Oct 12, 2026 · @Yoon Hang Kim MD

Yoon Hang Kim, MD, MPH — Board-Certified in Preventive Medicine | Integrative & Functional Medicine Physician

About Dr. Kim: Dr. Yoon Hang Kim is a board-certified preventive medicine physician with more than 20 years of experience in integrative and functional medicine. He has served as Medical Director of Integrative Oncology at Miami Cancer Institute and as Director of the Integrative Medicine Program at the University of Kansas Medical Center. He completed an Osher Fellowship at the University of Arizona Center for Integrative Medicine under Dr. Andrew Weil, holds medical acupuncture certification through UCLA, and is a recipient of the Institute for Functional Medicine Scholarship. He is the author of eight books and more than 25 peer-reviewed articles. His clinical interests include integrative oncology, low dose naltrexone (LDN), autoimmune conditions, chronic pain, fibromyalgia, chronic fatigue syndrome, mast cell activation syndrome (MCAS), and mold toxicity. He practices by telemedicine and at Hill Country Integrative Medicine in Fredericksburg, Texas. Professional: www.yoonhangkim.com | Clinical: www.directintegrativecare.com

Dr. Kim develops the clinical application of these and other supportive therapies at greater length in Integrative Oncology: Evidence-Based Strategies to Support Cancer Treatment. He is also the author of MCAS: Epidemic in Plain Sight and LDN Primer, and the founder of the LDN Support Group.

Introduction: The Pro-Oxidant Paradox in Cancer Therapeutics

The dominant narrative in both conventional and integrative medicine has long positioned oxidation as the villain and antioxidants as the hero. Patients take antioxidant supplements to neutralize free radicals. Oncologists worry about antioxidant interference with chemotherapy and radiation. The entire conversation assumes a single polarity: oxidation harms, reduction protects.

But this framing obscures a more nuanced biochemical reality. Cancer cells are not simply victims of oxidative stress — they are cells that have fundamentally reorganized their metabolism in ways that leave them uniquely vulnerable to precisely targeted oxidative challenges. The Warburg effect, the most famous metabolic hallmark of cancer, is not just about energy production. It is about the downstream consequences of metabolic reprogramming on antioxidant reserves, reducing equivalent availability, and the capacity to withstand acute oxidative insults.

Three therapies currently used in integrative oncology settings — high-dose intravenous vitamin C (IVC), methylene blue (MB), and major autohemotherapy with medical ozone (MAH) — converge on this vulnerability through distinct but mechanistically related biochemical pathways. Each generates reactive oxygen species (ROS) by a different route. Each exploits the same depleted antioxidant defense systems in cancer cells. And at least two of them — IVC and MB — share an identical clinical safety checkpoint: mandatory screening for glucose-6-phosphate dehydrogenase (G6PD) deficiency before treatment.

This chapter argues that the shared G6PD requirement is not a coincidence of unrelated side-effect profiles. It is a mechanistic fingerprint — a clinical signal that both therapies engage the same NADPH-dependent antioxidant defense network that cancer cells have already depleted through metabolic reprogramming. Ozone autohemotherapy, though it does not carry the same formal G6PD contraindication, engages an overlapping redox system through hormetic Nrf2 activation and ex vivo hydrogen peroxide generation.

The purpose of this chapter is not to argue that any of these therapies constitutes a proven cancer treatment. Rather, it presents a unified mechanistic framework that connects their biochemistry, explains their selectivity toward cancer cells, identifies where the translational evidence is strong and where it remains incomplete, and suggests that the most productive path forward is investigating them as members of a pharmacological class — pro-oxidant cancer therapies — rather than as isolated curiosities.

The central question each raises is the same: can a tolerable pro-oxidant exposure produce a sustained, tumor-selective metabolic or oxidative effect that either directly impairs cancer cell viability or enhances the effectiveness of conventional treatments? The existing literature supports asking that question. It does not yet answer it definitively for any of the three.

The Shared Vulnerability: Constrained Antioxidant Reserve and the Pentose Phosphate Pathway

To understand why pro-oxidant therapies can selectively target cancer cells, we must examine the metabolic architecture that makes those cells fundamentally different from normal tissue in their capacity to manage oxidative stress.

The Warburg Effect: More Than an Energy Strategy

Otto Warburg's observation — that cancer cells preferentially ferment glucose to lactate even in the presence of adequate oxygen — has been extensively revisited in modern cancer biology. What was once understood primarily as an energy production strategy is now recognized as a comprehensive metabolic reprogramming that alters biosynthetic capacity, signaling, and critically, redox homeostasis.

Glycolysis-dependent cancer cells upregulate glucose transporters (GLUTs), accumulate labile iron and copper in the tumor microenvironment, generate elevated baseline levels of intracellular ROS through accelerated metabolism and mitochondrial dysfunction, and — this is the critical point — run down their antioxidant reserves in the process of managing that constitutively elevated oxidative burden.

Cancer cells usually display higher levels of intracellular ROS than normal cells as a consequence of their accelerated metabolism (Patra and Hay, Trends Biochem Sci, 2014). This baseline oxidative stress is not incidental — it drives proliferative signaling through pathways including NF-κB, HIF-1α, and PI3K/AKT. But it comes at a cost: the antioxidant systems that buffer this stress are running near capacity.

The Antioxidant Enzyme Deficit

Research on murine hepatoma cells compared to freshly isolated normal hepatocytes revealed a striking disparity: cancer cells retained less than 10% of the antioxidant enzyme activity found in their normal counterparts. This depletion spanned the entire major antioxidant enzyme battery — catalase, superoxide dismutase (SOD), glutathione peroxidase (GPx), and reduced glutathione (GSH) content. In certain tumor lines, catalase was nearly undetectable by western blot analysis (Verrax et al., Curr Med Chem, 2009). This degree of depletion is a phenotype of particular tumor lines, not a universal law of cancer — antioxidant capacity varies widely, even among lines of the same tumor type — and the argument that follows needs only a relative shortfall in reserve, not a wholesale collapse.

Catalase is particularly important for the pro-oxidant therapy framework because it is the primary enzyme responsible for decomposing hydrogen peroxide (H₂O₂) — the ROS species generated by both IVC and ozone therapy, and the downstream product of MB-induced superoxide formation. University of Iowa researchers demonstrated that catalase activity is a central determinant of a cancer cell's sensitivity to pharmacological ascorbate: across a panel of lines, normal cells cleared hydrogen peroxide roughly twice as fast on average as cancer cells — with real variability within both groups — and low-catalase cells were killed by concentrations of H₂O₂ that high-catalase cells neutralized (Doskey et al., Redox Biol, 2016). When catalase expression was experimentally silenced in the resistant breast carcinoma cell line BT-20 using short hairpin RNA, the cells became susceptible to ascorbate-mediated oxidative stress — providing direct functional proof that catalase deficiency is the gatekeeper of pro-oxidant vulnerability (Klingelhoeffer et al., BMC Complement Altern Med, 2012).

The Pentose Phosphate Pathway: The Upstream Supply Line

Catalase, SOD, and GPx are the frontline enzymes, and most of them lean on a continuous supply of NADPH — the cell's primary reducing equivalent, produced predominantly through the pentose phosphate pathway (PPP). The dependence is not uniform, and the distinction matters: glutathione peroxidase relies on NADPH indirectly, through the glutathione-reductase step that regenerates reduced glutathione; catalase uses bound NADPH to protect itself from inactivation; while SOD, which disposes of superoxide using its own metal center, requires no NADPH at all.

The PPP branches from glycolysis at the first committed step of glucose metabolism. Its oxidative branch, catalyzed by glucose-6-phosphate dehydrogenase (G6PD) as the rate-limiting enzyme, irreversibly converts glucose-6-phosphate to 6-phosphogluconolactone while reducing NADP⁺ to NADPH. A second NADPH molecule is generated by 6-phosphogluconate dehydrogenase (6PGD) downstream. The non-oxidative branch interconverts sugar phosphates to provide ribose-5-phosphate for nucleotide synthesis and can feed intermediates back into glycolysis (Patra and Hay, 2014; Jiang et al., Protein Cell, 2014).

NADPH is essential for three critical antioxidant processes:

1. Glutathione reductase uses NADPH to regenerate reduced glutathione (GSH) from its oxidized form (GSSG). GSH is the cell's most abundant thiol antioxidant and a cofactor for glutathione peroxidase.
2. Thioredoxin reductase uses NADPH to maintain thioredoxin in its active (reduced) state. The thioredoxin system is a parallel antioxidant defense to the glutathione system.
3. Catalase activation requires NADPH as a bound cofactor to prevent the enzyme from being inactivated by its own substrate. Without NADPH, catalase undergoes compound II formation and loses activity.

G6PD therefore sits upstream of this NADPH-dependent machinery. In most nucleated cells it is one of several NADPH sources — malic enzyme, the IDH isoenzymes, one-carbon folate metabolism, and mitochondrial transhydrogenase also contribute — but it is typically the rate-limiting one and the first drawn on under acute oxidative demand. In the red blood cell, which has no mitochondria and none of these alternatives, G6PD is essentially the sole source, which is exactly what makes the erythrocyte so informative later in this discussion.

The Cancer Cell's Paradoxical NADPH Dependency

Here is where the picture becomes nuanced. Cancer cells do not simply have low PPP activity. Many tumors actually upregulate G6PD and PPP flux — not for antioxidant defense, but to generate the NADPH required for fatty acid synthesis, nucleotide production, and the biosynthetic demands of rapid proliferation (Jiang et al., 2014; Patra and Hay, 2014).

This creates a metabolic tension: the cancer cell needs NADPH for both biosynthesis and antioxidant defense, but the same pool of NADPH cannot serve both purposes simultaneously. Under conditions of acute oxidative challenge — which is precisely what pro-oxidant therapies deliver — the cell faces a choice between sustaining proliferation and sustaining survival. It typically cannot do both.

The elevated PPP in cancer cells may seem like it should provide adequate NADPH for antioxidant defense. But much of that NADPH may already be committed to biosynthesis rather than held in reserve — a metabolic competition that is plausible and partly evidenced, though not an established universal state, since tumors can also redistribute flux and raise NADPH output when challenged. When an exogenous pro-oxidant surge arrives — whether from extracellular H₂O₂ (IVC), mitochondrial NADPH consumption (MB), or ex vivo ozone-derived ROS (MAH) — the cancer cell's NADPH supply is stretched beyond its capacity.

Normal cells, which are not diverting their NADPH into biosynthetic overdrive, retain enough uncommitted NADPH to regenerate glutathione, maintain thioredoxin, and sustain catalase activity. They survive the same oxidative insult that overwhelms the cancer cell.

This is the fundamental basis of pro-oxidant selectivity: not that cancer cells "have no antioxidant defenses," but that their defenses are already operating at capacity, with no reserve buffer for an acute challenge.

IV Vitamin C: The Hydrogen Peroxide Pro-Drug

High-dose intravenous vitamin C (IVC), now more precisely termed pharmacological ascorbate (PAA) in the research literature, is the most extensively studied pro-oxidant therapy in integrative oncology. Its mechanism illustrates the core principle that connects all three interventions, and recent clinical data have elevated it from a mechanistic curiosity to a therapy with preliminary phase II efficacy signals.

The Pharmacokinetic Revolution: Why Route of Administration Changes Everything

The history of vitamin C in cancer is, in many ways, a story of pharmacokinetic misunderstanding. In the 1970s, Ewan Cameron and Linus Pauling reported retrospective survival benefits using 10 grams of ascorbate daily — given both intravenously and orally — in patients with advanced cancer. In the 1980s, two Mayo Clinic trials led by Charles Moertel failed to replicate these results using oral ascorbate alone. Vitamin C was declared ineffective, and the case was considered closed for nearly two decades.

It was not until Mark Levine and colleagues at the NIH undertook detailed pharmacokinetic studies in healthy volunteers that the critical distinction emerged (Levine et al., 1996; Padayatty et al., 2004). Their findings were striking:

• Oral ascorbate is under tight physiological control. Intestinal absorption saturates, renal excretion increases, and plasma concentrations rarely exceed 200 µM even at oral doses exceeding 100 times the recommended daily allowance. Maximum tolerated oral doses (~18 g daily) produce plasma levels of approximately 100–200 µM.
• Intravenous ascorbate bypasses intestinal absorption entirely. A dose of 1.25 g IV achieves peak plasma concentrations of approximately 1 mM. Clinical doses of 50–100 g produce sustained plasma levels of 10–20 mM or higher for up to 4 hours. After 75 g infusions, average plasma concentrations of 16.4 ± 0.5 mM have been measured.

This represents a 70- to 100-fold difference in achievable plasma concentration. The Cameron patients received IV ascorbate; the Moertel patients received only oral ascorbate. The two studies were not comparable, but this was not recognized for 15 years.

The Pro-Drug Mechanism: Extracellular Hydrogen Peroxide Generation

At pharmacological concentrations (above approximately 1 mM), ascorbate undergoes a fundamental shift from antioxidant to pro-oxidant behavior. The mechanism, elucidated primarily by Chen and colleagues at the NIH, proceeds as follows (Chen et al., Proc Natl Acad Sci U S A, 2005; Chen et al., 2007):

1. Ascorbate donates an electron to free transition metal ions — particularly catalytic iron (Fe³⁺) and copper (Cu²⁺) — which are present at elevated concentrations in the tumor microenvironment due to altered iron metabolism, angiogenesis, and tissue disruption.
2. This electron donation reduces Fe³⁺ to Fe²⁺ (the Fenton-active form) while simultaneously oxidizing ascorbate to the ascorbyl radical.
3. Fe²⁺ then reduces molecular oxygen to superoxide (O₂⁻) — an autoxidation step, not the Fenton reaction itself — and the superoxide dismutates to hydrogen peroxide (H₂O₂).
4. Additional H₂O₂ is generated through the autoxidation of ascorbate itself in the metal-rich extracellular fluid.
5. The net result: pharmacological ascorbate functions as a pro-drug that generates sustained extracellular H₂O₂ at concentrations sufficient to damage cancer cells.

Crucially, this H₂O₂ generation occurs in the extracellular space, not intracellularly. The hydrogen peroxide then diffuses into cells, where its fate depends on the cell's peroxide-clearing capacity — and where, in the presence of labile iron, the classical Fenton reaction (Fe²⁺ + H₂O₂) generates the far more damaging hydroxyl radical.

The Selectivity Mechanism: A Three-Part Advantage

The selectivity of IVC toward cancer cells is not a single phenomenon — it arises from the convergence of three independent biochemical advantages:

First, enhanced ROS generation at the tumor site. The tumor microenvironment concentrates the transition metals that catalyze H₂O₂ production. Labile iron pools are elevated in many tumors due to transferrin receptor upregulation, hepcidin dysregulation, and macrophage-mediated iron recycling. More metal catalyst at the tumor means more H₂O₂ generated per mole of ascorbate delivered.

Second, preferential cancer cell uptake. Cancer cells upregulate glucose transporters (GLUT1, GLUT3) as part of their glycolytic reprogramming. Dehydroascorbic acid (DHA), the oxidized form of ascorbate, enters cells through these same GLUT transporters. At pharmacological ascorbate concentrations, DHA levels also rise, enabling enhanced uptake precisely where GLUTs are overexpressed — in cancer cells. This is a biochemical irony: the Warburg effect, which protects cancer cells by providing metabolic flexibility, simultaneously opens the door to pro-oxidant therapy by upregulating the transporters that admit the oxidized pro-drug.

Third, catalase-dependent vulnerability. Normal cells express sufficient catalase to rapidly decompose the H₂O₂ that diffuses in from the extracellular space. The hydrogen peroxide is converted to water and oxygen before it can cause significant damage. Cancer cells with depleted catalase — and catalase matters most precisely under the high peroxide fluxes that pharmacological ascorbate creates, where it becomes the rate-limiting clearance enzyme, while peroxiredoxins and glutathione peroxidase carry more of the load at lower fluxes (Doskey et al., Redox Biol, 2016) — accumulate H₂O₂ intracellularly, leading to oxidative damage to DNA, mitochondrial membranes, and other critical structures.

Chen and colleagues demonstrated in vivo that intravenous ascorbate at doses of 0.25–0.5 mg/g body weight increased both blood and extracellular fluid ascorbate concentrations to above 8 mM and measurably increased H₂O₂ formation in the extracellular space. In more aggressive dosing models, intraperitoneal doses of 4 g/kg produced blood concentrations of 40 mM and tumor extracellular fluid peaks of approximately 20 mM sustained for up to 3 hours.

The Epigenetic Dimension: TET Enzyme Activation

Beyond its pro-oxidant cytotoxicity, pharmacological ascorbate has a second, mechanistically independent anticancer action: it serves as a cofactor for the ten-eleven translocation (TET) family of dioxygenases, which catalyze the oxidation of 5-methylcytosine (5mC) to 5-hydroxymethylcytosine (5hmC) — the first step in active DNA demethylation.

Aberrant DNA hypermethylation silences tumor suppressor genes in many cancers. TET2 loss-of-function mutations are common in hematologic malignancies. Cimmino and colleagues demonstrated that pharmacological ascorbate could enhance residual TET2 activity — not substitute for a wholly absent enzyme — and that doing so blocked aberrant self-renewal and leukemia progression in TET2-haploinsufficient models (Cell, 2017). This epigenetic mechanism is concentration-dependent and iron-dependent (TET enzymes require Fe²⁺ and 2-oxoglutarate as cofactors), making it complementary to rather than competitive with the pro-oxidant mechanism.

The epigenetic action also provides a partial explanation for why IVC may enhance chemotherapy and radiation effectiveness beyond simple oxidative stress potentiation: by demethylating silenced loci, ascorbate may restore sensitivity to apoptotic signaling pathways that chemotherapy and radiation engage.

The Phase II Breakthrough: Pancreatic Cancer 2024

The most significant clinical development for pharmacological ascorbate came in November 2024, when researchers at the University of Iowa published results from a randomized phase II trial in patients with stage IV metastatic pancreatic cancer — one of the most treatment-resistant malignancies.

In this trial, 36 participants with stage IV pancreatic cancer were randomized 1:1 to receive either standard chemotherapy (gemcitabine and nab-paclitaxel) alone, or the same chemotherapy plus high-dose IV vitamin C at 75 g three times weekly; 34 went on to receive their assigned treatment. The reported outcomes were striking:

• Median overall survival nearly doubled — 16 months with added ascorbate versus 8.3 months with chemotherapy alone (HR 0.46; 90% CI 0.23–0.92; p = 0.030).
• Progression-free survival was also prolonged in the ascorbate arm.
• No added toxicity was attributed to the ascorbate infusions, and quality of life was not adversely affected.

These results build on earlier phase I data from the same group, including long-term survivors from a trial combining IV vitamin C with radiation for pancreatic cancer.

Separate work in ovarian cancer demonstrated that IVC combined with carboplatin and paclitaxel reduced chemotherapy-related toxicity and showed signals of improved progression-free survival. Additional preclinical data suggested that PAA combined with PARP inhibitors (olaparib or veliparib) had enhanced efficacy in ovarian cancer cells, including those with wild-type BRCA status where PARP inhibitor response is typically limited.

These results should be read with their limits in view: the trial was small (36 randomized), single-institution, and powered around a 90% rather than the customary 95% confidence interval — a strong signal that invites confirmation, not a definitive result. They do not establish IVC as a standalone cancer treatment. But they do represent the strongest clinical evidence to date that pharmacological ascorbate — administered at pro-oxidant concentrations, through intravenous delivery, in combination with conventional therapy — can meaningfully alter outcomes in one of oncology's most difficult cancers. The mechanistic foundation (H₂O₂ generation → catalase-dependent selectivity → oxidative cancer cell injury) is consistent with the observed clinical benefit.

Methylene Blue: The Mitochondrial Electron Thief

Methylene blue (MB; methylthioninium chloride) is the oldest synthetic drug still in clinical use — synthesized by Heinrich Caro at BASF in 1876, deployed by Paul Ehrlich for biological staining, FDA-approved for methemoglobinemia, and now under investigation as a mitochondrial-targeted pro-oxidant agent. Its anticancer mechanism is mechanistically distinct from IVC while converging on the same underlying vulnerability.

Pharmacokinetics: What Reaches the Tumor

Before examining mechanism, the pharmacokinetic realities of MB are essential — because they explain both its promise and its most critical limitation.

MB is markedly lipophilic. Its oral bioavailability ranges from 53% to 97% depending on gastric pH and formulation, with plasma concentrations peaking 30–60 minutes after ingestion. It is extensively tissue-distributed: the current PROVAYBLUE label reports a steady-state volume of distribution of about 255 L and an elimination half-life of roughly 24 hours — considerably longer than the 5–7 hours older sources described, which the drug's multiphasic kinetics reconcile — and it is approximately 94% plasma protein bound (PROVAYBLUE label, 2024; Peter et al., Eur J Clin Pharmacol, 2000).

After a 2 mg/kg IV infusion, the FDA prescribing information (PROVAYBLUE®, 2024 label) reports a mean peak plasma concentration of approximately 2,917 ng/mL — roughly 9 µM using the anhydrous molecular weight. Peak tissue levels are reached at approximately 1 hour, with a tissue half-life of about 12 hours. MB exists in equilibrium between its charged, oxidized blue form and its uncharged, reduced leucomethylene blue (MBH₂) form. Only the reduced form crosses the blood-brain barrier (Müller, Acta Anat, 1992), and it is the leuco form that appears preferentially active in mitochondrial electron transfer.

The pharmacokinetic challenge for oncology is this: the 9 µM peak plasma concentration superficially overlaps with the 1–10 µM range that produced anticancer effects in glioblastoma cell culture (Poteet et al., 2013). But this comparison is misleading. With 94% protein binding, the free (active) plasma concentration is approximately 0.5 µM — substantially below the effective cell-culture range. Tissue distribution adds further complexity: MB concentrates in kidney, liver, lung, and brain but may not achieve equivalent concentrations in solid tumors. And even with a 24-hour half-life, a single systemic dose produces a declining exposure quite unlike the fixed, sustained concentrations maintained in cell-culture experiments — with roughly 94% of the drug protein-bound, the free fraction available to drive any redox effect is small.

The Alternative Electron Transfer Hypothesis

MB's most studied anticancer mechanism operates inside the mitochondria. The "alternative electron transfer" hypothesis, advanced most rigorously by Wen and colleagues (J Biol Chem, 2011), proposes that MB functions as a catalytic redox cycler that reroutes electron flow through the respiratory chain:

1. MB accepts electrons from NADH at Complex I (NADH-ubiquinone oxidoreductase), becoming reduced to leucomethylene blue (MBH₂).
2. MBH₂ donates these electrons directly to cytochrome c, bypassing Complexes I through III and the ubiquinone pool.
3. The oxidized MB is regenerated and re-enters the cycle — functioning as a continuously recycling electron shuttle.

Wen and colleagues demonstrated this mechanism using mitochondrial activity assays: MB rescued the effects of rotenone (Complex I inhibitor) and antimycin A (Complex III inhibitor) on complex I-III activity and free radical overproduction. This established MB as a functional bypass of the conventional electron transport chain.

A critical debate. The alternative electron transfer hypothesis is not universally accepted. Gureev and colleagues (2019), working with mouse brain mitochondria, challenged the notion that MB can fully bypass Complex III. Their data suggested that reduced MB donates electrons to the Qo ubiquinol-binding site of Complex III rather than directly to cytochrome c — meaning that Complex III inhibition by antimycin A could not be completely bypassed. They concluded that "the acceptor of electrons from MB is a Qo ubiquinol-binding site of Complex III; thus, MB-based drugs might not be helpful in mitochondrial dysfunctions involving Complex III inhibition."

A 2021 study (Sváb et al., Antioxidants, 2021) added further nuance, demonstrating species-dependent differences in how MB interacts with Complex III across rat, mouse, and guinea pig mitochondria.

The resolution may be that MB provides a partial bypass of Complex I and can function at the Complex III level through the Qo site, but does not provide a complete bypass of the entire I-III chain as originally proposed. For cancer applications, the most relevant aspect is not the bypass per se, but the metabolic consequences of rerouting electron flow.

Anticancer Consequences: Warburg Reversal and NADPH Depletion

Regardless of the precise electron transfer destination, MB's rerouting of mitochondrial electron flow produces three mechanistic consequences relevant to cancer cells:

1. Warburg reversal. Poteet and colleagues (J Biol Chem, 2013) demonstrated in glioblastoma cell lines that MB at 1–10 µM increased oxygen consumption rate, reduced extracellular acidification rate (a measure of glycolytic output), decreased lactate production, and — in a finding that distinguishes metabolic disruption from simple energy starvation — increased total cellular ATP. Proliferation was simultaneously inhibited, with S-phase cell cycle arrest and increased late apoptosis/necrosis at prolonged exposures.

The proposed interpretation: MB forces the cancer cell back toward oxidative phosphorylation and away from glycolytic dependence. The cell generates more ATP through respiration, but loses the metabolic intermediates and reducing equivalents that glycolysis provides for biosynthetic programs. It is metabolically disciplined but proliferatively impaired.

2. NADPH consumption — with caveats. MB is reduced to its leuco form by NADPH-dependent reductases, a reaction best characterized in the red cell's methemoglobin-reductase system; its mitochondrial electron shuttle, by contrast, draws primarily on NADH. Because cytosolic and mitochondrial NADPH are compartmentalized rather than one freely shared pool, the claim that MB drains the same NADPH that powers glutathione and thioredoxin reductase inside a tumor cell is a reasonable hypothesis rather than an established fact. To whatever extent it holds, it would form a bridge to IVC — both pressuring NADPH-dependent antioxidant defense, one by raising demand through extracellular H₂O₂, the other by competing for supply — but this convergence is best read as proposed, not proven.

3. Dose-dependent pro-oxidant behavior. At low concentrations (0.5–2 µM), MB may function as a net antioxidant by efficiently shuttling electrons through the chain and reducing electron leakage from dysfunctional complexes. At higher concentrations, the redox cycling rate exceeds cellular buffering capacity, and MB itself becomes a source of superoxide and downstream ROS. This biphasic dose-response curve — antioxidant at low doses, pro-oxidant at high doses — is characteristic of hormetic agents and complicates the therapeutic narrative.

In prostate cancer PC3 cells, concentration-dependent dark cytotoxicity (without light activation) was observed above 30 µM. At those concentrations, decreased antioxidant potential, reduced catalase levels, and increased lipid peroxidation were documented — but these findings were more pronounced with photoactivation (de Melo Gomes et al., 2023).

Photodynamic Activation: The Clearest Pro-Oxidant Case

MB absorbs red light in the 630–680 nm wavelength range. Upon photoactivation, it generates singlet oxygen (¹O₂) and other ROS directly — a mechanism identical in principle to conventional photodynamic therapy (PDT). A 2023 systematic review of preclinical anticancer MB-PDT studies identified tumor-size reductions in seven included articles across multiple cancer models (Taldaev et al., Front Pharmacol, 2023).

MB-PDT represents the mechanistically cleanest pro-oxidant application of methylene blue: the light provides spatial and temporal control over ROS generation, singlet oxygen production is directly measurable, and the oxidative injury is localized to illuminated tissue. However, MB-PDT results cannot be extrapolated to systemic oral or IV methylene blue without light activation. These are distinct interventions with different pharmacodynamic profiles.

The Critical Negative Result

Intellectual honesty requires foregrounding what did not work. Poteet and colleagues — the same investigators who demonstrated Warburg reversal in glioblastoma cell culture — tested oral MB at 10 and 50 mg/kg/day in two patient-derived glioblastoma xenograft models in mice:

• No tumor regressions occurred.
• Tumor-growth delay was not statistically significant.
• Tumor tissue did not show the AMPK activation observed in cell culture.

These are research doses (10–50 mg/kg/day in mice), not suggested human doses. But from a translational perspective, this negative result is highly informative: the investigators could not reproduce tumor target engagement under in-vivo conditions. The cell-culture mechanism — however elegant — did not translate to the animal model under the conditions tested.

This does not eliminate MB as a candidate. It does establish that the translational gap — from demonstrated cell-culture mechanism to in-vivo tumor effect — remains the most critical unresolved question for systemic methylene blue in oncology.

Major Autohemotherapy (Ozone): The Hormetic Oxidative Signal

Ozone therapy — specifically major autohemotherapy (MAH) — represents the most mechanistically distinct of the three pro-oxidant approaches. Unlike IVC, which delivers a chemical agent that generates ROS at pharmacological concentrations, and unlike MB, which reroutes mitochondrial electron flow, MAH works through an ex vivo oxidative exposure that generates reactive intermediates in blood and primes the body's antioxidant and immune systems through hormesis.

The Procedure and Immediate Biochemistry

In MAH, a measured volume of the client's blood (typically 100–250 mL) is withdrawn into a sterile collection system and exposed to a precisely controlled mixture of medical-grade oxygen and ozone, typically at concentrations of 10–80 µg O₃/mL. Ozone (O₃) is a triatomic allotrope of oxygen, far more reactive than O₂. The third oxygen atom makes it a powerful oxidizing agent — but also extremely unstable. Medical ozone has a half-life measured in minutes under physiological conditions.

When ozone contacts the biological matrix of blood, it reacts almost instantaneously with plasma components, generating two categories of downstream mediators (Tricarico and Travagli, Antioxidants, 2021):

Reactive Oxygen Species (ROS). Ozone dissolved in the aqueous phase of plasma generates several ROS, of which hydrogen peroxide (H₂O₂) is the most readily measured and the most relevant to the pro-oxidant framework. Bocci's group demonstrated that ozone dissolved in plasma or physiological saline generates H₂O₂ that can be unequivocally documented using specific detection methods. The concentrations are dose-dependent and, at therapeutic ozone concentrations, remain within the range that blood's antioxidant capacity can buffer.

Lipid Oxidation Products (LOPs). Ozone reacts with unsaturated fatty acids in plasma lipoproteins, generating a second class of mediators — lipid ozonation products including 4-hydroxynonenal (4-HNE) and other reactive aldehydes. These LOPs have longer half-lives than ozone itself or the primary ROS, allowing them to function as second messengers after the treated blood is reinfused. LOPs can traverse cell membranes and activate intracellular signaling cascades.

The treated blood is then reinfused intravenously. What returns to the circulation is not ozone — it has been completely consumed — but blood carrying these reactive intermediates, modified albumin, activated immune cells, and a redox-shifted cellular environment.

The Nrf2 Hormetic Pathway: Mechanisms and Evidence

The therapeutic rationale for MAH centers on hormesis — the principle that a controlled, sub-toxic oxidative stress triggers an adaptive antioxidant response that exceeds baseline protection. The word derives from the same conceptual framework as exercise physiology: physical exertion generates muscle damage and oxidative stress that, in the appropriate dose range, triggers adaptive strengthening rather than injury.

The key molecular mediator is nuclear factor erythroid 2-related factor 2 (Nrf2), the master regulator of the cellular antioxidant response.

Under resting conditions, Nrf2 is sequestered in the cytoplasm by its negative regulator Keap1 (Kelch-like ECH-associated protein 1), which continuously ubiquitinates Nrf2, targeting it for proteasomal degradation. Nrf2 has a half-life of approximately 20 minutes under basal conditions — it is made and destroyed continuously.

When oxidative stress occurs — including the ROS and LOPs generated during blood ozonation — reactive electrophiles modify specific cysteine residues on Keap1. This conformational change disrupts the Keap1-Nrf2 interaction, stabilizing Nrf2 and allowing it to accumulate in the cytoplasm. Stabilized Nrf2 translocates to the nucleus, dimerizes with small Maf proteins, and binds to antioxidant response elements (ARE) on DNA, upregulating a battery of cytoprotective genes.

The Nrf2-driven gene products include:

• Superoxide dismutase (SOD) — converts superoxide to H₂O₂
• Catalase — decomposes H₂O₂ to water and oxygen
• Glutathione peroxidase (GPx) — reduces H₂O₂ and lipid peroxides using GSH
• Glutathione S-transferase (GST) — conjugates reactive electrophiles to glutathione
• Heme oxygenase-1 (HO-1) — anti-inflammatory and cytoprotective enzyme
• NAD(P)H quinone oxidoreductase 1 (NQO1) — two-electron reduction of quinones, preventing one-electron redox cycling
• Multidrug resistance-associated proteins (MRPs) — cellular efflux pumps

Galiè and colleagues (Int J Mol Sci, 2019) published a detailed review establishing that low-concentration ozone exposure promotes nuclear translocation of Nrf2 to chromatin sites of active transcription and increases ARE-driven gene expression. They further demonstrated that ectopic expression of Keap1 reversed the ozone-induced ARE activation — providing mechanistic proof that the antioxidant response operates through the Keap1/Nrf2 pathway specifically.

Re and colleagues (2014) provided the first in vivo confirmation: Nrf2 levels in peripheral blood mononuclear cells (PBMCs) increased immediately after ozone/oxygen exposure at 35 µg/mL, prior to reinfusion, during MAH treatment in human subjects. After a series of three MAH sessions, Nrf2 returned to baseline — consistent with an adaptive feedback mechanism (hormesis) rather than sustained overactivation.

The Nrf2 Paradox in Cancer: Protection vs. Resistance

This is where the ozone therapy discussion must confront a significant biological complication, and where intellectual honesty is essential.

Nrf2 has a well-established dual role in cancer biology. The landmark review by Rojo de la Vega and colleagues — "NRF2 and the Hallmarks of Cancer" (Cancer Cell, 2018) — catalogued the evidence that Nrf2 participates in essentially every hallmark of cancer when constitutively activated:

• In normal cells: Transient, controlled Nrf2 activation is unambiguously cytoprotective. It prevents cancer initiation by neutralizing carcinogens, reducing oxidative DNA damage, and supporting genomic stability. This is the chemopreventive face of Nrf2.
• In established cancer cells: Constitutive or non-canonical NRF2 activation promotes tumor progression, metabolic reprogramming, resistance to apoptosis, enhanced proliferation, angiogenesis, invasion and metastasis, and — critically — resistance to chemotherapy and radiation therapy.

Constitutive NRF2 activation in tumors upregulates drug efflux transporters (MRPs), enhances glutathione-mediated drug detoxification, protects against chemotherapy-induced oxidative damage, and promotes metabolic flexibility that sustains tumor survival under therapeutic stress. NRF2 overexpression is associated with poor prognosis in lung, breast, esophageal, and other cancers.

This creates a genuine concern for ozone therapy in oncology clients: if MAH activates Nrf2, could it inadvertently strengthen the cancer cell's defenses against conventional treatment?

The counterargument — and this is where the pro-oxidant framework provides context — rests on the differential capacity for Nrf2-mediated adaptation:

Normal cells have intact Keap1-Nrf2 regulatory machinery and can mount a full adaptive response to hormetic oxidative stress. The Nrf2 upregulation strengthens their already-adequate defenses, making them more resistant to collateral damage from chemotherapy and radiation.

Cancer cells with constitutive Nrf2 activation are already running this pathway at or near maximum. They have less adaptive headroom. An additional Nrf2 stimulus may not produce proportional additional protection because the downstream targets (GSH synthesis, MRP expression, NQO1 activity) are already upregulated.

Cancer cells without constitutive Nrf2 activation could potentially benefit from ozone-induced Nrf2 upregulation — and this represents a legitimate safety concern that requires investigation.

Costanzo and colleagues (Eur J Histochem, 2020) directly tested this concern by exposing HeLa cancer cells to low ozone concentrations used in clinical practice. They found that ozone at therapeutic concentrations did not alter cytomorphology, motility, or proliferation features of the cancer cells — providing some reassurance that clinical-dose ozone does not actively promote tumor behavior, even if it does not directly harm it either.

Ozone and Chemotherapy Synergy: Preclinical Evidence

The preclinical evidence for ozone-chemotherapy combinations is older but consistent. In 1990, Zänker and Kroczek reported in Chemotherapy that ozone acted synergistically with 5-fluorouracil (5-FU) in breast cancer and colon cancer cell lines — and, notably, ozone overcame chemoresistance to 5-FU in cell lines previously resistant to the drug alone. The proposed mechanism was that ozone-generated ROS and the ROS generated by 5-FU converge on the same oxidative injury pathway.

Simonetti and colleagues (Evid Based Complement Alternat Med, 2017) confirmed and extended these findings in human colon cancer HT29 cells, showing that ozone combined with 5-FU or cisplatin produced enhanced cytotoxicity and significant inhibition of pro-survival cytokines (IL-6, IL-8, VEGF) — cytokines with central roles in colon cancer cell survival and chemoresistance.

In neuroblastoma cell cultures, ozone potentiated the effect of cisplatin and etoposide but not gemcitabine — a selectivity that suggests the synergy is pathway-dependent rather than universal.

A Critical Structural Limitation

When MAH is performed systemically, ozone does not enter the bloodstream. It is consumed entirely in the ex vivo reaction with blood components. The anticancer effect, if present in vivo, must therefore be mediated indirectly — through circulating ROS/LOP messengers, Nrf2-mediated enhancement of normal tissue resilience, immune modulation (NK cell activation, cytokine profile shifts), improved oxygen delivery to hypoxic tumor regions via enhanced erythrocyte 2,3-DPG and improved deformability, or combination effects with concurrent conventional therapy.

This is mechanistically honest: MAH is a systemic conditioning intervention, not a targeted cytotoxic therapy. Its potential value in oncology — if it proves to have value — is most likely as an adjuvant that strengthens normal tissue tolerance and enhances the effectiveness of treatments that do reach the tumor directly.

The G6PD Checkpoint: A Mechanistic Fingerprint

The most revealing convergence between IVC and MB — and the one that points most directly to a unified redox framework — is the shared clinical requirement for glucose-6-phosphate dehydrogenase (G6PD) screening before treatment.

Why G6PD matters. G6PD is the rate-limiting enzyme of the pentose phosphate pathway (PPP). Its essential function is generating NADPH — the primary cellular reducing equivalent that powers the glutathione reductase system, the thioredoxin reductase system, and catalase activation. Without adequate NADPH, cells cannot regenerate reduced glutathione (GSH) from its oxidized form (GSSG), cannot maintain thioredoxin in its active state, and cannot sustain catalase-mediated H₂O₂ detoxification.

In other words, G6PD is the upstream gatekeeper of the entire antioxidant defense network.

The hemolysis mechanism. Red blood cells are uniquely vulnerable because they lack mitochondria and depend entirely on the PPP for NADPH production. In G6PD-deficient individuals, erythrocytes cannot regenerate enough NADPH to neutralize oxidative stress. When IVC generates a surge of extracellular H₂O₂, or when MB's redox cycling consumes NADPH during its conversion between oxidized and reduced forms, G6PD-deficient erythrocytes accumulate oxidative damage, hemoglobin denatures into Heinz bodies, and hemolysis follows.

Methylene blue is particularly instructive here: its therapeutic action in methemoglobinemia depends on NADPH-dependent methemoglobin reductase to convert MB to leucomethylene blue. Without adequate G6PD activity, this reduction cannot occur — and the unreduced MB itself becomes an additional oxidant, worsening hemolysis rather than correcting methemoglobinemia (McDonagh et al., Pharmacogenet Genomics, 2013).

The mechanistic insight. The fact that both IVC and MB require G6PD screening is not a coincidence of unrelated side effects. It points to a shared dependence: both therapies place a load on the NADPH-powered machinery that manages oxidative stress — the same broad system whose reserve is constrained in many cancer cells. The screening requirement itself exists to protect erythrocytes, so the link to tumor killing is an inference from shared terrain, not a direct demonstration.

Consider the logic:

• Cancer cells have depleted catalase, SOD, GPx, and GSH.
• All of these defenses depend on NADPH for regeneration.
• NADPH comes primarily from G6PD and the pentose phosphate pathway.
• IVC overwhelms the H₂O₂-detoxifying capacity that NADPH supports.
• MB directly consumes NADPH through its redox cycling and diverts electron flow.
• G6PD-deficient erythrocytes are harmed because they mirror — in exaggerated form — the same NADPH-starved vulnerability that cancer cells exhibit constitutively.

The G6PD safety screen, then, is more than a precaution — it is a visible signature of the redox terrain these therapies work in. The G6PD-deficient erythrocyte, with no mitochondria and no alternative NADPH supply, is an accessible if imperfect model of what acute oxidative stress does to a cell whose reductive reserve cannot keep up — the same predicament, in exaggerated form, that the pro-oxidant strategy aims to create in a tumor. It is an illuminating analogy, not proof that the two mechanisms are identical.

Ozone's relationship to this pathway. Ozone MAH in fact shares the same contraindication: G6PD deficiency (favism) is listed as an absolute contraindication to ozone therapy in the ISCO3/AEPROMO Madrid Declaration and across published ozone protocols — precisely because the ozone-derived H₂O₂ generated ex vivo loads the same NADPH-dependent glutathione system in the red cells being treated. So the G6PD screen is common to all three, which makes their shared reliance on NADPH-powered oxidative-stress handling more striking still — even as the caveat above, analogy rather than proof, continues to apply.

Mechanistic Convergence: Side-by-Side

Translational Gaps and Honest Limitations

A mechanistic framework is only as useful as its honesty about what remains unproven. Each therapy carries specific translational gaps that the framework alone cannot close.

The exposure-target engagement problem. For all three therapies, the central unsolved question is identical: does the pro-oxidant effect demonstrated in cell culture or ex vivo blood translate into sustained, tumor-selective oxidative stress in a living human with cancer? This requires demonstrating four things in sequence — adequate drug delivery to the tumor, measurable metabolic or redox changes in tumor tissue, evidence that those changes exceed the cancer cell's (but not normal tissue's) tolerance, and clinical tumor response.

IVC has progressed the farthest along this chain, with pharmacokinetic data showing that IV dosing achieves plasma concentrations in the range associated with pro-oxidant activity, and with early-phase clinical trials reporting signals of activity. But even for IVC, the bridge between plasma concentration and intratumoral oxidative injury is incompletely characterized.

MB's translational problem is more severe. The strongest mechanistic study (Poteet et al., 2013) includes a built-in negative control: the same investigators who demonstrated Warburg reversal in glioblastoma cell culture could not reproduce target engagement or tumor control in xenograft models. Plasma protein binding (approximately 94%), rapid partitioning into blood cells, and the transient nature of an IV or oral peak all argue that nominal concentration matching between a cell-culture dish and a tumor is insufficient.

For ozone MAH, the translational challenge is structural: ozone does not enter the circulation. Any anticancer effect must be mediated by downstream signaling molecules — ROS, LOPs, cytokine shifts, immune cell priming — that themselves have limited half-lives and uncertain tumor access.

What would change the picture. For each therapy, the decisive next evidence is not another cell-culture demonstration of ROS-mediated killing. It is sustained tumor target engagement at tolerable exposure — pharmacodynamic proof that the proposed mechanism is operating in a human tumor. Specifically: tumor-tissue measurements of NADPH/NADP⁺ ratios, glutathione status, catalase activity, and oxidative damage markers, ideally paired with matched normal-tissue controls to confirm selectivity. These remain research endpoints, not validated tests for selecting a therapy — and a blood redox measurement cannot be assumed to reflect what is happening inside a tumor.

Until that evidence exists, these therapies remain mechanistically interesting research candidates — not established cancer treatments.

Clinical Implications for the Integrative Practitioner

Understanding the unified redox framework has practical value for clinicians using or considering these therapies, even in the absence of definitive efficacy data.

G6PD testing is non-negotiable for IVC and MB. This is not optional caution — it reflects direct engagement with the NADPH-dependent pathway that makes these therapies work. A G6PD-deficient individual's erythrocytes model the same vulnerability being targeted in cancer cells. The test protects the client while simultaneously confirming that the therapy is engaging its intended mechanism.

Consider G6PD testing for MAH as well. While not universally mandated, the mechanistic logic is clear: ozone-derived H₂O₂ in the treatment bag taxes the same NADPH-glutathione system in erythrocytes. At higher ozone concentrations, the biochemistry argues for screening.

Serotonin syndrome risk with MB. Methylene blue inhibits monoamine oxidase A. Concurrent use with SSRIs, SNRIs, TCAs, or opioids carrying serotonergic activity (tramadol, meperidine, fentanyl) poses a risk of potentially fatal serotonin syndrome. A thorough medication reconciliation is essential before any MB protocol.

Do not conflate mechanisms. The three therapies share a redox framework but operate through different doors: IVC generates extracellular H₂O₂; MB reroutes mitochondrial electron flow and depletes NADPH; ozone MAH works through hormetic Nrf2 activation and indirect oxidative signaling. Clinical results from one cannot be extrapolated to another. MB-PDT results (with directed light) cannot be assumed for systemic MB without light. Ozone MAH cannot be assumed to deliver direct tumor cytotoxicity.

Frame these accurately with clients. These are mechanistically plausible interventions under active investigation — not proven cancer treatments. The strongest clinical position is: the biology is real, the vulnerability is real, the safety requirements are understood, and the translational evidence is incomplete. That honest framing serves clients better than either premature enthusiasm or reflexive dismissal.

Important: This material is for educational purposes only and does not constitute medical advice. Pro-oxidant therapies should only be administered under the supervision of a qualified clinician experienced in integrative oncology. Always discuss treatment options with your healthcare team before making clinical decisions.

References

1. Poteet E, Choudhury GR, Winters A, et al. Reversing the Warburg effect as a treatment for glioblastoma. J Biol Chem. 2013;288(13):9153–9164. doi:10.1074/jbc.M112.440354 (see published correction, 2018: doi:10.1074/jbc.AAC118.005625)
2. Chen Q, Espey MG, Krishna MC, et al. Pharmacologic ascorbic acid concentrations selectively kill cancer cells: action as a pro-drug to deliver hydrogen peroxide to tissues. Proc Natl Acad Sci U S A. 2005;102(38):13604–13609.
3. Chen Q, Espey MG, Sun AY, et al. Ascorbate in pharmacologic concentrations selectively generates ascorbate radical and hydrogen peroxide in extracellular fluid in vivo. Proc Natl Acad Sci U S A. 2007;104(21):8749–8754.
4. Levine M, Conry-Cantilena C, Wang Y, et al. Vitamin C pharmacokinetics in healthy volunteers: evidence for a recommended dietary allowance. Proc Natl Acad Sci U S A. 1996;93(8):3704–3709.
5. Padayatty SJ, Sun H, Wang Y, et al. Vitamin C pharmacokinetics: implications for oral and intravenous use. Ann Intern Med. 2004;140(7):533–537.
6. Cimmino L, Dolgalev I, Wang Y, et al. Restoration of TET2 function blocks aberrant self-renewal and leukemia progression. Cell. 2017;170(6):1079–1095.e20. doi:10.1016/j.cell.2017.07.032
7. Verrax J, Curi Pedrosa R, Beck R, et al. In situ modulation of oxidative stress: a novel and efficient strategy to kill cancer cells. Curr Med Chem. 2009;16(15):1821–1830.
8. Doskey CM, Buranasudja V, Wagner BA, et al. Tumor cells have decreased ability to metabolize H₂O₂: implications for pharmacological ascorbate in cancer therapy. Redox Biol. 2016;10:274–284. doi:10.1016/j.redox.2016.10.010
9. Klingelhoeffer C, Kämmerer U, Koospal M, et al. Natural resistance to ascorbic acid induced oxidative stress is mainly mediated by catalase activity in human cancer cells and catalase-silencing sensitizes to oxidative stress. BMC Complement Altern Med. 2012;12:61. doi:10.1186/1472-6882-12-61
10. Wen Y, Li W, Poteet EC, et al. Alternative mitochondrial electron transfer as a novel strategy for neuroprotection. J Biol Chem. 2011;286(18):16504–16515.
11. Poteet E, Winters A, Yan LJ, et al. Neuroprotective actions of methylene blue and its derivatives. PLoS One. 2012;7(10):e48279. doi:10.1371/journal.pone.0048279
12. Gureev AP, Shaforostova EA, Popov VN, Starkov AA. Methylene blue does not bypass Complex III antimycin block in mouse brain mitochondria. FEBS Lett. 2019;593(5):499–503. doi:10.1002/1873-3468.13332
13. Sváb G, Kokas M, Sipos I, Ambrus A, Tretter L. Methylene blue bridges the inhibition and produces unusual respiratory changes in Complex III-inhibited mitochondria: studies on rats, mice and guinea pigs. Antioxidants (Basel). 2021;10(2):305. doi:10.3390/antiox10020305
14. Peter C, Hongwan D, Küpfer A, Lauterburg BH. Pharmacokinetics and organ distribution of intravenous and oral methylene blue. Eur J Clin Pharmacol. 2000;56(3):247–250.
15. U.S. Food and Drug Administration. PROVAYBLUE (methylene blue) injection, USP: prescribing information. 2024 label, NDA 204630/S-021.
16. Bužga M, Machytka E, Dvořáčková E, et al. Methylene blue: a controversial diagnostic acid and medication? Toxicol Res (Camb). 2022;11(5):711–717. doi:10.1093/toxres/tfac050
17. Taldaev A, Terekhov R, Nikitin I, et al. Methylene blue in anticancer photodynamic therapy: systematic review of preclinical studies. Front Pharmacol. 2023;14:1264961. doi:10.3389/fphar.2023.1264961
18. de Melo Gomes LC, de Oliveira Cunha AB, Peixoto LFF, et al. Photodynamic therapy reduces cell viability, migration and triggers necroptosis in prostate tumor cells. Photochem Photobiol Sci. 2023;22(6):1341–1356.
19. McDonagh EM, Bautista JM, Youngster I, Altman RB, Klein TE. PharmGKB summary: methylene blue pathway. Pharmacogenet Genomics. 2013;23(9):498–508. doi:10.1097/FPC.0b013e32836498f4
20. Ata F, Javed S, Muthanna B, et al. Favism-induced methemoglobinemia in a G6PD deficient male with a subsequent hemolytic cascade: case report and review of the literature. Clin Case Rep. 2021;9(4):2048–2052. doi:10.1002/ccr3.3941
21. Galiè M, Covi V, Tabaracci G, Malatesta M. The role of Nrf2 in the antioxidant cellular response to medical ozone exposure. Int J Mol Sci. 2019;20(16):4009. doi:10.3390/ijms20164009
22. Re L, Martínez-Sánchez G, Bordicchia M, et al. Is ozone pre-conditioning effect linked to Nrf2/EpRE activation pathway in vivo? A preliminary result. Eur J Pharmacol. 2014;742:158–162.
23. Galiè M, Costanzo M, Nodari A, et al. Mild ozonisation activates antioxidant cell response by the Keap1/Nrf2 dependent pathway. Free Radic Biol Med. 2018;124:114–121.
24. Costanzo M, Romeo A, Cisterna B, et al. Ozone at low concentrations does not affect motility and proliferation of cancer cells in vitro. Eur J Histochem. 2020;64(2):3119. doi:10.4081/ejh.2020.3119
25. Rojo de la Vega M, Chapman E, Zhang DD. NRF2 and the hallmarks of cancer. Cancer Cell. 2018;34(1):21–43. doi:10.1016/j.ccell.2018.03.022
26. Tricarico G, Travagli V. The relationship between ozone and human blood in the course of a well-controlled, mild, and transitory oxidative eustress. Antioxidants (Basel). 2021;10(12):1946. doi:10.3390/antiox10121946
27. Clavo B, Santana-Rodríguez N, Llontop P, et al. Ozone therapy as adjuvant for cancer treatment: is further research warranted? Evid Based Complement Alternat Med. 2018;2018:7931849. doi:10.1155/2018/7931849
28. Yanchu L, Rong P, Rong C, et al. Ozone therapy for high-grade glioma: an overview. Front Oncol. 2023;13:1161206. doi:10.3389/fonc.2023.1161206
29. Zänker KS, Kroczek R. In vitro synergistic activity of 5-fluorouracil with low-dose ozone against a chemoresistant tumor cell line and fresh human tumor cells. Chemotherapy. 1990;36(2):147–154. doi:10.1159/000238761
30. Simonetti V, Quagliariello V, Giustetto P, Franzini M, Iaffaioli RV. Association of ozone with 5-fluorouracil and cisplatin in regulation of human colon cancer cell viability: in vitro anti-inflammatory properties of ozone in colon cancer cells exposed to lipopolysaccharides. Evid Based Complement Alternat Med. 2017;2017:7414083. doi:10.1155/2017/7414083
31. Patra KC, Hay N. The pentose phosphate pathway and cancer. Trends Biochem Sci. 2014;39(8):347–354. doi:10.1016/j.tibs.2014.06.005
32. Jiang P, Du W, Wu M. Regulation of the pentose phosphate pathway in cancer. Protein Cell. 2014;5(8):592–602. doi:10.1007/s13238-014-0082-8
33. Alfarouk KO, Ahmed SBM, Elliott RL, et al. The pentose phosphate pathway dynamics in cancer and its dependency on intracellular pH. Metabolites. 2020;10(7):285.
34. Bodeker KL, Smith BJ, Berg DJ, et al. A randomized trial of pharmacological ascorbate, gemcitabine, and nab-paclitaxel for metastatic pancreatic cancer. Redox Biol. 2024;77:103375. doi:10.1016/j.redox.2024.103375
35. Levine M, Espey MG, Chen Q. Losing and finding a way at C: new promise for pharmacologic ascorbate in cancer treatment. Free Radic Biol Med. 2009;47(1):27–29. doi:10.1016/j.freeradbiomed.2009.04.001
36. Hoffer LJ, Levine M, Assouline S, et al. Phase I clinical trial of i.v. ascorbic acid in advanced malignancy. Ann Oncol. 2008;19(11):1969–1974. doi:10.1093/annonc/mdn377
37. DiSanto AR, Wagner JG. Pharmacokinetics of highly ionized drugs. I. Methylene blue — whole blood, urine, and tissue assays. J Pharm Sci. 1972;61(4):598–602. doi:10.1002/jps.2600610422
38. Müller T. Light-microscopic demonstration of methylene blue accumulation sites in mouse brain after supravital staining. Acta Anat (Basel). 1992;144(1):39–44.
39. Re L, Mawsouf MN, Menéndez S, et al. Ozone therapy: clinical and basic evidence of its therapeutic potential. Arch Med Res. 2008;39(1):17–26.
40. Elvis AM, Ekta JS. Ozone therapy: a clinical review. J Nat Sci Biol Med. 2011;2(1):66–70. doi:10.4103/0976-9668.82319