Mercaptopurine

Limits on what this shows

• Null findings: Menor 2004 [14]: azathioprine caused mitochondrial swelling and increased oxygen consumption in rat hepatocytes; 6-MP showed no effect on GSH content or metabolic activity (mitochondrial endpoints for 6-MP not stated in the summary read).

Regulatory status

Mitochondria are not mentioned in the label text retrieved [1][2][3].

Population evidence

ACMS 2016 [15]: isolated rat hepatocytes; 6-MP IC50 400 micromol/L at 2 hours; ROS formation, collapse of mitochondrial membrane potential and lysosomal damage; L-carnitine, chloroquine, diphenyleneiodonium and DMSO each reduced the effects. Tapner 2004 [12]: both drugs at 0.5 to 5 micromol/L caused mitochondrial damage and ATP depletion leading to necrosis; the ultrastructural mitochondrial abnormalities and mitochondrial GSH fall are described in the azathioprine arm in the summary read.

Mechanism

Named: ROS-driven mitochondrial and lysosomal injury in hepatocytes [15]; xanthine oxidase involvement via allopurinol protection [12].

Class cohort test

CLASS EXTRAPOLATION: azathioprine mitochondrial details [14][23] (azathioprine in cohort, mercaptopurine not or not stated). Mercaptopurine-specific data are Tier B hepatocyte models only.

Counter-evidence held on record

Counter-evidence and confounding In vitro hepatocyte models, no human mitochondrial biomarker, and the highest concentration is about 650 times the peak plasma level after 50 mg (arithmetic by the reviewer). Mitochondrial burden, oxidative stress and ATP share the same hepatocyte experiments (see Overlap map). Framing caution Do not frame this as mitochondrial damage in patients. No mitochondrial supplement, including L-carnitine, is recommended.

Limits on what this shows

• Null findings: Menor 2004 [14]: in rat hepatocytes azathioprine depleted GSH and reduced metabolic activity while 6-MP showed no effect on GSH content or metabolic activity. Pelin 2015 [19]: azathioprine, not 6-MP, induced concentration-dependent superoxide in human hepatic and intestinal cell lines (GST-M1 dependent), as read from a PubMed listing summary. N-acetylcysteine did not prevent thiopurine liver injury in the 9-patient trial [17].

Regulatory status

Oxidative stress is not mentioned in the label text retrieved [1][2][3].

Population evidence

Tulumen 2015 [16]: rats, 14 days; 6-MP alone lowered SOD, GPx and catalase and raised MDA (P under 0.005) and raised transaminases. Tapner 2004 [12]: rat hepatocytes, 6-MP or azathioprine 0.5 to 5 micromol/L; viability fell 5 to 34% by day 1 and 42 to 92% by day 4; allopurinol (20 micromol/L) and Trolox (2 mmol/L) gave near-complete protection; toxicity attributed to oxidative stress, mitochondrial damage and ATP depletion. Petit 2008 [13]: human primary hepatocytes and HepaRG cells; azathioprine and 6-MP induced glutathione synthesis and antioxidant enzymes. Misdaq 2015 [18]: Jurkat T cells with TPMT knockdown; 6-MP and 6-thioguanine altered SOD1, SOD2 and catalase. ACMS 2016 [15]: rat hepatocytes; ROS formation at 400 micromol/L, reduced by protective agents. A systematic review of 69 studies on thiopurine biotransformation and oxidative stress exists [20] (listing summary only; not used for grading). Human: van Asseldonk 2024 [17]: 9 IBD patients on thiopurines (agent not specified in the abstract), F2-isoprostanes rose from 101 to 157 ng/mmol on thiopurine reintroduction (p = 0.038).

Mechanism

Named: ROS formation and antioxidant enzyme change in hepatocytes and T cells [12][15][18]. The in vitro range 0.5 to 5 micromol/L brackets the peak plasma level after a 50 mg dose (about 0.6 micromol/L, see Pharmacokinetic basis); 400 micromol/L is far above it.

Class cohort test

CLASS EXTRAPOLATION applies to [17] (thiopurine agent not named in the abstract, so mercaptopurine in the cohort is unconfirmed, Tier C), to azathioprine data in [14][19] and to Lee and Farrell 2001 [23] (azathioprine in cohort, mercaptopurine not). Mercaptopurine-specific support is Tier B only (animal and cell studies).

Counter-evidence held on record

Counter-evidence and confounding No antioxidant-capacity measure was found in people taking mercaptopurine, and the data conflict on glutathione. Direction is a pro-oxidant load, not a depleted antioxidant nutrient. Disease inflammation in IBD and leukemia raises oxidative markers independent of the drug. Framing caution This is not a signal that antioxidant supplements are needed. N-acetylcysteine did not protect the liver in the small human study [17]. No supplement dose is recommended.

Limits on what this shows

• Null findings: None located.
• At-risk subgroups: Not defined; the model is pregnant rats at a teratogenic injected dose.

Regulatory status

Iron is not mentioned in the US label or SmPC text retrieved [1][2][3].

Population evidence

Amemiya, Keen and Hurley 1986, Teratology [29]: pregnant rats on diets with 4.5, 100 or 1000 micrograms zinc per gram received 6-MP injections on gestational day 11. Tissue mineral analysis showed pronounced reductions in zinc, copper, iron, calcium and magnesium in maternal and fetal specimens. Abstract read via OpenAlex; the iron values and dose were not captured. No human iron data were located: PubMed listing "mercaptopurine copper" surfaced this study only.

Mechanism

Not established; the authors suggest the drug disrupts mineral metabolism and that this contributes to teratogenesis [29].

Counter-evidence held on record

Counter-evidence and confounding Single study, animal, injected, a different exposure from oral dosing, and the same source supports copper, zinc, calcium and magnesium (see Overlap map). People on mercaptopurine for IBD or ALL have independent causes of iron deficiency and anemia. Framing caution Iron deficiency is not anemia and anemia is not iron deficiency (Hard Rule 2). No iron supplementation is recommended; excess iron carries its own risks. The oxygen delivery entry covers the marrow effect.

Limits on what this shows

• Null findings: None located.
• At-risk subgroups: Not defined.

Regulatory status

Copper is not mentioned in the US label or SmPC text retrieved [1][2][3].

Population evidence

Amemiya 1986 [29]: pronounced reductions in tissue copper, with zinc, iron, calcium and magnesium, in maternal and fetal rat tissues after 6-MP (details as under Iron). Kela and Vijayvargiya 1981, Biochem J [30]: 6-MP formed a stable 1:1 complex with copper (log K 6.74) and reversed copper inhibition of xanthine oxidase in vitro; read from a PubMed listing summary. No human serum copper or ceruloplasmin data were located for mercaptopurine.

Mechanism

In vitro chelation is demonstrated; whether it changes copper status in people is not.

Counter-evidence held on record

Counter-evidence and confounding Chelation in a test tube and a teratogenic rat model do not show a human copper deficit. Copper is kept separate from iron here although one rat study supports both (Hard Rule 7). Framing caution No copper supplementation is recommended from this entry.

Limits on what this shows

• Null findings: Maier 2018 main text does not name the drug [26]. No human microbiome study isolating mercaptopurine was found.

Regulatory status

The microbiome is not mentioned in the label text retrieved [1][2][3].

Population evidence

Xu 2024 [24]: beagles and mice, 2 weeks of 6-MP; microbiota perturbation with predicted impaired short-chain fatty acid production and altered amino acid synthesis; antibiotic microbiota depletion lowered S-adenosylmethionine (SAM) and increased intestinal damage and permeability; methylated 6-MP intermediates correlated negatively with SAM in pseudo-germ-free rats; Alistipes abundance correlated with SAM. Abstract read via OpenAlex. Becker 2023 review [25], read through a summarizing fetch: thiopurines were not among the IBD drugs the abstract says substantially alter microbial diversity; the body text described reduced diversity in thiopurine-treated patients and dose-dependent growth inhibition of some strains by azathioprine and 6-MP in vitro, but the reference attributions returned by the summarizer did not match and were discarded. Maier 2018 [26] screened more than 1,000 drugs against 40 gut strains; neither mercaptopurine nor azathioprine appears in the main text (supplementary tables not opened).

Mechanism

Named in the animal data: altered SCFA and amino acid pathways and a SAM link [24].

Reverse direction (Hard Rule 4)

Microbes act on the drug: Blautia wexlerae raised xanthine dehydrogenase activity and converted 6-MP to an inactive metabolite in azathioprine failure [27]; Veillonella parvula metabolizes thiopurines through xdhA [28]. Both read from PubMed listing summaries.

Class cohort test

Human diversity statements in [25] concern thiopurines as a group (azathioprine and mercaptopurine) and are CLASS EXTRAPOLATION: cohorts not verified to contain mercaptopurine. Capped at yellow.

Counter-evidence held on record

Counter-evidence and confounding Animal study of 2 weeks; IBD itself alters the microbiome and patients take other agents. Direction: composition and functional change (excess disturbance), not a depletion. Framing caution No probiotic or prebiotic is recommended by this entry. The microbiome-to-drug findings are pharmacokinetic and relate to dosing decisions made by prescribers.

Limits on what this shows

• Null findings: Bernstein 1994 describes low-dose therapy as associated with minimal hematologic toxicity, with no leukopenia under 2.8 [6]. No hemoglobin-specific null was retrievable.
• At-risk subgroups: TPMT or NUDT15 deficiency; allopurinol co-use; aminosalicylates (TPMT inhibition); methotrexate (raises 6-MP AUC 31 to 93%); trimethoprim-sulfamethoxazole; renal or hepatic impairment [1][2].

Regulatory status

Purixan label: myelosuppression "manifested by anemia, leukopenia, thrombocytopenia" is the most serious adverse effect; adverse reactions over 20% are myelosuppression including anemia, neutropenia, lymphopenia and thrombocytopenia (anemia is not separated out in the text retrieved) [1][2]. Patients homozygous deficient in TPMT or NUDT15 typically need 10% or less of the recommended dose [1]. The SmPC says full blood counts must be taken daily during remission induction, that suppression is reversible if the drug is withdrawn early enough, and gives NUDT15 variant frequencies of about 10% in East Asians, 4% in Hispanics, 0.2% in Europeans and 0% in Africans [3].

Population evidence

Bernstein 1994 [6]: 57 inflammatory bowel disease patients on low-dose 6-MP (mostly 50 mg/day), prospective hematologic monitoring. 16 (28%) developed mild leukopenia (WBC under 4.5 x 10³/mm³), none under 2.8; leukopenia appeared no earlier than 3 months and as late as 42 months; 9 (16%) developed macrocytosis (MCV above 101 fL). The title reports minimal hematologic toxicity. Hemoglobin values were not retrievable from the abstract text read. Heerasing 2016 [7]: 67 IBD patients on azathioprine or 6-MP (pooled); MCV above 101 fL excluded a subtherapeutic 6-TGN level with positive predictive value 92%; 6-TGN and MCV correlated (R = 0.42, P = 0.0005).

Mechanism

Named. TGNs incorporate into DNA and RNA and inhibit de novo purine synthesis, killing rapidly dividing cells, including marrow progenitors [1]. Reduced TPMT or NUDT15 activity lets cells accumulate excess 6-TGNs and raises the risk of severe myelosuppression [1] sec 12.5.

Counter-evidence held on record

Counter-evidence and confounding Confounding by indication is presumed (Hard Rule 6): anemia in leukemia and in inflammatory bowel disease also arises from the disease, other chemotherapy, blood loss, iron deficiency and chronic inflammation. The label figure is a pooled myelosuppression rate, not a hemoglobin rate attributable to mercaptopurine alone. Not red: only the label is a Tier A source with an anemia endpoint, and no independent primary hemoglobin study was opened. A raised MCV is an exposure marker, not B12 or folate deficiency [6][7]. Framing caution This is a toxic effect of the drug, not a nutrient deficiency. Anemia is not iron, B12 or folate deficiency (Hard Rule 2). Do not recommend iron, folate or B12 from this entry, and do not alter the dose without the prescriber.

Limits on what this shows

• Null findings: None located. No human ATP, lactate or energy expenditure study of mercaptopurine was found.

Regulatory status

ATP is not mentioned in the label text retrieved [1][2][3]. The label lists symptomatic hypoglycemia in children under 6 [1]; that is a clinical outcome and is not evidence of reduced ATP.

Population evidence

Petit 2008 [13]: human primary hepatocytes and HepaRG cells; ATP depletion after 24 hours with azathioprine and 6-MP, toxicity at 96 hours. Tapner 2004 [12]: ATP depletion with mitochondrial injury in rat hepatocytes; adding glucose restored ATP and arrested injury in the azathioprine arm. Liliemark 1990 [21]: murine WEHI-3b leukemia cells; 6-MP 10 nM to 50 micromol/L; ATP depletion accompanied 6-thioinosinate accumulation and inhibited purine synthesis, which may limit activation. Stet 1994 [22]: human Molt F4 lymphoblasts; 6-MP and methylmercaptopurine riboside lowered S-adenosylmethionine through ATP depletion from purine synthesis inhibition.

Mechanism

Named: inhibition of de novo purine synthesis depletes adenine nucleotides; ATP is needed to convert methionine to S-adenosylmethionine [22].

Class cohort test

Azathioprine ATP data (Lee and Farrell 2001 [23]) is class extrapolation (azathioprine in cohort, mercaptopurine not) and was not used to raise this grade.

Counter-evidence held on record

Counter-evidence and confounding Cell models at varying concentrations and durations; ATP depletion also follows any cell death. Kept separate from Mitochondrial burden and Oxygen delivery per Hard Rule 7, though sources overlap. Framing caution Not a claim of low energy in patients. No energy-support supplement is recommended.