Published Ahead of Print on June 23, 2014 as 10.1200/JCO.2014.55.3024 The latest version is at http://jco.ascopubs.org/cgi/doi/10.1200/JCO.2014.55.3024

JOURNAL OF CLINICAL ONCOLOGY

O R I G I N A L

R E P O R T

Calcium, Vitamin D, Dairy Products, and Mortality Among Colorectal Cancer Survivors: The Cancer Prevention Study-II Nutrition Cohort Baiyu Yang, Marjorie L. McCullough, Susan M. Gapstur, Eric J. Jacobs, Roberd M. Bostick, Veronika Fedirko, W. Dana Flanders, and Peter T. Campbell See accompanying editorial doi: 10.1200/JCO.2014.56.6299 Baiyu Yang, Marjorie L. McCullough, Susan M. Gapstur, Eric J. Jacobs, and Peter T. Campbell, American Cancer Society; Roberd M. Bostick, Veronika Fedirko, and W. Dana Flanders, Winship Cancer Institute, Emory University; Baiyu Yang, Emory University, Atlanta, GA. Published online ahead of print at www.jco.org on June 23, 2014. Supported by the American Cancer Society. Presented in part at the 2013 American Institute for Cancer Research Annual Meeting, Bethesda, MD, November 7-8, 2013. Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article. Corresponding author: Peter T. Campbell, PhD, Epidemiology Research Program, American Cancer Society National Home Office, 250 Williams St NW, Atlanta, GA 30303; e-mail: [email protected] © 2014 by American Society of Clinical Oncology 0732-183X/14/3299-1/$20.00 DOI: 10.1200/JCO.2014.55.3024

A

B

S

T

R

A

C

T

Purpose Higher calcium, vitamin D, and dairy product intakes are associated with lower colorectal cancer incidence, but their impacts on colorectal cancer survival are unclear. We evaluated associations of calcium, vitamin D, and dairy product intakes before and after colorectal cancer diagnosis with all-cause and colorectal cancer-specific mortality among colorectal cancer patients. Patients and Methods This analysis included 2,284 participants in a prospective cohort who were diagnosed with invasive, nonmetastatic colorectal cancer after baseline (1992 or 1993) and up to 2009. Mortality follow-up was through 2010. Prediagnosis risk factor information was collected on the baseline questionnaire. Postdiagnosis information was collected via questionnaires in 1999 and 2003 and was available for 1,111 patients. Results A total of 949 participants with colorectal cancer died during follow-up, including 408 from colorectal cancer. In multivariable-adjusted Cox proportional hazards regression models, postdiagnosis total calcium intake was inversely associated with all-cause mortality (relative risk [RR] for those in the highest relative to the lowest quartiles, 0.72; 95% CI, 0.53-0.98; Ptrend ⫽ .02) and associated with marginally statistically significant reduced colorectal cancer-specific mortality (RR, 0.59; 95% CI, 0.33 to 1.05; Ptrend ⫽ .01). An inverse association with all-cause mortality was also observed for postdiagnosis milk intake (RR, 0.72; 95% CI, 0.55 to 0.94; Ptrend ⫽ .02), but not vitamin D intake. Prediagnosis calcium, vitamin D, and dairy product intakes were not associated with any mortality outcomes. Conclusion Higher postdiagnosis intakes of total calcium and milk may be associated with lower risk of death among patients with nonmetastatic colorectal cancer. J Clin Oncol 32. © 2014 by American Society of Clinical Oncology

INTRODUCTION

The overall 5-year relative survival for colorectal cancer is 64% in the United States but decreases to 12% for distant metastatic disease.1 The associations of dietary factors with colorectal cancer incidence have been extensively reported,2 but their roles for colorectal cancer survival are largely unknown.3 Current dietary guidelines for cancer survivors are primarily based on incidence studies.4 Empirical knowledge of modifiable prognostic factors, including diet, for patients with colorectal cancer is needed for the more than 3.5 million colorectal cancer survivors worldwide.5 Higher intakes of calcium, vitamin D, and dairy products are generally associated with lower risk of

colorectal cancer incidence in observational studies.6,7 In addition, a major randomized, clinical trial of 1,200 mg of supplemental calcium versus placebo among 930 colorectal adenoma patients reported a 19% reduced risk of adenoma recurrence.8 In contrast, the Women’s Health Initiative clinical trial reported no effect of calcium plus vitamin D supplementation on colorectal cancer incidence,9-11 but suggestive benefits were observed among those not taking personal calcium or vitamin D supplements12 and those not concurrently randomly assigned to estrogen therapies.13 Two studies reported null associations of prediagnostic calcium intake with colorectal cancer survival.14,15 The main circulating biomarker of vitamin D, 25(OH)D, was associated © 2014 by American Society of Clinical Oncology

Information downloaded from jco.ascopubs.org and provided by at UCSF LIBRARY & CKM on November 19, 2014 from Copyright © 2014 American Society of Clinical Oncology. All rights reserved. 169.230.243.252

Copyright 2014 by American Society of Clinical Oncology

1

Yang et al

with lower risk of mortality among patients with colorectal cancer.16-19 To our knowledge, no study has examined whether total dairy or milk are associated with survival among patients with colorectal cancer. We investigated associations of pre- and postdiagnosis calcium (total, dietary, and supplemental), vitamin D (total and dietary), and dairy product (total and milk only) intakes with all-cause and colorectal cancer–specific mortality in a prospective study of men and women diagnosed with invasive, nonmetastatic colorectal cancer. PATIENTS AND METHODS Study Cohort Men and women in this study were selected from among the 184,000 participants in the Cancer Prevention Study II (CPS-II) Nutrition Cohort, a prospective study of cancer incidence that began in 1992.20 A 10-page, selfadministered questionnaire was used to collect information at baseline regarding demographics, medical history, physical activity, body size, cancer screening and early detection, diet, and other factors. Follow-up questionnaires were sent to participants biennially, beginning in 1997, to update exposure information and to learn of new cancer diagnoses. The CPS-II Nutrition Cohort is approved by the Institutional Review Board of Emory University. By the end of incidence follow-up on June 30, 2009, 3,832 of the 181,293 participants who had no personal history of the disease at baseline had been diagnosed with invasive colon or rectal cancer. Of these 3,832 patients with colorectal cancer, 2,188 were first self-reported on a follow-up questionnaire and then verified by review of medical records, and 865 patients had their diagnoses confirmed after self-report via linkage with state cancer registries. An additional 779 patients were initially identified as cancer deaths through linkage to the National Death Index (NDI)21; among those 779 patients, 531 colorectal cancer diagnoses were confirmed, either through linkage with state cancer registries (n ⫽ 529) or by examination of medical records (n ⫽ 2). Among the 3,832 patients with colorectal cancer, the following exclusions were applied: deaths determined through NDI that were not verified through medical records or cancer registries (n ⫽ 248), prevalent cancers (except for nonmelanoma skin cancer) at baseline (n ⫽ 387), implausible diagnosis date (n ⫽ 11), missing or unknown stage at diagnosis (n ⫽ 136), TNM summary stage IV or distant SEER stage at diagnosis (n ⫽ 421), nonadenocarcinoma histology (n ⫽ 50), implausible death date (n ⫽ 2), and poorquality dietary data at baseline (n ⫽ 293). We decided, a priori, to exclude patients with distant metastatic disease, consistent with previous studies from this cohort,22-25 because the 5-year relative survival in this group is so poor that it is unlikely that diet would substantially affect long-term mortality. After exclusions, 2,284 participants (1,274 men and 1,010 women) were included in this analysis. Among them, 1,682 were diagnosed with colon cancer (International Classification of Diseases for Oncology [ICD-O]: C18.0, C18.2–C18.9) and 602 with rectal cancer (ICD-O: C19.9, C20.9). By SEER summary stage, 1,154 participants were diagnosed with localized disease (malignant tumors limited to the colon or rectum) while 1,130 participants had regional disease (tumors that spread to adjacent tissue or regional lymph nodes through the bowel wall). Study Outcomes All participants were followed through December 31, 2010 to ascertain their vital status and cause of death (if applicable) through linkage to the NDI. Cause of death was obtained for 99.3% of all known deaths in the cohort. The primary outcome in this study was all-cause mortality. The secondary outcome was mortality specifically resulting from colorectal cancer (International Classification of Diseases Ninth Revision [ICD-9]: 153, 154; International Classification of Diseases Tenth Revision [ICD-10]: C18, C19, C20), defined from the singular underlying cause of death from NDI records. Other major causes of death in this cohort include cardiovascular diseases, neurodegenera2

© 2014 by American Society of Clinical Oncology

tive disease, other types of cancer (primarily lung and pancreas cancer), and respiratory system diseases. Pre- and Postdiagnosis Diet Prediagnosis diet was assessed at baseline (1992 or 1993) using a modified brief Block Food Frequency Questionnaire (FFQ).20,26,27 Postdiagnosis diet, where available, was assessed in 1999 and 2003, using a modified Willett FFQ.20,28-30 Both FFQs used similar questions on usual intake of dairy foods (major sources of dietary calcium and vitamin D, calculated by summing up total servings of milk, yogurt, ice cream, and cheese) and on calcium supplements and multivitamins (the major source of supplemental vitamin D during this time period; Appendix Table A1). For patients diagnosed after baseline and before the date of the 1999 survey completion, the 1999 survey was used for postdiagnosis diet. For patients diagnosed after 1999 and before the date of the 2003 survey completion, the 2003 survey was used for postdiagnosis diet. No postdiagnosis diet data are available from participants who did not return an eligible 1999 or 2003 postdiagnosis survey or from participants who were diagnosed after 2003. Of the 2,284 patients included in the prediagnosis analysis, 1,111 (48.6%) reported postdiagnosis diet. Statistical Analysis Sex- and questionnaire-specific quartiles were created for total calcium (ie, diet plus supplements), dietary calcium, total vitamin D (ie, diet plus supplements), dietary vitamin D, dairy, and milk. Questionnaire-specific categories were created for supplemental calcium (three levels) among men and women combined on the basis of visually inspecting the distribution and selecting interpretable cutoff points. We used multivariable Cox proportional hazards models to calculate relative risks (RRs) and 95% CIs. The underlying time axis for all Cox models was time since diagnosis. For prediagnosis models, person-time began on the date of diagnosis. For postdiagnosis models, we used delayed entry Cox models, wherein person-time started on the date they returned their postdiagnosis FFQ. In all analyses, person-time ended on the date of death or the end of follow-up (December 31, 2010), whichever came first. The proportional hazards assumption was evaluated for the main exposures with a likelihood ratio test by comparing models with and without an interaction term between an exposure and time; no violations were detected. All analyses were adjusted for age at diagnosis and tumor stage at diagnosis by stratifying within models. For prediagnosis models, we chose a priori to adjust for sex and baseline energy intake, and we in addition adjusted for baseline total folate intake because it changed the RR estimates by approximately 10%. Other demographic, lifestyle, and clinical covariates were evaluated, but none changed the RR estimates by more than 10%. Covariates in the basic postdiagnosis models also included sex and postdiagnosis energy intake, and in addition included postdiagnostic total folate in the multivariable model, to be consistent with the prediagnosis models. Baseline dietary intakes were evaluated as covariates in corresponding postdiagnosis models but did not materially change the RRs, so they were excluded. For each model the linear trend between exposure and mortality risk was assessed using the Wald test and modeling exposure as a continuous variable. In sensitivity analyses, we excluded participants with a history of diabetes, myocardial infarction, stroke at baseline, and death within 2 years of diagnosis. In addition, because treatment or serious illness may influence diet, we conducted sensitivity analyses excluding participants who completed FFQs within 1 year of diagnosis (1 year before diagnosis for prediagnostic models and 1 year after diagnosis for postdiagnosis models) and deaths within 2 years of the postdiagnosis questionnaire for postdiagnosis models. We tested for statistical interaction of each diet variable with age at diagnosis, sex, tumor stage, tumor sub-site, pre- or postdiagnosis body mass index, physical activity, total energy, and total folate intakes using likelihood ratio tests. All analyses were conducted using SAS version 9.3 (SAS Institute, Cary, NC).

RESULTS

Participants were, on average, age 64 years at baseline and 73 years at diagnosis. Fifty-six percent of participants were men, and most JOURNAL OF CLINICAL ONCOLOGY

Information downloaded from jco.ascopubs.org and provided by at UCSF LIBRARY & CKM on November 19, 2014 from Copyright © 2014 American Society of Clinical Oncology. All rights reserved. 169.230.243.252

Calcium, Vitamin D, Dairy, and Colorectal Cancer Survival

reported their race as white. There were no differences across quartiles of prediagnostic total calcium intake in the distributions of year of diagnosis, sex, tumor stage, grade, sub-site, treatment, or history of hypertension, myocardial infarction, diabetes, and stroke (Table 1). High calcium consumers were slightly older, better educated, more physically active, leaner, more likely to use nonsteroidal anti-inflammatory drugs and postmenopausal hormones (women only), less likely to smoke, and more likely to have a healthier overall diet. Among the 2,284 patients included in the prediagnosis analyses, 949 deaths occurred (408 from colon or rectal cancer) during a mean follow-up of 7.5 years (standard deviation, 4.6 years; range, 2 days to 18.1 years). No statistically significant associations were observed for any of the prediagnosis diet variables with any of the mortality outcomes (Table 2 and Appendix Table A2). The results were not meaningfully different after additional adjusting for other covariates or after additional sensitivity analyses (data not shown). The results were also null after we included patients with metastatic or unknown tumor stage (Appendix Table A3). In analyses restricted to the 1,111 participants who were included in the postdiagnosis analyses, prediagnosis use of supplemental calcium ⱖ 250 mg/d was statistically significantly associated with higher risk of all-cause mortality (RR, 1.65; 95% CI, 1.16 to 2.35; Appendix Table A4); this risk was primarily because of an increased RR for cardiovascular disease mortality (RR, 1.83; 95% CI, 0.82 to 4.09). Among the 1,111 patients included in the postdiagnosis analyses, 429 deaths occurred (143 from colon or rectal cancer) during a mean follow-up of 7.6 years (standard deviation, 3.4 years; range, 20 days to 11.3 years). The mean time between diagnosis and completion of the postdiagnosis questionnaire was 2.6 years. As shown in Table 3, comparing the highest to the lowest quartiles, total calcium (RR, 0.72; 95% CI, 0.53 to 0.98; Ptrend ⫽ .02) and milk (RR, 0.72; 95% CI, 0.55 to 0.94; Ptrend ⫽ .02) intakes were associated with lower all-cause mortality. Additional adjustment for prediagnostic total calcium and milk had no discernible effect on the results. A marginally statistically significant inverse association with all-cause mortality was observed for total dairy (RR, 0.75; 95% CI, 0.56 to 1.01; Ptrend ⫽ .05). Total calcium was also inversely associated with colorectal cancer–specific mortality (highest v lowest quartile RR, 0.59; 95% CI, 0.33 to 1.05; Ptrend ⫽ .01; Appendix Table A5). Because postdiagnosis diet and supplement use may be influenced by serious illness preceding death (reverse causation), we conducted a sensitivity analysis excluding deaths within the first 2 years of follow-up after completion of the postdiagnosis questionnaire. The results after this exclusion seemed similar to the original results. The RRs for the highest compared with the lowest quartile of total calcium and milk, respectively, were 0.69 (95% CI, 0.48 to 0.98; Ptrend ⫽ .03) and 0.68 (95% CI, 0.50 to 0.93; Ptrend ⫽ .02) for all-cause mortality; and 0.53 (95% CI, 0.24 to 1.19; Ptrend ⫽ .02) for total calcium and colorectal cancer–specific mortality. There was no evidence that the inverse associations of postdiagnosis total calcium and milk intakes with all-cause mortality were modified by age at diagnosis (⬍ 70 years v ⱖ 70 years), sex, tumor stage (localized v regional), tumor sub-site (colon v rectum), postdiagnosis body mass index (obese v not obese), physical activity (⬍ median v ⱖ median), total energy (⬍ median v ⱖ median), or total folate (⬍ median v ⱖ median) intakes (results stratified by stage shown in Appendix Table A6; other data not shown). www.jco.org

DISCUSSION

This study suggests that higher intakes of total calcium and milk after colorectal cancer diagnosis are associated with lower risk of mortality. These associations persisted after adjusting for important covariates, such as sex and tumor stage, and after several sensitivity analyses. We found no evidence that calcium, vitamin D, or dairy product intakes before colorectal cancer diagnosis were associated with mortality. To our knowledge, this is the first study to report associations of dairy and milk (both pre- and postdiagnosis) with colorectal cancer survival, and also the first to assess the role of postdiagnosis calcium and vitamin D intakes. Calcium, 25(OH)D (the major circulating form of vitamin D), and dairy products are associated with lower risk of incident colorectal cancer based on several meta-analyses.6,7,31 An earlier CPS-II Nutrition Cohort study reported inverse associations of colorectal cancer incidence with total calcium and total vitamin D intakes.32 The World Cancer Research Fund and American Institute for Cancer Research Continuous Update Project33 concluded in 2011 that calcium and milk were both probable factors associated with lower colorectal cancer risk. In contrast to the substantial evidence of a role for calcium, vitamin D, and dairy products in colorectal cancer primary prevention, the role of these factors in colorectal cancer survival is less studied.3 In two cohort studies, prediagnosis dietary calcium intake was not associated with all-cause mortality among patients with colorectal cancer, consistent with our findings.14,15 25(OH)D, either pre- or postdiagnosis, was associated with longer colorectal cancer survival in four previous studies16-19; in the current study, we observed no association with dietary vitamin D intake, which may not optimally reflect serum vitamin D status. In a large, pooled analysis, vitamin D intake was positively associated with serum 25(OH)D level, but the associations were relatively weak (Spearman correlation was 0.22 for dietary vitamin D and 0.29 for total vitamin D).34 In the current study, we found a statistically significant lower risk of death among patients with higher postdiagnostic intakes of total calcium and milk. Although not completely understood, several possible biologic mechanisms might underlie these associations. Clinical trials conducted among patients with previous colorectal adenoma suggested that daily treatment with calcium, compared with placebo, was associated with lower risk of colorectal adenoma recurrence.8,35,36 Potential mechanisms include calcium’s ability to bind to bile and fatty acids and prevent or lower toxicity37-39; direct effects on colonocyte proliferation,40,41 differentiation,42 and apoptosis43; and, alterations in K-ras mutations.44 Although these mechanisms were originally proposed in the primary prevention context, it is reasonable to hypothesize that calcium may also act through these mechanisms after diagnosis to reduce the risk of cancer recurrence, thus ultimately improving chance of survival. Direct clinical or epidemiologic evidence of calcium in colorectal cancer progression is limited, but in vitro evidence suggests that calcium may promote E-cadherin expression and suppress ␤-catenin/T-cell factor activation through the calcium sensing receptor, and restrain their malignant behaviors.45 Thus, calcium may be capable of limiting growth and distant metastasis from cancer cells that escaped the colon at the time of treatment. In our data, the strong inverse association of postdiagnosis total calcium © 2014 by American Society of Clinical Oncology

Information downloaded from jco.ascopubs.org and provided by at UCSF LIBRARY & CKM on November 19, 2014 from Copyright © 2014 American Society of Clinical Oncology. All rights reserved. 169.230.243.252

3

Yang et al

Table 1. Baseline Characteristics of Patients With Colorectal Cancer by Quartiles of Prediagnostic Total Calcium Intake in the CPS-II Nutrition Cohort Quartile of Total Calcium Intake, mg/dⴱ Q1 Characteristic No. Age at colorectal cancer diagnosis, years ⬍ 65 65 to ⬍ 70 70 to ⬍ 75 75 to ⬍ 80 ⱖ 80 Year of colorectal cancer diagnosis 1992-1996 1997-2000 2001-2004 2005-2009 Sex Male Female Race/ethnicity White/white-Hispanic Black/black-Hispanic Other/missing Education Less than high school High school degree Some college/trade school College graduate SEER summary stage Localized Regional Tumor grade at diagnosis Well differentiated Moderately differentiated Poorly differentiated Undifferentiated Colorectal cancer diagnosis site Colon Rectum First course of cancer treatment Surgery No Yes Chemotherapy No Yes Radiation No Yes Family history of colorectal cancer in 1982 No Yes History of diabetes No Yes History of stroke No Yes History of myocardial infarction No Yes

4

© 2014 by American Society of Clinical Oncology

No.

Q2 %

No.

570

Q3 %

No.

572

Q4 %

No.

570

%

P†

572 ⬍ .01

93 141 134 131 71

16.3 24.7 23.5 23.0 12.5

69 117 169 141 76

12.1 20.5 29.5 24.7 13.3

61 112 148 151 98

10.7 19.6 26.0 26.5 17.2

47 97 172 147 109

8.2 17.0 30.1 25.7 19.1

121 173 148 128

21.2 30.4 26.0 22.5

142 165 151 114

24.8 28.8 26.4 19.9

129 167 151 123

22.6 29.3 26.5 21.6

134 175 144 119

23.4 30.6 25.2 20.8

318 252

55.8 44.2

319 253

55.8 44.2

318 252

55.8 44.2

319 253

55.8 44.2

551 11 8

96.7 1.9 1.4

563 6 3

98.4 1.0 0.5

561 6 3

98.4 1.1 0.5

562 2 8

98.3 0.3 1.4

60 189 153 164

10.5 33.2 26.8 28.8

35 173 168 193

6.1 30.2 29.4 33.7

29 149 176 215

5.1 26.1 30.9 37.7

34 131 176 228

5.9 22.9 30.8 39.9

294 276

51.6 48.4

302 270

52.8 47.2

270 300

47.4 52.6

288 284

50.3 49.7

68 364 74 7

11.9 63.9 13.0 1.2

68 338 108 7

11.9 59.1 18.9 1.2

73 353 92 9

12.8 61.9 16.1 1.6

69 340 107 6

12.1 59.4 18.7 1.0

407 163

71.4 28.6

431 141

75.3 24.7

421 149

73.9 26.1

423 149

74.0 26.0

12 415

2.1 72.8

12 411

2.1 71.9

13 409

2.3 71.8

11 425

1.9 74.3

258 169

45.3 29.6

239 184

41.8 32.2

243 179

42.6 31.4

261 175

45.6 30.6

386 41

67.7 7.2

386 37

67.5 6.5

373 49

65.4 8.6

395 41

69.1 7.2

533 37

93.5 6.5

536 36

93.7 6.3

537 33

94.2 5.8

538 34

94.1 5.9

520 50

91.2 8.8

513 59

89.7 10.3

517 53

90.7 9.3

519 53

90.7 9.3

558 12

97.9 2.1

557 15

97.4 2.6

561 9

98.4 1.6

557 15

97.4 2.6

529 41

92.8 525 91.8 7.2 47 8.2 (continued on following page)

526 44

92.3 7.7

529 43

92.5 7.5

.95

1.00

.08

⬍ .01

.29

.43

.50

.97

.83

.79

.96

.84

.58

.93

JOURNAL OF CLINICAL ONCOLOGY

Information downloaded from jco.ascopubs.org and provided by at UCSF LIBRARY & CKM on November 19, 2014 from Copyright © 2014 American Society of Clinical Oncology. All rights reserved. 169.230.243.252

Calcium, Vitamin D, Dairy, and Colorectal Cancer Survival

Table 1. Baseline Characteristics of Patients With Colorectal Cancer by Quartiles of Prediagnostic Total Calcium Intake in the CPS-II Nutrition Cohort (continued) Quartile of Total Calcium Intake, mg/dⴱ Q1 Characteristic History of hypertension No Yes Physical activity, MET-h/wk Q1 Q2 Q3 Q4 BMI (kg/m2) ⬍ 18.5 18.5 to ⬍ 25 25 to ⬍ 30 ⱖ 30 Cigarette-smoking status Never Current Former NSAID use, No. of pills/mo 0 1 to ⬍ 15 15 to ⬍ 30 30 to ⬍ 60 ⱖ 60 HRT use among post-menopausal women None Current Former Dietary characteristics Alcohol intake, drinks/d Nondrinker ⬍1 ⱖ1

Energy intake, kcal/d Dietary folate intake, ␮g/d Total folate intake, ␮g/d Fruit/vegetable intake, servings/d Red/processed meat intake, servings/wk Whole grain intake, g/d

Q2

Q3

Q4

No.

%

No.

%

No.

%

No.

%

P†

342 228

60.0 40.0

345 227

60.3 39.7

366 204

64.2 35.8

336 236

58.7 41.3

90 204 144 123

15.8 35.8 25.3 21.6

61 192 173 138

10.7 33.6 30.2 24.1

60 180 172 148

10.5 31.6 30.2 26.0

49 180 172 166

8.6 31.5 30.1 29.0

6 187 259 112

1.1 32.8 45.4 19.6

5 190 258 108

0.9 33.2 45.1 18.9

3 218 237 103

0.5 38.2 41.6 18.1

9 253 218 84

1.6 44.2 38.1 14.7

228 74 268

40 13 47

210 43 315

36.7 7.5 55.1

222 45 302

38.9 7.9 53.0

230 26 309

40.2 4.5 54.0

279 96 32 96 46

48.9 16.8 5.6 16.8 8.1

258 92 47 99 54

45.1 16.1 8.2 17.3 9.4

241 79 67 112 50

42.3 13.9 11.8 19.6 8.8

246 61 69 127 51

43.0 10.7 12.1 22.2 8.9

119 50 65

49.0 20.6 26.8

124 64 52

50.4 26.0 21.1

110 78 50

44.5 31.6 20.2

94 78 58

38.5 32.0 23.8

207 178 173

36.3 31.2 30.4

229 212 124

40.0 37.1 21.7

227 234 98

39.8 41.1 17.2

255 209 99

44.6 36.5 17.3

.26

⬍ .01

⬍ .01

⬍ .01

⬍ .01

.03

⬍ .01

Mean

SD

Mean

SD

Mean

SD

Mean

SD

P

1,543.2 214.8 257.2 2.8 6.4 44.7

605.6 81.3 161.2 1.5 4.1 52.6

1,617.8 254.5 361.1 3.2 5.7 59.8

606.5 90.4 210.4 1.6 3.6 63.2

1,611.1 268.0 424.7 3.5 5.2 65.7

583.0 88.1 238.9 1.8 3.9 63.8

1,577.6 288.0 563.7 3.5 4.4 69.4

594.3 95.4 389.3 1.7 3.4 60.5

.01 ⬍ .01 ⬍ .01 ⬍ .01 ⬍ .01 ⬍ .01

NOTE. Some percentages do not add up to 100% because of missing data or rounding. Abbreviations: BMI, body mass index; CPS-II, Cancer Prevention Study-II; HRT, hormone replacement therapy; MET, metabolic equivalent; NSAID, nonsteroidal anti-inflammatory drug; Q, quartile; SD, standard deviation. ⴱ Quartiles in men:⬍ 578, 578 to⬍ 776, 776 to ⬍ 1,044, ⱖ 1,044; quartiles in women:⬍ 553, 553 to ⬍ 776, 776 to ⬍ 1,156, ⱖ 1,156. †P values derived from ␹2 test for differences in frequencies across total calcium strata for categorical predictors, and t test for continuous predictors with continuous total calcium intake.

intake with colorectal cancer–specific mortality was consistent with these mechanisms. Whereas postdiagnosis calcium intake was associated with lower risk of all-cause and colorectal cancer–specific mortality, there were no such associations with prediagnosis diet. Reasons for these discrepant findings are unclear. It is possible that calcium may have short-term rather than long-term effects on colorectal cancer progression and survival, and therefore only postdiagnosis diet is relevant in this context. It is also important to note that different FFQs were used to assess pre- and postdiagnosis diet. The www.jco.org

correlation coefficient (Pearson) between pre- and postdiagnosis total calcium intake was 0.37: this moderate correlation might suggest that participants changed their calcium intake after cancer diagnosis or, alternatively, this could reflect differences in the dietary assessment instruments. The FFQs used in this study included the major food and beverage sources of calcium and vitamin D, and validation studies have shown good agreements between estimates from diet recall and these FFQs (eg, Pearson correlation coefficients ranged from 0.57 to 0.66 for calcium and from 0.52 to 0.88 for dairy products).27,29,30 Therefore, we believe © 2014 by American Society of Clinical Oncology

Information downloaded from jco.ascopubs.org and provided by at UCSF LIBRARY & CKM on November 19, 2014 from Copyright © 2014 American Society of Clinical Oncology. All rights reserved. 169.230.243.252

5

Yang et al

Table 2. Associations of 1992 Prediagnostic Calcium, Vitamin D, and Dairy Intakes With All-Cause Mortality Among Patients With Nonmetastatic Colorectal Cancer in the CPS-II Nutrition Cohort

Men

Women

Total No. of Deaths

⬍ 578 578 to ⬍ 776 776 to ⬍ 1,044 ⱖ 1,044

⬍ 553 553 to ⬍ 776 776 to ⬍ 1,156 ⱖ 1,156

227 227 234 261

4,288 4,460 4,255 4,235

1.00 0.91 0.85 0.96

— 0.75 to 1.10 0.70 to 1.03 0.80 to 1.15 0.99

1.00 0.93 0.88 0.99

— 0.77 to 1.13 0.72 to 1.08 0.81 to 1.21 0.68

⬍ 548 548 to ⬍ 729 729 to ⬍ 949 ⱖ 949

⬍ 486 486 to ⬍ 629 629 to ⬍ 849 ⱖ 849

243 222 237 247

4,146 4,496 4,272 4,325

1.00 0.84 0.84 0.85

— 0.70 to 1.02 0.69 to 1.01 0.70 to 1.02 0.13

1.00 0.86 0.86 0.86

— 0.71 to 1.04 0.71 to 1.04 0.71 to 1.04 0.21

575 229 145

10,733 3,891 2,615

1.00 1.01 1.12

— 0.86, 1.19 0.92, 1.36 0.34

1.00 1.13 1.22

— 0.90, 1.42 0.96, 1.54 0.10

Rangeⴱ Exposure Total calcium, mg/d Q1§ Q2 Q3 Q4 Ptrend储 Dietary calcium, mg/d Q1§ Q2 Q3 Q4 Ptrend储 Supplemental calcium, mg/d C1§¶ C2 C3 Ptrend储 Total vitamin D, IU/d Q1§ Q2 Q3 Q4 Ptrend储 Dietary vitamin D, IU/d Q1§ Q2 Q3 Q4 Ptrend储 Total dairy, servings/wk Q1§ Q2 Q3 Q4 Ptrend储 Milk, servings/wk Q1§ Q2 Q3 Q4 Ptrend储

0 0.1 to ⬍ 250 ⱖ 250

Base Model†

PersonYears

RR

95% CI

MV Model‡ RR

95% CI

⬍ 122 122 to ⬍ 191 191 to ⬍ 425 ⱖ 425

⬍ 111 111 to ⬍ 201 201 to ⬍ 467 ⱖ 467

229 218 244 258

4,284 4,370 4,278 4,306

1.00 0.87 0.92 0.95

— 0.71 to 1.05 0.76 to 1.11 0.79 to 1.14 0.92

1.00 0.90 0.97 1.09

— 0.74 to 1.10 0.78 to 1.19 0.82 to 1.47 0.32

⬍ 105 105 to ⬍ 157 157 to ⬍ 226 ⱖ 226

⬍ 90 90 to ⬍ 136 136 to ⬍ 202 ⱖ 202

232 238 225 254

4,283 4,347 4,295 4,313

1.00 0.97 0.89 0.94

— 0.81 to 1.18 0.73 to 1.07 0.78 to 1.13 0.43

1.00 1.00 0.91 0.97

— 0.82 to 1.21 0.75 to 1.11 0.80 to 1.17 0.63

⬍ 5.5 5.5 to ⬍ 9.6 9.6 to ⬍ 14.5 ⱖ 14.5

⬍ 5.0 5.0 to ⬍ 8.9 8.9 to ⬍ 13.4 ⱖ 13.4

243 213 244 249

4,044 4,610 4,351 4,233

1.00 0.78 0.85 0.86

— 0.64 to 0.95 0.70 to 1.03 0.71 to 1.06 0.47

1.00 0.79 0.87 0.88

— 0.65 to 0.96 0.72 to 1.05 0.72 to 1.09 0.62

0 0.1 to ⬍ 5.7 5.7 to ⬍ 10.5 ⱖ 10.5

0 0.1 to ⬍ 5.1 5.1 to ⬍ 10.1 ⱖ 10.1

262 204 237 246

4,895 3,771 4,293 4,280

1.00 1.01 0.97 0.94

— 0.84 to 1.23 0.81 to 1.17 0.78 to 1.13 0.36

1.00 1.01 0.99 0.95

— 0.84 to 1.23 0.82 to 1.19 0.79 to 1.15 0.46

Abbreviations: C, category; CPS-II, Cancer Prevention Study-II; MV, multivariable; Q, quartile; RR, relative risk. ⴱ Range obtained from each sex-specific quartile of all exposures, except for supplemental calcium, which was obtained from each category for both sexes combined. †Base model adjusted for age at diagnosis, sex, tumor stage, and 1992 total energy intake. ‡Multivariable model adjusted for age at diagnosis, sex, tumor stage, and 1992 total energy and total folate intakes. §Reference group. 储Ptrend was calculated by using the median exposure in each quartile, specific to sex, for all exposures except for supplement calcium, which was calculated using the actual categories (ie, 1, 2, and 3) for both sexes. ¶Supplemental calcium was categorized based on visually inspecting the distribution of the variable.

that the low correlations between pre- and postdiagnosis diet are more likely resulting from real changes in diet after cancer diagnosis. We observed a potential higher risk of all-cause mortality from prediagnosis supplemental calcium intake, especially when restricting the analysis to the 1,111 participants who were included in the postdiagnosis analyses. Additional research should address whether this is a real potential harm to colorectal cancer patients. Milk may be associated with improved survival among patients with colorectal cancer because it is a rich source of dietary 6

© 2014 by American Society of Clinical Oncology

calcium and vitamin D. In addition, milk is a primary dietary source of conjugated linoleic acid, which was found to inhibit colorectal cancer cell growth in vitro.46,47 Other potentially beneficial components in dairy products include butyric acid, lactoferrin, and fermentation products.46 The strengths of our study include its large sample size, prospective design, and detailed pre- and postdiagnosis questionnaire information. We were also able to examine cause-specific mortality. Limitations include the lack of information on adverse effects from JOURNAL OF CLINICAL ONCOLOGY

Information downloaded from jco.ascopubs.org and provided by at UCSF LIBRARY & CKM on November 19, 2014 from Copyright © 2014 American Society of Clinical Oncology. All rights reserved. 169.230.243.252

Calcium, Vitamin D, Dairy, and Colorectal Cancer Survival

Table 3. Associations of Postdiagnosis Calcium, Vitamin D, and Dairy Intakes With All-Cause Mortality Among Patients With Nonmetastatic Colorectal Cancer in the CPS-II Nutrition Cohort 1999 Rangeⴱ Exposure Total calcium, mg/d Q1§ Q2 Q3 Q4 Ptrend储 Dietary calcium, mg/d Q1§ Q2 Q3 Q4 Ptrend储 Supplemental calcium, mg/d C1§¶ C2 C3 Ptrend储 Total vitamin D, IU/d Q1§ Q2 Q3 Q4 Ptrend储 Dietary vitamin D, IU/d Q1§ Q2 Q3 Q4 Ptrend储 Total dairy, servings/wk Q1§ Q2 Q3 Q4 Ptrend储 Milk, servings/wk Q1§ Q2 Q3 Q4 Ptrend储

Men

Women

2003 Rangeⴱ Men

Women

Total No. of PersonDeaths Years

Base Model† RR

95% CI

MV Model‡ RR

95% CI

⬍ 581 ⬍ 713 ⬍ 683 ⬍ 773 581 to ⬍ 775 713 to ⬍ 1,170 683 to ⬍ 882 773 to ⬍ 1,131 775 to ⬍ 1,105 1,170 to ⬍ 1,598 882 to ⬍ 1,162 1,131 to ⬍ 1,591 ⱖ 1,105 ⱖ 1,598 ⱖ 1,162 ⱖ 1,591

112 118 100 99

1,903 2,098 2,078 2,325

1.00 — 0.89 0.67 to 1.18 0.72 0.54 to 0.96 0.72 0.54 to 0.97 0.01

1.00 0.89 0.72 0.72

— 0.67, 1.18 0.53, 0.98 0.53, 0.98 0.02

⬍ 532 532 to ⬍ 683 683 to ⬍ 885 ⱖ 885

105 118 90 116

1,804 2,277 2,100 2,223

1.00 — 0.84 0.63 to 1.12 0.69 0.51 to 0.92 0.85 0.64 to 1.12 0.17

1.00 0.84 0.69 0.86

— 0.63, 1.11 0.51, 0.93 0.65, 1.14 0.21

221 108 100

3,966 2,001 2,437

1.00 0.90 0.94

1.00 — 0.95 0.72 to 1.27 0.98 0.73 to 1.31 0.88

⬍ 525 525 to ⬍ 671 671 to ⬍ 892 ⱖ 892

0 0.1 to ⬍ 500 ⱖ 500

⬍ 613 613 to ⬍ 765 765 to ⬍ 968 ⱖ 968

⬍ 609 609 to ⬍ 766 766 to ⬍ 990 ⱖ 990

0 0.1 to ⬍ 500 ⱖ 500

— 0.70, 1.16 0.72, 1.23 0.55

⬍ 164 164 to ⬍ 302 302 to ⬍ 559 ⱖ 559

⬍ 151 151 to ⬍ 379 379 to ⬍ 588 ⱖ 588

⬍ 194 194 to ⬍ 389 389 to ⬍ 603 ⱖ 603

⬍ 219 219 to ⬍ 509 509 to ⬍ 685 ⱖ 685

105 101 108 115

1,894 2,087 2,167 2,256

1.00 — 0.79 0.59 to 1.07 0.90 0.67 to 1.21 0.80 0.60 to 1.07 0.16

1.00 0.81 0.97 0.88

— 0.59, 1.10 0.67, 1.40 0.57, 1.35 0.35

⬍ 122 122 to ⬍ 178 178 to ⬍ 245 ⱖ 245

⬍ 100 100 to ⬍ 155 155 to ⬍ 229 ⱖ 229

⬍ 132 132 to ⬍ 188 188 to ⬍ 267 ⱖ 267

⬍ 103 103 to ⬍ 178 178 to ⬍ 257 ⱖ 257

103 115 105 106

2,065 2,144 2,109 2,085

1.00 — 0.99 0.75 to 1.31 0.94 0.70 to 1.25 0.89 0.67 to 1.19 0.29

1.00 0.99 0.95 0.90

— 0.75, 1.31 0.71, 1.27 0.67, 1.21 0.33

⬍ 4.7 4.7 to ⬍ 8.2 8.2 to ⬍ 11.7 ⱖ 11.7

⬍ 3.9 3.9 to ⬍ 7.7 7.7 to ⬍ 11.6 ⱖ 11.6

⬍ 5.1 5.1 to ⬍ 8.8 8.8 to ⬍ 12.3 ⱖ 12.3

⬍ 4.8 4.8 to ⬍ 8.1 8.1 to ⬍ 12.4 ⱖ 12.4

115 109 98 107

1,931 2,106 2,068 2,299

1.00 — 0.91 0.69 to 1.20 0.73 0.54 to 0.98 0.75 0.55 to 1.00 0.05

1.00 0.91 0.73 0.75

— 0.69, 1.21 0.54, 0.98 0.56, 1.01 0.05

⬍ 1.1 1.1 to ⬍ 5.6 5.6 to ⬍ 7.0 ⱖ 7.0

⬍ 1.0 1.0 to ⬍ 3.5 3.5 to ⬍ 7.0 ⱖ 7.0

⬍ 1.0 1.0 to ⬍ 5.6 5.6 to ⬍ 7.0 ⱖ 7.0

⬍ 1.0 1.0 to ⬍ 3.3 3.3 to ⬍ 7.0 ⱖ 7.0

106 109 41 173

1,844 2,135 881 3,543

1.00 — 0.84 0.64 to 1.12 0.76 0.52 to 1.11 0.71 0.55 to 0.93 0.01

1.00 0.85 0.76 0.72

— 0.64, 1.13 0.52, 1.12 0.55, 0.94 0.02

Abbreviations: C, category; CPS-II, Cancer Prevention Study-II; MV, multivariable; Q, quartile; RR, relative risk. ⴱ Range obtained from each questionnaire- and sex-specific quartile of all exposures, except for supplemental calcium, which was obtained from each category for both sexes combined. †Base model adjusted for age at diagnosis, sex, tumor stage, and postdiagnosis total energy intake. ‡Multivariable model adjusted for age at diagnosis, sex, tumor stage, and postdiagnosis total energy and total folate intakes. §Reference group. 储Ptrend calculated by using the median exposure in each quartile, specific to sex, for all exposures except for supplement calcium, which was calculated using the actual categories (ie, 1, 2, and 3) for both sexes. ¶Supplemental calcium was categorized based on visually inspecting the distribution of the variable.

treatment and tumor recurrence. FFQs may underestimate diet-disease associations compared with more objective biomarker measurements because of nondifferential misclassification. For large cohort studies, however, FFQs offer a feasible method to detect potential associations (especially when using energyadjusted nutrients) in the absence of biomarker measurements.48 As in most studies of this type, estimates of the effects of prediagnosis exposures are potentially biased because of selecting patients www.jco.org

who survived until the occurrence of colorectal cancer or the first postdiagnosis questionnaire.49-51 In conclusion, higher intakes of total calcium and milk after, but not before, colorectal cancer diagnosis may be associated with lower overall mortality. Our findings, if replicated in future observational studies and randomized trials, will provide important guidance for cancer survivors who are actively seeking diet and lifestyle changes to improve their prognosis. © 2014 by American Society of Clinical Oncology

Information downloaded from jco.ascopubs.org and provided by at UCSF LIBRARY & CKM on November 19, 2014 from Copyright © 2014 American Society of Clinical Oncology. All rights reserved. 169.230.243.252

7

Yang et al

AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST The authors indicated no potential conflicts of interest.

AUTHOR CONTRIBUTIONS Conception and design: Baiyu Yang, Marjorie L. McCullough, Peter T. Campbell

REFERENCES 1. American Cancer Society: Cancer Facts & Figures 2014. Atlanta: American Cancer Society, 2014 2. Chan AT, Giovannucci EL: Primary prevention of colorectal cancer. Gastroenterology 138:20292043, e10, 2010 3. Vrieling A, Kampman E: The role of body mass index, physical activity, and diet in colorectal cancer recurrence and survival: A review of the literature. Am J Clin Nutr 92:471-490, 2010 4. Rock CL, Doyle C, Demark-Wahnefried W, et al: Nutrition and physical activity guidelines for cancer survivors. CA Cancer J Clin 62:243-274, 2012 5. Bray F, Ren JS, Masuyer E, et al: Global estimates of cancer prevalence for 27 sites in the adult population in 2008. Int J Cancer 132:11331145, 2013 6. Cho E, Smith-Warner SA, Spiegelman D, et al: Dairy foods, calcium, and colorectal cancer: A pooled analysis of 10 cohort studies. J Natl Cancer Inst 96:1015-1022, 2004 7. Touvier M, Chan DS, Lau R, et al: Metaanalyses of vitamin D intake, 25-hydroxyvitamin D status, vitamin D receptor polymorphisms, and colorectal cancer risk. Cancer Epidemiol Biomarkers Prev 20:1003-1016, 2011 8. Baron JA, Beach M, Mandel JS, et al: Calcium supplements for the prevention of colorectal adenomas. Calcium Polyp Prevention Study Group. N Engl J Med 340:101-107, 1999 9. Wactawski-Wende J, Kotchen JM, Anderson GL, et al: Calcium plus vitamin D supplementation and the risk of colorectal cancer. N Engl J Med 354:684-696, 2006 10. Cauley JA, Chlebowski RT, WactawskiWende J, et al: Calcium plus vitamin D supplementation and health outcomes five years after active intervention ended: The Women’s Health Initiative. J Womens Health (Larchmt) 22:915-929, 2013 11. Prentice RL, Pettinger MB, Jackson RD, et al: Health risks and benefits from calcium and vitamin D supplementation: Women’s Health Initiative clinical trial and cohort study. Osteoporos Int 24:567580, 2013 12. Bolland MJ, Grey A, Gamble GD, et al: Calcium and vitamin D supplements and health outcomes: A reanalysis of the Women’s Health Initiative (WHI) limited-access data set. Am J Clin Nutr 94:1144-1149, 2011 13. Ding EL, Mehta S, Fawzi WW, et al: Interaction of estrogen therapy with calcium and vitamin D supplementation on colorectal cancer risk: Reanalysis of Women’s Health Initiative randomized trial. Int J Cancer 122:1690-1694, 2008 14. Slattery ML, French TK, Egger MJ, et al: Diet and survival of patients with colon cancer in Utah: Is there an association? Int J Epidemiol 18:792-797, 1989 8

© 2014 by American Society of Clinical Oncology

Financial support: Susan M. Gapstur, Peter T. Campbell Administrative support: Susan M. Gapstur, Peter T. Campbell Provision of study materials or patients: Marjorie L. McCullough, Susan M. Gapstur, Peter T. Campbell Collection and assembly of data: Marjorie L. McCullough, Susan M. Gapstur, Eric J. Jacobs, Peter T. Campbell Data analysis and interpretation: All authors Manuscript writing: All authors Final approval of manuscript: All authors

15. Zell JA, McEligot AJ, Ziogas A, et al: Differential effects of wine consumption on colorectal cancer outcomes based on family history of the disease. Nutr Cancer 59:36-45, 2007 16. Fedirko V, Riboli E, Tjønneland A, et al: Prediagnostic 25-hydroxyvitamin D, VDR and CASR polymorphisms, and survival in patients with colorectal cancer in western European ppulations. Cancer Epidemiol Biomarkers Prev 21:582-593, 2012 17. Ng K, Meyerhardt JA, Wu K, et al: Circulating 25-hydroxyvitamin d levels and survival in patients with colorectal cancer. J Clin Oncol 26:2984-2991, 2008 18. Ng K, Wolpin BM, Meyerhardt JA, et al: Prospective study of predictors of vitamin D status and survival in patients with colorectal cancer. Br J Cancer 101:916-923, 2009 19. Mezawa H, Sugiura T, Watanabe M, et al: Serum vitamin D levels and survival of patients with colorectal cancer: Post-hoc analysis of a prospective cohort study. BMC Cancer 10:347, 2010 20. Calle EE, Rodriguez C, Jacobs EJ, et al: The American Cancer Society Cancer Prevention Study II Nutrition Cohort: Rationale, study design, and baseline characteristics. Cancer 94:2490-2501, 2002 21. Calle EE, Terrell DD: Utility of the National Death Index for ascertainment of mortality among cancer prevention study II participants. Am J Epidemiol 137:235-241, 1993 22. McCullough ML, Gapstur SM, Shah R, et al: Association between red and processed meat intake and mortality among colorectal cancer survivors. J Clin Oncol 31:2773-2782, 2013 23. Dehal AN, Newton CC, Jacobs EJ, et al: Impact of diabetes mellitus and insulin use on survival after colorectal cancer diagnosis: The Cancer Prevention Study-II Nutrition Cohort. J Clin Oncol 30:53-59, 2012 24. Campbell PT, Patel AV, Newton CC, et al: Associations of recreational physical activity and leisure time spent sitting with colorectal cancer survival. J Clin Oncol 31:876-885, 2013 25. Campbell PT, Newton CC, Dehal AN, et al: Impact of body mass index on survival after colorectal cancer diagnosis: The Cancer Prevention Study-II Nutrition Cohort. J Clin Oncol 30:42-52, 2012 26. Block G, Hartman AM, Naughton D: A reduced dietary questionnaire: Development and validation. Epidemiology 1:58-64, 1990 27. Flagg EW, Coates RJ, Calle EE, et al: Validation of the American Cancer Society Cancer Prevention Study II Nutrition Survey Cohort Food Frequency Questionnaire. Epidemiology 11:462468, 2000 28. Willett WC, Sampson L, Stampfer MJ, et al: Reproducibility and validity of a semiquantitative food frequency questionnaire. Am J Epidemiol 122: 51-65, 1985 29. Rimm EB, Giovannucci EL, Stampfer MJ, et al: Reproducibility and validity of an expanded selfadministered semiquantitative food frequency ques-

tionnaire among male health professionals. Am J Epidemiol 135:1114-1126, 1992 30. Feskanich D, Rimm EB, Giovannucci EL, et al: Reproducibility and validity of food intake measurements from a semiquantitative food frequency questionnaire. J Am Diet Assoc 93:790796, 1993 31. Aune D, Lau R, Chan DS, et al: Dairy products and colorectal cancer risk: A systematic review and meta-analysis of cohort studies. Ann Oncol 23:3745, 2012 32. McCullough ML, Robertson AS, Rodriguez C, et al: Calcium, vitamin D, dairy products, and risk of colorectal cancer in the Cancer Prevention Study II Nutrition Cohort (United States). Cancer Causes Control 14:1-12, 2003 33. World Cancer Research Fund/American Institute for Cancer Research: Colorectal cancer 2011 report: Food, nutrition, physical activity, and the prevention of colorectal cancer. http://www.aicr.org/ continuous-update-project/reports/Colorectal-Cancer2011-Report.pdf 34. McCullough ML, Weinstein SJ, Freedman DM, et al: Correlates of circulating 25-hydroxyvitamin D: Cohort Consortium Vitamin D Pooling Project of Rarer Cancers. Am J Epidemiol 172:21-35, 2010 35. Bonithon-Kopp C, Kronborg O, Giacosa A, et al: Calcium and fibre supplementation in prevention of colorectal adenoma recurrence: A randomised intervention trial. European Cancer Prevention Organisation Study Group. Lancet 356:1300-1306, 2000 36. Hofstad B, Almendingen K, Vatn M, et al: Growth and recurrence of colorectal polyps: A double-blind 3-year intervention with calcium and antioxidants. Digestion 59:148-156, 1998 37. Newmark HL, Wargovich MJ, Bruce WR: Colon cancer and dietary fat, phosphate, and calcium: A hypothesis. J Natl Cancer Inst 72:1323-1325, 1984 38. Wargovich MJ, Eng VW, Newmark HL: Calcium inhibits the damaging and compensatory proliferative effects of fatty acids on mouse colon epithelium. Cancer Lett 23:253-258, 1984 39. Govers MJ, Termont DS, Lapre´ JA, et al: Calcium in milk products precipitates intestinal fatty acids and secondary bile acids and thus inhibits colonic cytotoxicity in humans. Cancer Res 56:32703275, 1996 40. Buset M, Lipkin M, Winawer S, et al: Inhibition of human colonic epithelial cell proliferation in vivo and in vitro by calcium. Cancer Res 46:54265430, 1986 41. Bostick RM, Fosdick L, Wood JR, et al: Calcium and colorectal epithelial cell proliferation in sporadic adenoma patients: A randomized, doubleblinded, placebo-controlled clinical trial. J Natl Cancer Inst 87:1307-1315, 1995 42. Fedirko V, Bostick RM, Flanders WD, et al: Effects of vitamin D and calcium on proliferation and differentiation in normal colon mucosa: A randomized clinical trial. Cancer Epidemiol Biomarkers Prev 18:2933-2941, 2009 JOURNAL OF CLINICAL ONCOLOGY

Information downloaded from jco.ascopubs.org and provided by at UCSF LIBRARY & CKM on November 19, 2014 from Copyright © 2014 American Society of Clinical Oncology. All rights reserved. 169.230.243.252

Calcium, Vitamin D, Dairy, and Colorectal Cancer Survival

43. Fedirko V, Bostick RM, Flanders WD, et al: Effects of vitamin D and calcium supplementation on markers of apoptosis in normal colon mucosa: A randomized, double-blind, placebo-controlled clinical trial. Cancer Prev Res (Phila) 2:213-223, 2009 44. Llor X, Jacoby RF, Teng BB, et al: K-ras mutations in 1,2-dimethylhydrazine-induced colonic tumors: Effects of supplemental dietary calcium and vitamin D deficiency. Cancer Res 51:4305-4309, 1991 45. Chakrabarty S, Radjendirane V, Appelman H, et al: Extracellular calcium and calcium sensing

receptor function in human colon carcinomas: Promotion of E-cadherin expression and suppression of beta-catenin/TCF activation. Cancer Res 63:67-71, 2003 46. Norat T, Riboli E: Dairy products and colorectal cancer. A review of possible mechanisms and epidemiological evidence. Eur J Clin Nutr 57:1-17, 2003 47. Shultz TD, Chew BP, Seaman WR, et al: Inhibitory effect of conjugated dienoic derivatives of linoleic acid and beta-carotene on the in vitro growth of human cancer cells. Cancer Lett 63:125-133, 1992

48. Kipnis V, Subar AF, Midthune D, et al: Structure of dietary measurement error: Results of the OPEN biomarker study. Am J Epidemiol 158:14-21; discussion 22-26, 2003 49. Flanders WD, Klein M: Properties of 2 counterfactual effect definitions of a point exposure. Epidemiology 18:453-460, 2007 50. Dahabreh IJ, Kent DM: Index event bias as an explanation for the paradoxes of recurrence risk research. JAMA 305:822-823, 2011 51. Flanders WD, Eldridge RC, McClellan W: A nearly unavoidable mechanism for collider bias with index-event studies. Epidemiology (in press)

■ ■ ■

www.jco.org

© 2014 by American Society of Clinical Oncology

Information downloaded from jco.ascopubs.org and provided by at UCSF LIBRARY & CKM on November 19, 2014 from Copyright © 2014 American Society of Clinical Oncology. All rights reserved. 169.230.243.252

9

Yang et al

Acknowledgment The American Cancer Society funds the creation, maintenance, and updating of the Cancer Prevention Study-II (CPS-II) cohort. We thank the CPS-II participants and Study Management Group for their invaluable contributions to this research. The authors also acknowledge the contribution to this study from central cancer registries supported through the Centers for Disease Control and Prevention National Program of Cancer Registries and cancer registries supported by the National Cancer Institute SEER program. Appendix

Table A1. Comparison of Questions on Each Food Frequency Questionnaire on Usual Intake of Dairy Foods, Calcium Supplements, and Multivitamins: Cancer Prevention Study II Nutrition Cohort 1992ⴱ

1999†

2003†

Whole milk and beverages with whole milk 2% milk and beverages with 2% milk Skim milk, 1%, or buttermilk Cheeses and cheese spreads (regular and low fat) Ice cream (regular and low fat)

Whole milk 2% milk Skim or 1% milk Cheese (cottage or ricotta, and other) Ice cream (regular and nonfat/sherbet) Yogurt (plain or artificially sweetened, frozen, and other) Pizza Calcium Regular use: yes/no Amount per day (ⱕ 900 mg 关calculated as 500 mg兴, ⱖ 901 mg 关calculated as 1,000 mg兴, unknown) Currently yes/no Frequency per week Brand (write-in)

Whole milk 2% milk Skim or 1% milk Cheese (cottage or ricotta, and other) Ice cream (regular and nonfat/sherbet) Yogurt (plain or artificially sweetened, frozen, and other) Pizza Calcium Regular use: yes/no Pills per week Amount in each pill (ⱕ 350 mg 关calculated as 250 mg兴, ⱖ 400 mg 关calculated as 500 mg兴, unknown) Currently yes/no Frequency per week Brand (write-in)

Food Milk

Other dairy products

Yogurt (regular and low fat, including frozen)

Calcium supplements

Multivitamins

Restaurant pizza Calcium or dolomite Frequency per week or per day Amount in each tablet (250 mg, 500 mg, 600 mg, or 750 mg)

Use at least once per week yes/no Type Stress-tabs type Therapeutic, Theragran type One-a-day type or Centrum No. of tablets per day or per week

ⴱ Dairy in 1992 was calculated as all types of milk (8-ounce glass serving) plus cheese and cheese spreads (2-ounce serving) plus ice cream (11⁄2-cup serving) plus yogurt (1-cup serving) plus cheese on pizza (11⁄2-ounce serving). †Dairy in 1999 and 2003 was calculated as all types of milk (8-ounce glass serving) plus ice cream (11⁄2-cup serving) plus yogurt (1-cup serving) plus cottage cheese (2-cup serving) plus processed cheese (2-ounce serving) plus hard cheese (11⁄2-ounce serving) plus cheese on pizza (11⁄2-ounce serving).

10

© 2014 by American Society of Clinical Oncology

JOURNAL OF CLINICAL ONCOLOGY

Information downloaded from jco.ascopubs.org and provided by at UCSF LIBRARY & CKM on November 19, 2014 from Copyright © 2014 American Society of Clinical Oncology. All rights reserved. 169.230.243.252

Calcium, Vitamin D, Dairy, and Colorectal Cancer Survival

Table A2. Associations of 1992 Prediagnostic Calcium, Vitamin D, and Dairy Intakes With Colorectal Cancer Mortality Among Patients With Nonmetastatic Colorectal Cancer in the CPS-II Nutrition Cohort Rangeⴱ Exposure Total calcium, mg/d Q1§ Q2 Q3 Q4 Ptrend储 Dietary calcium, mg/d Q1§ Q2 Q3 Q4 Ptrend储 Supplemental calcium, mg/d C1§¶ C2 C3 Ptrend储 Total vitamin D, IU/d Q1§ Q2 Q3 Q4 Ptrend储 Dietary vitamin D, IU/d Q1§ Q2 Q3 Q4 Ptrend储 Total dairy, servings/wk Q1§ Q2 Q3 Q4 Ptrend储 Milk, servings/wk Q1§ Q2 Q3 Q4 Ptrend储

Base Model†

Men

Women

Total No. of Deaths

⬍ 578 578 to ⬍ 776 776 to ⬍ 1,044 ⱖ 1,044

⬍ 553 553 to ⬍ 776 776 to ⬍ 1,156 ⱖ 1,156

101 103 102 102

4,288 4,460 4,255 4,235

1.00 0.99 0.88 0.95

⬍ 548 548 to ⬍ 729 729 to ⬍ 949 ⱖ 949

⬍ 486 486 to ⬍ 629 629 to ⬍ 849 ⱖ 849

106 99 107 96

4,146 4,496 4,272 4,325

250 95 63

0 0.1 to ⬍ 250 ⱖ 250

PersonYears

RR

95% CI

MV Model‡ RR

95% CI

— 0.75 to 1.32 0.66 to 1.17 0.72 to 1.26 0.78

1.00 1.04 0.94 1.01

— 0.77 to 1.39 0.69 to 1.26 0.74 to 1.38 0.91

1.00 0.93 0.95 0.82

— 0.70 to 1.24 0.72 to 1.25 0.62 to 1.10 0.18

1.00 0.96 0.99 0.86

— 0.72 to 1.28 0.74 to 1.31 0.64 to 1.16 0.30

10,733 3,891 2,615

1.00 0.98 1.07

— 0.77, 1.25 0.79, 1.43 0.76

1.00 1.11 1.18

— 0.79, 1.57 0.83, 1.66 0.36

⬍ 122 122 to ⬍ 191 191 to ⬍ 425 ⱖ 425

⬍ 111 111 to ⬍ 201 201 to ⬍ 467 ⱖ 467

96 108 99 105

4,284 4,370 4,278 4,306

1.00 1.06 0.90 0.98

— 0.80 to 1.40 0.67 to 1.21 0.73 to 1.30 0.77

1.00 1.12 1.00 1.14

— 0.83 to 1.49 0.72 to 1.38 0.73 to 1.78 0.61

⬍ 105 105 to ⬍ 157 157 to ⬍ 226 ⱖ 226

⬍ 90 90 to ⬍ 136 136 to ⬍ 202 ⱖ 202

103 109 93 103

4,283 4,347 4,295 4,313

1.00 1.01 0.87 0.91

— 0.76 to 1.33 0.65 to 1.16 0.69 to 1.21 0.39

1.00 1.03 0.90 0.96

— 0.78 to 1.36 0.67 to 1.22 0.72 to 1.28 0.61

⬍ 5.5 5.5 to ⬍ 9.6 9.6 to ⬍ 14.5 ⱖ 14.5

⬍ 5.0 5.0 to ⬍ 8.9 8.9 to ⬍ 13.4 ⱖ 13.4

110 91 107 100

4,044 4,610 4,351 4,233

1.00 0.83 0.89 0.86

— 0.62 to 1.10 0.67 to 1.18 0.63 to 1.17 0.55

1.00 0.84 0.92 0.89

— 0.63 to 1.13 0.69 to 1.23 0.65 to 1.22 0.73

0 0.1 to ⬍ 5.7 5.7 to ⬍ 10.5 ⱖ 10.5

0 0.1 to ⬍ 5.1 5.1 to ⬍ 10.1 ⱖ 10.1

110 88 114 96

4,895 3,771 4,293 4,280

1.00 1.07 1.07 0.95

— 0.80 to 1.43 0.81 to 1.40 0.71 to 1.28 0.62

1.00 1.06 1.08 0.98

— 0.79 to 1.42 0.82 to 1.42 0.73 to 1.32 0.80

Abbreviations: C, category; MV, multivariable; Q, quartile; RR, relative risk. ⴱ Range obtained from each sex-specific quartile of all exposures, except for supplemental calcium, which was obtained from each category for both sexes combined. †Base model adjusted for age at diagnosis, sex, tumor stage, and 1992 total energy intake. ‡Multivariable model adjusted for age at diagnosis, sex, tumor stage, and 1992 total energy and total folate intakes. §Reference group. 储Ptrend calculated by using the median exposure in each quartile, specific to sex, for all exposures except for supplement calcium, which was calculated using the actual categories (ie, 1, 2, and 3) for both sexes. ¶Supplemental calcium was categorized based on visually inspecting the distribution of the variable.

www.jco.org

© 2014 by American Society of Clinical Oncology

Information downloaded from jco.ascopubs.org and provided by at UCSF LIBRARY & CKM on November 19, 2014 from Copyright © 2014 American Society of Clinical Oncology. All rights reserved. 169.230.243.252

11

Yang et al

Table A3. Associations of 1992 Prediagnostic Calcium, Vitamin D, and Dairy Intakes With All-Cause Mortality Among Patients With Colorectal Cancer of All Stages in the CPS-II Nutrition Cohort Rangeⴱ Exposure Total calcium, mg/d Q1§ Q2 Q3 Q4 Ptrend储 Dietary calcium, mg/d Q1§ Q2 Q3 Q4 Ptrend储 Supplemental calcium, mg/d C1§¶ C2 C3 Ptrend储 Total vitamin D, IU/d Q1§ Q2 Q3 Q4 Ptrend储 Dietary vitamin D, IU/d Q1§ Q2 Q3 Q4 Ptrend储 Total dairy, servings/wk Q1§ Q2 Q3 Q4 Ptrend储 Milk, servings/wk Q1§ Q2 Q3 Q4 Ptrend储

Base Model†

Men

Women

⬍ 579 579 to ⬍ 775 775 to ⬍ 1,033 ⱖ 1,033

⬍ 545 545 to ⬍ 773 773 to ⬍ 1,152 ⱖ 1,152

337 335 344 366

4,562 4,874 4,563 4,628

1.00 0.93 0.89 0.96

— 0.79 to 1.09 0.76 to 1.05 0.82 to 1.12 0.85

1.00 0.96 0.94 1.00

— 0.81 to 1.13 0.79 to 1.11 0.84 to 1.19 0.75

⬍ 548 548 to ⬍ 724 724 to ⬍ 944 ⱖ 944

⬍ 481 481 to ⬍ 620 620 to ⬍ 848 ⱖ 848

350 337 342 353

4,427 4,859 4,640 4,701

1.00 0.86 0.92 0.87

— 0.73 to 1.01 0.78 to 1.07 0.75 to 1.03 0.22

1.00 0.88 0.95 0.91

— 0.75 to 1.04 0.81 to 1.12 0.77 to 1.07 0.45

851 321 210

11,618 4,219 2,790

1.00 1.01 1.06

— 0.88 to 1.16 0.90 to 1.26 0.52

1.00 1.11 1.14

— 0.91 to 1.35 0.94 to 1.40 0.19

0 0.1 to ⬍ 250 ⱖ 250

PersonYears

RR

MV Model‡

Total No. of Deaths

95% CI

RR

95% CI

⬍ 123 123 to ⬍ 191 191 to ⬍ 420 ⱖ 420

⬍ 108 108 to ⬍ 193 193 to ⬍ 461 ⱖ 461

343 324 350 365

4,534 4,728 4,660 4,704

1.00 0.87 0.94 0.95

— 0.74 to 1.03 0.80 to 1.11 0.81 to 1.11 0.99

1.00 0.92 1.02 1.08

— 0.78 to 1.09 0.85 to 1.22 0.84 to 1.40 0.33

⬍ 106 106 to ⬍ 158 158 to ⬍ 227 ⱖ 227

⬍ 90 90 to ⬍ 135 135 to ⬍ 203 ⱖ 203

345 338 331 368

4,562 4,758 4,737 4,571

1.00 0.88 0.88 0.97

— 0.75 to 1.03 0.75 to 1.03 0.83 to 1.14 0.98

1.00 0.90 0.92 1.02

— 0.77 to 1.06 0.78 to 1.09 0.86 to 1.20 0.52

⬍ 5.5 5.5 to ⬍ 9.6 9.6 to ⬍ 14.6 ⱖ 14.6

⬍ 4.9 4.9 to ⬍ 8.8 8.8 to ⬍ 13.3 ⱖ 13.3

350 321 352 359

4,394 4,886 4,761 4,585

1.00 0.81 0.89 0.92

— 0.68 to 0.95 0.76 to 1.05 0.77 to 1.09 0.88

1.00 0.82 0.92 0.95

— 0.70 to 0.97 0.78 to 1.09 0.80 to 1.14 0.79

0 0.1 to ⬍ 5.5 5.5 to ⬍ 10.4 ⱖ 10.4

0 0.1 to ⬍ 4.8 4.8 to ⬍ 10.0 ⱖ 10.0

440 286 344 368

5,365 3,847 4,859 4,677

1.00 0.97 0.92 0.95

— 0.82 to 1.14 0.78 to 1.08 0.81 to 1.12 0.48

1.00 0.96 0.93 0.98

— 0.82 to 1.14 0.80 to 1.09 0.83 to 1.15 0.72

Abbreviations: C, category; CPS-II, Cancer Prevention Study-II; MV, multivariable; Q, quartile; RR, relative risk. ⴱ Range obtained from each sex-specific quartile of all exposures, except for supplemental calcium, which was obtained from each category for both sexes combined. †Base model adjusted for age at diagnosis, sex, tumor stage, and 1992 total energy intake. ‡Multivariable model adjusted for age at diagnosis, sex, tumor stage, and 1992 total energy and total folate intakes. §Reference group. 储Ptrend calculated by using the median exposure in each quartile, specific to sex, for all exposures except for supplement calcium, which was calculated using the actual categories (ie, 1, 2, and 3) for both sexes. ¶Supplemental calcium was categorized based on visually inspecting the distribution of the variable.

12

© 2014 by American Society of Clinical Oncology

JOURNAL OF CLINICAL ONCOLOGY

Information downloaded from jco.ascopubs.org and provided by at UCSF LIBRARY & CKM on November 19, 2014 from Copyright © 2014 American Society of Clinical Oncology. All rights reserved. 169.230.243.252

Calcium, Vitamin D, Dairy, and Colorectal Cancer Survival

Table A4. Associations of 1992 Prediagnostic Calcium, Vitamin D, and Dairy Intakes With All-Cause Mortality Among Patients With Colorectal Cancer With Postdiagnosis Data (n ⫽ 1,111) in the CPS-II Nutrition Cohort Rangeⴱ Exposure Total calcium, mg/d Q1§ Q2 Q3 Q4 Ptrend储 Dietary calcium, mg/d Q1§ Q2 Q3 Q4 Ptrend储 Supplemental calcium, mg/d C1§¶ C2 C3 Ptrend储 Total vitamin D, IU/d Q1§ Q2 Q3 Q4 Ptrend储 Dietary vitamin D, IU/d Q1§ Q2 Q3 Q4 Ptrend储 Total dairy, servings/wk Q1§ Q2 Q3 Q4 Ptrend储 Milk, servings/wk Q1§ Q2 Q3 Q4 Ptrend储

Base Model†

Men

Women

Total No. of Deaths

⬍ 578 578 to ⬍ 776 776 to ⬍ 1,044 ⱖ 1,044

⬍ 553 553 to ⬍ 776 776 to ⬍ 1,156 ⱖ 1,156

100 101 107 121

2,733 2,909 2,765 2,913

1.00 0.84 0.88 0.91

⬍ 548 548 to ⬍ 729 729 to ⬍ 949 ⱖ 949

⬍ 486 486 to ⬍ 629 629 to ⬍ 849 ⱖ 849

109 98 106 116

2,556 3,031 2,676 3,058

251 110 68

0 0.1 to ⬍ 250 ⱖ 250

PersonYears

RR

95% CI

MV Model‡ RR

95% CI

— 0.63 to 1.12 0.65 to 1.17 0.68 to 1.20 0.87

1.00 0.87 0.91 0.96

— 0.64 to 1.17 0.67 to 1.23 0.70 to 1.30 0.79

1.00 0.69 0.81 0.74

— 0.52 to 0.92 0.61 to 1.08 0.56 to 0.97 0.13

1.00 0.69 0.82 0.75

— 0.51 to 0.92 0.61 to 1.10 0.56 to 1.00 0.19

6,987 2,682 1,652

1.00 1.00 1.34

— 0.79 to 1.27 0.99 to 1.80 0.12

1.00 1.31 1.65

— 0.94 to 1.83 1.16 to 2.35 0.01

⬍ 122 122 to ⬍ 191 191 to ⬍ 425 ⱖ 425

⬍ 111 111 to ⬍ 201 201 to ⬍ 467 ⱖ 467

109 88 114 118

2,718 2,843 2,792 2,968

1.00 0.69 0.84 0.79

— 0.51 to 0.93 0.64 to 1.12 0.60 to 1.05 0.49

1.00 0.70 0.85 0.91

— 0.52 to 0.95 0.63 to 1.16 0.58 to 1.43 0.85

⬍ 105 105 to ⬍ 157 157 to ⬍ 226 ⱖ 226

⬍ 90 90 to ⬍ 136 136 to ⬍ 202 ⱖ 202

105 102 107 115

2,679 2,826 2,901 2,915

1.00 0.88 0.85 0.84

— 0.66 to 1.18 0.64 to 1.12 0.63 to 1.11 0.22

1.00 0.89 0.86 0.85

— 0.66 to 1.19 0.65 to 1.15 0.63 to 1.13 0.27

⬍ 5.5 5.5 to ⬍ 9.6 9.6 to ⬍ 14.5 ⱖ 14.5

⬍ 5.0 5.0 to ⬍ 8.9 8.9 to ⬍ 13.4 ⱖ 13.4

105 92 116 116

2,490 3,028 2,941 2,862

1.00 0.72 0.83 0.80

— 0.54 to 0.98 0.63 to 1.10 0.59 to 1.09 0.42

1.00 0.73 0.84 0.82

— 0.54 to 0.99 0.63 to 1.12 0.60 to 1.13 0.52

0 0.1 to ⬍ 5.7 5.7 to ⬍ 10.5 ⱖ 10.5

0 0.1 to ⬍ 5.1 5.1 to ⬍ 10.1 ⱖ 10.1

118 99 100 112

3,259 2,304 2,883 2,874

1.00 1.23 0.89 0.89

— 0.93 to 1.64 0.67 to 1.19 0.67 to 1.19 0.14

1.00 1.23 0.91 0.90

— 0.93 to 1.64 0.69 to 1.21 0.68 to 1.20 0.17

Abbreviations: C, category; CPS-II, Cancer Prevention Study-II; MV, multivariable; Q, quartile; RR, relative risk. ⴱ Range obtained from each sex-specific quartile of all exposures, except for supplemental calcium, which was obtained from each category for both sexes combined. †Base model adjusted for age at diagnosis, sex, tumor stage, and 1992 total energy intake. ‡Multivariable model adjusted for age at diagnosis, sex, tumor stage, and 1992 total energy and total folate intakes. §Reference group. 储Ptrend calculated by using the median exposure in each quartile, specific to sex, for all exposures except for supplement calcium, which was calculated using the actual categories (ie, 1, 2, and 3) for both sexes. ¶Supplemental calcium was categorized based on visually inspecting the distribution of the variable.

www.jco.org

© 2014 by American Society of Clinical Oncology

Information downloaded from jco.ascopubs.org and provided by at UCSF LIBRARY & CKM on November 19, 2014 from Copyright © 2014 American Society of Clinical Oncology. All rights reserved. 169.230.243.252

13

Yang et al

Table A5. Associations of Postdiagnosis Calcium, Vitamin D, and Dairy Intakes With Colorectal Cancer Mortality Among Patients With Nonmetastatic Colorectal Cancer in the CPS-II Nutrition Cohort 1999 Rangeⴱ Exposure Total calcium, mg/d Q1§ Q2 Q3 Q4 Ptrend储 Dietary calcium, mg/d Q1§ Q2 Q3 Q4 Ptrend储 Supplemental calcium, mg/d C1§¶ C2 C3 Ptrend储 Total vitamin D, IU/d Q1§ Q2 Q3 Q4 Ptrend储 Dietary vitamin D, IU/d Q1§ Q2 Q3 Q4 Ptrend储 Total dairy, servings/wk Q1§ Q2 Q3 Q4 Ptrend储 Milk, servings/wk Q1§ Q2 Q3 Q4 Ptrend储

Men

Women

2003 Rangeⴱ

Total No. of PersonDeaths Years

Men

Women

⬍ 581 ⬍ 713 581 to ⬍ 775 713 to ⬍ 1,170 775 to ⬍1,105 1,170 to ⬍ 1,598 ⱖ 1,105 ⱖ 1,598

⬍ 683 683 to ⬍ 882 882 to ⬍ 1,162 ⱖ 1,162

⬍ 773 773 to ⬍ 1,131 1,131 to ⬍ 1,591 ⱖ 1,591

37 49 33 24

1,903 2,098 2,078 2,325

⬍ 532 532 to ⬍ 683 683 to ⬍ 885 ⱖ 885

⬍ 613 613 to ⬍ 765 765 to ⬍ 968 ⱖ 968

⬍ 609 609 to ⬍ 766 766 to ⬍ 990 ⱖ 990

35 36 35 37

⬍ 525 525 to ⬍ 671 671 to ⬍ 892 ⱖ 892

0 0.1 to ⬍ 500 ⱖ 500

0 0.1 to ⬍ 500 ⱖ 500

Base Model† RR

95% CI

MV Model‡ RR

95% CI

1.00 — 1.11 0.69 to 1.78 0.75 0.46 to 1.24 0.54 0.31 to 0.94 ⬍ 0.01

1.00 1.15 0.81 0.59

— 0.71 to 1.86 0.48 to 1.38 0.33 to 1.05 0.01

1,804 2,277 2,100 2,223

1.00 — 0.86 0.52 to 1.42 0.90 0.55 to 1.48 0.91 0.56 to 1.47 0.53

1.00 0.85 0.98 1.00

— 0.51 to 1.41 0.59 to 1.62 0.61 to 1.63 0.83

74 42 27

3,966 2,001 2,437

1.00 — 0.92 0.61 to 1.38 0.58 0.35 to 0.95 0.04

1.00 1.04 0.65

— 0.65 to 1.69 0.38 to 1.11 0.13

⬍ 164 164 to ⬍ 302 302 to ⬍ 559 ⱖ 559

⬍ 151 151 to ⬍ 379 379 to ⬍ 588 ⱖ 588

⬍ 194 194 to ⬍ 389 389 to ⬍ 603 ⱖ 603

⬍ 219 219 to ⬍ 509 509 to ⬍ 685 ⱖ 685

33 38 34 38

1,894 2,087 2,167 2,256

1.00 — 0.84 0.51 to 1.38 0.85 0.51 to 1.42 0.90 0.54 to 1.49 0.45

1.00 0.99 1.31 1.74

— 0.59 to 1.66 0.66 to 2.58 0.80 to 3.77 0.52

⬍ 122 122 to ⬍ 178 178 to ⬍ 245 ⱖ 245

⬍ 100 100 to ⬍ 155 155 to ⬍ 229 ⱖ 229

⬍ 132 132 to ⬍ 188 188 to ⬍ 267 ⱖ 267

⬍ 103 103 to ⬍ 178 178 to ⬍ 257 ⱖ 257

32 34 37 40

2,065 2,144 2,109 2,085

1.00 — 0.76 0.45 to 1.28 1.01 0.61 to 1.68 1.18 0.72 to 1.93 0.31

1.00 0.78 1.11 1.28

— 0.46 to 1.32 0.67 to 1.85 0.77 to 2.10 0.19

⬍ 4.7 4.7 to ⬍ 8.2 8.2 to ⬍ 11.7 ⱖ 11.7

⬍ 3.9 3.9 to ⬍ 7.7 7.7 to ⬍ 11.6 ⱖ 11.6

⬍ 5.1 5.1 to ⬍ 8.8 8.8 to ⬍ 12.3 ⱖ 12.3

⬍ 4.8 4.8 to ⬍ 8.1 8.1 to ⬍ 12.4 ⱖ 12.4

37 31 41 34

1,931 2,106 2,068 2,299

1.00 — 0.73 0.44 to 1.22 0.87 0.53 to 1.44 0.71 0.42 to 1.19 0.26

1.00 0.73 0.92 0.73

— 0.44 to 1.23 0.56 to 1.52 0.44 to 1.23 0.32

⬍ 1.1 1.1 to ⬍ 5.6 5.6 to ⬍ 7.0 ⱖ 7.0

⬍ 1.0 1.0 to ⬍ 3.5 3.5 to ⬍ 7.0 ⱖ 7.0

⬍ 1.0 1.0 to ⬍ 5.6 5.6 to ⬍ 7.0 ⱖ 7.0

⬍ 1.0 1.0 to ⬍ 3.3 3.3 to ⬍ 7.0 ⱖ 7.0

33 33 14 63

1,844 2,135 881 3,543

1.00 — 0.88 0.53 to 1.45 0.85 0.43 to 1.65 0.87 0.55 to 1.38 0.61

1.00 0.90 0.85 0.93

— 0.54 to 1.49 0.44 to 1.67 0.59 to 1.49 0.81

Abbreviations: C, category; CPS-II, Cancer Prevention Study-II; MV, multivariable; Q, quartile; RR, relative risk. ⴱ Range obtained from each questionnaire- and sex-specific quartile of all exposures, except for supplemental calcium, which was obtained from each category for both sexes combined. †Base model adjusted for age at diagnosis, sex, tumor stage, and post-diagnosis total energy intake. ‡Multivariable model adjusted for age at diagnosis, sex, tumor stage, and post-diagnosis total energy and total folate intakes. §Reference group. 储Ptrend calculated by using the median exposure in each quartile, specific to sex, for all exposures except for supplement calcium, which was calculated using the actual categories (ie, 1, 2, and 3) for both sexes. ¶Supplemental calcium was categorized based on visually inspecting the distribution of the variable.

14

© 2014 by American Society of Clinical Oncology

JOURNAL OF CLINICAL ONCOLOGY

Information downloaded from jco.ascopubs.org and provided by at UCSF LIBRARY & CKM on November 19, 2014 from Copyright © 2014 American Society of Clinical Oncology. All rights reserved. 169.230.243.252

Calcium, Vitamin D, Dairy, and Colorectal Cancer Survival

Table A6. Associations of Postdiagnosis Calcium, Vitamin D, and Dairy Intakes With All-Cause Mortality Among Patients With Colorectal Cancer in the CPS-II Nutrition Cohort Stratified by Tumor Stage at Diagnosis Localized Stage Exposure Total calcium, mg/d Q1ⴱ Q2 Q3 Q4 Ptrend† Dietary calcium, mg/d Q1ⴱ Q2 Q3 Q4 Ptrend† Supplemental calcium, mg/d C1ⴱ‡ C2 C3 Ptrend† Total vitamin D, IU/d Q1ⴱ Q2 Q3 Q4 Ptrend† Dietary vitamin D, IU/d Q1ⴱ Q2 Q3 Q4 Ptrend† Total dairy, servings/wk Q1ⴱ Q2 Q3 Q4 Ptrend† Milk, servings/wk Q1ⴱ Q2 Q3 Q4 Ptrend†

Regional Stage

Total No. of Deaths

PersonYears

RR

95% CI

55 60 60 46

1,150 1,237 1,190 1,332

1.00 1.00 0.97 0.69

— 0.67 to 1.49 0.64 to 1.46 0.45 to 1.07 0.12

57 58 40 53

— 0.61 to 1.35 0.47 to 1.11 0.55 to 1.25 0.30

55 53 43 57

— 0.59 to 1.29 0.67 to 1.48 0.92

99 60 49

— 0.57 to 1.34 0.60 to 1.60 0.36 to 1.13 0.13

48 49 48 63

— 0.64 to 1.36 0.58 to 1.27 0.47 to 1.08 0.07

42 57 52 57

— 0.63 to 1.38 0.57 to 1.28 0.46 to 1.05 0.06

53 55 49 51

— 0.51 to 1.13 0.43 to 1.28 0.46 to 0.98 0.05

48 54 20 86

50 65 47 59

122 48 51

57 52 60 52

61 58 53 49

62 54 49 56

58 55 21 87

1,031 1,351 1,211 1,316

2,404 1,179 1,326

1,186 1,096 1,270 1,358

1,254 1,241 1,149 1,266

1,196 1,208 1,096 1,409

1,100 1,282 464 2,063

1.00 0.91 0.72 0.83

1.00 0.87 1.00

1.00 0.88 0.98 0.64

1.00 0.93 0.86 0.71

1.00 0.93 0.85 0.69

1.00 0.76 0.75 0.67

Total No. of Deaths

PersonYears 753 861 888 993

773 926 889 907

1,562 822 1,111

708 991 897 897

811 903 960 820

735 898 971 889

744 853 417 1,480

RR

95% CI

Pinteraction

1.00 0.79 0.52 0.74

— 0.52 to 1.20 0.33 to 0.82 0.47 to 1.16 0.09

.08

— 0.50 to 1.14 0.43 to 1.04 0.60 to 1.34 0.49

.75

— 0.69 to 1.61 0.64 to 1.54 0.96

.68

— 0.48 to 1.19 0.55 to 1.77 0.68 to 2.59 0.79

.23

— 0.73 to 1.74 0.69 to 1.67 0.77 to 1.82 0.57

.53

— 0.61 to 1.38 0.40 to 0.98 0.53 to 1.27 0.37

.42

— 0.66 to 1.52 0.43 to 1.31 0.54 to 1.18 0.17

.84

1.00 0.75 0.67 0.90

1.00 1.05 0.99

1.00 0.76 0.99 1.33

1.00 1.13 1.07 1.18

1.00 0.91 0.63 0.82

1.00 1.00 0.75 0.80

NOTE. Only showing multivariable model results adjusted for age at diagnosis, sex, tumor stage, and postdiagnosis total energy and total folate intakes. Abbreviations: C, category; CPS-II, Cancer Prevention Study-II; MV, multivariable; Q, quartile; RR, relative risk. ⴱ Reference group. †Ptrend calculated by using the median exposure in each quartile, specific to sex, for all exposures except for supplement calcium, which was calculated using the actual categories (ie, 1, 2, and 3) for both sexes. ‡Supplemental calcium was categorized based on visually inspecting the distribution of the variable.

www.jco.org

© 2014 by American Society of Clinical Oncology

Information downloaded from jco.ascopubs.org and provided by at UCSF LIBRARY & CKM on November 19, 2014 from Copyright © 2014 American Society of Clinical Oncology. All rights reserved. 169.230.243.252

15