© 2000 by Oxford University Press
Journal of the National Cancer Institute, Vol. 92, No. 23, 1949,
December 6, 2000
© 2000 Oxford University Press
CORRESPONDENCE |
RESPONSE: More About: Prospective Study of Colorectal Cancer Risk in Men and Plasma Levels of Insulin-Like Growth Factor (IGF)-I and IGF-Binding Protein-3
Affiliations of authors: A. G. Renehan, S. T. O'Dwyer (Department of Surgery), S. M. Shalet (Department of Endocrinology), Christie Hospital National Health Service Trust, Manchester, U.K.
Correspondence to: Andrew G. Renehan, F.R.C.S., F.D.S., Department of Surgery, Christie Hospital National Health Service Trust, Wilmslow Rd., Manchester M20 4BX, U.K. (e-mail: arenehan{at}picr.man.ac.uk).
The IGFs are multifunctional regulatory peptides with characteristics of both classic "endocrine" hormones and tissue growth factors. There is considerable interindividual variation in circulating levels of IGF-I and its main binding protein, IGFBP-3, and it is postulated that this heterogeneity of IGF physiology may determine cancer risk. A number of recent prospective and case–control studies [reviewed in (1)] have supported this hypothesis for many common malignancies, including colorectal cancer. For colorectal cancer, the implication is that tumorigenesis may be hormone related, thus challenging conventional understanding. The contribution to the debate by Paterson et al. is, therefore, timely. This cross-sectional study compared two "at-risk" groups, namely, individuals with a family history of colorectal cancer and individuals with adenomas (known precursors of the disease), and found no association with plasma IGF-I levels compared with control subjects confirmed via colonoscopy to be free of colorectal cancer.
The findings by Paterson et al. raise a number of interesting issues. The failure to measure circulating levels of IGFBP-3 is an important omission in this study, since cancer risk appears to be positively associated with IGF-I and negatively associated with IGFBP-3. Since IGF-I and IGFBP-3 are generally well correlated with one another, this apparent paradox suggests that it is important to pay attention to "outliers" on the IGF-I versus IGFBP-3 plot (1). Laboratory evidence also becomes relevant: The mitogenic and antiapoptotic properties of IGF-I are well known and, in addition, IGF-independent proapoptotic actions of IGFBP-3 are now recognized and have been demonstrated for colorectal adenoma cell lines (2).
Circulating IGF-I and IGFBP-3 are considerably influenced by age and sex, and adjustments for these are complex but very relevant. It is well recognized that levels of both peptides decline with age, but the curvilinear nature of these changes and the differences between males and females at different ages are seldom appreciated (Fig. 1
). Normal manufacturers' reference ranges for IGF-I test kits (as presumably used by Paterson et al.) seldom fully reflect these relationships. Furthermore, adjustments to statistical relationships between variables by use of multivariate linear regression analysis may not always be appropriate, since it assumes a linear relationship among the variables. In the present study, subjects with adenomatous polyps were, on average, 14 years older than the control subjects, and the lower (rather than elevated) mean IGF-I level observed in these patients may simply reflect this age difference.
|
The distinction between early and advanced adenomas also appears to be important. In 365 healthy individuals (aged 55–64 years old) undergoing screening flexible sigmoidoscopy, we initially found no association between IGF-I, IGFBP-3, and the presence of all adenomas (3). We subsequently revisited this cohort with extended numbers of subjects with high-risk adenomas (
1 cm, tubulovillous/villous histology, severe dysplasia) (4). Our reanalysis revealed that IGF-I was positively (relative risk [RR] per standard deviation change = 4.39 [95% confidence interval = 1.31–14.7]) and IGFBP-3 was negatively (RR per standard deviation of change = 0.41 [95% confidence interval = 0.20–0.82]) associated with the presence of advanced adenomas. Giovannucci et al. (5) have recently reported similar relationships with advanced adenomas in women from the Nurses' Health Study. Finally, a family history of colorectal cancer is an important risk factor for the development of the disease (6). As a whole, it is unlikely that circulating IGF-I and/or IGFBP-3 levels differ substantially in this group from the general population. However, among those with a family history, we found that the ability of IGF-I/IGFBP-3 to predict advanced adenomas remains (4).
REFERENCES
1 Pollak M. Insulin-like growth factor physiology and cancer risk. Eur J Cancer 2000;36:1224–8.cancerlit;20340774
2
Williams AC, Collard TJ, Perks CM, Newcomb P, Moorghen M, Holly JM, Paraskava C. Increased p53-dependent apoptosis by the insulin-like growth factor binding protein IGFBP-3 in human colonic adenoma-derived cells. Cancer Res 2000;60:22–7.
3 Renehan AG, O'Dwyer ST, Shalet SM. Re: Prospective study of colorectal cancer risk in men and plasma levels of insulin-like growth factor (IGF)-I and IGF-binding protein-3. J Natl Cancer Inst 1999;23:2051–2.cancerlit;20048025
4 Renehan AG, Painter JE, Atkin WS, Potten CS, Shalet SM, O'Dwyer ST. "High-risk" colorectal adenomas and insulin-like growth factors. Br J Surg. In press 2001.
5
Giovannucci E, Pollak MN, Platz EA, Willett WC, Stampfer MJ, Majeed N, et al. A prospective study of plasma insulin-like growth factor-1 and binding protein-3 and risk of colorectal neoplasia in women. Cancer Epidemiol Biomarkers Prev 2000;9:345–9.
6
Fuchs CS, Giovannucci EL, Colditz GA, Hunter DJ, Speizer FE, Willett WC. A prospective study of family history and the risk of colorectal cancer. N Engl J Med 1994;331:1669–74.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||