UBT251, a Novel Triple GLP-1, GIP, and Glucagon Receptor Agonist, in Adults with Overweight or Obesity: A Multicenter, Randomized Phase 2 Trial
Cell Metabolism, 2026 · paper · Clinical trials
A randomized, double-blind, placebo-controlled phase 2 trial of UBT251, a triple GLP-1, GIP and glucagon receptor agonist, in 205 adults with overweight or obesity. At 24 weeks, placebo-adjusted body weight change ranged from −11.5% to −17.7% across doses.
5 findings
Substance Table S1, 4 mg (ID 1 mg): "Reached the maximum dose" is "21 (84.0)" of 25 in the full analysis population but "22 (88.0)" of 25 among study completers, although all 25 completed the study, so the two rows describe the same participants.
- p. 24 Table S1, row 'Reached the maximum dose' under Full analysis population, 4 mg (ID 1 mg)
21 (84.0)
- p. 24 Table S1, row 'Reached the maximum dose' under Completed the study, 4 mg (ID 1 mg)
22 (88.0)
- p. 3 Figure 1, 4 mg (ID 1 mg) arm
25 completed the study
Only the authors can say what is right here.
Judge's note: Page 24 image: full analysis population 25 with 21 (84.0) reached; completed the study 25 with 22 (88.0) reached. Figure 1 (page 3 image) shows "25 completed the study" for this arm, so completers are the full group and cannot contain more people reaching the dose than the full set. The 4 mg total column inherits the problem (45 in both rows, from 24+21 and 23+22). The percentages are each consistent with their own counts. No footnote or text (results page 4, table footnote) gives a different definition for the two rows. Only the authors can say which count is right.
Second check: another model agreed
Report a problem with this finding
Substance Table S3, HOMA-IR, N (missing), 2 mg (N=51) column reads "51 (1)": analysed plus missing is 52 in a group of 51; every other 2 mg row sums to 51 (e.g. "50 (1)").
- p. 27 Table S3, HOMA-IR, N (missing), 2 mg (N=51) column
51 (1)
Only the authors can say what is right here.
Judge's note: Read on the page 27 image: the cell is printed as 51 (1). Fasting insulin, which HOMA-IR is derived from, is 50 (1) in the same column. The other HOMA-IR cells sum to their group sizes (24+1, 25+0, 49+1, 46+8, 45+5). No text explains a different denominator. Probably 50 (1), but only the authors can say whether the count or the missing figure is wrong.
Second check: another model agreed
Report a problem with this finding
Reference or label STAR Methods point to the supplement for content it does not contain: "Detailed eligibility criteria are provided in the Supplemental information", "Detailed dose-escalation and dose-adjustment algorithms are provided in the Supplemental information", "Protocol-specified prohibited medications are listed in the Supplemental information", "(see Supplemental information for detailed definitions)".
- p. 14 STAR Methods, Human participants
Detailed eligibility criteria are provided in the Supplemental information.
- p. 18 STAR Methods, Dose escalation and adjustment
Detailed dose-escalation and dose-adjustment algorithms are provided in the Supplemental information.
- p. 18 STAR Methods, Dose escalation and adjustment
Protocol-specified prohibited medications are listed in the Supplemental information.
- p. 19 STAR Methods, Statistical analysis populations
(see Supplemental information for detailed definitions)
Judge's note: The supplement's table of contents (page 22 image) lists only the chemical structure, Tables S1–S5 and Figures S1–S4, with no supplementary methods. The inclusion and exclusion criteria (e2–e4), dose adjustment criteria and prohibited medications (e4–e5) and the complete per-protocol definition (e5) are printed in the STAR Methods themselves. Only the dose-escalation schedule (Table S5) is in the supplement. The pointers should direct to the corresponding STAR Methods subsections. The text layer of page 22 also carries a hidden older table of contents listing "Supplementary methods", but that section is not on the page image or in the supplement pages.
Second check: another model agreed
Report a problem with this finding
Reference or label Figure S3 panels A to D (response rates) head the between-group difference column "Diff. LS Mean (95%CI)", but the values are differences in response proportions (e.g. 90.2 vs 22.0 gives 68.2), not least-squares mean differences.
- p. 35 Figure S3A, right-hand column header
Diff. LS Mean (95%CI)
- p. 20 STAR Methods, Secondary and exploratory analyses
between-group differences in response rates were estimated with corresponding 95% CIs calculated by the Miettinen-Nurminen method
Judge's note: Page images 34, 35, 36 and 38: the group columns in panels A to D are headed 'n/N rate (95%CI)' while the right-hand column is headed 'Diff. LS Mean (95%CI)'. Its values equal the simple rate differences. The methods (page 20) say that 'between-group differences in response rates were estimated with corresponding 95% CIs calculated by the Miettinen-Nurminen method'. The header is a mislabel carried over from the continuous-outcome panels.
Second check: another model agreed
Report a problem with this finding
Wording Methods assign non-responder imputation to the treatment-regimen estimand ("For binary efficacy endpoints, a nonresponder imputation approach was used") and multiple imputation to the primary estimand, but the Figure 2 legend says "Analyses were based on the efficacy estimand ... non-responder imputation for (C)" and Figure S3 presents NRI results as "Subgroup analysis of the primary estimand".
- p. 20 STAR Methods, Secondary and exploratory analyses
For binary efficacy endpoints, a nonresponder imputation approach was used.
- p. 6 Figure 2 legend (continued)
Analyses were based on the efficacy estimand
- p. 34 Figure S3 legend
Subgroup analysis of the primary estimand were conducted using the same model as the main analysis.
- p. 6 Figure 2 legend
non-responder imputation for (C)
Only the authors can say what is right here.
Judge's note: Page 20 image: binary endpoints under the primary (efficacy) estimand use multiple imputation, and the nonresponder imputation sentence sits under the additional treatment-regimen estimand. Page 6 image: the Figure 2 legend applies the efficacy estimand to all panels, including NRI panel C. Page 34 image: Figure S3 panels A–D use NRI and are labelled primary estimand. No statement was found in the methods or supplement describing NRI under the efficacy estimand. The authors should say which estimand the NRI analyses used.
Second check: another model agreed
Report a problem with this finding
25 minor (typos and the like; most checked by the judge only)
Wording Table S4, 2 mg group, Week 28 N (missing) reads "51 (0)", although Figure 1 says "1 withdrew from the study" and "50 completed the study" and Table S4 shows "50 (1)" at Weeks 12 to 24.
- p. 29 Table S4, Change from baseline to Week 28, N (missing), 2 mg
51 (0)
- p. 3 Figure 1, 2 mg arm
50 completed the study
- p. 24 Table S1, Completed the study, 2 mg
50
Only the authors can say what is right here.
Judge's note: Page 29 image: 2 mg is 50 (1) at Weeks 12, 16, 20 and 24, then 51 (0) at Week 28. Figure 1 (page 3) and Table S1 (page 24) give 50 study completers in the 2 mg group. The Week 28 counts for 4 mg ID 0.5 (23), 6 mg (47) and placebo (45) equal their completer counts; 4 mg ID 1 is 24 against 25 completers. The methods describe a 4-week post-treatment follow-up but say nothing about how early-termination visits are windowed, so the 51 is unexplained. It could be a typo for 50 (1) or an early-termination visit mapped to Week 28; this is a question for the authors.
Second check: another model agreed
Report a problem with this finding
Wording Table 1 gives baseline "Pulse rate, beats per minute" as 81.5 (9.0), 80.4 (8.4), 81.2 (6.8), 80.8 (7.6), 78.4 (8.4), 79.1 (8.3); Table S4 gives baseline heart rate for the same 205 participants as "76.2 (10.2) 74.5 (8.3)" 74.3 (9.8), 74.4 (9.0), 72.5 (9.4), 71.8 (9.1), about 5 to 7 bpm lower, with no stated difference in source.
- p. 29 Table S4, Baseline heart rate row
76.2 (10.2) 74.5 (8.3)
- p. 4 Table 1, Pulse rate row
81.5 (9.0)
Only the authors can say what is right here.
Judge's note: Both tables read on the page images (pages 4 and 29). The full analysis and safety populations are the same randomised participants (page 3). Searched the text for heart rate, pulse rate, ECG and vital signs: nothing says the two come from different measurements (for example vital-sign pulse versus ECG heart rate) or timepoints. They may well be legitimately different measurements, so this is a query asking the authors to state the source of each, not a demonstrated error.
Second check: another model agreed
Report a problem with this finding
Wording Results report no AESIs of "hepatic dysfunction, or injection-site reactions", but the Methods list of pre-specified AESIs has no injection-site reactions; "hypoglycemic events" is in the Methods list and is not reported in the Results text or Table 3.
- p. 6 Results, Safety
hepatic dysfunction, or injection-site reactions were observed during the study
- p. 19 STAR Methods, Endpoints and assessments
Pre-specified AESIs included severe gastrointestinal events, hypoglycemic events, acute pancreatitis, musculoskeletal pain
Judge's note: Viewed the Safety text on PDF page 6 and Table 3 on page 9, and read the Methods AESI list in the text layer (page 19). The Methods list is severe gastrointestinal events, hypoglycemic events, acute pancreatitis, musculoskeletal pain, supraventricular arrhythmias, cardiac conduction disorders or QTc prolongation, liver function abnormality, acute kidney injury, systemic hypersensitivity reactions and cholelithiasis. The list is introduced by "included", so it may not be exhaustive, and hypoglycemia may simply have had no events. A minor consistency point.
Second check: another model agreed
Report a problem with this finding
Wording Table 3 lists treatment-related "Severe gastrointestinal reactions" as "0 1 (4.0) 1 (4.0) 2 (4.0) 2 (3.7) 4 (2.6)" while the TRAE "Grade ≥3" row is "1 (2.0) 0 0 0 4 (7.4) 5 (3.2)": severe GI reactions are reported in both 4 mg groups, where grade ≥3 TRAEs are zero.
- p. 9 Table 3, TRAE, Grade ≥3 row, 4 mg (ID 0.5 mg), 4 mg (ID 1 mg), 4 mg total
Grade ≥3 1 (2.0) 0 0 0 4 (7.4) 5 (3.2)
- p. 9 Table 3, Treatment-related AESIs, Severe gastrointestinal reactions
0 1 (4.0) 1 (4.0) 2 (4.0) 2 (3.7) 4 (2.6)
- p. 6 Figure S4, Diarrhoea row, 4 mg (ID 1 mg) group
Grade 3
- p. 11 Discussion, safety paragraph (end)
with severe diarrhea observed only during treatment initiation
Only the authors can say what is right here.
Judge's note: Viewed Table 3 on PDF page 9 and Figure S4 on page 43; both rows are printed as quoted. Figure S4 shows a grade 3 diarrhoea segment in the 4 mg (ID 1 mg) panel and none in the 4 mg (ID 0.5 mg) panel, so "severe" in the AESI row may not mean CTCAE grade 3. I found no definition of the AESI "severe gastrointestinal reactions" in the article or supplement text. Only the authors can say how the two rows fit together. The second reviewer note, on the timing of severe diarrhoea, is the same point as C006 and is judged there.
Second check: another model agreed
Report a problem with this finding
Wording Discussion says "another phase 2 trial (NCT07163624) of UBT251 in type 2 diabetes confers clinically meaningful improvements in glycemic control alongside weight loss" (no citation), while Limitations says UBT251 "is also being evaluated in a separate clinical trial in participants with type 2 diabetes (NCT07163624), which will provide complementary evidence".
- p. 11 Discussion, first column
another phase 2 trial (NCT07163624) of UBT251 in type 2 diabetes confers clinically meaningful improvements in glycemic control
- p. 11 Limitations of the study
UBT251 is also being evaluated in a separate clinical trial in participants with type 2 diabetes (NCT07163624), which will provide complementary evidence
Only the authors can say what is right here.
Judge's note: Viewed the page image for PDF page 11; both passages are printed as quoted. The same trial is presented once as having shown a result, with no source, and once as evidence still to come. The authors need to clarify the trial's status and give a source for the result claim.
Second check: another model agreed
Report a problem with this finding
Typo Results subheading reads "Participants deposition and baseline characteristics"; the section describes participant flow, so "disposition" is the intended word.
- p. 3 Results, first subheading
Participants deposition and baseline characteristics
Judge's note: Viewed the page image for PDF page 3. The heading is printed as quoted, and the paragraph under it covers screening, randomization and analysis populations.
Second check: none (judge only)
Report a problem with this finding
Typo "25 to UBT251 4 mg (ID 0.5 mg), 25 to UBT251 4 mg (initial dose [ID] 1 mg)": the abbreviation ID is used one mention before it is defined in the body text.
- p. 3 Results, first paragraph
25 to UBT251 4 mg (ID 0.5 mg), 25 to UBT251 4 mg (initial dose [ID] 1 mg)
Judge's note: Viewed the page image for PDF page 3; the text is as quoted. ID is defined in the Figure 1 legend on the same page, but in the running text the definition comes at the second use. Cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo "These findings suggest that triple agonist may further enhance weight-loss efficacy beyond that achieved with single- or dual-receptor agonists": singular count noun without article (should be "triple agonists" or "triple agonism").
- p. 3 Introduction, end of second paragraph
that triple agonist may further enhance weight-loss efficacy
- p. 3 Introduction, last paragraph before UBT251 paragraph
These findings suggest that triple agonist may further enhance weight-loss efficacy
Judge's note: Viewed the page image for PDF page 3, end of the Introduction's second paragraph. The wording is printed as quoted. Grammar only; the meaning is clear.
Second check: none (judge only)
Report a problem with this finding
Typo Table 1 prints "34. 1 (9.1)" and "34. 1 (8.4)" in the Age row (4 mg total, Total) and "122. 1 (11.5)" in the SBP row (Placebo), with a stray space after the decimal point.
- p. 4 Table 1, Age row (4 mg total and Total) and SBP row (Placebo)
34. 1 (9.1)
- p. 4 Table 1, SBP, Placebo
122. 1 (11.5)
Judge's note: Viewed the Table 1 crop and the full page image. The spaces are in the typeset table in all three cells, not only in the text layer.
Second check: none (judge only)
Report a problem with this finding
Typo Row labels read "Difference vs., placebo" throughout Table 2 and "Difference vs, placebo" throughout Table S2; the comma is stray.
- p. 7 Table 2, every difference row
Difference vs., placebo
- p. 25 Table S2, every difference row
Difference vs, placebo
Judge's note: Viewed both crops and the full Table 2 page. The comma is in the typeset labels on every difference row of both tables.
Second check: none (judge only)
Report a problem with this finding
Typo Table 2 footnote prints "HDLcholesterol, high-density lipoprotein cholesterol" (missing space), and the key resources table prints "Meso Scale Diagnostics,LLC" (missing space after the comma).
- p. 8 Table 2 footnote
HDLcholesterol, high-density lipoprotein cholesterol
- p. 14 Key resources table
Meso Scale Diagnostics,LLC
Judge's note: Viewed both crops; both are in the typeset page. The Table 1 footnote on page 4 writes "HDL cholesterol" with a space.
Second check: none (judge only)
Report a problem with this finding
Typo Table 3 footnote defines "IQR, interquartile range", but no interquartile range appears in Table 3.
- p. 10 Table 3 footnote
IQR, interquartile range
Judge's note: Viewed the footnote crop from page 10 and the first part of Table 3 on page 9. All entries are number (%), and the footnote itself says "Data are presented as number (%)". A leftover abbreviation.
Second check: none (judge only)
Report a problem with this finding
Typo Acknowledgments state the funding twice: "This study was supported by The United Bio-Technology (Hengqin) Co., Ltd., which was involved in the study design..." and again as the last sentence "This study was supported by The United Bio-Technology (Hengqin) Co., Ltd."
- p. 12 Acknowledgments
This study was supported by The United Bio-Technology (Hengqin) Co., Ltd.
Judge's note: Read the Acknowledgments on the page 12 image. The funding sentence appears mid-paragraph (with a continuation clause) and is repeated as the closing sentence. Redundant duplication, cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Declaration of interests prints malformed initials "SB.S. and T.H.-H. are employees of Novo Nordisk"; the same author is "S.B.S." in Author contributions.
- p. 12 Declaration of Interests
SB.S. and T.H.-H. are employees of Novo Nordisk
- p. 12 Author Contributions
Q.D. Xiaoxia Shi, Xiaoguang Shi
- p. 12 Declaration of interests
SB.S. and T.H.-H. are employees of Novo Nordisk.
Judge's note: Page 12 image: the Declaration of interests reads "SB.S.", while Author contributions uses "S.B.S." three times (author Sofie Bay Simony). Typographic slip. The missing comma after "Q.D." raised in the same cluster is handled under C019.
Second check: none (judge only)
Report a problem with this finding
Typo Data and code availability, last bullet: "Any additional information required to reanalyze the data reported in this work paper is available from the lead contact upon request." ("work paper" has a superfluous noun).
- p. 12 Data and code availability, third bullet
reported in this work paper is available
- p. 12 Data and code availability, last bullet
reanalyze the data reported in this work paper
Judge's note: Page 12 image shows "reported in this work paper" in the final bullet of Data and code availability. Either "work" or "paper" is redundant. Both quoted locations are the same single sentence.
Second check: none (judge only)
Report a problem with this finding
Typo Author contributions: "Z.Z., Z.C., P.J., G.Y., H.W., B.S., Q.D. Xiaoxia Shi, Xiaoguang Shi, H.S., ..." lacks a comma after "Q.D.".
- p. 12 Author contributions
Q.D. Xiaoxia Shi, Xiaoguang Shi
Judge's note: Page 12 image confirms "Q.D. Xiaoxia Shi, Xiaoguang Shi" with no comma after Q.D. The full names for the two Shi authors plausibly disambiguate identical initials (X.S.), so only the missing comma is a defect.
Second check: none (judge only)
Report a problem with this finding
Typo The prohibited-medication list item "● Growth hormone and its analogs" is typeset as a coloured section heading on e4 and so appears as a subsection "○ ● Growth hormone and its analogs" in the STAR Methods outline.
- p. 17 STAR Methods, Prohibited concomitant medications/treatments
● Growth hormone and its analogs
- p. 12 STAR★METHODS outline
○ ● Growth hormone and its analogs
- p. 12 STAR Methods contents list
Growth hormone and its analogs
Judge's note: Page 17 image: under "Prohibited concomitant medications/treatments" the second bullet, "Growth hormone and its analogs", is set in red bold heading style and splits the bulleted list. Page 12 image: the STAR Methods outline lists "○ ● Growth hormone and its analogs" as a method heading with a double bullet. Visible in the images, so this is a typesetting error and not an extraction artefact.
Second check: none (judge only)
Report a problem with this finding
Typo Reference 10 title printed as "Glucagon-like receptor agonists and next-generation incretin-based medications: metabolic, cardiovascular, and renal benefits"; "peptide-1" appears to be missing.
- p. 13 Reference list, ref 10
Glucagon- like receptor agonists and next-generation incretin-based medications
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: the true title of the cited Lancet article (Nauck et al.), believed to begin "Glucagon-like peptide-1 receptor agonists". The page 13 image shows the entry as quoted. "Glucagon-like receptor agonists" is not a term used elsewhere in the paper, which writes "glucagon-like peptide-1 (GLP-1) receptor agonists". The title cannot be checked from the files, so the authors should verify.
Second check: none (judge only)
Report a problem with this finding
Typo Reference 14 title has a stray space before the colon: "Among Adults Without Diabetes : A Systematic Review of Randomized Controlled Trials".
- p. 13 Reference list, ref 14
Among Adults Without Diabetes : A Systematic Review
Judge's note: Page 13 image shows "Diabetes : A Systematic Review" with a space before the colon. Cosmetic.
Second check: none (judge only)
Report a problem with this finding
Typo Key resources table lists "CAMP GS Dynamic Kit (NAb)" (Revvity, 62AM4PEB); the conventional capitalisation would be "cAMP Gs".
- p. 14 Key resources table
CAMP GS Dynamic Kit (NAb)
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: the vendor's product name is believed to be "cAMP Gs Dynamic" kit, cAMP being cyclic AMP and Gs the G-protein subtype. The page 14 image shows "CAMP GS Dynamic Kit (NAb)" as quoted. Capitalisation only.
Second check: none (judge only)
Report a problem with this finding
Typo Key resources table prints "Wuxi New Drug Development Co., Ltd. Shanghai" (lipid panel, HbA1c) and "WuXi AppTec" (anti-UBT251 antibody), with different capitalisation of Wuxi/WuXi.
- p. 14 Key resources table
Wuxi New Drug Development Co., Ltd.
- p. 14 Key resources table
WuXi AppTec
Only the authors can say what is right here.
Judge's note: Rests on outside knowledge: that both entries are the same WuXi AppTec company family, whose branding is "WuXi". The page 14 image shows both spellings as quoted. They may be different legal entities and either capitalisation could be intended, so this is a minor consistency query for the authors.
Second check: none (judge only)
Report a problem with this finding
Typo Randomization and blinding: "blinding was not maintained between the active-dose group" uses the singular where the plural "groups" is required.
- p. 15 STAR Methods, Randomization and blinding
blinding was not maintained between the active-dose group
Judge's note: Page 15 image shows the sentence as quoted. "Between" several UBT251 arms requires the plural; the preceding clause refers to "UBT251 treatment groups". Grammatical slip.
Second check: none (judge only)
Report a problem with this finding
Typo Table S2, Supplementary analysis 2, placebo: "-2.30 (-4.2, -0.4)" is printed to two decimals while every other estimate in the table uses one decimal.
- p. 25 Table S2, Supplementary analysis 2, Placebo
-2.30 (-4.2, -0.4)
Judge's note: Page 25 image confirms "-2.30 (-4.2, -0.4)" in the placebo column. All other least-squares means and confidence limits in Table S2 have one decimal. Formatting inconsistency; the value agrees with its interval.
Second check: none (judge only)
Report a problem with this finding
Typo Figure S1 and Figure S3 legends define "FPG, fasting plasma glucose", but every forest-plot panel labels the subgroup "Baseline FBG", an abbreviation that is not defined.
- p. 32 Figure S1 forest plot row label
Baseline FBG
- p. 31 Figure S1 legend
FPG, fasting plasma glucose
- p. 34 Figure S3 legend and panel A
FPG, fasting plasma glucose
Judge's note: Seen on the page images: the legends on pages 31 and 34 define FPG; the panels on pages 31, 32, 34, 35, 36 and 38 all read 'Baseline FBG'. The meaning is not in doubt; this is an abbreviation mismatch between legend and figure.
Second check: none (judge only)
Report a problem with this finding
Typo Figure S3 legend: "Subgroup analysis of the primary estimand were conducted using the same model as the main analysis." The singular subject takes a plural verb.
- p. 34 Figure S3 legend, penultimate paragraph
Subgroup analysis of the primary estimand were conducted using the same model as the main analysis.
Judge's note: Read on the page 34 image; the sentence is as quoted. The Figure S1 legend uses 'Subgroup analyses were performed'. Grammar only.
Second check: none (judge only)
Report a problem with this finding
1 more finding is waiting for a second check and not shown.