Biologic effects of equilin sulfate in postmenopausal women

Transcription

1 FERTILITY AND STERILITY Copyright The American Fertility Society Vol. 49, No. 2, February 1988 Printed in U.S.A. Biologic effects of equilin sulfate in postmenopausal women Rogerio A. Lobo, M.D.*t Hoa N. Nguyen, M.D.* Peter Eggena, Ph.D.:j: Paul F. Brenner, M.D.* University of Southern California School of Medicine, Los Angeles County/University of Southern California Medical Center, Women's Hospital, Los Angeles, and Veterans Administration Medical Center, Sepulveda, California In order to determine the relative potency of equilin sulfate (EqS), a major constituent of conjugated equine estrogens, 15 women received oral doses ofeqs (0.15, 0.31, and mg) for 25 days. Doses of 0.31 and mg significantly stimulated hepatic globulins. This stimulatory effect ranged from being 1.5 to 8 times greater than the effects of comparable doses of estrone sulfate and conjugated equine estrogen~. A significant stimulation in high-density lipoprotein-cholesterol occurred with as little as 0.15 mg of EqS. Elevations in the high-density lipoprotein/low-density lipoprotein-cholesterol ratio occurred with EqS, which resulted in an approximately 4-fold greater response than that achieved with comparable doses of conjugated equine estrogens. The fasting urinary calcium/creatinine ratio was only significantly lowered with mg of EqS and was less potent than conjugated equine estrogens in this regard. It is concluded that EqS is a potent estrogen that contributes significantly to the hepatic stimulatory effects of conjugated equine estrogens. These data also provide support for the suggestion that there may be a dissociation in potency between estrogenic effects on liver and bone. Fertil Steril 49:234, 1988 Equilin sulfate (EqS) is an equine estrogen that constitutes approximately 25% of conjugated equine estrogens (CEE). Although equine estrogens are known to be biologically active, 1 little is known of their clinical effects. EqS is known to be metabolized to equilin and 17~-dihydroequilin, and both have great affinity for the endometrial estrogen receptor. In particular, 17~-dihydroequilin has been thought to have a 15-fold greater affinity for the estrogen receptor compared with native 17~- Received August 3, 1987; revised and accepted October 21, *Department of Obstetrics and Gynecology, University of Southern California School of Medicine, Los Angeles County I University of Southern California Medical Center, and Women's Hospital, Los Angeles, California. t Reprint requests: Rogerio A. Lobo, M.D., Women's Hospital, Room 1M2, 1240 North Mission Road, Los Angeles, California :j: Veterans Administration Medical Center. 234 Lobo et al. Biologic effects of equilin sulfate estradiol (E 2). 2 It has been suggested, therefore, that these components in CEE impart a greater stimulatory effect on the endometrium and these observations have been drawn upon to explain the excess in cases of endometrial cancer associated with unopposed estrogen therapy with CEE. 3 4 Oral estrogens undergo rapid splanchnic metabolism that may result in a relative hepatic toxicity because of the effect of its "first passage." 5 We have observed previously that oral CEE therapy results in greater stimulation of hepatic globulins compared with native estrogens like estrone (E 1) and E 2 yet significantly less than that which occurs with synthetic estrogens such as ethinyl estradiol (EE 2). 6 This hepatic stimulation, while potentially harmful, has been viewed as beneficial, given the increases observed in levels of high-density lipoprotein (HDL)-cholesterol (C). 7 In order to more fully understand the biologic effects of CEE therapy, we compared the effects of EqS with E 1-sulfate (E 1-S) and CEE, which contains both EqS and

2 E1-S. We measured changes in serum hepatic globulins and lipoproteins as well as changes in urinary calcium excretion with various doses of these estrogens. ~ 100 Subjects MATERIALS AND METHODS Fifteen women, aged 34 to 60 (mean, 47.2 years), were studied 4 weeks after hysterectomy and bilateral salpingo-oophorectomy for benign disease. Each woman was within 20% of ideal body weight. Protocol EqS was purchased from Research Plus, Inc. (Bayonne, NJ) and formulated with dextrose in gelatin capsules by the School of Pharmacy at the University of Southern California (Los Angeles, CA). Capsules containing 0.156, 0.312, and mg of EqS were formulated to represent the constituents of EqS contained in frequently prescribed doses of CEE (0.625, 1.25, and 2.5 mg). Five women were ascribed to each of these three doses and EqS was ingested at the hour of sleep for 25 days. Prior to treatment, fasting blood was obtained, as was urine for assessment of the calcium/creatinine ratio. These measurements were repeated at 2 weeks and on the last day of treatment. From fasting blood, serum was obtained and analyzed for total C, HDL-C, and low density lipoproteins (LDL-C), as well as sex hormone-binding globulin (SHBG) binding capacity, corticosteroid-binding globulin (CBG) binding capacity, and angiotensinogen, as previously described In addition, we measured the high-molecular weight moiety of angiotensinogen, which migrates with an rf of 0.16 on polyacrilamide gel electrophoresisy Urine was analyzed for calcium and creatinine, as previously described All samples from the same patient were included in the same assay. In order to compare the effects of EqS with those of E1-S and CEE, we used our previously published data, which was obtained using the same methodology and study protocols.6 16 Specifically, the effects of various estrogens on hepatic globulins were assessed after 2 weeks,4 and the effects on lipids and calcium excretion after 25 days.16 Data were analyzed using paired "t" analyses with and without log transformation, and dose-response curve relationships were compared, as previously described.6 The linear response curve to the Vol. 49, No.2, February !. ~ :: n... II ' " ~- o :.::~~ O.lS,.,..., Ul Figure 1 Effect of different doses of EqS on SHBG. Asterisks denote significant changes occurring with 0.31 mg (P < 0.05) and mg (P < 0.01). Effect of different doses of EqS on CBG. The asterisk denotes the only significant change (P < 0.05) occurring with mg. Effect of different doses ofeqs on angiotensinogen. Asterisks denote significant changes from baseline (P < 0.05) for EqS 0.31 mg and (P < 0.01) for EqS mg. log dose for each of the estrogen preparations was calculated by the least squares regression method. For each of the parameters, relative estrogen potencies were estimated by normalizing parallel responses of the three estrogen preparations studied. RESULTS All patients tolerated the doses of EqS without side effects. One patient receiving the mg dose did not return after beginning treatment. Subjective relief of vasomotor symptoms was achieved in all patients, but was not quantified because of the nature of this study. Although paired data from each patient were used for statistical analyses, in all subsequent figures, baseline data comprised the mean values for all 15 women. All baseline data were similar for each parameter in the three groups. A dose-response increase in SHBG occurred with EqS treatment. Changes were significant with and mg doses (Fig. 1). SHBG levels in the three groups were 63.7 ± 8 before and 80 ± 14 nm with 0.15 mg, 65.8 ± 14 before and 95 ± 15 nm with 0.31 mg, and 72.9 ± 24 before and 142 ± 11 nm with mg of EqS. Similar responses occurred with CBG and angiotensinogen. In the case of CBG, only the mg dose resulted in a significant change (P < 0.05). CBG values in l!g/dl were 19.9 ± 1 before and 15.8 ± 2 after with 0.15 mg, 18.6 ± 0.7 and 18.4 ± 1.2 with 0.31 mg, and 18.8 ± 1.8 before and 24.5 ± 2 with mg of EqS. Lobo et al. Biologic effects of equilin sulfate 235

3 .. I. ~ E u... ' 0 I :] 0 I : 0.14 ~ 0.10 a u 0.06 NS pc0.05 ~=1n l I Jf L...J 0 pc0.05 Figure 2 Effect of different doses of EqS on HDL-C (upper panel) and the HDL/LDL-C ratio (lower panel). Asterisks denote significant changes (P < 0.05) from baseline. Angiotensinogen levels were 2185 ± 161, 2589 ± 298, and 2673 ± 483 ng/ml before EqS and increased to 2702 ± 478, 3613 ± 404, and 4608 ± 773 ng/ml with 0.15, 0.31, and mg ofeqs, respectively. Changes with 0.31 mg and mg of EqS were statistically significant (P < 0.05 and p < 0.01). No appreciable changes occurred in high-molecular weight angiotensinogen until doses of mg of EqS were ingested. Although not all samples were measured, all values for high-molecular weight angiotensinogen were undetectable, except in two patients after they had received mg of EqS. The percentage of the high-molecular weight moiety increased in these two patients from an undetectable level to 4.2 and 5.2%, respectively. These values must be viewed as representing a significant increase in that changes which occurred with 1.2 mg of CEE in a previous study were <2%.13 Total C decreased significantly (P < 0.05) in all three groups: 180 ± 20 to 174 ± 15 mg/dl with EqS 0.156, 238 ± 20 to 222 ± 17 mg/dl with EqS 0.31, and 212 ± 7.8 to 206 ± 6.9 mg/dl with EqS As with other parameters, baseline levels were not statistically different between groups. Changes in LDL-C were similar, decreasing from 120 ± 10 to 101 ± 12 mg/dl, 152 ± 6 to 143 ± 15 mg/dl, and 109 ± 7 to 102 ± 2 mg/dl in the three groups. Each dose of EqS resulted in significant (P < 0.05) increases in HDL-C and in the HDL/LDL-C ratio (Fig. 2). The urinary calcium/creatinine ratios were similar in the three groups and were not significantly altered by the lower doses of EqS (0.156 and 0.31 mg). A significant decrease occurred with mg (P < 0.05) (Fig. 3) Figure 3 Effect of different doses of EqS on fasting Ca 2 + /Cr ratios. The asterisk denotes the only dose (0.625 mg) that achieved significance (P < 0.05). Dose-response curves were constructed from these data and from data previously obtained for the effects of E1-S and CEE on hepatic globulins. Figure 4 illustrates the parallel dose responses occurring in SHBG. Data for E1-S and CEE were obtained from our previous work. 6 The increases in HDL-C occurring with treatment were compared. The increase associated with mg of EqS in this study was similar to changes obtained with mg doses of CEE and E1-S.16 This suggests a 4-fold potency difference in this parameter between EqS and both CEE and E1-S (Fig. 5). It required mg of EqS to significantly reduce the calcium/creatinine ratio. This significantly reduced level was achieved by 0.3 mg of CEE.16 This suggests that, for the calcium/creati- :E c \.:) <0 J: "' <: ESTROGEN, mg Figure 4 Dose responses of EqS (closed circles), CEE (opened circles), and E1-S (triangles). SHBG, which represents the increases achieved (~SHBG) with each dose Lobo et al. Biologic effects of equilin sulfate

4 ~ 10 ao E 8 y :r:: <I 2 ~ J ~ 0.06 o u EqS CEE E1S EqS CEE E 1S Figure 5 Upper panel: The increase in HDL-C with EqS 0.15 mg.and the doses of CEE (0.625 mg) and E1-S (0.6 mg) that achieved similar changes. Lower panel The Ca 2 /Cr ratio that occurred with EqS mg, and the doses of C:EE (0.3 mg) and E1-S (0.625 mg) achieved similar ratios. nine ratio; CEE is twice as potent as EqS (Fig. 5). Although mg E 1-S also reduced this ratio by a similar magnitude, we had not studied lower doses of E 1-S 16 (e.g., 0.3 mg) and therefore cannot give accurate potency ratios. Table 1 summarizes estimates on the potency differences between EqS, CEE, and E 1-S drawn from our data. EqS is considered more potent on a weight basis than C:EE and E 1-S for ali hepatic globulins measured. EqS appears to be one-half as potent as CEE in reducing the calcium/creatinine ratio and at least equipotent to the effects of E 1-S. Accurate equivalency for lipoprotein determinations is not possible, but the data suggest that EqS is more potent on a weight basis than either CEE or E1-S. DISCUSSION Our data provide evidence that EqS is a potent estrogen. A dose of 0.31 mg of EqS significantly increased SHBG and angiotensinogen levels. This quantity of EqS, which is the amount contained in 1.25 mg of CEE, explains in part the stimuiatory effects of CEE ori hepatic globulins. An increase in the high-molecular weight moiety of angiotensinogen has been correlated with the occurrence of hypertension 13 and was a useful parameter for us to evaluate. No changes occurred with EqS until patiemts received mg. This dose would be equivalent to that of patients receiving 2.5 mg of CEE, but was not evaluated separately by us. Although Vol. 49, No. 2, February 1988 increases in this moiety of angiotensinogen were observed with mg of EqS, no increase in blood pressure was observed. These preliminary data reinforce the concept that changes in hepatic globulins do not necessarily correlate with clinical findings, such as the development of hypertension. Overall, our data on hepatic globulins suggest that EqS is 1.5 to 8 times more potent than the other estrogens studied. The data on lipoproteins is not as clear-cut because of our study design. Here we have attempted to show the sensitivity of HDL-C to short-term oral estrogenic stimulation. Nevertheless, it requires up to 3 months to appreciate the full range of estrogenic effects on lipoprotein metabolism. In our original study, 16 we were unable to demonstrate significant changes in HDL-C with mg of CEE over 25 days. Nevertheless, we have shown that this dose significantly increases HDL-C in a study of 1 year's durationp That as little as 0.15 mg EqS for 25 days was able to raise HDL-C in this small sample size suggests that EqS is more potent than CEE in raising this lipoprotein by a factor of at least 4. There is limited data on the effects of lower doses of E 1-S on lipoproteins. 18 However, we might predict that, because ofthe additional stimulatory effect afforded by the EqS constituent of CEE, E 1 -S is less potent than CEE in altering lipoproteins. Although EqS appears to be a potent estrogen; its effect of inhibiting bone resorption may be less significant. As suggested by the urinary calcium/ creatinine ratio data, the only dose that significantly reduced bone resorption was mg of EqS. This may be compared with the significant reductions with 0.3 mg of CEE observed by us and others E 1-S mg also significantly reduced calcium excretion, although doses lower than this were not studied by usp This suggests that, on a weight basis, EqS may be equipotent to E 1-S in altering the calcium/creatinine ratio. Table 1 Relative Potency of EqS, CEE, and E1-S" Angiotensinogen SHBG CBG HDL/LDL Ca 2 tcr (4) ND (1). Changes in the HDL/i..DL ratio represent an approximation for the comparison between EqS and CEE (in parentheses) and was not done (ND) in the case of E1-S. The change in the Ca 2 I Cr ratio with E1-S also represents an approximation. Lobo et al. Biologic effects of equilin sulfate 237

5 It is well appreciated that the calcium/creatinine ratio is a crude marker of bone resorption. In that it has been shown that mg of CEE is required to inhibit bone resorption.in all women by bone ~easurement techniques, 20 the calcium/creatinine ratio may be significantly lowered with less estrogen (e.g., 0.3 mg of CEE). From our dat~. we could therefore hypothesize that there is potentially a dissociation in the effects of oral EqSbetween liver and bone. An analogy may be dr!lwn here to similar effects ob~erved with a E!ynthetic estrogen, EE2 While as little as 5 ILg of EE 2 significantly stimulated SHBG, 21 it requires at least 10 ILg to alter the calcium/creatinine ratio. Thus, this di.ssociated effect observed by us for EqS may not be related to the type of estrogen studied. REFERENCES 1. Howard RP, Keaty EC: Evaluation of equilin-3-mono-sulphate and other estrogens. Arch Intern Med 128:229, Bhavnimi BR, Woolever CA, Pan CC: Interaction of equine estrogens with estrogen receptors in human endometrium and rat uterus (Abstr 22). Presented at the 33rd annual meeting of the Society for Gynecologic Investigation, Toronto, Ontarip, Canada, March 19 to 22, 1986, p Mack TM, Pike MC, Henderson BE, Pfeffer RI, Gerkins VR, Arthur M, Brown SE: Estrogens and endometrial cancer in a retirement community. N Engl J Med 294:1262, Jick H, Watkins RN, Hunter JR, Dinan BJ, Madsen S, Rothman KJ, Walker AM: Replacement estrogens and endometrial cancer. N Engl J Med 300:218, Siddle N, Whitehead M: Flexible prescribing of estrogens. Cop.temp Obstet Gynecol 22:137, Mashchak CA, Lobo RA, Dozono-Takano R, Eggeria P, Nakamura RM, Brenner PF, Mishell DR Jr: Comparison of pharmacodynamic properties of various estrogen formulations. Am J Obstet Gynecol144:511, Bush TL, Barrett Coriner E: Noncontraceptive estrogens and cardiovascular disease. Epidemiol Rev 7:80, Lopes-Virella J\1F, Stone P, Ellis S, Colwell JA: Choiesterol determination in high-density lipoproteins separated by three different methods. Clin Chern 23:882, Friedewald WT, Levy RI, Fredrickson DS: Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chern 18:499, Rosner W: A simplified method for the quantitative determination of testosterone-estradiol-binding globulin activity in humim plasma. J Clin Endocrinol Metab 34:983, Moore DE, Kawagoe S, Davajan V, Mishell DR Jr, Nakamura RM: An in vivo system in man for quantitation of estrogenicity. I. Physiologic changes in binding capacity of serum corticosteroid-binding globulin. Am J Obstet Gynecol 130:475, Eggena P, Barrett JD, Sambhi MP, Wiedemal;i. CE: The validity of comparing the measurements of angiotensin. I. Generated in human plasma by radioimmunoassay and bioassay. J Clin Endocrinol Metab 39:865, Shionoiri H, Eggena P, Barrett JD, Thananopavarn C, Golub MS, Eggena Z, Nakamura R, Judd HL, Sambhi Mp: An increase in. high-molecular weight renin substrate associated with estrogenic hypertension. Biochem Med 29:14, Nordin BEC, Gallagher jc, Aaron JE, Horsman A: Postmenopausal osteopenia and osteoporosis. In Estrogens in the Postmenopause, Edited by P A van Keep, C Lauritzen. Basel, S Karger, Front Hormone Res 3:133, Frumar AM, Meldrum DR, Geola F, Shamohki IM, Tataryn IV, Deftos LJ, Judd HL: Relationship of fasting urinary calcium to circulating estrogen and body weight in postmenopausal women. J Clin Endocrinol Metab 50:70, Lobo RA, Brenner P, Mishell DR Jr: Metabolic parameters and steroid levels in postmenopausal women receiving lower doses of natural estrogen replacement. Obstet Gynecol 62:94, Barnes RB, Roy S, Lobo RA: Comparison of iipid and androgen levels after conjugated estrogen or depomedroxyprogesterone acetate treatment in postmenopausal women. Obstet Gynecol 66:216, Bush TL, Miller VT: Effects of pharmacologic agents used during menopause: impact on lipids and lipoproteins. In Menopause: Physiology and Pharmacology, Edited by DR Mishell Jr. Chicago, Year Book Medical Publishers, 1987, p Geola FL, Frumar AM, Tataryn IV, Lu JKH, Hershman JM, Eggena P, Sambhi MP, Judd HL: Biological effects of various doses of conjugated equine estrogens in postmenopausal women. J Clin Endocrinol Metab 51:620, Lindsay R, Hart DM, Clark DM: The minimum effective dose of estrogen for prevention of postmenopausal bone loss. Obstet Gynecol 63:759, Mandel FP, Geola FL, Lu JKH, Eggena P, Sambhi MP, Hershman JM, Judd HL: Biologic effects of various doses of ethinyl estradiol in postmenopausal women. Obstet Gynecol 59:673, Lobo et al. Biologic effects of equilin sulfate