|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 1999 American Society for Clinical Oncology Age, Sex, and Smoking Are Predictors of Circulating Insulin-Like Growth Factor 1 and Insulin-Like Growth Factor–Binding Protein 3From the Division of Endocrinology, Department of Internal Medicine, Beth Israel Deaconess Medical Center, Boston, MA; Department of Internal Medicine, Newton Wellesley Hospital, Newton, MA; and Department of Epidemiology, Athens University Medical School, Athens, Greece. Address reprint requests to Christos Mantzoros, MD, DSc, Endocrinology, RN 325, Beth Israel Deaconess Medical Center, 99 Brookline Ave, Boston, MA 02215; email cmantzor{at}bidmc.harvard.edu
PURPOSE: Insulin-like growth factor (IGF-1) and its major binding protein (IGF-BP3) have recently been implicated in the pathogenesis of several malignancies. However, anthropometric and lifestyle predictors of these hormones have not been elucidated. Here we report the results of a cross-sectional study. SUBJECTS AND METHODS: This cross-sectional study examines the relationship of a series of epidemiologic parameters (age, sex, height, body mass index, smoking, alcohol consumption, and coffee drinking) with IGF-1 and IGF-BP3 in a sample of 130 healthy adults. RESULTS: We observed that serum levels of IGF-1 are higher, whereas levels of IGF-BP3 are lower, in men than in women. In addition, serum levels of IGF-1 are independently and negatively associated with age and positively associated with pack-year history of smoking. Finally, serum levels of IGF-BP3 are independently and negatively associated with the number of cigarettes smoked per day or pack-year history of smoking. CONCLUSION: Age, sex, and smoking are independent predictors of IGF-1 and/or IGF-BP3. The influence of these epidemiologic variables on the pathogenesis of disease states associated with IGF-1 and IGF-BP3 warrants further exploration.
INSULIN-LIKE GROWTH factor 1 (IGF-1) is a peptide hormone that promotes growth and regulates metabolism.1,2 IGF-1 is secreted mainly by the liver and circulates in two states, free and bound to one of six binding proteins.3 One of these, IGF-binding protein 3 (IGF-BP3), accounts for approximately 95% of the total binding activity in plasma.3 IGF-1 has been implicated in the pathophysiology of diabetes mellitus, osteoporosis, and renal disease1-6 and has recently been associated with increased risk for the development of breast and prostate cancer.2,7-11 In addition, serum IGF-BP3 has been associated inversely, and independently of IGF-1, with risk for the development of breast and prostate cancer.8,10 Thus, the identification of modifiable determinants of IGF-1 and IGF-BP3 levels in healthy individuals could potentially lead to the identification of modifiable risk factors for these disease states and could have pathophysiologic and therapeutic implications. However, only two previous epidemiologic studies have evaluated potential determinants of IGF-1 jointly.12,13 In addition, no previous studies have investigated lifestyle predictors of IGF-BP3, whereas the two previous epidemiologic studies have reported conflicting results with respect to the role of lifestyle factors in predicting serum IGF-1 levels.12,13 To better understand the relationship between anthropometric, common sociodemographic, and lifestyle parameters, on one hand, and IGF-1 and IGF-BP3, on the other, we have undertaken a study among apparently healthy, adult Greek subjects.
In the context of an ongoing study on the role of diet in rheumatoid arthritis, we collected blood samples from 130 apparently healthy subjects, 30 to 84 years old, living in the greater Athens area. Fewer than 5% of the randomly selected individuals declined to participate and were replaced. Participants responded to an interviewer-administered questionnaire, giving information on their exact age, height, weight, cigarette smoking habits (number and duration), and usual consumption of coffee (in cups per month) and alcohol (in glasses per month), as previously described.14-17 Although detailed physical examinations were not performed, none of the participants had a history or symptoms of any disease. Specifically, subjects were not malnourished and did not have diabetes mellitus, thyroid pathology, or liver or kidney failure. Blood samples were collected before 10:00 AM and were immediately centrifuged. The frozen samples, packed in dry ice, were shipped from Athens to Beth Israel Deaconess Medical Center in Boston, MA. The coded samples arrived unthawed and were analyzed blindly, in a single run, by laboratory personnel under the supervision of one of the authors (C.S.M.). IGF-1 and IGF-BP3 levels were measured by a commercially available immunoradiometric kit (Diagnostic System Laboratories, Webster, TX), as previously described.9
Student's t test for independent samples was used to examine possible differences in mean IGF-1 and IGF-BP3 levels between men and women. Analysis of IGF-1 and IGF-BP3 levels by tertiles of age, body mass index (BMI), and coffee intake was also performed to check for potential nonlinear associations. Both IGF-1 and IGF-BP3 were approximately normally distributed. Simple and multiple linear regression was used to estimate the association between IGF-1, IGF-BP3, and the study predictor variables (entered as continuous variables, as indicated in the relevant tables). Similar linear regression models were constructed with the variables smoking, alcohol consumption, and coffee intake entered as categorical variables. We also analyzed data using similar regression models after stratifying by age (
Table 1 presents descriptive statistics of the study variables, ie, the mean, SD, median, and first and fourth quartiles of IGF-1, IGF-BP3, age, height, BMI, cigarette smoking (cigarettes/d), alcohol drinking (glasses/mo), and coffee drinking (cups/mo). Table 2 lists the frequency distribution of the predictors of IGF-1 considered in the study. These distributions and descriptive statistics are fairly similar to those previously reported in the Greek population.14
Tables 3 and 4 present the means ± SD of IGF-1 and IGF-BP3 by age and sex and by sex and smoking. We compared IGF-1 and IGF-BP3 levels in men and women using Student's t test. IGF-1 levels are higher among men. The mean ± SD of IGF-1 levels in men was 273.57 ± 131.10 ng/mL, whereas in women it was 221.24 ± 126.40 ng/mL (P = .04). Similarly, IGF-BP3 levels are lower in men than in women. The mean ± SD of IGF-BP3 levels in men was 2,171.94 ± 478.67 ng/mL, whereas in women it was 2,414.33 ± 621.01 ng/mL (P = .02). In addition, the IGF-1/IGF-BP3 ratio is higher in men than in women (mean ± SD, 0.14 ± 0.08 v 0.10 ± 0.06; P = .003). Finally, it appears that the mean IGF-1 may decline by age in both sexes (Table 3). Since most of the studied predictors are interrelated, we constructed multiple regression analysis models.
Tables 5 and 6 present simple age-adjusted and multiple linear regression models. Simple regression of IGF-1 on the study predictors (entered as continuous variables) reveals that only age and sex are significant predictors of circulating IGF-1 levels (Table 5). Similarly, results derived from the multivariate linear regression models indicate that IGF-1 levels are associated inversely, significantly, and independently with age; a significant decline of IGF-1 levels by 36 ng/mL was noted for each 11-year increment (approximately 1 SD) (Table 5). Moreover, similar associations were observed between IGF-1 and age when the other predictors were entered as dichotomous variables in the models. In contrast, no notable relationship was observed between IGF-1 levels and any of the other study variables when entered as continuous, dichotomous, or polychotomous variables. However, when pack-year history of smoking was entered in a similar backward regression analysis model instead of current cigarettes smoked, a significant positive association between IGF-1 levels and this indicator of smoking history was observed (P = .038).
Simple regression of IGF-BP3 on the study predictor variables reveals that only smoking and sex are significant predictors of circulating IGF-BP3 levels (Table 6). The inverse association between smoking and IGF-BP3 persists after adjusting for the other study variables. Importantly, when, instead of current numbers of cigarettes, the pack-year history of smoking is considered as the predictor variable, the negative and independent association between smoking and IGF-BP3 remains significant (P = .03). Stratification by age (
IGF-1 is a peptide hormone secreted mainly by the liver but also produced by several other tissues in response to growth hormone (GH).1,3 In addition to its well-known role in metabolism, IGF-1 has important autocrine and paracrine actions to promote normal growth by acting to multiply and differentiate chondrocytes in the epiphyses of long bones.1,3,18 Moreover, recent in vitro and in vivo data indicate that IGF-1 is a growth factor that can induce malignant cellular proliferation.1,3 Three recent reports have shown that circulating IGF-1 is positively and strongly associated with prostate cancer risk,7-9 and two recent studies have demonstrated that serum IGF-1 levels are associated with breast cancer.10,11 Interestingly, IGF-BP3, the main IGF-1–binding protein, has been inversely and independently associated with risk for developing prostate and breast cancer.8,10,19 Others have suggested that acromegaly, a disease characterized by increased levels of IGF-1, is also associated with increased risk for colon cancer.20 Finally, IGF-1 has been implicated in the pathophysiology of several disease states, such as diabetes mellitus, osteoporosis, and renal disease.1-6 Thus, since IGF-1 and IGF-BP3 are potentially important determinants of disease, especially cancer, epidemiologic identification of factors that influence levels of serum IGF-1 could be of extreme importance. In our study, there is evidence for sexual dimorphism with respect to IGF-1 levels, with men having higher levels than women. In addition, serum IGF-1 levels are negatively and significantly associated with age. Thus, our study, which was performed in a Mediterranean population, is in agreement with earlier epidemiologic investigations in an American12 and a Northern European13 population and is in accord with previous clinical studies reporting similar findings in relation to IGF and age.21-24 Although the underlying mechanism remains unknown, it has been suggested that the association between age and IGF-1 could be related to age-induced changes of GH, the main regulator of IGF-1 synthesis and secretion.3,23 One previous epidemiologic study has reported a positive association between alcohol intake and serum IGF-1,12 whereas the other study reported a negative association between IGF-1 and smoking among men13 and coffee consumption among women.13 Our study did not confirm any of these associations that were shown by only one of the two previous studies, although it did confirm the association with age shown by both previous studies.12,13 In addition, our study demonstrates that, although IGF-1 is not associated with the number of cigarettes currently smoked, there is a positive association between IGF-1 levels and pack-year history of smoking after adjusting for the other study variables. Finally, we found no significant association between IGF-1 and alcohol or coffee consumption. Smoking, alcohol intake, and coffee consumption are interrelated in most populations.15 Thus, it is possible that the discrepant results in previous studies could have been produced because these studies adjusted only for some, but not all, of these lifestyle variables. It appears that mutual adjustment for all three lifestyle variables is a prerequisite of validity as far as the relationships of these variables with IGF-1 are concerned. Interestingly, endogenous IGF-1 levels were not associated with BMI, the index for obesity used in this study. This finding is in agreement with previous observations demonstrating that IGF-1 is not associated with any index of body mass, total body fat, or lean body mass used in previous epidemiologic studies.12,13 Thus, the results of this and the previous two epidemiologic studies are in contrast to findings of experimental studies. Specifically, it has been clearly shown that exogenously administered therapeutic doses of GH that increase IGF-1 levels significantly affect body composition.3 It is possible that therapeutic doses of IGF-1 may have an effect on BMI, whereas IGF-1 within the normal range plays no role of comparable significance in determining body mass as indicated by self-reported BMI. Although alcohol may affect the GH/IGF-1/IGF-BP3 axis in alcoholics with liver cirrhosis,25,26 no epidemiologic studies have previously studied alcohol intake in relation to IGF-1 levels or any lifestyle variables in relation to IGF-BP3 in normal adults. This study has evaluated jointly several predictors of IGF-BP3, including lifestyle factors (ie, smoking, coffee intake, and alcohol consumption). It appears that sex and smoking are significantly related to IGF-BP3 levels. Moreover, in addition to the number of cigarettes currently smoked, the pack-year history of smoking was significantly and inversely associated with IGF-BP3 levels. The inverse and independent association between smoking and IGF-BP3, if confirmed by future studies, may have important pathophysiologic implications. IGF-17-11 and IGF-BP38,11 have recently been implicated in the pathogenesis of breast and prostate cancer. Moreover, increased circulating levels of GH, which regulate both IGF-1 and IGF-BP3, may increase risk for colon cancer20 and have recently been associated with increased mortality from several malignancies, including colon cancer.27 Colon cancer, as well as several other malignancies, has also been associated with smoking.28-30 Interestingly, although an association between current smoking and cancer has been shown by some but not all studies,31-37 it appears that there may be a stronger association between cumulative exposure to smoking and cancer.38,39 It is possible that smoking may increase risk for developing malignancy by decreasing IGF-BP3, a protein that has been implicated in carcinogenesis, acting either directly, by binding to cancer cells and inhibiting cell growth,40 or indirectly, through binding to IGF-1.1-3 This study provides evidence that sex seems to affect IGF-1 and IGF-BP3 levels in healthy adults, as previously described.26 In addition, this study indicates that age is an important predictor of IGF-1, whereas smoking is a significant predictor of IGF-BP3 and possibly IGF-1 levels. Most of the studied potential determinants were not associated with either IGF-1 or IGF-BP3 levels, however. This limited confounding could simplify the design, analysis, and interpretation of future studies on IGF-1 and disease, as previously suggested.12 This study has been undertaken among controls participating in a larger ongoing case-control study of diet in relation to rheumatoid arthritis. Thus, although the distribution of the study variables in this sample does not differ substantially from the corresponding distributions in previously published case-control studies performed in Greece,14 the study subjects do not represent a strictly representative sample of any defined population. However, representativeness is not a prerequisite for validity in this epidemiologic context.41 Some misclassification of laboratory measurements is possible due to variability inherent in the methods used to measure the hormonal variables. However, this random misclassification is expected to suppress effect estimates toward the null and should not have influenced the significant results of this study. In addition, this epidemiologic study is of moderate size. However, we have considered a wide range of lifestyle factors, we have performed appropriate control for mutual confounding, and the analysis has been based on state-of-the-art laboratory measurements. Finally, the present study has not evaluated exercise as a predictor of IGF-1. Neither of the previous population-based epidemiologic studies found an association between exercise/physical activity and IGF-1 levels, however.12,13 In conclusion, we present evidence that smoking, age, and sex, ie, variables frequently used in epidemiologic investigation, may affect levels of IGF-1 and/or IGF-BP3. Inasmuch as these demographic and lifestyle variables could influence the occurrence of IGF-1–related disease states, studies on the role of IGF-1 and IGF-BP3 in the pathophysiology of these diseases should adjust for the potential confounding effects of smoking, age, and sex. Finally, the association between smoking and IGF-1 and IGF-BP3 may have important implications in the pathogenesis of malignancies and should be further explored.
Supported by the Clinical Associate Physician Award, National Institutes of Health and Beth Israel Deaconess Medical Center; the Junior Investigator and the Hershey Family Awards, Beth Israel Deaconess Medical Center and Harvard Medical School (CSM); the Onassis Foundation Scholarship (VGK); and the University of Athens Research Program (EK).
1. LeRoith D, Clemmons D, Nissley P, et al: NIH conference: Insulin-like growth factors in health and disease. Ann Intern Med 116:854-862, 1992 2. Daughaday WH: The possible autocrine/paracrine and endocrine roles of insulin-like growth factors of human tumors. Endocrinology 127:1-4, 1990[Medline] 3. Tritos N, Mantzoros CS: Recombinant human growth hormone: Old and novel uses. Am J Med 105:44-57, 1998[Medline] 4. Bach LA, Rechler MM: Insulin-like growth factors and diabetes. Diabetes Metab Rev 8:229-257, 1992[Medline]
5.
Rudman D, Feller AG, Hangraj HS, et al: Effect of human growth hormone in men over 60 years old. N Engl J Med 323:1-6, 1990 6. Hirschberg R, Kopple JD: Effects of growth hormone and IGF-I on renal function. Kidney Int Suppl 27:S20-S26, 1989 7. Mantzoros CS, Tzonou A, Signorello LB, et al: Insulin-like growth factor 1 in relation to prostate cancer and benign prostatic hyperplasia. Br J Cancer 76:1115-1118, 1997[Medline]
8.
Chan J, Stampfer MJ, Giovannucci E, et al: Plasma insulin-like growth factor-I and prostate cancer risk: A prospective study. Science 279:563-566, 1998
9.
Wolk A, Mantzoros CS, Andersson SO, et al: Insulin-like growth factor 1 and prostate cancer risk: A population-based, case-control study. J Natl Cancer Inst 90:911-915, 1998 10. Hankinson SE, Willett WC, Colditz GA, et al: Circulating concentrations of insulin-like growth factor-I and risk of breast cancer. Lancet 351:1393-1396, 1998[Medline] 11. Bohlke K, Cramer DW, Trichopoulos D, et al: Insulin like growth factor-I in relation to premenopausal ductal carcinoma in situ of the breast. Epidemiology 9:570-573, 1998[Medline]
12.
Goodman-Gruen D, Barrett-Connor E: Epidemiology of insulin-like growth factor-I in elderly men and women: The Rancho Bernardo Study. Am J Epidemiol 145:970-976, 1997 13. Landin-Wilhelmsen K, Wilhelmsen L, Lappas G, et al: Serum insulin-like growth factor I in a random population sample of men and women: Relation to age, sex, smoking habits, coffee consumption and physical activity, blood pressure and concentrations of plasma lipids, fibrinogen, parathyroid hormone and osteocalcin. Clin Endocrinol 41:351-357, 1994[Medline] 14. Signorello LB, Tzonou A, Mantzoros CS, et al: Serum steroids in relation to prostate cancer risk in a case-control study (Greece). Cancer Causes Control 8:632-636, 1997[Medline] 15. Mantzoros CS, Georgiadis EI: Body build and physical activity are important predictors of serum androgen concentrations in young healthy men. Epidemiology 6:432-435, 1995[Medline] 16. Giovannucci E, Rimm EB, Stampfer MJ, et al: Height, body weight, and risk of prostate cancer. Cancer Epidemiol Biomarkers Prev 6:557-563, 1997[Abstract]
17.
Andersson SO, Wolk A, Bergstrom R, et al: Body size and prostate cancer: A 20-year follow-up study among 135,006 Swedish construction workers. J Natl Cancer Inst 89:385-389, 1997 18. Underwood LE, Van Wyk JJ: Normal and aberrant growth, in Wilson JD, Foster DW (eds): Williams Textbook of Endocrinology (ed 8). Philadelphia, PA, Saunders, 1992, pp 1079-1139 19. Bruning PF, Van Doorn J, Boufrer JM, et al: Insulin-like growth factor binding protein 3 is decreased in early-stage operable pre-menopausal breast cancer. Int J Cancer 62:266-270, 1995[Medline] 20. de Herder WW, van der Lely AJ, Lamberts SWJ: Colorectal cancer screening in acromegaly: Still many unresolved questions. Clin Endocrinol 47:644-646, 1997[Medline] 21. Jorgensen JO, Vahl N, Hansen TB, et al: Determinants of serum insulin-like growth factor I in growth hormone deficient adults as compared to healthy subjects. Clin Endocrinol 48:479-486, 1998[Medline] 22. Copras E, Harman SM, Blackman MR: Human growth hormone and human aging. Endocr Rev 14:20-39, 1993[Medline] 23. Cohick WS, Clemmons DR: The insulin-like growth factors. Annu Rev Physiol 55:131-153, 1993[Medline]
24.
Tritos N, Mantzoros CS: Clinical Review: Syndromes of severe insulin resistance. J Clin Endocrinol Metab 83:3025-3030, 1998
25.
Santolaria F, Gonzales-Gonzales G, Gonzales-Reimers E, et al: Effects of alcohol and liver cirrhosis on the GH-IGF-1 axis. Alcohol Alcohol 30:703-708, 1995 26. Gelato MC, Frost RA: IGFBP3: Functional and structural implications in aging and wasting syndromes. Endocrine 7:81-85, 1997[Medline]
27.
Orme SM, McNally RJQ, Cartwright RA, et al for the United Kingdom Acromegaly Study Group: Mortality and cancer incidence in acromegaly: Retrospective cohort study. J Clin Endocrinol Metab 83:2730-2734, 1998 28. Hsing AW, McLaughlin JK, Chow WH, et al: Risk factors for colorectal cancer in a prospective study among U.S. white men. Int J Cancer 77:549-553, 1998[Medline] 29. Heineman EF, Zahm SH, McLaughlin JK, et al: Increased risk of colorectal cancer among smokers: Results of a 26-year follow up of US veterans and a review. Int J Cancer 59:728-738, 1994[Medline] 30. Boyle P: Cancer, cigarette smoking and premature death in Europe: A review including the recommendations of the European Cancer Experts Consensus Meeting, Helsinki, October 1996. Lung Cancer 17:1-60, 1997[Medline] 31. Doll R, Peto R: Mortality in relation to smoking: 20 years' observations on male British doctors. BMJ 2:1525-1536, 1976 32. Hammond EC: Smoking in relation to mortality and morbidity: Findings in first thirty-four months of follow-up in a prospective study started in 1959. J Natl Cancer Inst 32:1161-1188, 1964 33. Weir JM, Dunn JE Jr: Smoking and mortality: A prospective study. Cancer 25:105-112, 1970[Medline] 34. Cederlof R, Friberg L, Hrubec Z, et al: The relationship of smoking and some social covariables to mortality and cancer morbidity: A ten year follow-up in a probability sample of 55,000 Swedish subjects age 18 to 69 (2 parts). Stockholm, Sweden: Karolinska Institute, 1975
35.
Castersen JM, Pershagen G, Eklund G: Mortality in relation to cigarette and pipe smoking: 16 years' observation of 25,000 Swedish men. J Epidemiol Community Health 41:166-172, 1987
36.
Severson RK, Nomura AMY, Grove JS, et al: A prospective study of demographics, diet, and prostate cancer among men of Japanese ancestry in Hawaii. Cancer Res 49:1857-1860, 1989 37. Palmer JR, Rosenberg L: Cigarette smoking and the risk of breast cancer. Epidemiol Rev 5:145-156, 1993
38.
Bennicke K, Conrad C, Sabroe S, et al: Cigarette smoking and breast cancer. BMJ 310:1431-1433, 1995
39.
Ishibe N, Hankinson SE, Colditz GA, et al: Cigarette smoking, cytochrome P450 1A1 polymorphisms, and breast cancer risk in the Nurses' Health Study. Cancer Res 58:667-671, 1998
40.
Oh Y, Muller HL, Lamson G, et al: Insulin-like growth factor (IGF)-independent action of IGF-binding protein-3 in Hs578T human breast cancer cells: Cell surface binding and growth inhibition. J Biol Chem 268:14964-14971, 1993 41. Trichopoulos D, MacMahon B: Epidemiology: Principles and Methods (ed 2). Boston, MA, Little Brown, 1996 Submitted September 3, 1998; accepted November 18, 1998. This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 1999 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
50 and > 50 years). All P values presented in this paper are two tailed. The SPSS statistical package (SPSS for Windows, release 5.0.1; SPSS, Inc, Chicago, IL) was used throughout. 
