Neo0297

Comparative study of blood insulin levels in breast and endometrial V.B. GAMAYUNOVA, YU. F. BOBROV, E. V. TSYRLINA, T. P. EVTUSHENKO, L. M. BERSTEIN* Laboratory of Endocrinology, Prof. N.N. Petrov Research Institute of Oncology, 189 646 St.Petersburg, Russia Blood insulin level was measured in 113 breast cancer (BC) patients, 18 endometrial cancer (EC) patients, and 35 women with benign breast disease (BBD), after fasting and after 120 min of oral glucose tolerance test (OGTT). A significant increase in reactive insulin level was shown in postmenopausal BC patients with abdominal obesity (waist/hip ratio > 0.85) and no differences in insulin level were found between BC and BBD patients. Menstrual status and overweight (Quetelet index) did not influence significantly blood insulin concentration in BC patients, but the basal insulin level was lower in those patients who had been moderate smokers. In EC patients, the level of insulin after fasting and following 120 min OGTT was much higher than in BC and BBD patients although they had a similar body mass to these groups of patients. The effect of age on insulin secretion in BC patients is discussed as well as the possible causes and consequences of hyperin- sulinemia developing in EC and BC patients.
Key words: Insulin, breast cancer, endometrial cancer, benign breast disease, aging, upper type of obesity, smoking.
Insulin may directly or indirectly promote somatic cell early cancer of the endometrium or ovary but levels fell to division, enhance the activity of lipoprotein lipase, increase normal after surgery or in advanced disease [18].
body fat concentration, and take part in modification of cel- In earlier studies [3, 4, 7, 8, 17] we have reported that hy- lular immunity and metabolic immunodepression (for re- perinsulinemia is a key element leading to hyperlipidemia, view see [8, 11]). In a case-control study of 223 women, obesity, and age-associated disturbances in tolerance to car- BRUNING et al. [5] have concluded that hyperinsulinemia and bohydrates and it was postulated that in breast and endome- insulin resistance are risk markers for breast cancer and are trial cancer patients, the changes in blood insulin and glucose independent of general obesity and body fat distribution.
levels might promote stimulation of tumor growth, or modi- These authors found increased C-peptide concentration in fications in immune defence mechanisms. The present work the serum of breast cancer patients (higher in postmeno- reports the insulin response to the oral glucose tolerance test pausal than premenopausal women) compared to that of (OGTT) in patients with early breast cancer and relates it to patients with other types of cancer (melanoma, cervical can- menstrual status, body mass (Quetelet index), type of fat to- cer, malignant lymphoma). DEL GIUDICE et al. [6] also report pography and smoking. Patients with endometrial cancer or hyperinsulinemia to be a risk factor for premenopausal breast benign breast disease were investigated as control groups.
cancer development but details are not available. It has been suggested that overrich diet and insufficient physical exer- cise may favor the development of hyperinsulinemia and increase the risk of breast cancer in women, especially in those with genetic susceptibility to neoplastic development 113 patients with breast cancer, 18 patients with early [15]. Hyperinsulinemia has also been noted in women with endometrial cancer, and 35 women with benign breast dis- ease were investigated at our Institute between 1992 and 1995. None had received any therapy before the examina- *Author to whom correspondence should be sent.
tion and their ages ranged from 25 to 77. Of 113 breast can- GAMAYUNOVA, BOBROV, TSYRLINA, EVTUSHENKO, BERSTEIN Table 1. Blood glucose and insulin levels (M ± m) in breast cancer (BC), spectively. Also Quetelet index (weight, kg/height (m2)) was endometrial cancer (EC) and benign breast disease (BBD) patients calculated, and as a criterion of fat distribution, the waist/hip ratio was measured. Differences between the studied groups were evaluated by Student’s t-test. Besides mean values (M), standard deviations (σ) and standard errors (m), linear corre- lation coefficients (γ) were also calculated. The computer- 52.1 ± 0.5 69.7 ± 1.2 90.5 ± 1.6 111.7 ± 3.8 ized statistical analysis was used for estimation of the results.
BBD 50.0 ± 2.1 71.8 ± 2.5 94.6 ± 3.3 118.1 ± 6.4 60.1 ± 2.2 70.0 ± 3.5 100.5 ± 2.9b 134.9 ± 9.8a 14.3 ± 2.0b,c 67.1±10.9b,d n = 18a – Difference from BC group is significant (p < 0.05), b – difference from BC group is significant (p < 0.01), c – difference from BBD group is signi- Table 1 shows the results of blood glucose and insulin ficant (p < 0.05), d – difference from BBD group is significant (p < 0.01).
measurement in BC, EC and BBD patients after night fast- ing, and 120 min following the oral glucose load. It is seen from the table, that glucose and insulin levels in BC patients cer (BC) patients (all of them at an early clinical stage), 57 are similar to those in BBD patients and their mean values of patients aged from 25 to 52 were premenopausal, and 56 body weight and age are practically the same. Average values patients aged from 47 to 77 were postmenopausal for more of body weight in EC patients were similar to those in BC and than 12 months. Among 18 patients with early endometrial BBD patients. In EC patients, blood insulin concentrations cancer (EC), 3 women were premenopausal and 15 women sampled following night fasting and by the 120-min of OGTT postmenopausal for at least a year. The third group consist- were significantly higher than those in BC and BBD patients, ed of 35 women with either diffuse or nodular benign breast while their levels of basal and reactive glycemia exceeded disease (BBD) and comprised 22 premenopausal and 13 postmenopausal women. The total number of cases was 166 When the data were analyzed according to whether the and none had a history or evidence of overt diabetes melli- patients were pre- or postmenopausal (Table 2), blood insulin and glucose levels after night fasting and 2 h following the Blood insulin and glucose levels were determined after glucose load were found to be similar in pre- and postmeno- night fasting and during oral glucose tolerance test - peroral pausal BC patients. Mean values of body weight of these pa- glucose load (40 g/m2 of body surface). Glucose concentra- tients were also similar. Postmenopausal BBD patients showed tion in serum sampled from the cubital vein was measured by a significant rise in glycemia level after 120 min OGTT as enzymo-colorimetrical method using special kits (Boehring- compared to premenopausal BBD patients. Postmenopausal er Mannheim Gmbh, FRG). Commercial RIA kits (Nonor- BBD patients also demonstrated a tendency to a larger body ganic Chemistry Institute, Minsk, Belarus) were used for in- mass and an enhanced reactive insulin level. However, an in- sulin determination. Intraassay coefficients of variation for creased blood insulin concentration was most pronounced in insulin and glucose concentrations were 5.0% and 4.3%, re- postmenopausal EC patients although they did not differ sig- Table 2. Blood glucose and insulin levels (M ± m) in pre- and postmenopausal BC, EC and BBD patients a – Difference from BBD group of reproductive age is significant (p < 0.02), b – difference from BC group of reproductive and postmenopausal age is significant (p < 0.02), c – difference from BBD group of postmenopausal age is significant (p < 0.05).
COMPARATIVE STUDY OF BLOOD INSULIN LEVELS IN BREAST AND ENDOMETRIAL CANCER PATIENTS Table 3. Blood glucose and insulin levels (M ± m) in BC patients with different Quetelet index values and different types of fat topography a – Difference from postmenopausal patients with Quetelet index <30 is significant (p < 0.05), b – difference from postmenopausal patients with waist/hip (W/H) ratio <0.85 is significant (p < 0.05).
nificantly in their body mass and glycemia level from the EC Table 4. Coefficients of linear correlation between basal and reactive patients of reproductive age (Table 2).
(120 min after glucose load) blood insulin level and some indices in Table 3 shows that an increase of Quetelet index over 30 was not accompanied by a significant rise in blood insulin concentration either in premenopausal or in postmenopausal BC patients. However, postmenopausal BC patients with the abdominal (or android) type of fat topography (waist/hip ra- insulin insulin insulin insulin insulin insulin tio > 0.85) demonstrated a pronounced hyperinsulinemia by 0 min 120 min 0 min 120 min 0 min 120 min 120 min OGTT as compared with patients with a waist/hip Correlation analysis in breast and endometrial cancer pa- tients (Table 4) shows that anthropometric parameters exert a greater effect on insulin levels in EC patients than in BC pa- tients. In postmenopausal BC patients, the correlation between insulin level and age was stronger than that between insulin Table 5. Blood glucose and insulin levels (M ± m) in pre- and postmeno- pausal smoking and non-smoking BC patients Among BC patients, 35 women (31.0%) were moderate smokers (they smoked on the average 7 cigarettes per day).
When the relationship between the studied parameters and smoking habits was analyzed (Table 5), blood insulin levels non-smoking 90.0 ± 2.7 112.6 ±7.1 7.4 ± 0.9 38.9 ± 7.6 were found to be lowest in postmenopausal BC patients who had smoked over 5 cigarettes a day, whereas glucose concen- 89.2 ± 3.5 100.5 ±5.7 8.9 ± 1.4 29.0 ± 3.5 trations were practically identical in all other groups.
88.2 ± 6.5 101.3 ±9.4 7.9 ± 2.7 32.8 ± 5.6 non-smoking 93.9 ± 2.5 116.5 ±6.6 7.9 ± 0.7 37.6 ± 3.9 82.2 ± 6.6 113.7 ±16.1 8.2 ± 1.6 32.3 ± 12.3 Our study shows that blood insulin levels in BC patients are similar to those in BBD patients and are not affected by 84.1 ± 9.5 100.2 ± 24.9 5.8 ± 0.5* 23.0 ± 10.0 menstrual status or by overweight (Quetelet index). However, only postmenopausal women with the abdominal obesity Difference from non-smoking patients of the same group is significant (waist/hip ratio > 0.85) show a higher level of hyperinsuline- mia by 120 min of OGTT, than those with lower body type of obesity (Tables 1–3). Many authors [3, 12, 16] have noted that hyperinsulinemia and insulin resistance are associated In our study we failed to observe any differences in blood with abdominal obesity, a decreased blood level of sex hor- insulin concentrations between patients with early stage mone-binding globulin (SHBG) and corresponding increase BC, and BBD cases of the same age group and the same of free blood estradiol concentrations.
menstrual status (Table 2). The progression of hyperin- GAMAYUNOVA, BOBROV, TSYRLINA, EVTUSHENKO, BERSTEIN sulinemia and insulin resistance in BC patients with age was noted in our earlier work [4, 8]. In this study, the age- dependent character of insulinemia in BC patients is most [1] BARON, J.A., BYERS, I., GREENBERG, E.R. et al.: Cigarette smok- clearly demonstrated in the group of postmenopausal pa- ing in women with cancer of the breast and reproductive organs. J. Natl. Cancer Inst., 77, 1986, 677–680.
Of a special interest is the level of hyperinsulinemia [2] BERSTEIN, L.M.: Newborn macrosomy and cancer. Adv. Cancer found in the group of EC patients (Tables 1, 2). In these patients, basal blood insulin concentration is more than 1.5 [3] BERSTEIN, L.M.: Increased risk of breast cancer in women with times that of BC and BBD patients. According to the aver- central obesity: Additional considerations. J. Natl. Cancer age data, in EC patients, hyperinsulinemia was not related to the excess of body mass (just as in BC and BBD pa- [4] BOBROV, YU.F., VASILJEVA, I.A., GAMAYUNOVA, V.B. et al.: Blood tients). However, the correlation between insulin level de- insulin level in oncological patients. Vopr. Oncol., 28, 1982, termined after fasting and the Quetelet index, and also be- tween insulin concentration by 120 min of OGTT and waist/ [5] BRUNING, P.F., BONFRER, J.M.G., VAN NOORD, P.A.H. et al.: Insulin resis- tance and breast cancer risk. Int. J. Cancer, 52, 1992, 511–516.
hip ratio, was higher in patients with EC than in those with [6] DEL GIUDICE, M.E., FANTUS, I.G., EYSSEN, G. et al.: Insulin as a risk factor for premenopausal breast cancer. Proc. 18th In postmenopausal patients with EC NAGAMANI et al. [10] Annual San Antonio Breast Cancer Symp., Dec. 1995.
reported the levels of hyperinsulinemia which were more than [7] DILMAN, V.M.: The Law of Deviation of Homeostasis and Dis- 3 times the control figures, and also luteinization of the ova- eases of Aging. Boston, John Wright, PSG, Inc. 1981.
rian stroma, while their glycemia levels did not differ from [8] DILMAN, V.M.: Development, Aging and Disease. Langhorne, those in the control group. The cause of such pronounced PA, Harwood Academic Publishers 1994.
hyperinsulinemia in EC patients is unclear and needs further [9] GODSLAND, I., WYNN, V., WALTON, C., STEVENSON, J.C.: Insulin resis- research. According to recent data, hyperinsulinemia suppress- tance and cigarette smoking. Lancet, 339, 1992, 1619–1620.
es the expression of insulin-like growth factor binding pro- [10] NAGAMANI, M., HANNIGAN, E.V., TUNG, VAN DINH, STUART, C.A.: tein-1 and thus may lead to the excessive and continuous stim- Hyperinsulinemia and stromal luteinization of the ovaries in postmenopausal women with endometrial cancer. J. Clin.
ulation of endometrium by IGFs [13]. On the other hand, it Endocrinol. Metab., 67, 1988, 144–148.
should be taken into account that giving birth to a large baby [11] PLEDGER, W.J.: Regulation of cell proliferation. In: Control of (as an early sign of development of the lowered carbohydrate Cell Growth and Proliferation. Ed.: Baserga, R. New York, tolerance and of resistance to insulin) is much more frequent in case histories of patients with EC than in those of BC pa- [12] PUGEAT, M., GRAVE, J.C., ELMIDANI, M. et al.: Pathophysiology tients [2], and therefore, the factors promoting hyperinsuline- of SHBG: Relation to insulin. J. Steroid Biochem. Mol.
mia development may be formed long before the disease is [13] RUTANEN, E.M., NUMAN, T., LEHTOVIRTA, P. et al.: Suppressed It has been reported that in people smoking more than expression of insulin-like growth factor binding protein-1 20 cigarettes a day, the basal and the reactive insulin levels mRNA in the endometrium: A molecular mechanism as- were elevated in response to glucose load [19], although sociating endometrial cancer with its risk factors. Int. J.
this was not observed in people who smoked moderately [14] SAAH, M.I., RAYGODA, M., GRUNBERG, N.E.: Effect of nicotine [9]. Nicotine infusion to rats during 2–14 days, has led to on body weight and plasma insulin in female and male the decline of blood insulin level [14]. These findings may rats. Life Sci., 55, 1994, 925–932.
be relevant to our data concerning the decrease of basal [15] STOLL, B.A.: Timing of weight gain in relation to breast cancer blood insulin levels in BC patients who smoke moderately risk. Ann. Oncol., 6, 1995, 245–248.
(Table 5). Since, as a rule, BC patients smoke more often [16] STRAIN, G., ZUMOFF, B., ROSNER, W., PI-SUNYER, X.: The relationship than EC patients [1], it cannot be excluded that in BC pa- between serum levels of insulin and sex hormone binding glob- tients, smoking may influence the incidence of hyperin- ulin. J. Clin. Endocrinol. Metab., 79, 1994, 1173–1176.
[17] VISHNEVSKY, A.S., BOBROV, JU.F., TSYRLINA, E.V., DILMAN, V.M.: Confirmation is needed that hyperinsulinemia is a marker Hyperinsulinemia as a factor modifying sensitivity of en- of increased risk to hormone-related cancers such as those of dometrial carcinoma to hormonal influences. Eur. J. Gy- breast and endometrium. Our observations suggest that hy- [18] YAM, D., BEN-HUR, H., FINK, A. et al.: Insulin and glucose sta- perinsulinemia is as common in BBD as in BC cases and there- tus, tissue and plasma lipids in patients with tumours of fore, that it does not merely reflect the presence of cancer but the ovary or endometrium: Possible dietary implications.
Br. J. Cancer, 70, 1994, 1186–1187.
[19] ZAVARONI, I., BONINI, L., CASPARANI, P., REAVEN, G.M.: Cigareate We are very grateful to Prof. B. STOLL (London, UK) for his ad- rmlativ aregrulote iny glucose intolerant, hyperinsuline- vice and help with preparation of this manuscript.
mic and dyslipidemic. Am. J. Cardiol., 73, 1994, 904–905.

Source: http://www.aepress.sk/neo/full/neo297h.pdf