Cancer · VITD-CANC-02 · v2
Across these trials vitamin D consistently points in a helpful direction for cancer death, but almost no single trial reaches a clear, reliable result on its own. The largest and most relevant trial [1] found fewer cancer deaths with vitamin D, but the result was not firm enough to rule out chance. A related analysis of that same trial [4] did reach a firm result, but it counted cancers that had spread as well as cancers that killed, so it does not isolate death alone. Several trials that looked directly at survival [7], [19], [20] found no clear difference, partly because many participants already had healthy vitamin D levels before starting. The most striking benefit [21] came from a study that modelled all deaths (not just cancer deaths) and only in people who were short of vitamin D. The panel concluded the signal is real in direction and biologically plausible, but depends heavily on who is treated, at what dose, and for how long.
The largest general-population trial [1] reported death from cancer with a hazard ratio of 0.83 (95% CI, 0.67 to 1.02), favourable but with a confidence interval crossing 1.0. A pre-specified secondary analysis of the same cohort [4] found a significant reduction in advanced (metastatic or fatal) cancer, HR 0.83 (95% CI, 0.69-0.99; P = .04). Dedicated survival endpoints were null elsewhere: digestive-tract overall survival [7] gave HR for death 0.95 (95% CI, 0.57-1.57), the Finnish trial [19] showed no significant secondary-endpoint or mortality difference, and the WHI calcium-plus-vitamin-D analysis [20] gave HM-specific mortality HR 0.77 (95% CI, 0.53-1.11). An observational trial emulation [21] projected a major all-cause mortality reduction confined to insufficient (HR 0.85, 95% CI 0.79-0.91) and deficient (HR 0.79, 95% CI 0.72-0.87) subsets. Positive effects otherwise concentrate in post-hoc subgroups [17], [18], [23]. The contested point: The division is whether a directionally favourable but non-significant cancer-death estimate [1], plus a significant composite that bundles non-fatal metastatic disease with death [4], amounts to evidence that vitamin D lowers cancer-specific mortality. The Sceptic holds that no primary RCT isolates a significant cancer-death reduction in its main population, while the favourable signals are composite [4], subgroup [17], [18], [23], or all-cause and observational [21].
All 24 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from Vitamin D, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
The single most direct cancer-mortality endpoint, VITAL death-from-cancer HR 0.83 (0.67-1.02) [1], is not statistically significant; the confidence interval crosses 1.0.
The only significant mortality-adjacent RCT result [4] is a composite of metastatic OR fatal cancer, so it does not isolate cancer death; non-fatal metastatic events inflate the signal (indirectness).
Where it remains uncertain. No primary RCT shows a significant cancer-specific mortality reduction in an unselected population; dedicated survival endpoints [7], [19], [20] are null and the strongest effect [21] is all-cause and confined to deficient subsets.
Aleatory
Even the largest trial [1] rests on 341 cancer deaths, and small survival trials [7] (N=417), [12] (N=155), [14] (N=72) have intervals wide enough that chance alone can produce or hide a difference.
Epistemic
Benefit may depend on starting vitamin D level; replete populations (mean 75 nmol/L in [19]) show null results, while deficient subsets [21] show large effects, but the corpus does not resolve the true effect in the deficient.
Inferential
The one significant RCT result [4] counts metastatic OR fatal cancer, so a death-specific inference cannot be drawn from it directly.
The load-bearing papers from a corpus of 24, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Largest, most direct cancer-mortality estimate; favourable point estimate but non-significant
“for death from cancer (341 deaths), 0.83 (95% CI, 0.67 to 1.02)”
Panel auditInternal validity: high · scope of inference: US men ≥50 and women ≥55, generally vitamin D replete, 5.3-year follow-up.Key limitations: cancer death secondary endpoint · CI crosses 1 · largely replete population
Pre-specified composite including fatal cancer is significant, but mixes non-fatal metastatic disease with death
“a significant reduction in advanced cancers (metastatic or fatal) was found for those randomized to vitamin D compared with placebo (226 of 12 927 assigned to vitamin D [1.7%] and 274 of 12 944 assigned to placebo [2.1%]; HR, 0.83 [95% CI, 0.69-0.99]; P = .04)”
Panel auditInternal validity: high · scope of inference: US men ≥50 and women ≥55, with strongest effect in normal-BMI individuals.Key limitations: composite endpoint not fatal cancer alone · secondary/post hoc analysis
Dedicated OS endpoint in digestive-tract cancer null
“The 5-year overall survival in the vitamin D vs placebo groups was 82% vs 81% (HR for death, 0.95; 95% CI, 0.57-1.57; P = .83)”
Panel auditInternal validity: high · scope of inference: Postoperative stage I-III digestive tract cancer patients at one Japanese hospital.Key limitations: single center · OS secondary endpoint · subgroup interactions exploratory
The abstract reports all-type cancer incidence, not cancer death, so the mortality contrast is not given.
Panel auditInternal validity: medium · scope of inference: Healthy postmenopausal rural women ≥55 with mean baseline 25(OH)D 32.8 ng/mL.Key limitations: incidence not mortality · combined with calcium · underpowered
4 of 24 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
Consistent directional benefit with a near-significant flagship estimate and a significant composite, but no significant cancer-specific mortality reduction in any overall RCT population, warrants an Uncertain verdict.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
RCT powered for cancer death in deficient populations. The largest projected benefit [21] is confined to insufficient or deficient people, yet no randomized cancer-mortality trial in this corpus enrolled such a population. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A randomized placebo-controlled trial recruiting only participants with low baseline 25(OH)D, with cancer-specific mortality as a pre-specified primary endpoint and adequate follow-up.
Open question: Does vitamin D supplementation reduce cancer-specific mortality in people with baseline vitamin D deficiency? Why it is still open: The benefit in [21] is modelled, all-cause, and observational, and the direct RCT [1] enrolled a largely replete population.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 26 Oct 2026 · protocol v0.19