Vitamin Anti-Cancer Evidence: A Systematic Review Anchored on Vitamin B6 / PLP and Pancreatic Cancer

Date: 2026-08-14 Methodology: Six-stage deep-research framework (scoping → retrieval → synthesis → drafting → review → finalization) Evidence Grading: Level I–VII pyramid (peer-reviewed RCTs / meta-analyses > cohort & case-control > in vitro & animal > narrative reviews > grey literature)

---

Key Points

• **Anchor study (Feehan et al., 2026, *Molecular Nutrition & Food Research*, DOI: 10.1002/mnfr.70333): PLP at 125–500 μg/mL produced dose-dependent growth inhibition in BxPC3 pancreatic adenocarcinoma cells; apoptosis rose from 2.38% to 45.4%; mitochondrial membrane potential was depolarized; superoxide generation increased; KRAS, E2F, G2M checkpoint, and mTOR transcripts were downregulated; mitochondrial aspartate/glutamate carriers (AGC1/2) were inhibited, cutting glutamine dependence. Evidence level: Level III (in vitro).
• In vitro–in vivo translational gap: Effective PLP concentrations (mmol/L range) are 2–3 orders of magnitude above serum PLP achievable by oral supplementation (μmol/L). The study's value is mechanistic, not therapeutic.
• Vitamin B6 broader evidence:** Drosophila in vivo work (Pilesi et al., 2024, *Cell Death and Disease*) supports PLP-deficiency + Ras cooperation in tumorigenesis; epidemiological meta-analyses (≈1.96 million subjects, ≈99,000 cancer cases) link high B6 intake/PLP status to lower risk across multiple cancers. No RCT confirmation exists.
• Vitamin C: Oral doses act as antioxidants; intravenous pharmacologic doses (25–30 mmol/L vs. oral ceiling ≈250 μmol/L) act as pro-oxidants, generating H₂O₂ that selectively kills tumor cells. Meta-analysis (Qu et al., 2025) suggests overall survival benefit with moderate-certainty evidence; a phase III trial in metastatic colorectal cancer found no overall PFS benefit, with possible signal in RAS-mutant subgroups. Not FDA-approved for cancer.
• Vitamin D: Three meta-analyses (Keum & Giovannucci 2018; Zhang & Niu 2019; *Annals of Oncology* 2019, PMCID PMC6821324) consistently show no reduction in cancer incidence (RR ≈ 0.99–1.03); mortality reduction signal is weak and low-quality.
• Vitamin E (SELECT trial): 35,000+ men, 7–12 year follow-up. Trial stopped early in 2008; 2011 update (JAMA) showed high-dose vitamin E (400 IU/day) increased prostate cancer risk by 17% (HR 1.17, P=0.008)—a 17% relative increase—approximately 11 extra cases per 1,000 men. Earlier ATBC trial in Finnish smokers showed 32% reduction, indicating population-dependent effects.
• Folate: U-shaped / stage-dependent. Low folate intake increases colorectal cancer risk (observational); high-dose folic acid (1 mg/day) in the Cole et al. 2007 JAMA trial showed no overall adenoma prevention; in patients with high baseline folate, supplementation may increase advanced adenoma risk. Benefit restricted to low-baseline subgroups (Wu et al., 2009).
• Vitamin A / ATRA (all-trans retinoic acid): The only RCT-validated, guideline-standard vitamin-derived cancer therapy. Induces terminal differentiation in acute promyelocytic leukemia (APL) by targeting the PML-RARα fusion gene; combined with anthracyclines and arsenic trioxide (ATO), APL cure rates reach ~70–95%. Exceeding dietary vitamin A intake does not confer this benefit and risks teratogenicity and hepatotoxicity.

Evidence Strength Map

| Vitamin | Core claim | Best evidence | Translation status | Risk | |---|---|---|---|---| | A / ATRA | APL differentiation therapy | Level I (RCT) | Standard of care | Benefit (drug-grade) | | C (IV high-dose) | Pro-oxidant tumor killing | Level III mechanism + Level I/II early clinical | Experimental | Awaiting phase III | | B6 / PLP | In vitro pancreatic cancer suppression | Level III in vitro + Level II epidemiology | Preclinical | In vitro effect; oral dose unattainable | | D | Reduce cancer incidence / mortality | Level I RCT meta-analyses | Prevention not established | No incidence reduction | | E (high-dose) | Cancer prevention | Level I RCT | — | Net harm (+17% prostate cancer) | | Folate / B9 | Prevent colorectal adenoma | Level I RCT (heterogeneous) | Stage-dependent | Benefit only at low baseline |

Cross-Vitamin Contradictions

• Vitamin E: ATBC (smokers benefited) vs. SELECT (overall harm)—population stratification governs direction.
• Folate: Observational deficiency-risk vs. RCT supplementation-no-benefit / potential harm—timing and baseline folate are key moderators.
• Vitamin D: Observational low-25(OH)D association vs. RCT null effects—suggests lifestyle/UV exposure confounders not reproducible by isolated supplementation.
• Vitamin B6: In vitro potency vs. unattainable human oral concentrations—mechanistic plausibility, unproven translation.

The In Vitro–In Vivo Translational Chasm

The most important calibration for the widely circulated "~50% cancer cell apoptosis" headline. Feehan et al. (2026) required 125–500 μg/mL PLP (approximately mmol/L scale) to achieve the reported apoptotic effect. Human oral supplementation produces serum PLP in the μmol/L (nmol/mL) range—roughly 100- to 1,000-fold lower than the in vitro threshold. Even intravenous pharmacologic intervention has not demonstrated sustained tumor-tissue delivery at this scale. Claiming oral vitamin B6 supplementation mimics this apoptotic effect constitutes an in vitro–in vivo extrapolation fallacy. Vitamin C faces a similar translational reality: intravenous administration is required to reach pharmacologic pro-oxidant concentrations.

Biphasic Dose Risk and Safety Ceilings

Vitamins exhibit a near-universal concentration-dependent duality:

• Deficiency → genomic instability (e.g., B6 deficiency → uracil misincorporation; folate deficiency → aberrant methylation).
• Excess → net harm (vitamin E 400 IU: +17% prostate cancer; chronic high folate: potential adenoma promotion; excess vitamin A: teratogenicity, hepatotoxicity; chronic high-dose B6: peripheral neuropathy).

Adult Tolerable Upper Intake Level (UL) for vitamin B6 is approximately 100 mg/day; the recommended daily intake is roughly 1.2–1.5 mg/day—a nearly 100-fold gap between recommended intake and the upper limit. The safe operating window lies between adequate dietary intake and the UL, not toward the UL.

Conclusions

1. "Vitamin anti-cancer" claims cannot be generalized. Evidence ranges from RCT-validated standard therapy (vitamin A/ATRA in APL) to in vitro phenomena without human translation (vitamin B6/PLP, oral vitamin C). 2. The PLP/pancreatic cancer study is rigorous and mechanistically valuable—mitochondrial depolarization → oxidative stress → KRAS downregulation + AGC1/2 glutamine blockade → apoptosis—but the effective concentrations are unattainable by oral supplementation. It does not support "supplement B6 to treat pancreatic cancer." 3. High-dose intravenous vitamin C is the most clinically advanced oxidative-stress / mitochondrial anti-cancer pathway beyond ATRA, with pharmacologic feasibility and RAS-mutant subgroup signals, but it remains non-standard. 4. Vitamin D supplementation does not reduce cancer incidence; high-dose vitamin E causes net harm; folate is biphasic and stage-dependent. Together, these expose the "more is better" misconception about antioxidant supplements. 5. ATRA in APL is the sole RCT-validated, guideline-standard vitamin-derived cancer therapy, and its success derives from targeted fusion-gene pharmacology and formal drug development, not dietary supplementation logic.

Public-Facing Recommendations

Obtain vitamins through a balanced diet (lean meats, fatty fish, whole grains, legumes, nuts, vegetables, fruits). Outside of physician-directed prescription contexts such as ATRA, do not pursue high-dose vitamin supplementation for cancer prevention. All cancer prevention and treatment decisions must follow medical consultation; this report does not constitute medical advice.

Limitations

• Narrative/systematic assessment; no PRISMA-compliant per-study risk-of-bias scoring (RoB 2 / ROBINS-I) or forest-plot pooling was performed.
• The precise 45.4% apoptosis figure for the Feehan et al. study originates from lay-summary coverage of the original article; the published abstract confirms apoptosis induction but does not list the exact percentage. Verification against the full text is advised.
• Vitamins K, B12, and niacin were not included in the current scope.
• Literature search cutoff: 2026-08-14; subsequent publications may update conclusions.

Selected References (APA 7.0)

Feehan, J., Kitchen, B., Fraser, S., Tripodi, N., Dargahi, N., & Apostolopoulos, V. (2026). Pyridoxal 5′-phosphate suppresses growth of pancreatic adenocarcinoma cells in vitro via regulation of oncogenic RAS. *Molecular Nutrition & Food Research, 70*, e70333. https://doi.org/10.1002/mnfr.70333

Pilesi, E., et al. (2024). Vitamin B6 deficiency cooperates with oncogenic Ras to induce malignant tumors in Drosophila. *Cell Death and Disease, 15*, 388. https://doi.org/10.1038/s41419-024-06787-3

Cao, X., et al. (2025). The dual role of vitamin C in cancer: From antioxidant prevention to prooxidant therapeutic applications. *Frontiers in Medicine, 12*, 1633447. https://doi.org/10.3389/fmed.2025.1633447

Qu, J., et al. (2025). Overall and progression-free survival of patients with malignant neoplasm following intravenous vitamin C: A systematic review and meta-analysis. *International Journal for Vitamin and Nutrition Research, 95*(3), 37372. https://doi.org/10.31083/IJVNR37372

Keum, N., & Giovannucci, E. (2018). Cancer and vitamin D supplementation: A systematic review and meta-analysis. PMID: 29635490.

Klein, E. A., et al. (2011). Vitamin E and the risk of prostate cancer: The SELECT trial. *JAMA, 306*(14), 1549–1556.

Cole, B. F., et al. (2007). Folic acid for the prevention of colorectal adenomas: A randomized clinical trial. *JAMA, 297*(21), 2351–2359.

Differentiation therapy with all-trans retinoic acid (ATRA) in acute promyelocytic leukemia. (2016). PMID: 27265273.

National Cancer Institute. (n.d.). *Acute Myeloid Leukemia Treatment (PDQ®) – Health Professional Version*. https://cancer.gov/cancertopics/pdq/treatment/adultAML/healthprofessional

Harvard T.H. Chan School of Public Health. (n.d.). *Vitamin E – The Nutrition Source*. https://www.hsph.harvard.edu/nutritionsource/vitamin-e

---

⚠️ Translational Risk Warning: In vitro efficacy does not equal in vivo or clinical efficacy. Feehan et al. (2026) effective PLP concentrations (125–500 μg/mL) are 2–3 orders of magnitude above physiologically achievable serum PLP via oral intake. The scientific contribution of the anchor study is mechanistic insight, not a basis for high-dose vitamin B6 supplementation in cancer. Any high-dose vitamin supplementation sought for cancer prevention or treatment should be undertaken only under medical supervision, and—outside of prescription contexts such as ATRA—currently lacks clinical evidence of benefit. This report does not constitute medical advice.