| 1 |
OR13G1–CaMKK2–AMPK and ferroptosis |
↑ OR13G1 expression; ↑ CaMKK2/AMPK signaling; ↑ ferroptosis |
Tumor-bearing mice showed reduced colonic inflammation and pathological injury; normal-cell selectivity remains incompletely established |
Suppression of colorectal tumor development and cancer-cell growth |
Direct evidence for both whole oil and PA in a 2026 study using colorectal cancer cells and an inflammation-associated mouse cancer model. Independent replication is needed. |
| 2 |
Mitochondrial apoptosis and cell-cycle control |
↓ Mitochondrial membrane potential; ↑ caspase activation; ↑ cell-cycle arrest |
Some studies included nonmalignant comparator cells; a human therapeutic window is unestablished |
Apoptotic death and reduced proliferation |
Predominantly PA evidence. Mitochondrial depolarization does not by itself establish mitochondrial permeability transition pore opening. |
| 3 |
NF-κB–MCL-1 survival signaling |
↓ NF-κB activation; ↓ MCL-1; ↑ apoptosis |
Comparable modulation and safety at anticancer exposure are unestablished |
Reduced antiapoptotic signaling |
PA evidence from DU145 and PC-3 prostate cancer cells and mouse xenografts. MCL-1 overexpression partly reversed the apoptotic effect. |
| 4 |
GNAI1–ARRB1 and autophagic cell death |
↓ GNAI1–ARRB1 interaction; ↓ ERK/JAK2–STAT3/mTOR signaling; ↑ autophagic death |
Effects on normal tissues require further characterization |
Disruption of survival signaling and induction of lethal autophagy |
Mechanistic PA evidence in non-small-cell lung cancer. Autophagy can support survival in other settings; this is not a universal direction of effect. |
| 5 |
EGFR and MAPK signaling |
↓ EGFR-associated signaling; altered MAPK activity; ↑ apoptosis and cell-cycle arrest |
Insufficient evidence for a consistent normal-cell response |
Reduced growth signaling |
PA evidence in A549 lung cancer models. Individual MAPK branches should be recorded separately rather than assigning one direction to all MAPK signaling. |
| 6 |
ROS-mediated DNA damage |
↑ ROS; ↑ DNA-damage response; ↑ apoptosis |
No established selective pro-oxidant effect across normal tissues |
Oxidative stress-associated growth inhibition |
PA evidence in parental and vincristine-resistant lung cancer cells. This mechanism is distinct from the whole-oil ferroptosis finding. |
| 7 |
HDAC2 and c-MYC |
↓ HDAC activity; ↓ HDAC2 expression; ↓ c-MYC |
Normal-cell epigenetic effects are insufficiently characterized |
Reduced proliferation and altered cell-cycle regulation |
PA evidence in colorectal cancer cells. Reduced enzyme activity and expression do not establish selective direct binding to HDAC2. |
| 8 |
Migration and invasion |
↓ Migration and invasion; ↓ selected MMPs; ↑ E-cadherin in melanoma models |
Effects on normal migration and wound repair cannot be inferred from tumor assays |
Reduced invasive phenotype |
Mostly PA evidence. Cell migration assays do not establish prevention of metastasis in patients. |
| 9 |
Cisplatin combination |
↑ Combination inhibition of colony formation and migration in melanoma experiments |
Human combination toxicity and pharmacokinetic interactions are unknown |
Experimental chemosensitization |
PA evidence from B16F10 melanoma research. No clinical combination regimen is established; whole oil should not be assumed equivalent. |
| 10 |
Clinical Translation Constraint |
Experimental activity depends on preparation, concentration, and tumor model |
Systemic safety at proposed anticancer exposures is unestablished |
Limits clinical interpretation |
Variable oil composition, formulation-dependent absorption, high experimental concentrations, limited selectivity data, and absence of human anticancer trials prevent assignment of a therapeutic dose. |