
Funding
not stated in the source

The changes from baseline and the placebo-adjusted differences are different numbers. The 120 mg arm lost less than the 60 mg arm.

Read the original source
Company statement, 7 July 2026: HARBOR-1 in 556 adults, not peer reviewed

Regulatory record

Kailera Therapeutics press release, 7 July 2026, matching a Hengrui release of the same date, read on Yahoo Finance.

The phase 3 HARBOR-1 trial (NCT06904105, completed, no results posted in the registry) randomized 556 adults with obesity or overweight in China (mean weight 94.1 kg, BMI 34.0, 62% female) 2:2:1 to once-daily HRS-7535 120 mg, 180 mg or placebo. At week 44, mean weight loss was 9.5% (120 mg) and 10.9% (180 mg) versus 2.5% on placebo under the efficacy estimand, and 8.0% and 9.8% versus 2.4% under the treatment policy estimand. Under the treatment policy estimand, at least 5% weight loss was reached by 58.6% and 68.2%, at least 10% by 39.6% and 46.6%, and at least 15% by 18.5% and 26.0%. A week 50 ad hoc analysis (efficacy estimand) gave 9.5% and 11.1% versus 2.6%. Nausea occurred in 70.3% (120 mg) and 70.0% (180 mg) versus 16.2% on placebo, vomiting in 66.7% and 68.6% versus 4.5%, and diarrhea in 36.9% and 35.9% versus 15.3%. Discontinuation due to treatment-emergent adverse events was 4.1% (120 mg) and 3.1% (180 mg) versus 2.7% on placebo. The company states no liver safety signal was observed, consistent with prior HRS-7535 trials. Most events were mild to moderate and gastrointestinal.

Participants / model
556 Chinese adults with obesity or overweight (mean weight 94.1 kg, BMI 34.0, 62% female)
Treatment
HRS-7535 120 or 180 mg orally once daily, or placebo, randomized 2:2:1
Follow-up
Week 44 primary time point; week 50 ad hoc analysis
Study design
Company press release on a completed phase 3 trial; not peer reviewed
Funding
not stated in the source

The Kailera and Hengrui releases label two estimands without saying which is primary in the text read.

Read the original source
OUTSTAND-2: 810 Chinese adults with type 2 diabetes on metformin, HbA1c versus dapagliflozin

Randomized controlled trial

Guo L et al. Nat Med. 2026. PMID 42834229.

Differences versus dapagliflozin in HbA1c at week 32 were -0.30% (95% CI -0.51 to -0.09), -0.22% (-0.44 to -0.01) and -0.40% (-0.62 to -0.19) at 30, 60 and 90 mg, establishing non-inferiority for all three doses. Under the treatment policy estimand the 90 mg dose met superiority (difference -0.25%; 95% CI -0.45 to -0.05; one-sided P=0.0067) and the 60 mg dose did not (P=0.0789). Mean body weight changes were -2.35%, -3.55% and -4.17% versus -3.60% on dapagliflozin. Adverse events led to discontinuation in 4.0%, 4.0% and 3.9% versus 1.5%.

Participants / model
810 Chinese adults with type 2 diabetes on metformin (mean HbA1c 8.60%), at 98 sites
Treatment
Safiglipron 30, 60 or 90 mg orally once daily, or dapagliflozin 10 mg
Follow-up
32 weeks plus a 20-week extension
Study design
Phase 3 double-dummy active-comparator trial (NCT06589765), reported in Nature Medicine
Funding
not stated in the source

The comparator was dapagliflozin, not placebo.

Read the original source
OUTSTAND-1: 284 Chinese adults with type 2 diabetes, placebo-adjusted HbA1c -1.20% to -1.45%

Randomized controlled trial

Yu M et al. Nat Med. 2026. PMID 42811187.

Placebo-adjusted HbA1c differences were -1.22% (95% CI -1.53 to -0.91), -1.20% (-1.52 to -0.89) and -1.45% (-1.75 to -1.14) at 30, 60 and 90 mg, all P<0.0001 (treatment policy estimand). Body weight differences were modest (-0.65%, -2.23% and -3.56% versus placebo). Adverse events led to discontinuation in 1.4%, 2.9% and 6.9% on the three doses and 0% on placebo.

Participants / model
284 Chinese adults with type 2 diabetes managed with diet and exercise (mean HbA1c 7.95%), at 46 sites
Treatment
Safiglipron 30, 60 or 90 mg orally once daily, or placebo
Follow-up
32 weeks
Study design
Phase 3 placebo-controlled trial (NCT06672172), reported in Nature Medicine
Funding
not stated in the source

HbA1c and weight figures are placebo-adjusted differences under the treatment policy estimand.

Read the original source
Company statement, 7 July 2026: NDA plans in China; no US application or label found

Regulatory record

Jiangsu Hengrui Pharmaceuticals announcement, 7 July 2026, on hengrui.com; Kailera release of 26 May 2026; openFDA Drugs@FDA and DailyMed searches run on 10 October 2026.

Hengrui says it plans to submit NDAs for HRS-7535 for the treatment of type 2 diabetes and obesity in China. The release does not state that an application has been accepted. Kailera's release of 26 May 2026 likewise says it plans to submit an NDA for type 2 diabetes. On 10 October 2026 an openFDA Drugs@FDA search by generic name safiglipron returned no records and a DailyMed search returned 0 labels, so there is no US application or label.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Statements dated 26 May and 7 July 2026; searches run on 10 October 2026
Study design
Company statement and dated regulatory search
Funding
not stated in the source

A plan to submit is not a submission, and a submission is not an approval. The US search shows what the databases held on that date.

Read the original source
Repost, 2 September 2026: NMPA acceptance notice for the type 2 diabetes application

Regulatory record

Brokerage repost dated 2 September 2026 of a Hengrui announcement, read on iTiger.

A brokerage repost dated 2 September 2026 of a Hengrui announcement says its subsidiary, Shandong Shengdi Pharmaceutical Co., Ltd., received an NMPA Acceptance Notice for the marketing authorization application for HRS-7535 tablets in adult type 2 diabetes, based on OUTSTAND-1 and OUTSTAND-2. The repost read does not mention an obesity application; one aggregator reported an obesity acceptance that was not confirmed. Acceptance of a filing is not approval, and no approval exists in any source reviewed.

Participants / model
Adults with type 2 diabetes, planned label
Treatment
HRS-7535 tablets
Follow-up
Statement dated 2 September 2026
Study design
News repost of a company announcement; not a regulator record
Funding
not stated in the source

The announcing subsidiary Shandong Shengdi Pharmaceutical Co., Ltd. is not tied to Hengrui by the registry itself. Acceptance for obesity is unconfirmed.

Read the original source
Registry: Kailera phase 2 of KAI-7535 outside China, 306 enrolled, data expected 2027

Trial registry record

ClinicalTrials.gov record NCT07497880, read on 10 October 2026; Kailera release of 7 July 2026.

Kailera's global phase 2 obesity trial of KAI-7535 (NCT07497880, sponsor Kailera, United States and Australia) is registered as Active, not recruiting with 306 participants enrolled (actual), registry start 2026-04-06, primary completion 2027-07-07, and primary outcome percent change in body weight at week 44 in participants without diabetes. Kailera's 7 July 2026 release says the trial tests doses up to 180 mg (up to 360 mg in a higher-dose cohort), with data expected in 2027. A registration is not an outcome; there are no results.

Participants / model
Adults with obesity or overweight without diabetes, planned and enrolled
Treatment
KAI-7535 oral tablets once daily
Follow-up
Primary outcome at week 44; primary completion 7 July 2027
Study design
Trial registry record; no results
Funding
not stated in the source

This is the only study outside China in the sources reviewed, and it has no results.

Read the original source
PubMed search, 10 October 2026: 6 records, none reports HARBOR-1

PubMed search record

PubMed search for HRS-7535 OR safiglipron OR HRS7535, run on 10 October 2026.

A PubMed search for HRS-7535 OR safiglipron OR HRS7535 returned 6 records, namely the phase 1 report, the phase 2 obesity report, a phase 2 diabetic kidney disease trial, a randomized trial in type 2 diabetes inadequately controlled on metformin, and the two Nature Medicine phase 3 type 2 diabetes reports (OUTSTAND-1 and OUTSTAND-2). None reports HARBOR-1, the obesity phase 3 trial, so the 9.8% and 10.9% weight-loss figures remain company statements.

Participants / model
Records reviewed, not a population
Treatment
Not applicable
Follow-up
Search run on 10 October 2026
Study design
Dated literature search
Funding
not stated in the source

A search result shows what PubMed returned on that date. It is not proof that no unindexed report exists.

Read the original source

Why is Safiglipron in B tier?

B fits a compound with a published phase 1, a published phase 2 in obesity and two published phase 3 trials in type 2 diabetes. It is not A because the obesity phase 3 is reported only by the company, trials outside China are limited to an ongoing phase 2, weight loss is below the injectable class, nausea and vomiting were frequent in HARBOR-1, and nothing is approved.