Nabumetone

Limits on what this shows

• Null findings: No uncoupling in isolated mitochondria [15][16]; no succinate-driven inhibition [15]. The 2026 study screened cells at concentrations not read here, so relevance to therapeutic exposure (6MNA more than 99% protein bound [1]) is unestablished.

Regulatory status

No mitochondrial statement in the label text retrieved [1].

Population evidence

Moreno-Sanchez 1999 [15] (Tier B, in vitro, abstract read via PubMed): diclofenac, piroxicam, indomethacin, naproxen, nabumetone, nimesulide and meloxicam were compared. Nabumetone and naproxen did not behave as uncouplers; nabumetone inhibited respiration, ATP synthesis and membrane potential with glutamate plus malate but not with succinate, inhibited NADH oxidation in submitochondrial particles and inhibited O2 uptake in intact cells and in whole heart, whereas the other five drugs stimulated respiration; the authors concluded nabumetone exerted a specific inhibition of site 1 (complex I). Concentrations and species were not in the abstract text read. Mahmud 1996 [16] (Tier B, isolated rat liver mitochondria, reconstructed abstract): acidic NSAIDs stimulated respiration at micromolar concentrations (uncoupling potency inversely correlated with pKa, r = -0.87), while nabumetone and selective COX-2 inhibitors did not. Huang 2026 [17] (Tier B, adult human primary cardiomyocytes, 18 agents screened, reconstructed abstract): nabumetone was described as a representative NSAID causing severe mitochondrial damage, with abnormal fission, suppressed mitophagy and impaired electrophysiological and metabolic function, via the PGE2-EP4 pathway; SIRT3 expression fell and the activator honokiol protected cells. A rat intestinal study (Gut 1997) [28] reportedly found nabumetone did not cause the mitochondrial changes seen with indomethacin (PubMed listing summary only; abstract not opened).

Mechanism

Named: complex I (site 1) inhibition [15]; PGE2-EP4 signaling with SIRT3 loss and mitochondrial fission [17]. Absent: protonophoric uncoupling, which requires an ionizable acidic group [16].

Hard Rule 2 and 3 checks

This is the drug's own toxicity in vitro, not a nutrient deficiency. No functional marker in nabumetone users was found, so the entry cannot be graded above yellow.

Counter-evidence held on record

Counter-evidence and confounding The two findings are not mutually consistent in direction of mechanism (no uncoupling versus complex I inhibition versus fission and mitophagy defects) and use different preparations. No human tissue, biopsy or lactate study with nabumetone exists in the retrieved literature. The 2026 abstract was read only as a reconstructed OpenAlex abstract. Framing caution Cell and isolated-organelle findings do not show mitochondrial harm in people taking nabumetone. No coenzyme Q10, carnitine or antioxidant supplement is implied.

Limits on what this shows

• Null findings: Antioxidant activity was not seen after in vivo rat dosing [18]; microsomal lipid peroxidation was not increased by the metabolite [18]. A PubMed listing for nabumetone with oxidative, antioxidant, glutathione, lipid or superoxide terms returned 129 records; the 10 shown were reviews, formulation and environmental degradation papers and a pseudoporphyria review, with no human redox marker study.

Regulatory status

No oxidative stress statement in the label text retrieved [1].

Population evidence

Barth 1994 [18] (Tier B, rat liver microsomes and rats; abstract read via Europe PMC): nabumetone and its main metabolite BRL 10,720 inhibited iron-stimulated lipid peroxidation in vitro at 10 micromolar to 1 millimolar, nabumetone the strongest and also in liver slices; the antioxidant effects were not found after oral dosing of rats (5 to 300 mg/kg for three days); nabumetone increased H2O2 formation both in vitro and in vivo, and BRL 10,720 increased H2O2 in vivo and enhanced chemiluminescence, interpreted as more superoxide and other reactive oxygen species, though lipid peroxidation in liver microsomes was not enhanced. Canudas 2000 [19] (Tier B, in vitro, reconstructed abstract): light-activated nabumetone generates singlet oxygen and radicals and photohemolysis was reduced by glutathione and azide. Qureshi 2021 [20] (Tier B, HaCaT keratinocytes): title states UV-induced photogenotoxicity mediated by ROS generation; abstract could not be obtained and a PubMed listing summary describes mitochondrial membrane disruption and DNA damage. Huang 2026 [17]: mitochondrial dysfunction (no ROS statement in the abstract read).

Mechanism

Named in vitro: radical scavenging chemistry of the naphthyl ring (antioxidant direction) and cytochrome P450-dependent H2O2 production (pro-oxidant direction) [18]; light-dependent singlet oxygen generation [19].

Hard Rule 2 check

This is a redox signal of the drug's own chemistry and toxicity. It is not an antioxidant nutrient depletion.

Counter-evidence held on record

Counter-evidence and confounding Rat and in vitro data only; UV-dependent results require light exposure; the in vitro concentrations (10 micromolar to 1 millimolar) were not compared with free plasma 6MNA in the abstract read, and 6MNA is more than 99% protein bound [1]. Framing caution No antioxidant supplement is implied. The entry is a mechanistic and context flag. Photosensitivity concerns belong with the prescriber or pharmacist.

Limits on what this shows

• Null findings: Fecal blood loss equal to placebo at 2,000 mg for 21 days [5]. Fewer endoscopic lesions than naproxen [1]. Both are short-exposure comparisons in healthy volunteers or short trials, not older arthritis patients on years of therapy, and the comparator-based trials [7][8][10] do not test nabumetone against non-use. The 1987 Am J Med companion 51Cr report [6] could not be read beyond a PubMed listing summary. A 2004 Drugs review of nabumetone safety [30] was not opened (no abstract available) and is not relied on.
• At-risk subgroups: Older adults (half-life 29.8 hours), people on warfarin (label: synergistic GI bleeding) or aspirin (label: not generally recommended), people with prior GI bleeding, and longer or higher-dose exposure [1].

Regulatory status

Label warns that NSAIDs cause serious GI bleeding, ulceration and perforation, and states: "Anemia is sometimes seen in patients receiving NSAIDS, including nabumetone tablets. This may be due to fluid retention, occult or gross GI blood loss, or an incompletely described effect upon erythropoiesis." It advises a hemoglobin or hematocrit check if signs or symptoms of anemia appear. In controlled trials of 1,677 nabumetone patients the cumulative incidence of peptic ulcers was 0.3% (95% CI 0%, 0.6%) at 3 to 6 months, 0.5% (0.1%, 0.9%) at 1 year and 0.8% (0.3%, 1.3%) at 2 years. Five endoscopy trials compared a total of 194 patients on nabumetone 1,000 mg daily or naproxen and found fewer endoscopically detected lesions with nabumetone. Iron is not mentioned [1].

Population evidence

Lussier 1989 [5] (Tier A, double-blind parallel, healthy volunteers): 30 subjects randomized to nabumetone 2,000 mg, aspirin 3.6 g or placebo for 21 days after a 7-day placebo run-in, plus 6 untreated controls; 51Cr-labeled red cells; fecal blood loss on nabumetone was not significantly different from placebo, while aspirin was significantly higher than all other groups (p<0.0001). Huang 1999 [7] (Tier A, meta-analysis, abstract-level): 13 studies, 49,501 patients; significantly more comparator-NSAID patients had GI adverse events (P = 0.007); the abstract states perforations, ulcers and bleeds were "10 to 36 times more likely" with comparators and GI hospitalizations 3.7-fold higher; sponsor, heterogeneity and weighting were not read. Scott 2000 [8] (Tier A, randomized double-blind, 6 months, osteoarthritis, nabumetone 1,500 to 2,000 mg/day versus diclofenac SR and piroxicam): fewer ulcer and bleeding events with nabumetone, no serious ulcer complications versus 1.4% with comparators (reconstructed abstract; one author was affiliated with the manufacturer). GI-REASONS [10] (Tier A, PROBE trial, 8,067 patients): nabumetone was 14% of the nonselective NSAID arm, which had more GI events than celecoxib (OR 1.82, 95% CI 1.31 to 2.55); nabumetone was not reported separately.

Mechanism

Present as a toxicity pathway: NSAID inhibition of prostaglandin synthesis can injure gastric and small-bowel mucosa and bleed (label class statement [1]). The non-acidic prodrug design was proposed to lower topical mucosal irritation [5]. Iron loss itself (ferritin, transferrin saturation, serum iron) was not measured in any retrieved nabumetone study.

Hard Rule 2 and 6 checks

GI bleeding and anemia are clinical outcomes, not iron deficiency; iron deficiency needs measured iron indices. Confounding by indication is presumed: rheumatoid and other chronic inflammatory disease cause anemia independent of the drug. Moore 2008 (systematic review of 51Cr fecal blood loss) was opened [9] but no nabumetone data were visible in the text retrieved, so it failed the cohort test and was not used.

Counter-evidence held on record

Counter-evidence and confounding The only direct blood-loss measurement is a 3-week healthy-volunteer study, the label itself lists fluid retention (dilution) as one cause of anemia, and the favorable meta-analytic ratios come from comparator trials with a possible sponsor influence. Large observational estimates against non-use could not be retrieved (the SOS meta-analysis abstract read listed no nabumetone figure [31]), so the real-world bleeding risk versus non-use is not established here. Framing caution This is a toxicity pathway (blood loss), not an absorption or intake deficiency. New iron deficiency, black stool or unexplained anemia on nabumetone calls for evaluating the bleeding source and the prescription, not an iron supplement by default. No iron dose is recommended here.

Limits on what this shows

• Why no grade: Both conditions of Step 7 are met: no Tier B or C evidence exists for nabumetone and copper in what was retrievable, and no circulating consumer claim was found to rebut as white. This is a disclosed gap, not a finding of safety.

What was searched

PubMed listing (parenthesized form) "nabumetone AND (copper OR zinc OR iron OR magnesium)": 4 records (a rat urine and electrolyte output study, a pediatric pain guideline review, an NSAID R-enantiomer GTPase paper mentioning magnesium, and the Barth rat microsome iron-peroxidation paper [18]); none on copper. Earlier requests in the forms "nabumetone copper", "nabumetone AND copper", "nabumetone copper complex" and "nabumetone AND (copper OR zinc OR ceruloplasmin OR metal complex)" returned HTTP 429 or 404 (not interpretable). Crossref title queries for "nabumetone copper" and "nabumetone metal complex" returned only formulary and reference entries titled Nabumetone. The Europe PMC and OpenAlex search endpoints for copper and zinc were rate-limited (HTTP 429) and not retrieved. Repair pass (2026-10-07, after a wait): a wider PubMed listing "nabumetone AND (copper OR zinc OR iron OR magnesium OR metal)" returned 10 listing summaries (J Rheumatol Suppl 1992 GI safety; CADTH pediatric pain guideline review; Am J Med 1999 NSAID renal function; salt forms of 6MNA for skin permeation; R-enantiomer GTPase paper naming magnesium; rat urine and electrolyte output; renal hemodynamics in osteoarthritis; fecal porphyrin GI blood loss; Barth-type rat microsome free radical paper; NSAID taste study with "metallic" taste), none about copper, ceruloplasmin or copper complexation of nabumetone [33]. A PubMed query for the active metabolite, "6-methoxy-2-naphthylacetic acid" AND (copper OR zinc OR metal), returned 2 unrelated records (salt formation for skin permeation; rat mast cell mediator release) [33]. A query adding "trace elements", ceruloplasmin, "metal complex", chelate and superoxide dismutase returned unrelated records (nut consumption meta-analysis, 2026 guidelines) and was read as a failed query, not a zero-hit result. Crossref bibliographic query "nabumetone zinc" returned only formulary and reference entries titled Nabumetone [33]. A second WebSearch attempt returned "budget used up" (no search performed). Europe PMC, Semantic Scholar, OpenAlex and Crossref copper queries returned HTTP 429 again and were not retrieved. The label text retrieved mentions no copper [1]; the Drugs.com nabumetone food and nutrient interaction page lists no copper [25]; the hellopharmacist nutrient depletion page lists only folic acid [24]; Antidote Research nabumetone pages returned 404.

Channel outcomes

No Tier B or Tier C source for copper with nabumetone was located. Coordination-chemistry papers for other NSAIDs were graded yellow in the finished naproxen and meloxicam dossiers, but those sources were not opened here and were not carried over (Hard Rule 6), and a nabumetone copper complex paper was not found.

Gap type

Likely a genuine literature gap for PubMed-indexed titles and abstracts (two valid parenthesized queries, 4 and 10 unrelated hits, plus a metabolite-name query with 2 unrelated hits, none concerning copper), but partly tooling-limited because Europe PMC and OpenAlex search endpoints were rate-limited, WebSearch was exhausted, and Semantic Scholar was not tried. Re-check when search access returns.

Limits on what this shows

• Null findings: A PubMed listing for nabumetone with microbiota, microbiome, dysbiosis, gut flora or intestinal bacteria terms returned exactly 3 records, all on bacterial metabolism of nabumetone. No fecal 16S, metagenomic, diversity or C. difficile study dosing nabumetone was found. The 1996 permeability review [32] (listing summary only) states nabumetone and aspirin are the exceptions among NSAIDs that did not increase intestinal permeability.

Regulatory status

No microbiome statement in the label text retrieved. Diarrhea 14%, dyspepsia 13% and abdominal pain 12% are listed GI adverse reactions (clinical trial incidence at 1% or more) with no mechanism attributed [1].

Population evidence

Reverse direction (Hard Rule 4). Jourova 2019 [21] (Tier B, in vitro and rodents; abstract opened via a BVS record): nabumetone is metabolized by bacteria to non-active metabolites, more under anaerobic conditions, with non-significant trends toward higher AUC, Cmax and half-life in rats with reduced microbiota and germ-free mice. Jourova 2020 [22] (Tier B, germ-free versus specific-pathogen-free mice, abstract via Europe PMC and a PMC page): germ-free mice formed more 6MNA, intestinal CYP expression differed, and nabumetone increased CYP1A in germ-free mice. A third paper on Clostridium bolteae keto-reductases acting on nabumetone [32] was seen only as a PubMed listing summary. CLASS EXTRAPOLATION: meta-analysis of NSAID use and Clostridium difficile-associated diarrhea [26] (Tier C, nine observational studies, more than 39,000 participants, pooled OR 1.41, 95% CI 1.06 to 1.87; abstract via DOAJ): individual NSAIDs were not named, so whether nabumetone was in the constituent cohorts cannot be determined; capped at yellow. Tiihonen 2010 [27] (human elderly NSAID users versus non-users, 16S): the abstract did not name the NSAIDs, so the cohort test could not be completed and it was not used.

Mechanism

Named for the drug effect on microbiota: none. Named for microbiota effect on the drug: bacterial enzymatic conversion to inactive metabolites [21][22].

Counter-evidence held on record

Counter-evidence and confounding The graded signal is a context flag, not a drug-to-microbiome finding. The class meta-analysis rests on retrospective exposure history, confounded by hospitalization and antibiotic use, and does not isolate nabumetone. Framing caution This entry does not support a probiotic, fiber or antibiotic decision. Persistent diarrhea or black stools on nabumetone belongs with the prescriber.

Limits on what this shows

• Null findings: No retrieved study showed a hemoglobin decline on nabumetone, and none excluded one. The hematologic reactions are labeled as unknown causal relationship [1].

Regulatory status

Adverse reactions: anemia, leukopenia, granulocytopenia and thrombocytopenia appear under incidence below 1% with "Causal Relationship Unknown"; edema 3 to 9%. Warnings carry the class anemia sentence quoted in the Iron entry and advise checking hemoglobin or hematocrit if signs or symptoms of anemia appear [1]. The NSAID class warning that these drugs inhibit platelet aggregation and prolong bleeding time in some patients is also on the label [1].

Population evidence

No human hemoglobin, hematocrit, methemoglobin or red-cell mass study of nabumetone against a control was retrieved. A PubMed listing for nabumetone with hemolytic, methemoglobinemia or erythrocyte terms returned 5 records, none a human hematologic outcome study: two phototoxicity papers (Pharmazie 2000 and 2005, the second on 6MNA), a gastric cytotoxicity paper, a nabumetone versus diclofenac rheumatoid arthritis trial (listing summary: endoscopic ulcer outcome) and a leukemia cell-line paper. Canudas 2000 [19] (Tier B, in vitro, reconstructed abstract): nabumetone photodegradation was faster in oxygenated conditions; in red blood cell studies the phototoxic response was reduced by oxygen quenchers and radical scavengers including glutathione and sodium azide, implicating singlet oxygen and free radicals. The Eversmeyer 1993 comparative safety report [29] reportedly noted fewer hemoglobin decreases with nabumetone than diclofenac, but only a PubMed listing summary was seen (abstract not opened), so it is not relied on.

Mechanism

Anemia is routed through blood loss (Iron entry), fluid dilution (Sodium entry) and an "incompletely described effect upon erythropoiesis" per the label [1]. The in vitro photohemolysis requires light exposure and says nothing about oxygen-carrying capacity in people on therapy.

Hard Rule 2 check

Anemia is an outcome, not a deficiency of oxygen delivery or of iron. A drug's own hemodynamic effect is not at issue for nabumetone. This is an outcome entry. It shares its label sentence and mechanism with the Iron and Sodium entries (Overlap map).

Counter-evidence held on record

Counter-evidence and confounding Chronic inflammatory disease, age and concurrent aspirin or anticoagulant use all lower hemoglobin independent of nabumetone, and the one human hemoglobin comparison (Eversmeyer) was not opened. The photohemolysis result is a UV-dependent in vitro finding. Framing caution Do not read this as proof that nabumetone impairs oxygen carrying or causes iron deficiency. Unexplained fatigue, pallor or breathlessness on an NSAID calls for a hemoglobin check and bleeding evaluation, not a supplement.

Limits on what this shows

• Null findings: No effect with succinate [15]; no uncoupling [15][16]. No human or animal in vivo ATP, phosphocreatine or lactate measurement with nabumetone was found. No circulating "low energy" claim for nabumetone was found.

Regulatory status

No ATP or energy-metabolism statement in the label text retrieved [1].

Population evidence

Moreno-Sanchez 1999 [15] (Tier B): nabumetone "inhibited respiration, ATP synthesis, and membrane potential in the presence of glutamate + malate, but not with succinate"; the other uncoupling drugs in the study (diclofenac, piroxicam, indomethacin, nimesulide, meloxicam) inhibited ATP synthesis by collapsing the membrane potential, whereas nabumetone acted by inhibiting respiration. Mahmud 1996 [16]: no uncoupling by nabumetone, so the ATP effect is not via a proton leak. Huang 2026 [17]: impaired metabolic function in human cardiomyocytes; ATP content was not stated in the reconstructed abstract.

Mechanism

Named in vitro: site 1 inhibition lowers NADH-linked respiration and thereby ATP synthesis [15].

Hard Rule 7 and 9 checks

ATP production is a separate entry from Mitochondrial burden and from Oxygen delivery/utilization, but it shares its only source ([15], with [16][17]) with Mitochondrial burden; the Overlap map records this so the two are not counted as independent confirmation.

Counter-evidence held on record

Counter-evidence and confounding One in vitro preparation; the abstract did not give drug concentrations, so relevance to plasma 6MNA is unknown. The label text retrieved carries no energy-metabolism statement. Framing caution Not evidence of fatigue or low energy caused by nabumetone. Fatigue on an NSAID is far more likely to reflect the underlying condition, anemia or sleep, and belongs with the prescriber.