Holy basil (tulsi): what the evidence says
For her partner, 10 September 2026, after a friend's papers in the research channel. Every claim carries its evidence level (L1 randomised trials, guidelines or meta-analyses; L2 single trials or prospective cohorts; L3 retrospective series; L4 case reports; L5 cells, animals and reasoning) and, in brackets, a friend's category: established, frontier, ancient or anecdotal. Sources are in the shared-papers thread and the addendum to research thread 11.
What it is
Holy basil, tulsi, Ocimum sanctum (also Ocimum tenuiflorum), is a leaf used in Ayurvedic and Siddha medicine for at least three thousand years, as a tea, a paste and a religious plant. Its oil is mostly eugenol (the clove compound) with caryophyllene, ursolic acid, rosmarinic acid, apigenin and carvacrol. (Ancient.)
What human studies show
A 2017 systematic review found 24 human studies, most small, on blood sugar, cholesterol, blood pressure, stress and immune markers, every one reporting benefit and none reporting an adverse event (Jamshidi 2017, L1 review of L2 to L3 studies; frontier). A literature that is positive in every study and free of side effects in every study is the signature of publication and design bias, so the findings should be read as "plausible modest effects on metabolism and stress", not as proof. Two effects in those studies matter for the patient: tulsi lowers blood sugar and reduces platelet stickiness. The trials registered today are for dyspepsia, stress and cognition, type 2 diabetes, sensitive skin, gum disease and frailty (ClinicalTrials.gov, 10 September 2026).
The cancer claim
- Leaf extracts and the essential oil slowed the growth, migration and invasion of pancreatic cancer cells in dishes and switched down the growth signals ERK, FAK and NF-kB; a water extract injected into the abdomen of immune-deficient mice carrying human pancreatic tumours, 6 mg per mouse twice a week (about 270 mg/kg), reduced tumour growth (Shimizu 2013, L5; anecdotal-lab).
- Eugenol kills osteosarcoma cells in dishes at 0.5 to 10 millimolar, concentrations a thousand-fold above anything a person can reach from tea or capsules (reviewed in Molecules 2023 and 2021, L5; anecdotal-lab).
- No trial of tulsi or eugenol has been run in cancer patients anywhere, for any cancer, sarcoma included; PubMed and ClinicalTrials.gov both checked on 10 September 2026. Nothing exists for chordoma.
So the cancer case for tulsi is real laboratory work that has not moved into people in thirteen years. That is the honest tier: a hypothesis.
The COX-2 connection
Tulsi's compounds inhibit COX-2, the enzyme of inflammatory pain, in the test tube. The patient already takes diclofenac, which blocks COX-2 about as selectively as celecoxib, round the clock. Chordoma does express COX-2 (two tissue series, L3), but no trial has treated chordoma by blocking it, and the randomised trials of celecoxib as cancer treatment in lung and breast cancer were negative (L1). Whatever COX-2 effect exists, she has it already; a herbal inhibitor on top adds bleeding and kidney load without adding effect. (Established.)
Safety
- Eugenol's accepted daily intake is 2.5 mg per kg (the WHO and FAO food-additive limit), about 130 mg a day for the patient. Above that it turns pro-oxidant and damages DNA in cells, and it is a contact allergen. Clove oil overdose causes liver failure in children (case reports, L4). A cup of tulsi tea is well inside the limit; concentrated oils, tinctures and capsule extracts are not standardised and can exceed it. (Established for the limit; anecdotal for the harms.)
- Platelets: the same antiplatelet effect that makes diclofenac stop before surgery applies to tulsi, so it is held for the same window: about a week before any biopsy or operation and one to two weeks after, until the nose and the flap have sealed. (Clinical reasoning from the trials, L5.)
- Blood sugar: it lowers glucose modestly; fine, but worth knowing on days she eats little.
- Kidney, stomach and heart: nothing found in the 24 studies, which is worth something even allowing for their bias.
Where it lands for the patient
- Now, before biopsy or surgery: not started, because of the platelet effect and the interaction with the diclofenac she needs for pain.
- After surgery, once the bleeding window has passed and the team knows: two or three cups a day of the leaf as tea is a reasonable comfort with a plausible small benefit on stress and appetite; the leaf, not oils or extracts.
- Not as a treatment for the tumour: there is no human evidence, and no trial in progress.