WHAT WOULD IT LOOK LIKE

IF EVERYONE HAD SUFFICIENT

VITAMIN D?

Robert P. Heaney, MD, FACP, FASN

Creighton University Osteoporosis Research Center

VIT D & CARDIOVASCULAR DISEASE
3.5
 1739 Framingham
Offspring members 80 % increase
3.0
 age: 59 yrs in risk
2.5 53 % increase

Hazard Ratio
 follow-up: 5.4 yrs in risk
 120 individuals developed 2.0
a CV event 1.5
 HR calculated against
25(OH)D values > 15 ng/mL 1.0

 Wang et al. Circulation 0.5

2008
0.0
< 10 < 15 > 15
ng/mL ng/mL ng/mL

CU ORC

2
BREAST CANCER RISK
 Case-control study 1.2

 1394 cases
1.0
 1365 controls
69 % decrease

Hazard Ratio
 Odds ratio for CA 0.8
in risk
inversely
0.6
associated with vit
D status [25(OH)D]
0.4
 Abbas et al.,
Carcinogenesis (2008)
0.2
29:93–99

0.0
30 –4
5
–6
0
–7
5 75
< 3 0 4 5 6 0 >

Serum 25(OH)D (nmol/L)

VITAMIN D & INFLUENZA*
35
 208 African-American,
postmenopausal women 30

 3 yr DB-RCT 25
 placebo or vit D3
20
 800 IU/d – 2 yrs
 2000 IU/d – 3rd yr 15

 basal 25(OH)D: 18.8 ± 7.5 10

 P < 0.002 5

0
Placebo Vitamin D

CU ORC
*Aloia & U-Ng (2007) Epidemiol & Infect
4
VITAMIN D & THE COMMON COLD*
25
 18,883 individuals
in NHANES-III
20
 tested association

% with URTI
between serum 15
25(OH)D & recent
URTI 10 29 % reduction
 P < 0.001
5
 association stronger
for those with 0
< 10 10–29.9 30+
asthma & COPD
Serum 25(OH)D (ng/mL)
CU ORC
Ginde et al., Arch Int Med 2009 169:
5
DIABETES & 25(OH)D
1.0
White
 Scragg et al., 2004 0.9 Hispanic
Diabetes Care 0.8
27:2813–18

Relative Risk
0.7
 NHANES-III
0.6
 6,228 adults
0.5
 plasma glucose
0.4
independently
predicted by BMI 0.3
& serum 25OHD 0.2
(fasting and 2 hr 0.1
post load) 0.0
1st 2nd 3rd 4th
25(OH)D Quartiles
CU ORC

6
NEONATAL VIT D & DIABETES*
 10,366 northern 3-fold higher risk
10
Finnish children
 2000 IU Vit D/d 1st
year of life 1

Relative Risk
 prevalence of
type I diabetes
assessed at age 31 0.1

 RR calculated vs. no
supplementation
0.01
88% lower risk u l ar u l ar k e ts
R eg re g ri c
Ir ?

*Hypponen et al., Lancet 2001;358:1500–03 Vitamin D Administration

Similar clinical study
results are being
published weekly
Two questions: –
How can a single nutrient have
such diverse effects in so many
different tissues ?
If these effects are correct, why
haven’t they been apparent
previously ?
Two questions: –
How can a single nutrient have
such diverse effects in so many
different tissues ?
If these effects are correct, why
haven’t they been apparent
previously ?
THE VITAMIN D ICEBERG

Ca economy

cell cycle regulation

gene control
THE VITAMIN D ICEBERG

endocrine

autocrine
VIT D – CANONICAL SCHEME

skin liver kidney gut

D3 25(OH)D3 1,25(OH)2D3 CaBP

CU ORC

13
 VIT D – EXPANDED SCHEME

endocrine kidney gut

1,25(OH)2D3 CaBP
skin liver
D3 25(OH)D3 various
tissues
periphery
autocrine 1,25(OH)2D3 cell
signals

CU ORC

14
Autocrine functions
AUTOCRINE ACTION
25(OH)D

1,25D
VDR

1,25D
VDR VDRE
RXR

Transcription

CU ORC

16
AUTOCRINE ACTION
25(OH)D

 cell proliferation
 cell differentiation
 apoptosis
 immune response VDRE
 24-hydroxylase

Transcription

CU ORC

17
AUTOCRINE ACTION
25(OH)D

~ 800 genes
have VDREs
VDRE

Transcription

CU ORC

18
AUTOCRINE ACTION
25(OH)D

25OHD 25OHD

1,25D 1,25D
VDR

VDRE
1,25D
VDR
RXR

Transcription

CU ORC

19
AUTOCRINE ACTION
25(OH)D

25OHD 25OHD

1,25D 1,25D
VDR

VDRE
1,25D
VDR
RXR

Transcription

CU ORC

20
This scheme means that each tissue
 has the amount of 1,25(OH)2D it needs
 when it needs it
 and is not dependent upon a “one-size-
fits all” systemic level of circulating
1,25(OH)2D
VITAMIN D & INNATE IMMUNITY*

activated
Toll-like
receptor

CU ORC
*Liu et al., Science 2006
22
VITAMIN D & INNATE IMMUNITY*
25(OH)D

bactericidal Cathelicidin
peptide

CU ORC
*Liu et al., Science 2006
23
VITAMIN D & INNATE IMMUNITY*
25(OH)D
 human
monocytes
in fetal calf
serum

the Vit D
1-a hydroxylase
Cyp27B1
VDR
the Vit D
receptor

CU ORC
*Liu et al., Science 2006
24
VITAMIN D & INNATE IMMUNITY*
25(OH)D
 human
monocytes
in fetal calf
serum
 fetal calf
serum is low
in both
Cyp27B1
25(OH)D & VDR
1,25(OH)2D
…………

CU ORC
*Liu et al., Science 2006
25
VITAMIN D & INNATE IMMUNITY*
25(OH)D
 human
monocytes
in fetal calf
serum
 add
1,25(OH)2D
to the Cyp27B1
system 1,25D VDR

Cathelicidin
Cyp24

CU ORC
*Liu et al., Science 2006
26
VITAMIN D & INNATE IMMUNITY*
25(OH)D
 human
monocytes
in fetal calf
serum
 add
25(OH) D to 25OHD

the system Cyp27B1

1,25D VDR

Cathelicidin
Cyp24

CU ORC
*Liu et al., Science 2006
27
VITAMIN D & TUBERCULOSIS
 human monocytes Cathelicidin mRNA
activated with M.
4
Tuberculosis and
incubated in human
3
serum serum 25(OH)D:
78 nmol/L
 African-American
2
 White serum 25(OH)D:
22 nmol/L
 African-American 1
with added
25(OH)D 0
A- A W A- A
+ 25D
CU ORC
*Liu et al., Science 2006
28
VITAMIN D & TUBERCULOSIS
these experiments show that:
 vit D is an essential mediator in the innate
immune response
 serum 25(OH)D is the critical variable
 at least some of the increased sensitivity to
infection in vit D-deficiency is due to reduction
in response to infectious agents because
25(OH)D is rate-limiting
 the greater tuberculosis susceptibility of blacks
is due in part to their low vit D status

CU ORC

29
If this new understanding is correct,
 do we see evidence of impaired function in
patients with low vitamin D status?
VITAMIN D & TUBERCULOSIS*
 67 pts with pulmonary Sputum Conversion (%)
TB 100

 standard treatment 90
for all
 in addition, 80

randomized to either
vit D 10,000 IU/d or 70

placebo 60

 P = 0.002
50
Placebo Vit D

CU ORC
*Nursyam et al., Acta Med Indones 2006
31
CELL MODELS

old: DNA in somatic cells functions

mainly to make faithful
copies for tissue repair or
replacement

DNA functions constantly in

new:
synthesis of needed cellular
apparatus
CU
CU ORC
ORC
HOW A CELL RESPONDS
Signal/
. . . but I do have
Demand
the plans for what
I need in my DNA
I don’tlibrary.
have . . .
the equipment
I need . . . .

newly synthesized
Response cellular equipment

CU ORC

33
HOW A CELL RESPONDS
25(OH)D
Signal/
Demand

1,25(OH)2D is the key

that unlocks the DNA library

newly synthesized
Response cellular equipment

CU ORC

34
PERSPECTIVE
 vitamin D is an integral component of the
mechanism whereby cells control gene
transcription in response to a variety of
extracellular stimuli
 adequate vitamin D status enables optimal
response to a broad variety of signals
 deficiency will manifest itself differently,
depending upon the tissue being stressed,
thus explaining the diversity of responses

CU ORC

35
Two questions: –
How can a single nutrient have
such diverse effects in so many
different tissues ?
If these effects are real, why
haven’t they been apparent
previously ?
VITAMIN D & INFLUENZA*
35
 208 African-American,
postmenopausal women 30

 3 yr DB-RCT 25
 placebo or vit D3
20
 800 IU/d – 2 yrs
 2000 IU/d – 3rd yr 15

 basal 25(OH)D: 18.8 ± 7.5 10

 P < 0.002 5

0
Placebo Vitamin D

CU ORC
*Aloia & U-Ng (2007) Epidemiol & Infect
37
Such differences would not be apparent in
ordinary medical practice because people
who don’t get sick do not see the doctor –
are not tracked and would not be recognized
as having been protected.

The protection is seen only when a cohort of

well individuals is followed prospectively.
Endocrine mechanism
A VITAMIN D THRESHOLD
0.5

ABSORPTION FRACTION
0.4

0.3

0.2

0.1

0.0
0 20 40 60 80 100 120 140 160

SERUM 25(OH)D (nmol/L)

CU ORC

42
A VITAMIN D THRESHOLD
0.5

physiological

ABSORPTION FRACTION
0.4
regulation no longer
limited by vit D
0.3
availability

0.2

0.1

0.0
0 20 40 60 80 100 120 140 160

SERUM 25(OH)D (nmol/L)

CU ORC

43
A VITAMIN D THRESHOLD
0.5

ABSORPTION FRACTION
0.4

0.3

0.2

0.1

0.0
0 20 40 60 80 100 120 140 160

SERUM 25(OH)D (nmol/L)

CU ORC

44
VITAMIN D – Sources

Body D3
? stores 25(OH)D
VITAMIN D – Sources

Body D3
? stores 25(OH)D
VITAMIN D – Sources

Body D3
150 stores 25(OH)D
VITAMIN D – Sources

Body D3
150 stores 25(OH)D

typical input,
all sources:
~2350 iu
VITAMIN D – Sources

Body D3
150 stores 25(OH)D

needed input,
all sources:
~4000 iu
 VIT D – EXPANDED SCHEME

endocrine kidney gut

1,25(OH)2D3 CaBP
skin liver
D3 25(OH)D3 various
tissues
periphery
autocrine cell
1,25(OH)2D3
signals

CU ORC

50
 VIT D – EXPANDED SCHEME

endocrine kidney gut

1,25(OH)2D3 CaBP
skin liver
D3 25(OH)D3 various
tissues
periphery
autocrine 1,25(OH)2D3 cell
signals
 Won’t calcitriol meet the
body’s need for vitamin D?

NO!
VIT D – CANONICAL SCHEME

skin liver kidney gut

D3 25(OH)D3 1,25(OH)2D3 CaBP

Why not Answer:

This is the value that just give 1,25(OH)
you can’t2give
D? enough
It’s the active
needs to be optimized to achieve
agent, isn’t
needed
it? levels.

CU ORC

53
Bone strength
Serum 25(OH)D and Hip BMD

 NHANES-III
Non-Hispanic whites
 Adults Age
20 – 49 yrs
 LOWESS plot
of difference Hispanics
from lowest
quantile

African-Americans

Bischoff-Ferrari HA. Am J Med 2004; 116: 634-9.

VITAMIN D & FRACTURE RISK
1.0
 N = 2,686
FRACTURE RELATIVE RISK –33%  ages 65–85
(hip, forearm, spine) 0.8

 5 yr RCT
0.6
 Vit D  800
0.4
IU/d
 Trivedi et al.
0.2 BMJ 2003;
326:469
0.0

0 25 50 75 100 125 150

(nmol/L)

CU ORC

56
 VITAMIN D & FRACTURES
Relative Risk
 meta-analysis 1.0
 9 RCTs
0.8
 Vit D doses
> 400 IU (but 0.6

none > 800 IU) 0.4

 n = ~ 32,000
0.2
 Bischoff-Ferrari et al.
Arch Int Med 0.0
(2009);169:551
Non-vertebral Hip

CU ORC

57
VITAMIN D & RISK OF FALLING*
1.0
 122 women
 Age: 63–99 0.8
–49%
 DB-RCT

Fall Risk
0.6
 Ca 1,200 mg/d
 Ca + 800 IU Vit D 0.4

 12 week duration
0.2
 25(OH)D 12 ng/mL
at baseline 0.0
Ca only Ca + D

CU ORC
*Bischoff et al. JBMR. 2003;18:343–351.
58
VIT D & NEUROMUSCULAR FUNCTION*
Performance Score
 1359 men & women; 9
mean age 75.5 8
 Amsterdam longitud. 7
aging study
6
 neuromuscular
5
performance
measured on a scale 4
of 0 to 12 (higher is 3
better) 2
 each step statistically 1
significant
0
<25 25–50 50–75 >75
*Wicherts et al. JBMR. 2005.
SERUM 25(OH)D
In brief, raising serum 25(OH)D
from 50 to ~80 nmol/L
improves Ca absorption, raises
BMD, and reduces falls and
osteoporotic fracture risk
OTHER CHRONIC DISEASES?
Disease Status of Evidence
 osteoporosis ++++
 osteoarthritis +
 falls/neuromusc. fcn ++++
 multiple sclerosis ++
 fibromyalgia-like syndrome ++
 type I diabetes ++
 insulin sensitivity ++
 cardiovascular disease +++
 periodontal disease ++++
 various cancers ++++
 tuberculosis ++++
 hypertension ++++

CU ORC

61
Cardiovascular effects
VIT D & BLOOD PRESSURE*
P < 0.02
 148 women, aged 150
P < 0.01
74 ± 1 P < 0.01
 DB–RCT

Systolic BP (mm Hg)

 baseline 25(OH)D < –5.7 –13.1
50 nmol/L
 treated for 8 wks
with:
Ca 1200 mg/d or 125
Ca + 800 IU vit D/d

0
Ca only Ca+D
*Pfeifer et al., JCEM 2001; 86:1633–37
INTERVENTION
VIT D & BLOOD PRESSURE*
 1811 men & women
10
with measured
25(OH)D levels**
 4 yrs’ observation

Relative Risk
 97 cases of incident 3.18
hypertension 1
 RR computed for
25(OH)D <15ng/mL
vs. >30 ng/mL

0.1
>30 <15
*Forman at al., 2007;Hypertension 49:1063
** Health Profs Follow-up Study & Nurses Health Study
Anti-promotion for cancer
 VITAMIN D & PROSTATE CA*
2.5

 13 yr
longitudinal 2.0

RELATIVE RISK
study
1.5
 19,000 men
 149 cases 1.0

prostate CA
0.5

0.0
1 2 3 4
*Ahonen et al., CancerCauses&Control 11:847-852 (2000) 25(OH)D QUARTILES
 VITAMIN D & PROSTATE CA*
2.5
 those below
the median 2.0
25(OH)D level

RELATIVE RISK
were 70% 1.5
more likely to
develop 1.0
prostate CA
than those 0.5
above
0.0
1 2 3 4
*Ahonen et al., CancerCauses&Control 11:847-852 (2000) 25(OH)D QUARTILES
COLORECTAL CANCER
1.0
 Nurses’ Health Study
 ages 46–78
0.8
 nested case-control

Odds Ratio
study 0.6
 193 incident cases
 25(OH)D measured 0.4
twice, prior to
diagnosis 0.2
 Feskanich et al., Cancer
Epidemiol Biomarkers Prev
0.0
2004 13:1502–08 16

0
7
2

4
2
2
t–

h–

h–
d–
d–
1s

4t

5t
3r
2n

25(OH)D Quintiles (with medians*)

*ng/mL
COLORECTAL CANCER
 5 prospective
1.0
studies
P < 0.001
 > 200,000 0.8

individuals

Odds Ratio
0.6
 430 cases
 ORs computed 0.4

for 25(OH)D
quantiles 0.2

 Garland et al, 0.0

2005 0 20 40 60 80 100 120

Serum 25(OH)D (nmol/L)

CU ORC

69
MAMMOGRAPHIC DENSITIES
1.0
 543 women aged 40–60
Vit D
 1989–90 Ca
0.8
 dietary intakes assessed

Odds Ratio
with FFQ 0.6
 odds ratios developed
for <30% vs. >70% of 0.4
film with densities
 [Berube et al., 2004; Cancer 0.2
Epidemiol Biomarkers Prev
13:1466–72]
0.0
1st 2nd 3rd 4th
Quartiles
CU ORC

70
VITAMIN D & CANCER*
 1179 healthy women
 aged 66.7 ± 7.3
 four year trial
 1032 finished (87.5%)
 baseline 25(OH)D: 71.8 nmol/L ± 20.3
 three treatment groups:
 control
 Ca (1400–1500 mg/d)
 Ca plus D3 (1100 IU/d)
 achieved 25(OH)D: 96 nmol/L ± 21.4

CU ORC
*Lappe et al. AJCN 2007
71
VITAMIN D & CANCER*
1.00
Fraction Cancer-Free

0.98
Ca+D

0.96
Ca-only

0.94
RR
P<= 0.01
0.402

0.92
Placebo

0.90
0 1 2 3 4

Time (yrs) *Lappe et al. AJCN 2007

VITAMIN D & CANCER*
1.00
Fraction Cancer-Free

0.98 Ca+D

0.96
Ca-only RR = 0.232

0.94
Placebo

0.92

0.90
0 1 2 3 4 5

Time (yrs) *Lappe et al. AJCN 2007

VITAMIN D & CANCER*
1.00

96 nmol/L
Fraction Cancer-Free

0.98 Ca+D

0.96
Ca-only

0.94 71.8 nmol/L

Placebo

0.92

0.90
0 1 2 3 4 5

Time (yrs) *Lappe et al. AJCN 2007

CANCERS BY TREATMENT (YRS 2–4)
Placebo Ca+D
Site
(n=266) (n = 403)
Breast 7 (2.6%) 4 (1.0%)
Colon 2 (0.7%) 0 (0.0%)
Lung 3 (1.1%) 1 (0.2%)
Marrow/Lymphoma 4 (1.5%) 2(0.5%)
Other 2 (0.7%) 1 (0.2%)
Total 18 (6.8%) 8 (2.0%)*
* P < 0.05
Safety
 VITAMIN D INTAKE & TOXICITY*

1,800
Serum 25(OH)D (nmol/L)

1,600 below
no toxicity 15 studies of adults
30,000
1,400 IU/d receiving vitamin D
1,200 supplementation
(means)
1,000
800 8 studies reporting
toxicity (individual
600 values)
400
200 no toxicity below 500
nmol/L (200 ng/mL)
0
1,000 10,000 100,000 1,000,000 10,000,000

Vitamin D Intake (IU/day)

* Hathcock JN et al. Am J Clin Nutr. 2007;85:6–18.

TUIL: 10,000 IU/d*

*Hathcock et al.,(2007) AJCN 85:6–18

TWO KEY QUESTIONS
assuming a target value of 80 nmol/L:
 how much of an increase in daily inputs
would be required to ensure that no
more than 2.5% of the population fell
below the target value?
 what , if anything, is the risk of raising
their 25(OH)D in those who already are
at or above the target value?

CU ORC

79
25(OH)D IN OLDER WOMEN*
*Lappe et al., JACN 2006
 1168 women 100

aged 55 &
older 80

 latitude 41º N
 25(OH)D Frequency 60

values
adjusted for 40
season
 median vit D 20
supplement
dose = 200 IU 0
0 40 80 120 160

25(OH)D (nmol/L)
CU ORC

80
 SHIFTING THE DISTRIBUTION
0.025

 improving

RELATIVE FREQUENCY
vitamin D 0.020

status at a
population 0.015
level means
raising
0.010
everybody’s
value, i.e.,
moving the 0.005
distribution
to the right 0.000
0 20 40 60 80 100 120 140 160 180

25(OH)D (nmol/L)
CU ORC

81
SHIFTING THE DISTRIBUTION
 using an effect 0.025
size of
1 nmol/L/mg/d

RELATIVE FREQUENCY
0.020
 it would require
~2000 IU of 0.015
additional D each
day to shift the
distribution 0.010

sufficiently to
ensure that no 0.005
more than 2.5 %
fell below 80
0.000
nmol/L 0 50 100 150 200

25(OH)D (nmol/L)
CU ORC

82
SHIFTING THE DISTRIBUTION
 taking an effect 0.025
size of
1 nmol/L/mg/d

RELATIVE FREQUENCY
0.020
 it would require
~2000 IU of 0.015
additional D each
day to shift the
distribution 0.010

sufficiently to
ensure that no 0.005
more than 2.5 %
fell below 80
0.000
nmol/L 0 50 100 150 200

25(OH)D (nmol/L)
CU ORC

83
SHIFTING THE DISTRIBUTION
0.025
 what about those
already 2 SD above

RELATIVE FREQUENCY
the mean? 0.020

 the rise with an

0.015
extra ~2000 IU/d
would be
predicted to bring 0.010

them to no more
than 170–180 0.005
nmol/L – well
below the toxic 0.000
range 0 50 100 150 200

25(OH)D (nmol/L)
CU ORC

84
CONCLUSIONS
 vitamin D acts in multiple systems
 serum 25(OH)D levels below 80 nmol/L
are not adequate for any body system
 levels of as high as 120 nmol/L may be
closer to optimal
 inputs from all sources combined
(needed to sustain 80 nmol/L) are in
the range of ~4,000 IU/d and higher

CU ORC

86
WHAT IS THE OPERATIVE MODEL?

 for the media?

 for regulators?
 for nutritional policy makers?
 for nutritional physiologists?

CU ORC

87
WHAT IS THE OPERATIVE MODEL?

 for the media and for regulators

 nutrition is about killing yourself

with a fork
 it’s about avoiding risks
 it’s about warnings & cautions

CU ORC

88
For a package of
macaroni & cheese

http://vm.cfsan.fda.gov/~dms/foodlab.html
Limit these
nutrients

Get enough of
these
nutrients
MEDIA REPORTING
 the overwhelming majority of media
reports about nutrition emphasizes harm
and risk
 while the explanation is partly that harm
is more newsworthy than benefit (and the
media battens on controversy)
 still the impression unwittingly conveyed
to the general public is one of concern
and danger
CU ORC

91
WHAT IS THE OPERATIVE MODEL?
 for nutritional policy makers

 nutrition is about determining

the least one can get by on
without suffering overt disease
 (once called MDRs)

CU ORC

92
WHAT IS THE OPERATIVE MODEL?
 for nutritional physiologists

 adult nutrition is about preventive

maintenance of tissues and organs
 it’s about keeping them from wearing
out or breaking down prematurely
 its referent is the intake that prevailed
when human physiology evolved

CU ORC

93
THE PREVENTIVE MAINTENANCE MODEL
 foundational premises:
 all tissues need all nutrients
 shortages impair the functioning of all
body systems
 premature organ/system “wearing out”, as
a consequence of nutrient deficiency, will
vary from person to person, depending on
variable genetic composition; and
 therefore, expression of nutrient deficiency
will usually be pluriform – both between
and within individuals
CU ORC

94
THE PREVENTIVE MAINTENANCE MODEL
 also recognizes that:
 the organism will work perfectly well
without maintenance – for a while . . .
 it thus reconciles the seeming paradox that
an organism can be “deficient” without
being clinically “sick”
– for a while . . .
 it’s also about squaring the morbidity/
mortality curve

CU ORC

95
THEORETICAL MORTALITY CURVE

0 20 40 60 80 100

AGE (yrs)
CU ORC

96
THEORETICAL MORTALITY CURVE
100

80
SURVIVAL (%)

60

40

20

0
0 10 20 30 40 50 60 70 80 90 100

AGE (yrs)
CU ORC

97
SQUARING THE MORTALITY CURVE
100

80
Percent alive/well

Optimal nutrition has the

potential
60 to contribute to

this improvement
40

Certainly, NCEP and DGA

20 take this for granted

0
0 10 20 30 40 50 60 70 80 90 100

Age (yrs)
CU ORC

98
WHAT WOULD IT BE LIKE?
 fewer cancers
 less diabetes
 fewer osteoporotic fractures
 less hypertension & CV disease
 less periodontal disease
 less multiple sclerosis
 less severe infectious disease

CU ORC

99
We don’t really know the true
burden of chronic disease.
And we won’t, until everyone
has enough vitamin D.
Thank you . . .