Aim of this post:
- to try to trash the Nordic studies with their J-/U-shaped risk curves.
- to look at the only studies that really matter - vitamin D & survival.
There are two main sets of PCa vitamin D study findings. Many studies have noted a reduction in PCa risk as blood levels increase. (However, there are not enough men with levels >40 ng/mL to draw any conclusions from that group. They get lost in the upper quartiles/quintiles, where most men are barely D-sufficient.) The second, less numerous set of studies, have reported a U- or J-shaped risk curve. These have mostly come out of Scandinavia, where D levels are very low for most of the year. The sweet spot on the curves includes men who would normally be considered D-insufficient. & the risk begins to appear at the low end of sufficiency. There is a third set of studies that found D to be largely irrelevant.
The earliest report of a U-shaped curve came from Finland in 2004 [1], but the study also included data from Norway & Sweden. Big study. Since the concept of D as a risk factor has not lost traction, it's worth looking into the innards.
"... using serum banks of 200,000 samples. We studied serum 25(OH)-vitamin D levels of 622 prostate cancer cases and 1,451 matched controls and found that both low (</=19 nmol/l) and high (>/=80 nmol/l) 25(OH)-vitamin D serum concentrations are associated with higher prostate cancer risk."
Using the more common U.S. blood test units: "low (</=7.6 ng/mL) and high (>/=32 ng/mL)" were "associated with higher prostate cancer risk."
Deficiency is defined as <20 ng/mL, but there was no excess risk between 7.6 & 20. More remarkable is that "high" included everyone who was neither deficient nor insufficient. Sufficiency begins at 32 ng/mL. Very hard to take the results seriously. How can mere sufficiency be risky, & should one therefore aim for insufficiency?
The author, Tuohimaa, suggested that:
"a high vitamin D level might lead to vitamin D resistance through increased inactivation by enhanced expression of 24-hydroxylase" At least he didn't suggest that D stimulates PCa growth.
In the endocrine system, there is usually a feedback loop to stop secretion of a hormone when enough has been received. An exception is melatonin, which is going to stop at the end of night anyway. Prostate cells become increasingly resistant to melatonin after exposure, & this seems to be the kind of thing that Tuohimaa has in mind. But Vitamin D is quite different, in that conversion to the active hormonal form is controled from within the cell (i.e. is under intracrine control). Calcidiol is converted to calcitriol only as needed. Sure, the production of calcitriol triggers production of the enzyme that will ultimately clear it from the cell. But having more calcidiol outside does not mean that prostate cells will make more calcitriol inside. It is hard to see how the clearance enzyme could be chronically over-expressed, leading to resistance.
A potential problem in the study design is the way the blood samples were taken. Finland took no samples in the summer (very wise) & not many (18) in the fall (also good); mostly (352/588) in the winter. Norway took most (508/1,077) in the fall. Sweden took more (145/408) in the spring. Without serious supplementation, Scandinavian men would be deficient by the end of winter / start of spring. Many of those same men might achieve sufficiency during the summer - enough, perhaps, to take them through the fall. In other words, timing is particularly crucial at those lattitudes, in populations that mostly do not supplement.
Had the study taken two readings (late summer & late winter), the data would have been more useful, IMO. A single reading doesn't allow one to back into an average for an individual. With only one reading, an effort should have been made to bunch the samples into a single collection month, say. Problems appeared right away:
"In the Finnish study group, an increased risk was seen for the lowest compared to the highest quintile [odds ratio (OR) = 1.9 ...], whereas in the Norwegian and Swedish study groups, the pattern of risk was reversed: increased risks were seen for the highest compared to the lowest quintiles (OR = 1.4 ... & ... 1.7, ... respectively)."
To some extent, the team got around this glitch by settling on a reference range of 16-23.6 ng/mL. Odd to have a reference range where the mid-point is the dividing line between insufficiency & deficiency. But, in fact, this worked well for Sweden, which had 52% of cases in that range. For Norway too. But not Finland, where 52% of cases fell into the 8-15.6 ng/mL range. Anyway, by fudging things, they got a nice U-curve:
... D ... ... relative risk ...
<=7.6 1.5
8-15.6 1.3
16-23.6 1.0
24-31.6 1.2
>=32 1.7
As might be expected from the different collection patterns, the data for individual countries did not behave.
The safest spot for Swedes was 8-15.6, with 30% less risk than the reference range.
For Norwegians, it was better to be <=7.6; a 10% advantage over the reference. (!)
& yet the study was taken seriously.
More studies followed. From 2014 [2]:
"Our meta-analysis, for the first time, suggested significant positive relationship between high level of 25-hydroxyvitamin D and increased risk of prostate cancer"
"We identified 21 relevant publications from databases of PubMed and MEDLINE and included 11,941 cases and 13,870 controls in the meta-analysis. Overall studies revealed a significant 17 % elevated risk of prostate cancer for individuals with higher level of 25-hydroxyvitamin D".
The study that may have done most harm in the U.S. was the SELECT (Selenium and Vitamin E Cancer Prevention Trial) paper of 2014 [3]. Relative risks:
Vitamin D: <18 <23 <29 <36 >36
Gleason 7-10 1.00 0.63 0.66 0.79 0.88
Gleason 8-10 1.00 0.68 0.36 0.85 0.78
"It is notable that not a single, large (n cases>200) prospective study has reported a linear, inverse association between blood vitamin D concentrations and prostate cancer risk. Our results are similar to those from a study in European Nordic countries (18), which reported the lowest risk of prostate cancer among men with vitamin D concentrations of {16-24 ng/m)L, with higher risk among men with lower and higher values."
What is lacking, of course, is an examination of risk in the 32-100 ng/mL range. Not easy to do in a country that spends $400 million annually on sunscreen products, & considers 400 IU vitamin D to be an adequate supplement. Insufficiency is rampant.
In January, a paper from the Finnish Alpha-Tocopherol, Beta-Carotene Cancer Prevention (ATBC) Study was published [4]. Actually a U.S. NIH study using ATBC data.
"Recent epidemiologic evidence suggests that higher circulating vitamin D does not protect against prostate cancer and, in fact, may increase the risk of developing this malignancy. However, few studies have examined the most clinically relevant outcome, prostate cancer mortality."
"Men with higher serum 25(OH)D were less likely to die from their prostate cancer ..." A 28% decrease in mortality risk, comparing high to low quintiles.
From 2014 [5]:
"We found that veterans who were initially vitamin D deficient were significantly less likely to survive than those who were not initially deficient, and that both initial and follow-up vitamin D deficiency were associated with decreased likelihood of survival after prostate cancer diagnosis."
& a review paper from 2013 [6]:
"The relationship between vitamin D and cancer appears to be stronger for studies of cancer mortality than incidence."
"Here we review analytic epidemiologic studies investigating the relation between vitamin D, measured by circulating levels of 25-hydroxyvitamin D (25-OHD), and cancer survival. A relationship between low 25-OHD levels and poor survival is shown by most of the reviewed studies."
What we are unlikely to discover anytime soon, is the survial benefit of being at 70 ng/mL, say, versus 50 or 90 ng/mL.
-Patrick
[1] ncbi.nlm.nih.gov/pubmed/146...
[2] ncbi.nlm.nih.gov/pubmed/248...
[3] ncbi.nlm.nih.gov/pmc/articl...
[4] ncbi.nlm.nih.gov/pubmed/268...
Supplementing Vitamin D has no effect on fracture risk
More on Vitamin D, COVID-19
