Methotrexate Toxicity

Complex MONDO:0034212 Pathograph 47 Show in embeddings browser Drug Toxicity

Methotrexate is a counterfeit folate. It is close enough to the real vitamin that the cell's own folate importer carries it inside and the cell's own ligase chains glutamates onto it, which traps it there; then it jams the enzyme that regenerates tetrahydrofolate, and the cell can no longer make thymidine or purines. Every cell the drug enters gets the same lesion, so which tissue is injured is not a question about the lesion. It is a question about which tissues are replacing themselves right now: marrow, gut epithelium, liver. What makes the poisoning distinctive is that the second half of the mechanism is not pharmacodynamic at all. Over 90% of the drug leaves by the kidney, and it is poorly soluble in acid urine, so at high dose it crystallises in the tubules and injures them. An injured kidney clears less methotrexate, a higher methotrexate concentration crystallises more, and the loop tightens on itself. That is why the two halves of the entry are asymmetric: the folate arm decides what the injury looks like, and the renal arm decides how bad and how long it gets. It is also why the countermeasures split the same way. Hydration and urine alkalinisation act on the renal arm and leucovorin on the folate arm, while glucarpidase and dialysis act upstream of both, on the circulating pool -- which is exactly why neither of those two touches the drug already inside cells.

Ask OpenScientist

Ask a research question about Methotrexate Toxicity. OpenScientist will conduct autonomous deep research using the Disorder Mechanisms Knowledge Base and PubMed literature (typically 10-30 minutes).

Submitting...

Do not include personal health information in your question. Questions and results are cached in your browser's local storage.

◈

Mechanistic Hypotheses

Adenosine accumulation as the cause of acute methotrexate encephalopathy

adenosine_neurotoxicity EMERGING

Evidence balance 1 support

The proposal is that inhibition of AICAR transformylase backs up AICA-ribotide, which raises extracellular adenosine, and that adenosine acting on central receptors produces the vasoactive and neuromodulatory changes seen as acute methotrexate encephalopathy. What supports it is a measurement and a response: central adenosine is raised in affected patients, and aminophylline, which displaces adenosine from central receptors, has reversed symptoms in a small uncontrolled series. What it lacks is any controlled demonstration, and the source that reports it calls it potential rather than established. It is recorded here so that the edges depending on it are visibly conditional.

Pathograph links

De Novo Purine Synthesis Blockade -> Adenosine Accumulation INDIRECT KNOWN INTERMEDIATES Substrate accumulating behind the AICAR transformylase block is the proposed source of the adenosine surge. The intermediates are known and named, but the step is part of a hypothesis rather than a settled mechanism in this disease.

Adenosine Accumulation -> Cerebral White Matter Injury INDIRECT UNKNOWN INTERMEDIATES The proposed route from adenosine to white matter injury is vasoactive and neuromodulatory rather than a demonstrated chain, so the intermediates are recorded as unknown.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"A potential mechanism of neurotoxicity is the accumulation of adenosine after MTX-induced reductions in purine synthesis"

The hedged wording is the point: the source advances this as a candidate mechanism, which is why the hypothesis is EMERGING rather than CANONICAL.

Discussions and Knowledge Gaps

What produces methotrexate pneumonitis, given that it is commoner on weekly low-dose therapy than after high-dose infusion?

KNOWLEDGE GAP pneumonitis_mechanism_unknown

Attached to

phenotypes#Interstitial pneumonitis

Every other toxicity in this entry scales with concentration and duration, which is why the pathograph is built around exposure and clearance. Pneumonitis inverts that relationship: it is reported in about 0.5% of patients per year on weekly low-dose methotrexate and is rare after high-dose infusion. An inverted dose-response is not a minor discrepancy, it is evidence that the lesion does not belong to the folate arm at all, and hypersensitivity has long been the favoured alternative. The entry deliberately leaves the phenotype unattached to any pathophysiology node rather than inventing an edge from the folate block, because an edge drawn to tidy up the graph would assert exactly the concentration dependence the epidemiology contradicts.

Do NSAIDs and proton pump inhibitors clinically delay methotrexate elimination in humans, or is the interaction a mouse finding that has been carried into practice without human confirmation?

KNOWLEDGE GAP nsaid_interaction_contested

Attached to

environmental#Co-administration of non-steroidal anti-inflammatory drugs

The two evidence items on this exposure's mechanism link disagree, and the disagreement is recorded rather than resolved. The mechanistic demonstration is a mouse study showing indomethacin and ketoprofen inhibit renal tubular methotrexate transporters. A 2023 transporter-focused review of the clinical literature concludes there is no firm evidence of an interaction with NSAIDs or proton pump inhibitors in humans, and notes that most reports are case reports or series. Protocols continue to substitute these drugs before high-dose infusion, which is defensible as asymmetric risk but is not the same as the interaction being established. What would settle it is a controlled pharmacokinetic study in patients rather than more case reports.

How large is the MTHFR C677T effect on methotrexate toxicity, and does it replicate outside single-centre oncology cohorts?

KNOWLEDGE GAP mthfr_effect_size_uncertain

Attached to

genetic#MTHFR

The association cited here comes from 148 courses in 32 patients with one tumour type, and the nephrotoxicity odds ratio carries a confidence interval from 1.65 to 103.86. An interval spanning nearly two orders of magnitude is compatible with a large effect and with a barely detectable one. The entry records the gene as a susceptibility modifier on that basis but does not support using the genotype to predict toxicity in practice, which is the use the source proposes.

⚙

Pathophysiology

Systemic Methotrexate Exposure

The initiating event is the presence of methotrexate in the circulation at a concentration the patient cannot clear on the expected schedule. This is a single node because everything downstream follows from concentration and duration rather than from route: the same lesion is reached by a high-dose oncology infusion, by a weekly low-dose regimen taken daily in error, and by an ordinary dose given to a patient whose kidney cannot excrete it.

SLCO1B1 hgnc:10959

Downstream

Cellular Methotrexate Uptake by the Reduced Folate Carrier Circulating drug reaches the intracellular compartment only through carrier-mediated import, so the systemic concentration sets the rate at which the intracellular lesion is established.

Cerebral White Matter Injury Central nervous system toxicity is an established consequence of the drug independently of any account of how it happens. This edge carries no `hypothesis_groups` on purpose: without it the whole neurologic arm would hang only on the adenosine hypothesis, which would assert that methotrexate neurotoxicity is explicable solely by that route.

Renal Tubular Methotrexate Crystallisation The same circulating pool is delivered to the tubular lumen by glomerular filtration and tubular secretion, where its solubility rather than its pharmacology decides what happens next.

Cellular Methotrexate Uptake by the Reduced Folate Carrier

Methotrexate does not cross the membrane freely. It is imported by SLC19A1, the reduced folate carrier, which exists to bring physiological folates in and cannot distinguish them from the antifolate. Uptake capacity is therefore also the drug's selectivity: cells that import folate avidly import methotrexate avidly, and loss of carrier function is a recognised route to methotrexate resistance.

SLC19A1 hgnc:10937

reduced folate carrier antiport activity GO:0008518

Downstream

Intracellular Methotrexate Polyglutamation Import delivers the monoglutamate form to the cytosolic enzyme that extends it.

Show evidence (2 references)

PMID:36071163 SUPPORT BACKGROUND In Vitro

"The human reduced folate carrier (hRFC, also known as SLC19A1) is the major importer of folates into the cell"

Identifies the transporter that carries methotrexate into the cell. Quoted from the paper's opening framing rather than from its own structural result, hence BACKGROUND.

PMID:36071163 SUPPORT BACKGROUND In Vitro

"it is a major determinant in methotrexate (antifolate) sensitivity"

Supports treating uptake as a determinant of the lesion rather than a preliminary step, since carrier function sets how sensitive a cell is. Also background rather than this paper's own finding: its abstract reports a methotrexate-bound structure, but the body states that no cryo-EM density for the drug was observed in the central cavity.

Intracellular Methotrexate Polyglutamation

Once inside, methotrexate is treated as a folate by folylpolyglutamate synthetase, which adds a chain of glutamate residues. The chain is the reason toxicity outlives the plasma level: the polyglutamate is too charged to leave the cell, so it accumulates and is retained long after the extracellular drug has gone. Retention increases sharply with chain length, which is why an exposure that looks brief in plasma is not brief inside the cell.

Downstream

Dihydrofolate Reductase Inhibition Polyglutamation both concentrates the drug at its target and raises its affinity for the folate-dependent enzymes.

Show evidence (2 references)

PMID:2416193 SUPPORT In Vitro

"Methotrexate is glutamylated in cultured hepatoma cells to derivatives that contain a total of 2 to 5 gamma-glutamyl residues."

Establishes that the intracellular drug is converted to polyglutamate derivatives, and the chain lengths reached.

PMID:2416193 SUPPORT In Vitro

"The importance of the longer-chain-length polyglutamates is apparent from the 6-hr retention of the polyglutamate species: Glu2, 15%; Glu3, 21%; Glu4, 50%; and Glu5, 83%."

Quantifies the retention claim directly: the longest chains are still largely present at six hours, which is the measured basis for treating polyglutamation as the trapping step.

Dihydrofolate Reductase Inhibition

The central molecular lesion. Methotrexate occupies the folate site of dihydrofolate reductase and blocks the reduction of dihydrofolate to tetrahydrofolate. The enzyme is not damaged, only occupied, which is exactly why a competing reduced folate given from outside can rescue the cell without the enzyme ever being repaired.

DHFR hgnc:2861

dihydrofolate reductase activity GO:0004146 ↓ DECREASED

Downstream

Tetrahydrofolate Depletion Blocking the reductase removes the only route by which the cell regenerates tetrahydrofolate from the dihydrofolate produced each time thymidylate is made.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS In Vitro

"After entry into the cell, methotrexate is polyglutamated, binds dihydrofolate reductase (DHFR) with an affinity 1,000-fold greater than that of folate, and competitively inhibits conversion of dihydrofolate to tetrahydrofolate"

States the mechanism and the affinity margin over the natural substrate, which is what makes the block near-complete at therapeutic concentrations.

Tetrahydrofolate Depletion

Tetrahydrofolate is the carrier for every one-carbon transfer the cell performs. Depleting it does not disable one pathway but the shared currency of several, which is why a single enzyme block produces a synthesis failure on two separate nucleotide branches at once.

MTHFR hgnc:7436

tetrahydrofolate metabolic process GO:0046653 ↓ DECREASED

Downstream

Thymidylate Synthesis Blockade

De Novo Purine Synthesis Blockade

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS In Vitro

"Tetrahydrofolate is essential for biosynthesis of thymidine and purines, which are needed for synthesis of DNA."

Names the two downstream branches that tetrahydrofolate supplies, which is the reason this node has two outgoing edges rather than one.

Thymidylate Synthesis Blockade

Thymidylate synthase needs 5,10-methylenetetrahydrofolate as its one-carbon donor, so the folate block starves it of cofactor. This is the cell's only de novo source of thymidine monophosphate, and there is no salvage route that can replace it at the rate a dividing cell consumes it.

TYMS hgnc:12441

dTMP biosynthetic process GO:0006231 ↓ DECREASED

Downstream

Uracil Misincorporation and DNA Strand Breakage Blocking the reaction does not only remove thymidylate; it leaves the unconsumed precursor pool to accumulate as dUTP, which is the substrate for the next step.

Uracil Misincorporation and DNA Strand Breakage

With thymidylate scarce and dUTP abundant, the polymerase puts uracil into DNA. Uracil-DNA glycosylase excises it, the repair patch is resynthesised from the same depleted pool, and uracil goes back in. The futile cycle is what converts a shortage of one nucleotide into physical strand breakage. It is the uracil-DNA arm of what is classically called thymineless death, alongside the thymidylate pool depletion that the term originally named.

Downstream

Replicating Cell Death

Show evidence (1 reference)

PMID:12374095 SUPPORT REVIEW SYNTHESIS In Vitro

"cytotoxicity results from a process known as "thymineless death""

Names the cytotoxic mechanism attributed to thymidylate synthase inhibition in this branch.

De Novo Purine Synthesis Blockade

Two steps of purine ring assembly, catalysed by GART and by AICAR transformylase, are formyl-tetrahydrofolate-dependent and stall for the same reason thymidylate synthesis does. The branch matters twice over: it contributes to the synthesis failure that kills dividing cells, and the substrate that accumulates behind the ATIC block is the starting point of the adenosine hypothesis of neurotoxicity.

Adenosine accumulation as the cause of acute methotrexate encephalopathy

ATIC hgnc:794

de novo purine nucleotide synthesis GO:0006189 ↓ DECREASED

Downstream

Replicating Cell Death

Adenosine Accumulation Substrate accumulating behind the AICAR transformylase block is the proposed source of the adenosine surge. The intermediates are known and named, but the step is part of a hypothesis rather than a settled mechanism in this disease.

Show evidence (1 reference)

PMID:32066940 SUPPORT REVIEW SYNTHESIS Other

"multiple mechanisms potentially contribute to the anti-inflammatory actions of methotrexate, including the inhibition of purine and pyrimidine synthesis"

Confirms that purine synthesis inhibition is an established action of the drug, cited here for the pathway and not for the anti-inflammatory conclusion the sentence is making.

Adenosine Accumulation

Substrate accumulating behind the ATIC block is followed by a rise in extracellular adenosine; the intervening enzymology is usually given as inhibition of adenosine and AMP deaminases, which none of the sources cited here states, so it is named as the usual account rather than asserted. In rheumatoid arthritis this is the favoured explanation of why a weekly low dose is anti-inflammatory. In the toxicity setting the same molecule is the leading candidate for acute methotrexate encephalopathy, on the strength of raised central adenosine in affected patients and symptom reversal with an adenosine antagonist. It is a hypothesis, and the entry treats it as one.

Adenosine accumulation as the cause of acute methotrexate encephalopathy

Downstream

Cerebral White Matter Injury The proposed route from adenosine to white matter injury is vasoactive and neuromodulatory rather than a demonstrated chain, so the intermediates are recorded as unknown.

Show evidence (1 reference)

PMID:32066940 SUPPORT REVIEW SYNTHESIS Other

"as well as the promotion of adenosine release and expression of certain long non-coding RNAs"

Records adenosine release as one of the drug's recognised downstream actions, which is the premise the neurotoxicity hypothesis builds on. Quoted from a review of the drug's anti-inflammatory action, so it establishes that the drug promotes adenosine release and not that adenosine is neurotoxic.

Replicating Cell Death

The convergence point of the folate arm. A cell that cannot finish replication and is accumulating strand breaks dies or arrests, and the selectivity of the whole drug is contained in the word replicating: the lesion is identical everywhere and the consequence is confined to the tissues that are dividing.

DNA replication GO:0006260 ↓ DECREASED

Downstream

Bone Marrow Hematopoietic Failure

Gastrointestinal Mucosal Barrier Breakdown

Hepatocellular Injury

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS In Vitro

"Blockade of tetrahydrofolate synthesis by methotrexate leads to inability of cells to divide and to produce proteins."

States the cellular consequence of the folate block that this node represents.

Bone Marrow Hematopoietic Failure

Haematopoiesis is the most continuously proliferative tissue in the adult, so it is the first to fail and the last to recover. Without leucovorin rescue, marrow suppression is the dose-limiting toxicity of high-dose methotrexate; with pharmacokinetically guided rescue it is uncommon unless elimination is delayed.

hematopoietic precursor cell CL:0008001

Downstream

Sepsis Loss of the neutrophil defence is one of the two inputs to the septic outcome; the other is the breached mucosal barrier, and the entry draws both edges because the deaths follow the two arriving together.

Pancytopenia

Decreased total neutrophil count

Thrombocytopenia

Bone marrow hypocellularity

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"A dose-limiting toxicity of HDMTX in the absence of leucovorin is severe, prolonged myelosuppression"

Establishes marrow failure as the dose-limiting consequence of the unrescued folate block.

Gastrointestinal Mucosal Barrier Breakdown

The crypt is among the fastest-cycling compartments in the body, and the villus epithelium above it is replaced from there. Skin is the third such compartment and fails the same way, which is why the dermatologic toxicity hangs off this node too. Killing the crypt removes the replacement without removing the loss, so the villus shortens and the barrier opens. Mucositis is therefore not only painful in its own right but a portal: it is the route by which a neutropenic patient becomes a septic one.

intestinal crypt stem cell CL:0002250 enterocyte CL:0000584

Animal models

Methotrexate-induced small intestinal mucositis in the rat Rat recapitulates Produces the crypt and villus lesion and the tight-junction changes that define the human node.

Downstream

Sepsis The barrier failure is the portal of entry. It is not sufficient alone, which is why this edge and the one from marrow failure converge on the same node rather than either carrying the outcome by itself.

Mucocutaneous toxicity

Oral ulcer

Diarrhea

Nausea

Hepatocellular Injury

The liver is injured in two different ways by the same drug on two different schedules, and conflating them is the commonest error about methotrexate hepatotoxicity. After high-dose exposure the injury is an acute, transient, largely inconsequential transaminase rise. Under years of weekly low-dose exposure the concern is instead cumulative fibrosis, and how that risk divides between the drug and coexisting fatty liver disease -- which both raises the risk of methotrexate hepatotoxicity and is itself worsened by the drug -- is actively contested.

hepatocyte CL:0000182

Downstream

Elevated circulating hepatic transaminase concentration

Hepatic fibrosis Fibrosis follows repeated low-grade injury over years rather than any single exposure, so the link runs through a chronic injury-repair cycle rather than directly.

Show evidence (2 references)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Hepatotoxicity after HDMTX is much less common than with the lower, long-term oral methotrexate dosing that is used in patients with rheumatoid arthritis, who are at risk for liver fibrosis"

Supports separating the two hepatic phenotypes by dose regimen, which is what this node's two downstream targets represent.

PMID:36564450 SUPPORT REVIEW SYNTHESIS Human Clinical

"patients with NAFLD are at increased risk for methotrexate hepatotoxicity, and methotrexate can worsen the course of NAFLD"

Records the interaction with fatty liver disease that qualifies how much of the chronic fibrosis risk is attributable to the drug alone.

Cerebral White Matter Injury

Methotrexate neurotoxicity presents as a stroke mimic: hemiparesis, facial droop, altered consciousness or seizure, with restricted diffusion in deep white matter on MRI. The imaging resemblance to infarction is the clinical danger, because it invites thrombolysis for a lesion that is not thrombotic. The mechanism is not settled; the entry records the adenosine account as a hypothesis rather than as the explanation.

oligodendrocyte CL:0000128

Downstream

Leukoencephalopathy

Encephalopathy

Seizure

Show evidence (2 references)

PMID:36582751 SUPPORT Human Clinical

"MTX has been linked to adverse neurologic effects that mimic acute stroke, including facial drooping, hemiplegia, impaired consciousness, and seizures, as well as changes on imaging-known as MTX-induced leukoencephalopathy (LE)."

Describes the clinical and radiological syndrome this node represents, including the stroke-mimicry that makes it a diagnostic trap.

PMID:36582751 SUPPORT Human Clinical

"Six months afterward, he was rechallenged with MTX and developed recurrence of symptoms."

A within-patient rechallenge recurrence, which is the strongest attribution available from a single case that the drug and not a coincident event produced the lesion.

Renal Tubular Methotrexate Crystallisation

This is the arm that has nothing to do with folate. Methotrexate and its hepatic metabolite 7-hydroxymethotrexate are weak acids and are poorly soluble when the urine is acid; concentrated in the tubular lumen by water reabsorption, they come out of solution. The entire supportive-care ritual around high-dose methotrexate exists to prevent this one physical event.

Downstream

Tubular Obstruction by Methotrexate Crystals

Direct Tubular Epithelial Toxicity

Afferent Arteriolar Vasoconstriction The source names the vasoconstriction as a consequence of intrarenal crystal formation but does not say by what route, so the intermediates are recorded as unknown.

Show evidence (2 references)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Because methotrexate is acidic, drug crystals are not present in urine with an alkaline pH, as alkalinization greatly increases methotrexate solubility and excretion."

States the pH dependence that makes this a solubility event rather than a pharmacological one, and therefore one that alkalinisation can prevent.

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Precipitation of methotrexate crystals occurs in acidic urine (pH < 5.5) when the concentration of methotrexate in the renal tubules exceeds 2 × 10−3 molar."

Gives the two quantitative conditions, pH and luminal concentration, under which precipitation occurs.

Tubular Obstruction by Methotrexate Crystals

The mechanical limb. Precipitated crystals block the tubular lumen. This is one of three insults the source states can each worsen the injury on its own, which is why they are modelled as separate nodes rather than one.

Downstream

Acute kidney injury

Impaired Renal Methotrexate Elimination

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Intrarenal crystal formation can lead to tubule obstruction, direct toxic damage to the renal tubular epithelium (due to prolonged contact with methotrexate), and hypoperfusion from afferent arteriolar vasoconstriction, each of which independently can worsen AKI"

Names tubule obstruction as the first of three independently sufficient injury routes.

Direct Tubular Epithelial Toxicity

The cytotoxic limb. Prolonged contact between the tubular epithelium and concentrated luminal methotrexate poisons the cells directly, independently of any obstruction. This is the only one of the three limbs with a cell type to bind.

renal tubular epithelial cell CL:1000507

Downstream

Acute kidney injury

Impaired Renal Methotrexate Elimination

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

Names direct toxic damage to the tubular epithelium, attributed to prolonged contact, as an independently sufficient injury route.

Afferent Arteriolar Vasoconstriction

The haemodynamic limb, and the one that is not a tubular event at all: constriction of the afferent arteriole drops glomerular perfusion. It carries no cell type because the source names a vessel response rather than a cell population.

Downstream

Acute kidney injury

Impaired Renal Methotrexate Elimination

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

Names hypoperfusion from afferent arteriolar vasoconstriction as the third independently sufficient injury route.

Impaired Renal Methotrexate Elimination

The hinge of the whole entry. Because over 90% of the drug leaves by the kidney, injuring the kidney raises the concentration of the agent that injured it. Higher concentration means more crystallisation and a longer intracellular exposure everywhere else, so the renal arm does not merely add a phenotype: it sets the severity and duration of the folate arm. Clinically this loop is named delayed methotrexate elimination, and it is what turns a routine infusion into a fatal one.

Downstream

Renal Tubular Methotrexate Crystallisation The returning limb of the vicious cycle. A rising concentration in a kidney that can no longer excrete it precipitates further, which is why this edge closes a loop rather than continuing a chain.

Bone Marrow Hematopoietic Failure Amplification rather than initiation. The marrow lesion is produced by the folate arm; delayed elimination decides how deep and how prolonged it is.

Gastrointestinal Mucosal Barrier Breakdown

Hepatocellular Injury

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Renal toxicity leads to impaired methotrexate clearance and prolonged exposure to toxic concentrations, which further worsen renal function and exacerbate nonrenal adverse events"

States the feedback explicitly, in both directions: impaired clearance worsens the kidney further and amplifies the non-renal toxicities.

⬡

Pathograph

Use the checkboxes to hide or show graph categories. Hover nodes for evidence and cross-linked metadata.

●

Phenotypes

Blood 4

Pancytopenia FREQUENT HP:0001876

Show evidence (1 reference)

PMID:25172240 SUPPORT PRIMARY RESULT Human Clinical

"Pancytopenia was the most common manifestation of low dose MTX toxicity detected in 78.5% of the patients."

Quantifies both the association and the frequency band in the series' own hospitalised low-dose cohort.

Decreased total neutrophil count HP:0001875

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"When elimination is delayed because of third spacing and fluid accumulations or as a result of renal injury, neutropenia and thrombocytopenia may be severe"

Reports neutropenia as a consequence of delayed elimination, and notes the conditions under which it becomes severe.

Thrombocytopenia HP:0001873

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"When elimination is delayed because of third spacing and fluid accumulations or as a result of renal injury, neutropenia and thrombocytopenia may be severe"

Reports thrombocytopenia in the same clause and under the same precipitating conditions.

Bone marrow hypocellularity HP:0005528

Digestive 3

Diarrhea HP:0002014

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Volume depletion is perhaps the most important and can result from fluid losses due to vomiting or diarrhea, adrenal insufficiency, or renal salt wasting"

Cited for the second claim in this description: diarrhoeal fluid loss is named as a route to the volume depletion that raises renal risk.

Nausea OCCASIONAL Nausea and vomiting HP:0002017

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Emesis occurs in 10%–30% of patients receiving HDMTX even when appropriate antiemetics are used; in this subgroup, antiemetics should be escalated to completely control vomiting and additional hydration provided to replace lost fluid"

Gives the frequency band and states the reason it is managed aggressively, namely replacement of the lost fluid.

Hepatic fibrosis HP:0001395

Course: PROGRESSIVE

Show evidence (1 reference)

PMID:36564450 SUPPORT REVIEW SYNTHESIS Human Clinical

"Liver fibrosis and the mechanisms of fibrogenesis also need to be considered in relation to chronic exposure to methotrexate."

Associates fibrosis specifically with chronic rather than acute exposure, which is the distinction this phenotype records.

Genitourinary 1

Acute kidney injury HP:0001919

Show evidence (2 references)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"it can cause significant toxicity, including acute kidney injury (AKI) in 2%-12% of patients"

Gives the incidence range across high-dose methotrexate patients. No frequency band is recorded because 2%-12% straddles two of them.

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Crystal-induced nephropathy initially manifests as an asymptomatic elevation in serum creatinine and then progresses to tubular necrosis and more severe renal injury."

Describes the clinical course from silent creatinine rise to tubular necrosis, which is the progression claimed here.

Head and Neck 1

Oral ulcer HP:0000155

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Oral mucositis can become a dose-limiting toxicity, require the use of opioids, increase infectious risk, and lead to chemotherapy delays."

Establishes oral mucositis as a dose-limiting clinical problem and links it to the infectious risk that makes it more than a local lesion.

Immune 1

Sepsis HP:0100806

Show evidence (1 reference)

PMID:25172240 SUPPORT PRIMARY RESULT Human Clinical

"Seven (25%) patients died, all from pancytopenia followed by sepsis."

Establishes sepsis as the terminal event in every death in the series, and the sequence leading to it.

Integument 1

Mucocutaneous toxicity Abnormality of the skin HP:0000951

Coarse binding: variable spectrum

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"exacerbate nonrenal adverse events, including myelosuppression, mucositis, dermatologic toxicity, and hepatotoxicity"

Names dermatologic toxicity alongside the marrow, mucosal and hepatic arms this entry already models, which is why skin belongs beside them.

Metabolism 1

Elevated circulating hepatic transaminase concentration VERY_FREQUENT HP:0002910

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"almost all patients have elevations of serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) values after HDMTX, but these laboratory findings have no clinical significance"

Supports both halves of the description: the near-universal frequency after high-dose exposure, and its lack of clinical significance there.

Nervous System 3

Leukoencephalopathy HP:0002352

Show evidence (1 reference)

PMID:36582751 SUPPORT Human Clinical

"Brain magnetic resonance imaging (MRI) revealed bilateral deep white matter T2 hyperintense foci, increased on the right, with associated diffusion restriction in the right centrum semiovale-consistent with MTX-induced LE."

Documents the specific imaging findings, including the diffusion restriction that produces the resemblance to infarction.

Encephalopathy OCCASIONAL HP:0001298

Temporal: TRANSIENT

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"up to 11% of patients may have CNS events, including confusion, seizures, somnolence, and headaches with or without radiographic evidence of leukoencephalopathy"

Gives the upper bound on frequency and enumerates the range of presentations grouped under this phenotype.

Seizure HP:0001250

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"up to 11% of patients may have CNS events, including confusion, seizures, somnolence, and headaches with or without radiographic evidence of leukoencephalopathy"

Lists seizures among the central nervous system events attributed to the drug, with or without imaging change.

Respiratory 1

Interstitial pneumonitis HP:0006515

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Pulmonary toxicity is observed in 0.5% of patients per year who receive weekly low-dose methotrexate, but rarely with HDMTX"

Supports both the frequency and the inverted dose relationship that is the reason this phenotype is left unattached to the pathograph.

🧬

Genetic Associations

SLCO1B1

Gene: SLCO1B1 hgnc:10959 relationship_type: SUSCEPTIBILITY

Show evidence (2 references)

PMID:32961024 SUPPORT PRIMARY RESULT Human Clinical

"The addition of the SLCO1B1 variants individually further improved the model"

The study's own pharmacokinetic modelling result: SLCO1B1 genotype adds explanatory power for clearance variability beyond clinical covariates.

PMID:32961024 SUPPORT BACKGROUND Human Clinical

"Common function-altering polymorphisms in SLCO1B1 (encodes OATP1B1, which transports MTX) may contribute to clearance variability."

Identifies the gene product and its relation to methotrexate. Quoted from the paper's framing rather than from its results, hence BACKGROUND.

MTHFR

Gene: MTHFR hgnc:7436 relationship_type: SUSCEPTIBILITY

Show evidence (2 references)

PMID:33069634 SUPPORT PRIMARY RESULT Human Clinical

"The patients carrying the homozygous variant genotype had a higher risk of developing nephrotoxicity than those carrying the wild-type genotype (odds ratio [OR] 13.08; 95% confidence interval [CI], 1.65-103.86; P = .002)"

The study's own effect estimate for nephrotoxicity by genotype. The very wide confidence interval is why the note calls the association unsettled rather than established.

PMID:33069634 SUPPORT PRIMARY RESULT Human Clinical

"Significant differences were observed in hepatotoxicity (OR 9.33; 95% CI, 2.54-34.27; P < .001) and hematologic toxicity (OR 3.09; 95% CI, 1.18-8.07; P = .024) in addition to nephrotoxicity between the homozygous variant genotype and the wild-type genotype."

Extends the genotype association to the hepatic and haematologic toxicities, which is what makes this a modifier of the whole toxidrome rather than of one organ.

💊

Medical Actions

Hyperhydration and Urinary Alkalinisation

Action: fluid therapy with urinary alkalinisation Ontology label: Fluid Therapy NCIT:C116537

Platform: Other

The first-line intervention and the only one that prevents rather than rescues. High urine flow dilutes the drug in the tubular lumen and bicarbonate raises urinary pH above 7, which together keep methotrexate and 7-hydroxymethotrexate in solution. The effect size is physical and large: a pH rise from 6.0 to 7.0 increases solubility five- to eightfold. Because it acts on the renal arm only, it does nothing about the folate block and is always given alongside leucovorin rather than instead of it.

Mechanism Target:

INHIBITS Renal Tubular Methotrexate Crystallisation — Acts directly on the precipitation step by changing the two variables it depends on, luminal concentration and pH.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"An increase in urine pH from 6.0 to 7.0 increases the solubility of methotrexate and its metabolites by five- to eightfold"

Quantifies the solubility change alkalinisation produces, which is the mechanism by which this treatment inhibits the crystallisation node.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"The use of fluids to promote high urinary flow rates and alkalinize the urine protects the kidney from injury during treatment with HDMTX"

States the protective effect of the combined hydration and alkalinisation strategy on the kidney.

Leucovorin Rescue

Action: Pharmacotherapy NCIT:C15986

Agent: leucovorin NCIT:C71631

Platform: Small molecule

Leucovorin is a reduced folate, which means it enters the pathway downstream of the enzyme methotrexate is blocking. It does not displace the drug or repair the enzyme; it makes the blocked enzyme unnecessary. That is a genuine bypass rather than an antidote, and it has two consequences. It must not start too early, because rescuing the tumour is as easy as rescuing the marrow. And it competes with methotrexate for the same reduced folate carrier, so when the plasma methotrexate concentration is very high the dose must rise steeply to win that competition.

Mechanism Target:

BYPASSES Dihydrofolate Reductase Inhibition — Supplies reduced folate downstream of the blocked reductase, so the pathway runs while the enzyme remains inhibited.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS In Vitro

"LV enters cells through the RFC and allows formation of FH4 despite the presence of MTX, which effectively rescues cells."

States the bypass explicitly: tetrahydrofolate is formed despite the drug still being present, which is what distinguishes this from inhibition or displacement.

Show evidence (2 references)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Leucovorin is particularly effective in the prevention of myelosuppression, gastrointestinal toxicity, and neurotoxicity during treatment with HDMTX."

Names the three toxicities leucovorin rescue prevents, which correspond to the three tissue arms downstream of the folate block.

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"leucovorin must compete with MTX to enter cells via the reduced folate carrier"

Supports the dose-escalation claim: because rescue and drug share one transporter, the rescue dose is set by the methotrexate concentration it must outcompete.

Glucarpidase

Action: Pharmacotherapy NCIT:C15986

Agent: glucarpidase NCIT:C346

Platform: Other

A recombinant bacterial carboxypeptidase that cuts methotrexate in the bloodstream into two inactive fragments. It is fast and near-complete: a single dose removes 97% or more of circulating drug within fifteen minutes. Its limitation is the one that matters mechanistically, and it is not a dosing detail. The enzyme stays in the plasma, so it does nothing to the methotrexate already inside cells. It removes the cause of further accumulation and leaves the existing lesion untouched, which is precisely why it is always given with leucovorin rather than in place of it.

Mechanism Target:

INHIBITS Systemic Methotrexate Exposure — Destroys the circulating pool enzymatically, acting on the exposure node itself rather than on any downstream mechanism.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Glucarpidase cleaves methotrexate into DAMPA and glutamate, two nontoxic metabolites, and thus provides an enzymatic method to rapidly remove methotrexate in patients with renal dysfunction"

States the reaction and its product, and that its purpose is removal of circulating drug when the renal route has failed.

Show evidence (2 references)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"A single dose of glucarpidase (50 U/kg i.v. over 5 minutes) reduces plasma methotrexate concentrations by 97% or more within 15 minutes"

Quantifies the magnitude and speed of the plasma reduction claimed in the description.

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"despite the decrease in the magnitude and duration of systemic exposure to methotrexate after glucarpidase, it has no effect on intracellular methotrexate concentrations"

States the compartmental limitation that makes concurrent leucovorin mandatory rather than optional.

Drainage of Third-Space Fluid Collections

Action: drainage of a third-space fluid collection Ontology label: Surgical Procedure NCIT:C15329

Platform: Surgery

Draining a pleural effusion or ascites before the infusion, or deferring the infusion until it resolves, removes the reservoir. It is listed alongside avoiding interacting medications as one of only two named strategies for preventing myelosuppression, which places it well above its usual billing as a procedural afterthought.

Mechanism Target:

INHIBITS Impaired Renal Methotrexate Elimination — Removes the compartment that sustains the systemic concentration after the infusion ends.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"draining effusions before treatment (or delaying HDMTX until effusions resolve)"

Names drainage, or deferral until resolution, as the intervention directed at this mechanism.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"The only known strategies to prevent myelosuppression are to prevent delayed methotrexate elimination by avoiding interacting medications around the time of infusion and draining effusions before treatment (or delaying HDMTX until effusions resolve) and to ensure optimal leucovorin dosing."

Places drainage among the only named strategies for preventing myelosuppression, which supports the description's claim about its standing.

Haemodialysis

Action: Hemodialysis NCIT:C15248

Platform: Other

Used when methotrexate remains high despite hydration, alkalinisation and rescue. The evidence is weak by the standards of the rest of this entry, and one specific failure mode is documented: plasma concentrations rebound substantially after the procedure as drug redistributes out of tissue, so a single session is rarely sufficient.

Mechanism Target:

INHIBITS Systemic Methotrexate Exposure — Removes circulating drug extracorporeally when renal clearance has failed.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"even when hemodialysis is effective, many patients experience a rebound in serum methotrexate concentrations of 10%–220% of the postprocedure values"

Documents the rebound that limits the procedure's usefulness, with the reported range.

🌍

Environmental Factors

Therapeutic methotrexate administration

The exposure is the treatment. That is what separates this entry from a poisoning: nobody encounters methotrexate by accident, and the dose that injures is usually the dose that was prescribed, delivered to a patient whose clearance was not what the protocol assumed.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Although HDMTX is safely administered to most patients, it can cause significant toxicity"

Frames the exposure correctly: the same administration is tolerated by most patients and injures a minority, which is why the entry's risk factors do most of the work.

Mechanism Target:

TRIGGERS Systemic Methotrexate Exposure — Administration establishes the systemic pool from which both arms of the mechanism follow.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"High-dose methotrexate (HDMTX), defined as a dose higher than 500 mg/m2, is used to treat a range of adult and childhood cancers."

Establishes the therapeutic administration that constitutes the exposure, and the dose threshold at which the high-dose toxicity profile applies.

Co-administration of non-steroidal anti-inflammatory drugs

Scoped to NSAIDs rather than to the whole interacting-drug list because that is where both the mechanistic and the contradicting evidence sit, and a binding should not name a wider exposure than its sources establish. ECTO was searched on 2026-09-22: `runoak -i sqlite:obo:ecto info "l~NSAID"` returns ECTO:2000011 exposure to NSAIDs, defined as an exposure to NSAID therapy and descending from ECTO:2000057 exposure to pharmacotherapy, which is the co-medication sense this entry means; `l~anti-inflammatory` also returns ECTO:9001720 exposure to non-steroidal anti-inflammatory drug, the substance-exposure counterpart, which was not chosen because the entry describes a drug given as therapy alongside methotrexate rather than an environmental encounter with the compound. The other agents named in the description (omeprazole, piperacillin/tazobactam, ciprofloxacin, cotrimoxazole, probenecid, imatinib) are recorded in prose; no ECTO class covers that heterogeneous set, and inventing one binding for all of them would assert a grouping the ontology does not make.

Weak organic acids and transporter substrates given alongside methotrexate can slow its renal elimination and push a patient into the delayed-elimination loop. The size of this effect is genuinely disputed. A transporter-focused review finds the best-documented interactions to be with indomethacin, ketoprofen, omeprazole, piperacillin/tazobactam, ciprofloxacin, cotrimoxazole, probenecid and imatinib, while explicitly declining to endorse a firm NSAID or proton-pump-inhibitor interaction on the available evidence. Standard high-dose protocols nonetheless reconcile and substitute these drugs before infusion, on the reasoning that the cost of avoiding them is small and the cost of being wrong is not.

Show evidence (1 reference)

PMID:36946211 SUPPORT REVIEW SYNTHESIS Human Clinical

"most interaction records with methotrexate occurred with coadministration of indomethacin, ketoprofen, omeprazole, piperacillin/tazobactam, ciprofloxacin, cotrimoxazole, probenecid, and imatinib, mainly due to the role of transporters"

Names the agents with the most interaction records and attributes them to transporter competition. Only indomethacin and ketoprofen fall inside this exposure's NSAID binding; the rest are recorded in the description and sit outside it.

Mechanism Target:

EXACERBATES Impaired Renal Methotrexate Elimination — The proposed route is competition at renal tubular transporters, so the intermediates are named but the clinical effect size is contested.

Show evidence (2 references)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Model Organism

"Studies in mice documented inhibition of renal tubule methotrexate transporters by indomethacin and ketoprofen, with reduced methotrexate elimination and prolonged elevated levels and toxicity"

The mechanistic demonstration behind the interaction, in mice. Graded MODEL_ORGANISM because that is the evidence the quoted sentence describes, and REVIEW_SYNTHESIS because this review is relaying it rather than reporting its own experiment.

PMID:36946211 REFUTE REVIEW SYNTHESIS Human Clinical

"Our findings showed no firm evidence of interactions of proton pump inhibitors (PPIs), levetiracetam, and NSAIDS with MTX."

Directly contradicts the clinical interaction claim for the two drug classes most often named. Recorded as REFUTE rather than dropped, because the disagreement with the preceding item is the honest state of this edge.

Volume depletion and acidic urine during high-dose infusion

No exposure term was bound. ECTO was searched on 2026-09-22 with `runoak -i sqlite:obo:ecto info` over `l~dehydration` and `l~water deprivation`, both of which return nothing at all; `l~fluid`, which returns only exposures to named liquids such as ECTO:0500001 exposure to brake fluid and ECTO:7000109 exposure to bodily fluid; and `l~acid`, which returns CDNO concentration terms and chemical-acid classes, not a urinary-pH state. None of these describes volume depletion or urine acidity in a patient, so the slot carries a free-text `preferred_term` only.

Two physical conditions, both modifiable, that decide whether the filtered drug stays in solution. They are listed as an exposure rather than a phenotype because the supportive-care protocol treats them as the thing to be controlled, and controlling them is the single most effective intervention in the entry.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Because volume depletion and acidic urine are major risk factors for AKI, hyperhydration and urine alkalinization are mandatory during HDMTX treatment"

Supports the claim that these two conditions are the modifiable risk factors the supportive-care protocol targets.

Mechanism Target:

TRIGGERS Renal Tubular Methotrexate Crystallisation — Low urine flow concentrates the drug in the tubular lumen and low pH reduces its solubility, which are the two conditions precipitation requires.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Because volume depletion and acidic urine are major risk factors for AKI, hyperhydration and urine alkalinization are mandatory during HDMTX treatment"

Names both conditions as the major risk factors and states that the protocol exists to correct them.

Third-space fluid sequestration

No exposure term was bound. This is a patient's own fluid compartment rather than an external exposure, and ECTO was searched on 2026-09-22 with `runoak -i sqlite:obo:ecto info "l~fluid"`, which returns only exposures to named external liquids (ECTO:0500001 exposure to brake fluid, ECTO:7000109 exposure to bodily fluid and similar), and `l~effusion` and `l~ascites`, which return nothing. The slot carries a free-text `preferred_term` only.

A pleural effusion or ascites acts as a reservoir. Methotrexate partitions into the collection while the plasma concentration is high and leaks back afterwards, so elimination looks complete and then is not. It is the one route to delayed elimination that does not involve the kidney at all, which is why it has its own countermeasure: drain the collection, or defer the infusion until it resolves.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Delayed methotrexate excretion has been associated with extravascular fluid collections, including ascites, pleural effusions, or intracranial fluid"

Names the specific collections that behave as reservoirs, which is the entry-level claim this exposure makes.

Mechanism Target:

EXACERBATES Impaired Renal Methotrexate Elimination — Redistribution out of the collection sustains the systemic concentration after the infusion has finished, which is delayed elimination by a non-renal route.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"When elimination is delayed because of third spacing and fluid accumulations or as a result of renal injury, neutropenia and thrombocytopenia may be severe"

Names third spacing as a cause of delayed elimination distinct from renal injury, which is exactly the claim this link makes.

Folic acid supplementation during chronic low-dose therapy

The link is drawn to the tissue-injury nodes rather than to Dihydrofolate Reductase Inhibition, which was the first draft. The trials measure clinical adverse events, not enzyme occupancy, so an edge to the molecular node would attach a clinical outcome to a biochemical claim the source does not make.

Replacing folate alongside weekly methotrexate reduces the gastrointestinal and hepatic adverse effects without abolishing the therapeutic effect. It is the clearest protective exposure in the entry, and the protection is selective in an informative way: pooled trial data show a clear reduction in gastrointestinal effects and a large one in transaminase elevation, while the reduction in stomatitis does not reach significance and the haematologic outcome could not be assessed at all.

Show evidence (2 references)

PMID:23728635 SUPPORT PRIMARY RESULT Human Clinical

"The results support a protective effect of supplementation with either folic or folinic acid for patients with rheumatoid arthritis during treatment with MTX."

The review's own overall conclusion, which is the entry-level claim this exposure makes.

PMID:23728635 SUPPORT PRIMARY RESULT Human Clinical

"It does not appear that supplementation with either folic or folinic acid has a statistically significant effect on the efficacy of MTX in treating RA"

Supports the second half of the description: the protection does not come at the cost of the therapeutic effect, which is what makes supplementation routine rather than a trade-off.

Mechanism Target:

PROTECTS_AGAINST Hepatocellular Injury — The largest and best-evidenced protective effect: supplementation cuts the incidence of methotrexate-associated transaminase elevation by roughly three quarters in relative terms.

Show evidence (1 reference)

PMID:23728635 SUPPORT PRIMARY RESULT Human Clinical

"Folic and folinic acid also appear to be protective against abnormal serum transaminase elevation caused by MTX, with a 76.9% relative (16% absolute) risk reduction (RR 0.23, 95% CI 0.15 to 0.34; P < 0.00001)"

Quantifies the protective effect on the hepatic arm specifically, from a pooled analysis of randomised placebo-controlled trials.

PROTECTS_AGAINST Gastrointestinal Mucosal Barrier Breakdown — Smaller than the hepatic effect and, notably, driven by the general gastrointestinal symptom outcome rather than by stomatitis, which did not separate from placebo.

Show evidence (2 references)

PMID:23728635 SUPPORT PRIMARY RESULT Human Clinical

"a 26% relative (9% absolute) risk reduction was seen for the incidence of GI side effects such as nausea, vomiting or abdominal pain (RR 0.74, 95% CI 0.59 to 0.92; P = 0.008)"

Gives the effect size for the gastrointestinal arm, which is the magnitude this link claims.

PMID:23728635 REFUTE PRIMARY RESULT Human Clinical

"whilst showing a trend towards reduction in risk, the results were not statistically significant (RR 0.72, 95% CI 0.49 to 1.06)"

The stomatitis outcome specifically did not reach significance, which qualifies rather than supports a blanket protective claim over the whole mucosal node. Recorded as REFUTE against that broader reading.

🔬

Diagnosis

Serial plasma methotrexate concentration

The central monitoring test of high-dose infusion, and the one that decides management there. Levels are drawn at protocol-defined intervals after the infusion starts and read against a nomogram; a level above the expected curve triggers leucovorin escalation and, if high enough, glucarpidase. Two limits are worth stating rather than assuming. It tracks the renal arm well and the other toxicities poorly. And it does not transfer to the low-dose setting at all: in a series of hospitalised low-dose toxicity, serum concentrations correlated with neither the neutropenia nor the thrombocytopenia, and the authors conclude there is no rationale for monitoring there.

Show evidence (2 references)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Plasma methotrexate monitoring is a reliable indicator specifically of nephrotoxicity but may be a limited predictor of other toxicities"

States exactly the scope claim made here: the assay tracks the renal arm well and the rest of the toxidrome less well.

PMID:25172240 REFUTE PRIMARY RESULT Human Clinical

"There is no rationale for MTX therapeutic drug monitoring in the setting of low-dose toxicity."

Directly contradicts any reading of this node as a test for methotrexate toxicity in general. Recorded as REFUTE against that broader claim, and the description is scoped to high-dose infusion accordingly.

Urine pH monitoring during infusion

Checked at every void, because a short period of acid urine is enough to start precipitation. A falling pH is acted on with bolus bicarbonate rather than observed.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"It is also important to check urine pH values with each void during the infusion to ensure no extended periods of time with acidic urine, which could increase the risk for precipitation, nephrotoxicity, and delayed methotrexate elimination."

States the monitoring frequency and the three consequences it is intended to prevent.

Serum creatinine and urine output

Standard renal monitoring, with a documented limitation: creatinine lags the injury, so tubular damage may already be irreversible by the time it rises. Falling urine output or a positive fluid balance identifies early injury sooner.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"the rise in serum creatinine values lags behind progressive intrinsic renal damage, such that precise measurement of function at a specific moment is difficult"

Supports the stated limitation of creatinine as a real-time measure of the injury this entry cares about.

📈

Progression

High-dose infusion and immediate post-infusion period

Hydration and alkalinisation are running, levels are drawn on protocol, and the question being asked is only whether elimination is on the expected curve. Nothing is yet clinically apparent.

Delayed elimination

A level above the nomogram, a rising creatinine, or a falling urine output. This is the decision point of the whole entry: intervention here is leucovorin escalation and possibly glucarpidase, and the alternative is the feedback loop closing.

Established organ toxicity

Mucositis and the cytopenia nadir overlap, and that overlap is when sepsis risk peaks. Specific day ranges are deliberately not given: the figures in circulation for them are not stated by any reference this entry cites.

Recovery

Usually complete once the drug is cleared and the proliferative compartments repopulate, and subsequent high-dose courses can generally be given safely after recovery even when the episode was severe.

Show evidence (1 reference)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"However, even when toxicity is severe, subsequent HDMTX courses can generally be administered safely after the patient recovers"

Supports the claim that a severe episode does not normally preclude further treatment, which is the prognostic point of this phase.

⚖️

Clinical Burden

Variable

The burden is concentrated in a minority of exposures and is almost entirely preventable, which is what makes it a supportive-care problem rather than a pharmacological one. Acute kidney injury complicates 2% to 12% of high-dose courses. The deaths are mostly not renal: they follow from the cytopenia and the open mucosal barrier arriving together, so the proximate cause is usually sepsis. Beyond the acute episode, an interrupted or delayed chemotherapy course is itself a harm, because the regimens methotrexate belongs to are curative ones.

Show evidence (2 references)

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"AKI and other toxicities of high-dose methotrexate can lead to significant morbidity, treatment delays, and diminished renal function."

Names the three burdens claimed: acute morbidity, interruption of the cancer treatment, and residual loss of renal function.

PMID:27496039 SUPPORT REVIEW SYNTHESIS Human Clinical

"Grade IV mucositis is an oncologic emergency and is associated with infections, the need for parenteral nutrition, increased use of health care resources, delayed chemotherapy, and even death"

Supports treating the mucosal lesion as a driver of mortality rather than a comfort issue.

🐁

Animal Models

Methotrexate-induced small intestinal mucositis in the rat

Repeated intraperitoneal methotrexate in rats produces small intestinal mucosal injury with loss of crypt proliferation and disrupted tight-junction protein expression, and is a standard preparation for testing protective agents against the gastrointestinal arm.

Species

Rat

Publication

PMID:36996098

Pathograph links

Gastrointestinal Mucosal Barrier Breakdown recapitulates fidelity: moderate observes: tissue Produces the crypt and villus lesion and the tight-junction changes that define the human node.

Limitations: Dosed intraperitoneally on a three-day cycle rather than by the human routes and schedules, and studied without leucovorin rescue, so it models the unrescued lesion rather than clinical mucositis under supportive care.

Readouts

Small intestinal mucosal injury on histopathology

Direction: INCREASED

Interpretation: Histological confirmation that the preparation produces the mucosal lesion it is used to model.

Show evidence (1 reference)

PMID:36996098 SUPPORT PRIMARY RESULT Model Organism

"The pathology results showed that small intestinal mucosal injury in the DKT-MTX group was less severe than that in the MTX group."

Reports the histological comparison. The methotrexate-only arm is the more injured one, which is the measurement this readout records.

Crypt Ki-67-positive proliferating cell count

Direction: DECREASED

Interpretation: A direct measure of the crypt proliferative compartment, which is the cell population the folate block kills.

Show evidence (1 reference)

PMID:36996098 SUPPORT PRIMARY RESULT Model Organism

"The crypts in the DKT-MTX group contained more Ki-67-positive cells than MTX group."

The methotrexate-only arm has fewer proliferating crypt cells, which is the loss of crypt proliferation this readout measures.

Show evidence (1 reference)

PMID:36996098 SUPPORT PRIMARY RESULT Model Organism

"Intraperitoneal injection of 10 mg/kg methotrexate (MTX) every 3 days for a total of 3 doses was used for induction of CIM in a rat model."

Documents the induction protocol, which is what establishes this as a methotrexate model of the mucosal node rather than a general mucositis model.

Show evidence (1 reference)

PMID:36996098 SUPPORT PRIMARY RESULT Model Organism

"Intraperitoneal injection of 10 mg/kg methotrexate (MTX) every 3 days for a total of 3 doses was used for induction of CIM in a rat model."

Documents the induction protocol that defines this preparation as a methotrexate model.

{ }

Source YAML

click to show

name: Methotrexate Toxicity
creation_date: '2026-09-22T02:15:00Z'
description: >-
  Methotrexate is a counterfeit folate. It is close enough to the real vitamin
  that the cell's own folate importer carries it inside and the cell's own
  ligase chains glutamates onto it, which traps it there; then it jams the
  enzyme that regenerates tetrahydrofolate, and the cell can no longer make
  thymidine or purines. Every cell the drug enters gets the same lesion, so
  which tissue is injured is not a question about the lesion. It is a question
  about which tissues are replacing themselves right now: marrow, gut
  epithelium, liver.

  What makes the poisoning distinctive is that the second half of the mechanism
  is not pharmacodynamic at all. Over 90% of the drug leaves by the kidney, and
  it is poorly soluble in acid urine, so at high dose it crystallises in the
  tubules and injures them. An injured kidney clears less methotrexate, a higher
  methotrexate concentration crystallises more, and the loop tightens on itself.
  That is why the two halves of the entry are asymmetric: the folate arm decides
  what the injury looks like, and the renal arm decides how bad and how long it
  gets. It is also why the countermeasures split the same way. Hydration and
  urine alkalinisation act on the renal arm and leucovorin on the folate arm,
  while glucarpidase and dialysis act upstream of both, on the circulating
  pool -- which is exactly why neither of those two touches the drug already
  inside cells.
categories:
- Treatment Toxicity
category: Complex
parents:
- Drug Toxicity
synonyms:
- methotrexate intoxication
- methotrexate overdose
- delayed methotrexate elimination
- methotrexate-induced pancytopenia
disease_term:
  preferred_term: methotrexate toxicity
  term:
    id: MONDO:0034212
    label: methotrexate toxicity
references:
- reference: PMID:27496039
  title: Preventing and Managing Toxicities of High-Dose Methotrexate.
- reference: PMID:36071163
  title: Methotrexate recognition by the human reduced folate carrier SLC19A1.
- reference: PMID:12374095
  title: The role of dUTPase and uracil-DNA repair in cancer chemotherapy.
- reference: PMID:25172240
  title: "Clinical characteristics and risk factors for low dose methotrexate toxicity: a cohort of 28 patients."

pathophysiology:
- name: Systemic Methotrexate Exposure
  description: >-
    The initiating event is the presence of methotrexate in the circulation at a
    concentration the patient cannot clear on the expected schedule. This is a
    single node because everything downstream follows from concentration and
    duration rather than from route: the same lesion is reached by a high-dose
    oncology infusion, by a weekly low-dose regimen taken daily in error, and by
    an ordinary dose given to a patient whose kidney cannot excrete it.
  biological_scale: ORGANISM
  genes:
  - preferred_term: SLCO1B1
    term:
      id: hgnc:10959
      label: SLCO1B1
  chemical_entities:
  - preferred_term: methotrexate
    term:
      id: CHEBI:44185
      label: methotrexate
    modifier: INCREASED
  downstream:
  - target: Cellular Methotrexate Uptake by the Reduced Folate Carrier
    causal_link_type: DIRECT
    description: >-
      Circulating drug reaches the intracellular compartment only through
      carrier-mediated import, so the systemic concentration sets the rate at
      which the intracellular lesion is established.
  - target: Cerebral White Matter Injury
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      Central nervous system toxicity is an established consequence of the drug
      independently of any account of how it happens. This edge carries no
      `hypothesis_groups` on purpose: without it the whole neurologic arm would
      hang only on the adenosine hypothesis, which would assert that
      methotrexate neurotoxicity is explicable solely by that route.
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "CNS toxicity may occur after HDMTX"
      explanation: >-
        Establishes the drug-to-central-nervous-system link as a recognised
        consequence, with no mechanism attached, which is exactly what this
        mechanism-free edge claims.
  - target: Renal Tubular Methotrexate Crystallisation
    causal_link_type: DIRECT
    description: >-
      The same circulating pool is delivered to the tubular lumen by glomerular
      filtration and tubular secretion, where its solubility rather than its
      pharmacology decides what happens next.
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "More than 90% of methotrexate is eliminated by the kidneys"
      explanation: >-
        Establishes that the renal route carries essentially the whole systemic
        dose, which is what makes the tubular lumen a site of exposure rather
        than an incidental one.

- name: Cellular Methotrexate Uptake by the Reduced Folate Carrier
  description: >-
    Methotrexate does not cross the membrane freely. It is imported by SLC19A1,
    the reduced folate carrier, which exists to bring physiological folates in
    and cannot distinguish them from the antifolate. Uptake capacity is
    therefore also the drug's selectivity: cells that import folate avidly
    import methotrexate avidly, and loss of carrier function is a recognised
    route to methotrexate resistance.
  biological_scale: MOLECULAR
  genes:
  - preferred_term: SLC19A1
    term:
      id: hgnc:10937
      label: SLC19A1
  molecular_functions:
  - preferred_term: reduced folate carrier antiport activity
    term:
      id: GO:0008518
      label: folate:monoatomic anion antiporter activity
  evidence:
  - reference: PMID:36071163
    reference_title: Methotrexate recognition by the human reduced folate carrier SLC19A1.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: BACKGROUND
    snippet: "The human reduced folate carrier (hRFC, also known as SLC19A1) is the major importer of folates into the cell"
    explanation: >-
      Identifies the transporter that carries methotrexate into the cell.
      Quoted from the paper's opening framing rather than from its own
      structural result, hence BACKGROUND.
  - reference: PMID:36071163
    reference_title: Methotrexate recognition by the human reduced folate carrier SLC19A1.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: BACKGROUND
    snippet: "it is a major determinant in methotrexate (antifolate) sensitivity"
    explanation: >-
      Supports treating uptake as a determinant of the lesion rather than a
      preliminary step, since carrier function sets how sensitive a cell is.
      Also background rather than this paper's own finding: its abstract
      reports a methotrexate-bound structure, but the body states that no
      cryo-EM density for the drug was observed in the central cavity.
  downstream:
  - target: Intracellular Methotrexate Polyglutamation
    causal_link_type: DIRECT
    description: >-
      Import delivers the monoglutamate form to the cytosolic enzyme that
      extends it.

- name: Intracellular Methotrexate Polyglutamation
  description: >-
    Once inside, methotrexate is treated as a folate by folylpolyglutamate
    synthetase, which adds a chain of glutamate residues. The chain is the
    reason toxicity outlives the plasma level: the polyglutamate is too charged
    to leave the cell, so it accumulates and is retained long after the
    extracellular drug has gone. Retention increases sharply with chain length,
    which is why an exposure that looks brief in plasma is not brief inside the
    cell.
  biological_scale: MOLECULAR
  evidence:
  - reference: PMID:2416193
    reference_title: "Glutamylation of methotrexate in hepatoma cells in vitro: regulation and the development of specific inhibitors."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Methotrexate is glutamylated in cultured hepatoma cells to derivatives that contain a total of 2 to 5 gamma-glutamyl residues."
    explanation: >-
      Establishes that the intracellular drug is converted to polyglutamate
      derivatives, and the chain lengths reached.
  - reference: PMID:2416193
    reference_title: "Glutamylation of methotrexate in hepatoma cells in vitro: regulation and the development of specific inhibitors."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "The importance of the longer-chain-length polyglutamates is apparent from the 6-hr retention of the polyglutamate species: Glu2, 15%; Glu3, 21%; Glu4, 50%; and Glu5, 83%."
    explanation: >-
      Quantifies the retention claim directly: the longest chains are still
      largely present at six hours, which is the measured basis for treating
      polyglutamation as the trapping step.
  downstream:
  - target: Dihydrofolate Reductase Inhibition
    causal_link_type: DIRECT
    description: >-
      Polyglutamation both concentrates the drug at its target and raises its
      affinity for the folate-dependent enzymes.

- name: Dihydrofolate Reductase Inhibition
  description: >-
    The central molecular lesion. Methotrexate occupies the folate site of
    dihydrofolate reductase and blocks the reduction of dihydrofolate to
    tetrahydrofolate. The enzyme is not damaged, only occupied, which is exactly
    why a competing reduced folate given from outside can rescue the cell
    without the enzyme ever being repaired.
  biological_scale: MOLECULAR
  genes:
  - preferred_term: DHFR
    term:
      id: hgnc:2861
      label: DHFR
  molecular_functions:
  - preferred_term: dihydrofolate reductase activity
    term:
      id: GO:0004146
      label: dihydrofolate reductase activity
    modifier: DECREASED
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: REVIEW_SYNTHESIS
    snippet: "After entry into the cell, methotrexate is polyglutamated, binds dihydrofolate reductase (DHFR) with an affinity 1,000-fold greater than that of folate, and competitively inhibits conversion of dihydrofolate to tetrahydrofolate"
    explanation: >-
      States the mechanism and the affinity margin over the natural substrate,
      which is what makes the block near-complete at therapeutic concentrations.
  downstream:
  - target: Tetrahydrofolate Depletion
    causal_link_type: DIRECT
    description: >-
      Blocking the reductase removes the only route by which the cell
      regenerates tetrahydrofolate from the dihydrofolate produced each time
      thymidylate is made.

- name: Tetrahydrofolate Depletion
  description: >-
    Tetrahydrofolate is the carrier for every one-carbon transfer the cell
    performs. Depleting it does not disable one pathway but the shared currency
    of several, which is why a single enzyme block produces a synthesis failure
    on two separate nucleotide branches at once.
  biological_scale: MOLECULAR
  genes:
  - preferred_term: MTHFR
    term:
      id: hgnc:7436
      label: MTHFR
  chemical_entities:
  - preferred_term: tetrahydrofolic acid
    term:
      id: CHEBI:26907
      label: tetrahydrofolic acid
    modifier: DECREASED
  biological_processes:
  - preferred_term: tetrahydrofolate metabolic process
    term:
      id: GO:0046653
      label: tetrahydrofolate metabolic process
    modifier: DECREASED
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: REVIEW_SYNTHESIS
    snippet: "Tetrahydrofolate is essential for biosynthesis of thymidine and purines, which are needed for synthesis of DNA."
    explanation: >-
      Names the two downstream branches that tetrahydrofolate supplies, which is
      the reason this node has two outgoing edges rather than one.
  downstream:
  - target: Thymidylate Synthesis Blockade
    causal_link_type: DIRECT
  - target: De Novo Purine Synthesis Blockade
    causal_link_type: DIRECT

- name: Thymidylate Synthesis Blockade
  description: >-
    Thymidylate synthase needs 5,10-methylenetetrahydrofolate as its
    one-carbon donor, so the folate block starves it of cofactor. This is the
    cell's only de novo source of thymidine monophosphate, and there is no
    salvage route that can replace it at the rate a dividing cell consumes it.
  biological_scale: MOLECULAR
  genes:
  - preferred_term: TYMS
    term:
      id: hgnc:12441
      label: TYMS
  biological_processes:
  - preferred_term: dTMP biosynthetic process
    term:
      id: GO:0006231
      label: dTMP biosynthetic process
    modifier: DECREASED
  downstream:
  - target: Uracil Misincorporation and DNA Strand Breakage
    causal_link_type: DIRECT
    description: >-
      Blocking the reaction does not only remove thymidylate; it leaves the
      unconsumed precursor pool to accumulate as dUTP, which is the substrate
      for the next step.
    evidence:
    - reference: PMID:12374095
      reference_title: The role of dUTPase and uracil-DNA repair in cancer chemotherapy.
      supports: SUPPORT
      evidence_source: IN_VITRO
      quote_role: REVIEW_SYNTHESIS
      snippet: "Upon TS inhibition, dUTP pools may accumulate, inducing repeated cycles of uracil misincorporation into DNA and repair-mediated DNA damage."
      explanation: >-
        States the causal step from thymidylate synthase inhibition to uracil
        misincorporation, which is the specific edge this link asserts.

- name: Uracil Misincorporation and DNA Strand Breakage
  description: >-
    With thymidylate scarce and dUTP abundant, the polymerase puts uracil into
    DNA. Uracil-DNA glycosylase excises it, the repair patch is resynthesised
    from the same depleted pool, and uracil goes back in. The futile cycle is
    what converts a shortage of one nucleotide into physical strand breakage.
    It is the uracil-DNA arm of what is classically called thymineless death,
    alongside the thymidylate pool depletion that the term originally named.
  biological_scale: MOLECULAR
  evidence:
  - reference: PMID:12374095
    reference_title: The role of dUTPase and uracil-DNA repair in cancer chemotherapy.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: REVIEW_SYNTHESIS
    snippet: "cytotoxicity results from a process known as \"thymineless death\""
    explanation: >-
      Names the cytotoxic mechanism attributed to thymidylate synthase
      inhibition in this branch.
  downstream:
  - target: Replicating Cell Death
    causal_link_type: DIRECT

- name: De Novo Purine Synthesis Blockade
  description: >-
    Two steps of purine ring assembly, catalysed by GART and by AICAR
    transformylase, are formyl-tetrahydrofolate-dependent and stall for the same
    reason thymidylate synthesis does. The branch matters twice over: it
    contributes to the synthesis failure that kills dividing cells, and the
    substrate that accumulates behind the ATIC block is the starting point of
    the adenosine hypothesis of neurotoxicity.
  biological_scale: MOLECULAR
  genes:
  - preferred_term: ATIC
    term:
      id: hgnc:794
      label: ATIC
  biological_processes:
  - preferred_term: de novo purine nucleotide synthesis
    term:
      id: GO:0006189
      label: 'de novo'' IMP biosynthetic process'
    modifier: DECREASED
  evidence:
  - reference: PMID:32066940
    reference_title: Methotrexate and its mechanisms of action in inflammatory arthritis.
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "multiple mechanisms potentially contribute to the anti-inflammatory actions of methotrexate, including the inhibition of purine and pyrimidine synthesis"
    explanation: >-
      Confirms that purine synthesis inhibition is an established action of the
      drug, cited here for the pathway and not for the anti-inflammatory
      conclusion the sentence is making.
  downstream:
  - target: Replicating Cell Death
    causal_link_type: DIRECT
  - target: Adenosine Accumulation
    causal_link_type: INDIRECT_KNOWN_INTERMEDIATES
    hypothesis_groups:
    - adenosine_neurotoxicity
    description: >-
      Substrate accumulating behind the AICAR transformylase block is the
      proposed source of the adenosine surge. The intermediates are known and
      named, but the step is part of a hypothesis rather than a settled
      mechanism in this disease.

- name: Adenosine Accumulation
  description: >-
    Substrate accumulating behind the ATIC block is followed by a rise in
    extracellular adenosine; the intervening enzymology is usually given as
    inhibition of adenosine and AMP deaminases, which none of the sources cited
    here states, so it is named as the usual account rather than asserted. In
    rheumatoid arthritis this is the favoured explanation of why a weekly
    low dose is anti-inflammatory. In the toxicity setting the same molecule is
    the leading candidate for acute methotrexate encephalopathy, on the strength
    of raised central adenosine in affected patients and symptom reversal with
    an adenosine antagonist. It is a hypothesis, and the entry treats it as one.
  biological_scale: MOLECULAR
  evidence:
  - reference: PMID:32066940
    reference_title: Methotrexate and its mechanisms of action in inflammatory arthritis.
    supports: SUPPORT
    evidence_source: OTHER
    quote_role: REVIEW_SYNTHESIS
    snippet: "as well as the promotion of adenosine release and expression of certain long non-coding RNAs"
    explanation: >-
      Records adenosine release as one of the drug's recognised downstream
      actions, which is the premise the neurotoxicity hypothesis builds on.
      Quoted from a review of the drug's anti-inflammatory action, so it
      establishes that the drug promotes adenosine release and not that
      adenosine is neurotoxic.
  downstream:
  - target: Cerebral White Matter Injury
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    hypothesis_groups:
    - adenosine_neurotoxicity
    description: >-
      The proposed route from adenosine to white matter injury is vasoactive and
      neuromodulatory rather than a demonstrated chain, so the intermediates are
      recorded as unknown.
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "A potential mechanism of neurotoxicity is the accumulation of adenosine after MTX-induced reductions in purine synthesis"
      explanation: >-
        The source states this as a potential mechanism, which is the epistemic
        status this edge is recording rather than a settled causal claim.

- name: Replicating Cell Death
  description: >-
    The convergence point of the folate arm. A cell that cannot finish
    replication and is accumulating strand breaks dies or arrests, and the
    selectivity of the whole drug is contained in the word replicating: the
    lesion is identical everywhere and the consequence is confined to the
    tissues that are dividing.
  biological_scale: CELLULAR
  biological_processes:
  - preferred_term: DNA replication
    term:
      id: GO:0006260
      label: DNA replication
    modifier: DECREASED
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: IN_VITRO
    quote_role: REVIEW_SYNTHESIS
    snippet: "Blockade of tetrahydrofolate synthesis by methotrexate leads to inability of cells to divide and to produce proteins."
    explanation: >-
      States the cellular consequence of the folate block that this node
      represents.
  downstream:
  - target: Bone Marrow Hematopoietic Failure
    causal_link_type: DIRECT
  - target: Gastrointestinal Mucosal Barrier Breakdown
    causal_link_type: DIRECT
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "Mucositis after HDMTX is caused by cellular damage to rapidly dividing epithelial cells along the entire gastrointestinal tract"
      explanation: >-
        Attributes the mucosal lesion specifically to death of rapidly dividing
        epithelium, which is the edge asserted here.
  - target: Hepatocellular Injury
    causal_link_type: DIRECT

- name: Bone Marrow Hematopoietic Failure
  description: >-
    Haematopoiesis is the most continuously proliferative tissue in the adult,
    so it is the first to fail and the last to recover. Without leucovorin
    rescue, marrow suppression is the dose-limiting toxicity of high-dose
    methotrexate; with pharmacokinetically guided rescue it is uncommon unless
    elimination is delayed.
  biological_scale: TISSUE
  cell_types:
  - preferred_term: hematopoietic precursor cell
    term:
      id: CL:0008001
      label: hematopoietic precursor cell
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "A dose-limiting toxicity of HDMTX in the absence of leucovorin is severe, prolonged myelosuppression"
    explanation: >-
      Establishes marrow failure as the dose-limiting consequence of the
      unrescued folate block.
  downstream:
  - target: Sepsis
    causal_link_type: INDIRECT_KNOWN_INTERMEDIATES
    description: >-
      Loss of the neutrophil defence is one of the two inputs to the septic
      outcome; the other is the breached mucosal barrier, and the entry draws
      both edges because the deaths follow the two arriving together.
    evidence:
    - reference: PMID:25172240
      reference_title: "Clinical characteristics and risk factors for low dose methotrexate toxicity: a cohort of 28 patients."
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "Seven (25%) patients died, all from pancytopenia followed by sepsis."
      explanation: >-
        States the sequence directly in the series' own mortality data:
        pancytopenia first, then sepsis, in every death.
  - target: Pancytopenia
    causal_link_type: DIRECT
  - target: Decreased total neutrophil count
    causal_link_type: DIRECT
  - target: Thrombocytopenia
    causal_link_type: DIRECT
  - target: Bone marrow hypocellularity
    causal_link_type: DIRECT

- name: Gastrointestinal Mucosal Barrier Breakdown
  description: >-
    The crypt is among the fastest-cycling compartments in the body, and the
    villus epithelium above it is replaced from there. Skin is the third such
    compartment and fails the same way, which is why the dermatologic toxicity
    hangs off this node too. Killing the crypt
    removes the replacement without removing the loss, so the villus shortens
    and the barrier opens. Mucositis is therefore not only painful in its own
    right but a portal: it is the route by which a neutropenic patient becomes
    a septic one.
  biological_scale: TISSUE
  cell_types:
  - preferred_term: intestinal crypt stem cell
    term:
      id: CL:0002250
      label: intestinal crypt stem cell
  - preferred_term: enterocyte
    term:
      id: CL:0000584
      label: enterocyte
  downstream:
  - target: Sepsis
    causal_link_type: INDIRECT_KNOWN_INTERMEDIATES
    description: >-
      The barrier failure is the portal of entry. It is not sufficient alone,
      which is why this edge and the one from marrow failure converge on the
      same node rather than either carrying the outcome by itself.
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "Oral mucositis can become a dose-limiting toxicity, require the use of opioids, increase infectious risk, and lead to chemotherapy delays."
      explanation: >-
        Names increased infectious risk as a consequence of the mucosal lesion,
        which is the step this edge asserts.
  - target: Mucocutaneous toxicity
    causal_link_type: DIRECT
  - target: Oral ulcer
    causal_link_type: DIRECT
  - target: Diarrhea
    causal_link_type: DIRECT
  - target: Nausea
    causal_link_type: DIRECT

- name: Hepatocellular Injury
  description: >-
    The liver is injured in two different ways by the same drug on two different
    schedules, and conflating them is the commonest error about methotrexate
    hepatotoxicity. After high-dose exposure the injury is an acute, transient,
    largely inconsequential transaminase rise. Under years of weekly low-dose
    exposure the concern is instead cumulative fibrosis, and how that risk
    divides between the drug and coexisting fatty liver disease -- which both
    raises the risk of methotrexate hepatotoxicity and is itself worsened by
    the drug -- is actively contested.
  biological_scale: TISSUE
  cell_types:
  - preferred_term: hepatocyte
    term:
      id: CL:0000182
      label: hepatocyte
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Hepatotoxicity after HDMTX is much less common than with the lower, long-term oral methotrexate dosing that is used in patients with rheumatoid arthritis, who are at risk for liver fibrosis"
    explanation: >-
      Supports separating the two hepatic phenotypes by dose regimen, which is
      what this node's two downstream targets represent.
  - reference: PMID:36564450
    reference_title: Busting the myth of methotrexate chronic hepatotoxicity.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "patients with NAFLD are at increased risk for methotrexate hepatotoxicity, and methotrexate can worsen the course of NAFLD"
    explanation: >-
      Records the interaction with fatty liver disease that qualifies how much
      of the chronic fibrosis risk is attributable to the drug alone.
  downstream:
  - target: Elevated circulating hepatic transaminase concentration
    causal_link_type: DIRECT
  - target: Hepatic fibrosis
    causal_link_type: INDIRECT_KNOWN_INTERMEDIATES
    description: >-
      Fibrosis follows repeated low-grade injury over years rather than any
      single exposure, so the link runs through a chronic injury-repair cycle
      rather than directly.

- name: Cerebral White Matter Injury
  description: >-
    Methotrexate neurotoxicity presents as a stroke mimic: hemiparesis, facial
    droop, altered consciousness or seizure, with restricted diffusion in deep
    white matter on MRI. The imaging resemblance to infarction is the clinical
    danger, because it invites thrombolysis for a lesion that is not
    thrombotic. The mechanism is not settled; the entry records the adenosine
    account as a hypothesis rather than as the explanation.
  biological_scale: TISSUE
  cell_types:
  - preferred_term: oligodendrocyte
    term:
      id: CL:0000128
      label: oligodendrocyte
  evidence:
  - reference: PMID:36582751
    reference_title: "Recurrence of methotrexate-induced leukoencephalopathy after methotrexate rechallenge: A case report and literature review."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "MTX has been linked to adverse neurologic effects that mimic acute stroke, including facial drooping, hemiplegia, impaired consciousness, and seizures, as well as changes on imaging-known as MTX-induced leukoencephalopathy (LE)."
    explanation: >-
      Describes the clinical and radiological syndrome this node represents,
      including the stroke-mimicry that makes it a diagnostic trap.
  - reference: PMID:36582751
    reference_title: "Recurrence of methotrexate-induced leukoencephalopathy after methotrexate rechallenge: A case report and literature review."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Six months afterward, he was rechallenged with MTX and developed recurrence of symptoms."
    explanation: >-
      A within-patient rechallenge recurrence, which is the strongest
      attribution available from a single case that the drug and not a
      coincident event produced the lesion.
  downstream:
  - target: Leukoencephalopathy
    causal_link_type: DIRECT
  - target: Encephalopathy
    causal_link_type: DIRECT
  - target: Seizure
    causal_link_type: DIRECT

- name: Renal Tubular Methotrexate Crystallisation
  description: >-
    This is the arm that has nothing to do with folate. Methotrexate and its
    hepatic metabolite 7-hydroxymethotrexate are weak acids and are poorly
    soluble when the urine is acid; concentrated in the tubular lumen by water
    reabsorption, they come out of solution. The entire supportive-care ritual
    around high-dose methotrexate exists to prevent this one physical event.
  biological_scale: TISSUE
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Because methotrexate is acidic, drug crystals are not present in urine with an alkaline pH, as alkalinization greatly increases methotrexate solubility and excretion."
    explanation: >-
      States the pH dependence that makes this a solubility event rather than a
      pharmacological one, and therefore one that alkalinisation can prevent.
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Precipitation of methotrexate crystals occurs in acidic urine (pH < 5.5) when the concentration of methotrexate in the renal tubules exceeds 2 × 10−3 molar."
    explanation: >-
      Gives the two quantitative conditions, pH and luminal concentration, under
      which precipitation occurs.
  downstream:
  - target: Tubular Obstruction by Methotrexate Crystals
    causal_link_type: DIRECT
  - target: Direct Tubular Epithelial Toxicity
    causal_link_type: DIRECT
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "Nephrotoxicity results from crystallization of methotrexate in the renal tubular lumen, leading to tubular toxicity"
      explanation: >-
        Asserts the causal step from luminal crystallisation to tubular injury,
        which is exactly this edge.
  - target: Afferent Arteriolar Vasoconstriction
    causal_link_type: INDIRECT_UNKNOWN_INTERMEDIATES
    description: >-
      The source names the vasoconstriction as a consequence of intrarenal
      crystal formation but does not say by what route, so the intermediates
      are recorded as unknown.

- name: Tubular Obstruction by Methotrexate Crystals
  description: >-
    The mechanical limb. Precipitated crystals block the tubular lumen. This is
    one of three insults the source states can each worsen the injury on its
    own, which is why they are modelled as separate nodes rather than one.
  biological_scale: TISSUE
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Intrarenal crystal formation can lead to tubule obstruction, direct toxic damage to the renal tubular epithelium (due to prolonged contact with methotrexate), and hypoperfusion from afferent arteriolar vasoconstriction, each of which independently can worsen AKI"
    explanation: >-
      Names tubule obstruction as the first of three independently sufficient
      injury routes.
  downstream:
  - target: Acute kidney injury
    causal_link_type: DIRECT
  - target: Impaired Renal Methotrexate Elimination
    causal_link_type: DIRECT

- name: Direct Tubular Epithelial Toxicity
  description: >-
    The cytotoxic limb. Prolonged contact between the tubular epithelium and
    concentrated luminal methotrexate poisons the cells directly, independently
    of any obstruction. This is the only one of the three limbs with a cell type
    to bind.
  biological_scale: TISSUE
  cell_types:
  - preferred_term: renal tubular epithelial cell
    term:
      id: CL:1000507
      label: kidney tubule cell
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Intrarenal crystal formation can lead to tubule obstruction, direct toxic damage to the renal tubular epithelium (due to prolonged contact with methotrexate), and hypoperfusion from afferent arteriolar vasoconstriction, each of which independently can worsen AKI"
    explanation: >-
      Names direct toxic damage to the tubular epithelium, attributed to
      prolonged contact, as an independently sufficient injury route.
  downstream:
  - target: Acute kidney injury
    causal_link_type: DIRECT
  - target: Impaired Renal Methotrexate Elimination
    causal_link_type: DIRECT

- name: Afferent Arteriolar Vasoconstriction
  description: >-
    The haemodynamic limb, and the one that is not a tubular event at all:
    constriction of the afferent arteriole drops glomerular perfusion. It
    carries no cell type because the source names a vessel response rather than
    a cell population.
  biological_scale: TISSUE
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Intrarenal crystal formation can lead to tubule obstruction, direct toxic damage to the renal tubular epithelium (due to prolonged contact with methotrexate), and hypoperfusion from afferent arteriolar vasoconstriction, each of which independently can worsen AKI"
    explanation: >-
      Names hypoperfusion from afferent arteriolar vasoconstriction as the third
      independently sufficient injury route.
  downstream:
  - target: Acute kidney injury
    causal_link_type: DIRECT
  - target: Impaired Renal Methotrexate Elimination
    causal_link_type: DIRECT

- name: Impaired Renal Methotrexate Elimination
  description: >-
    The hinge of the whole entry. Because over 90% of the drug leaves by the
    kidney, injuring the kidney raises the concentration of the agent that
    injured it. Higher concentration means more crystallisation and a longer
    intracellular exposure everywhere else, so the renal arm does not merely add
    a phenotype: it sets the severity and duration of the folate arm. Clinically
    this loop is named delayed methotrexate elimination, and it is what turns a
    routine infusion into a fatal one.
  biological_scale: ORGANISM
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Renal toxicity leads to impaired methotrexate clearance and prolonged exposure to toxic concentrations, which further worsen renal function and exacerbate nonrenal adverse events"
    explanation: >-
      States the feedback explicitly, in both directions: impaired clearance
      worsens the kidney further and amplifies the non-renal toxicities.
  downstream:
  - target: Renal Tubular Methotrexate Crystallisation
    causal_link_type: DIRECT
    description: >-
      The returning limb of the vicious cycle. A rising concentration in a
      kidney that can no longer excrete it precipitates further, which is why
      this edge closes a loop rather than continuing a chain.
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "Renal toxicity leads to impaired methotrexate clearance and prolonged exposure to toxic concentrations, which further worsen renal function and exacerbate nonrenal adverse events"
      explanation: >-
        States that impaired clearance raises the concentration and further
        worsens renal function. It establishes the loop but not the
        precipitation step specifically, which the next item supplies.
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "Precipitation of methotrexate crystals occurs in acidic urine (pH < 5.5) when the concentration of methotrexate in the renal tubules exceeds 2 × 10−3 molar."
      explanation: >-
        Supplies the missing step that actually closes the loop: precipitation
        is concentration-dependent, so the higher luminal concentration
        produced by impaired clearance precipitates more.
  - target: Bone Marrow Hematopoietic Failure
    causal_link_type: DIRECT
    description: >-
      Amplification rather than initiation. The marrow lesion is produced by the
      folate arm; delayed elimination decides how deep and how prolonged it is.
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "Prolonged renal dysfunction with increased systemic methotrexate exposure can cause myelosuppression, mucositis, hepatotoxicity, and, in severe cases, multiorgan failure"
      explanation: >-
        Names myelosuppression among the consequences of prolonged exposure
        caused by renal dysfunction, which is this edge.
  - target: Gastrointestinal Mucosal Barrier Breakdown
    causal_link_type: DIRECT
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "Prolonged renal dysfunction with increased systemic methotrexate exposure can cause myelosuppression, mucositis, hepatotoxicity, and, in severe cases, multiorgan failure"
      explanation: >-
        Names mucositis among the same set of consequences of prolonged
        exposure.
  - target: Hepatocellular Injury
    causal_link_type: DIRECT
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "Prolonged renal dysfunction with increased systemic methotrexate exposure can cause myelosuppression, mucositis, hepatotoxicity, and, in severe cases, multiorgan failure"
      explanation: >-
        Names hepatotoxicity among the same set of consequences of prolonged
        exposure.

mechanistic_hypotheses:
- hypothesis_group_id: adenosine_neurotoxicity
  hypothesis_label: Adenosine accumulation as the cause of acute methotrexate encephalopathy
  status: EMERGING
  description: >-
    The proposal is that inhibition of AICAR transformylase backs up AICA-ribotide,
    which raises extracellular adenosine, and that adenosine acting on central
    receptors produces the vasoactive and neuromodulatory changes seen as acute
    methotrexate encephalopathy. What supports it is a measurement and a
    response: central adenosine is raised in affected patients, and aminophylline,
    which displaces adenosine from central receptors, has reversed symptoms in a
    small uncontrolled series. What it lacks is any controlled demonstration, and
    the source that reports it calls it potential rather than established. It is
    recorded here so that the edges depending on it are visibly conditional.
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "A potential mechanism of neurotoxicity is the accumulation of adenosine after MTX-induced reductions in purine synthesis"
    explanation: >-
      The hedged wording is the point: the source advances this as a candidate
      mechanism, which is why the hypothesis is EMERGING rather than CANONICAL.

phenotypes:
- name: Pancytopenia
  category: Hematologic
  description: >-
    The dominant manifestation of low-dose methotrexate toxicity, and the one
    that kills, because a patient with an open mucosal barrier and no
    neutrophils becomes septic. In a hospitalised low-dose toxicity series it
    was present in over three-quarters of patients.
  frequency: FREQUENT
  phenotype_term:
    preferred_term: Pancytopenia
    term:
      id: HP:0001876
      label: Pancytopenia
  evidence:
  - reference: PMID:25172240
    reference_title: "Clinical characteristics and risk factors for low dose methotrexate toxicity: a cohort of 28 patients."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Pancytopenia was the most common manifestation of low dose MTX toxicity detected in 78.5% of the patients."
    explanation: >-
      Quantifies both the association and the frequency band in the series'
      own hospitalised low-dose cohort.

- name: Decreased total neutrophil count
  category: Hematologic
  description: >-
    Neutropenia is the component of the cytopenia that determines the immediate
    risk, and it arrives alongside mucositis rather than after it, so the portal
    of entry and the loss of defence coincide.
  phenotype_term:
    preferred_term: Neutropenia
    term:
      id: HP:0001875
      label: Decreased total neutrophil count
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "When elimination is delayed because of third spacing and fluid accumulations or as a result of renal injury, neutropenia and thrombocytopenia may be severe"
    explanation: >-
      Reports neutropenia as a consequence of delayed elimination, and notes
      the conditions under which it becomes severe.

- name: Thrombocytopenia
  category: Hematologic
  description: >-
    Falls with the neutrophil count for the same reason, and carries the
    bleeding risk of the toxidrome.
  phenotype_term:
    preferred_term: Thrombocytopenia
    term:
      id: HP:0001873
      label: Thrombocytopenia
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "When elimination is delayed because of third spacing and fluid accumulations or as a result of renal injury, neutropenia and thrombocytopenia may be severe"
    explanation: >-
      Reports thrombocytopenia in the same clause and under the same
      precipitating conditions.

- name: Bone marrow hypocellularity
  category: Hematologic
  description: >-
    The marrow correlate of the peripheral cytopenias, reflecting loss of the
    proliferating precursor compartment rather than peripheral destruction.
  phenotype_term:
    preferred_term: Bone marrow hypocellularity
    term:
      id: HP:0005528
      label: Bone marrow hypocellularity

- name: Oral ulcer
  category: Gastrointestinal
  description: >-
    Ulcerative stomatitis is the earliest visible sign of the folate block and
    is used as a bedside monitor for it. Where serum methotrexate assays are
    unavailable, twice-daily inspection of the mucous membranes substitutes for
    them.
  phenotype_term:
    preferred_term: Ulcerative stomatitis
    term:
      id: HP:0000155
      label: Oral ulcer
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Oral mucositis can become a dose-limiting toxicity, require the use of opioids, increase infectious risk, and lead to chemotherapy delays."
    explanation: >-
      Establishes oral mucositis as a dose-limiting clinical problem and links
      it to the infectious risk that makes it more than a local lesion.

- name: Diarrhea
  category: Gastrointestinal
  description: >-
    The lower-tract expression of the same barrier failure, and a compounding
    risk factor in its own right: the fluid loss produces the volume depletion
    that promotes tubular crystallisation.
  phenotype_term:
    preferred_term: Diarrhea
    term:
      id: HP:0002014
      label: Diarrhea
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Volume depletion is perhaps the most important and can result from fluid losses due to vomiting or diarrhea, adrenal insufficiency, or renal salt wasting"
    explanation: >-
      Cited for the second claim in this description: diarrhoeal fluid loss is
      named as a route to the volume depletion that raises renal risk.

- name: Nausea
  category: Gastrointestinal
  description: >-
    Emesis is common after high-dose infusion even under antiemetic cover, and
    matters beyond comfort because vomiting removes the oral fluid intake that
    the renal protection depends on.
  frequency: OCCASIONAL
  phenotype_term:
    preferred_term: Nausea and vomiting
    term:
      id: HP:0002017
      label: Nausea and vomiting
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Emesis occurs in 10%–30% of patients receiving HDMTX even when appropriate antiemetics are used; in this subgroup, antiemetics should be escalated to completely control vomiting and additional hydration provided to replace lost fluid"
    explanation: >-
      Gives the frequency band and states the reason it is managed
      aggressively, namely replacement of the lost fluid.

- name: Acute kidney injury
  category: Renal
  description: >-
    Present in a minority of high-dose courses but decisive when it occurs,
    because it converts a self-limited exposure into a self-sustaining one. It
    begins as an asymptomatic creatinine rise and progresses to tubular
    necrosis.
  phenotype_term:
    preferred_term: Acute kidney injury
    term:
      id: HP:0001919
      label: Acute kidney injury
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "it can cause significant toxicity, including acute kidney injury (AKI) in 2%-12% of patients"
    explanation: >-
      Gives the incidence range across high-dose methotrexate patients. No
      frequency band is recorded because 2%-12% straddles two of them.
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Crystal-induced nephropathy initially manifests as an asymptomatic elevation in serum creatinine and then progresses to tubular necrosis and more severe renal injury."
    explanation: >-
      Describes the clinical course from silent creatinine rise to tubular
      necrosis, which is the progression claimed here.

- name: Elevated circulating hepatic transaminase concentration
  category: Hepatic
  description: >-
    Nearly universal after high-dose methotrexate and, in that setting, of no
    clinical consequence. Recorded because the contrast with chronic low-dose
    exposure is the point: the same enzyme rise means something different
    depending on which regimen produced it.
  frequency: VERY_FREQUENT
  phenotype_term:
    preferred_term: Elevated serum transaminases
    term:
      id: HP:0002910
      label: Elevated circulating hepatic transaminase concentration
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "almost all patients have elevations of serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) values after HDMTX, but these laboratory findings have no clinical significance"
    explanation: >-
      Supports both halves of the description: the near-universal frequency
      after high-dose exposure, and its lack of clinical significance there.

- name: Hepatic fibrosis
  category: Hepatic
  description: >-
    The concern of chronic weekly dosing rather than of acute poisoning, and the
    reason long-term methotrexate patients undergo periodic liver monitoring.
    How much of the risk is the drug and how much is coexisting fatty liver
    disease is actively contested.
  phenotype_term:
    preferred_term: Hepatic fibrosis
    term:
      id: HP:0001395
      label: Hepatic fibrosis
    clinical_course: PROGRESSIVE
  evidence:
  - reference: PMID:36564450
    reference_title: Busting the myth of methotrexate chronic hepatotoxicity.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Liver fibrosis and the mechanisms of fibrogenesis also need to be considered in relation to chronic exposure to methotrexate."
    explanation: >-
      Associates fibrosis specifically with chronic rather than acute exposure,
      which is the distinction this phenotype records.

- name: Leukoencephalopathy
  category: Neurologic
  description: >-
    Deep white matter signal change on MRI with restricted diffusion, after
    intrathecal or high-dose systemic exposure. Often reversible, and the
    imaging can be indistinguishable from acute infarction.
  phenotype_term:
    preferred_term: Leukoencephalopathy
    term:
      id: HP:0002352
      label: Leukoencephalopathy
  evidence:
  - reference: PMID:36582751
    reference_title: "Recurrence of methotrexate-induced leukoencephalopathy after methotrexate rechallenge: A case report and literature review."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Brain magnetic resonance imaging (MRI) revealed bilateral deep white matter T2 hyperintense foci, increased on the right, with associated diffusion restriction in the right centrum semiovale-consistent with MTX-induced LE."
    explanation: >-
      Documents the specific imaging findings, including the diffusion
      restriction that produces the resemblance to infarction.

- name: Encephalopathy
  category: Neurologic
  description: >-
    Transient central nervous system disturbance is reported after a meaningful
    minority of high-dose courses, ranging from confusion and somnolence to
    cortical blindness and hemiparesis. Onset is typically within 24 hours of
    the dose and symptoms usually resolve spontaneously.
  frequency: OCCASIONAL
  phenotype_term:
    preferred_term: Encephalopathy
    term:
      id: HP:0001298
      label: Encephalopathy
    temporality: TRANSIENT
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "up to 11% of patients may have CNS events, including confusion, seizures, somnolence, and headaches with or without radiographic evidence of leukoencephalopathy"
    explanation: >-
      Gives the upper bound on frequency and enumerates the range of
      presentations grouped under this phenotype.

- name: Seizure
  category: Neurologic
  description: >-
    Part of the acute neurotoxic spectrum rather than a separate syndrome, and
    reported both with and without radiological leukoencephalopathy.
  phenotype_term:
    preferred_term: Seizure
    term:
      id: HP:0001250
      label: Seizure
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "up to 11% of patients may have CNS events, including confusion, seizures, somnolence, and headaches with or without radiographic evidence of leukoencephalopathy"
    explanation: >-
      Lists seizures among the central nervous system events attributed to the
      drug, with or without imaging change.

- name: Sepsis
  category: Infectious
  description: >-
    The mortality mechanism. Neither input kills on its own: the cytopenia
    removes the defence and the mucositis opens the door, and the deaths in the
    low-dose toxicity series followed the two arriving together. This is the
    node the whole supportive-care effort is ultimately trying to prevent.
  phenotype_term:
    preferred_term: Sepsis
    term:
      id: HP:0100806
      label: Sepsis
  evidence:
  - reference: PMID:25172240
    reference_title: "Clinical characteristics and risk factors for low dose methotrexate toxicity: a cohort of 28 patients."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Seven (25%) patients died, all from pancytopenia followed by sepsis."
    explanation: >-
      Establishes sepsis as the terminal event in every death in the series,
      and the sequence leading to it.

- name: Mucocutaneous toxicity
  category: Dermatologic
  description: >-
    Skin is the third fast-turnover compartment, alongside marrow and gut, and
    it fails for the same reason. It is named in the source's own summary of
    the toxicities that delayed elimination exacerbates, and at the severe end
    it reaches toxic epidermal necrolysis.
  phenotype_term:
    preferred_term: Mucocutaneous toxicity
    term:
      id: HP:0000951
      label: Abnormality of the skin
    coarse_binding_basis: VARIABLE_SPECTRUM
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "exacerbate nonrenal adverse events, including myelosuppression, mucositis, dermatologic toxicity, and hepatotoxicity"
    explanation: >-
      Names dermatologic toxicity alongside the marrow, mucosal and hepatic
      arms this entry already models, which is why skin belongs beside them.

- name: Interstitial pneumonitis
  category: Respiratory
  description: >-
    The one toxicity in this entry that does not belong to either arm. It occurs
    in patients on weekly low-dose methotrexate and is rare after high-dose
    infusion, which is the reverse of every other toxicity here and argues
    against a concentration-driven mechanism. Its pathogenesis is not
    established and is recorded as an open question rather than assigned to a
    node.
  phenotype_term:
    preferred_term: Interstitial pneumonitis
    term:
      id: HP:0006515
      label: Interstitial pneumonitis
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Pulmonary toxicity is observed in 0.5% of patients per year who receive weekly low-dose methotrexate, but rarely with HDMTX"
    explanation: >-
      Supports both the frequency and the inverted dose relationship that is the
      reason this phenotype is left unattached to the pathograph.

genetic:
- name: SLCO1B1
  relationship_type: SUSCEPTIBILITY
  gene_term:
    preferred_term: SLCO1B1
    term:
      id: hgnc:10959
      label: SLCO1B1
  notes: >-
    Encodes OATP1B1, a hepatic uptake transporter that handles methotrexate.
    Common function-altering variants account for part of the between-patient
    spread in clearance, which in this entry means they move a patient along
    the delayed-elimination axis rather than causing a distinct lesion. This is
    a modifier of exposure duration, not a cause of toxicity. The cited study
    establishes that genotype improves a clearance model; it does not state the
    direction of that effect, and neither does this entry.
  evidence:
  - reference: PMID:32961024
    reference_title: Effect of SLCO1B1 Polymorphisms on High-Dose Methotrexate Clearance in Children and Young Adults With Leukemia and Lymphoblastic Lymphoma.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The addition of the SLCO1B1 variants individually further improved the model"
    explanation: >-
      The study's own pharmacokinetic modelling result: SLCO1B1 genotype adds
      explanatory power for clearance variability beyond clinical covariates.
  - reference: PMID:32961024
    reference_title: Effect of SLCO1B1 Polymorphisms on High-Dose Methotrexate Clearance in Children and Young Adults With Leukemia and Lymphoblastic Lymphoma.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: BACKGROUND
    snippet: "Common function-altering polymorphisms in SLCO1B1 (encodes OATP1B1, which transports MTX) may contribute to clearance variability."
    explanation: >-
      Identifies the gene product and its relation to methotrexate. Quoted from
      the paper's framing rather than from its results, hence BACKGROUND.

- name: MTHFR
  relationship_type: SUSCEPTIBILITY
  gene_term:
    preferred_term: MTHFR
    term:
      id: hgnc:7436
      label: MTHFR
  notes: >-
    The C677T variant reduces methylenetetrahydrofolate reductase activity and
    shifts intracellular folate pools. In a prospective series of 148 high-dose
    courses in 32 patients, homozygotes had substantially higher odds of
    nephrotoxicity, hepatotoxicity and haematologic toxicity than wild type.
    Note the cohort is small and confined to one tumour type; the wider
    MTHFR-toxicity literature is heterogeneous, and the entry does not treat
    the association as settled across all regimens.
  evidence:
  - reference: PMID:33069634
    reference_title: Influence of MTHFR C677T Polymorphism on High-Dose Methotrexate-Related Toxicity in Patients With Primary Central Nervous System Diffuse Large B-Cell Lymphoma.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The patients carrying the homozygous variant genotype had a higher risk of developing nephrotoxicity than those carrying the wild-type genotype (odds ratio [OR] 13.08; 95% confidence interval [CI], 1.65-103.86; P = .002)"
    explanation: >-
      The study's own effect estimate for nephrotoxicity by genotype. The very
      wide confidence interval is why the note calls the association
      unsettled rather than established.
  - reference: PMID:33069634
    reference_title: Influence of MTHFR C677T Polymorphism on High-Dose Methotrexate-Related Toxicity in Patients With Primary Central Nervous System Diffuse Large B-Cell Lymphoma.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "Significant differences were observed in hepatotoxicity (OR 9.33; 95% CI, 2.54-34.27; P < .001) and hematologic toxicity (OR 3.09; 95% CI, 1.18-8.07; P = .024) in addition to nephrotoxicity between the homozygous variant genotype and the wild-type genotype."
    explanation: >-
      Extends the genotype association to the hepatic and haematologic
      toxicities, which is what makes this a modifier of the whole toxidrome
      rather than of one organ.

environmental:
- name: Therapeutic methotrexate administration
  description: >-
    The exposure is the treatment. That is what separates this entry from a
    poisoning: nobody encounters methotrexate by accident, and the dose that
    injures is usually the dose that was prescribed, delivered to a patient
    whose clearance was not what the protocol assumed.
  exposure_term:
    preferred_term: exposure to methotrexate
    term:
      id: ECTO:0000172
      label: exposure to methotrexate
  influences_mechanisms:
  - target: Systemic Methotrexate Exposure
    environmental_effect: TRIGGERS
    causal_link_type: DIRECT
    description: >-
      Administration establishes the systemic pool from which both arms of the
      mechanism follow.
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "High-dose methotrexate (HDMTX), defined as a dose higher than 500 mg/m2, is used to treat a range of adult and childhood cancers."
      explanation: >-
        Establishes the therapeutic administration that constitutes the
        exposure, and the dose threshold at which the high-dose toxicity
        profile applies.
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Although HDMTX is safely administered to most patients, it can cause significant toxicity"
    explanation: >-
      Frames the exposure correctly: the same administration is tolerated by
      most patients and injures a minority, which is why the entry's risk
      factors do most of the work.

- name: Co-administration of non-steroidal anti-inflammatory drugs
  description: >-
    Weak organic acids and transporter substrates given alongside methotrexate
    can slow its renal elimination and push a patient into the
    delayed-elimination loop. The size of this effect is genuinely disputed. A
    transporter-focused review finds the best-documented interactions to be with
    indomethacin, ketoprofen, omeprazole, piperacillin/tazobactam,
    ciprofloxacin, cotrimoxazole, probenecid and imatinib, while explicitly
    declining to endorse a firm NSAID or proton-pump-inhibitor interaction on
    the available evidence. Standard high-dose protocols nonetheless reconcile
    and substitute these drugs before infusion, on the reasoning that the cost
    of avoiding them is small and the cost of being wrong is not.
  exposure_term:
    preferred_term: exposure to NSAIDs
    term:
      id: ECTO:2000011
      label: exposure to NSAIDs
  notes: >-
    Scoped to NSAIDs rather than to the whole interacting-drug list because that
    is where both the mechanistic and the contradicting evidence sit, and a
    binding should not name a wider exposure than its sources establish. ECTO
    was searched on 2026-09-22: `runoak -i sqlite:obo:ecto info "l~NSAID"`
    returns ECTO:2000011 exposure to NSAIDs, defined as an exposure to NSAID
    therapy and descending from ECTO:2000057 exposure to pharmacotherapy, which
    is the co-medication sense this entry means; `l~anti-inflammatory` also
    returns ECTO:9001720 exposure to non-steroidal anti-inflammatory drug, the
    substance-exposure counterpart, which was not chosen because the entry
    describes a drug given as therapy alongside methotrexate rather than an
    environmental encounter with the compound. The other agents named in the
    description (omeprazole, piperacillin/tazobactam, ciprofloxacin,
    cotrimoxazole, probenecid, imatinib) are recorded in prose; no ECTO class
    covers that heterogeneous set, and inventing one binding for all of them
    would assert a grouping the ontology does not make.
  influences_mechanisms:
  - target: Impaired Renal Methotrexate Elimination
    environmental_effect: EXACERBATES
    causal_link_type: INDIRECT_KNOWN_INTERMEDIATES
    description: >-
      The proposed route is competition at renal tubular transporters, so the
      intermediates are named but the clinical effect size is contested.
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      quote_role: REVIEW_SYNTHESIS
      snippet: "Studies in mice documented inhibition of renal tubule methotrexate transporters by indomethacin and ketoprofen, with reduced methotrexate elimination and prolonged elevated levels and toxicity"
      explanation: >-
        The mechanistic demonstration behind the interaction, in mice. Graded
        MODEL_ORGANISM because that is the evidence the quoted sentence
        describes, and REVIEW_SYNTHESIS because this review is relaying it
        rather than reporting its own experiment.
    - reference: PMID:36946211
      reference_title: "A critical review of methotrexate clinical interactions: role of transporters."
      supports: REFUTE
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "Our findings showed no firm evidence of interactions of proton pump inhibitors (PPIs), levetiracetam, and NSAIDS with MTX."
      explanation: >-
        Directly contradicts the clinical interaction claim for the two drug
        classes most often named. Recorded as REFUTE rather than dropped,
        because the disagreement with the preceding item is the honest state of
        this edge.
  evidence:
  - reference: PMID:36946211
    reference_title: "A critical review of methotrexate clinical interactions: role of transporters."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "most interaction records with methotrexate occurred with coadministration of indomethacin, ketoprofen, omeprazole, piperacillin/tazobactam, ciprofloxacin, cotrimoxazole, probenecid, and imatinib, mainly due to the role of transporters"
    explanation: >-
      Names the agents with the most interaction records and attributes them to
      transporter competition. Only indomethacin and ketoprofen fall inside this
      exposure's NSAID binding; the rest are recorded in the description and sit
      outside it.

- name: Volume depletion and acidic urine during high-dose infusion
  description: >-
    Two physical conditions, both modifiable, that decide whether the filtered
    drug stays in solution. They are listed as an exposure rather than a
    phenotype because the supportive-care protocol treats them as the thing to
    be controlled, and controlling them is the single most effective
    intervention in the entry.
  exposure_term:
    preferred_term: dehydration and acidic urine during methotrexate infusion
  notes: >-
    No exposure term was bound. ECTO was searched on 2026-09-22 with
    `runoak -i sqlite:obo:ecto info` over `l~dehydration` and
    `l~water deprivation`, both of which return nothing at all; `l~fluid`,
    which returns only exposures to named liquids such as ECTO:0500001
    exposure to brake fluid and ECTO:7000109 exposure to bodily fluid; and
    `l~acid`, which returns CDNO concentration terms and chemical-acid
    classes, not a urinary-pH state. None of these describes volume depletion
    or urine acidity in a patient, so the slot carries a free-text
    `preferred_term` only.
  influences_mechanisms:
  - target: Renal Tubular Methotrexate Crystallisation
    environmental_effect: TRIGGERS
    causal_link_type: DIRECT
    description: >-
      Low urine flow concentrates the drug in the tubular lumen and low pH
      reduces its solubility, which are the two conditions precipitation
      requires.
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "Because volume depletion and acidic urine are major risk factors for AKI, hyperhydration and urine alkalinization are mandatory during HDMTX treatment"
      explanation: >-
        Names both conditions as the major risk factors and states that the
        protocol exists to correct them.
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Because volume depletion and acidic urine are major risk factors for AKI, hyperhydration and urine alkalinization are mandatory during HDMTX treatment"
    explanation: >-
      Supports the claim that these two conditions are the modifiable risk
      factors the supportive-care protocol targets.

- name: Third-space fluid sequestration
  description: >-
    A pleural effusion or ascites acts as a reservoir. Methotrexate partitions
    into the collection while the plasma concentration is high and leaks back
    afterwards, so elimination looks complete and then is not. It is the one
    route to delayed elimination that does not involve the kidney at all, which
    is why it has its own countermeasure: drain the collection, or defer the
    infusion until it resolves.
  exposure_term:
    preferred_term: third-space fluid collection acting as a methotrexate reservoir
  notes: >-
    No exposure term was bound. This is a patient's own fluid compartment rather
    than an external exposure, and ECTO was searched on 2026-09-22 with
    `runoak -i sqlite:obo:ecto info "l~fluid"`, which returns only exposures to
    named external liquids (ECTO:0500001 exposure to brake fluid, ECTO:7000109
    exposure to bodily fluid and similar), and `l~effusion` and `l~ascites`,
    which return nothing. The slot carries a free-text `preferred_term` only.
  influences_mechanisms:
  - target: Impaired Renal Methotrexate Elimination
    environmental_effect: EXACERBATES
    causal_link_type: DIRECT
    description: >-
      Redistribution out of the collection sustains the systemic concentration
      after the infusion has finished, which is delayed elimination by a
      non-renal route.
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "When elimination is delayed because of third spacing and fluid accumulations or as a result of renal injury, neutropenia and thrombocytopenia may be severe"
      explanation: >-
        Names third spacing as a cause of delayed elimination distinct from
        renal injury, which is exactly the claim this link makes.
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Delayed methotrexate excretion has been associated with extravascular fluid collections, including ascites, pleural effusions, or intracranial fluid"
    explanation: >-
      Names the specific collections that behave as reservoirs, which is the
      entry-level claim this exposure makes.

- name: Folic acid supplementation during chronic low-dose therapy
  description: >-
    Replacing folate alongside weekly methotrexate reduces the gastrointestinal
    and hepatic adverse effects without abolishing the therapeutic effect. It is
    the clearest protective exposure in the entry, and the protection is
    selective in an informative way: pooled trial data show a clear reduction in
    gastrointestinal effects and a large one in transaminase elevation, while
    the reduction in stomatitis does not reach significance and the
    haematologic outcome could not be assessed at all.
  exposure_term:
    preferred_term: exposure to folic acid
    term:
      id: ECTO:9000123
      label: exposure to folic acid
  influences_mechanisms:
  - target: Hepatocellular Injury
    environmental_effect: PROTECTS_AGAINST
    causal_link_type: DIRECT
    description: >-
      The largest and best-evidenced protective effect: supplementation cuts the
      incidence of methotrexate-associated transaminase elevation by roughly
      three quarters in relative terms.
    evidence:
    - reference: PMID:23728635
      reference_title: Folic acid and folinic acid for reducing side effects in patients receiving methotrexate for rheumatoid arthritis.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "Folic and folinic acid also appear to be protective against abnormal serum transaminase elevation caused by MTX, with a 76.9% relative (16% absolute) risk reduction (RR 0.23, 95% CI 0.15 to 0.34; P < 0.00001)"
      explanation: >-
        Quantifies the protective effect on the hepatic arm specifically, from a
        pooled analysis of randomised placebo-controlled trials.
  - target: Gastrointestinal Mucosal Barrier Breakdown
    environmental_effect: PROTECTS_AGAINST
    causal_link_type: DIRECT
    description: >-
      Smaller than the hepatic effect and, notably, driven by the general
      gastrointestinal symptom outcome rather than by stomatitis, which did not
      separate from placebo.
    evidence:
    - reference: PMID:23728635
      reference_title: Folic acid and folinic acid for reducing side effects in patients receiving methotrexate for rheumatoid arthritis.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "a 26% relative (9% absolute) risk reduction was seen for the incidence of GI side effects such as nausea, vomiting or abdominal pain (RR 0.74, 95% CI 0.59 to 0.92; P = 0.008)"
      explanation: >-
        Gives the effect size for the gastrointestinal arm, which is the
        magnitude this link claims.
    - reference: PMID:23728635
      reference_title: Folic acid and folinic acid for reducing side effects in patients receiving methotrexate for rheumatoid arthritis.
      supports: REFUTE
      evidence_source: HUMAN_CLINICAL
      quote_role: PRIMARY_RESULT
      snippet: "whilst showing a trend towards reduction in risk, the results were not statistically significant (RR 0.72, 95% CI 0.49 to 1.06)"
      explanation: >-
        The stomatitis outcome specifically did not reach significance, which
        qualifies rather than supports a blanket protective claim over the whole
        mucosal node. Recorded as REFUTE against that broader reading.
  evidence:
  - reference: PMID:23728635
    reference_title: Folic acid and folinic acid for reducing side effects in patients receiving methotrexate for rheumatoid arthritis.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "The results support a protective effect of supplementation with either folic or folinic acid for patients with rheumatoid arthritis during treatment with MTX."
    explanation: >-
      The review's own overall conclusion, which is the entry-level claim this
      exposure makes.
  - reference: PMID:23728635
    reference_title: Folic acid and folinic acid for reducing side effects in patients receiving methotrexate for rheumatoid arthritis.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "It does not appear that supplementation with either folic or folinic acid has a statistically significant effect on the efficacy of MTX in treating RA"
    explanation: >-
      Supports the second half of the description: the protection does not come
      at the cost of the therapeutic effect, which is what makes supplementation
      routine rather than a trade-off.
  notes: >-
    The link is drawn to the tissue-injury nodes rather than to Dihydrofolate
    Reductase Inhibition, which was the first draft. The trials measure clinical
    adverse events, not enzyme occupancy, so an edge to the molecular node would
    attach a clinical outcome to a biochemical claim the source does not make.

treatments:
- name: Hyperhydration and Urinary Alkalinisation
  description: >-
    The first-line intervention and the only one that prevents rather than
    rescues. High urine flow dilutes the drug in the tubular lumen and
    bicarbonate raises urinary pH above 7, which together keep methotrexate and
    7-hydroxymethotrexate in solution. The effect size is physical and large: a
    pH rise from 6.0 to 7.0 increases solubility five- to eightfold. Because it
    acts on the renal arm only, it does nothing about the folate block and is
    always given alongside leucovorin rather than instead of it.
  therapeutic_modality: OTHER
  treatment_term:
    preferred_term: fluid therapy with urinary alkalinisation
    term:
      id: NCIT:C116537
      label: Fluid Therapy
  target_mechanisms:
  - target: Renal Tubular Methotrexate Crystallisation
    treatment_effect: INHIBITS
    description: >-
      Acts directly on the precipitation step by changing the two variables it
      depends on, luminal concentration and pH.
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "An increase in urine pH from 6.0 to 7.0 increases the solubility of methotrexate and its metabolites by five- to eightfold"
      explanation: >-
        Quantifies the solubility change alkalinisation produces, which is the
        mechanism by which this treatment inhibits the crystallisation node.
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "The use of fluids to promote high urinary flow rates and alkalinize the urine protects the kidney from injury during treatment with HDMTX"
    explanation: >-
      States the protective effect of the combined hydration and alkalinisation
      strategy on the kidney.
  notes: >-
    The binding covers only the hydration half. NCIT:C28342 Alkalinization
    exists and was checked on 2026-09-22, but it is absent from the
    TreatmentActionTerm enum cache -- it is not reachable from NCIT:C25218
    Clinical Intervention or Procedure -- so it cannot sit in this slot. The
    two halves are in any case inseparable in practice, since the bicarbonate
    is delivered in the fluid, and the alkalinisation concept is carried in the
    treatment name and preferred_term.

- name: Leucovorin Rescue
  description: >-
    Leucovorin is a reduced folate, which means it enters the pathway
    downstream of the enzyme methotrexate is blocking. It does not displace the
    drug or repair the enzyme; it makes the blocked enzyme unnecessary. That is
    a genuine bypass rather than an antidote, and it has two consequences. It
    must not start too early, because rescuing the tumour is as easy as rescuing
    the marrow. And it competes with methotrexate for the same reduced folate
    carrier, so when the plasma methotrexate concentration is very high the dose
    must rise steeply to win that competition.
  therapeutic_modality: SMALL_MOLECULE
  treatment_term:
    preferred_term: Pharmacotherapy
    term:
      id: NCIT:C15986
      label: Pharmacotherapy
    therapeutic_agent:
    - preferred_term: leucovorin
      term:
        id: NCIT:C71631
        label: Leucovorin
  target_mechanisms:
  - target: Dihydrofolate Reductase Inhibition
    treatment_effect: BYPASSES
    description: >-
      Supplies reduced folate downstream of the blocked reductase, so the
      pathway runs while the enzyme remains inhibited.
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: IN_VITRO
      quote_role: REVIEW_SYNTHESIS
      snippet: "LV enters cells through the RFC and allows formation of FH4 despite the presence of MTX, which effectively rescues cells."
      explanation: >-
        States the bypass explicitly: tetrahydrofolate is formed despite the
        drug still being present, which is what distinguishes this from
        inhibition or displacement.
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Leucovorin is particularly effective in the prevention of myelosuppression, gastrointestinal toxicity, and neurotoxicity during treatment with HDMTX."
    explanation: >-
      Names the three toxicities leucovorin rescue prevents, which correspond to
      the three tissue arms downstream of the folate block.
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "leucovorin must compete with MTX to enter cells via the reduced folate carrier"
    explanation: >-
      Supports the dose-escalation claim: because rescue and drug share one
      transporter, the rescue dose is set by the methotrexate concentration it
      must outcompete.
  notes: >-
    The shared-transporter competition has a hard limit. At sufficiently high
    plasma methotrexate concentrations no achievable plasma leucovorin
    concentration delivers enough intracellular rescue, which is the situation
    glucarpidase exists for.

- name: Glucarpidase
  description: >-
    A recombinant bacterial carboxypeptidase that cuts methotrexate in the
    bloodstream into two inactive fragments. It is fast and near-complete: a
    single dose removes 97% or more of circulating drug within fifteen minutes.
    Its limitation is the one that matters mechanistically, and it is not a
    dosing detail. The enzyme stays in the plasma, so it does nothing to the
    methotrexate already inside cells. It removes the cause of further
    accumulation and leaves the existing lesion untouched, which is precisely
    why it is always given with leucovorin rather than in place of it.
  therapeutic_modality: OTHER
  treatment_term:
    preferred_term: Pharmacotherapy
    term:
      id: NCIT:C15986
      label: Pharmacotherapy
    therapeutic_agent:
    - preferred_term: glucarpidase
      term:
        id: NCIT:C346
        label: Glucarpidase
  target_mechanisms:
  - target: Systemic Methotrexate Exposure
    treatment_effect: INHIBITS
    description: >-
      Destroys the circulating pool enzymatically, acting on the exposure node
      itself rather than on any downstream mechanism.
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "Glucarpidase cleaves methotrexate into DAMPA and glutamate, two nontoxic metabolites, and thus provides an enzymatic method to rapidly remove methotrexate in patients with renal dysfunction"
      explanation: >-
        States the reaction and its product, and that its purpose is removal of
        circulating drug when the renal route has failed.
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "A single dose of glucarpidase (50 U/kg i.v. over 5 minutes) reduces plasma methotrexate concentrations by 97% or more within 15 minutes"
    explanation: >-
      Quantifies the magnitude and speed of the plasma reduction claimed in the
      description.
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "despite the decrease in the magnitude and duration of systemic exposure to methotrexate after glucarpidase, it has no effect on intracellular methotrexate concentrations"
    explanation: >-
      States the compartmental limitation that makes concurrent leucovorin
      mandatory rather than optional.
  notes: >-
    Two practical consequences follow from the enzyme's specificity. Leucovorin
    is also a glucarpidase substrate, so the two cannot be given within two
    hours of each other. And the product DAMPA cross-reacts with the standard
    methotrexate immunoassay, so for several days afterwards only a
    chromatographic method reports the true level.

- name: Drainage of Third-Space Fluid Collections
  description: >-
    Draining a pleural effusion or ascites before the infusion, or deferring the
    infusion until it resolves, removes the reservoir. It is listed alongside
    avoiding interacting medications as one of only two named strategies for
    preventing myelosuppression, which places it well above its usual billing as
    a procedural afterthought.
  therapeutic_modality: SURGERY
  treatment_term:
    preferred_term: drainage of a third-space fluid collection
    term:
      id: NCIT:C15329
      label: Surgical Procedure
  target_mechanisms:
  - target: Impaired Renal Methotrexate Elimination
    treatment_effect: INHIBITS
    description: >-
      Removes the compartment that sustains the systemic concentration after
      the infusion ends.
    evidence:
    - reference: PMID:27496039
      reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      quote_role: REVIEW_SYNTHESIS
      snippet: "draining effusions before treatment (or delaying HDMTX until effusions resolve)"
      explanation: >-
        Names drainage, or deferral until resolution, as the intervention
        directed at this mechanism.
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "The only known strategies to prevent myelosuppression are to prevent delayed methotrexate elimination by avoiding interacting medications around the time of infusion and draining effusions before treatment (or delaying HDMTX until effusions resolve) and to ensure optimal leucovorin dosing."
    explanation: >-
      Places drainage among the only named strategies for preventing
      myelosuppression, which supports the description's claim about its
      standing.

- name: Haemodialysis
  description: >-
    Used when methotrexate remains high despite hydration, alkalinisation and
    rescue. The evidence is weak by the standards of the rest of this entry,
    and one specific failure mode is documented: plasma concentrations rebound
    substantially after the procedure as drug redistributes out of tissue, so a
    single session is rarely sufficient.
  therapeutic_modality: OTHER
  treatment_term:
    preferred_term: Hemodialysis
    term:
      id: NCIT:C15248
      label: Hemodialysis
  target_mechanisms:
  - target: Systemic Methotrexate Exposure
    treatment_effect: INHIBITS
    description: >-
      Removes circulating drug extracorporeally when renal clearance has
      failed.
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "even when hemodialysis is effective, many patients experience a rebound in serum methotrexate concentrations of 10%–220% of the postprocedure values"
    explanation: >-
      Documents the rebound that limits the procedure's usefulness, with the
      reported range.

diagnosis:
- name: Serial plasma methotrexate concentration
  description: >-
    The central monitoring test of high-dose infusion, and the one that decides
    management there. Levels are drawn at protocol-defined intervals after the
    infusion starts and read against a nomogram; a level above the expected
    curve triggers leucovorin escalation and, if high enough, glucarpidase. Two
    limits are worth stating rather than assuming. It tracks the renal arm well
    and the other toxicities poorly. And it does not transfer to the low-dose
    setting at all: in a series of hospitalised low-dose toxicity, serum
    concentrations correlated with neither the neutropenia nor the
    thrombocytopenia, and the authors conclude there is no rationale for
    monitoring there.
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Plasma methotrexate monitoring is a reliable indicator specifically of nephrotoxicity but may be a limited predictor of other toxicities"
    explanation: >-
      States exactly the scope claim made here: the assay tracks the renal arm
      well and the rest of the toxidrome less well.
  - reference: PMID:25172240
    reference_title: "Clinical characteristics and risk factors for low dose methotrexate toxicity: a cohort of 28 patients."
    supports: REFUTE
    evidence_source: HUMAN_CLINICAL
    quote_role: PRIMARY_RESULT
    snippet: "There is no rationale for MTX therapeutic drug monitoring in the setting of low-dose toxicity."
    explanation: >-
      Directly contradicts any reading of this node as a test for methotrexate
      toxicity in general. Recorded as REFUTE against that broader claim, and
      the description is scoped to high-dose infusion accordingly.

- name: Urine pH monitoring during infusion
  description: >-
    Checked at every void, because a short period of acid urine is enough to
    start precipitation. A falling pH is acted on with bolus bicarbonate rather
    than observed.
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "It is also important to check urine pH values with each void during the infusion to ensure no extended periods of time with acidic urine, which could increase the risk for precipitation, nephrotoxicity, and delayed methotrexate elimination."
    explanation: >-
      States the monitoring frequency and the three consequences it is intended
      to prevent.

- name: Serum creatinine and urine output
  description: >-
    Standard renal monitoring, with a documented limitation: creatinine lags the
    injury, so tubular damage may already be irreversible by the time it rises.
    Falling urine output or a positive fluid balance identifies early injury
    sooner.
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "the rise in serum creatinine values lags behind progressive intrinsic renal damage, such that precise measurement of function at a specific moment is difficult"
    explanation: >-
      Supports the stated limitation of creatinine as a real-time measure of
      the injury this entry cares about.

animal_models:
- name: Methotrexate-induced small intestinal mucositis in the rat
  species: Rat
  publication: PMID:36996098
  description: >-
    Repeated intraperitoneal methotrexate in rats produces small intestinal
    mucosal injury with loss of crypt proliferation and disrupted tight-junction
    protein expression, and is a standard preparation for testing protective
    agents against the gastrointestinal arm.
  modeled_mechanisms:
  - target: Gastrointestinal Mucosal Barrier Breakdown
    relationship: RECAPITULATES
    fidelity: MODERATE
    model_scale: TISSUE
    description: >-
      Produces the crypt and villus lesion and the tight-junction changes that
      define the human node.
    limitations: >-
      Dosed intraperitoneally on a three-day cycle rather than by the human
      routes and schedules, and studied without leucovorin rescue, so it models
      the unrescued lesion rather than clinical mucositis under supportive care.
    readouts:
    - name: Small intestinal mucosal injury on histopathology
      target: Gastrointestinal Mucosal Barrier Breakdown
      direction: INCREASED
      interpretation: >-
        Histological confirmation that the preparation produces the mucosal
        lesion it is used to model.
      evidence:
      - reference: PMID:36996098
        reference_title: Therapeutic effect and mechanism of Daikenchuto in a model of methotrexate-induced acute small intestinal mucositis.
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        quote_role: PRIMARY_RESULT
        snippet: "The pathology results showed that small intestinal mucosal injury in the DKT-MTX group was less severe than that in the MTX group."
        explanation: >-
          Reports the histological comparison. The methotrexate-only arm is the
          more injured one, which is the measurement this readout records.
    - name: Crypt Ki-67-positive proliferating cell count
      target: Gastrointestinal Mucosal Barrier Breakdown
      direction: DECREASED
      interpretation: >-
        A direct measure of the crypt proliferative compartment, which is the
        cell population the folate block kills.
      evidence:
      - reference: PMID:36996098
        reference_title: Therapeutic effect and mechanism of Daikenchuto in a model of methotrexate-induced acute small intestinal mucositis.
        supports: SUPPORT
        evidence_source: MODEL_ORGANISM
        quote_role: PRIMARY_RESULT
        snippet: "The crypts in the DKT-MTX group contained more Ki-67-positive cells than MTX group."
        explanation: >-
          The methotrexate-only arm has fewer proliferating crypt cells, which
          is the loss of crypt proliferation this readout measures.
    evidence:
    - reference: PMID:36996098
      reference_title: Therapeutic effect and mechanism of Daikenchuto in a model of methotrexate-induced acute small intestinal mucositis.
      supports: SUPPORT
      evidence_source: MODEL_ORGANISM
      quote_role: PRIMARY_RESULT
      snippet: "Intraperitoneal injection of 10 mg/kg methotrexate (MTX) every 3 days for a total of 3 doses was used for induction of CIM in a rat model."
      explanation: >-
        Documents the induction protocol, which is what establishes this as a
        methotrexate model of the mucosal node rather than a general mucositis
        model.

  evidence:
  - reference: PMID:36996098
    reference_title: Therapeutic effect and mechanism of Daikenchuto in a model of methotrexate-induced acute small intestinal mucositis.
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    quote_role: PRIMARY_RESULT
    snippet: "Intraperitoneal injection of 10 mg/kg methotrexate (MTX) every 3 days for a total of 3 doses was used for induction of CIM in a rat model."
    explanation: >-
      Documents the induction protocol that defines this preparation as a
      methotrexate model.
discussions:
- discussion_id: pneumonitis_mechanism_unknown
  kind: KNOWLEDGE_GAP
  attaches_to:
  - phenotypes#Interstitial pneumonitis
  prompt: >-
    What produces methotrexate pneumonitis, given that it is commoner on weekly
    low-dose therapy than after high-dose infusion?
  rationale: >-
    Every other toxicity in this entry scales with concentration and duration,
    which is why the pathograph is built around exposure and clearance.
    Pneumonitis inverts that relationship: it is reported in about 0.5% of
    patients per year on weekly low-dose methotrexate and is rare after
    high-dose infusion. An inverted dose-response is not a minor discrepancy,
    it is evidence that the lesion does not belong to the folate arm at all,
    and hypersensitivity has long been the favoured alternative. The entry
    deliberately leaves the phenotype unattached to any pathophysiology node
    rather than inventing an edge from the folate block, because an edge drawn
    to tidy up the graph would assert exactly the concentration dependence the
    epidemiology contradicts.

- discussion_id: nsaid_interaction_contested
  kind: KNOWLEDGE_GAP
  attaches_to:
  - environmental#Co-administration of non-steroidal anti-inflammatory drugs
  prompt: >-
    Do NSAIDs and proton pump inhibitors clinically delay methotrexate
    elimination in humans, or is the interaction a mouse finding that has been
    carried into practice without human confirmation?
  rationale: >-
    The two evidence items on this exposure's mechanism link disagree, and the
    disagreement is recorded rather than resolved. The mechanistic
    demonstration is a mouse study showing indomethacin and ketoprofen inhibit
    renal tubular methotrexate transporters. A 2023 transporter-focused review
    of the clinical literature concludes there is no firm evidence of an
    interaction with NSAIDs or proton pump inhibitors in humans, and notes that
    most reports are case reports or series. Protocols continue to substitute
    these drugs before high-dose infusion, which is defensible as asymmetric
    risk but is not the same as the interaction being established. What would
    settle it is a controlled pharmacokinetic study in patients rather than
    more case reports.

- discussion_id: mthfr_effect_size_uncertain
  kind: KNOWLEDGE_GAP
  attaches_to:
  - genetic#MTHFR
  prompt: >-
    How large is the MTHFR C677T effect on methotrexate toxicity, and does it
    replicate outside single-centre oncology cohorts?
  rationale: >-
    The association cited here comes from 148 courses in 32 patients with one
    tumour type, and the nephrotoxicity odds ratio carries a
    confidence interval from 1.65 to 103.86. An interval spanning nearly two
    orders of magnitude is compatible with a large effect and with a barely
    detectable one. The entry records the gene as a susceptibility modifier on
    that basis but does not support using the genotype to predict toxicity in
    practice, which is the use the source proposes.

clinical_burden:
  burden_level: VARIABLE
  rationale: >-
    The burden is concentrated in a minority of exposures and is almost entirely
    preventable, which is what makes it a supportive-care problem rather than a
    pharmacological one. Acute kidney injury complicates 2% to 12% of high-dose
    courses. The deaths are mostly not renal: they follow from the cytopenia and
    the open mucosal barrier arriving together, so the proximate cause is
    usually sepsis. Beyond the acute episode, an interrupted or delayed
    chemotherapy course is itself a harm, because the regimens methotrexate
    belongs to are curative ones.
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "AKI and other toxicities of high-dose methotrexate can lead to significant morbidity, treatment delays, and diminished renal function."
    explanation: >-
      Names the three burdens claimed: acute morbidity, interruption of the
      cancer treatment, and residual loss of renal function.
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "Grade IV mucositis is an oncologic emergency and is associated with infections, the need for parenteral nutrition, increased use of health care resources, delayed chemotherapy, and even death"
    explanation: >-
      Supports treating the mucosal lesion as a driver of mortality rather than
      a comfort issue.

progression:
- phase: High-dose infusion and immediate post-infusion period
  notes: >-
    Hydration and alkalinisation are running, levels are drawn on protocol, and
    the question being asked is only whether elimination is on the expected
    curve. Nothing is yet clinically apparent.
- phase: Delayed elimination
  notes: >-
    A level above the nomogram, a rising creatinine, or a falling urine output.
    This is the decision point of the whole entry: intervention here is
    leucovorin escalation and possibly glucarpidase, and the alternative is the
    feedback loop closing.
- phase: Established organ toxicity
  notes: >-
    Mucositis and the cytopenia nadir overlap, and that overlap is when sepsis
    risk peaks. Specific day ranges are deliberately not given: the figures in
    circulation for them are not stated by any reference this entry cites.
- phase: Recovery
  notes: >-
    Usually complete once the drug is cleared and the proliferative
    compartments repopulate, and subsequent high-dose courses can generally be
    given safely after recovery even when the episode was severe.
  evidence:
  - reference: PMID:27496039
    reference_title: Preventing and Managing Toxicities of High-Dose Methotrexate.
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    quote_role: REVIEW_SYNTHESIS
    snippet: "However, even when toxicity is severe, subsequent HDMTX courses can generally be administered safely after the patient recovers"
    explanation: >-
      Supports the claim that a severe episode does not normally preclude
      further treatment, which is the prognostic point of this phase.

notes: >-
  Scope. This entry covers methotrexate toxicity as a drug-induced condition
  across its dose regimens, not the therapeutic use of the drug. Where a claim
  is regimen-specific it says so, because several of them invert between the
  two: hepatic fibrosis and pneumonitis belong to chronic weekly low-dose
  exposure, while crystal nephropathy and delayed elimination belong to
  high-dose infusion.

  No GeneReviews chapter exists for this disease, which is expected for a drug
  toxicity rather than a Mendelian disorder; `just check-genereviews --online`
  reports NO_CHAPTER for both GeneReviews and StatPearls against the entry's
  current name.

  Myocarditis was dropped after the deep-research report offered HP:0001636 for
  it, a CURIE that is in fact Tetralogy of Fallot; the phenotype itself is
  case-report level and was not re-sourced.

  An earlier draft of this note claimed third-space fluid sequestration had been
  left out for want of a citable source. That was false, and the entry's own
  cached reference refutes it: PMID:27496039 names third spacing as a cause of
  delayed elimination and effusion drainage as a countermeasure. Both are now
  curated, as an environmental exposure and a treatment respectively. The note
  is left here rather than deleted because a false negative-sourcing claim tells
  the next reader not to look, which is the specific harm it does.

📚

References & Deep Research

References

PMID:27496039

Preventing and Managing Toxicities of High-Dose Methotrexate.

No top-level findings curated for this source.

PMID:36071163

Methotrexate recognition by the human reduced folate carrier SLC19A1.

No top-level findings curated for this source.

PMID:12374095

The role of dUTPase and uracil-DNA repair in cancer chemotherapy.

No top-level findings curated for this source.

PMID:25172240

Clinical characteristics and risk factors for low dose methotrexate toxicity: a cohort of 28 patients.

No top-level findings curated for this source.

Deep Research

Deep research results are used as seeds for research; they do not undergo the same validation as the main records and may contain errors. How we use deep research.

Evaluations and curation notes (2)

Record notes

Scope. This entry covers methotrexate toxicity as a drug-induced condition across its dose regimens, not the therapeutic use of the drug. Where a claim is regimen-specific it says so, because several of them invert between the two: hepatic fibrosis and pneumonitis belong to chronic weekly low-dose exposure, while crystal nephropathy and delayed elimination belong to high-dose infusion. No GeneReviews chapter exists for this disease, which is expected for a drug toxicity rather than a Mendelian disorder; `just check-genereviews --online` reports NO_CHAPTER for both GeneReviews and StatPearls against the entry's current name. Myocarditis was dropped after the deep-research report offered HP:0001636 for it, a CURIE that is in fact Tetralogy of Fallot; the phenotype itself is case-report level and was not re-sourced. An earlier draft of this note claimed third-space fluid sequestration had been left out for want of a citable source. That was false, and the entry's own cached reference refutes it: PMID:27496039 names third spacing as a cause of delayed elimination and effusion drainage as a countermeasure. Both are now curated, as an environmental exposure and a treatment respectively. The note is left here rather than deleted because a false negative-sourcing claim tells the next reader not to look, which is the specific harm it does.

Create: Methotrexate Toxicity · 2026-09-22T02:46:29Z · View source

New entry for methotrexate toxicity (MONDO:0034212), curated de novo. Deep research: claude_code provider, 17 web searches, 57 citations (research/Methotrexate_Toxicity-deep-research-claude_code.md). The run was launched with --validate-references and --validate-terms but emitted neither section, so both were retro-fitted with just validate-research-reference and just validate-research-terms. References: 38/38 resolved, 0 unresolved, 0 off topic. Terms: the report offered HP:0001636 for "Myocarditis", which is Tetralogy of Fallot. just preflight-dr returned SKIP because MONDO records no causal gene for a drug toxicity, so identity was checked manually against the MONDO record (parent MONDO:0800390 chemotherapy-induced toxicity) and the report's gene profile (MTHFR, DHFR, SLCO1B1, ABCB1, ATIC). Structure. Two arms off one exposure node. The folate arm runs SLC19A1 uptake -> polyglutamation -> DHFR inhibition -> tetrahydrofolate depletion, branching to thymidylate blockade (thence uracil misincorporation) and to de novo purine blockade, converging on replicating cell death and then on marrow, gut, skin and liver. The pharmacokinetic arm runs tubular crystallisation -> three independently sufficient injury limbs (obstruction, direct epithelial toxicity, afferent arteriolar vasoconstriction) -> impaired renal elimination, which closes a deliberate cited cycle back onto crystallisation and amplifies the tissue nodes. 22 pathophysiology nodes, 16 phenotypes (15/16 causally connected), 6 treatments, 3 diagnostics, 2 susceptibility genes, 5 environmental exposures, 1 animal model, 1 mechanistic hypothesis, 3 knowledge gaps. Evidence. 91/91 snippets verified; each was additionally checked character-exact against references_cache before being written, which is stricter than the validator. Every reference_title was read off the cache frontmatter. Backbone is PMID:27496039 (full text cached); also PMID:36071163, PMID:2416193, PMID:12374095, PMID:25172240, PMID:32961024, PMID:32066940, PMID:36946211, PMID:36996098, PMID:36582751, PMID:36564450, PMID:33069634 and PMID:23728635. Three disagreements are recorded rather than resolved, each as a KNOWLEDGE_GAP discussion: pneumonitis (dose-response inverted relative to every other toxicity here, so the phenotype is deliberately left unattached to any node); the NSAID interaction (a SUPPORT from a mouse transporter study and a REFUTE from PMID:36946211 on the same link); and the MTHFR effect size (nephrotoxicity OR 13.08, CI 1.65-103.86). A fourth and fifth REFUTE record the folate/stomatitis null result and the absence of any rationale for therapeutic drug monitoring in low-dose toxicity. Errors found and fixed during curation. The DR report's HP:0002317 for leukoencephalopathy (correct: HP:0002352) and HP:0001394 for hepatic fibrosis (correct: HP:0001395) were wrong; caught by looking every CURIE up. A draft notes: field claimed ECTO had no term for the interacting-drug exposure; re-running the searches per dismech-terms step 3a returned ECTO:2000011 exposure to NSAIDs, so the entry was rescoped and bound rather than shipping a false negative-existence claim. A draft waiver misreported what l~fluid and l~acid return and was rewritten to match the actual output. Four frequency bands were corrected or removed against the figures cited beside them (Pancytopenia VERY_FREQUENT -> FREQUENT on 78.5%; bands dropped on Oral ulcer, Acute kidney injury and Interstitial pneumonitis). PMID:33799927 turned out to be about serum amyloid A in sarcoidosis and was discarded. HP:0000951's label was written from memory as "Abnormal skin morphology"; it is "Abnormality of the skin". A pre-PR adversarial review by a fresh-context subagent found 18 further defects, all fixed in this same pass. Three were blocking. First, a notes: paragraph claimed third-space fluid sequestration had been omitted for want of a citable source - false, and refuted by a reference the entry already cited 50 times; third spacing is now an environmental exposure and effusion drainage a treatment, and the note records the correction rather than being deleted. Second, sepsis was named twice in prose as the mortality mechanism with no node, no phenotype and no edge; it is now a phenotype with converging edges from marrow failure and mucosal breakdown, evidenced by PMID:25172240's own mortality data. Third, this details field was a placeholder: a regex substitution failed silently and validate-history passed it because the schema only requires a string. Of the non-blocking findings: GO:0008517 (folic acid transporter) was the wrong sibling for SLC19A1 and was rebound to GO:0008518, whose definition names both the carrier and methotrexate; CL:1000838 manufactured a proximal convoluted segment the source does not name and was widened to CL:1000507; the three-limb nephropathy node was split into three, as its own description conceded it should be; a direct exposure-to-white-matter edge was added so the neurologic arm no longer hangs solely on the adenosine hypothesis; the cycle edge gained the concentration-dependence quote that actually closes the loop; dermatologic toxicity was added; an uncited apoptosis binding was dropped; two BACKGROUND quote_roles were added; and several prose claims were corrected where they asserted more than the source (crypt turnover interval, encephalopathy resolution timeframe, single-centre attribution, NAFLD risk reattribution, the definition of thymineless death, and progression day ranges that appear in no cited reference). Validation. just validate (91/91 snippets), validate-terms, check-entity-refs, check-causal-targets, check-duplicate-keys, check-qualifier-terms, check-enum-values, check-coarse-phenotypes, check-reference-titles, check-snippet-grading, check-environmental-evidence, prose-figure-audit and list-disconnected-phenotypes all run and read. No GeneReviews chapter exists (check-genereviews --online: NO_CHAPTER).

Claude Code ▸

Methotrexate Toxicity — Comprehensive Research Report

claude-haiku-4-5-20251001, claude-sonnet-5 57 citations 2026-09-22T02:23:21.988310

Methotrexate Toxicity — Comprehensive Research Report

1. Disease Information

Overview. Methotrexate (MTX) toxicity is a drug-induced condition arising from excess systemic or local exposure to methotrexate, an antifolate antimetabolite used at low doses for rheumatoid arthritis, psoriasis, inflammatory bowel disease, and ectopic pregnancy, and at intermediate-to-high doses for hematologic malignancies (acute lymphoblastic leukemia, primary CNS lymphoma, osteosarcoma) and as an abortifacient/antineoplastic agent. Toxicity is classified as a complex, iatrogenic/pharmacogenomic disorder rather than a single-gene Mendelian disease: the phenotype depends on dose, route (oral, IV, intrathecal), renal function, hydration status, third-space fluid accumulation, drug interactions, and inherited variation in folate-pathway and drug-transporter genes. It ranges from mild mucositis/cytopenia in low-dose chronic regimens to fulminant multiorgan failure after high-dose (HD-MTX) or intrathecal overdose. "It particularly affects the bone marrow, skin, liver, kidney, bone, lung, and brain" (MalaCards synthesis) (MalaCards: Methotrexate Toxicity).

Key identifiers: - MONDO: MONDO:0034212 - Orphanet: ORPHA:565782 ("Methotrexate toxicity") (Orphanet) - ICD-10-CM: T45.1X1A (accidental poisoning by antineoplastic/immunosuppressive drugs, initial encounter), T45.1X5A (adverse effect), Y14 (poisoning by drug/medicament, unspecified intent) (ICD10Data T45.1X1A, T45.1X5A) - MeSH: related to D008727 (Methotrexate) with adverse-effect qualifiers - Pharmacogenomic entries: ClinVar carries MTHFR c.665C>T (p.Ala222Val, rs1801133) annotated for "methotrexate response – Toxicity" (ClinVar RCV001847567.2)

Synonyms: methotrexate intoxication, methotrexate overdose, high-dose methotrexate (HD-MTX) delayed elimination/delayed methotrexate elimination (DME), methotrexate-induced pancytopenia, methotrexate-induced nephrotoxicity/crystal nephropathy, methotrexate-associated leukoencephalopathy.

Data source note: Understanding of this entity is derived from a mix of aggregated resources (FAERS pharmacovigilance database, national adverse-drug-reaction registries) and individual-patient sources (case reports/series, pediatric oncology cohort studies, single-center retrospective reviews) — there is no dedicated population disease registry, so most quantitative "risk factor" data come from oncology or rheumatology cohorts rather than population-representative EHR data.

2. Etiology

Disease causal factors

Methotrexate toxicity is fundamentally mechanistic/iatrogenic: it results from methotrexate (or its metabolite 7-hydroxymethotrexate) accumulating to supratherapeutic intracellular or plasma concentrations, disproportionate to the intended therapeutic antifolate effect. It is not caused by an intrinsic disease process but by a mismatch between dose/route and the patient's capacity to distribute, metabolize, and excrete the drug.

Risk factors

Genetic risk factors (pharmacogenomic modifiers of clearance/toxicity, not causal mutations): - SLCO1B1 (encodes OATP1B1, a hepatic uptake transporter) — the SNP rs4149056 (c.521T>C) is the most replicated genome-wide hit for delayed MTX clearance; it was independently replicated in a genome-wide study of clearance and associated with higher serum MTX at 48/72 h and elimination delay (Ramsey et al., ScienceDirect; Schulte 2021, Clin Transl Sci). - MTHFR (methylenetetrahydrofolate reductase) — rs1801133 (677C>T): T-allele carriers have significantly higher risk of hematopoietic toxicity vs. CC genotype; rs1801131 (1298A>C): CC/AC genotypes show lower toxicity frequency than AA (Frontiers in Oncology 2021; Sci Rep 2025). - ABCB1 (P-glycoprotein/MDR1, an efflux transporter that pumps MTX out of cells) — polymorphisms influence hematopoietic toxicity risk in adults with hematologic malignancies receiving HD-MTX (PMC8739189). - ATIC (AICAR transformylase, rs2372536, 347C>G) — included in meta-analyses of MTX toxicity association, relevant because MTX polyglutamates inhibit ATIC, driving adenosine accumulation. - A case report links MTHFR polymorphism combined with a drug interaction to severe hematologic toxicity after low-dose MTX for ectopic pregnancy, illustrating gene–drug interaction in a non-oncology setting (PMC12528151). - Genotyping of MTHFR 677C>T and/or ABCB1 pretreatment is proposed as a means to tailor HD-MTX dosing, though clinical adoption is not standardized.

Environmental / clinical risk factors: - Renal impairment / reduced GFR — the single strongest determinant of MTX pharmacokinetics; MTX is cleared predominantly by renal excretion, and its own nephrotoxicity can self-perpetuate a vicious cycle of reduced clearance → higher levels → more nephrotoxicity (PMC10642233). - Dehydration / inadequate hydration and urine alkalinization during HD-MTX. - Third-space fluid accumulation (pleural effusion, ascites, edema) acts as a reservoir causing delayed/biphasic clearance and redistribution toxicity, including cases of encephalopathy from a loculated parapneumonic effusion and cardiotoxicity-associated pleural effusions recapturing MTX after apparent clearance (CHEST 2024; PMC13417858). - Drug–drug interactions: NSAIDs (inhibit OAT-mediated renal tubular secretion), PPIs (compete for BCRP efflux transporter, reducing renal/biliary elimination), trimethoprim-sulfamethoxazole (dual DHFR inhibition + tubular secretion competition), penicillins, and other weak organic acids (SPS NHS; PMC13235338). - Advanced age, hypoalbuminemia, folate deficiency, dosing errors (daily instead of weekly dosing in low-dose regimens), infection, dialysis-dependence — consistently identified across low-dose MTX pancytopenia cohorts (ScienceDirect cohort of 28 patients; PMC13511332 — dialysis patients had ~1 in 3 risk of serious adverse event or death within 90 days of starting low-dose MTX). - Concurrent renal-toxic agents and extremes of urine pH favoring MTX/7-OH-MTX crystallization.

Protective factors: - Folic acid supplementation in chronic low-dose regimens: reduces GI adverse effects by ~9%, elevated LFTs by ~16%, and drug discontinuation by ~15%, without abrogating efficacy (Medscape; PubMed 15965822). - Leucovorin (folinic acid) rescue after HD-MTX — see Treatment section. - Vigorous hyperhydration and urinary alkalinization (urine pH >7) during HD-MTX — reduces crystal nephropathy risk by keeping MTX and 7-OH-MTX in solution. - Possible protective MTHFR genotypes (e.g., 1298 CC/AC) associated with lower toxicity risk in some cohorts.

Gene–environment interactions: The clearest example is MTHFR/ABCB1/SLCO1B1 genotype combined with concurrent nephrotoxic or transporter-competing co-medications (NSAIDs, PPIs, trimethoprim) or with pre-existing renal impairment — genetic slow-clearance variants convert an otherwise tolerable interaction into severe or fatal toxicity, as demonstrated in the ectopic-pregnancy case report combining MTHFR variant status with a drug interaction (PMC12528151).

3. Phenotypes

Methotrexate toxicity is a multi-system toxidrome. Below, phenotypes are grouped by organ system with approximate onset timing, severity range, and suggested HPO terms.

Phenotype	Type	Onset/Course	Frequency/Severity	Suggested HPO term
Pancytopenia / bone marrow suppression	Laboratory abnormality	Days after dose (low-dose chronic) or ~1–2 weeks (HD-MTX); can be delayed and prolonged	Most common manifestation of low-dose toxicity — reported in ~78.5% of low-dose toxicity cases in one series; mortality up to 25–44% in fatal cohorts, chiefly via sepsis	HP:0001876 (Pancytopenia)
Oral/GI mucositis, ulcerative stomatitis	Sign/symptom	Onset 3–7 days post HD-MTX; self-limited, <2 weeks	Very frequent (leading FAERS-reported reaction)	HP:0000164 (Abnormality of the oral cavity) / HP:0034754-like ulcerative stomatitis
Nausea, vomiting, diarrhea, abdominal distress	Symptom	Acute, dose-related	Common	HP:0002015 (Vomiting), HP:0002014 (Diarrhea)
Acute kidney injury / crystal nephropathy	Clinical sign/lab abnormality	Acute, within hours to days of HD-MTX	Central to delayed elimination cascade	HP:0001919 (Acute kidney injury)
Hepatotoxicity (acute transaminitis; chronic fibrosis/cirrhosis with prolonged low-dose use)	Laboratory abnormality / clinical sign	Acute: days; Chronic: after ≥2 years and cumulative dose ≥1.5 g	Acute reversible chemical hepatitis common; chronic fibrosis rarer but potentially fatal	HP:0002910 (Elevated hepatic transaminase), HP:0001394 (Hepatic fibrosis)
Pneumonitis / interstitial lung disease	Clinical sign	Variable, can be subacute	Incidence 0.3–14% across cohorts	HP:0006515 (Interstitial pneumonitis)
Leukoencephalopathy / stroke-like episodes / encephalopathy	Clinical sign, often on MRI	Acute-subacute (intrathecal or HD-MTX with CNS penetration; can mimic stroke)	Reported particularly in pediatric ALL and intrathecal overdose	HP:0002317 (Leukoencephalopathy) — see also mimicking HP:0001297 (Stroke)
Skin/mucosal toxicity (rash, mucocutaneous ulceration)	Physical sign	Days, can be severe in low-dose error cases	Reported with invasive secondary infection (e.g., mucormycosis) in severe cases	HP:0000988 (Skin rash)
Myocarditis / cardiotoxicity, transudative pleural effusion	Clinical sign	Case-report level, rare	Rare	HP:0001636 (Myocarditis)
Chemical conjunctivitis	Sign	Acute	Uncommon	HP:0000509 (Conjunctivitis)
Fever	Sign	Accompanies pancytopenia/sepsis	Common in severe presentations	HP:0001945 (Fever)
Bleeding/purpura (from thrombocytopenia)	Sign	Subacute	Seen in severe pancytopenia	HP:0000978 (Bruising susceptibility)

Quality-of-life impact: Mucositis and GI symptoms substantially impair oral intake and daily function during acute episodes; pancytopenia-associated infection/sepsis is the dominant driver of morbidity and mortality; chronic hepatic fibrosis and pulmonary fibrosis (in long-term low-dose regimens) can produce lasting functional impairment. Neurotoxic leukoencephalopathy, when it occurs, can leave residual cognitive/motor deficits, particularly in pediatric ALL survivors, and diagnostic ambiguity with acute stroke can trigger inappropriate thrombolysis, itself a risk (PMC10834224).

4. Genetic/Molecular Information

Because methotrexate toxicity is not a monogenic disease, this section covers pharmacogenomic modifiers rather than disease-causing mutations.

Causal "gene": none in the Mendelian sense; the causal agent is the drug itself acting on its target enzyme, DHFR (dihydrofolate reductase, HGNC:2861), and downstream folate-pathway enzymes.
Modifier genes and variant detail:
MTHFR (HGNC:7436) — rs1801133 (677C>T, p.Ala222Val) and rs1801131 (1298A>C, p.Glu429Ala); missense variants reducing enzyme activity and altering intracellular folate pools, modifying (not causing) toxicity risk. ClinVar records rs1801133 under "methotrexate response – Toxicity" (ClinVar).
SLCO1B1 (HGNC:10959) — rs4149056 (c.521T>C, p.Val174Ala), a missense variant reducing OATP1B1 hepatic uptake transporter function, associated with elevated 48h/72h MTX levels and elimination delay in genome-wide analyses (ScienceDirect GWAS).
ABCB1/MDR1 (HGNC:40) — encodes P-glycoprotein efflux transporter; polymorphisms (e.g., in exons 21/26, commonly studied SNPs 1236C>T, 2677G>T/A, 3435C>T) modify hematopoietic toxicity risk.
ATIC (HGNC:794) — rs2372536 (347C>G); MTX polyglutamates directly inhibit the ATIC gene product, so variants here modulate the resulting adenosine surge implicated in toxicity/efficacy.
Allele frequency / population data: Not systematically reported for this entity; MTHFR 677T allele frequency and SLCO1B1 521C allele frequency vary by ancestry and would be sourced from gnomAD/1000 Genomes for population context (not disease-specific).
Somatic vs. germline: All pharmacogenomic variants relevant here are germline; toxicity is not driven by somatic mutation.
Functional consequences: DHFR inhibition by MTX is the central molecular lesion (competitive, high-affinity inhibition blocking dihydrofolate → tetrahydrofolate conversion); polyglutamated MTX additionally inhibits thymidylate synthase (TYMS) and AICAR transformylase (ATIC), and to a lesser extent other folate-dependent enzymes, compounding antifolate effects at high intracellular concentration.
Epigenetic information: Not a primary driver of this entity; folate-pathway disruption itself affects DNA methylation capacity (via impaired S-adenosylmethionine regeneration, a downstream consequence of folate/homocysteine cycle disruption) — this is a consequence pathway rather than a causal epigenetic lesion, and is invoked in proposed mechanisms of MTX neurotoxicity (homocysteine/NMDA-receptor toxicity).
Chromosomal abnormalities: Not applicable; this is a toxic/pharmacologic disorder rather than a structural genomic disease.

5. Environmental Information

Environmental/pharmacological factors: The "environmental exposure" here is the drug itself and co-administered agents. Key interacting exposures:
NSAIDs — inhibit renal organic anion transporters (OAT), reducing MTX renal clearance, especially at high MTX doses (SPS NHS).
Proton pump inhibitors (PPIs) — compete with MTX for BCRP efflux transporters, reducing renal and biliary elimination (Pharmacy Times; PMC13235338).
Trimethoprim-sulfamethoxazole — additive DHFR inhibition plus tubular secretion competition, a well-documented cause of severe pancytopenia when combined with MTX.
Penicillins and other weak organic acids — compete for renal tubular secretion.
Alcohol — compounds hepatotoxicity risk in chronic low-dose regimens.
Lifestyle factors: Poor oral hydration/fluid intake during HD-MTX cycles increases nephrotoxicity risk; adherence errors (e.g., inadvertent daily dosing instead of weekly dosing of low-dose oral MTX) are a well-recognized and preventable cause of severe/fatal toxicity.
Infectious agents: Not a cause of the toxicity itself, but secondary opportunistic infection (bacterial sepsis, and rarely invasive fungal infection such as pulmonary mucormycosis) is a major downstream consequence of MTX-induced myelosuppression/mucosal barrier breakdown and a leading cause of death (Frontiers case report — pulmonary mucormycosis after low-dose MTX toxicity).

6. Mechanism / Pathophysiology

Causal chain (ordered)

Methotrexate (and active polyglutamated metabolites) enter cells via reduced folate carrier/folate receptors and are retained intracellularly as methotrexate polyglutamates (MTX-PG).
MTX and MTX-PG competitively and with high affinity inhibit dihydrofolate reductase (DHFR), blocking the reduction of dihydrofolate (DHF) to tetrahydrofolate (THF) — this is the primary, directly demonstrated molecular lesion (Frontiers/PMC synthesis).
THF depletion leads to (a) impaired one-carbon transfer reactions required for thymidylate synthesis (via inhibited thymidylate synthase cofactor supply) and purine synthesis (via inhibited glycinamide ribonucleotide transformylase and AICAR transformylase/ATIC), and (b) impaired methionine/homocysteine cycling (via inhibited methionine synthase cofactor supply) — this branch point produces two downstream consequences.
Branch A — impaired DNA/RNA synthesis → arrests cells in S-phase, disproportionately affecting rapidly dividing tissues: bone marrow precursors, GI mucosal epithelium, hair follicles → this leads to myelosuppression (pancytopenia) and mucositis/enterocolitis (this is directly demonstrated in rodent models and human marrow/mucosal biopsies).
Branch A also produces locally elevated oxidative stress (reactive oxygen species), which leads to apoptosis via caspase-3/caspase-8 activation and nitrosative stress in intestinal epithelium — demonstrated in rat intestinal mucositis models showing villus atrophy, crypt loss, and elevated cleaved caspase-3/8 (Naunyn-Schmiedeberg's Arch Pharmacol 2024; PMC — Wnt/β-catenin enterocyte turnover rat model).
Branch B — ATIC inhibition → accumulation of AICA-ribotide → leads to increased extracellular adenosine release, an anti-inflammatory but also neuro-vasoactive mediator implicated both in MTX's therapeutic anti-inflammatory action in rheumatoid arthritis and in acute methotrexate-related encephalopathy via adenosine-mediated vasodilation and cerebral edema (mechanism is inferred from clinical/pharmacologic correlation, not fully demonstrated causally in humans) (Frontiers/BJP 2003).
Branch B also causes impaired remethylation of homocysteine → elevated homocysteine, which is proposed (inferred, not conclusively proven) to exert direct endothelial toxicity and excitatory NMDA-receptor-mediated neurotoxicity, contributing to white-matter injury and leukoencephalopathy (Neurotoxic effects review, PMC11981439).
In parallel, independent of the folate-inhibition cascade — a pharmacokinetic branch: MTX and its hepatic metabolite 7-hydroxymethotrexate (7-OH-MTX) are renally excreted largely unchanged; at high concentrations and especially in acidic or low-volume urine, both precipitate as crystals within renal tubular lumens, causing direct mechanical tubular obstruction and direct pharmacological/cytotoxic tubular cell injury (S-phase arrest of tubular epithelial cells, plus 7-OH-MTX-induced oxidative stress and apoptosis in tubular cells) → this leads to acute kidney injury/crystal nephropathy (NEJM correspondence; PMC9293278; Impact of MTX/7-OH-MTX on renal toxicity, PMC11739129).
Acute kidney injury then reduces renal MTX clearance, which feeds back to raise systemic MTX exposure further — a self-amplifying vicious cycle that is the pathophysiologic core of "delayed methotrexate elimination (DME)," the syndrome underlying most severe/fatal HD-MTX toxicity.
Independently, third-space fluid collections (pleural effusion, ascites) act as a redistribution reservoir: MTX partitions into these compartments during the infusion/high-plasma-level phase and releases back into the circulation later, producing delayed or biphasic rebound toxicity even after apparent initial clearance — demonstrated in multiple case reports where thoracentesis/effusion drainage produced rapid clinical and biochemical improvement (CHEST 2024; PMC13417858).
The combined consequences of Branches A, B, and the renal/pharmacokinetic branch converge on the clinical syndrome: pancytopenia + sepsis risk, mucositis, hepatotoxicity, pneumonitis, and (in CNS-exposed patients) leukoencephalopathy, with AKI-driven impaired clearance determining overall severity and duration.

Category detail

Molecular pathways: One-carbon/folate metabolism (KEGG hsa00670), purine biosynthesis, pyrimidine (thymidylate) biosynthesis, methionine/homocysteine cycle.
Cellular processes: Cell-cycle S-phase arrest; apoptosis (caspase-3/-8 dependent); oxidative/nitrosative stress; microglial activation and oligodendrocyte precursor depletion (proposed mechanism for demyelination in leukoencephalopathy) (PMC11981439).
Protein dysfunction: DHFR is not structurally altered but is pharmacologically inhibited (competitive antagonism at the folate-binding site) — a functional rather than structural lesion.
Metabolic changes: Reduced tetrahydrofolate pools; elevated homocysteine; accumulation of AICA-ribotide and extracellular adenosine; possible hyperammonemia in severe cases (via purine-cycle disruption, less well established).
Immune system involvement: Adenosine-mediated immunosuppression is central to MTX's therapeutic anti-inflammatory action in RA but is mechanistically linked (via ROS generation in monocytoid/cytotoxic T-cell lines) to lymphocyte toxicity (BJP 2003).
Tissue damage mechanisms: Oxidative stress and apoptosis (GI mucosa); crystal-mediated tubular obstruction plus direct cytotoxicity (kidney); presumed blood-brain-barrier cumulative toxic injury (CNS).
Biochemical abnormalities: DHFR inhibition; secondary thymidylate synthase and ATIC pathway disruption; hyperhomocysteinemia.
Molecular profiling: Gut microbiota composition is significantly altered in a time-dependent manner following MTX administration in mouse models, implicating dysbiosis in GI toxicity (PubMed 30049384).
Suggested ontology terms: GO:0004146 (dihydrofolate reductase activity), GO:0006281 (DNA repair)/GO:0006260 (DNA replication) for S-phase arrest context, GO:0006915 (apoptotic process), GO:0034599 (cellular response to oxidative stress), GO:0006144 (purine nucleobase metabolic process); CL:0000075 (epithelial cell — intestinal/renal tubular), CL:0000097 (mast cell — not primary but relevant to mucosal injury context), CL:0000451 (dendritic cell)/CL:0000542 (lymphocyte) for immune involvement.

7. Anatomical Structures Affected

Organ level (primary): Bone marrow (hematopoietic suppression), gastrointestinal tract (mucosal ulceration/mucositis), kidney (crystal nephropathy/AKI), liver (hepatocellular injury/fibrosis), lung (pneumonitis/interstitial lung disease), central nervous system (leukoencephalopathy, encephalopathy). Secondary involvement: skin/mucous membranes (rash, ulceration), heart (rare myocarditis/pericardial-pleural effusion), eye (chemical conjunctivitis).
Body systems: Hematologic, digestive, renal/urinary, hepatic, respiratory, nervous, integumentary — essentially a multisystem toxidrome. UBERON terms: UBERON:0002371 (bone marrow), UBERON:0001007 (digestive system), UBERON:0002113 (kidney), UBERON:0002107 (liver), UBERON:0002048 (lung), UBERON:0000955 (brain), especially UBERON:0002316 (white matter of central nervous system) for leukoencephalopathy.
Tissue/cell level: Bone marrow hematopoietic progenitor cells; small intestinal crypt/villus enterocytes (villus atrophy, crypt loss documented in rat models); renal proximal tubular epithelial cells; hepatocytes; alveolar epithelium/interstitium; oligodendrocytes and microglia in CNS white matter. Cell Ontology: CL:0000015 (hematopoietic stem cell), CL:0000584 (enterocyte), CL:1000838 (kidney proximal convoluted tubule epithelial cell), CL:0000182 (hepatocyte), CL:0000128 (oligodendrocyte), CL:0000129 (microglial cell).
Subcellular level: Nucleus (DNA/RNA synthesis blockade); mitochondria (implicated in oxidative-stress-driven apoptosis). GO Cellular Component: GO:0005634 (nucleus), GO:0005739 (mitochondrion).
Localization: Renal toxicity is typically bilateral (both kidneys via systemic circulation); CNS leukoencephalopathy is often bilateral, periventricular/deep white matter — imaging pattern that can mimic bilateral acute ischemic stroke, prompting inappropriate thrombolysis in reported cases (PMC10834224).

8. Temporal Development

Onset: Depends entirely on regimen. Low-dose weekly oral/SC MTX toxicity (e.g., in RA) is typically insidious/subacute, emerging over weeks to months, often precipitated by an acute trigger (dosing error, new interacting drug, intercurrent renal impairment). HD-MTX toxicity is acute, unfolding over hours to days after infusion, with mucositis at day 3–7 and nephrotoxicity/pancytopenia nadir around day 7–14.
Progression: Follows a dose-and-clearance-dependent course rather than fixed disease stages. "Delayed methotrexate elimination" is the operative staging concept: normal clearance vs. mild/moderate delay (managed with leucovorin dose escalation) vs. severe delay with organ dysfunction (requiring glucarpidase). Progression rate is variable — rapid deterioration is possible with renal injury feedback loops or third-space redistribution.
Course pattern: Generally self-limited and reversible with prompt supportive care/rescue therapy in most cases (mucositis "usually...lasts less than 2 weeks"); can be relapsing/biphasic when third-space fluid redistributes drug back into circulation after apparent clearance, or when a second high-dose cycle is given without addressing the prior cause of delayed clearance. Chronic hepatotoxicity/pulmonary fibrosis from long-term low-dose regimens follows an indolent, cumulative-dose-dependent course over years.
Remission: Spontaneous resolution occurs once the drug is cleared/inactivated (endogenously or via glucarpidase) and marrow/mucosal/renal recovery proceeds; treatment-induced remission via leucovorin rescue and, when needed, glucarpidase.
Critical periods/windows for intervention: Leucovorin rescue must begin within 24–48 hours of MTX administration (or within 24 hours of completing an infusion) to be maximally effective, and continues until MTX level falls below 0.1 µM (BJCP review). Glucarpidase is most effective when given promptly once severe toxicity/AKI with markedly elevated MTX levels is recognized, reducing plasma MTX by ≥97% within 15 minutes (FDA approval summary via PMC).

9. Inheritance and Population

Epidemiology: No single population-wide prevalence/incidence figure exists because this is a treatment-emergent toxicity, not an independent disease with fixed population prevalence; incidence is reported per exposed cohort:
Interstitial lung disease/pneumonitis: 0.3–14% of MTX-exposed patients depending on cohort and dose.
Pancytopenia is the dominant manifestation of low-dose toxicity, detected in ~78.5% of cases in one low-dose toxicity case series (n=28) (ScienceDirect).
FAERS pharmacovigilance analysis (2004–2024) captured 130,818 MTX-related adverse event reports, with females representing 64.2% of reporters and adults 18–64.9 years the most frequent age group (Frontiers in Immunology 2025).
Among dialysis patients newly started on low-dose MTX, approximately 1 in 3 experienced a serious adverse event or death within 90 days (PMC13511332).
Mortality in major MTX-toxicity cohorts: reported figures range from ~12.9% (27/31 discharged, 4 died) to markedly higher (25–44%) in specifically pancytopenia-driven low-dose toxicity series, chiefly from sepsis.
Inheritance pattern: Not applicable in the Mendelian sense — this is an acquired, drug-induced condition modulated by common pharmacogenomic variants (MTHFR, SLCO1B1, ABCB1) inherited in the general population as ordinary polymorphisms rather than disease alleles. No single-gene inheritance pattern, penetrance, expressivity, anticipation, mosaicism, or founder-effect concept applies directly, though the modifying alleles themselves (e.g., MTHFR 677T) have population-specific allele frequencies (higher in some European and East Asian populations) that could stratify risk.
Population demographics: Affected populations mirror the populations receiving MTX therapeutically: RA/psoriasis/IBD patients (predominantly adult, female-skewed given RA epidemiology), pediatric and young-adult ALL/lymphoma patients (HD-MTX protocols), and dialysis/renal-impaired patients (disproportionately high-risk subgroup). No distinct geographic endemicity; distribution follows MTX prescribing patterns globally.
Sex ratio: FAERS data show a notable female predominance among adverse-event reporters (64.2%), likely reflecting the female predominance of rheumatoid arthritis rather than a sex-specific toxicity mechanism.
Age distribution: Bimodal in practice — pediatric/young-adult oncology patients receiving HD-MTX, and adult/elderly patients on chronic low-dose regimens (with elderly ≥65 years patients also showing increased risk of drug hypersensitivity per the FAERS analysis).

10. Diagnostics

Laboratory tests: - Serum/plasma MTX concentration via immunoassay or LC-MS/MS, monitored at defined intervals (commonly 24/48/72 hours) post-HD-MTX to guide leucovorin dosing and trigger escalation of care (JALM 2025). LOINC-coded methotrexate serum level assays exist for this purpose. - Complete blood count (for pancytopenia/myelosuppression), liver function tests (AST/ALT/bilirubin), serum creatinine/BUN and eGFR (renal function and monitoring for AKI), urine pH (to confirm alkalinization adequacy). - Critical assay caveat: after glucarpidase administration, a non-toxic MTX metabolite (DAMPA) cross-reacts with standard immunoassays, producing artifactually elevated apparent MTX levels; LC-MS/MS is required for accurate post-glucarpidase monitoring (PubMed 29994985; Pediatr Blood Cancer 2024). - Biomarkers: Homocysteine level (proposed neurotoxicity correlate); no FDA-qualified specific biomarker panel beyond MTX level itself and standard organ-function labs.

Imaging: Brain MRI (T2/FLAIR/diffusion-weighted imaging) for suspected leukoencephalopathy — shows restricted diffusion in white matter that can mimic acute ischemic stroke (Ann Child Neurol 2024; PMC10834224); chest radiograph/CT for pneumonitis; abdominal imaging to detect ascites/effusions as third-space reservoirs before HD-MTX.

Functional/electrophysiologic tests: EEG may show nonspecific slowing in MTX encephalopathy; nerve conduction studies have been explored for MTX-related peripheral/central neurotoxicity diagnostic challenges (PMC12269288).

Genetic testing: Not part of routine standard-of-care diagnosis of toxicity itself, but pretreatment MTHFR 677C>T / 1298A>C, SLCO1B1 rs4149056, and ABCB1 genotyping have been proposed (research/selected-center use) to risk-stratify patients before HD-MTX, particularly in hematologic malignancy protocols.

Clinical criteria: No formal DSM/ICD diagnostic algorithm exists beyond recognizing the clinical/laboratory toxidrome (cytopenia, mucositis, rising creatinine, elevated MTX level) in the context of known exposure. Diagnosis is fundamentally exposure-based (poisoning/adverse-effect coding under ICD-10 T45.1).

Differential diagnosis: For pancytopenia — other drug-induced marrow suppression, aplastic anemia, myelodysplastic syndrome (a reported longer-term hematologic sequela of chronic low-dose MTX, per a 2025 case report of therapy-related MDS (Oxford Med Case Reports)); for AKI — other nephrotoxic drug exposures, tumor lysis syndrome; for leukoencephalopathy — acute ischemic stroke (a well-documented diagnostic pitfall leading to inappropriate thrombolysis), PRES, other toxic/metabolic leukoencephalopathies.

Screening: Baseline CBC, renal function, liver function, and (in women of reproductive age) pregnancy testing before initiating MTX at any dose; ongoing periodic CBC/LFT/renal monitoring throughout chronic low-dose therapy is the principal "screening" strategy for evolving toxicity (AAFP guide to monitoring methotrexate).

11. Outcome / Prognosis

Mortality: Highly variable by setting. In severe low-dose MTX-induced pancytopenia cohorts, mortality has been reported as high as 25–44%, almost always due to sepsis/infection in the setting of neutropenia. In a broader major-toxicity cohort, 4/31 patients (12.9%) died. Dialysis patients starting low-dose MTX had markedly elevated 90-day risk of serious adverse events or death (~1 in 3) compared with hydroxychloroquine comparators.
Morbidity/functional outcomes: Most acute toxicity (mucositis, cytopenia, mild-moderate AKI) is reversible with prompt supportive care; chronic hepatic fibrosis/cirrhosis and pulmonary fibrosis from long-term low-dose therapy can cause lasting organ impairment. CNS leukoencephalopathy may leave residual neurocognitive deficits, particularly relevant in pediatric ALL survivors.
Complications: Secondary bacterial sepsis (leading cause of death in pancytopenic patients), invasive fungal infection (e.g., pulmonary mucormycosis reported after severe low-dose toxicity), myelodysplastic syndrome as a longer-term hematologic sequela, and diagnostic/iatrogenic complications such as inappropriate thrombolytic therapy given for MTX-leukoencephalopathy misdiagnosed as stroke.
Recovery potential: Excellent with early recognition and appropriate rescue therapy (leucovorin ± glucarpidase, dialysis for severe renal failure); recovery potential drops sharply once profound, prolonged pancytopenia with sepsis or severe multiorgan AKI supervenes.
Prognostic factors: WBC count at hospital admission has been identified as the most important prognostic factor for survival in low-dose MTX pancytopenia; degree/duration of renal impairment and time-to-initiation of rescue therapy (leucovorin/glucarpidase) are key prognostic determinants in HD-MTX toxicity; presence of third-space fluid collections predicts a more protracted/relapsing course.

12. Treatment

Pharmacotherapy / rescue agents

Leucovorin (folinic acid) rescue — the mainstay. Provides a reduced folate (tetrahydrofolate precursor) that bypasses the DHFR block. Standard HD-MTX rescue dosing: 10–25 mg/m² IV or IM every 6 hours for 72 hours; in renal compromise, higher dosing (150 mg/m² every 3 hours) is used. For overdose/unknown MTX level: up to 100 mg/m² IV. Must begin within 24–48 hours of MTX start (or ≤24 hours after infusion completion) and continue until MTX level <0.1 µM (J Clin Pharm Ther review; BJCP 2025). Caution: excessive leucovorin can reduce MTX's antitumor efficacy; repeated high-dose calcium folinate can cause hypercalcemia.
Glucarpidase (carboxypeptidase G2) — recombinant bacterial enzyme, FDA-approved 2012, cleaves ~99% of circulating MTX to inactive metabolites (DAMPA and glutamate) within 15 minutes; FDA-recommended dose ≥50 U/kg, reducing plasma MTX by ≥97% (PMC8756618). Indicated for HD-MTX patients with markedly elevated MTX and/or renal impairment refractory to hydration/alkalinization/leucovorin. Leucovorin should not be given within 2 hours before or after glucarpidase (leucovorin is also a glucarpidase substrate). NCIT: NCIT:C1872-style specific-agent binding pattern; glucarpidase itself would bind to an NCIT drug/agent term (e.g., analogous to therapeutic_agent slot) under treatment_term NCIT:C15986 (Pharmacotherapy) with therapeutic_agent for glucarpidase and separately for leucovorin/folinic acid.
NCIT clinical-intervention terms applicable: NCIT:C15986 (Pharmacotherapy) as the treatment_term for both leucovorin and glucarpidase administration; NCIT:C49236 (Therapeutic Procedure) may apply to hyperhydration/alkalinization protocols.

Supportive and procedural measures

Aggressive IV hyperhydration and urinary alkalinization (target urine pH ≥7) to maintain MTX/7-OH-MTX solubility and prevent crystal nephropathy — foundational to all HD-MTX protocols (European consensus, PMC11446969; Middle East consensus 2025).
Drainage of third-space fluid collections (thoracentesis/chest tube for pleural effusion, paracentesis for ascites) before HD-MTX administration when feasible, and emergently if delayed clearance/redistribution toxicity is identified — demonstrated to rapidly improve MTX kinetics and clinical status in case reports.
Renal replacement therapy (high-flux hemodialysis, sometimes combined with high-dose folinic acid) for severe AKI with markedly elevated MTX refractory to other measures (PMC3636355).
Discontinuation of interacting drugs (NSAIDs, PPIs, trimethoprim-sulfamethoxazole) is a critical, low-tech but essential intervention.
Pegylated G-CSF and supportive transfusion for severe myelosuppression (J Cancer Detect Diagn Res 2024 case report).

Experimental / investigational

Routine low-dose glucarpidase following HD-MTX in adult CNS lymphoma patients has been studied as a preventive strategy in a phase I open-label multi-center study (PMC8756618).
Preclinical/animal-model protective strategies for MTX-induced intestinal mucositis include mesenchymal stem cells plus glutathione, lutein, butyrate, and olmesartan, targeting oxidative stress and inflammatory pathways (PMC13358174; PMC3765170; PMC8617074; PMC4273993) — these are model-organism/in-vitro findings, not yet standard clinical practice.

Treatment algorithm (escalation logic)

Baseline risk stratification (renal function, third-space fluid, drug list review) before HD-MTX.
Hyperhydration + urinary alkalinization throughout infusion.
Scheduled leucovorin rescue starting per protocol timing.
Serum MTX level monitoring at defined intervals (e.g., 24/48/72 h) with leucovorin dose escalation if levels exceed nomogram thresholds.
If MTX remains markedly elevated despite hydration/alkalinization/leucovorin, or AKI develops → glucarpidase, with leucovorin timing adjusted around dosing, and switch to LC-MS/MS for subsequent MTX monitoring.
Dialysis for refractory severe renal failure.
Supportive care (transfusion, G-CSF, antimicrobial prophylaxis/treatment) for cytopenia-related complications.

13. Prevention

Primary prevention: Careful patient selection (avoid MTX or use extreme caution in renal impairment — MTX is contraindicated at creatinine clearance <50 mL/min per common guidance), pre-treatment drug reconciliation to avoid NSAID/PPI/trimethoprim co-prescription, correct dosing education (especially preventing inadvertent daily dosing of weekly low-dose oral regimens — a recurrent, preventable cause of severe toxicity), and drainage of pre-existing third-space fluid collections before HD-MTX.
Secondary prevention (early detection): Scheduled CBC, renal function, and liver function monitoring throughout MTX therapy (baseline plus periodic monitoring per AAFP/rheumatology guidance); serum MTX level monitoring at fixed intervals during HD-MTX; prompt recognition of early mucositis/cytopenia signs to trigger dose modification or rescue escalation.
Tertiary prevention: Leucovorin rescue and glucarpidase administration to prevent progression to organ failure once toxicity is identified; drainage of accumulating effusions to prevent relapse.
Folic acid supplementation in chronic low-dose regimens reduces incidence of GI adverse effects, LFT elevation, and treatment discontinuation, though no consensus dosing regimen exists (Medscape).
Genetic/risk stratification: Pretreatment MTHFR/ABCB1/SLCO1B1 genotyping has been proposed as a personalized-prevention strategy in HD-MTX oncology protocols, though not yet part of standard guidelines.
Public health / systems-level: Standardized institutional protocols (structured risk stratification for delayed elimination, standardized monitoring intervals, pharmacokinetically guided leucovorin dose adjustment, and criteria for early glucarpidase use) were the focus of 2024–2025 regional consensus statements (Middle East and European consensus groups), reflecting an active area of guideline development (Frontiers in Oncology 2025; European consensus, PMC11446969).
Counseling: Patient education regarding weekly (not daily) dosing for low-dose oral regimens, avoidance of alcohol and interacting OTC medications (NSAIDs), and prompt reporting of mucositis/fever/bruising.

14. Other Species / Natural Disease

Methotrexate toxicity is not a naturally occurring disease in animals but is extensively induced experimentally; it is also encountered clinically in veterinary oncology when MTX or related antifolates are used therapeutically in companion animals, though this is not well represented in the search results retrieved. No OMIA entry or spontaneous veterinary disease model was identified in this search; the "natural disease" concept does not apply here (this is a pharmacologic/toxic disorder, not a genetic disease with cross-species natural occurrence). Orthologous relevance is instead expressed through the conservation of the DHFR/folate pathway across mammals, which is why rodent models faithfully reproduce human-relevant intestinal, renal, and marrow toxicity (see Model Organisms below). No zoonotic or transmission relevance applies, as this is a drug toxicity, not an infectious or heritable disease.

15. Model Organisms

Model types: Predominantly rodent (rat and mouse) induced-toxicity models; no established knockout/transgenic genetic models specific to this toxicity (since it is drug-induced, not gene-driven), though DHFR and folate-pathway gene knockouts/hypomorphs in mice have been used in related folate-biology research.
Specific model systems:
Rat intestinal mucositis model: intraperitoneal MTX injection (a commonly cited regimen is ~10 mg/kg IP every 3 days for 3 doses, or single higher-dose regimens) reproducibly induces villus atrophy, crypt loss, and decreased crypt cell proliferation, maximal around day 3–5 post-injection, with recovery mediated by HGF/c-Met upregulation (PMC2374397; PMC4222808).
Mouse models of GI toxicity and gut microbiota disruption: demonstrate time-dependent alteration of gut microbial diversity and immune/macrophage response following MTX administration (PubMed 30049384).
Rat nitrosative-stress small intestinal injury model (Arch Toxicol 2008).
IEC-6 rat intestinal epithelial cell line (in vitro) used to study ROS-mediated apoptosis and protective agents such as lutein (PMC3765170).
Osteosarcoma cell lines used to study MTX resistance mechanisms via ROS/p53-mediated apoptosis modulated by estrogen-related receptor alpha (PMC4055217) — relevant to efficacy/resistance rather than toxicity per se, but mechanistically overlapping.
Rabbit model for MTX-ketoprofen (NSAID) pharmacokinetic interaction studies, directly modeling the human drug-interaction toxicity risk (PMC1971536).
Phenotype recapitulation: Rodent models faithfully reproduce human intestinal mucositis (villus atrophy, crypt apoptosis) and renal/hepatic oxidative injury, and are used to test protective/mitigating agents (mesenchymal stem cells + glutathione, Daikenchuto herbal formula, butyrate, olmesartan). They do not fully recapitulate the human leukoencephalopathy phenotype, and CNS-toxicity mechanisms remain comparatively less modeled.
Model limitations: Acute high-bolus IP dosing regimens in rodents do not perfectly mirror the pharmacokinetics of human HD-MTX infusion protocols with leucovorin rescue, and species differences in folate metabolism and renal handling of MTX/7-OH-MTX limit direct extrapolation, particularly for nephrotoxicity thresholds and third-space redistribution kinetics.
Research applications: These models are primarily used to test GI-mucositis mitigation strategies (probiotics/microbiota modulation, antioxidants, growth factors) and to dissect the ROS/apoptosis/inflammatory signaling underlying mucosal and renal injury, informing candidate adjunctive therapies (none yet FDA-approved beyond leucovorin/glucarpidase).

Summary of Suggested Ontology Terms for KB Curation

Category	Examples
HP	HP:0001876 (Pancytopenia), HP:0001919 (Acute kidney injury), HP:0002910 (Elevated hepatic transaminase), HP:0006515 (Interstitial pneumonitis), HP:0002317 (Leukoencephalopathy), HP:0000164 (oral cavity abnormality/mucositis), HP:0001945 (Fever)
GO (BP)	GO:0006281/GO:0006260 (DNA synthesis/replication context), GO:0006915 (apoptotic process), GO:0034599 (cellular response to oxidative stress), GO:0006144 (purine nucleobase metabolic process)
GO (MF)	GO:0004146 (dihydrofolate reductase activity)
CL	CL:0000015 (hematopoietic stem cell), CL:0000584 (enterocyte), CL:1000838 (proximal tubule epithelial cell), CL:0000182 (hepatocyte), CL:0000128 (oligodendrocyte)
UBERON	UBERON:0002371 (bone marrow), UBERON:0002113 (kidney), UBERON:0002107 (liver), UBERON:0002048 (lung), UBERON:0002316 (CNS white matter)
CHEBI	CHEBI:44185 (methotrexate), folinic acid/leucovorin (CHEBI term for calcium folinate)
HGNC	DHFR (HGNC:2861), MTHFR (HGNC:7436), SLCO1B1 (HGNC:10959), ABCB1 (HGNC:40), ATIC (HGNC:794)
NCIT	NCIT:C15986 (Pharmacotherapy) for leucovorin/glucarpidase treatment_term
MONDO	MONDO:0034212

Sources

Reference Validation

Checked with linkml-reference-validator 0.2.1.

Outcome	Count
References checked	38
Resolved	38
Unresolved (possible confabulation)	0
Unverifiable	0
References weighed for topical relevance	38
On topic	24
Off topic	0

All extracted references resolved successfully.

Term Validation

Checked with linkml-term-validator 0.4.5, through the ols: adapter.

Outcome	Count
Terms checked	52
Resolved	46
Unresolved (possible confabulation)	0
Obsolete	0
Unverifiable	6
Terms whose name was checked	37
Terms named correctly	31
Terms named as a different term	4
Terms whose name is worth a second look	2

Terms the report names something else

These identifiers resolve, so nothing about them looks wrong, and the ontology calls them something unrelated to what the report calls them. That usually means the identifier is not the one the sentence needs:

MONDO:0034212 (2 mentions) - the report calls it "MONDO"; MONDO calls it methotrexate toxicity
HP:0001636 (1 mention) - the report calls it "Myocarditis"; HP calls it Tetralogy of Fallot
CL:0000075 (1 mention) - the report calls it "epithelial cell — intestinal/renal tubular"; CL calls it columnar/cuboidal epithelial cell
CL:0000097 (1 mention) - the report calls it "mast cell — not primary but relevant to mucosal injury context"; CL calls it mast cell

Terms whose name is worth a second look

The report's name for these is recognisably related to the term's own name without being one of them. A loose paraphrase reads the same way as a citation of the wrong sibling term - and so does a related synonym, which the ontology records precisely because it names something adjacent rather than the same thing - so these are listed rather than judged:

UBERON:0002316 (2 mentions) - the report calls it "white matter of central nervous system"; UBERON calls it white matter, and lists "white matter of neuraxis" among its other names
CL:0000015 (2 mentions) - the report calls it "hematopoietic stem cell"; CL calls it male germ cell

Prefixes with no resolver

Terms carrying these prefixes were not checked either way, because no configured ontology covers them. An unrecognised prefix may name an ontology this run could not reach as easily as one that does not exist, so nothing here is evidence of fabrication: ORPHA.

Ask OpenScientist

OpenScientist Report

Mechanistic Hypotheses

Discussions and Knowledge Gaps

Pathophysiology

Pathograph

Phenotypes

Genetic Associations

Medical Actions

Environmental Factors

Diagnosis

Progression

Clinical Burden

Animal Models

Source YAML

References & Deep Research

References

Deep Research

Methotrexate Toxicity — Comprehensive Research Report

1. Disease Information

2. Etiology

Disease causal factors

Risk factors

3. Phenotypes

4. Genetic/Molecular Information

5. Environmental Information

6. Mechanism / Pathophysiology

Causal chain (ordered)

Category detail

7. Anatomical Structures Affected

8. Temporal Development

9. Inheritance and Population

10. Diagnostics

11. Outcome / Prognosis

12. Treatment

Pharmacotherapy / rescue agents

Supportive and procedural measures

Experimental / investigational

Treatment algorithm (escalation logic)

13. Prevention

14. Other Species / Natural Disease

15. Model Organisms

Summary of Suggested Ontology Terms for KB Curation

Sources

Reference Validation

Term Validation

Terms the report names something else

Terms whose name is worth a second look

Prefixes with no resolver