Sulfasalazine

Limits on what this shows

• Key null / negative: Cell-type selectivity (synoviocytes and colon carcinoma cells unaffected) [32]. Rat doses are far above human doses and the study is under editorial concern [33].
• At-risk subgroups: None established.

Regulatory

No mitochondrial statement in the label text extracted [1][2].

Population evidence

None in patients. Tier B: Liptay 2002, Jurkat cells (ED50 about 1.0 mM) and primary human T-lymphocytes (ED50 about 0.5 mM); Bax accumulation at mitochondria, transmembrane potential collapse, cytochrome c release, AIF translocation, fully prevented by Bcl-2 overexpression; SW620 colon carcinoma cells and synoviocytes were unaffected at these concentrations [32]. A rat study (400 to 600 mg/kg/day for 14 days) reported kidney mitochondrial succinate dehydrogenase decrease, depolarization, reactive oxygen species, glutathione depletion and swelling [33]; the publisher issued an Expression of Concern for that article in 2026, the notice was read only as a title-level stub, and no grade rests on it.

Mechanism

Named (Bax/AIF pathway) in T cells [32]; in kidney, oxidative stress with mitochondrial depolarization proposed by the rat study's authors [33].

Counter-evidence held on record

Counter-evidence on record The authors of [32] frame the T-cell effect as therapeutic. Plasma concentrations in treated patients were not compared with 0.5 to 1 mM in this review. Framing caution Not a human mitochondrial toxicity finding and not a basis for CoQ10 or other supplement advice.

Limits on what this shows

• Key null / negative: The one human study read shows lipid peroxide falling with treatment, but it cannot separate drug effect from disease remission [34]. No human glutathione or malondialdehyde study stratified by sulfasalazine was opened.
• At-risk subgroups: None identified.

Regulatory

No oxidative-stress statement in the label text extracted [1][2].

Population evidence

Human, reverse direction: Ahnfelt-Ronne 1990 (cohort: sulfasalazine-treated inflammatory bowel disease patients with rheumatoid arthritis controls; sample size not stated in the abstract) found rectal biopsy lipid peroxide elevated before treatment and normal after 5 weeks alongside improved disease activity [34]. Tier B: sulfasalazine inhibits xCT and induces ferroptotic death in glioma cells, non-toxic below 200 µM [27]; it lowered GPX4, FTH1 and SLC7A11 in synoviocytes and in collagen-induced arthritis mice [28]; it depleted thiols and raised reactive oxygen species and lipid peroxidation in zebrafish larvae co-exposed to copper [30]. A review of toxicity mechanisms (Tier C, narrative) proposes oxidative stress with 5-ASA scavenging that may generate toxic radicals [26].

Mechanism

Named: xCT inhibition, reduced cystine uptake and glutathione synthesis, lower GPX4 activity [27][28]. Opposing: 5-ASA radical scavenging [34]. Relevance of the micromolar to millimolar in vitro range to tissue exposure was not established here.

Counter-evidence held on record

Counter-evidence on record Scavenging is intended pharmacology (Hard Rule 2). The cancer-cell ferroptosis work uses concentrations of 200 µM and above and an intended antitumor framing [27]. Framing caution Not an antioxidant deficiency. No antioxidant or glutathione-precursor supplement is supported.

Limits on what this shows

• Key null / negative: Hemoglobin rose on sulfasalazine in the rheumatoid arthritis comparison despite hemolysis markers, consistent with disease-driven anemia improving [10]. No ferritin or iron-absorption study stratified by sulfasalazine was found.
• At-risk subgroups: None. Anemia in colitis or rheumatoid arthritis is presumed disease-driven until shown otherwise (Hard Rule 6); the drug's hemolysis is handled in the Oxygen delivery entry.

Regulatory

No iron statement in the label text extracted [1][2].

Population evidence

None in patients. Tier B: in K7M2 osteosarcoma cells, ferric ammonium citrate or ferrous ammonium sulfate amplified, and iron depletion attenuated, sulfasalazine-induced loss of viability [29]; deferoxamine inhibited sulfasalazine-induced ferroptosis in synoviocytes and in collagen-induced arthritis mice [28].

Mechanism

Named in vitro (xCT inhibition with iron-dependent lipid peroxidation) [29][28]. The direction is nutrient affecting the drug's cellular effect (Hard Rule 4), not the drug affecting iron status.

Counter-evidence held on record

Counter-evidence on record The label has no iron interaction and no separation interval. Cancer-cell iron loading at supraphysiologic concentrations does not translate to dietary iron effects. Framing caution Kept separate from Copper. Do not tell patients sulfasalazine causes iron deficiency; no iron dose is recommended.

Limits on what this shows

• Key null / negative: In 74 patients with rheumatoid arthritis versus 30 controls, serum copper and ceruloplasmin were higher and erythrocyte copper lower, with "pharmaceutical interventions" showing no significant impact on copper levels; sulfasalazine was not analyzed separately [31]. Copper shifts there track inflammation.
• At-risk subgroups: None attributable to the drug.

Regulatory

No copper statement in the label text extracted [1][2].

Population evidence

None stratified by sulfasalazine. Tier B: Mariano 2026, zebrafish larvae pretreated with sulfasalazine then exposed to copper; thiols depleted and reactive oxygen species and lipid peroxidation raised versus copper alone; behavioral effects of copper were not intensified [30]. A PubMed listing also returned a 2023 Carcinogenesis paper combining sulfasalazine with disulfiram-copper in cell and mouse models; its record was not opened and is not relied on.

Mechanism

Named (glutathione depletion increases susceptibility to metal-induced oxidative stress) in a non-mammalian model [30]. No mechanism for copper loss, malabsorption or retention was found.

Counter-evidence held on record

Counter-evidence on record The larval finding is an exposure interaction with an environmental copper challenge, not copper depletion or excess in patients. Framing caution Kept separate from Iron. Not a copper deficiency or excess claim; no copper supplementation is supported.

Limits on what this shows

• Key null / negative: Overall community structure unchanged at week 12 [22]; no jejunal overgrowth [20]. The 2026 in vitro paper itself criticizes earlier fecal-flora studies for not controlling other factors affecting the microbiome [24].
• At-risk subgroups: None defined. No study measured a microbiome-derived nutrient (for example vitamin K or B-vitamin synthesis) on sulfasalazine.

Regulatory

The label states that in the intestine sulfasalazine is metabolized by intestinal bacteria to sulfapyridine and 5-ASA; it contains no microbiome safety statement in the text extracted [1][3].

Population evidence

Kanerud 1994: 17 rheumatoid arthritis patients before and after 16 weeks of treatment; fecal total aerobes, E. coli and Bacteroides counts fell significantly, with minor jejunal effects and no convincing jejunal bacterial overgrowth [20]. Neumann 1987: 26 outpatients with active rheumatoid arthritis randomized to sulfasalazine or D-penicillamine; the sulfasalazine group showed substantial decreases in C. perfringens and E. coli counts and the penicillamine group did not (abstract: first sentence only; findings from the fetch tool summary) [21]. Lima 2024: 33 sequential patients with inflammatory bowel disease and peripheral spondyloarthritis (22 on sulfasalazine, 11 controls) plus a 16-subject validation cohort; "no differences in the overall microbial structure between baseline and week 12," while baseline composition (Faecalibacterium prausnitzii, butyrate capacity) differed between responders and non-responders [22]. Tier B: TNBS colitis rats (6 per group) showed 16S shifts with more short-chain-fatty-acid producers, fewer Proteobacteria and more Mycoplasma [23]; in vitro MICs of 625 µg/mL or more for the parent against Proteus mirabilis and Klebsiella pneumoniae, and 78 to 625 µg/mL for sulfapyridine [24].

Mechanism

Named: sulfapyridine has antibacterial activity in vitro [24]; sulfapyridine induced folate stress and increased butyrate synthesis in F. prausnitzii in the Lima work [22]. Bacterial azoreduction is required for activation [1][25].

Boundary (customer-visible wording)

Sulfasalazine can lower counts of some fecal bacteria (shown in small rheumatoid arthritis studies, 16 weeks or less); overall community change was not seen in one inflammatory bowel disease study.

Counter-evidence held on record

Counter-evidence on record The culture-based counts are 1987 and 1994 uncontrolled or active-comparator designs in a different disease from ulcerative colitis, and the authors of [20] note antibacterial action remains unproven as the therapeutic mechanism. The review [25] only restates these findings (Tier C). Framing caution Composition change, not an infection-risk or deficiency claim. Do not suggest probiotics or supplements on this basis.

Limits on what this shows

• Key null / negative: In rheumatoid arthritis, hemoglobin improved on sulfasalazine 2 g/day over 24 weeks while MCV rose, which the authors attributed to hemolysis-driven reticulocytosis (30 patients, sulfasalazine versus penicillamine) [10]. No G6PD-stratified sulfasalazine cohort was found. A G6PD study of a different 5-ASA drug (mesalamine) exists in the sibling dossier set but was not opened here and is not transferred (Hard Rule 6).
• At-risk subgroups: Slow acetylators, higher doses and higher serum sulfapyridine; G6PD deficiency (label, close observation); breastfed newborns and G6PD-deficient infants (sulfapyridine appears in milk at about 30% to 60% of maternal serum per label) [1][39].

Regulatory

Label warnings and adverse reactions list agranulocytosis, aplastic anemia, Heinz body anemia, hemolytic anemia and methemoglobinemia; patients with G6PD deficiency "should be observed closely for signs of hemolytic anemia"; complete blood counts and liver function tests are required before starting and every second week for the first three months [2][3].

Population evidence

Pounder 1975: 52 ulcerative colitis patients on mean 2.5 g/day maintenance versus 50 controls and 10 untreated colitis patients; 35 treated patients had one or more drug-induced red-cell abnormalities, none of the controls; 23 had contracted red cells (attributed to mild hemolysis) correlating with dose, serum sulfapyridine and acetylator status; 21 had elevated methemoglobin; 11 macrocytosis; some Heinz bodies; contraction was seen at 1.5 g/day [16]. Van Hees 1978: 36 unselected ulcerative colitis or Crohn's patients on 4.5 to 6 g/day; hemolysis in 19, with higher free sulfapyridine (P<0.001); all patients at 37 µg/mL or above had hemolysis versus 4 of 21 below; 4 had hemolytic anemia; 18 of 19 with hemolysis were slow acetylators versus 6 of 17 without [17]. Goodacre 1978: 40 inflammatory bowel disease patients; hemolysis in 17 (43%) by starch gel electrophoresis; Heinz bodies in 8 (47% of hemolysis cases); hemoglobin significantly lower in the hemolysis group; 53% had reticulocytes above 5%; correlation with serum sulfapyridine but not parent drug level; no relation to disease extent or activity [18]. Jensen 1996 (25 arthritis outpatients): about half had decreased HbA1c indicating chronic hemolysis and 28% reticulocytosis [9].

Mechanism

Named as an exposure relationship: red-cell injury tracks the sulfapyridine moiety and slow acetylation (half-life 14.8 versus 10.4 hours) [3][17][18]. Heinz bodies indicate oxidative denaturation of hemoglobin; they were present in under half of hemolysis cases, so they are not required [18]. A G6PD-independent case is documented: Heinz body hemolytic anemia after about one year of treatment with no G6PD or other red-cell enzyme deficiency (single case report) [19]. The exact molecular step was not established in the sources opened.

Counter-evidence held on record

Counter-evidence on record Overt hemolytic anemia was a minority outcome (4 of 36 at 4.5 to 6 g/day) [17]; most hemolysis was laboratory-detected. Red was assigned for causal attribution to the drug across three independent Tier A cohorts with direct measurement, not for a nutrient deficit. Anemia in colitis or rheumatoid arthritis has many other causes and is not attributed to the drug without hemolysis markers. Framing caution Toxicity entry, kept separate from ATP production and from Iron. No supplement addresses it; nutrient repletion would not treat it. Dose and monitoring decisions belong to the prescriber.

Limits on what this shows

• Key null / negative: No clinical study of fatigue or energy output attributable to sulfasalazine was found. Crossref and PubMed searches for sulfasalazine with ATP and mitochondrial respiration returned no ATP-measuring study (see Ungraded section for the channels used).
• At-risk subgroups: None defined.

Regulatory

No energy-metabolism statement in the label text extracted [1][2].

Population evidence

None in patients. Tier B surrogates, shared with Mitochondrial burden: membrane-potential collapse in T cells [32]; succinate dehydrogenase decrease in rat kidney mitochondria, under an Expression of Concern [33]. The Liu 2022 osteosarcoma study reports altered mitochondrial morphology with iron supplementation during sulfasalazine exposure, without an ATP readout in the abstract [29].

Mechanism

Plausible only: loss of mitochondrial membrane potential would be expected to reduce ATP synthesis, but that step was not measured.

Counter-evidence held on record

Counter-evidence on record This entry adds no independent evidence beyond Mitochondrial burden; the shared origin is recorded in the Overlap map. Framing caution Kept separate from Mitochondrial burden and Oxygen delivery/utilization. Fatigue in colitis or rheumatoid arthritis is not an ATP finding for this drug and no supplement is supported.