Semaglutide

Limits on what this shows

• Key negative or null findings: The in vitro respiration reduction was transient and self-resolving within the same experiment (normalized by day 5), and tirzepatide, a related but structurally distinct incretin dual agonist, did not show the same dip, arguing against a class-wide direct mitochondrial-toxicity mechanism. Multiple rodent cardiomyocyte studies (ischemia-reperfusion and diabetic cardiomyopathy models) report semaglutide enhancing PINK1/Parkin-mediated mitophagy and mitochondrial quality control rather than impairing it, consistent with a protective rather than depleting direct effect on stressed tissue.
• At-risk subgroups: None established; the only human-derived data is a short-duration cell-culture experiment, not a patient population, so no subgroup can be identified from it.

Regulatory status

Not listed on the FDA or DailyMed label for any semaglutide formulation as an adverse reaction or monitoring parameter; no mitochondrial-toxicity warning of the kind seen with some antiretrovirals or statins.

Population evidence

No semaglutide-specific human clinical study measuring muscle mitochondrial density, respiration, or biogenesis (for example, muscle biopsy or 31P-MRS) in living patients was located. The closest human-derived evidence is in vitro: Old et al. (Journal of Cachexia, Sarcopenia and Muscle, 2026) exposed cultured human skeletal muscle cells to semaglutide, tirzepatide, and cagrilintide under healthy and lipotoxic conditions and found that semaglutide, like cagrilintide, produced a transient reduction in mitochondrial respiration at 48 hours that had resolved by day 5, in contrast to tirzepatide's sustained improvement. A 2026 systematic review and meta-analysis in Metabolism Open (Greenblatt et al.) specifically set out to test direct effects of GLP-1 receptor agonists on mitochondrial function in human-derived in vitro models; its full findings could not be independently verified beyond title and indexing metadata, so it is recorded here as a located, relevant, but not fully confirmed source (see Sources).

Mechanism

Two distinct pathways, the same separation applied to thiamine and B12 above. Direct pharmacology: GLP-1 receptor activation could plausibly touch mitochondrial signaling directly, which is exactly what the in vitro cell-culture studies test, isolated from any caloric or whole-body effect since cultured cells cannot themselves eat less. Caloric-restriction/weight-loss-mediated: in diet-induced obese mice, semaglutide-driven weight loss (reduced fat and lean mass) was associated with improved skeletal-muscle oxidative phosphorylation efficiency and altered complex III assembly proteins (Choi et al.), an adaptation more plausibly explained by the metabolic remodeling that accompanies substantial weight loss than by a direct drug effect on the electron transport chain.

Counter-evidence held on record

Every semaglutide-specific mitochondrial signal located, direct or weight-loss-mediated, either resolved within days (the in vitro respiration dip) or moved in the protective direction (improved oxidative phosphorylation efficiency with weight loss; enhanced mitophagy in stressed cardiomyocytes). No sustained mitochondrial impairment has been reported in any source found. Framing caution This is not evidence that semaglutide harms mitochondria. The consumer-facing framing should say plainly that the available direct-exposure data shows a small, temporary effect in cultured cells, not a demonstrated clinical mitochondrial injury, and should not be merged with unrelated claims about GLP-1-associated muscle loss, which is a body-composition finding, not a mitochondrial-function finding (see the ATP production entry below and the SEMALEAN study in Sources).

Limits on what this shows

• Key negative or null findings: No study, human or animal, measuring oxidative stress markers with semaglutide in the cohort was found showing an increase. This dossier specifically searched for evidence that rapid lipolysis or weight loss might transiently mobilize stored oxidative burden from adipose tissue, a theoretical concern raised elsewhere for rapid weight loss generally; no semaglutide-specific study measuring this was located, so it is not asserted here.
• At-risk subgroups: Not established from semaglutide-specific data.

Regulatory status

Not listed on the label as an adverse effect or monitoring parameter.

Population evidence

Omachi et al. (Diabetes Therapy, 2024), a prospective study of 37 adults with type 2 diabetes switched from dulaglutide to semaglutide or continued on dulaglutide over 24 weeks, measured oxidative stress alongside continuous glucose monitoring and reported that switching to semaglutide was superior to continued dulaglutide for reducing oxidative stress and glucose variability. This is Tier A: semaglutide is directly in the studied cohort and the outcome was directly measured, not inferred. Several rodent and cell-culture studies (diabetic neuropathy, MASLD, high-glucose cardiomyocyte injury models) separately report semaglutide reducing oxidative-stress and neuroinflammatory markers, including via Nrf2/Keap1 pathway activation, consistent in direction with the human finding.

Mechanism

No mechanism for semaglutide increasing oxidative stress was located in any tier of evidence. The reduction seen is generally attributed in this literature to improved glycemic stability (less glucose-variability-driven oxidative stress) and weight loss (reduced adipose-tissue-driven inflammatory and oxidative signaling), both indirect, metabolism-mediated pathways rather than a direct antioxidant action of the drug on a specific enzyme system.

Counter-evidence held on record

The single Tier A semaglutide-specific human study found a reduction in oxidative stress markers relative to an active comparator, not an increase. This is recorded here as a context flag, per the grading key's provision for interactions running the other direction, not as a burden finding. Framing caution Do not present this as "semaglutide protects against oxidative stress" as an established clinical claim either; one 37-patient active-comparator study without a placebo arm is a single data point, not a settled finding, and an active-comparator design cannot rule out that both drugs perform similarly against no treatment at all. What can be said is that no evidence of an oxidative-stress-increasing effect was found in any tier searched.

Limits on what this shows

• Key negative or null findings: None of the pooled or intake studies located include a non-GLP-1 comparator group matched for obesity or type 2 diabetes, so confounding by indication (both conditions are independently associated with lower iron stores, especially in premenopausal women and in inflammation-driven states) cannot be excluded. STEP 1's formal safety tables record no anemia or iron-deficiency adverse events.
• At-risk subgroups: Patients with the largest weight loss and longest treatment duration, premenopausal women, and anyone with a baseline diet already low in iron-rich foods.

Regulatory status

Not listed on the FDA or DailyMed label for any formulation.

Population evidence

The pooled systematic review already cited above for vitamin D and B-vitamin findings (Clinical Obesity, 2026, roughly 480,825 adults with diabetes and/or obesity on GLP-1 receptor agonists) reported iron deficiency in 3.2% of the largest contributing study and nutritional anemia in 4%; a smaller 69-person study within the same pooled analysis found 64% with insufficient iron intake. A separate 2026 narrative review (Urbina et al., Clinical Obesity) covering overlapping literature from the same window (January 2019 to May 2025, also approximately 480,825 adults) reports GLP-1 receptor agonist users showing ferritin levels 26 to 30% lower than comparators, with over 60% consuming insufficient dietary iron. The cross-sectional food-record study already cited above (Frontiers in Nutrition, 2025, semaglutide 53.6% of n=69) separately found iron intake below the dietary reference intake in that sample.

Mechanism

Two pathways, not to be merged. Direct: delayed gastric emptying could plausibly reduce the gastric-acid-dependent reduction of dietary non-heme iron from ferric to the more absorbable ferrous form, by loose analogy to proton-pump inhibitors, but this has not been directly tested for semaglutide. Indirect, and better supported by the data available: eating substantially less food, including less iron-rich meat and fortified food, mechanically reduces iron intake in proportion to reduced caloric intake, the same behavioral pathway already established for the general-adequacy entry above.

Counter-evidence held on record

Every iron-status study located is either a pooled, non-semaglutide-stratified GLP-1-class analysis or a small cross-sectional intake study without a pre-treatment baseline or non-drug comparator, the identical evidentiary limitation that kept the general-adequacy entry above at yellow rather than orange. The two pooled reviews cited here substantially overlap in underlying source studies and time window (see Overlap map below) and should be read as one body of evidence, not two independent confirmations. Framing caution This does not establish that semaglutide depletes iron directly. Obesity and type 2 diabetes are independently associated with altered iron status before any GLP-1 receptor agonist is started, and the behavioral, reduced-intake pathway already documented for other nutrients in this dossier is the best-supported explanation for what data exists.

Limits on what this shows

• Why no grade: UNGRADED requires both no Tier B or C evidence and no circulating consumer claim to rebut; both conditions are met here. This is printed as a gap in the evidence, not as reassurance that copper status is unaffected.

Regulatory status

Not addressed on the label.

Population evidence

None located at Tier A, B, or C. Searches specifically combining semaglutide or GLP-1 receptor agonists with copper or trace-element status returned no relevant results in PubMed, Crossref, or Semantic Scholar. The nutrient-adequacy studies already cited above (Frontiers in Nutrition, 2025; the pooled Clinical Obesity reviews) do not include copper among the nutrients they measured.

Mechanism

No mechanism, direct or caloric-restriction-mediated, has been proposed for semaglutide and copper specifically in the literature searched. Copper deficiency is a well-documented complication of malabsorptive bariatric surgery, particularly procedures that bypass the duodenum, the primary site of copper absorption, which is why this category was included in this expansion by analogy. Semaglutide has no anatomic malabsorptive component, so the bariatric-surgery mechanism does not directly transfer, and no study has tested whether the caloric-restriction-only pathway, without anatomic bypass, is sufficient to affect copper status on its own.

Counter-evidence held on record

None to hold; there is no finding to counter because no finding exists. Framing caution Absence of evidence is not evidence of absence. A consumer, clinician, or future researcher should not read this entry as confirmation that semaglutide is safe with respect to copper status; it means only that the question has not yet been studied and that no wellness claim needed rebutting at the time of this review.

Limits on what this shows

• Key negative or null findings: Obesity and type 2 diabetes are themselves independently associated with altered gut microbiota, reduced diversity, and an altered ratio of major bacterial phyla, before any GLP-1 receptor agonist is started; none of the studies located included a comparator arm that would let this baseline disease-driven pattern be separated from a semaglutide-specific effect, so confounding by indication is presumed, not excluded, per hard rule 2.
• At-risk subgroups: Not established from semaglutide-specific human data; a theoretical concern would track with the same reduced-intake and rapid-weight-loss subgroups already identified above.

Regulatory status

Not addressed on the label.

Population evidence

Gofron et al. (Nutrients, 2025), a systematic review of 38 studies of GLP-1 medications and gut microbiota, reports that semaglutide specifically increased metabolically beneficial bacterial genera but reduced overall microbial diversity, while noting that the underlying studies vary widely in population, duration, and design and calling for further research to confirm the pattern. A separate 2026 human, semaglutide-specific study (Wang et al., Diabetes Obesity and Metabolism) examined weight regain after semaglutide discontinuation in women with obesity and found associations with gut microbiota and bile acid metabolism; this describes a post-discontinuation weight-regain phenomenon rather than an on-drug deficiency and is noted here for completeness rather than as deficiency evidence, and its full findings could not be independently verified beyond title and indexing metadata. A 2026 mechanistic review (Kamath et al., British Journal of Clinical Pharmacology) frames the relationship as bidirectional: gut microbiota help regulate the body's own GLP-1 secretion via short-chain fatty acids and bile-acid derivatives, while GLP-1 receptor agonist drugs reciprocally reshape microbial composition, meaning cause and effect run in both directions and are not easily separated in observational data.

Mechanism

Two pathways, not to be merged, per the semaglutide-specific reminder for this expansion. Direct: delayed gastric emptying changes the timing and gastrointestinal-pH environment food is exposed to before reaching the colon, a genuinely pharmacological effect on the gut lumen distinct from how much is eaten. Indirect: eating substantially less food reduces the fermentable substrate (fiber, resistant starch) reaching the colon, which independently reduces microbial diversity regardless of drug mechanism, the same pattern seen with any large, sustained caloric restriction. The located evidence does not disentangle which of these two pathways drives the diversity reduction reported.

Counter-evidence held on record

The systematic review's own authors describe the evidence as preliminary and call for confirmation; a diversity reduction alongside an increase in some beneficial taxa is not a coherent "harm" signal, and no study excludes obesity or type 2 diabetes itself as the source of the baseline microbiota pattern being described. Framing caution This is not evidence that semaglutide should be avoided for gut-health reasons, nor that a probiotic or specific supplement is indicated; no dose or product recommendation follows from a diversity-metric change of unclear clinical significance.

Limits on what this shows

• Key negative or null findings: STEP 1's formal safety tables record no anemia adverse events. The STEP-HFpEF result runs opposite to any impairment narrative, showing improved rather than reduced functional capacity, most plausibly explained by reduced body weight lowering cardiac and ventilatory workload rather than by any nutrient-mediated oxygen-carrying-capacity change.
• At-risk subgroups: Not established from semaglutide-specific data; would theoretically track with the iron-deficiency-anemia subgroup already identified above if that pathway is real.

Regulatory status

Not addressed on the label; no hematologic monitoring requirement.

Population evidence

A targeted search for semaglutide paired with hemoglobin, hematocrit, tissue oxygenation, or VO2 returned no matching studies in the databases searched. The pooled GLP-1-class cohort already cited for iron above reports nutritional anemia in 4% of roughly 480,825 users, a downstream outcome of iron and other nutrient status rather than a distinct oxygen-delivery mechanism (see Overlap map; this figure is counted in the Iron entry, not recounted here as new evidence). Separately, the STEP-HFpEF trial (Kosiborod et al., New England Journal of Medicine, 2023), a semaglutide-specific randomized placebo-controlled trial in patients with obesity-related heart failure with preserved ejection fraction, found that semaglutide improved heart-failure symptoms, physical limitations, and functional capacity relative to placebo over 52 weeks.

Mechanism

No direct pharmacological mechanism by which semaglutide would impair oxygen delivery or utilization was identified in any tier of evidence. The only plausible indirect pathway is the iron-deficiency-anemia route already covered in the Iron entry above; oxygen delivery would only be secondarily affected if iron status were affected first, and that pathway is itself only class-level, non-stratified evidence.

Counter-evidence held on record

The one Tier A semaglutide-specific trial to measure a functional outcome dependent on oxygen delivery and utilization (exercise and symptom capacity in heart failure) found improvement. This entry exists to document an evidence gap and a reverse-direction data point, not a burden. Framing caution This should not be read as a general claim that semaglutide improves oxygen delivery either; the STEP-HFpEF result is specific to a heart-failure population and is most plausibly explained by weight loss reducing cardiac workload, not by a direct effect on oxygen-transport physiology in the general semaglutide-using population.

Limits on what this shows

• Key negative or null findings: No sustained reduction in mitochondrial respiration or ATP-generating capacity was found in any study located; the one direct-exposure signal (cultured cells) self-resolved within the same experiment.
• At-risk subgroups: Not established from human data; patients with the largest lean-mass loss during rapid weight loss would be the theoretical subgroup of interest, an outcome already flagged in general obesity-medicine guidance (see the joint advisory already cited above) rather than a new finding here.

Regulatory status

Not addressed on the label.

Population evidence

In diet-induced obese mice, semaglutide-driven weight loss was associated with improved skeletal-muscle mitochondrial oxidative phosphorylation efficiency (Choi et al., bioRxiv preprint 2024; the same group's work is also indexed to Obesity (Silver Spring), 2025, per PubMed, though that journal-of-record version's own abstract was not independently re-verified for this entry). Separately, in cultured human skeletal muscle cells, direct semaglutide exposure produced a transient reduction in mitochondrial respiration at 48 hours that resolved by day 5 (Old et al., Journal of Cachexia, Sarcopenia and Muscle, 2026, already cited in the Mitochondrial burden entry above; the same source and finding, not recounted as independent evidence, see Overlap map). Several rodent cardiac studies report semaglutide improving cardiac energy-substrate utilization and mitochondrial quality control in disease models, for example Ma et al. (Nature Communications, 2024), reporting improved cardiac remodeling via a Creb5/NR4a1 energy-substrate axis, and Tian et al. (Acta Pharmacologica Sinica, 2025), reporting energetic improvement in diabetic cardiomyocytes; both were confirmed only at the title/author/journal/PMID level via search-listing metadata, not by independently opened full abstracts. A semaglutide-specific human study of body composition (the SEMALEAN study, Alissou et al., Diabetes Obesity and Metabolism, 2025 to 2026) examined fat mass, lean mass, and muscle function during semaglutide treatment; its full findings could not be independently retrieved beyond the study's existence and design, so no specific effect size is asserted here, but lean-mass loss during large-magnitude weight loss is a well-established, caloric-restriction-mediated phenomenon relevant to total-body ATP-generating capacity even where individual-cell efficiency is unaffected or improved.

Mechanism

The same two pathways as the Mitochondrial burden entry above, deliberately not merged: direct drug exposure, tested only in short-duration human cell culture, showing a small and temporary effect; and caloric-restriction and weight-loss-mediated remodeling, tested in mice, showing improved per-mitochondrion efficiency alongside a separate, non-mitochondrial consideration, reduced total muscle mass from lean-mass loss during weight loss, which would reduce whole-body ATP-generating tissue even if each remaining mitochondrion works efficiently.

Counter-evidence held on record

This entry and the Mitochondrial burden entry above draw on an overlapping evidence base (the same in vitro study, the same mouse weight-loss study); they are graded and presented separately, per this expansion's brief, but represent one body of evidence about the same organelle system, not two independent confirmations of a deficit. See the Overlap map below. Framing caution Lean-mass loss during weight loss is a real, well-documented phenomenon that deserves clinical attention, but it is a body-composition outcome, not a demonstrated defect in ATP production at the cellular level, and must not be conflated with the mitochondrial-efficiency data above, which if anything trends toward improvement, not impairment.