calcium/vitamin D decreases cancer but increases cardiac mortality

 A recent analysis from the Women’s Health Initiative found that calcium with vitamin D supplementation was associated with a decrease in cancer mortality, but an increase in cardiovascular mortality (see vit d dec cancer wom health initiative AIM2024 in dropbox, or doi:10.7326/M23-2598)

 Details:

-- 36,282 postmenopausal women in 40 trial sites across the United States were randomized to 1000 mg of calcium carbonate (400 mg of elemental calcium) plus  400 IU (10 mcg) of vitamin D₃ (CaD) daily versus placebo

--this study represents the long-term follow-up of the initial 7-year randomized intervention trial of calcium and vitamin D supplementation in postmenopausal women who had no prior history of breast or colorectal cancer (see commentary below for more details about the original study)

    -- the current study is a subgroup analysis that includes personal supplement usage of multivitamins (with or without minerals), other supplement mixtures, and single supplement calcium or vitamin D

    -- this data was assessed through interviewing at the trial screening, and years 1, 3, 6 and  9.

  -- medication adherence was assessed by pill count in the clinic

 

-- Mean age 62, 82% white/13% Black/6% Latina, 34% had a college degree or higher, median family income $43,000

-- smoking: never 52%/past 38%/current 10%, hysterectomy 41%, hormone replacement therapy: never 55%/past user 18%/current user 26%

-- self-reported health excellent 17%/very good 41%/good 34%/poor 8%

-- diabetes 7%, history of cancer 4%, mean waist circumference 91 cm, BMI 28

-- mean healthy eating index 64 (0-100, the higher the better)

-- median dietary calcium 710 mg

-- median dietary vitamin D 144 IU

-- mean blood pressure 128/76, median physical functioning 90 (0-100, the higher the better)

-- geometric mean serum 25(OH)D level 38 nmol/L (15 ng/mL)

 

-- These demographic and health-related baseline values were stratified to those not taking versus taking prior supplements, with the following significant differences between these groups

    -- personal (non-study) supplement use at follow-up:

        -- year 3: 27% versus 77%

        -- year 6: 43% versus 81%

        -- year 9: 55% versus 84%

    -- more supplements at baseline were taken by white individuals, those taking hormone replacement therapy, those eating more healthfully, those having more dietary calcium intake (median 750 versus 710 mg),  mean serum 25(OH) D level  (59 versus 39 nmol/L, equivalent to 24 ng/mL versus 16 ng/mL), median episodes of moderate to strenuous physical activity per week (8 versus 5)

 

-- Measurements: incidental colorectal, invasive breast and total cancer; disease-specific and all-cause mortality; total cardiovascular disease; and hip fractures

-- analysis was stratified by personal supplement use at baseline

-- cardiovascular events were defined as myocardial infarction, coronary death, stroke, heart failure, angina, peripheral vascular disease, carotid artery disease, coronary revascularization, and other cardiovascular death; these cardiac events were centrally adjudicated through 2010

 

--median follow-up of 22.3 years

 

Results:

-- women randomized to CaD versus placebo:

    -- cancer mortality: 1817 versus 1943 deaths, 7% decrease, HR 0.93 (0.87-0.99)

        -- cancer mortality in those who took at least 80% of study pills: 11% decrease, HR 0.89 (0.80-0.99)

    -- cardiovascular mortality: 2621 versus 2420 deaths, 6% increase, HR 1.06 (1.01-1.12)

        -- cardiovascular mortality: did not change much in those who had higher adherence, HR 1.07 (0.97-1.18), though this did not reach statistical significance

    -- no significant effects on the other measures:

        -- no difference was found in exploratory analyses in cancer-specific mortality rates including deaths from colorectal, breast, and other cancers

        -- all-cause mortality: 7834 versus 7748, HR 1.00 (0.97-1.03)

        -- hip fracture: could not be assessed given that only a subset of participants were evaluated

        -- total CVD:  randomization was insufficient for results

 

-- Results stratified by women who did versus did not take supplements:

    -- results were pretty consistent between baseline and year one pre-randomization: 29% initiated and 19% ceased taking supplements between these visits

    -- women who did not use supplements prior to the intervention and who had been randomly assigned to receive CaD had:

        -- 32% decrease in colorectal cancer, HR 0.68 (0.46-0.99)

        -- 25% decrease in invasive breast cancer, HR 0.75 (0.60-0.93)

        -- 15% decreased risk of total cancer, HR 0.85 (0.75-0.96)

        -- no effect on CVD or mortality (total, cancer, or CVD-related)

    -- results during cumulative follow-up of these women:

        -- 31% decreased risk of colorectal cancer, HR 0.69 (0.54-0.87)

        -- 19% decreased risk of invasive breast cancer, HR 0.81 (0.71-0.94)

        -- 11% decreased risk of total cancer, HR 0.89 (0.82-0.96)

        -- no significant change in cancer mortality, HR 0.93 (0.82-1.04)

    -- BUT, women who had taken personal supplements prior to being assigned to CaD versus placebo:

        -- 9% increase CVD mortality (1800 versus 1633 deaths), HR 1.09 (1.02-1.16)

 

Commentary:

-- the original Women’s Health Initiative study was designed to assess the effectiveness of CaD supplementation in  reducing the incidence of bone fracture and colorectal cancer

    -- medication adherence was assessed by pill count in the clinic

    -- median trial-specified intervention was 7.0 years

-- the current study was a subgroup analysis involving personal supplement usage including multivitamins (with or without minerals), other supplement mixtures, and single supplement calcium or vitamin D; this data was assessed through interviewing administrated at trial screening, and years 1, 3, 6 and 9. And there was extended follow-up at 13.5 and 16.2 years

 

-- Overall the study found a significant 7% decrease in cancer mortality but an overall 6% increase in cardiovascular mortality in postmenopausal women who were stratified to taking calcium and vitamin D supplementation

    -- also, those who took >80% of their meds had an 11% decreased cancer mortality, more than the 7% decrease in the whole group

-- however, in their post hoc analysis stratified by the use of supplementation before randomization, they found that those not taking supplements at baseline had an even more impressive 11% decrease in total cancer as well as a 31% decrease in colorectal cancer and a 19% decrease in invasive in breast cancer. And, there was no increase in cardiovascular mortality

 

-- A meta-analysis of 10 RCTs (see vit d dec total cancer incidence and mortality AnnOnc2019 in dropbox, or doi:10.1093/annonc/mdz059) found the following
    -- total cancer incidence (10 trials, 6537 cases): not statistically significant difference by vitamin D intervention

    -- total cancer mortality (5 trials, 1591 deaths): 25(OH)D levels attained were 54–135 nmol/l (22-54 ng/mL)  in the intervention group], finding:

        -- 13% decrease in cancer mortality, RR 0.87 (0.79–0.96), which was largely attributable to interventions with daily dosing (as opposed to infrequent bolus dosing)

        -- but no significant difference by the achieved 25(OH)D levels: 0.88 (95% CI, 0.78–0.98) for <100 nmol/l  (40ng/mL) and 0.85 (0.70–1.03) for >100 nmol/l

            -- note: there were no granular data here: what was the array of achieved 25(OH)D levels? The 2 buckets of <100 vs >100 nmol/L (40ng/mL) does not provide really useful information (was the difference between 99 vs 100 nmol/L???)

 

-- the VITAL study (https://www.nejm.org/doi/full/10.1056/NEJMoa1809944) of 25,871 men and women randomized to 2000 IU vitamin D and omega-3 fatty acids 1 g/d and followed 5.3 yrs did not find a statistical benefit in the intervention group

    -- a secondary analysis of the VITAL study, however, that assessed a composite incidence of metastatic and fatal invasive total cancer, found a significant 17% reduction in these advanced cancers, HR 0.83 (0.69-0.99), p=0.04. However, in those with BMI<25, there was a 38% reduction, HR 0.62 (0.45-0.86), but no significant difference in those with higher BMIs  (https://pubmed.ncbi.nlm.nih.gov/33206192/ )

        -- even with “sufficient” vitamin D levels at baseline (mean 25(OH)D levels for all were 31 ng/mL), there was a significant decrease in metastatic/fatal cancer with vitamin D supplementation in those with lower BMI

            --but, older observational, non-RCT studies have suggested that cancer benefit was associated with higher 25(OH)D levels, in the >40-50 ng/mL range]

            --would the cancer benefits have been more impressive if people who were actually vitamin D deficient had their 25(OH)D levels raised to the more than sufficient level?

-- another meta-analysis, this one with 50 trials and 74,655 participants, found a statistically significant 15% reduction in cancer deaths, RR 0.85 (0.74-0.97),  though there were better results in studies using vitamin D₃ vs vitamin D₂ for all-cause mortality (https://www.bmj.com/content/bmj/366/bmj.l4673.full.pdf )

 

-- as noted, those people on CaD in this WHI study had an increase in CVD mortality, but by subgroup analysis, this was confined to women who had been taking supplements prior to the study. Not sure why this is the case, but:

    --There is some evidence that calcium supplements, but not calcium from food, might acutely raise plasma calcium levels, and thereby increase the risk of calcium artery calcifications, as found in the MESA study (also finding more Mis); other studies have associated calcium supplementation with abdominal aorta calcifications. Calcium in food, by contrast, is associated with decreasing blood pressure, vascular disease, strokes, and coronary artery disease

    -- one study found that taking 1 g of calcium as either citrate or lactate gluconate raises ionized calcium from a mean of 1.22 to 1.30 mmol/L. And the supposition is that these increased plasma calcium might alter the calcium homeostasis and lead to ectopic calcifications

    -- for a systematic review, see calcium supplements and heart disease Athero2020 in dropbox, or doi.org/10.1016/j.atherosclerosis.2020.01.008 )

-- in this WHI study it was specifically the women who had previously been on calcium supplements were the ones who had increased cardiovascular events; perhaps the issue here is the extent of supplement taking, which likely was lower for those women not on supplements at the start of the study. But there is no information on that.

 

--the information in the medical literature on the benefits of vitamin D supplementation are quite mixed, and several of the studies were not so great (eg, not much change in 25(OH)D levels in the intervention group, or both groups with quite good 25(OH)D levels at baseline, etc), but a few points here:

    -- vitamin D receptors are found in essentially all of the cells in the body, including pretty much all of the cells in the immune system, both innate and adaptive parts: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5031366/

    -- observational studies have found vitamin D deficiency to be associated with an array of autoimmune diseases, as well as increased susceptibility to infection

        -- lower rates of upper respiratory infections in one study (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3543548/ )

        -- association of vitamin D with an array of autoimmune diseases: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6047889/

            -- and maps suggest that areas of little sunlight tend to have higher incidences of these problems

    -- there is a reasonable (I think) evolutionary argument: humans evolved in the rather sunny fertile crescent, then (for perhaps not-so-clear reasons why so many of us live in areas of the world like Boston where there is insufficient sunlight most of the time to get adequate vitamin D (see https://www.bu.edu/articles/2010/sunny-dispositions-and-vitamin-d/#:~:text=%E2%80%9CWe%20showed%20that%20in%20Boston,at%20five%20in%20the%20afternoon.%E2%80%9D , per a local vitamin D researcher), making the following points:

        -- there are lots of genes (approx 2000, up to 1/6 of the human genome) that are directly or indirectly regulated by vitamin D

        -- half the people in the US do not get adequate vitamin D, with estimates that 1.5-2 billion people in the world are vitamin D deficient

        -- in Boston, we cannot make vitamin D from November through March. And, even in the summertime, there is insufficient UV light to make vitamin D at 8AM or 5PM.  Best getting sun between 10AM and 3PM for 10-15 minutes, depending on the season

    -- and, of course, there are negatives to sunlight in terms of skin cancer (especially too much)

    -- so, a reasonable option is taking the vitamin D pills. Which has the added advantage of being really safe, with rare concern for overdoses. And they are really cheap

    -- my own anecdotal experience in Boston (based on my uniform testing of all of my patients for their 25(OH)D levels for several decades) is that it is pretty rare to find anyone not on supplements who has a vitamin D level >30ng/L, about 20-30% have levels between 20-30 ng/dL, and the rest  have levels <20 ng/mL, with about 10% <12 ng/mL

 

Limitations

-- this was a quite healthy group of predominantly white women in the WHI, limiting the generalizability of the results to the larger population

-- there was an increase in women taking supplements over the course of the study: 27% of women in the placebo group were taking supplements at the initiation of the study, but this increased to 55% by year 9.  We do not know how this affected the results. Did those women starting supplements but in the placebo group have benefits (eg, were those women not taking supplements in both groups even worse off?)

-- we also do not know exactly what supplements they were taking. Calcium with vitamin D? just calcium or just vitamin D? other supplements? No information provided in the article

-- we do not have any information about what the participants were doing after the 9-year mark (13 years later). Which ones were on supplements over the 23 year assessment? How much? What were their vitamin D levels?

-- also, the cardiovascular mortality might have been skewed by hormone replacement therapy, which was 38% in those already on supplements at baseline vs 27% in those not on baseline supplements. Perhaps ironically, it was the WHI that found increased cardiovascular risk  in women on hormone replacement therapy, though there was no mention of that in this current article…..

 

So,

-- as is clear in my multitude of prior blogs on vitamin D (https://gmodestmedblogs.blogspot.com/search?q=vitamin+d), I do think there is really only benefit and not much harm in suggesting vitamin D supplements in those deficient (unless one is taking on the order of 60,000IU per day); and if there is any question, just check a serum calcium at some point)

    -- I also check calcium levels in those with hyperparathyroidism on vitamin D supplements when indicated, though often the PTH levels decrease with vitamin D supplementation https://pubmed.ncbi.nlm.nih.gov/24423366/ )

-- it is physiologically clear that there needs to be enough calcium intake, though best from foods

-- this WHI study tangentially further strengthens the argument for vitamin D supplementation, and argues that we do not suggest calcium supplementation

 

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

  

to get access to all of the blogs:  go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

  -- click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

  -- or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them

 

if you would the article, please email me.

 

please feel free to circulate this to others. also, if you send me their emails (gmodest@bidmc.harvard.edu), i can add them to the list