Evidence desk · Supplement
Vitamin D3 + K2
The verdict
Vitamin D is real medicine for one job: correcting a deficiency. Raise a low level toward the normal range and you fix the thing that was actually broken. Push past repletion and the benefit curve goes flat — the big prevention trials (notably VITAL, ~25,871 people) found no reduction in heart disease or cancer, at any baseline level.
So the rule is dose to the number, not above it. You're already at 59 ng/mL — replete. Hold there; don't chase a bigger number for a payoff that isn't in the data. The "must pair with K2" pitch is mostly marketing: plausible mechanism, thin human outcomes. Optional, low-value, not required.
Before you read on
Your blood vitamin D is 59 ng/mL. Your friend's is 18 and feels run-down. You both start the same 5,000 IU pill. Whose level should you expect to do more — and why might that be the whole story of this supplement?
The claims, sorted
Vitamin D gets sold for a long list of things. They don't share a verdict, so split them before you judge any single headline.
| Claim | Best evidence | Holds up? |
|---|---|---|
| Fix deficiency (bone, muscle, fatigue) | Decades of repletion data | Yes — where it's low |
| Prevent heart disease | VITAL RCT (n≈25,871) | No |
| Prevent cancer | VITAL + meta-analyses | No (incidence) |
| Higher level = more benefit | No threshold effect in VITAL | No |
| D needs K2 to "work safely" | Mechanism + small surrogate trials | Weak / optional |
Where the benefit actually lives
This is the same shape as the fish-oil story, drawn in blood levels. Vitamin D is genuinely useful when you're deficient — bone health, muscle function, the classic deficiency syndromes. Lift a low level (say, under 20 ng/mL, the NIH/IOM sufficiency cutoff) up toward 30+ and you've corrected a real shortfall. That part isn't controversial.
The error is assuming the line keeps climbing. It doesn't. Once you're repleted, more vitamin D buys little or nothing — the curve flattens. Many labs and influencers quote a "40–60 ng/mL optimal" range, but that target comes from observational data (people with higher D tend to be healthier), which is exactly the kind of association that doesn't survive a randomized test. When you actually randomize people to extra D, the hard outcomes don't move.
VITAL is the trial that settled the prevention question: tens of thousands of adults randomized to 2,000 IU/day of D3 or placebo for ~5 years. No drop in cardiovascular events, no drop in cancer incidence, and — importantly — no benefit even in people who started with lower levels. That last point is the one marketing skips. If raising D prevented disease, the deficient would have gained the most. They didn't.
The K2 pairing: plausible, not proven
The pitch goes: vitamin D pulls calcium into the blood, and vitamin K2 (usually MK-7) directs that calcium into bone instead of arteries — so you "need" K2 to take D safely. The biology is real enough on paper. But the human evidence for outcomes — fewer fractures, less arterial calcification, fewer cardiac events from adding K2 — is thin and mostly rests on surrogate markers, not endpoints that matter.
So K2 lands as optional and low-value: harmless, biologically plausible, but not a must-have, and not something to add for its own sake. If your D3 already comes bundled with K2, fine. If it doesn't, you're not missing a proven benefit by skipping it.
A note on dose, toxicity, and the wrong vitamin
Vitamin D toxicity is real but takes sustained very high intake — think tens of thousands of IU daily for a long stretch — pushing blood levels far above yours and causing hypercalcemia. At 59 ng/mL you're nowhere near that line; it's a reason not to keep escalating, not a reason to fear your current dose. And one disambiguation: your tretinoin is topical vitamin A, a completely separate molecule. It has nothing to do with vitamin D dosing — don't let the "vitamin" label blur them together.
By phase
Vitamin D is essentially phase-invariant. Unlike collagen or a cut-specific tool, the logic doesn't change with where you are in the fat-loss arc: keep your level in range, recheck periodically, don't chase. The phase boxes below say almost the same thing on purpose. Your specific call is in For you.
Phase 1 · Aggressive cut (Zepbound)
Maintain. Lower food intake can shave nutrient buffers, so this is the phase to actually verify the level rather than assume it. You're already replete (59), so the job is holding range, not adding. No cut-specific reason to push the dose up.
Phase 2 · Building muscle
Same. Adequate D supports bone and muscle function, which matters when training hard — but "adequate" is the whole point. Being above range doesn't add an anabolic edge the data can find. Hold the number.
Phase 3 · Maintenance
Same again. Keep a steady dose that holds you mid-range, recheck once or twice a year, adjust only if the level drifts. This is a set-and-monitor supplement, not one to titrate by feel.
For you
Your last panel put vitamin D at 59 ng/mL — replete, comfortably inside even the aggressive 40–60 target. Your VDR reduced-function genotype is a fair reason to aim for a sensible mid-range rather than scraping the 20 ng/mL floor; receptors that respond a bit less to the same level make "comfortably sufficient" the smarter target. But that argues for landing in 40–60, which you already have — it does not argue for climbing higher. The genotype changes your target, not the shape of the curve.
My call — hold the number, spend attention elsewhere
Keep whatever dose holds you in the ~40–60 band, and no more. You're at 59; that's the goal line, not a checkpoint on the way to 80. VITAL is explicit that more D doesn't buy fewer cardiac or cancer events, deficient or not — so escalating chases a payoff that isn't there, and at high enough doses you'd eventually trade it for hypercalcemia risk. Recheck once or twice a year to confirm you haven't drifted.
On K2: optional. If it's already bundled into your D3 softgel, keep it — it's harmless and the mechanism is plausible. Don't go buy a separate K2 for its own sake; the outcome evidence isn't there. And the bigger point for your cardiovascular risk: the lever that matters isn't vitamin D at all — it's ApoB 89. That's where real prevention attention belongs, not in nudging a D level that's already fine. Note too that Rhonda Patrick overstates the 40–60 target and leans on a confounded dementia association — useful to know whose framing you're hearing when "optimize your D" comes up.
Profile used: 41, South Asian male, 5′9″, Phase 1 cut on Zepbound; vitamin D 59 ng/mL (replete), ApoB 89, HOMA-IR 3.65, hs-CRP 1.9, VDR reduced-function genotype; tretinoin (topical vit A, unrelated) on board. Tell me if any of that moves and I'll update.
Recall check
- What is the one job vitamin D is clearly proven to do — and what shape does the benefit curve take past that point?
- VITAL randomized ~25,871 people. What did it find for heart disease and cancer, and what did it find by baseline D level?
- Where does the "40–60 ng/mL optimal" target come from, and why is that a weaker kind of evidence?
- Is the D + K2 pairing a must-have? What's the strongest and weakest part of the K2 case?
Explain it back
In one sentence: explain why someone at 59 ng/mL gains nothing from doubling their vitamin D dose, using the words "deficiency," "flat," and "VITAL."
Check it yourself
This verdict is just the toolkit applied. Re-derive it with the evidence ladder, surrogate vs hard outcomes, publication bias & funding, and is it strong enough to act? See it sit in the wider stack via the protocol and operating filters, and compare the same "maintain, don't escalate" logic in fish oil.
Sources · Manson et al., VITAL trial, NEJM 2019 (vitamin D and cardiovascular/cancer outcomes) · NIH Office of Dietary Supplements — Vitamin D fact sheet · Reviews of vitamin K2 (menaquinone) for bone and arterial outcomes report mainly surrogate-marker effects with limited hard-outcome data. Evidence summaries, not medical advice — confirm anything that touches your medication or labs with your prescriber.