Phytosterols and Cholesterol in Malignant and Benign Breast Tumors1

[CANCER RESEARCH 37, 3034-3036. September 1977]

Margot J. Mellies, Terry T. Ishikawa, Charles J. Glueck, and John D. Crissman

Lipid Research and General Clinical Research Centers [M. J. M., T. T. I., C. J. G.j and the Department of Pathology [J. D. C], University of Cincinnati, College of Medicine, Cincinnati. Ohio 45267

SUMMARY of phytosterols in the tissue and plasma was endogenous or exogenous. No direct analyses of the tissue phytosterol and Tissue phytosterol and cholesterol levels in 10 benign and cholesterol content of normal breast or benign breast tu 8 malignant breast tumors were quantitated to reexamine mors were made as controls (3). Subsequently, in 1969, Day the hypothesis that malignant tumors had distinctive phy ef al. (1) examined tissue from 8 women with breast carci tosterol content. Phytosterols were present in 9 of 10 be noma, finding "traces of substances with retention times nign and 7 of 8 malignant breast tumors. Mean (Â±S.E.) similar to those of the phytosterols reported to be present cholesterol, campesterol, stigmasterol, and 0-sitosterol in by Gordan" in 1 subject. No data on normal breast tissue or malignant and benign tumors (^9/9 wet weight) did not benign adenomas were presented. The hypothesis of Gor significantly differ (p > 0.1): dan ef al, (3) and Day ef a/. (1) of distinctive phytosterol content in malignant tumors has not been further evaluated Cholesterol Campesterol Stigmasterol /3-Sitosterol in the 10-year interval since its proposal, so that neither Malignant 756 Â±244 13 Â±9 10 Â±6 17 Â±9 validation nor nonconfirmation are available. We became interested in breast tumor sterols (1, 3) after Benign 1205 Â±848 14 Â±6 18 Â±10 28 Â±12 reporting that phytosterol-rich diets led to 3- to 5-fold incre ments in plasma phytosterols in infants (6) and the subse In the malignant tumors, tissue cholesterol correlated quent observation that infants fed phytosterol-rich formulae with campesterol (r = 0.97) and ÃŸ-sitosterol (r = 0.97) (p < rapidly accrue phytosterols in their aortic tissues (7). The 0.01), but not with stigmasterol (r = -0.06). In benign current study was designed to examine malignant and be tumors, tissue cholesterol correlated with campesterol (r = nign breast tumors with a focus on their sterol content, to 0.43), stigmasterol (r = 0.64), and /3-sitosterol (r = 0.94), reexamine the hypothesis that malignant tumors had dis with p < 0.01 for the latter two. Phytosterols were present in tinctive elevations of the phytosterols (1, 3). four samples of normal breast tissue with mean (Â±S.E.) campesterol, stigmasterol, and /3-sitosterol (2 Â±0.8, 15 Â±9, MATERIALS AND METHODS 7 Â±5 Â¿ig/gwet weight) slightly but not significantly lower than in benign and malignant breast tumors, p > 0.1. The Patients. Breast tissue samples were provided fresh from comparability of tissue phytosterols in benign and malig the surgical pathology laboratory after frozen section. Ap nant breast tumors and in normal breast tissue appears to proximately 1 g of tissue was obtained (in the sequential render unlikely any putative etiological relationship be order of submission to the pathology department) and was tween phytosterols and breast carcinoma. analyzed with a blinded code number. Tissue diagnoses were provided for the coded samples only after completion INTRODUCTION of the sterol analyses. Of the 22 samples, 10 were from women with benign fibroadenoma or fibrocystic disease, 8 In 1967 Gordan ef al. (3) reported the presence of choles were from women with carcinomas of the breast, and 4 were terol, campesterol, ÃŸ-sitosterol, stigmasterol, and a stig- from normal breast tissue. Analysis of Phytosterols in Tissues, with Gas-Liquid masteryl ester in 11 of 12 human breast cancers. In plasma samples from breast cancer patients, Gordan ef al. ob Chromatographie Conditions. The breast tumor samples served relative elevations of stigmasteryl ester, 7-dehydrosi- were washed in iced 0.9% NaCI solution to remove traces tosteryl ester, or both (3). Similar plasma peaks were found of blood and coagulated material. The tissue was then in 2 controls who were nursing mothers (3). There were no blotted dry, weighed, and finely minced, and a Folch ex data to indicate whether the apparent excess accumulation traction (2) was carried out by an initial homogenization step in cold chlorofornrmethanol (2:1) with a mechanical blender. An internal recovery standard of [14C]cholesterol ' Supported in part by the General Clinical Research Center Grant RR00068-14. A portion of this work was done during Dr. Glueck s tenure as was added at the beginning of the extraction process. After an Established Investigator of the American Heart Association, 1971 to 1976. Folch extraction the sample was saponified, extracted Studies approved by Institutional Human Experimentation Committee. In with heptane, and concentrated; 5-a-cholestane internal formed consent obtained. Rsearch carried out according to the Declaration of Helsinki. standard was added, and the total tissue sterols were quan Received March 8, 1977; accepted May 27, 1977. titated by gas-liquid chromatography (6, 7) (Table 1). Free

3034 CANCER RESEARCH VOL. 37

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1977 American Association for Cancer Research. Phytosterols in Breast Tumors and esterified sterols were not quantitated separately. Whitney nonparametric test for differences (Ref. 8, p. 128). Mean Â±S.D. recovery of [14C]cholesterol in the 18 breast Relationships between tumor cholesterol and phytosterols tumor tissue samples was 94 Â±7%. Mean Â±S.D. recovery were studied by regression analysis (Ref. 8, p. 135). of ^-['"Clsitosterol in 6 breast tumor samples was 95 Â± 6%. The sterol analyses were made on Perkin-Elmer 900 and RESULTS Hewlett-Packard 402B gas chromatographs with flame ioni- zation detectors. The instruments were equipped with silan- Cholesterol and Phytosterol Levels in Benign and Malig ized glass columns 1.8 x 2 mm packed with 3% OV-17 on nant Breast Tumors, and in Normal Breast Tissue. Phyto 100/120 mesh Gas Chrom Q (Ohio Valley Speciality Chemi sterols were present in 9 of 10 benign breast tumors. Des- cal, Inc., Marietta, Ohio). The column temperature was mosterol was present in 3 of the 10 samples (Table 1). There maintained at 250Â°,the injection port was held at 290Â°,and was a wide range of variation in both cholesterol and phy 300Â°were used for the detector temperature. The flow rate tosterol levels from sample to sample. of 30 ml/min was used for the nitrogen carrier gas, the hy Phytosterols were present in 7 of the 8 breast carcinomas, drogen was maintained at 25 ml/min, and nitrogenioxygen and desmosterol was present in 3 of them (Table 1). As was (60:40) was set at 300 ml/min. The columns were precondi the case for the benign lesions, there was wide variation in tioned in situ with Col-Treet (Regis No. 970069) before the both cholesterol and phytosterol levels from sample to sam sterol analyses. All of the gas-liquid Chromatographie con ple. ditions, as above, were as previously described (7). The In a comparison of the benign and malignant tumors, sterols were quantitated directly in their alcohol forms with there were no significant differences in mean levels of tis out prior derivatization (4, 5). The tissue sterols gave elution sue cholesterol, campesterol, stigmasterol, or ÃŸ-sitosterol times similar to those previously reported (4, 5). The sterol (Table 1). The mean levels and variance for the 3 phytoste fraction was not subjected to silver nitrate thin-layer chro- rols were similar for the benign and malignant tumor sam matography to separate the possible presence of 5-u-satu- ples, whereas mean cholesterol was somewhat higher (but rated stanols and A7-sitosterol (3). Under these experi highly variable) in the benign lesions (Table 1). mental conditions, in the presence of 288 M9 of tissue Phytosterols were present in all 4 normal breast tissues, cholesterol per g wet weight, as little as 0.4, 0.7, and 0.8 p.g and desmosterol was present in 3. Mean (Â±S.E.) choles of campesterol, stigmasterol, and ÃŸ-sitosterol were quan terol, campesterol, stigmasterol, and ÃŸ-sitosterol levels titated (Table 1). were, respectively, 206 Â±59, 2 Â±0.8,15 Â±9, and 7 Â±5 /Â¿g/g Statistical Analyses. The phytosterol and cholesterol le wet weight. In comparison with both benign and malignant vels in the benign and malignant tumors and in normal tumors, normal breast tissue campesterol, stigmasterol, breast tissue were compared by using the Wilcoxon-Mann- and ÃŸ-sitosterol content was slightly but not significantly

Table 1 Content of cholesterol, desmosterol, campesterol, stigmasterol, and ÃŸ-sitosterol(ng/g wet weight) in benign and malignant breast tumors Sample No. Diagnosis Cholesterol Desmosterol Campesterol Stigmasterol 0-Sitosterol tumors245891314161718FibroadenomaFibroadenomaFibroadenomaFibroadenomaFibrocysticFibrocysticFibroadenomaFibrocysticFibroadenomaFibrocystic13774188759495157Benign

7.7169364

.266169 1

1.27117.82.63.716.63.61.40.62.648.911.02.229.45.01.81.62.587.727.52.589.223.010.15.83.72.587.7

Mean Â±S.E. 1205 Â±848 14.2 Â±5.7 17.6 Â±10.5 28.0 Â±11.8 Malignant tumors 136710111519Mean 0.5 1.6 1.0 1.6 1.8 10.2 7.4 16.2 67.6 44.0 14.7 8.0 3.8 3.1 12.5 0.4 0.7 0.8 4.7 4.9 4.4 4.2 6.6 66.2

Â±S.E.CarcinomaCarcinomaCarcinomaCarcinomaCarcinomaCarcinomaCarcinomaCarcinoma2788132372747288781412353756Â±244 (I 13.2 Â±9.1(NS) 9.7 Â±5.8(NS) 16.6 Â±8.6(NS) NS, no significant difference when compared with benign tumors (p > 0.1).

SEPTEMBER 1977 3035

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1977 American Association for Cancer Research. M. J. Mellies et al. lower (p > 0.1). Mean cholesterol content of the normal rols are probably accumulated in breast tissue and breast breast tissues was lower than that of malignant (p < 0.05) tumors in a fashion similar to those in aorta, by deposition but not benign tumors (p > 0.1). from exogenous sources (7). The comparability of tissue Relationships between Tissue Cholesterol and Phytos- phytosterols in aorta, in normal breast tissue, and in benign terols. In the benign tumors, tissue cholesterol levels corre and malignant breast tumors appears to render unlikely any lated with those of campesterol (r = 0.43), stigmasterol (r = putative etiological relationship between the phytosterols 0.64), and jÃ¶-sitosterol (r = 0.94) (p < 0.01 for the latter 2). In and carcinoma of the breast (1, 3). the carcinomas, tissue cholesterol correlated with campes Unfortunately, we were not able to obtain equitemporal terol (r = 0.97) and /S-sitosterol (r = 0.97) (p < 0.01) but not plasma samples from the women with benign and malignant with stigmasterol (r = -0.06). tumors of the breast to reexamine the finding of Gordan ef al. (3) of elevated plasma plant sterols in women with malig nant breast tumors. The content and composition of sterols DISCUSSION in the plasma of various patients are critical for comparison with respective tumor tissues to determine whether, like When compared to benign breast tumors, malignant tu aorta (7), they simply reflect plasma composition. The com mors of the breast do not have a distinctive phytosterol parable tissue levels of phytosterols in benign and malig content. In addition, the phytosterol content of benign and nant breast tumors and in normal breast tissues suggest malignant tumors is not significantly higher than that of that plasma phytosterol levels would have little predictive normal breast tissues. The suggestion by Gordan ef al. (3) specificity in regard to the malignant and nonmalignant and Day et al. (1) of increased phytosterol content of malig nature of the breast lesion. nant breast tumors was not, then, validated when adequate controls, e.g., benign breast tumors, were also analyzed. In the 10 benign breast tumors, cholesterol, campesterol, stig REFERENCES masterol, and /3-sitosterol levels (mean Â±S.E.) were, re spectively, 1205 Â±848, 14 Â±6, 18 Â±10, and 28 Â±12 ng/g, 1. Day, E A., Malcolm, G. T., and Beeler, M. F. Tumor Sterols. Metabolism, J8. 646-651, 1969. with values for the 8 malignant tumors being, respectively, 2. Folch, J.. Lees, M., and Sloane-Stanley, G. H. A Simple Method for the 756 Â± 244, 13 Â± 9, 10 Â± 6, and 17 Â± 9 M9/9- Tissue Isolation and Purification of Total Lipids from Animal Tissue. J. Biol. Chem.,226. 497-509, 1957. cholesterol and phytosterol levels in both benign and malig 3. Gordan, G. X., Fitzpatrick, M. E., and Lubich, W. P. Identification of nant tumors were highly correlated. Osteolytic Sterols in Human Breast Cancer. Trans. Assoc. Am. Physi The levels of the phytosterols in the breast tumors were cians, 80. 182-189, 1967. 4. Ishikawa, T. T., Brazier, J. B., Stewart, L. E., Fallal, R. W., and Glueck, similar to those in normal aortic tissue samples. In 11 sam C. J. Direct quantitation of plasma cholesterol by gas-liquid chromatog- ples of normal aorta, cholesterol, campesterol, stigmas raphy. J. Lab. Clin. Med., 87. 345-353, 1976. terol, and ÃŸ-sitosterol levels (mean Â±S.E.) were, respec 5. Ishikawa, T. T., MacGee, J., Morrison, J. A., and Glueck, C. J. Quantita tive analysis of cholesterol in five to twenty microliters of plasma. J. Lipid tively, 3442 Â±687, 14 Â±4, 13 Â±4, and 16 Â±4 /Â¿g/gwet Res., 15: 286-291, 1974. weight (7). 6. Mellies. M., Glueck, C. J., Sweeny, C., Fallal, R., Tsang, R., and Ishi kawa. T. T. Plasma and Dietary Phytosterols in Children. Pediatrics, 57. The remarkable comparability of tissue phytosterols in 60-67, 1976. normal aortic tissue (7), in normal breast tissue, and in 7. Mellies, M. J., Ishikawa, T. T., Glueck, C. J., Bove, K., and Morrison, J. benign and malignant breast tumors indicates the lack of Phytosterols in Aortic Tissue in Adults and Infants. J. Lab. Clin. Med., 88. 914-921, 1976. any distinctive (1, 3) concentration of phytosterols in breast 8. Snedecor. G. W., and Cochran. W. G. Statistical Methods, Ed. 6. pp. 128, tumors, malignant or benign. Moreover, the similarities in 135. Ames, Iowa: Iowa State College Press, 1967. aortic and breast tissue phytosterols suggest that phytoste

3036 CANCER RESEARCH VOL. 37

Margot J. Mellies, Terry T. Ishikawa, Charles J. Glueck, et al.

Cancer Res 1977;37:3034-3036.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/37/9/3034

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/37/9/3034. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.