Dietary intakes of calcium and vitamin D and

Dietary intakes of calcium and vitamin D and risk of colorectal cancer in women Jennifer Lin, Ph. D Division of Preventive Medicine Brigham and Women’s Hospital Harvard Medical School, USA jhlin@rics. bwh. harvard. edu

2004 Estimated US Cancer Cases Prostate Lung & bronchus Colon & rectum Urinary bladder Melanoma of skin Non-Hodgkin lymphoma Kidney Oral Cavity Leukemia Pancreas All Other Sites 33% 11% 6% 4% Men 699, 560 Women 668, 470 4% 3% 3% 3% 2% 18% Source: American Cancer Society, 2004. 32% 11% 6% 4% 4% 4% 3% 2% 2% 20% Breast Lung & bronchus Colon & rectum Uterine corpus Ovary Non-Hodgkin lymphoma Melanoma of skin Thyroid Pancreas Urinary bladder All Other Sites

2004 Estimated US Cancer Deaths Lung & bronchus Prostate Colon & rectum Pancreas Leukemia Non-Hodgkin lymphoma Esophagus Liver & bile duct intrahepatic Urinary bladder Kidney All other sites 32% 10% 5% 5% 4% Men 290, 890 Women 272, 810 4% 3% 3% 3% 21% Source: American Cancer Society, 2004. 25% 10% 6% 6% 4% 3% 3% 2% 2% 24% Lung & bronchus Breast Colon & rectum Ovary Pancreas Leukemia Non-Hodgkin lymphoma Uterine corpus Multiple myeloma Brain/ONS All other sites

CRC Incidence rate by age & sex Source: National Cancer Institute for 1997 -2001

Anatomy of the Colon and Rectum

Calcium and vitamin D and colorectal cancer incidence • Calcium may reduce risk of colorectal cancer by binding to carcinogenic bile acids and ionized fatty acids. • Vitamin D has the potential to regulate cell proliferation and is important in calcium absorption in the gut. • The RDA recommends calcium intake of 10001200 mg/day and vitamin D of 5 -15 μg/day (200400 IU/day) for healthy adults.

Calcium and colorectal cancer incidence—(1) • Extensive efforts have been made in epidemiologic studies to evaluate calcium intake and colorectal cancer risk. However, findings have been inconclusive. • A meta-analysis of 10 cohort studies observed a 14% risk reduction in the highest quintile of total calcium intake relative to the lowest quintile (Cho et al, 2004), while the other metal-analysis study reported no association (Bergsma-Kadijk et al 1996).

Calcium and colorectal cancer incidence—(2) • Clinical trials of calcium supplementation for the prevention of recurrence of colorectal adenomas: (1) one observed a 30% risk reduction after 1 year of follow-up among participants taking 1200 mg/day of calcium (Baron et al, 1999); (2) another trial found that 2000 mg/day of calcium after 3 years of follow-up was associated with a modest risk reduction (Bonithon-Kopp et al, 2000).

Vitamin D and colorectal cancer incidence • Observational findings on the association between vitamin D intake and colorectal cancer risk have also been inconclusive. • Only three studies have reported a doseresponse trend of lower colorectal cancer risk with higher vitamin D intake (La Vecchia et al 1992; Garland et al 1995; Mc. Cullough et al 2003).

Possible explanations of discrepant findings on calcium and vitamin D intake • Methodological limitations: calcium and vitamin D supplements were not available in some studies, which may have attenuated the associations. • Other dietary factors: Intakes of fat, phosphorus, fiber may have reduced calcium absorption in the gut.

In this study (Lin et al, 2005, Am J Epi) … • We prospectively evaluated baseline intakes of calcium and vitamin D using both dietary sources and supplements from a large female cohort in the Women’s Health Study. • We also evaluated whether the effect of calcium on colorectal cancer risk was modified by intakes of total fat, phosphorus, total fiber and vitamin D.

Statistical Analysis • Relative risks (RR) and 95% confidence intervals (CI) were estimated from Cox proportional hazards regression. Tests for trend were performed by fitting the median value of each category as a continuous variable in the models. All p values were two sided. • The models were controlled for age and randomized trial assignment, and additionally for body mass index, physical activity, family history of colorectal cancer in a first degree relative, history of colon or rectal polyps, smoking status, multivitamin use, postmenopausal hormone therapy, saturated fat intake, red meat intake, and total energy intake.

The Women’s Health Study • A completed randomized trial evaluating low-dose aspirin and vitamin E for the prevention of cardiovascular disease and cancer among 39, 876 women aged 45 years. • At baseline, participants provided information on lifestyle exposures, dietary intake, and medical history characteristics. • During an average follow-up of 9 years, 212 women had a confirmed diagnosis of colorectal cancer (85 cases had primary tumor occurred in the proximal colon, 73 in the distal colon, and 40 cases in the rectum).

Results– (1) • Table 1 presents the baseline distributions of risk factors for colorectal cancer according to intakes of total calcium and vitamin D. Women who consumed greater amounts of calcium and vitamin D tended to be older and health conscious. • Intake of total calcium intake was not associated with risk of colorectal cancer (Table 2). • Intake of total vitamin D intake was also not shown to be related to risk of colorectal cancer (Table 3).

Results – (2) • Major food sources of calcium and vitamin D were also not associated with colorectal cancer risk (Tables 4 a and 4 b). • Test for the interaction of calcium intake with intakes of total fat, phosphorus, vitamin D, and total fiber in relation to colorectal cancer was not statistically significant (p values for interaction >0. 2). No significant associations could be observed in the stratified analyses according to these four nutrients (Tables 5 a and 5 b). • The overall results were also not changed according to tumor location (colon, rectum) (Data not shown).

Table 1. Baseline characteristics according to quintiles of calcium and vitamin D intake in the Women’s Health Study (Q=quintile).

Table 2. Relative risks (RRs) and 95% confidence intervals (CIs) of colorectal cancer according to quintile intakes of calcium in the Women Health’s Study.

Table 3. Relative risks (RRs) and 95% confidence intervals (CIs) of colorectal cancer according to quintile intakes of vitamin D in the Women Health’s Study

Table 4 a. Relative risks and 95% confidence intervals (CIs) of colorectal cancer according to quintile intakes of specific dietary sources of calcium and vitamin D in the Women Health’s Study.

Table 4 b. Relative risks and 95% confidence intervals (CIs) of colorectal cancer according to quintile intakes of specific dietary sources of calcium and vitamin D in the Women Health’s Study.

Table 5 a. Relative risks and 95% confidence intervals of colorectal cancer according to quintile of total calcium intake by modifying factors.

Table 5 b. Relative risks and 95% confidence intervals of colorectal cancer according to quintile of total calcium intake by modifying factors.

Limitations of our study • Residual confounding. • Insufficient power for subgroup analysis. • Inadequate measures of vitamin D intake: the synthesis of vitamin D in the skin from the sun is an important source, however, we had no information on sun exposure. • Intakes of calcium and vitamin were only measured once at baseline.

Conclusion • Our findings do not support a protective role of calcium and vitamin D intake against colorectal cancer incidence. • However, given the evidence from both animal and in vitro studies, the benefits of these two nutrients cannot be ruled out.

References • • Cho et al (2004). J Natl Cancer Inst, 96: 1015 -22. Bergsma-Kadijk et al (1996). Epidem, 7: 590 -7. La Vecchia et al (1997). Int J Cancer, 73: 525 -30. Garland C et al (1985). Lancet, 1: 307 -9. Mc. Cullough ML et al (2003). Cancer Causes Control, 14: 1 -12. Bonithon-Kopp et al (2000). Lancet, 3561300 -6. Baron et al (1999). N Engl J Med, 340: 101 -7. Lin et al (2005). Am J Epi, 161: 755 -64.