CHAPTER

───────────────────────

Paracelsus and the Dose That Makes the Poison

Yoon Hang "John" Kim, MD, MPH

Board Certified in Preventive Medicine

Integrative & Functional Medicine

Abstract

Vitamin C (L-ascorbic acid) occupies a singular place in human medicine: an essential micronutrient at nutritional doses and a candidate pharmacologic agent at concentrations achievable only by intravenous infusion. This chapter frames that duality through the Paracelsian principle—that the dose makes the poison. At physiologic plasma concentrations (below roughly 200 μmol/L), ascorbate acts as a chain-breaking antioxidant and an obligate cofactor for the hydroxylase enzymes of collagen synthesis. Because intestinal absorption saturates and renal clearance accelerates, oral dosing cannot exceed a plasma ceiling of approximately 220 μmol/L, however aggressive the regimen. Intravenous administration bypasses this limit, reaching millimolar concentrations at which ascorbate reduces catalytic transition metals and generates hydrogen peroxide through Fenton chemistry. In cancer cells—characterized by elevated labile iron, diminished catalase activity, and mutation-dependent dehydroascorbate uptake—this pro-oxidant activity can be selectively cytotoxic, and a 2024 randomized phase II trial in metastatic pancreatic cancer reported a doubling of median overall survival when pharmacologic ascorbate was added to chemotherapy. The same pro-oxidant mechanism, however, offers no clear benefit and signals possible harm in sepsis: the large LOVIT trial found increased death or persistent organ dysfunction, and current critical-care guidelines advise against routine use. Vitamin C thus illustrates how route, dose, and host biology—not the molecule alone—determine whether ascorbate protects or poisons. Its oncologic promise remains investigational and tumor-specific; its role in critical illness remains cautionary.

Keywords: ascorbic acid; vitamin C; pharmacologic ascorbate; pro-oxidant; hydrogen peroxide; Fenton reaction; pancreatic cancer; sepsis; pharmacokinetics; G6PD deficiency.

"All things are poison, and nothing is without poison; the dosage alone makes it so a thing is not a poison."

— Paracelsus (1493–1541)

1  The Paracelsian Paradox of Ascorbic Acid

Five centuries ago, Theophrastus von Hohenheim—known to history as Paracelsus—upended classical medicine with a deceptively simple observation: the boundary between remedy and toxin is not drawn by the substance itself, but by the dose. That principle, dosis sola facit venenum, has survived every revolution in pharmacology since. Few molecules illustrate it more elegantly than L-ascorbic acid—vitamin C.

At nutritional concentrations measured in micromoles, vitamin C is an indispensable antioxidant: a cofactor for collagen synthesis, a scavenger of reactive oxygen species, and arguably the most cost-effective free-radical quencher available to human biology. At pharmacologic concentrations measured in millimoles—achievable only through intravenous infusion—the same molecule shifts toward pro-oxidant behavior, generating the very oxidative stress it was recruited to prevent. In cancer cells burdened with excess labile iron, this pro-oxidant shift can prove selectively lethal. In septic patients, it may prove selectively harmful.

This chapter traces vitamin C along the full arc of the Paracelsian curve: from the evolutionary accident that made it essential, through its biochemistry as an antioxidant, to the pharmacokinetic boundary that separates physiologic from pharmacologic dosing, and finally to the clinical evidence—both promising and cautionary—that defines its modern therapeutic frontier.

2  The Evolutionary Loss: Why Humans Need Vitamin C at All

Most mammals synthesize ascorbic acid endogenously from glucose via a four-enzyme pathway in the liver. Humans, along with other primates, guinea pigs, and certain bat and passerine species, cannot. The culprit is a nonfunctional GULO gene—encoding L-gulonolactone oxidase, the enzyme catalyzing the final step of the biosynthetic pathway.¹ The mutation appears to have occurred tens of millions of years ago in the primate lineage—commonly estimated at approximately 61 million years—long before dietary sources became unreliable.

This evolutionary loss carries a profound clinical implication: every microgram of ascorbate in human plasma arrived through the gut. The recommended dietary allowance—75 mg/day for women, 90 mg/day for men—was established to cover approximately 97–98% of healthy adults, based substantially on near-maximal neutrophil ascorbate concentration with minimal urinary loss.^2,3 The amount required merely to prevent frank scurvy is considerably lower (on the order of 10 mg/day). Whether the current RDA is optimal for broader health outcomes—particularly under conditions of heightened oxidative stress—remains an active area of investigation.

3  Vitamin C and Collagen: The Biochemical Foundation

The best-characterized physiologic role of ascorbic acid is as an obligate cofactor for the post-translational modification of collagen. Two iron- and 2-oxoglutarate–dependent dioxygenases—prolyl-4-hydroxylase and lysyl hydroxylase—require ascorbate to maintain their catalytic iron in the reduced (Fe²⁺) state. Without this reducing capacity, the enzymes stall, proline and lysine residues in nascent procollagen chains remain unhydroxylated, and the resulting protein cannot form the stable triple-helix structure that gives collagen its tensile strength.

The clinical consequences of this biochemical deficit are immediate and visible. Scorbutic patients develop bleeding gums not because ascorbate is directly hemostatic, but because the collagen matrix supporting gingival vasculature disintegrates. Wound healing arrests. Bone remodeling fails. At the molecular level, scurvy is a disease of structural protein collapse—and it resolves within days of ascorbate repletion.

Oral supplementation at doses of 200–400 mg/day is sufficient to achieve tightly regulated steady-state plasma concentrations in the range of 70–80 μmol/L—more than adequate to saturate the hydroxylase enzymes and maintain normal connective tissue turnover.^4,5 Higher oral doses produce diminishing returns in plasma concentration due to absorptive saturation and increased renal clearance, a pharmacokinetic constraint explored in detail below.

4  The Antioxidant Economics of Ascorbic Acid

Before examining vitamin C at pharmacologic doses, it is worth pausing to appreciate its unrivaled efficiency at physiologic ones. Ascorbic acid is a potent water-soluble antioxidant that operates through multiple mechanisms: direct scavenging of superoxide, hydroxyl radicals, and singlet oxygen; regeneration of oxidized α-tocopherol (vitamin E), thereby extending lipid-phase antioxidant protection; maintenance of the glutathione redox couple; and protection of LDL particles from oxidative modification.⁶

Ascorbic acid is also among the least expensive bioactive compounds in routine clinical use. It is produced at industrial scale, is chemically stable in dry form, and is available as a bulk commodity ingredient at a small fraction of the cost of most other antioxidant supplements. Precise figures are difficult to state durably—bulk prices vary with grade, purity, supplier, order quantity, and market conditions, and no single quotation represents the market. The qualitative point is nonetheless robust: cost is not a meaningful barrier to oral vitamin C supplementation anywhere in the world.

Two cautions attach to this observation. Bulk ingredient pricing should never be equated with the cost of sterile injectable preparations, which are manufactured to entirely different standards. And price per kilogram does not by itself establish antioxidant activity per dollar or clinical cost-effectiveness—that inference would require a specified activity assay or clinical outcome measure and a consistent basis for comparison.

This cost advantage applies exclusively to oral supplementation. Intravenous administration—with its requirements for sterile compounding, clinical supervision, and monitoring—introduces an entirely different cost structure, one that reflects the pharmacologic ambitions of high-dose therapy rather than the nutritional simplicity of daily supplementation.

5  The Pharmacokinetic Boundary: Where Nutrition Ends and Pharmacology Begins

This is where Paracelsus enters the conversation in earnest. The difference between vitamin C as antioxidant and vitamin C as pro-oxidant is not one of chemistry alone—it is fundamentally a question of achievable concentration, and the route of administration is the principal determinant of which concentrations become possible.

5.1  Oral Administration: The Ceiling

Intestinal absorption of ascorbate is mediated by sodium-dependent vitamin C transporter 1 (SVCT1), a saturable carrier that becomes rate-limiting at oral doses exceeding 500 mg. Bioavailability falls accordingly: approximately 70–90% at a 200 mg dose, but less than 50% at doses above 1,000 mg.^4,5 Unabsorbed ascorbate draws water osmotically into the intestinal lumen, producing the dose-limiting side effect of osmotic diarrhea.

Simultaneously, renal excretion accelerates once plasma concentrations exceed 70–85 μmol/L. The combined effect of absorptive saturation and renal clearance imposes an absolute pharmacokinetic ceiling: even with maximum tolerated oral dosing (3 g every four hours around the clock), peak plasma concentrations plateau at approximately 220 μmol/L (0.22 mM).⁴ No oral regimen, regardless of formulation or frequency, can breach this barrier.

5.2  Intravenous Administration: Beyond the Ceiling

Intravenous infusion bypasses the intestinal bottleneck entirely, delivering ascorbate directly into the vascular compartment with 100% bioavailability. The concentrations achieved are of a different order of magnitude (Table 1):⁴

* Pharmacokinetic model predictions, not direct measurements.

Table 1. Plasma ascorbate concentrations by route and dose. Data from Padayatty et al., Ann Intern Med, 2004.

This concentration gap—approximately 70-fold between maximum oral and high-dose IV administration—is the pharmacokinetic expression of the Paracelsian principle. At 220 μmol/L, ascorbate functions as an antioxidant. At millimolar concentrations, it shifts toward pro-oxidant behavior, generating hydrogen peroxide through iron-dependent chemistry. The molecule is identical; the biology is not. And while the route of administration is the principal reason pharmacologic plasma concentrations become achievable, the resulting redox biology depends on dose, infusion rate, compartment, catalytic metal availability, oxygen tension, and the antioxidant enzyme capacity of the target tissue.

6  Oxidative Stress and the Two Faces of Ascorbate

6.1  The Antioxidant Face (Physiologic Concentrations)

At plasma concentrations below 200 μmol/L, ascorbate functions as a classical chain-breaking antioxidant. It donates electrons sequentially—first forming the relatively stable ascorbyl radical, then dehydroascorbic acid—to neutralize superoxide, hydroxyl radicals, peroxyl radicals, and singlet oxygen. It regenerates vitamin E at the lipid-aqueous interface. It maintains intracellular glutathione in its reduced state. At these concentrations, the net effect is unambiguously protective.⁶

6.2  The Pro-Oxidant Face (Pharmacologic Concentrations)

At millimolar concentrations, ascorbate engages in a fundamentally different chemistry—one mediated by transition metals, particularly catalytic iron. Pharmacologic ascorbate promotes the formation of extracellular hydrogen peroxide through metal-catalyzed oxidation of ascorbate and reduction of molecular oxygen, with the ascorbate radical as an intermediate.^7,8 The hydrogen peroxide so generated can then react with ferrous iron through Fenton chemistry to yield the hydroxyl radical:

Fe²⁺  +  H₂O₂  →  Fe³⁺  +  •OH  +  OH⁻

The hydroxyl radical (•OH) is among the most reactive species in biology—capable of abstracting hydrogen atoms from virtually any organic molecule, initiating lipid peroxidation, oxidizing proteins, and inducing DNA strand breaks. The essential concept is that ascorbate at pharmacologic concentrations acts as a pro-drug for the delivery of hydrogen peroxide to the extracellular space, where downstream iron-dependent chemistry determines the extent of oxidative injury.^7,8,9,10

6.3  Selective Cytotoxicity: Tumor-Context-Dependent Mechanisms

The clinical relevance of pro-oxidant ascorbate hinges on a critical observation: cancer cells are disproportionately vulnerable. Several features of malignant cell biology converge to explain this selectivity, though their relative contributions vary by tumor type and genetic context:

Elevated labile iron pools. Cancer cells accumulate catalytically active iron at concentrations far exceeding those of normal tissue, providing abundant substrate for Fenton chemistry. This mechanism has been particularly well-characterized in glioblastoma, where MRI-based T2* mapping of tumor iron content has been shown to predict clinical response to pharmacologic ascorbate.^11,12

Reduced catalase activity. Many tumor lines express low levels of catalase, the enzyme responsible for decomposing H₂O₂ to water and oxygen. Tumor cells have been shown to have a systematically decreased ability to metabolize H₂O₂ compared to normal cells—a finding that provides a mechanistic basis for selective toxicity.¹³ Normal cells, with robust catalase expression, efficiently clear the peroxide before damage accumulates.

Metabolic vulnerability via GLUT transporters. This mechanism has been particularly demonstrated in KRAS- and BRAF-mutant colorectal cancer. Cancer cells with upregulated glucose transporters import dehydroascorbic acid (DHA) at high rates. Intracellular reduction of DHA back to ascorbate depletes NAD⁺ and ATP via GAPDH inactivation, triggering metabolic catastrophe independent of the extracellular pro-oxidant mechanism. This pathway is dependent on specific oncogenic mutations, underscoring that the mechanism is not universal to all cancers.¹⁴

Mitochondrial dysfunction. Compromised mitochondrial antioxidant systems in cancer cells may compound vulnerability to oxidative challenge, though this mechanism is less well-characterized than the iron and catalase pathways.

These mechanisms are best understood as complementary and tumor-context-dependent rather than as universal properties of all malignancies. Normal cells—with tightly regulated iron metabolism, competent catalase, intact mitochondria, and balanced redox homeostasis—are largely spared. This differential vulnerability is the biological basis for the therapeutic window of high-dose intravenous vitamin C in oncology, and it also explains why clinical responses are likely to vary by tumor type and genetic profile.

7  Adverse Effects: The Cost of Crossing the Pharmacokinetic Boundary

7.1  Osmotic Diarrhea

The most common adverse effect of high-dose oral vitamin C is self-limiting: osmotic diarrhea from unabsorbed ascorbate in the intestinal lumen. The "bowel tolerance" dose—typically 5–15 g/day in divided doses for healthy individuals—serves as a practical ceiling for oral supplementation and is rarely clinically significant.

7.2  Oxalate Nephropathy

Ascorbate is metabolized to oxalate and excreted renally. At high doses, urinary oxalate excretion rises proportionally, increasing the risk of calcium oxalate crystal deposition in renal tubules. Case reports document acute oxalate nephropathy with renal failure following very high-dose IV vitamin C (>10 g/day) in patients with pre-existing renal impairment or a history of nephrolithiasis. Long-term high-dose oral supplementation has been associated with increased kidney stone risk in epidemiologic studies, particularly in men. Baseline and periodic assessment of renal function is mandatory for any patient receiving high-dose IV protocols.

7.3  G6PD Deficiency: A Non-Negotiable Screen

Glucose-6-phosphate dehydrogenase (G6PD) deficiency is an X-linked enzymopathy affecting several hundred million people worldwide. G6PD catalyzes the first step of the pentose phosphate pathway, generating NADPH—the reducing currency that maintains glutathione in its active, reduced form. In erythrocytes, which lack mitochondria and depend entirely on the pentose phosphate pathway for NADPH, G6PD deficiency creates an antioxidant deficit with no compensatory backup.

When a G6PD-deficient patient receives high-dose intravenous ascorbate, the pro-oxidant hydrogen peroxide generated overwhelms the already depleted glutathione system. The result is acute oxidative hemolysis: Heinz body formation from denatured hemoglobin, red cell membrane fragmentation, and potentially life-threatening anemia. Case reports document hemolysis and methemoglobinemia at IV doses as low as 6–10 g in G6PD-deficient individuals. At least one fatality has been attributed to this mechanism.

Screening for G6PD deficiency before initiating high-dose IV vitamin C is not a recommendation—it is a clinical mandate. The National Cancer Institute states that patients with G6PD deficiency should not receive high-dose vitamin C because of hemolysis risk.¹⁵ The University of Iowa pancreatic cancer trial excluded G6PD-deficient patients from enrollment.¹⁶ A quantitative G6PD enzyme activity assay should be obtained prior to treatment initiation, and institutional or protocol-specific thresholds should guide dosing decisions.

8  Clinical Evidence: Pancreatic Cancer and the University of Iowa Trials

Pancreatic ductal adenocarcinoma (PDAC) carries one of the grimmest prognoses in oncology: median survival for stage IV disease hovers around 8 months with standard gemcitabine plus nab-paclitaxel chemotherapy, and 5-year survival remains below 10%. It is precisely in settings of such limited therapeutic options that the pro-oxidant mechanism of high-dose vitamin C has been tested most rigorously.

8.1  Phase 2 Randomized Trial (Bodeker et al., 2024)

In 2024, Bodeker, Smith, Berg, and colleagues at the University of Iowa published results of a phase 2 randomized controlled trial in Redox Biology, enrolling 36 patients with stage IV metastatic PDAC (34 of whom received treatment and were analyzed). Patients were randomized to standard chemotherapy alone versus chemotherapy plus high-dose intravenous vitamin C at 75 g IV, three times weekly (Table 2).¹⁶

Table 2. Phase 2 outcomes: IV vitamin C plus chemotherapy in metastatic pancreatic cancer. Bodeker KL et al., Redox Biol. 2024;77:103375.

A doubling of median survival in stage IV pancreatic cancer—from 8.3 to 16.0 months—is a result of considerable clinical significance. Notably, the addition of pharmacologic ascorbate did not worsen global quality of life or increase the frequency or severity of adverse events; some individual quality-of-life domains favored the ascorbate arm, though global health status did not differ significantly between groups.¹⁶ The mechanistic interpretation aligns with preclinical data: high-dose ascorbate generates sustained H₂O₂ in the tumor microenvironment, selectively damaging cancer cells with elevated labile iron and impaired catalase, while potentially sensitizing them to DNA damage from concurrent cytotoxic chemotherapy.

8.2  The Earlier Iowa Studies: Four Distinct Cohorts

The randomized trial did not arrive in isolation. It rests on a decade of earlier University of Iowa work that is frequently conflated in secondary reporting. Because these were separate cohorts with different diseases, treatment combinations, and designs, their results should not be pooled or presented interchangeably.

Glioblastoma, phase I. The first-in-human trial combined pharmacologic ascorbate with radiation and temozolomide in newly diagnosed glioblastoma. Ascorbate was infused three times weekly during radiation and twice weekly during the adjuvant phase, with dose cohorts escalated until the targeted plasma concentration of at least 20 mmol/L was achieved. The trial met its primary objective of safety and tolerability; no dose-limiting toxicities were observed. Survival outcomes were encouraging relative to historical expectation, but a phase I safety study cannot establish efficacy.¹⁷

Glioblastoma, phase II. The later single-arm phase II study enrolled 55 patients and reported median overall survival of 19.6 months against a historical benchmark of approximately 14.6 months. This study also demonstrated that T2* MRI mapping of tumor iron predicted response—the clearest human evidence to date linking the labile-iron mechanism to clinical outcome. It remains, however, a single-arm comparison against historical controls.¹²

Metastatic pancreatic cancer, phase I (PACMAN). This phase I study combined IV ascorbate with gemcitabine in stage IV pancreatic adenocarcinoma and reported mean survival of 13 ± 2 months among patients completing at least two treatment cycles, with no dose-limiting toxicity. The figure is a mean in a selected completer population, not a median in an intention-to-treat cohort, and is not directly comparable to the randomized trial above.¹⁸

Pancreatic chemoradiation, phase I. A separate phase I cohort combined ascorbate with gemcitabine and radiation; fourteen patients completed protocol treatment. The original publication reported median overall survival of 21.7 months versus 12.7 months in an institutional comparator cohort (P = .08) and median progression-free survival of 13.7 versus 4.6 months. These were nonrandomized comparisons, and the overall survival difference did not reach conventional significance.¹⁹

8.3  The Long-Term Survivors: What Can and Cannot Be Claimed

The most frequently repeated claim in popular coverage of this work concerns a small group of exceptional long-term survivors. The underlying observation is real, but its provenance and limitations deserve precision. Peer-reviewed follow-up of the phase I chemoradiation cohort documented three patients surviving beyond six years without evidence of recurrence as of March 30, 2021.²⁰ A November 2024 University of Iowa institutional report subsequently described these same three long-term survivors as being approximately nine years out from treatment.²¹

Two caveats govern how this observation should be read. First, the nine-year figure comes from institutional follow-up reporting, not from a peer-reviewed trial publication, and should be attributed and dated accordingly. Second—and more important clinically—all three survivors also received neoadjuvant FOLFIRINOX and underwent pancreaticoduodenectomy before or after protocol therapy.²⁰ These outcomes emerged from an uncontrolled, multimodality treatment context and cannot be attributed to ascorbate in isolation. They are hypothesis-generating observations, not evidence of an ascorbate-specific survival effect, and they should not be presented as long-term survival in metastatic disease.

Importantly, the University of Iowa program is NIH-funded (National Cancer Institute grant, 2018), lending institutional credibility to a line of investigation once dismissed as fringe. These are federally funded, IRB-approved, controlled trials conducted at a major academic medical center. A phase 2 trial in non-small cell lung cancer remains ongoing.

9  Clinical Evidence: Vitamin C in Sepsis—A Cautionary Tale

If the pancreatic cancer data illustrate the Paracelsian principle working in the clinician's favor, the sepsis literature illustrates the opposite: mechanistic plausibility and clinical reality diverging with potentially fatal consequences.

9.1  The Rationale

Sepsis generates massive, sustained oxidative stress. Critically ill septic patients exhibit markedly depleted plasma vitamin C levels, often meeting criteria for frank scurvy. The logic seemed compelling: replace the depleted antioxidant, mitigate oxidative organ damage, improve survival.

9.2  CITRIS-ALI (2019): Encouraging Secondary Signal, Negative Primary Endpoints

The CITRIS-ALI trial enrolled 167 patients with sepsis-induced acute respiratory distress syndrome (ARDS) and randomized them to IV vitamin C (50 mg/kg every 6 hours for 96 hours) versus placebo. The trial's primary endpoints—modified SOFA score, C-reactive protein, and thrombomodulin—were all negative.²² However, among 46 prespecified secondary outcomes analyzed without multiplicity adjustment, 28-day mortality was 29.8% in the vitamin C group versus 46.3% in placebo. This mortality signal is best regarded as exploratory rather than confirmatory, given the negative primary endpoints and the absence of correction for multiple comparisons.

9.3  LOVIT (2022): The Reckoning

Then came LOVIT—the largest and most methodologically rigorous trial of IV vitamin C in sepsis. Published in the New England Journal of Medicine, the trial enrolled 872 adults with sepsis requiring vasopressors across multiple centers. The intervention was IV vitamin C 50 mg/kg every 6 hours for up to 96 hours.²³

The results were unambiguous and adverse. The primary composite outcome of death or persistent organ dysfunction at 28 days occurred in 44.5% of the vitamin C group versus 38.5% in placebo—a risk ratio of 1.21 (95% CI 1.04–1.40; P = 0.01). Twenty-eight-day mortality alone was 35.4% versus 31.6%, a trend toward harm that did not independently reach conventional statistical significance.²³ By the composite measure, vitamin C was not merely unhelpful; it was actively harmful.

9.4  Making Sense of Discordant Results

HC = hydrocortisone. Note: CITRIS-ALI and LOVIT tested vitamin C as monotherapy; VITAMINS, ACTS, and VICTAS tested a triple-therapy combination. 

Table 3. Summary of major randomized trials of IV vitamin C in sepsis.

Several factors may explain the discordance. CITRIS-ALI selectively enrolled ARDS patients—potentially a subgroup in which oxidative stress is maximal and antioxidant replacement is physiologically appropriate—but its mortality finding was a secondary endpoint without multiplicity correction, and its primary endpoints were negative. LOVIT enrolled a broader sepsis population with a larger sample size providing substantially narrower confidence intervals. The design differences in co-interventions (vitamin C alone versus triple therapy) further complicate cross-trial comparisons, and meta-analyses to date have reached mixed conclusions.^24,25

One proposed explanation invokes the Paracelsian framework directly: 50 mg/kg every 6 hours (200 mg/kg/day) achieves plasma concentrations well into the millimolar range—concentrations at which ascorbate's pro-oxidant activity may dominate. In cancer, this pro-oxidant shift is therapeutic because tumor cells are selectively vulnerable. In sepsis, where the target is systemic oxidative stress across organs with intact (if stressed) antioxidant defenses, the same shift may compound rather than counteract the injury. This hypothesis has not been directly tested, and neither LOVIT nor CITRIS-ALI established that Fenton chemistry was the mechanism of the observed clinical results.

Current consensus: Based on LOVIT and subsequent meta-analyses, routine use of high-dose IV vitamin C in sepsis is not recommended and may be harmful. Contemporary Surviving Sepsis Campaign guidelines suggest against IV vitamin C, with low-certainty evidence.²⁶ Its use should be restricted to clinical trials pending further evidence on patient selection and dosing.

10  Discussion: The Therapeutic Window and Clinical Recommendations

Vitamin C thus occupies a position in therapeutics that Paracelsus would have recognized immediately: a substance whose benefit and harm are determined not by its nature, but by its quantity, its context, and the biology of its recipient.

10.1  The Therapeutic Spectrum

The full spectrum of ascorbate biology, from deficiency to pharmacologic dosing, is summarized in Table 4.

Note: There is no clinically validated universal concentration at which vitamin C abruptly switches from antioxidant to pro-oxidant. The transition depends on dose, iron availability, oxygen tension, compartment, and the antioxidant enzyme capacity of the target tissue.

Table 4. The Paracelsian spectrum of vitamin C.

10.2  Practical Recommendations

For general health and collagen support: Oral vitamin C at 200–400 mg/day is sufficient to saturate plasma and tissue stores; doses up to 1,000 mg/day are safe but offer diminishing returns in plasma concentration. No G6PD screening is required. The primary adverse effect—osmotic diarrhea—is self-limiting and dose-dependent.

For investigational cancer therapy: IV vitamin C at pharmacologic doses (e.g., 75 g per infusion in the Iowa pancreatic protocol), administered 2–3 times weekly, should be considered only in the context of a clinical trial or under the supervision of an experienced integrative oncology provider. Mandatory prerequisites include quantitative G6PD screening, baseline renal function assessment, and glucose monitoring (high-dose ascorbate can produce falsely elevated point-of-care glucometer readings). Concurrent use with standard chemotherapy or radiation appears synergistic in preclinical and early clinical data. Response is likely to be tumor-type- and genotype-dependent.

For sepsis: High-dose IV vitamin C is not recommended outside of clinical trials. The LOVIT trial—the largest and most rigorous investigation to date—demonstrated increased harm on its primary composite outcome, and Surviving Sepsis Campaign guidelines suggest against its use. Until subgroup analyses or novel trial designs clarify which patients (if any) might benefit, the default clinical position is restraint.

11  Conclusion

Paracelsus could not have known about SVCT1 transporters, Fenton chemistry, or randomized controlled trials. But he understood—with an intuition that has aged remarkably well—that the identity of a substance in the body is inseparable from the quantity in which it is delivered. Vitamin C is the modern proof of his principle: an essential nutrient at micromolar concentrations, a selective cytotoxin at millimolar ones, and potentially a source of harm when applied without regard for the biology of the recipient.

The pharmacokinetic ceiling of oral administration is not merely a dosing inconvenience—it is a biological firewall that separates two fundamentally different pharmacologic identities sharing the same structural formula. Recognizing that boundary, and respecting it, is the difference between evidence-based nutrition and evidence-based pharmacology.

The challenge ahead is not whether high-dose vitamin C "works." The more precise and clinically actionable questions are: in whom, at what dose, in which tumor types, at what stage of disease, and by what route? The University of Iowa trials suggest that for selected cancers—particularly pancreatic ductal adenocarcinoma—the pro-oxidant mechanism may be genuinely therapeutic. LOVIT supplies the necessary counterweight: it demonstrates that a biologically plausible intervention can worsen clinical outcomes in a different disease context, though it does not establish pro-oxidant chemistry as the mechanism responsible for that harm.

Dosis sola facit venenum. The dose alone makes the poison. Five centuries later, the old alchemist still has the last word.

References

1. Nishikimi M, Yagi K. Molecular basis for the deficiency in humans of gulonolactone oxidase, a key enzyme for ascorbic acid biosynthesis. Am J Clin Nutr. 1991;54(6 Suppl):1203S-1208S.

2. 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 USA. 1996;93(8):3704-3709.

3. Institute of Medicine (US) Panel on Dietary Antioxidants and Related Compounds. Dietary Reference Intakes for Vitamin C, Vitamin E, Selenium, and Carotenoids. Washington, DC: National Academies Press; 2000.

4. 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.

5. Lykkesfeldt J, Tveden-Nyborg P. The pharmacokinetics of vitamin C. Nutrients. 2019;11(10):2412.

6. Kaźmierczak-Barańska J, Boguszewska K, Adamus-Grabicka A, Karwowski BT. Two faces of vitamin C—antioxidative and pro-oxidative agent. Nutrients. 2020;12(5):1501.

7. 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 USA. 2005;102(38):13604-13609.

8. 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 USA. 2007;104(21):8749-8754. doi:10.1073/pnas.0702854104.

9. Chen Q, Espey MG, Sun AY, et al. Pharmacologic doses of ascorbate act as a prooxidant and decrease growth of aggressive tumor xenografts in mice. Proc Natl Acad Sci USA. 2008;105(32):11105-11109.

10. Du J, Cullen JJ, Buettner GR. Ascorbic acid: chemistry, biology, and the treatment of cancer. Biochim Biophys Acta. 2012;1826(2):443-457.

11. Du J, Wagner BA, Buettner GR, Cullen JJ. Role of labile iron in the toxicity of pharmacological ascorbate. Free Radic Biol Med. 2015;84:289-295.

12. Petronek MS, Monga V, Bodeker KL, et al. Magnetic resonance imaging of iron metabolism with T2* mapping predicts an enhanced clinical response to pharmacologic ascorbate in patients with GBM. Clin Cancer Res. 2024;30(2):283-293. doi:10.1158/1078-0432.CCR-22-3952.

13. 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.

14. Yun J, Mullarky E, Lu C, et al. Vitamin C selectively kills KRAS and BRAF mutant colorectal cancer cells by targeting GAPDH. Science. 2015;350(6266):1391-1396.

15. National Cancer Institute. Intravenous Vitamin C (PDQ®)–Health Professional Version. Bethesda, MD: National Cancer Institute. [Verify version and access date at publication.]

16. 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.

17. Allen BG, Bodeker KL, Smith MC, et al. First-in-human phase I clinical trial of pharmacologic ascorbate combined with radiation and temozolomide for newly diagnosed glioblastoma. Clin Cancer Res. 2019;25(22):6590-6597. doi:10.1158/1078-0432.CCR-19-0594.

18. Welsh JL, Wagner BA, van’t Erve TJ, et al. Pharmacological ascorbate with gemcitabine for the control of metastatic and node-positive pancreatic cancer (PACMAN): results from a phase I clinical trial. Cancer Chemother Pharmacol. 2013;71(3):765-775. doi:10.1007/s00280-013-2070-8.

19. Alexander MS, Wilkes JG, Schroeder SR, et al. Pharmacologic ascorbate reduces radiation-induced normal tissue toxicity and enhances tumor radiosensitization in pancreatic cancer. Cancer Res. 2018;78(24):6838-6851. doi:10.1158/0008-5472.CAN-18-1680.

20. Du J, Carroll RS, Steers GJ, et al. Catalase modulates the radio-sensitization of pancreatic cancer cells by pharmacological ascorbate. Antioxidants (Basel). 2021;10(4):614. doi:10.3390/antiox10040614.

21. University of Iowa Carver College of Medicine. High-dose IV vitamin C plus chemotherapy doubles survival in advanced pancreatic cancer. Institutional news report. November 11, 2024. [Institutional follow-up reporting; not a peer-reviewed trial publication.]

22. Fowler AA 3rd, Truwit JD, Hite RD, et al. Effect of vitamin C infusion on organ failure and biomarkers of inflammation and vascular injury in patients with sepsis and severe acute respiratory failure: the CITRIS-ALI randomized clinical trial. JAMA. 2019;322(13):1261-1270.

23. Lamontagne F, Masse MH, Menard J, et al. Intravenous vitamin C in adults with sepsis in the intensive care unit. N Engl J Med. 2022;386(25):2387-2398. doi:10.1056/NEJMoa2200644.

24. Wen C, Li Y, Hu Q, Liu H, Xu X, Lü M. IV vitamin C in sepsis: a latest systematic review and meta-analysis. Int J Clin Pract. 2023;2023:6733465. doi:10.1155/2023/6733465.

25. Tariq MA, Amin H, Ali U. Meta-analysis assessing the effectiveness of intravenous vitamin C in patients with sepsis and septic shock. Acute Crit Care. 2022;37(4):674-677. doi:10.4266/acc.2022.01088.

26. Evans L, Rhodes A, Alhazzani W, et al. Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock 2021. Crit Care Med. 2021;49(11):e1063-e1143.