- Email: [email protected]
Hyperinsulinemia in colon, stomach and breast cancer patients
ELSEVIER
LETTERS
Cancer Letters 104 (1996) 129-l 32
Hyperinsulinemia in colon, stomach and breast cancer patients Daniel Yama,*, Aaron Finkb, Aaron Mashiahc, Ertzel Ben-Hurd “Weizmann Institute of Science,Department of Membrane Research, bChemical Laboratories, Kaplan Hospital, Rehovot, ‘Department of Surgery, Kaplan Hospital, Rehovot, dDepartment of Obstetn’cs and Gynecology, Kaplan Hospital,
Rehovot 76100, Israel Israel Rehovot. Isrwl
lsruei
Received6 February1996;accepted5 March 1996
Abstract
The objective of this study was to collect more information on the intricate relationship between the presenceof a tumor, insulin status and blood lipids. We selectednon-obesesubjects suffering from colon, stomach and breast cancer and determined the concentration of fasting insulin, glucose, cholesterol, and triglycerides in blood before (BS) and after surgery (AS). Controls were healthy non-obese subjects. Insulin was also measuredin tumors and non-canceroustissues from the same organ. BS insulin and glucose (with the exception of glucose in colon patients) were significantly higher than the controls and fell to almost normal levels at AS. Serum cholesterol and triglycerides levels were reduced in stomachpatients, BS and AS and cholesterol in colon patients BS. Tumors had 1.9-3.0 times as much insulin, or insulin-like substances,as control tissues. These results are consistent with our previous studiesshowing hyperinsulinemia in the presenceof a tumor. Kqwords: Hyperinsulinemia; Tumor
1. Introduction The involvement of insulin in malignancies as growth factor or anabolic hormone (for a review see Yam [I] is inferred from various observations, such as increased uptake of glucose and other nutrients [2] and stimulation of DNA synthesis [3]. Disturbed glucose metabolism associated with hyperinsulinemia in endometrium, ovary [4] and lung cancer patients and insulin-resistance in other types of cancer is well documented [5,6]. In addition, hyperinsulinemia and insulin resistance, beside leading to impaired glucose and lipid metabolism [7], confer ____ * Corresponding author.
I-...-serious advantage to cancer cells, since insulin resistance is absent in these cells [8]. Hobbs and Miller [9] and Yam et al. [4] reported high insulin content in various tumor tissues. Furthermore, there is evidence attributing to certain human tumors the capacity of producing and secreting insulin and to other ‘insulindependent’ tumors the ability of enhancing host insulin secretion [IO]. Some conflicting observations have been reported regarding blood cholesterol and its possible relationship to cancer [ 11,12]. In order to obtain more information on the relationship between tumor. insulin status and blood lipids, we selected subjects among patients suffering from colon, stomach and breast cancer.
0304.3835/96/$12.000 1996ElsevierScienceIrelandLtd. All rightsreserved Pll: 50304.3835(96)0421 I-S
D. Yum et al. /Cancer Letters 104 (1996) 129-132
130
Cholesterol, Boehringer Diagnostica, GmBH, Mannheim, Germany).
2. Materials and methods 2. I. Patients blood and tissue samples The study population comprised non-obese subjects, 16 suffering from colon cancer, 9 from stomach cancer and 10 from breast cancer. Fasting venous blood was taken from 15 non-obese healthy subjects and cancer patients before and 4-6 months after surgery (BS and AS, respectively). Serum was separated and kept at -20°C. Biopsies of cancerous and noncancerous tissues were obtained from the same organ and kept frozen at -20°C. 2.2. Blood analyses Serum insulin concentration was determined by a using 125Iradioimmunoassay double-antibody labeled human insulin (Pharmacia Diagnosis AB, Uppsala, Sweden), triglycerides were measured by an enzymatic procedure according to Fossati and Principe [25] (Triglycerides Enzymatiques PAP 1000, Bio Merieux, Charbonnierres-les, Bains, France) and total cholesterol by an enzymatic calorimetric method according to Siedel et al. [26] (Monotest
2.3. Insulin concentration in tumor and adjacent tissues After being weighed and homogenized, using a fritted glass homogenizer, sonification and centrifugation, tissues were assayed for insulin as in serum. 3. Results Serum insulin levels (Table IA) were significantly elevated in all patients BS (19.4 in colon, 17.8 in stomach and 34.3pU/ml-’ in breast versus 7.9 in controls), and almost normal AS. The same tendency was observed in the patients’ serum glucose with the exception of the colon BS which showed normal levels. Reduced values of triglycerides and cholesterol (Table 1A) characterized stomach patients BS and AS and cholesterol in colon patients BS. The insulin content of cancerous tissues (Table 2) was 2-3 times higher than that of the non-cancerous tissues from the same organ.
Table 1 (A) Concentrations of serum insulin glucose, and (B) triglycerides and cholesterol in controls and cancer surgery (AS)
patients(pat)b&ore(BS)andafter
(A) Tumor
Colon Stomach Breast Control
Patient no.
16 9 10 15
Insulin *U/ml)
Glucose (mg/dl)
BS
AS
BS
AS
19.4 + 5.2* 17.8 *4.1* 34.3 2 11.4*
12.2 + 5.6 10.8 + 4.1 9.1 * 4.8
115.1 f 23.3 144.6 + 24.4* 136.2 + 18.4*
106.2 -c 18.1 95.1 -c 8.4 91.4 I 5.8
7.9 f 3.7
94.8 k 6.8
Triglycerides (mg/dl)
Colon Stomach Breast Control *Significantly
Cholesterol (mg/dl)
BS
AS
BS
AS
131.5 c 38.1 122.9 + 18.4* 158.6 + 36.6
160.4 + 36.3 78.3 f 12.4* 184.6 + 26.1
159.2 * 29.4* 129.4 + 32.1* 199.1 + 28.6
178.4 + 24.4 113.5 f 54.6* 183.5 zk34.3
164.6 f 14.3 different from control values P c 0.05.
204.6 -+ 26.8
D. Yam et al. /Cancer Table 2 Insulin PUlg content of cancerous and non-cancerous colon. stomach and breast cancer patients Cancerous Colon Stomach Breast *Significantly a.05.
tissues
122.6 + 46.2s 85.6 + 36. I * 47.2 -c 18.1* different
from
tissues from
Non-cancerous
tissues
64.5 + 25.9 28.4 r 16.1 22.0 + 9.2 non cancerous
tissues
values,
P <
4. Discussion In this study, colon, stomach and breast cancer blood patients BS, and their cancerous tumors, had significantly higher insulin or insulin-like substances levels compared to controls. These findings extend the results of our previous study with ovarian and endometrial cancer patients [4] in which hyperinsulinemia was accompanied by hyperglycemia, hyperlipidemia and hypercholesterolemia and all blood parameters returned to normal after surgical intervention in non-obese women. In the present study, hyperglycemia was observed only in stomach and breast cancer patients BS and stomach cancer patients both BS and AS were characterized by lower values of triglycerides and cholesterol. Hypocholesterolemia was also observed in colon cancer patients BS. Therefore, this study does not elucidate the conllictinp observation reported regarding hypo- and hypercholesterolemia in cancer [11,12]. However, it is possible that since cholesterol, lipids and glucose metabolism are profoundly affected by insulin [13], abnormal blood levels of these parameters in tumor bearing sub,jects may be secondary to abnormal insulin regulation and metabolism. In addition it is also possible that different types of tumor affect the concentration of fats and other nutrients in blood and normal tissues differently. The assumption of insulin involvement in cancer is reinforced by several studies reporting high insulin activity in several tumors [4,9], or elevated plasma concentrations of insulin or insulin-like substances, sometimes associated with insulin-resistance, in tumor-bearing animals and patients [S,i+l(,]. The role of insulin in cancer promotion is reviewed by Yam [ 1] and is illustrated by the results of studies with alloxan diabetic animals. Of interest
Letters
104 (I 996) 129-f
32
l?i
are the following observations: (1) certain tumors grew more slowly in these animals. (2) There was a lower tumor incidence than in non-diabetic animals. (3) Insulin administration enhanced the growth and tumor incidence. (4) Alloxan diabetic mice became hyperinsulinemic after successful transplantation of tumors [14-161. In this connection it was suggested that certain tumors produce factors that stimulate insulin secretion by pancreatic beta cells 1IO] while others (the most aggressive ones) are insulinproducing/secreting tumors [ 15,161. In view of the above evidence for the involvement of insulin in cancer, agents that reduce insulin secretion or action such as somatostatin 1171 or interferon [18], or block the tyrosine kinase receptor [19] or protein kinase C 1201 may be inhibitory for certain tumor development [ 18-241. Hyperinsulinemia is related to a cluster of metabolic and physiologic disorders such as dyslipidemia, obesity, atherosclerosis, cardiovascular diseases, grouped together under ‘syndrome X’ 1”7].Results in this study reinforce previous suggestions attributing a major role to hyperinsulinemia also in malignancies. Acknowledgements The authors are indebted to P. Budowski for helpful editorial assistance. References I]
Yam, D. ( 1992) Insulin cancer relationships possible dietary implication. Med. Hypotheses, 38, I I l-l 17 11 Jehl, J.A., Mayer. J. and Mckee, K.W. (19%) influence of the hereditary obese hyperinsulinemic syndrome of alloxandiabetes on the survival on mice with Ehrlich ascites carcinoma. Cancer Res., IS, 341-343. 31 Cohen, N.D. and Hilf, R. (19741 Influence of insulin on growth and metabolism of 7,12-dimethylbt:nz(a)anthracene induced mammary tumors. Cancer Res., 34, 3245-3252. [4] Yam, D., Ben-Hur, H., Fink, A., Dgani, R.. Sham, A.. Eliraz. A., Insler, V. and Berry, E.M. (1994) Insulin and glucose SIRIUS. tissue and plasma lipids in patients wtth tumors of the ovary or endometrium: possible dietary implication Br. J. Cancer, 70. I I X6-I 187. [5] Berstein, L.M., Bobrov, Y.F., Ostroumova. M.N. and Dilman, V.M. (1985) Relationship between lipidemia and insulmemia and body surface area and subcutaneous fat tissue condition in patients with cancer of hreaat and lung. Voprosi Onkologii (Leningrad), 3 I, 44. [6] Copeland, G P.. Leinsteri, S.J.. Davic. .I C end Hepkin, LJ
132
D. Yam et al. /Cancer Letters 104 (1996) 129-132
(1987) Insulin resistance in patients with colon cancer. Br. .I. slug., 74, 1031-1037. [7] Reaven, GM. (1988) Role of insulin resistance in human disease. Diabetes, 37, 1595-1607. [8] Mountjoy, K.G., Finlay, G.J. and Holdaway, I.M. (1987) Abnormal insulin receptor down regulation and dissociation of down regulation from insulin action in cultured human tumor cells. Cancer Res., 47,6500-6506. [9] Hobbs, C.B. and Miller, A.L. (1966) Review of endocrine syndromes associated with tumors of non endocrine origin. J. Clin. Pathol., 19, 119-124. [IO] Pavelic, K. and Slijepcevic, M. (1978) Growth of a thymoma in diabetic mice treated with insulin. Eur. J. Cancer, 14,675-682. [I l] Fernleib, M. (1983) Review of the epidemiological evidence for a possible relationship between hypocholesterolemia and cancer. Cancer Res., 43,2503-2507. [I21 De Waard, F. (1975) Breast cancer incidence and nutritional status with particular reference to body weight and height. Cancer Res., 35.3351-3356. [13] Stolar, M.W. (1988) Atherosclerosis in diabetes. The role of hyperinsuhnemia. Metab. Clin. Exp., 37, 11 l-1 17. [14] Yam, D., Fink, A., Nir, I. and Budowski, P. (1991) Insuhntumor interrelationships in Thymoma bearing mice. Effects of dietary glucose and fructose. Br. J. Cancer, 64, 10431049. 1151 Yam, D., Zilberstein, A., Fink, A. and Nir, I. (1990) Insulin tumor interrelationships in EL4lymphoma or Thymoma bearing mice I. Alloxan-diabetic or non diabetic mice. Br. J. Cancer, 61, 689-694. [16] Yam, D., Fink, A., Nir, I. and Budowski, P. (1990) lnsulintumor interrelationship in EL4 lymphoma or Thymoma bearing mice II. Effects of dietary omega-3 and omega-6 polyunsaturated fatty acids. Br. J. Cancer, 62, 897-903.
[17] Reichhn, S. (1983) Somatostatin. N. Engl. J. Med., 309, 1495-1590. [18] Shimizu, F., Shimizu, M. and Kamiyama, K. (1985) Inhibitory effect of interferon on the production of insulin. Endocrinology, I 17, 208 l-2084. [19] Reddy, K.B., Mangold, G.L. and Alberstein, A. (1992) Inhibition of breast cancer cell growth in vitro by a thyrosine kinase inhibitor. Cancer Res., 52, 3638-3641. [20] Kahl-Rainer, P. and Marian, B. (1994) Retinoids inhibit protein kinase C-dependent transduction of 1,Zdiglyceride signals in human colonic tumors. Nutr. Cancer, 21, 1577 168. [21] Setyono-Han, B., Henkelman, MS., Fockens, J.A. and Klijn, J.G.M. (1987) Direct inhibition of somatostatin on the growth of human breast cancer. Cancer Res., 47, 15661570. [22] Liebow, C., Reilly, C., Serrano, D. and Shally, A.V. (1989) Somatostatin analogues inhibit growth of pancreatic cancer by stimulating thyrosine phosphatase. Proc. Natl. Acad. Sci. USA, 86.2003-2007. [23] Urlich, B., Bell, J.R. and Chen, E.Y. (1985) Human insulin receptor and its relationship to the thyrosine kinase family of oncogenes. Nature (London), 3 13.753-761, [24] Dumont, J.A., Jones, W.D. and Bitonti, A.J. (1992) Inhibition of experimental metastasis and cell adhesion of B16 Fl melanoma cells by inhibitors of protein kinase C. Cancer Res., 52, 1195-1200 [25] Fossati, P. and Principe, L. (1982) Serum triglycerides determined calorimetrically with an enzyme that produces hydrogen peroxide. Clin. Chem., 28, 2077-2080. [26] Siedel, J., Hagele, E.O., Ziegenhorn, J. and Wahlefed, A.W. (1983) Reagent for the enzymatic determination of serum total cholesterol with improved lipolytic efficiency. Clin. Chem., 29, 1075-1077.
LETTERS
Cancer Letters 104 (1996) 129-l 32
Hyperinsulinemia in colon, stomach and breast cancer patients Daniel Yama,*, Aaron Finkb, Aaron Mashiahc, Ertzel Ben-Hurd “Weizmann Institute of Science,Department of Membrane Research, bChemical Laboratories, Kaplan Hospital, Rehovot, ‘Department of Surgery, Kaplan Hospital, Rehovot, dDepartment of Obstetn’cs and Gynecology, Kaplan Hospital,
Rehovot 76100, Israel Israel Rehovot. Isrwl
lsruei
Received6 February1996;accepted5 March 1996
Abstract
The objective of this study was to collect more information on the intricate relationship between the presenceof a tumor, insulin status and blood lipids. We selectednon-obesesubjects suffering from colon, stomach and breast cancer and determined the concentration of fasting insulin, glucose, cholesterol, and triglycerides in blood before (BS) and after surgery (AS). Controls were healthy non-obese subjects. Insulin was also measuredin tumors and non-canceroustissues from the same organ. BS insulin and glucose (with the exception of glucose in colon patients) were significantly higher than the controls and fell to almost normal levels at AS. Serum cholesterol and triglycerides levels were reduced in stomachpatients, BS and AS and cholesterol in colon patients BS. Tumors had 1.9-3.0 times as much insulin, or insulin-like substances,as control tissues. These results are consistent with our previous studiesshowing hyperinsulinemia in the presenceof a tumor. Kqwords: Hyperinsulinemia; Tumor
1. Introduction The involvement of insulin in malignancies as growth factor or anabolic hormone (for a review see Yam [I] is inferred from various observations, such as increased uptake of glucose and other nutrients [2] and stimulation of DNA synthesis [3]. Disturbed glucose metabolism associated with hyperinsulinemia in endometrium, ovary [4] and lung cancer patients and insulin-resistance in other types of cancer is well documented [5,6]. In addition, hyperinsulinemia and insulin resistance, beside leading to impaired glucose and lipid metabolism [7], confer ____ * Corresponding author.
I-...-serious advantage to cancer cells, since insulin resistance is absent in these cells [8]. Hobbs and Miller [9] and Yam et al. [4] reported high insulin content in various tumor tissues. Furthermore, there is evidence attributing to certain human tumors the capacity of producing and secreting insulin and to other ‘insulindependent’ tumors the ability of enhancing host insulin secretion [IO]. Some conflicting observations have been reported regarding blood cholesterol and its possible relationship to cancer [ 11,12]. In order to obtain more information on the relationship between tumor. insulin status and blood lipids, we selected subjects among patients suffering from colon, stomach and breast cancer.
0304.3835/96/$12.000 1996ElsevierScienceIrelandLtd. All rightsreserved Pll: 50304.3835(96)0421 I-S
D. Yum et al. /Cancer Letters 104 (1996) 129-132
130
Cholesterol, Boehringer Diagnostica, GmBH, Mannheim, Germany).
2. Materials and methods 2. I. Patients blood and tissue samples The study population comprised non-obese subjects, 16 suffering from colon cancer, 9 from stomach cancer and 10 from breast cancer. Fasting venous blood was taken from 15 non-obese healthy subjects and cancer patients before and 4-6 months after surgery (BS and AS, respectively). Serum was separated and kept at -20°C. Biopsies of cancerous and noncancerous tissues were obtained from the same organ and kept frozen at -20°C. 2.2. Blood analyses Serum insulin concentration was determined by a using 125Iradioimmunoassay double-antibody labeled human insulin (Pharmacia Diagnosis AB, Uppsala, Sweden), triglycerides were measured by an enzymatic procedure according to Fossati and Principe [25] (Triglycerides Enzymatiques PAP 1000, Bio Merieux, Charbonnierres-les, Bains, France) and total cholesterol by an enzymatic calorimetric method according to Siedel et al. [26] (Monotest
2.3. Insulin concentration in tumor and adjacent tissues After being weighed and homogenized, using a fritted glass homogenizer, sonification and centrifugation, tissues were assayed for insulin as in serum. 3. Results Serum insulin levels (Table IA) were significantly elevated in all patients BS (19.4 in colon, 17.8 in stomach and 34.3pU/ml-’ in breast versus 7.9 in controls), and almost normal AS. The same tendency was observed in the patients’ serum glucose with the exception of the colon BS which showed normal levels. Reduced values of triglycerides and cholesterol (Table 1A) characterized stomach patients BS and AS and cholesterol in colon patients BS. The insulin content of cancerous tissues (Table 2) was 2-3 times higher than that of the non-cancerous tissues from the same organ.
Table 1 (A) Concentrations of serum insulin glucose, and (B) triglycerides and cholesterol in controls and cancer surgery (AS)
patients(pat)b&ore(BS)andafter
(A) Tumor
Colon Stomach Breast Control
Patient no.
16 9 10 15
Insulin *U/ml)
Glucose (mg/dl)
BS
AS
BS
AS
19.4 + 5.2* 17.8 *4.1* 34.3 2 11.4*
12.2 + 5.6 10.8 + 4.1 9.1 * 4.8
115.1 f 23.3 144.6 + 24.4* 136.2 + 18.4*
106.2 -c 18.1 95.1 -c 8.4 91.4 I 5.8
7.9 f 3.7
94.8 k 6.8
Triglycerides (mg/dl)
Colon Stomach Breast Control *Significantly
Cholesterol (mg/dl)
BS
AS
BS
AS
131.5 c 38.1 122.9 + 18.4* 158.6 + 36.6
160.4 + 36.3 78.3 f 12.4* 184.6 + 26.1
159.2 * 29.4* 129.4 + 32.1* 199.1 + 28.6
178.4 + 24.4 113.5 f 54.6* 183.5 zk34.3
164.6 f 14.3 different from control values P c 0.05.
204.6 -+ 26.8
D. Yam et al. /Cancer Table 2 Insulin PUlg content of cancerous and non-cancerous colon. stomach and breast cancer patients Cancerous Colon Stomach Breast *Significantly a.05.
tissues
122.6 + 46.2s 85.6 + 36. I * 47.2 -c 18.1* different
from
tissues from
Non-cancerous
tissues
64.5 + 25.9 28.4 r 16.1 22.0 + 9.2 non cancerous
tissues
values,
P <
4. Discussion In this study, colon, stomach and breast cancer blood patients BS, and their cancerous tumors, had significantly higher insulin or insulin-like substances levels compared to controls. These findings extend the results of our previous study with ovarian and endometrial cancer patients [4] in which hyperinsulinemia was accompanied by hyperglycemia, hyperlipidemia and hypercholesterolemia and all blood parameters returned to normal after surgical intervention in non-obese women. In the present study, hyperglycemia was observed only in stomach and breast cancer patients BS and stomach cancer patients both BS and AS were characterized by lower values of triglycerides and cholesterol. Hypocholesterolemia was also observed in colon cancer patients BS. Therefore, this study does not elucidate the conllictinp observation reported regarding hypo- and hypercholesterolemia in cancer [11,12]. However, it is possible that since cholesterol, lipids and glucose metabolism are profoundly affected by insulin [13], abnormal blood levels of these parameters in tumor bearing sub,jects may be secondary to abnormal insulin regulation and metabolism. In addition it is also possible that different types of tumor affect the concentration of fats and other nutrients in blood and normal tissues differently. The assumption of insulin involvement in cancer is reinforced by several studies reporting high insulin activity in several tumors [4,9], or elevated plasma concentrations of insulin or insulin-like substances, sometimes associated with insulin-resistance, in tumor-bearing animals and patients [S,i+l(,]. The role of insulin in cancer promotion is reviewed by Yam [ 1] and is illustrated by the results of studies with alloxan diabetic animals. Of interest
Letters
104 (I 996) 129-f
32
l?i
are the following observations: (1) certain tumors grew more slowly in these animals. (2) There was a lower tumor incidence than in non-diabetic animals. (3) Insulin administration enhanced the growth and tumor incidence. (4) Alloxan diabetic mice became hyperinsulinemic after successful transplantation of tumors [14-161. In this connection it was suggested that certain tumors produce factors that stimulate insulin secretion by pancreatic beta cells 1IO] while others (the most aggressive ones) are insulinproducing/secreting tumors [ 15,161. In view of the above evidence for the involvement of insulin in cancer, agents that reduce insulin secretion or action such as somatostatin 1171 or interferon [18], or block the tyrosine kinase receptor [19] or protein kinase C 1201 may be inhibitory for certain tumor development [ 18-241. Hyperinsulinemia is related to a cluster of metabolic and physiologic disorders such as dyslipidemia, obesity, atherosclerosis, cardiovascular diseases, grouped together under ‘syndrome X’ 1”7].Results in this study reinforce previous suggestions attributing a major role to hyperinsulinemia also in malignancies. Acknowledgements The authors are indebted to P. Budowski for helpful editorial assistance. References I]
Yam, D. ( 1992) Insulin cancer relationships possible dietary implication. Med. Hypotheses, 38, I I l-l 17 11 Jehl, J.A., Mayer. J. and Mckee, K.W. (19%) influence of the hereditary obese hyperinsulinemic syndrome of alloxandiabetes on the survival on mice with Ehrlich ascites carcinoma. Cancer Res., IS, 341-343. 31 Cohen, N.D. and Hilf, R. (19741 Influence of insulin on growth and metabolism of 7,12-dimethylbt:nz(a)anthracene induced mammary tumors. Cancer Res., 34, 3245-3252. [4] Yam, D., Ben-Hur, H., Fink, A., Dgani, R.. Sham, A.. Eliraz. A., Insler, V. and Berry, E.M. (1994) Insulin and glucose SIRIUS. tissue and plasma lipids in patients wtth tumors of the ovary or endometrium: possible dietary implication Br. J. Cancer, 70. I I X6-I 187. [5] Berstein, L.M., Bobrov, Y.F., Ostroumova. M.N. and Dilman, V.M. (1985) Relationship between lipidemia and insulmemia and body surface area and subcutaneous fat tissue condition in patients with cancer of hreaat and lung. Voprosi Onkologii (Leningrad), 3 I, 44. [6] Copeland, G P.. Leinsteri, S.J.. Davic. .I C end Hepkin, LJ
132
D. Yam et al. /Cancer Letters 104 (1996) 129-132
(1987) Insulin resistance in patients with colon cancer. Br. .I. slug., 74, 1031-1037. [7] Reaven, GM. (1988) Role of insulin resistance in human disease. Diabetes, 37, 1595-1607. [8] Mountjoy, K.G., Finlay, G.J. and Holdaway, I.M. (1987) Abnormal insulin receptor down regulation and dissociation of down regulation from insulin action in cultured human tumor cells. Cancer Res., 47,6500-6506. [9] Hobbs, C.B. and Miller, A.L. (1966) Review of endocrine syndromes associated with tumors of non endocrine origin. J. Clin. Pathol., 19, 119-124. [IO] Pavelic, K. and Slijepcevic, M. (1978) Growth of a thymoma in diabetic mice treated with insulin. Eur. J. Cancer, 14,675-682. [I l] Fernleib, M. (1983) Review of the epidemiological evidence for a possible relationship between hypocholesterolemia and cancer. Cancer Res., 43,2503-2507. [I21 De Waard, F. (1975) Breast cancer incidence and nutritional status with particular reference to body weight and height. Cancer Res., 35.3351-3356. [13] Stolar, M.W. (1988) Atherosclerosis in diabetes. The role of hyperinsuhnemia. Metab. Clin. Exp., 37, 11 l-1 17. [14] Yam, D., Fink, A., Nir, I. and Budowski, P. (1991) Insuhntumor interrelationships in Thymoma bearing mice. Effects of dietary glucose and fructose. Br. J. Cancer, 64, 10431049. 1151 Yam, D., Zilberstein, A., Fink, A. and Nir, I. (1990) Insulin tumor interrelationships in EL4lymphoma or Thymoma bearing mice I. Alloxan-diabetic or non diabetic mice. Br. J. Cancer, 61, 689-694. [16] Yam, D., Fink, A., Nir, I. and Budowski, P. (1990) lnsulintumor interrelationship in EL4 lymphoma or Thymoma bearing mice II. Effects of dietary omega-3 and omega-6 polyunsaturated fatty acids. Br. J. Cancer, 62, 897-903.
[17] Reichhn, S. (1983) Somatostatin. N. Engl. J. Med., 309, 1495-1590. [18] Shimizu, F., Shimizu, M. and Kamiyama, K. (1985) Inhibitory effect of interferon on the production of insulin. Endocrinology, I 17, 208 l-2084. [19] Reddy, K.B., Mangold, G.L. and Alberstein, A. (1992) Inhibition of breast cancer cell growth in vitro by a thyrosine kinase inhibitor. Cancer Res., 52, 3638-3641. [20] Kahl-Rainer, P. and Marian, B. (1994) Retinoids inhibit protein kinase C-dependent transduction of 1,Zdiglyceride signals in human colonic tumors. Nutr. Cancer, 21, 1577 168. [21] Setyono-Han, B., Henkelman, MS., Fockens, J.A. and Klijn, J.G.M. (1987) Direct inhibition of somatostatin on the growth of human breast cancer. Cancer Res., 47, 15661570. [22] Liebow, C., Reilly, C., Serrano, D. and Shally, A.V. (1989) Somatostatin analogues inhibit growth of pancreatic cancer by stimulating thyrosine phosphatase. Proc. Natl. Acad. Sci. USA, 86.2003-2007. [23] Urlich, B., Bell, J.R. and Chen, E.Y. (1985) Human insulin receptor and its relationship to the thyrosine kinase family of oncogenes. Nature (London), 3 13.753-761, [24] Dumont, J.A., Jones, W.D. and Bitonti, A.J. (1992) Inhibition of experimental metastasis and cell adhesion of B16 Fl melanoma cells by inhibitors of protein kinase C. Cancer Res., 52, 1195-1200 [25] Fossati, P. and Principe, L. (1982) Serum triglycerides determined calorimetrically with an enzyme that produces hydrogen peroxide. Clin. Chem., 28, 2077-2080. [26] Siedel, J., Hagele, E.O., Ziegenhorn, J. and Wahlefed, A.W. (1983) Reagent for the enzymatic determination of serum total cholesterol with improved lipolytic efficiency. Clin. Chem., 29, 1075-1077.