Vitamin D Status and Survival of Metastatic Colorectal Cancer
Patients: Results From CALGB/SWOG 80405 (Alliance)
Ng K, Venook AP, Sato K, Hollis BW, Niedzwiecki D, Ye C,
Chang I-W, O’Neil BH, Innocenti F, Lenz H-J, Blanke CD, Mayer RJ,
Fuchs CS, Meyerhardt JA
Abstract 507
Background: Vitamin D and Colorectal Cancer
• Vitamin D inhibits cell proliferation and angiogenesis, induces
cell differentiation and apoptosis, and has anti-inflammatory
effects
• Vitamin D receptor (VDR) and 1-α-hydroxylase are expressed
in colorectal cancer (CRC) cells
– Anti-proliferative effects greatest in cell lines with high
VDR1
• Treatment of APCmin mice with vitamin D decreases tumor
burden,2 whereas adenoma numbers and size are increased in
VDR-null APCmin mice3
• Low plasma 25(OH)D levels associated with risk of CRC
1. Evans SR, et al. Clin Cancer Res. 1998;4(11):2869-2876. 2. Huerta S, et al. Cancer Res. 2002;62(3):741-746. 3. Zheng W, et al. Int J Cancer.
2011;130(1):10-19. 4. Ma Y, et al. J Clin Oncol. 2011;29(28):3775-3782.
.
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
Prospective Cohort Study of 304 CRC Patients Suggests Association Between Prediagnosis 25(OH)D and Survival
Ng, K et al. J Clin Oncol. 2008;26(18):2984-3991.
Adjusted for age, gender, stage, grade, site, year of diagnosis, season of blood draw, BMI, and post-diagnosis physical activity
Survival benefit
may be greater in
stage III and IV
patients
1.0
0.81 [0.49 – 1.35]
0.81 [0.48 – 1.37]
0.52 [0.29-0.94]
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
<22.8 22.8-27.1 27.2-33.1 >33.1
Hazard
Rati
o f
or
Death
P trend = .02
Plasma 25(OH)D (ng/mL)
People with highest
level of vitamin D have
48% improvement
in outcome
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
Study Objective
Are higher vitamin D levels associated
with improved survival in patients with
metastatic CRC?
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
CALGB/SWOG 80405: Final Design
n = 1140
Primary endpoint: Overall Survival
Chemo + cetuximab
Chemo + bevacizumab
mCRC
first line
KRAS wild type (WT)
(codons 12,13)
Strata: • FOLFOX/FOLFIRI
• Prior adjuvant
chemo
• Prior radiotherapy
FOLFIRI or
FOLFOX
MD choice
Chemo + bevacizumab and
cetuximab
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
CALGB/SWOG 80405: Final Design
n = 1140
Primary endpoint: Overall Survival
Chemo + cetuximab
Chemo + bevacizumab
mCRC
first line
KRAS wild type
(codons 12,13)
Strata: • FOLFOX/FOLFIRI
• Prior adjuvant
chemo
• Prior radiotherapy
FOLFIRI or
FOLFOX
MD choice
Chemo + bevacizumab and
cetuximab
2334
Original
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
Study Cohort
Randomized
(n = 2334)
Bevacizumab
(n = 899)
Cetuximab
(n = 902)
Both
(n = 533)
RAS WT
(n = 256)
RAS mutant (mut)
(n = 167)
Unknown
(n = 476)
Plasma 25(OH)D Available (n = 1043)
n = 172 n = 126 n = 123
RAS WT
(n = 270)
RAS mutant
(n = 180)
Unknown
(n = 452)
n = 173 n = 121 n = 124
RAS WT
(n = 0)
RAS
mutant
(n = 124)
Unknown
(n = 409)
n = 0 n = 62 n = 142
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
Statistical Methods
• Preplanned, prospective, observational cohort study
• Primary endpoint: Overall survival
– Kaplan-Meier method
– Log rank test
• Plasma 25(OH)D measured by radioimmunoassay prior to
treatment
• Validated diet and lifestyle questionnaires prior to treatment
• Multivariable analyses using Cox proportional hazards models
• All P values two-sided and considered significant at the .05 level
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
Vitamin D Cohort vs Final Trial Cohort
Vitamin D
(n = 1043)
Final Trial
(n = 1137)
Median age, years 60 59
Male, % 58 61
ECOG PS 0 / 1, % 61 / 39 58 / 42
Primary tumor in place, % 25 28
Palliative intent, % 82 84
FOLFOX / FOLFIRI, % 77 / 23 73 / 27
All P values >.05
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
Baseline Characteristics (1)
Q1
(n = 208)
Q2
(n = 209)
Q3
(n = 208)
Q4
(n = 210)
Q5
(n = 208) P
Median
25(OH)D,
ng/mL (range)
8.0
(2.2-10.8)
13.6
(10.9-15.4)
17.2
(15.4-19.2)
21.4
(19.3-24.0)
27.5
(24.1-72.7) --
Median age,
years 59 60 60 61 61 .07
Male, % 48 64 58 64 55 .004
Black, % 25 12 6 7 2 <.0001
ECOG 0 / 1, % 49 / 50 64 / 36 58 / 42 63 / 37 70 / 30 .002
RAS WT / mut /
unknown, %
33 / 30 /
37
31 / 30 /
39
26 / 39 /
35
38 / 29 /
33
37 / 21 /
42 .02
Median 25(OH)D = 17.2 ng/mL
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
Baseline Characteristics (2)
• No significant difference in chemotherapy backbone, history of
prior adjuvant therapy, or assigned biologic between quintiles of
25(OH)D
• Significantly lower 25(OH)D seen in:
– Patients living in the north and northeast (P<.0001)
– Patients with blood drawn in winter and spring (P = .03)
– Obese patients (P = .0006)
– Less physically-active patients (P = .004)
– Patients not reporting vitamin D supplement use (P<.0001)
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
Higher Vitamin D Levels Associated With Better Survival
417 328 227 117 56 27 5 1
418 332 237 125 64 34 11 2
208 171 137 76 41 22 1
Quintiles 1 & 2
Quintiles 3 & 4
Quintile 5
No. at Risk
0
0.0
0.2
0.4
0.6
0.8
1.0
0 1 2 3 4 5 6 7
Time (years)
Ove
rall
Su
rviv
al P
roba
bili
ty
Log-rank P = 0.01
Quintile
5
3
2
mOS (months)
30.0
28.4
32.6
1 24.5
95% CI
21.7-28.6
4 27.2
25.8-32.2
24.2-31.0
25.0-31.5
27.7-36.9
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
Time (Years)
Higher Vitamin D Levels Also Associated with Better Progression-Free Survival (PFS)
417 163 55 19 10 4 3 1
418 196 72 34 19 13 3
208 96 37 17 6 4
Quintiles 1 & 2
Quintiles 3 & 4
Quintile 5
No. at Risk
0
0
0
4
0.0
0.2
0.4
0.6
0.8
1.0
0 1 2 3 5 6 7
Time (Years)
Pro
gre
ssio
n-F
ree S
urv
ival P
robabili
ty
Log-rank P = .02
Quintile
5
3
2
mPFS (months)
10.9
11.4
12.2
1 10.1
95% CI
9.2-11.3
4 12.7
9.6-11.6
9.7-12.9
11.1-13.6
10.8-14.2
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
Multivariate Analysis • Final model adjusted for:
– Age
– Sex
– Race
– ECOG performance status
– Chemotherapy backbone
– Previous adjuvant therapy
– Assigned biologic
– RAS mutation status
– Season of blood draw
– Geographic region of residence
– Body-mass index
– Physical activity
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
Multivariate Hazard Ratios: Overall Survival
1.0
0.83
[0.66-1.03]
0.81
[0.65-1.02]
0.65
[0.51-0.83]
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
2.2-10.8 10.9-15.4 15.5-19.2 >24.1
Haza
rd R
ati
o f
or
Dea
th
P trend = .001
Plasma 25(OH)D (ng/mL)
Pa ti e n ts w i th th e h i g h e s t
l e v e l s o f v i ta m i n D h a v e
a 3 5 % i m p r o v e m e n t
in overall survival
19.3-24.0
0.79
[0.63-1.00]
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
Multivariate Hazard Ratios: PFS
1.0 0.99
[0.80-1.23]
0.84
[0.67-1.05] 0.79
[0.63-0.99]
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
2.2-10.8 10.9-15.4 15.5-19.2 >24.1
Ha
za
rd R
ati
o f
or
Pro
gre
ss
ion
or
De
ath
P trend = .01
Plasma 25(OH)D (ng/mL)
Pa ti e n ts w i th th e h i g h e s t
l e v e l s o f v i ta m i n D h a v e
a 2 1 % i m p r o v e m e n t
i n PF S
19.3-24.0
0.83
[0.67-1.04]
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
Subgroup Analyses of
Overall Survival, Comparing
Extreme Quintiles of 25(OH)D
Adjusted HR:
Age (years)
<60
>60
Race
White
Black
Sex
Male
Female
ECOG performance status
0
1
Planned chemotherapy
FOLFIRI
FOLFOX
Assigned treatment arm
Bevacizumab
Cetuximab
Both
RAS mutation status
Wild-type
Mutant
Body mass index (kg/m2)
<25
≥25
Physical activity (MET-hours/wk)
<3
≥3
P interaction
0.27
0.10
0.09
0.85
0.32
0.90
0.20
0.78
0.50
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
Favors Higher 25(OH)D
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
Randomized Double-Blind Phase II Trial of Vitamin D in Metastatic CRC
Participating sites:
DFCI
MGH
BIDMC
DF/HCC satellites
DF/HCC affiliates
Northwestern
Vanderbilt
MSTI (Boise, ID)
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
n = 120
Conclusions
• Metastatic CRC patients are frequently vitamin D
deficient
• Higher vitamin D levels are associated with significantly
improved overall survival and PFS
• This association persists across all patient subgroups
and after adjusting for multiple prognostic factors
• A phase II randomized trial to evaluate the impact of
vitamin D supplementation as an adjunct to
chemotherapy is currently ongoing
Ng K, et al. J Clin Oncol. 2015;33(suppl 3): Abstract 507.
Top Related