RESEARCH · February 20, 2013
TAKE-HOME MESSAGE
- Colorectal cancer in women was found to correlate with elevated levels of homocysteine and low levels of cysteine in a case-controlled analysis of postmenopausal women from the Women’s Health Initiative Observational Study.
SUMMARY
PracticeUpdate Editorial Team
Chronic inflammation of the large intestine and rectum can contribute to the development of colorectal cancer (CRC). Folate deficiency is a known risk factor for CRC and can lead to raised homocysteine levels. Whether hyperhomocysteinemia, an independent risk factor for vascular disease possibly mediated by inflammation, is linked to the development of CRC is uncertain. Cysteine level has also been shown to be a risk factor for vascular disease, but minimal information is available regarding associations between cysteine and CRC.
This retrospective, nested, case-control analysis used data from the Women’s Health Initiative Observational Study (WHI-OS) to investigate associations between homocysteine and cysteine and CRC risk. Cases comprising 988 postmenopausal women with confirmed invasive CRCs were matched to 988 controls. CRC cases had a higher mean body mass index, longer smoking duration, and were less physically active than controls. In addition, more cases than controls had a history of colon polyp removal, but more controls than cases had undergone colonoscopy.
CRC cases had a significantly higher geometric mean homocysteine level than did controls (9.42 µmol/L vs 9.12 µmol/L; P = .04), whereas there was no significant between-group difference in geometric mean plasma cysteine level (283.9 µmol/L vs 284.0 µmol/L). Among CRC cases, a greater proportion of women had homocysteine levels within the upper quartile (> 9.85 µmol/L) than did controls, but there was no difference between groups in the percentage distribution among quartiles for cysteine level.
As baseline homocysteine level increased, a significant trend (P = .02) toward increased likelihood of CRC was identified by adjusted multivariate analysis. Compared with the lowest quartile of homocysteine level (≤ 6.75 µmol/L), the odds ratios (ORs) for risk of CRC for women with homocysteine levels in the second (> 6.74 µmol/L to 7.85 µmol/L), third (> 7.85 µmol/L to 9.85 µmol/L), and fourth (> 9.85µmol/L) quartiles were 1.06 (95% CI, 0.79–1.41), 1.05 (95% CI, 0.79–1.40), and 1.46 (95% CI, 1.05–2.04), respectively. This significant trend was evident for all tumors combined but also independently for proximal colon tumors (P for trend = .008) and localized tumors (P for trend = .01).
Conversely, there was a significant trend (P = .01) toward a decreased likelihood of CRC as baseline cysteine level increased. Compared with the lowest quartile of cysteine level (≤ 260 µmol/L), the ORs for risk of CRC for women with cysteine levels in the second (> 260 µmol/L to 282 µmol/L), third (> 282 µmol/L to 309 µmol/L), and fourth (> 309 µmol/L) quartiles were 0.73 (95% CI, 0.54–0.98), 0.86 (95% CI, 0.63–1.16), and 0.57 (95% CI, 0.40–0.82), respectively. Again, this significant trend was evident for all tumors combined but also independently for rectal tumors (P for trend = .02) and localized tumors (P for trend = .003).
This analysis of the WHI-OS demonstrated that increasing homocysteine level and decreasing cysteine level are risk factors for CRC.
The American Journal of Clinical Nutrition
Homocysteine, Cysteine, and Risk of Incident Colorectal Cancer in the Women’s Health Initiative Observational Cohort
Am J Clin Nutr 2013 Feb 20;97(4)827-834, JW Miller, SA Beresford, ML Neuhouser, et al