D. Y. WANG, J. L. HAYWARD and R. D. BULBROOK
Division of Chemistry and Biochemistry, Imperial Cancer Research Fund, Lincoln’s Inn Fields, London, W.C.2
Introduction
THERE is a close relationship between the amounts of androgen metabolites in the urine and the clinical course of breast cancer. The majority of patients with widespread métastasés who excrete subnormal amounts of andro-sterone and aetiocholanolone do not respond to adrenalectomy or hypophysectomy [1] or to pituitary destruction by yttrium-90 [2]. A similar abnormality in steroid excretion is found in about half the patients with the early disease (before metastatic spread is found) and these patients also have a poor prognosis [3, 4].
The urinary androgen metabolites are derived from at least four precursors (déhydro-épiandrostérone and its sulphate, androstene-dione, and testosterone) which means that deductions about the concentration of precursors in the blood cannot be made from measurements of urinary metabolites [5]. In an attempt to investigate the physiological meaning of the correlation between urinary steroid levels and the clinical course of breast cancer direct measurements of plasma 17-oxosteroids and 17-hydroxycorticosteroids have been made [6] and preliminary investigations of secretion rates have been carried out [7].
This paper is concerned with the levels of free testosterone in the plasma of normal women and patients with benign breast disease or early breast cancer.
Methods
(a) Subjects
Blood (20 ml) was taken into heparinised syringes from the antecubital vein between 2 p.m. and 4 p.m. for normal women and between 10 a.m. and 11 a.m. for patients with breast disease. The time difference is unlikely to be important since there is no diurnal variation in plasma testosterone levels [8]. The plasma prepared from these samples was stored at —20°. Blood samples from premenopausal women were taken at random during the menstrual cycle. Minor variations in plasma testosterone levels have been reported during the cycle [9] but this variation would be unlikely to alter any of the conclusions in the present study since the proportion of premenopausal women in each of the three groups of subjects was approximately equal.
The patients with benign breast disease con-sisted of fourteen women admitted to hospital to await diagnosis and subsequent surgery. Their ages ranged from 24 to 59 years (mean 43 • 4). The patients with breast cancer were twenty-four women, in hospital awaiting radical mastectomy. The ages of these patients ranged from 25 to 84 years (mean 49-7). All diagnoses were confirmed eventually by histological examination. The normal women were nineteen volunteers, apparently healthy, who were housewives living in the North London area. Their ages ranged from 25 to 76 years (mean 5T9).
(b) Chemical
Plasma testosterone was estimated by the method of Burger, Kent and Kellie [10]. Since there was the possibility of contamination of the testosterone fraction by epitestosterone [11] an additional purification step was introduced into the method. This was a preliminary purification of the plasma testosterone extract by paper chromatography (2:2:4 trimethylpentane: toluene/methanol: water. 450:50/400: 100).
(c) Precision
Six estimations were carried out on pooled plasma from women. The mean value was O'069 μg/100 ml plasma with a standard deviation of 0-026 μg/100 ml. Similar experiments on plasma from males gave a mean value of T038 μg/100 ml with a standard deviation of 0-071 μg/100 ml. Errors of this order have been reported previously [12] and are due mainly to the relatively large amounts of marker 14C-testostcrone added to the plasma. Since the mean level of testosterone found in this study was 0-12 μg/100 ml the coefficient of variation would be of the order of ±20%.
(d) Specificity
The ratio of tritium to 14C in the last four chromatographic steps of the analytical method are shown in Fig. 1. The ratios are constant in the last two steps which implies that the testosterone acetate is radiochemically pure.
Figure 1. Ratio of tritium to carbon-H during purification of doubly labelled testosterone acetate.
Results
The distribution of plasma testosterone levels with age are shown in Fig. 2 for normal women and for patients with benign breast disease or breast cancer. There is no correlation with age in any of these groups of patients.
Figure 2. Plasma testosterone levels and age in normal women, patients with benign breast disease, and patients with early breast cancer.
O = normal women
Δ = women with benign breast disease
■ = women with early breast cancer
The individual values for each group are shown in Fig. 3: the mean values are almost identical. The mean value for normal women is 0-12±0-12 μg/100 ml (standard deviation). These values are in agreement with others in the literature [8, 12, 13]. The mean value for the breast cancer cases is 0-12±0-09 μg/100 ml and is 0-10±0-08 μg/100 ml for those with benign breast disease. There was no correlation between the plasma testosterone concentration and the amounts of androsterone or aetio-cholanolone in the urine (see also [8, 9]).
Figure 3. Comparison of plasma testosterone levels in normal women, patients with benign breast disease, and patients with early breast cancer.
Discussion
It has been assumed that the subnormal excretion of urinary androgen metabolites by patients with breast cancer who have a poor prognosis after mastectomy or after endocrine ablation was a reflection of a deficient production of androgens [14, 15]. The present results showing normal plasma testosterone levels in both benign and maligant breast disease indicate that this interpretation of the results of assays of urinary androgen metabolites must be regarded with caution.
The abnormally low urinary excretion, found in some patients, of androsterone and aetiocholanolone is probably related to a deficiency in the secretion of déhydroépi-androstérone and its sulphate. This would be expected since the major portion of the urinary 1 l-deoxy-17-oxosteroids are derived from these two precursors [5]. Furthermore, a good correlation between the levels of plasma 17-oxo-steroids (mainly the sulphates of déhydroépiandrostérone and androsterone) and those of the urinary 1 l-deoxy-17-oxosteroids has been reported [6].
this compound was the major androgen in women [5]. It has recently been shown [16, 17] that androstenedione is present in much higher concentrations than testosterone in the peripheral blood of normal women and that previous estimates of the amounts of testosterone secreted are probably over-estimates [5].
Before any statement can be made concerning androgenic status in women with breast cancer it would be desirable to obtain simultaneous measurements of both androstenedione and testosterone (and of the rates of their inter-conversion) in these patients. A more sensitive and precise method than that used in this study such as that employing 35S-labelled thiosemi-carbazide [18] would be essential in such an investigation. A second essential factor is the elucidation of the physiological roles of déhydroépiandrostérone and its sulphate in women. It seems unlikely that rapid progress can be made in the investigation of endocrinological aspects of human breast cancer until more is known about the androgenic stimulus in normal women.
Résumé
Chez les patientes atteintes de cancer du sein, il existe une corrélation entre le pronostic de Vaffection et les taux urinaires des métabolites des androgènes. Afin de voir si ces données peuvent être étendues aux taux des androgènes dans la circulation périphérique, la testostérone a été dosée dans le plasma de trois groupes de sujets: des femmes normales, des patientes atteintes d’une affection bénigne du sein et des patientes atteintes de cancer du sein à un stade précoce. Aucune différence n’a été observée. Les valeurs moyennes {et la déviation standard) du taux delà testostérone plasmatique furent respectivement dans les trois groupes: 0-12±0,12 μg\l00ml; O,10±0,08 μg/lOO ml et 0,12±0,09 μg/100 ml. On ne trouve aucune corrélation entre ces concentrations de testostérone et l’excrétion urinaire des métabolites des androgènes.
Summary
The prognosis of patients suffering from breast cancer is related to the urinary levels of androgen metabolites. In an attempt to extend thisfinding, in terms of the levels of androgen in the peripheral circulation, plasma testosterone levels have been determined in three groups of women : normal, patients with benign breast disease, and patients suffering from early breast cancer. No differences in plasma testosterone levels were found between any of the groups, the values for them being 0-12±0-12 μg/100 ml (.standard deviation), 0-10± 0-08 μg/100 ml and 0-12±0-09 μg/100 ml, respectively. No correlation was found to exist between these testosterone levels and the urinary excretion of androgen metabolites.
Zusammenfassung
Bei den von einem Brustkrebs befallenen Patientinnen besteht ein bestimmtes Verhältnis zwichen der Krankheitsprognose und der Konzentration der im Ham ausgeschiedenen Androgenen Stoffwechselprodukte. Um zu bestimmen, ob diese Angaben auch bezüglich des Androgenen-Spiegels in dem zirkulierenden Plasma gültig sind, hat man den Plasma-Testosteronspiegel bei 3 Gruppen von Patientinnen bestimmt: bei normalen Frauen, bei solchen, von einer gutartigen Brusterkrankung befallen, und bei Frauen an einem im Frühstadium Brustkrebs leidenden. Man konnte keinen Unterschied bezüglich des Plasma-Testosteronspiegels dieser 3 Gruppen fiststellen. Die gefundenen Werte {mit der Standardabweichung) waren wie folgt: 0,12 ±0,12 μg/100 ml ml; 0,10±0,08 μg/100 ml und 0,12±0,09 μg/100 ml. Es besteht kein Zusammenhang zwischen diesen Plasma-Testosteronwerte und der Konzentration der mit dem Harn ausgeschiedenen Androgenen Stqffwechselprodukte.
Acknowledgements
The authors wish to thank Miss L. Peacock for skilful technical assistance.
References
R. D. BULBROOK, J. L. HAYWARD, C. C. SPICER and B. S. THOMAS, A comparison between the urinary steroid excretion of normal women and women with advanced breast cancer. Lancet ii, 1235 (1962).
P. JURET, M. HAYEM and A. FLEISLER, A propos de 150 implantations d’yttrium radioactif intrahypophysaires dans le traitement du cancer du sein à un stade avancé. J. Chir. [Paris) 87, 409 (1964).
R. D. BULBROOK, J. L. HAYWARD and B. S. THOMAS, The relation between the urinary 17-hydroxycorticosteroids and 1 l-deoxy-17-oxosteroids and the fate of patients after mastectomy. Lancet i, 945 (1964).
R. D. BULBROOK and J. L. HAYWARD, The possibility of predicting the response of patients with early breast cancer to subsequent endrocine ablation. Cancer Res. 25, 1135 (1965).
R. L. VANDE WIELE, P. C. MACDONALD, E. GURPIDE and S. LIEBERMAN, Studies on the secretion and interconversion of the androgens. Recent Progr. Horn. Res. 19, 275 (1963).
N. DESHPANDE, J. L. HAYWARD and R. D. BULBROOK, Plasma 17-hydroxy-corticosteroids and 17-oxosteroids in patients with breast cancer and in normal women. J. Endocr. 32, 167 (1965),
li. D. BULBROOK, J. L. HAYWARD and R. A. A. SALOKANGAS, The secretion rate of dehydroepiandrosterone and cortisol in patients with advanced breast cancer. J. Endocr. 26,1-II (1963).
E. J. LAMB, W. J. DIGNAM, R. J. PION AND H. H. SIMMER, Plasma androgens in women. Acta Endrocr. 45, 243 (1964).
J. LOBOTSKY, H. I. WYSS, E. J. SEGRE and C. W. LLOYD, Plasma testosterone in the normal woman. J. din. Endocr. 24, 1261 (1964).
H. G. BURGER, J. R. KENT and A. E. KELLIE, Determination of testosterone in human peripheral and adrenal venous plasma. J. din. Endocr. 24, 432 (1964).
1L S. G. KORENMAN, H. WILSON, and M. LIPSETT, Isolation of 17α-hydroxyandrost-4-еп-З-опе (epitestosterone) from human urine. J. Biol. Chem. 239, 1004 (1964).
B. HUDSON, J. COGHLAN, A. DULMANIS, M. WINTOUR and I. EKKEL, The estimations of testosterone in biological fluids: I. Testosterone in plasma. Aust. J. exp. biol. med. Sci. 41, 235 (1963).
E. FORCHIELLI, G. SORCINI, M. S. NIGHTINGALE, N. BRUST and R. I. DORFMAN, Testosterone in human plasma. Analyt. Biochem. 5, 416 (1963).
R. D. BULBROOK, Hormonal factors in the aetiology and treatment of breast cancer. In Proc. UhCanadianCancerConf. (R. W. BEGG, Editor), pp. 36-49 (1966).
R. D. BULBROOK, Hormone assays in human breast cancer. Vitam. and Harm 23, 329 (1965).
R. HORION, Estimation of androstenedione in human peripheral blood with a5S-thiosemicarbazide. J. din. Endocr. 25, 1237 (1965).
R. HORTON and J. F. TAIT, Androstenedione production and interconversion rates measured in peripheral blood and studies on the possible sites of its conversion to testosterone. J. din. Invest. 45, 301 (1966)
A. RIONDEL, J. F. TAIT, M. GUT, S. A. S. TAIT, E. JOACHIM and B. LITTLE, Estimation of testosterone in human peripheral blood using S35-thiosemicar-bazide. J. din. Endocr. 23, 620 (1963).