Effects of Semaglutide Treatment on Psoriatic Lesions in Obese Patients with Type 2 Diabetes Mellitus: An Open-Label, Randomized Clinical Trial
- Design
- Prospective cohort · 31 participants · Biomarker outcome
- Match to healthy normal-weight adults aged 55–75
- Different population[Auto] Participants had type 2 diabetes and obesity/overweight (age/BMI not reported in abstract).
- Could weight loss explain it?
- Possibly[Auto] Participants had obesity and/or type 2 diabetes and the abstract does not separate direct drug effects from weight loss or glycaemic improvement.
- Study tier
- Study tier 3[Auto] Small or short randomized trial (auto-provisional).
- Assessment
- Version 4 · Checked against the source by AI · Oct 19, 2026
Study facts come from the paper. Population match, weight-loss explanation and study tier are our judgments, made against the reference group of healthy normal-weight adults aged 55–75.
In an open-label study of 31 people with psoriasis, obesity and type 2 diabetes on metformin, psoriasis severity fell over 12 weeks with semaglutide (median PASI 21 to 10) and also, less, in controls who stayed on metformin alone (mean PASI 20.6 to 15.9). BMI fell in both groups; IL-6 fell only with semaglutide. Small, unblinded and short.
01Findings
What the study reported
- Drugs
- Semaglutide
- Comparator
- continued metformin without semaglutide
- Dose
- 1.0 mg per week
- Treatment duration
- 12 weeks
- Primary outcome
- Not extracted
- Effect
- Not extracted
- 95% confidence interval
- Not extracted
- Follow-up
- 12 weeks
- Adverse events
- Not extracted
- Limitations
- Not extracted
Who was studied
- Obesity status
- obesity/overweight present (all or most)
- Diabetes status
- type 2 diabetes present (all or most)
- Sample size
- 31
Study quality details
- Study design
- Prospective cohort
- Sample size
- 31
- Randomization
- yes
- Blinding
- open-label
- Comparator
- continued metformin without semaglutide
- Follow up duration
- 12 weeks
- Outcome type
- biomarker
- Replication
- not assessed (auto)
- Consistency with other evidence
- not assessed (auto)
- Population applicability
- INDIRECT
- Statistical precision
- not extracted
- Peer review status
- yes
01bRisk of bias
How much the result can be relied on
Risk of bias is judged per result, not per paper: the same study can report one marker at low risk and another at high. Two reviewers assess each result independently against a written guide, and every domain judgment below carries its reasoning and the sentences it rests on.
Risk of bias is not the same as whether the numbers work. A trial can be well conducted and still print a standard deviation that is arithmetically impossible, or a table whose sign contradicts its own text. Risk of bias has no question for that, so where it was found it is recorded separately below, under Source integrity.
This study was not randomized. Who received the drug was decided some other way than by chance, so its results are judged with ROBINS-I, the tool for non-randomized studies, and not with RoB 2. A non-randomized result is recorded as such in any body of evidence and is never pooled with randomized trials without a written reason.
These words are not quality scores. On the ROBINS-I scale the levels run Low, Moderate, Serious, Critical: Critical risk of bias is the worst rating a non-randomized result can receive, and even Low means comparable to a well-conducted randomized trial, not better. The two scales are not interchangeable. Levels are always written out in full here for that reason.
C-reactive protein (CRP), serum · 12 weeks
Serious risk of bias · ROBINS-I · Two reviewers differed; settled by a recorded ruling
How the overall was reached: The worst domain judgment is Serious (D1, D2, D4, D7), and no domain is Critical, so the overall is Serious by the guide's rule, with no override. Allocation followed a clinician's recommendation and nothing is adjusted for it (D1); post-start exclusions are plausibly response-related and only completers are analysed (D2, D4); the planned between-group comparison is not the one reported (D7).
| Domain | Judgment and reasoning |
|---|---|
| D1 · Confounding Serious risk of bias | Allocation was by clinician choice (“upon the recommendation of an endocrinologist”), so allocation by indication is live, and none of the named confounders is adjusted for. Baseline CRP differs between the arms (3.8 against 4.8, or 9.6 in Table 3) and the only reported contrast is within-arm change, so regression to the mean and time trends are not separated from the drug. The authors attribute everything to the drug: “all the differences in parameters monitored were ascribed to semaglutide only.” Per the guide a study without randomization and without adjustment is at least Serious. Critical was considered (unspecified allocation criteria, internally inconsistent baseline tables); the case falls between the Serious and Critical anchors, and I take the lower, Serious, because baselines on age, sex, BMI and PASI are similar and the randomization claim cannot be excluded.upon the recommendation of an endocrinologist all the differences in parameters monitored were ascribed to semaglutide only |
| D2 · Selection of participants Serious risk of bias | Three of 31 allocated participants (9.7%) were dropped after allocation, below the 10% line, but the reasons are tied to the intervention and to disease course: “two of them due to drug side effects (nausea, vomiting), and one patient due to exacerbation of the disease.” The protocol of the paper also lists adverse reactions to semaglutide as a ground for exclusion. Exclusion for an adverse event or for disease exacerbation is plausibly related to intervention response, which the v1.1 anchor places at Serious. Critical was considered (analysis of those who stayed in); with three exclusions the set could not by itself produce a CRP effect, so, falling between Serious and Critical, I take the lower, Serious. Table 1 denominators (2/13 = 15.4% active smokers) confirm the semaglutide arm was analysed at n = 13.three patients were excluded from the study; two of them due to drug side effects (nausea, vomiting), and one patient due to exacerbation of the disease adverse reactions to semaglutide described in the available Patient information leaflet (PIL): nausea, vomiting, allergic reaction at the localization of drug administration, were excluded from the study |
| D3 · Classification of interventions Low risk of bias | Intervention status (semaglutide dispensed free of charge, controls on continued metformin) was defined at the start and verified by monthly phone contact in both groups; it could not have been affected by the outcome, which had not occurred. How the endocrinologist chose who got the drug is a confounding matter (D1), not a classification one.The patients treated with semaglutide received the drug free of charge the patients were contacted by phone, every month |
| D4 · Deviations from intended interventions Serious risk of bias | Only an analysis of participants who remained in the study is reported, with no assignment-based analysis; the v1.1 anchor places that at Serious. Co-interventions are partly described and similar on paper (same diet advice, topical salicylic acid for both, lipid-lowering therapy deferred), but the metformin dose is not given, the registry (not the paper) lets controls take other oral antidiabetics, and HbA1c and BMI fell in the control arm too, so care that moves metabolic and inflammatory markers was not held equal or documented. Between Moderate and Serious the explicit completers-only anchor fixes Serious.Both groups were allowed to take the topical keratolytic therapy of salicylic acid. the introduction of lipid lowering therapy was postponed due to possible impact of statins on interleukin levels |
| D6 · Measurement of outcomes Low risk of bias | CRP was measured by the hospital laboratory (registry: turbidimetric test) on fasting samples at the same two timepoints in both arms; nothing says the assay changed or differed between arms. It is not described as high-sensitivity. Open-label status is a limitation the paper itself flags, but for an objective laboratory marker that does not move this to Moderate. No statement says laboratory staff were blind.These analyses were done at the Institute for Laboratory Diagnostics of the UCC. Biochemical analyses-Turbid metric test |
| D5 · Missing data Moderate risk of bias | Outcome data are missing for 3 of 31 (9.7%), under the 20% threshold, but the arm difference is 2 of 15 (13.3%) against 1 of 16 (6.3%), 7.1 points, over the 5-point threshold, so data were not available for nearly all. The reasons differ by arm and no analysis addresses the missing patients. The per- marker N for Table 3 is not stated. With absolute numbers of 2 and 1, this sits between Moderate and Serious; the outcome-related character of the exclusions is already carried at Serious in D2, and the tie-break takes the lower severity. [D5 set to Moderate by a recorded owner ruling: IN-055, owner-decided 2026-10-05 (robins-i-guide.md v1.2: an exclusion judged in D2 is not judged again in D5). The other reviewer read Serious on the threshold alone.]two of them due to drug side effects (nausea, vomiting), and one patient due to exacerbation of the disease. |
| D7 · Selection of the reported result Serious risk of bias | Result assessed: CRP at 12 weeks (F1) against baseline (F0), Table 3, semaglutide arm p = 0.01 and control arm p = 0.5, with Table 4 comparing baseline CRP between arms. Sources checked: paper, registry record, new protocol, new SAP. (1) Can the protocol and SAP count as a retrieved source? They are retrieved and read, so they are not the unretrievable-protocol floor case. On content they do show the outcome and timepoint: the registry lists CRP as its own outcome with a timeframe up to 12 weeks, and the protocol lists CRP among serum markers measured at 0 and 3 months. (2) What weight can they carry? Less than a prespecification record. The registry entry was first submitted 2024-06-11, after the 2024-06-10 completion; the protocol and SAP are labelled May 2023 but were uploaded 2024-06-20 and their files were created on that date; the protocol also uses past tense for work already done. None can be shown to have existed before the data, so they show what the authors later said was planned, not that the choice was made blind to results. Low needs a source that can serve as the prespecified record, so Low is not reached on this ground alone. (3) Does the reported analysis match the plan? No. The SAP states the plan as a comparison between group 1 and group 2 with an independent-samples t-test or Wilcoxon Mann-Whitney test. For CRP the paper reports within-arm pre-post tests (paired t-test or Wilcoxon signed-rank, per Section 2.6) in Table 3, and its only between-arm CRP test is on baseline values in Table 4, so the planned between-group contrast at 12 weeks, or of change, is not reported anywhere. The paper gives no explanation. What is reported is the contrast that favours the result (a significant fall in the semaglutide arm, a non-significant fall in an arm that also fell), and the one that would test the difference is missing. Table 3 and Table 4 also disagree on control baseline CRP (9.6 ± 10.7 against 4.8 (9.7)), which is unexplained. (4) Anchors. None of the three Serious prongs fires literally (no exploratory or post hoc statement, no outcome list to omit from, timepoint matches the registry). The Critical limb (analysis contradicts the registered one with no explanation) is the nearest fit, but the SAP names no outcome and no timepoint, does not rule out paired tests, and the paper's Methods do name the independent-groups test, so this is a planned between-group contrast left unreported rather than a flat contradiction, and the SAP's own dating is too weak to carry Critical. Moderate is excluded because its definition requires that nothing indicates the analysis was unplanned, and here a retrieved SAP shows the reported contrast is not the planned one. Per the v1.1 tie-break between Serious and Critical I take the lower, Serious, because the omitted planned contrast with only the favouring contrast reported is the same selection problem as reporting a marker only at the favouring timepoint or subgroup. A reviewer who reads the SAP as too generic to show a planned contrast, or too late to count, could land at Moderate; I do not think that fits, but it is the nearest alternative. [Judged in a blind NEW_EVIDENCE round after the protocol and SAP were retrieved; both reviewers agreed.]"measure":"Change in inflammation marker level: CRP,","description":"Biochemical analyses-Turbid metric test","timeFrame":"up to 12 weeks" The second part of the serum was used to determine the concentration of serum TNF-α, IL-1β, IL-6, IL-17, CRP and IL23. |
Interleukin-6 (IL-6), serum · 12 weeks
Serious risk of bias · ROBINS-I · Two reviewers differed; settled by a recorded ruling
How the overall was reached: The worst domain judgment is Serious (D1, D2, D4, D7), and no domain is Critical, so the overall is Serious by the guide's rule, with no override. Allocation followed a clinician's recommendation and nothing is adjusted for it (D1); post-start exclusions are plausibly response-related and only completers are analysed (D2, D4); the planned between-group comparison is not the one reported (D7).
| Domain | Judgment and reasoning |
|---|---|
| D1 · Confounding Serious risk of bias | Same reasoning as O1: clinician-chosen allocation (“upon the recommendation of an endocrinologist”), no adjustment for any named confounder, and a within-arm-only contrast. The baseline IL-6 values differ (3.5 in the semaglutide arm, 5.6 in the control arm with a wide spread) and are never compared; the control arm's IL-6 fell further than the treated arm's (to 2.3 against 2.8), and the paper's "significant" IL-6 result is a within-arm p = 0.05 only. Between Serious and Critical I take the lower, Serious, for the same reasons as in O1.upon the recommendation of an endocrinologist IL-6 (pg/mL) 3.5 (2.3) 2.8 (1.1) 0.05 * 5.6 (12.2) 2.3 (3.6) 0.1 |
| D2 · Selection of participants Serious risk of bias | As for O1: 3 of 31 (9.7%) dropped after allocation for side effects and disease exacerbation, which the v1.1 anchor treats as plausibly related to intervention response, so Serious; Critical was considered and the lower anchor taken because three exclusions could not by themselves produce the IL-6 change.three patients were excluded from the study; two of them due to drug side effects (nausea, vomiting), and one patient due to exacerbation of the disease adverse reactions to semaglutide described in the available Patient information leaflet (PIL): nausea, vomiting, allergic reaction at the localization of drug administration, were excluded from the study |
| D3 · Classification of interventions Low risk of bias | Same as O1: intervention status was recorded at the start, supplied free of charge, and checked by monthly phone contact; it could not be affected by the IL-6 result.The patients treated with semaglutide received the drug free of charge the patients were contacted by phone, every month |
| D4 · Deviations from intended interventions Serious risk of bias | Same as O1: analysis of participants who stayed in only, with no assignment-based analysis, which the v1.1 anchor places at Serious; co-interventions on diet, topical therapy and lipid-lowering are described as equal, but metformin dose and other antidiabetics are not documented for both arms, and the paper itself notes that metformin is an anti-inflammatory that could affect cytokines.Both groups were allowed to take the topical keratolytic therapy of salicylic acid. many experimental and clinical studies confirming the anti-inflammatory properties of metformin that could influence the level of cytokines |
| D6 · Measurement of outcomes Low risk of bias | IL-6 was measured by one commercial ELISA kit (BioLegend ELISA MAX Deluxe Sets) in one research laboratory, on fasting samples at the same two timepoints in both arms, and the registry names the same kit; nothing says the assay changed or differed between arms. Objective marker; open-label status does not move this off Low. No statement says laboratory staff were blind. Values sit near the assay's low range (medians 2.3 to 5.6 pg/mL), which adds noise but not directional bias.The serum levels of IL-1β, IL-6, IL-17 and IL-23 were measured by means of the Enzyme Linked Immunosorbent Assay (ELISA) technique Bio Legend ELISA MAX Deluxe Sets |
| D5 · Missing data Moderate risk of bias | Same missingness as O1 (9.7% overall; 13.3% against 6.3% by arm, a 7.1-point difference over the 5-point threshold), reasons differing by arm, no analysis addressing it, and no per-marker N for the ELISA results. Between Moderate and Serious on 2 and 1 patients; the outcome-related character of the loss is carried in D2, and the tie-break takes the lower severity. [D5 set to Moderate by a recorded owner ruling: IN-055, owner-decided 2026-10-05 (robins-i-guide.md v1.2: an exclusion judged in D2 is not judged again in D5). The other reviewer read Serious on the threshold alone.]two of them due to drug side effects (nausea, vomiting), and one patient due to exacerbation of the disease. |
| D7 · Selection of the reported result Serious risk of bias | Result assessed: IL-6 at 12 weeks (F1) against baseline (F0), Table 3, semaglutide arm p = 0.05 and control arm p = 0.1. Sources checked: paper, registry record, new protocol, new SAP. (1) Retrieved source: the registry lists IL-6 inside a bundled outcome (TNFa, IL-1b, IL-6, IL-17 and IL-23) with a timeframe typed as up to12 weeks, and the protocol lists IL-6 among serum markers measured at 0 and 3 months, so outcome and timepoint are shown in the same way as for O1, though IL-6 is not a registered outcome of its own. (2) Weight: the same limits as O1 apply. The registry entry was submitted after completion, the protocol and SAP were uploaded 2024-06-20, and nothing shows any of them existed before the data, so they show a later account of the plan. (3) Analysis match: the SAP plans a between-group comparison with an independent-samples test; the paper reports only within-arm pre-post tests for IL-6 (Table 3) and, unlike CRP, does not report any between-arm IL-6 test, not even at baseline. The control arm's IL-6 also fell from 5.6 to 2.3, so the planned contrast is the one that would show whether the difference between arms holds, and it is absent with no explanation. The semaglutide-arm result sits at p = 0.05, the margin of the paper's stated threshold, so what is left unreported matters more here. (4) Anchors: as for O1. No Serious prong fires literally and the Critical limb is the nearest fit but not a flat contradiction, since the SAP names no outcome or timepoint and the paper's Methods name the independent-groups test. Moderate is excluded because a retrieved SAP shows the reported contrast is not the planned one. Tie-break between Serious and Critical takes the lower, Serious. Same note as O1 that a reviewer reading the SAP as too generic or too late to count could land at Moderate. [Judged in a blind NEW_EVIDENCE round after the protocol and SAP were retrieved; both reviewers agreed.]"measure":"Serum values of TNFa, IL-1b, IL-6, IL-17 and IL-23" "timeFrame":"up to12 weeks" |
02Funding
Funding and conflicts
- Funding
- No external funding (per full text).
- Industry funded
- Unclear
- Manufacturer
- None identified
- Sponsor role
- not reported in abstract
- Author conflicts
- The authors declare no conflicts of interest.
- Independent replication
- unknown
Funding is shown on every study and never used to score it.
04Source
The source, as retrieved
Abstract
Psoriasis is a chronic inflammatory skin disease with relapsing nature. Estimates are that approximately 2-3% of the world's population suffers from this disease. More severe forms of psoriasis are conditions of high inflammation, which is confirmed by the clinical picture and numerous inflammatory parameters such as C-reactive protein (CRP), cytokines and homocysteine, which vary with disease activity. The objective of this clinical study was to investigate the effect of GLP-1 receptor agonist semaglutide therapy on pro-inflammatory factors in the serum and the severity of the clinical picture of psoriasis in obese patients with type 2 diabetes mellitus (T2DM) on chronic metformin therapy. This randomized clinical study was conducted on 31 psoriatic patients with T2DM that were randomized into two groups: one that received semaglutide during the 12-week trial (n = 15), while the second was control (n = 16). The results demonstrated that the severity of the clinical picture of psoriasis, determined by the Psoriasis Area and Severity Index (PASI) score, was significantly better after the administration of semaglutide (the median baseline PASI score in patients treated with semaglutide was 21 (IQR = 19.8), while after 12 weeks of therapy the score was 10 (IQR = 6; p = 0.002). Also, the quality of life in the group of patients who received the drug, measured by the Dermatology Life Quality Index (DLQI), improved significantly after 3 months (a median baseline DLQI score in the semaglutide group was 14 (IQR = 5) at the beginning of the study, and after 12 weeks of treatment the median DLQI score was 4 (IQR = 4; p = 0.002)). The use of semaglutide led to a significant decrease in pro-inflammatory cytokines in the serum (IL6), as well as a significant decrease in CRP values (p < 0.05). A significant decrease in the body mass index (BMI) value in the semaglutide-treated group was also identified, as well as a significant decrease in the level of low-density cholesterol (LDL) (p < 0.05). In conclusion, semaglutide, based on its systemic anti-inflammatory characteristics, could contribute to the treatment of psoriatic obese patients with T2DM.