ADRENAL IN PATIENTS WITH METASTATIC BREAST CANCER

Charles D. West, … , Barbara Damast, O. H. Pearson

J Clin Invest. 1958;37(3):341-349. https://doi.org/10.1172/JCI103614.

Research Article

Find the latest version: https://jci.me/103614/pdf ADRENAL ESTROGENS IN PATIENTS WITH METASTATIC BREAST CANCER 1 By CHARLES D. WEST,2 BARBARA DAMIAST, AND O. H. PEARSON (From the Division of Clinical Investigation, Sloan-Kettering Institute for Cancer Research, and the James Ewing Hospital, New York, N. Y.) (Submitted for publication June 11, 1957; accepted October 31, 1957)

Numerous studies in both patients and experi- tomy, and the administration of ACTH upon mental animals have established that estrogens their excretion. play an important role in the growth of some breast cancers. It is generally agreed that the ANALYTICAL METHODS beneficial effect of bilateral oophorectomy in some The analytical methods used in this investigation have patients with metastatic breast cancer is due to the been reported in detail elsewhere (8). Urine specimens elimination of the principal source of estrogens. were hydrolyzed with p-glucuronidase and extracted con- tinuously with ether for 48 hours. The phenolic frac- Inevitably, patients who obtain castration remis- tions were prepared from the ether extracts by partition- sions relapse. Many will respond favorably to ing between toluene and 1 N NaOH. Unless otherwise subsequent adrenalectomy or hypophysectomy specified, the phenolic fractions were routinely fraction- (1-3). As one possible explanation for these ob- ated as outlined in Figure 1. , and es- servations, it has been suggested that the adrenal triol were separated from the phenolic fractions by mul- tiple sequential fractionations by countercurrent distri- cortex might function as an extragonadal source bution and paper chromatography. of estrogens. Since certain adrenal cortical tu- The technique used for countercurrent distribution was mors cause feminization in men and since other developed by Craig and Craig (9) and adapted to the hypercortical states are frequently associated with analysis of estrogens by Engel (10). Countercurrent the excretion of excessive amounts of estrogens distribution served not only for fractionation, but also to provide information on the identity, purity and quan- (4), it is conceivable that normal adrenal cortical tity of the isolated estrogens. tissue might function as a significant source of The fundamental techniques used for paper chromatog- estrogens in patients with breast cancer. The raphy were developed by Bush (11) and adapted to the demonstration of biologically active estrogens in analysis of estrogens by Mitchell and Davies (12). The the urine of ovariectomized women by Dao (5) solvent systems, time of development and temperature for the paper chromatographies are indicated in Figure 1. and Smith and Emerson (6) supports this con- The estrogens on the paper chromatograms were visu- cept. That the pituitary gland might be involved alized by staining with an aqueous solution of 1 per cent in the production of adrenal estrogens is suggested ferric chloride and 1 per cent potassium ferricyanide as by the work of Nathanson, Engel, and Kelley described by Axelrod (13). (and in the discussion of Reference 7 by Paschkis) Estrone, estradiol and were measured by photo- fluorometry using Engel's modification (10) of the who reported increased levels of urinary estrogens method of Bates and Cohen (14). To obtain biological in patients given adrenocorticotrophic hormone confirmation of the chemical methods, the isolated estro- (ACTH). gens were assayed for estrogenic activity by either the The purpose of this study was to investigate the rat uterine weight or the rat vaginal cornification meth- ods (15, 16). possible role of adrenal estrogens in breast can- To support the identification of the free estrogens by cer by attempting to isolate and identify estro- paper chromatography and countercurrent distribution, gens excreted by castrated patients and to deter- the acetate derivatives were prepared and analyzed by mine the effects of adrenalectomy, hypophysec- paper chromatography against authentic acetates. Acetylation was carried out in acetic anhydride and py- 1 This project was generously supported in part by a ridine overnight at room temperature. grant from the United States Public Health Service [C-2241 (C3)]. RESULTS 2 Scholar of the American Cancer Society. Present Address: Veterans Administration Hospital, Salt Lake Urines from patients without ovaries were City, Utah. analyzed for estrogens by the methods described 341 342 CHARLES D. WEST, BARBARA DAMAST, AND 0. H. PEARSON

PRENOLIC FRACTION

ciwA. (]C+ 12) 20 PR (13) uR- CE3CE, 120, CC14 (50:50:100) CH3C0OC2I5,C61l2C2C5OH, 1E20 nX99 (50:50:33:67) n99

THEO. OSTNERI TREO. UTRADIOL TRIO. ETI(L

PAPE CHROM. PAPER CHROM. PAPE CHROK PIT. ETHER: CH3OH(50: 50) rruT. CH3cU(50:50) c66, mPT. ETHRm, CHY3O,H20 VOC 20 ME 320C, 20 HRS. (50:50:90:10); 320C e8 MBS.

Cc JNZWURREIM CWnWsERCUrNT CH3ICOOC5,C6112,C2H50E-2 CHCOOC2Hs5C6Hl12,C2H5sCE,20 CH3COOC2O5,C6312,C2OE,5120 (40:60:60:40); nu150 (40:60:60:40); nu200 (50:50:50:50); nw299 FIG. 1. OUTLINE OF METHOD OF FRACTIONATION FOR URINARY ESTROGENS El, estrone; E2, estradiol-17 ,; E,, estriol. Solvent ratios are by volume; n equals number of transfers. both during control periods and while the patients In order to stimulate the adrenal cortices of this were on ACTH, in an attempt to increase the ex- hypophysectomized patient to maximal secretory cretion of urinary estrogens by stimulating the activity, 50 mg. of ACTH was administered in- adrenal cortex. The results obtained in one pa- tramuscularly every six hours. After one week tient will be presented in detail to afford an op- of ACTH therapy the adrenal cortices were ac- portunity for critical evaluation of the methods tively secreting, as evidenced by a striking in- used. crease in urinary 17-ketosteroids and 17-hydroxy- This 66 year old patient (A.S.) had a radical corticoids and the development of sodium reten- mastectomy for carcinoma of the breast six years tion and hypokalemic alkalosis. At this time a prior to this study, at which time she was 12 years three day urine specimen was collected for estro- postmenopausal. She developed osseous meta- gen analyses. stases two years later and was treated with both The phenolic fractions were prepared from the estrogens and without improvement. urine samples and analyzed for estrone, estradiol In April, 1956, a hypophysectomy was performed and estriol as outlined in Figure 1. The results and the patient had a remission lasting six months. of the initial countercurrent distribution of the Upon relapse, the metastatic osseous disease was benzene soluble material in the phenolic fraction treated with X-ray and a radiation dose of 2,000 r are shown in Figure 2A. The distribution curve was delivered to both ovaries. The present estro- exhibited two nonhomogeneous peaks composed gen excretion study was done three months later of mixtures of fluorescent phenols. It was theo- during the time the patient's neoplastic disease was retically possible that estrone could have been advancing rapidly. one of the fluorescent phenols making up the first ADRENAL ESTROGENS IN CANCER 343

IdESTRONE *-* ANALYZED peak, and estradiol one of those in the more polar

3 - -- - THEORETICAL peak. To explore these possibilities further the tubes which would theoretically have contained estrone were combined and chromatographed on paper as 2 indicated in Figure 1. The fraction from the pa- E2 per chromatogram corresponding in mobility to K 2.1 reference standards of authentic estrone was eluted and analyzed by countercurrent distribution for a second time in a different solvent system. This 4 i time a single fluorescent compound was obtained, ,_1 and the distribution curve agreed with the theo- 21 retical curve for estrone (Figure 2B). The 24 hour excretion of estrone was calculated from this A countercurrent distribution curve to be 10 micro- 6 ESTRONE 0-* ANALYZED grams. - - - THEORETICAL An aliquot of this material was analyzed by paper chromatography for the second time. As can be seen by the photograph of the finished pa- per chromatogram (Figure 3), the isolated ma- terial had the same mobility as the two reference standards of authentic estrone on each side. The acetate derivative of the isolated material was prepared and analyzed simultaneously with reference standards of and free 0.1 estrone by paper chromatography in petroleum ether and methanol (1: 1) at 240 C. The chro- matographic behavior of the acetate derivative was identical with that of estrone acetate, providing additional evidence for the identification of the isolated material as estrone. B The tubes from the initial countercurrent distri- IYESTRADIOL E2 bution of the benzene fraction (Figure 2A), which 0.3 KO.58 _-* ANALYZED theoretically would have contained estradiol, were combined and analyzed by paper chromatography

0.2- and again by countercurrent distribution. The 4 _/ ~~~ final distribution curve did not exhibit a peak in for A more 0.1I the region expected estradiol. polar fluorescent phenolic compound was observed but

I unidentified (Figure 2C). @-- - =W, I- I I- - I I I I 0 25 50 75 1001200 150 100

TUBE NO. I TRANSFER NO. 50: 100); n equals 99; partition coefficient (K) for E, NOFUAMENTAL SERIES WITHDRAWAL-- SERIES equals 0.33; KE2 equals 2.1. C B. Countercurrent distribution for estrone following paper chromatography in CH3COOCH,, C6H12, C2H6OH, FIG. 2. COUNTERCURRENT DISTRIBUTION CURVES FOR Es- H20 (40: 60: 60: 40) ; n equals 150; KE, equals 0.96. TRONE (E1) AND ESTRADIOL (E2) ON PATIENT A.S. C. Countercurrent distribution for estradiol following

0, analyzed curves; - - - -, theoretical curves. paper chromatography in CH3COOC2H5, C6H12, CH5OH, A. Initial countercurrent distribution of the benzene H20 (40: 60: 60: 40); n equals 200; KE2 equals 0.58. fraction of the phenolic extract in CHSOH, H20, CC14 (50: Single withdrawal procedure used in B and C. 344 CHARLES D. WEST, BARBARA DAMAST, AND 0. H. PEARSON

:,

f. -4.,. -4

FIG. 3. PHOTOGRAPHS OF THE PAPER CHROMATOGRAM OBTAINED ON THE ESTRONE FRACTION FROM THE FINAL COUNTERCURRENT DISTRIBUTION, PATIENT A.S. The middle spot is the isolated estrone with two reference standards of authentic estrone on each side. Developed in methanol: petroleum ether (1: 1 v/v) for nineteen and one-half hours at 320 C. An attempt to identify estriol was successful. a single fluorescent compound was obtained whose The initial countercurrent distribution of the wa- distribution curve fit the theoretical distribution ter soluble material in the phenolic fraction re- curve for estriol exactly (Figure 4B). From vealed a mixture of fluorescent phenols which these countercurrent data it was calculated that could have contained estriol (Figure 4A). The 51 micrograms of estriol was excreted daily by tubes which theoretically would have contained the patient. any estriol present were combined, fractionated One aliquot of the isolated material was again further by paper chromatography, and analyzed chromatographed on paper and identified as es- again by countercurrent distribution. This time triol (Figure SA). A second aliquot was acetyl-

6ESTRIOL *--* ANALYZED I ESTRIOL - - - THEORETICAL 4 r * *-- ANALYZED 8.4 - - - THEORETICAL 3 F

2 [ I

2.0

III I I I I I 0 50 1001299 250 200 150 100 40 60 TUBE NO. TRANSFER NO. TUBE NO. --FUNDAMENTAL SERIES-_- -- WITHDRAWAL SERI A B FIG. 4. COUNTERCURRENT DISTRIBUTION CURVES FOR ESTRIOL (E3) ON PATIENT A.S. A. Initial distribution of aqueous fraction of the phenolic extract in CH3COOC2Ho, C6H,2, C2HOH, H20 (50: 50: 33: 67) ; n equals 99; KE, equals 1.2. B. Countercurrent distribution for estriol following paper chromatography in CHCOOC2H, C6H,2, C2HOH, H20 (50: 50: 50: 50); n equals 299; KE, equals 0.35. Single withdrawal procedure used.b - ADRENAL ESTROGENS IN CANCER 345

A *:;.- ......

FIG. 5. PHOTOGRAPHS OF PAPER CHROMATOGRAMS OBTAINED ON THE ESTRIOL FRACTION FROM THE FINAL COUNTER- CURRENT DISTRIBUTION, PATIENT A.S. A. Middle spot is the isolated estriol with estriol reference standards on each side, developed in petroleum ether, benzene, methanol, H2O (50: 50: 90: 10) for 48 hours at 32° C. B. Middle spot is the acetate derivative of the isolated estriol with reference standards of and estriol on each side. Developed in methanol, H2O, petroleum ether (80: 20: 100 v/v) for three hours at 32° C. ated and identified as estriol triacetate by paper this 66 year old patient (A.S.) were within nor- chromatography (Figure 5B). mal limits for an actively menstruating woman. A third aliquot was assayed quantitatively for Perhaps of significance is the fact that she ex- estrogenic activity by the rat uterine weight creted no measurable amounts of estradiol. Since method. Doses of 0.2, 0.4, and 1.0 microgram of the patient was 12 years postmenopausal, had un- isolated estriol, as determined fluorometrically, dergone X-ray castration and hypophysectomy, were injected into each of three castrated 21 day and excreted the estrogens under the influence of old female rats at each dose level, in six divided ACTH, it would appear highly probable that these doses on three successive days. An estriol stand- urinary estrogens or their precursors originated ard dose response curve was run simultaneously, from the adrenal cortex. Nevertheless, the ovaries as well as controls. The isolated material gave a cannot be absolutely ruled out in this patient as positive bioassay equivalent to 41 micrograms of the site of origin, since they were still in situ estriol excreted daily, in comparison with a value and X-ray castration and hypophysectomy may of 51 micrograms per 24 hours which was calcu- not have destroyed their hormonal function. lated from the final countercurrent distribution Similar estrogen excretion studies were carried data, thus providing biological confirmation for out in two other patients with metastatic breast the chemical identification. cancer who had been surgically castrated (M.H. Under the influence of ACTH administration and E.Z.) and one other patient who had been the amounts of estrone and estriol excreted by castrated and hypophysectomized (M.G.), both 346 CHARLES D. WEST, BARBARA DAMAST, AND 0. H. PEARSON

TABLE I Estrogen excretion by castrated patients with metastatic breast cancer and the effect of adrenocorticotrophic hormone (A CTH) *

Control ACTH treatment El E2 Ea El Es Es Pt. Age Endocrine status (mg./24 hrs.) (;sg./24 hrs.) (;&g./24 hrs.) (ug./24 hrs.) (ug./24 hrs.) (ig./24 his.) M. H. 62 Bilat. oophorect. 2 0 1 19 0 38 (25)t E. Z. 42 Bilat. oophorect. 0 0 0 0 0 6 (3) A. S. 66 Spontaneous 10 0 51 menopause (41) X-ray cast. Hypophysectomy M. G. 29 Bilat. oophorect. 0 0 0 0 0 5 Hypophysectomy (6) M. M. 43 Bilat. oophorect. 0 0 0 L. S. 48 Bilat. oophorect. 0 0 (35) C. M. 35 Bilat. oophorect. 0 0 (17)

* El, estrone; E2, estradiol; E3, estriol. t Estrogen values in parentheses determined by bioassay and the other values by photofluorometry. before and while on ACTH therapy. The results the administration of ACTH she excreted 6 mi- are summarized in Table I. crograms of estriol daily. Neither estrone nor One castrated patient (M.H.) excreted 2 micro- estradiol were found in the urine during the grams of estrone and 1 microgram of estriol daily ACTH period. during the control period. The estrone was iso- No identifiable estrogens could be detected in lated as a pure fluorescent compound by a combi- the control urine of patient M.G. Who had under- nation of countercurrent distribution and paper gone bilateral oophorectomy and hypophysectomy. chromatography and measured quantitatively by Upon the administration of ACTH, she excreted 5 fluorometry as previously described. The isolated micrograms of estriol daily, but no estrone or estrone gave a positive biological assay for estro- estradiol. The results obtained on patient A.S. gens by causing complete cornification of the vagi- which were presented above in detail are included nal epithelium in 15 out of 15 castrated adult fe- in Table I. male rats in doses of 0.035 microgram per rat None of the four castrated patients (M.H., administered intravaginally. Appropriate con- E.Z., A.S. and M.G.) excreted detectable amounts trols and standards were run simultaneously. The of estradiol either during the control period or on estriol was also isolated in a homogeneous fluores- ACTH. However, all of them excreted a fluores- cent form and measured by fluorometry. It gave cent phenol which did not separate from estradiol a positive test for estrogenic activity by the vaginal on paper chromatography but did on subsequent cornification method in doses of 0.03 microgram countercurrent distribution. This compound is per rat. slightly more polar than estradiol and is so far With the administration of ACTH the excre- unidentified. tion of estrone by patient M.H. increased from 2 to Using a less rigorous method estrogen excre- 19 micrograms per 24 hours; and estriol, from tion studies were carried out in three additional 1 to 38 micrograms per 24 hours. ovariectomized patients with metastatic breast None of the three classical estrogens could be cancer (M.M., L.S. and C.M.). With this identified in the urine of the other castrated pa- method estrone, estradiol and estriol were sepa- tient (E.Z.) during the control period, but upon rated from each other by a single countercurrent ADRENAL ESTROGENS IN CANCER 347

1 distribution of the phenolic fraction using 100 ESTRIOL - ANALYZED transfers in a solvent system made up of methanol, 0.71 ---- THEORETICAL water, CHC13 and CCI4 (70: 30: 40: 60). If the distribution curve, as determined fluorometrically, was consistent with the presence of any of the 0.5- three estrogens, the total amount of the individual estrogen was measured by assaying the appropri- ate peak tubes for estrogenic activity by the rat uterine weight method. Although estrone, es- 0.3- tradiol and estriol can be separated from each other by the single countercurrent distribution used, they cannot be isolated in a fluorogenically 0.I - homogeneous form and accurate quantitative meas- dh 9 -EIf - 3 urement by fluorometry is impossible. In addi- 0 50 1001299 250 200 150 100 tion the three classical estrogens may not neces- TUBE NO. TRANSFER NO. -FUNOAMENTAL SERIES- t4 WITH-AWAL SERIES * sarily separate from other unknown estrogenic materials in the urinary extracts which would af- FIG. 6. COUNTERCURRE-NT DISTRIBUTION CURVE OB- fect the results obtained by bioassay. Neverthe- TAINED FOR ESTRIOL FOLLOWING AN INITIAL DISTRIBU- TION AND PAPER CHROMATOGRAPHY IN A CASTRATED, less, the results obtained by this less rigorous tech- ADRENALECTOMIZED, HYPOPHYSECTOMIZED PATIENT nique in the three castrated patients (C.M., M.M. * *, analyzed curve; -, theoretical curve for and L.S.) are included in Table I to demonstrate estriol; KE, equals 0.35; n equals 299; solvent system, further that patients without ovaries can excrete CH3COOC2H5, C6H12, C2HOH, H20 (50: 50: 50: 50). estrogens. Single withdrawal procedure used. None of the three patients excreted estrone or estradiol in amounts which were detectable by duction of these urinary estrogens or their pre- countercurrent distribution. However, the distri- cursors. If this interpretation is correct, patients bution curves were compatible with the excre- without both ovaries and adrenals should not ex- tion of estriol in two of the patients (C.M. and crete estrogens. L.S.). Therefore, the single peak tubes which To test this theory. urines from two castrated, theoretically contained the maximum amount of adrenalectomized patients and two other patients estriol were assayed by the rat uterine weight who had been surgically castrated, adrenalecto- method. In this way it was determined that pa- mized and hypophysectomized (for a total of tient C.M. excreted 17 microgram equivalents of four patients without either ovaries or adrenals) estriol in 24 hours; and patient L.S., 35 micro- were analyzed for estrogens by the complete pro- grams in 24 hours. The comparatively high cedure outlined in Figure 1. None of these four values for estriol in these two patients could have patients excreted estrone or estradiol, and estriol been due to contaminating estrogenic materials could not be demonstrated in the urine of three of which were not separated from estriol by the these patients. However good evidence for the single countercurrent distribution. excretion of estriol by the fourth patient, who had No estrogens were demonstrated in the urine undergone bilateral oophorectomy, adrenalectomy, of the third castrated patient (M.M.) by this and hypophysectomy, was obtained. By the pro- method. It may have been significant that this cedure outlined in Figure 1, a homogeneous fluor- patient differed from all the rest of the castrated escent compound was isolated whose distribution patients in that her neoplastic disease was in re- curve coincided with the theoretical curve for mission at the time the estrogen analyses were estriol (Figure 6). The amount of estriol ex- done. creted daily was calculated from the final counter- The demonstration of estrone and estriol in the current distribution curve to be 11 micrograms. urine of castrated patients and the increased ex- The acetate derivative of this material was pre- cretion upon ACTH administration strongly sug- pared and identified as estriol triacetate by paper gest that the adrenal cortex is involved in the pro- chromatography. In addition, it was shown that 348 CHARLES D. WEST, BARBARA DAMAST, AND 0. H. PEARSON the isolated material had estrogenic biological throughout the study, it would appear that physio- activity. logical levels of cortisone are not metabolized to estrogens to any great extent. To examine the DISCUSSION possibility of this conversion more rigorously 600 mg. of cortisol was given intravenously to a In this investigation it was demonstrated that castrated, adrenalectomized, hypophysectomized castrated with metastatic breast cancer patients patient. and the urine analyzed for estrogens. may excrete significant of amounts estrone and Estrone. estradiol and estriol could not be demon- the estriol in their urine, but not estradiol. With strated in the urine. These observations suggest additional observation that the administration of that corticosteroids are not converted to estrogens ACTH increased excretion, it seems highly prob- in any significant amounts. able that these or their urinary estrogens pre- It is of interest that estriol was found more con- cursors were produced by the adrenal cortex. sistently in the urine of castrated patients than any The observation that patients without both ovaries other estrogen. Estriol is usually considered to and adrenals usually failed to excrete these estro- be an estrogen metabolite rather than a secretory gens lends support to this conclusion. product. It cannot be ascertained from this study Whether the isolated urinary estrogens were whether the estriol isolated from urine was a se- actually secreted by the adrenal cortex or whether cretory product of the adrenal cortex or a meta- they represent metabolic products of some other bolic end product of some other precursor. adrenal precursor could not be ascertained from The demonstration that the adrenal cortex in this study. There is indirect evidence to sup- patients with metastaic breast cancer can be a port both of these possibilities. With the isolation source of significant amounts of estrogens is con- of estrone from adrenal cortical tissue by Beall sistent with the theory that relapses following (17) and the demonstration by Meyer (18) that castration remissions are due to adrenal estrogens 19-hydroxy-A4 androstene-3, 17-dione can be con- and that subsequent remissions following adrenal- verted to estrone by adrenal tissue, it would ap- ectomy are due to the removal of this source of pear that adrenal cortical tissue has the necessary estrogens, but does not prove it. Proof of this enzyme systems for synthesizing estrogens and theory will depend upon whether the growth of that secretion of estrogens by the adrenal gland is the metastatic cancer in these patients can be cor- a real possibility. On the other hand, it has been related with the level of urinary estrogens during demonstrated that certain C19 neutral of relapses and remissions. the androgenic series, some of which are pro- The increased excretion of urinary estrogens by duced by the adrenal cortex, are converted to es- castrated patients upon the administration of trogens both in vivo and in vitro by a variety of ACTH and the absence of estrogens in the urine tissues ( 18-21 ). Under the conditions of the of castrated, hypophysectomized patients suggest present experiments the adrenal cortex could that endogenous ACTH might be involved in the have secreted these C1, neutral steroids or other control of estrogen production by the adrenal unknown precursors which were metabolized in cortex. If this is the case, one of the possible part to estrogens elsewhere in the body and ex- mechanisms by which hypophysectomy could cause creted in the urine. remission in castrated patients is through sec- The possibility that corticosteroids, such as ondary suppression of estrogen production by the cortisol and cortisone, might be converted in part adrenal gland. to estrogens was considered since this conversion In view of the fact that one of the castrated, would play an important role in the management adrenalectomized, hypophysectomized patients ex- of adrenalectomized and hypophysectomized pa- creted appreciable amounts of estriol, other pos- tients with estrogen-dependent breast cancer. sible sources of estrogens besides the ovaries and Since the adrenalectomized and hypophysecto- adrenal glands, such as the diet or accessory ovar- mized patients in this study failed to excrete meas- ian or adrenal tissue, must be considered to ex- urable amounts of estrogens even though they plain relapses occurring in some patients with were receiving maintenance doses of cortisone estrogen-dependent tumors. ADRENAL ESTROGENS IN CANCER 349

SUMMARY treatment. Proc. Soc. exp. Biol. (N. Y.) 1954, 85, 264. Significant amounts of estrone and estriol have 7. Nathanson, I. T., Engel, L. L., and Kelley, R. M. been demonstrated in the urine of castrated pa- The effect of ACTH on the urinary excretion of tients with metastatic breast cancer by means of steroids in neoplastic disease in Proc. Second Clinical ACTH Conf. New York, The Blakiston countercurrent paper distribution, chromatog- Co., 1951, vol. I, p. 54. raphy, and bioassay methods. The excretion of 8. West, C. D., Damast, B., and Pearson, 0. H. Uri- these estrogens by castrated patients increased nary estrogen excretion in Cushing's syndrome. strikingly upon the administration of ACTH. J. clin. Endocr. 1958, 18, 28. With one exception, castrated and adrenalecto- 9. Craig, L. C., and Craig, D. Extraction and distribu- tion in of A. mized patients failed to excrete identifiable Technique Organic Chemistry, Weissberger, Ed. New York, Interscience Pub- amounts of estrogens. These observations sug- lishers, Inc., 1950, vol. 3, p. 171. gest that estrogens or their precursors are pro- 10. Engel, L. L. The chemical estimation of duced by the adrenal cortex and that endogenous hormone metabolites in Recent Progr. Hormone ACTH might be involved in the control of this Res. New York, Academic Press Inc., 1950, vol. 5, p. 335. estrogen production. Estriol was demonstrated 11. Bush, I. E. Methods of paper chromatography of in the urine of one patient who had been castrated, steroids applicable to the study of steroids in mam- adrenalectomized, and hypophysectomized, rais- malian blood and tissues. Biochem. J. 1952, 50, ing the possibility of additional sources of estro- 370. gens besides the ovaries and adrenals, such as 12. Mitchell, F. L., and Davies, R. E. The isolation and estimation of the steroid oestrogens in placental diet and accessory ovarian or adrenal tissue. tissue. Biochem. J. 1954, 56, 690. The possible pathophysiological significance of 13. Axelrod, L. R. The chromatographic fractionation these findings in breast cancer was discussed. and identification of compounds related to the estrogens in Recent Progr. Hormone Res. New ACKNOWLEDGM ENT York, Academic Press Inc., 1954, vol. 9, p. 69. 14. Bates, R. W., and Cohen, H. Experimental basis for The authors acknowledge with gratitude the help and selecting the optimal conditions for fluorometry of advice given by Drs. T. F. Gallagher, W. 0. Money, natural estrogens. Endocrinology 1950, 47, 166. M. D. Lipsett and Rulon W. Rawson; and the technical 15. Lauson, H. D., Heller, C. G., Golden, J. B., and assistance of Mrs. Louise Berman, Mr. Theodore Hall Sevringhaus, E. L. The immature rat uterus in and Miss Dorothy Sellitto. the assay of , and a com- parison of estradiol, estrone, and estriol. Endo- REFERENCES crinology 1939, 24, 35. 16. Emmens, C. S. Precursors of oestrogens. J. En- 1. Huggins, C., and Bergenstal, D. M. Influence of bi- docr. 1941, 2, 444. lateral adrenalectomy, adrenocorticotrophin, and 17. Beall, D. The isolation of oestrone from ox adre- cortisone acetate on certain human tumors. Science nals. J. Endocr. 1940, 2, 81. 1951, 114, 482. 18. Meyer, A. S. Conversion of 19-hydroxy-A4-andro- 2. Luft, R., and Olivercrona, H. Experiences with hy- stene-3,17-dione to estrone by endocrine tissue. pophysectomy in man. J. Neurosurg. 1953, 10, 301. Biochim. biophys. Acta 1955, 17, 441. 3. Pearson, 0. H., Ray, B. S., Harrold, C. C., West, 19. Heard, R. D. H., Jellinck, P. H., and O'Donnell, V. J. C. D., Li, M. C., Maclean, J. P., and Lipsett, M. B. Biogenesis of estrogens: The conversion of testos- Hypophysectomy in the treatment of advanced terone-4-C1' to estrone in the pregnant mare. cancer. Trans. Ass. Amer. Phycns 1955, 68, 101. Endocrinology 1955, 57, 200. 4. Soffer, L. J. Diseases of the Adrenal, 2nd ed. 20. West, C. D., Damast, B. L., and Pearson, 0. H. The Philadelphia, Lea and Febiger, 1948, pp. 249-252. conversion of to estrogens in cas- 5. Dao, T. L. Y. Estrogen excretion in women with trated, adrenalectomized human females. J. biol. mammary cancer before and after adrenalectomy. Chem. 1956, 218, 409. Science 1953, 118, 21. 21. Baggett, B., Engel, L. L., Savard, K., and Dorfman, 6. Smith, 0. W., and Emerson, K., Jr. Urinary estro- R. I. The conversion of testosterone-3-C'4 to C"4 gens and related compounds in postmenopausal estradiol-17,6 by human ovarian tissue. J. biol. women with mammary cancer: Effect of cortisone Chem. 1956, 221, 931.