| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Chair of Obstetrics and Gynecology (S.P., M.M., T.R., F.Z.), University of Catanzaro, 88100 Catanzaro; Department of Molecular and Clinical Endocrinology and Oncology (F.O., G.L.), University of Naples "Federico II", 80131 Naples; Department of Gynecology, Obstetrics, and Human Reproduction (M.P., C.N.), University of Naples "Federico II", 80131 Naples; Chair of Obstetrics and Gynecology (E.Z.), University "Tor Vergata" of Rome, 72100 Rome; and Chair of Obstetrics and Gynecology (P.L.B.P.), University of Rome "Campus Biomedico", 72100 Rome, Italy
Address all correspondence and requests for reprints to: Dr. Stefano Palomba, Via Nicolardi 188, 80131 Napoli, Italy. E-mail: . stefanopalomba{at}tin.it
Abstract
This prospective randomized single-blind placebo-controlled clinical trial was carried out to evaluate the effects of raloxifene administration on uterine and leiomyoma sizes in premenopausal women. Ninety healthy premenopausal women affected by asymptomatic uterine leiomyomas were enrolled and treated with raloxifene at the doses of 60 mg/d (group A), 180 mg/d (group B), or placebo (group C). The duration of the treatment was 6 cycles of 28 d each.
At entry and after three and six cycles, uterine and leiomyoma sizes were measured by transvaginal ultrasonography. The difference between uterine and leiomyoma volumes (
size) was calculated in all subjects. The characteristics of the menstrual cycles and the side effects of the treatments were recorded using a daily diary. The severity of the uterine bleedings was assessed using a rank scale.
Throughout the study, no significant changes were observed in uterine and leiomyoma size or in size among the three groups and within each group of treatment. The length and severity of uterine bleedings was not significantly different between the three groups and within each group.
In conclusion, raloxifene has no significant effect on uterine and leiomyoma size or on menstrual cycle in premenopausal women.
THE UTERINE LEIOMYOMAS are the most frequent benign tumors affecting women, and at least 20–25% of women in fertile age are affected by uterine leiomyomas (1). This gynecological disease has a high economic cost; in fact, one third of hysterectomies are performed in patients referred for uterine leiomyomas (1).
It has been demonstrated that low-dose oral contraceptives reduce the duration and the severity of menstrual cycle in women with symptomatic uterine leiomyomas (2). Furthermore, this treatment may be associated with a higher risk to develop uterine leiomyomas (3, 4). It has been reported that the use of RU486, an antiprogestagen compound, induces a significant reduction of leiomyoma size acting on progesterone receptors (5, 6). Other studies (7, 8, 9) showed that it is possible to reduce leiomyoma size using gestrinone or danazol. However, because of their androgenic properties, these treatments have significant side effects. De Leo et al. (10) recently reported reduction of uterine leiomyomas after administration of laureatide, a somatostatin analog acting with a no sex hormones-related effect.
Today, the more effective treatment for uterine leiomyomas may be considered the GnRH analog (GnRH-a) administration (1). The continuous administration of GnRH-a inhibits the release of gonadotropins inducing a state of hypogonadotropic hypogonadism by down-regulating pituitary GnRH receptors (11). Moreover, the hypoestrogenic state induced by GnRH-a frequently causes climacteric-like symptoms that can be prevented by adding steroid hormones (add-back therapy) (12, 13). The association of sex hormones at the start of the GnRH-a treatment induces a reduction in the effectiveness of the analog alone (14, 15), whereas only the use of tibolone does not reduce it (16, 17, 18). In addition, it seems that women previously treated with GnRH-a plus add-back therapy should be considered at risk to develop an accelerated bone loss in postmenopausal age similarly to surgical postmenopausal women (19).
Preclinical data (20, 21, 22, 23, 24) showed that raloxifene hydrochloride, a selective estrogen receptor modulator (SERM), may have a beneficial effect on leiomyomas. In a recent clinical study (25) performed in postmenopausal patients with leiomyomas, we reported a significant reduction in leiomyoma size after 12 months of treatment with raloxifene 60 mg daily.
The aim of the present study was to evaluate the effects of different doses of raloxifene compared with placebo on uterine and leiomyoma sizes in premenopausal women with asymptomatic uterine leiomyomas.
Materials and Methods
This study has been performed according to the guidelines of the Helsinki Declaration on human experimentation and was approved by the Institutional Review Board. Before entering the study, the purpose of the protocol was clearly explained to women attending the general ambulatories of departments and all subjects signed a written informed consent form.
Ninety premenopausal women affected by uterine leiomyomas were enrolled in the study.
Inclusion criteria were: healthy premenopausal women with ovulatory menstrual cycles ranging from 26–30 d, and presence of no more than two asymptomatic uterine leiomyomas measuring more than 20 mm at transvaginal ultrasonography (TV-USG). The healthy state was confirmed by medical history, physical and pelvic examination, and complete blood chemistry. The spontaneous ovulatory cycles were evaluated by serial TV-USG and by measuring plasma progesterone (P) during the luteal phase of the cycle.
Exclusion criteria were: neoplastic, metabolic, liver, hematological and infectious diseases; history of acute or recurrent vascular thrombosis; body mass index (BMI) more than 30 kg/m2; use of hormonal therapy in the previous 3 months; presence of an hypoechoic or calcified leiomyomas; and endometrial abnormalities detected at TV-USG. Women with symptomatic uterine leiomyomas were excluded from the study for ethical reasons.
Using a computer-generated randomization list, study population has been randomized in three groups of 30 women each. Group A was treated with raloxifene at the dose of 60 mg daily (1 tablet/d per os) plus polyvitamins (2 tablets/d per os). Group B was treated with raloxifene at the dose of 180 mg daily (3 tablets/d per os). Group C (placebo group) was treated only with polyvitamins (3 tablets/d per os). The duration of the treatment was six cycles of 28 d each.
At the beginning of the study, and after three and six cycles of treatment, uterine and leiomyoma size, number of tumors, and endometrial thickness were evaluated by TV-USG. Ultrasonographic scans were performed at the fifth day of the menstrual cycle by the same experienced operator using a Toshiba PowerVision 6000 (Toshiba Medical System, Rome, Italy) equipped with a 7.5 MHz transvaginal probe. The operator was unaware of treatment assignment. Uterine and leiomyoma sizes were evaluated, measuring the three main diameters (D1, D2, D3) and applying the formula of the ellipsoid (D1·D2·D3·0.52). An arithmetic mean of the sizes was used in presence of two leiomyomas. As previously described (25), to evaluate the effect of raloxifene on the myometrium, the difference between uterine and leiomyoma volumes ( size) was calculated in each subject. Endometrial thickness was measured as well.
During the study, all women agreed to use barrier contraception. The subjects were instructed to report on a daily diary the characteristics of menstrual cycle (length and severity of uterine bleedings) as well as the onset of side effects.
The severity of the uterine bleedings was carefully recorded by each woman using a rank scale ranging from 1–10. Serum hemoglobin levels were measured before, and after three and six cycles of treatment.
For every three cycles, a standard clinical evaluation and laboratory analyses, including hematological, renal function, and liver function tests, and microscopic examinations of sediment from midstream urine specimens, were performed.
At baseline and after six cycles of raloxifene administration, serum FSH, 17ß-estradiol (E2), and P levels were measured in each women. In the early follicular, periovulatory, and luteal phases of the cycle, blood samples were obtained in the morning between 0800 and 0900 h after overnight fasting and bed rest. All blood samples were assayed in duplicate and, after centrifugation, stored at -80 C until analysis. Biochemical results were expressed as the mean of two measurements.
In particular, the blood samples of early follicular, periovulatory, and luteal phases were taken 5 d after menses, the same day in which the ovulation pattern was observed at TV-USG, and 7 d after the echographic signs of ovulation, respectively. The scans were performed by the same experienced operator every 3 d, starting the fifth day after the onset of menses. When the follicular dimensions achieved at least 16 mm, the TV-USG was performed daily. The follicular development was studied measuring the two main diameters of the follicle. The day of ovulation was retrospectively defined by the decrease in follicular dimensions and by the liquid in the cul-de-sac.
Plasma FSH, E2, and P, levels were measured with commercial kits. Serum FSH levels were determined by RIA [DiaSorin, Inc. Saluggia, Italy; sensitivity, 0.2 IU/liter; intra-assay and interassay coefficients of variation (CV), 1.4 and 4.2%, respectively]. Serum E2 levels were determined using a RIA (Eurogenetics Italy, Turin, Italy; sensitivity, 0 pmol/liter; intra-assay and interassay CV, 3.7 and 5.8%, respectively). Serum P levels were determined using a RIA (Eurogenetics Italy; sensitivity, 0.3 nmol/liter; intra-assay and interassay CV, 6.5 and 4.5%, respectively).
On the basis of a previous study (26), the sample size was calculated to be of 30 subjects per group to detect an effect on the size of 1 SD with an 
value of 0.05 (two-sided) and a power 1-
of 0.8.
Statistical analysis was performed using the SPSS 9.0 (SPSS, Inc., Chicago, IL) package. Data were expressed as mean ± SD. ANOVA was used to evaluate differences between the three groups in age, BMI, parity, number of cigarettes smoked per day, endometrial thickness, uterine and leiomyoma sizes, and size between groups at entry and after three and six cycles of treatment. The differences in length and severity of menstrual cycles between and within groups were compared at entry, and after three and six cycles of treatment using ANOVA and Wilcoxon’s signed rank tests, respectively. The changes in mean plasma hormone concentrations during the three phases of cycle at different times between and within groups were also evaluated by ANOVA.
Results
Only two women (one in group A and one in group C) dropped out of the study. The dropouts were due to personal reasons. Two women of group B reported the onset of hot flushes. One woman of group C reported gastralgia. No other adverse events were reported. No dropouts for drug-related side effects were reported.
The characteristics of the subjects studied are reported in Table 1
. At the beginning of the study, there were no significant differences among the groups for age, BMI, parity, and number of cigarettes smoked per day.
|
size was observed within and between groups (Table 2). After six cycles of treatment, a significant (P < 0.05) increase in mean uterine and leiomyoma size was observed only in groups A and C in comparison with basal values (Table 2). At the end of the study, the size was unmodified within and between groups (Table 2). After three cycles of treatment, no new tumor was detected in any woman. Furthermore, after six cycles of treatment, we observed three (10.3%) and four (13.8%) cases of new tumoral formations in groups A and C, respectively. At the same time, no new tumor was observed in group B. The number and the percentage of women for the three groups with unmodified, decreased, and increased uterine and leiomyoma size after three and six cycles of treatment are showed in Table 3.
|
|
).
Throughout the study, no significant difference in length and severity of uterine bleedings or in serum hemoglobin levels was observed within and between groups (Table 2).
No significant difference was detected in mean plasma FSH, E2, and P levels between the three groups at entry (Table 4). The mean plasma FSH and E2 levels were significantly (P < 0.05) higher in the periovulatory phase than early follicular and luteal phases in each group of treatment without significant differences between groups (Table 4). A significant (P < 0.05) difference in mean plasma E2 levels was also detected in luteal phase vs. early follicular and periovulatory phases in each group without significant differences between groups (Table 4). The mean plasma P levels in the luteal phase were significantly (P < 0.05) higher than in the early follicular and periovulatory phases in each group without significant differences between groups (Table 4). After six cycles of treatment, during the different phases of the menstrual cycle no significant variation was observed between groups and in each group in comparison with baseline values (Table 4)
|
Raloxifene is a synthetic nonsteroidal compound that belongs to SERMs, agents that interact with estrogen receptors (ERs) eliciting tissue-specific responses (27). It is known that raloxifene acts on the metabolism, the central nervous system, the skeleton, and the cardiovascular system as an estrogenic agonist (27, 28, 29), whereas it shows an estrogenic antagonist effect on reproductive organs, such as the breast and the uterus (30, 31, 32).
These different actions could be explained by the different distribution of the specific ER subtypes that mediate gene transcription if activated by raloxifene (33). In fact, two ERs exist in humans, encoded by two independent ER genes (34). ER binds estrogens with high affinity and low capacity, whereas ERß binds estrogens with low affinity and high capacity (34). The E2 activation of ER and ß gives two different regulatory signals, inducing activation and inhibition of transcription, respectively (33).
Raloxifene inhibits the proliferation of rat leiomyoma cells in vitro (20). Data on animal models showed that raloxifene has a dose-related capacity of blocking estrogen-induced stimulation of uterine weight gain (21) and induces minimal effects on myometrial thickness and uterine weight in untreated ovariectomized rats as determined by histological examination of uteri (22). In another study, it has been demonstrated that the administration of raloxifene induces a fast regression of abdominal wall estrogen-induced leiomyomas in guinea pigs (23). A more recent rat study (24) has confirmed that the treatment with tamoxifen or with a raloxifene analog reduces the size of leiomyomas and the incidence by 40–60%. This action is due to inhibition of cell proliferation without a decrease in apoptotic index (24).
Other clinical studies (35, 36, 37) have underlined an unclear effect of tamoxifen on uterine leiomyomas in humans.
Our previous clinical data (25) have shown that 1 yr of raloxifene administration at doses of 60 mg daily significantly reduces the uterine and leiomyoma sizes in postmenopausal women. A relevant finding of this study was the selective action of raloxifene on the leiomyoma tissue highlighted by a significant increase in difference between uterine and leiomyoma sizes ( size). In particular, in postmenopausal women the raloxifene administration seems to induce a significant reduction of leiomyoma sizes, without any significant action on normal myometrium.
On the contrary, in our present study on premenopausal women, no significant effect on uterine and leiomyoma sizes was observed after six cycles of raloxifene administration at standard and/or high doses.
However, our results should not be considered completely negative. In fact, after six cycles of raloxifene treatment at high dosage, in only two women an increase in tumor size was detected, whereas in an high percentage of cases the leiomyoma size was unmodified. Indeed, it seems that the use of 180 mg daily of raloxifene acts more on preventing tumoral growth than reducing the leiomyoma sizes. A higher incidence of new leiomyomas was observed in the groups treated with raloxifene 60 mg daily or placebo, thus suggesting a dose-related response of raloxifene treatment. Unfortunately, it was not possible to perform an appropriate statistical analysis to evaluate the raloxifene effect on the prevention of leiomyomas for the small group of women and the short treatment period.
The possible explanation for our results may be 2-fold. Firstly, we believe that the raloxifene doses were too low to reduce or revert the proliferative effect of serum E2 in normal ovulatory women. In fact, in postmenopausal women the serum E2 levels are about 10-fold lower in comparison with normally cycled premenopausal women. Secondly, it is possible that in postmenopausal women ERs have a different intratumoral pattern in terms of concentration, expression, and affinity in comparison with premenopausal women. Data in literature about this last point are missing and no clear consensus exists about the presence of sex hormone receptors in leiomyomas (38, 39, 40). Recent data (41) showed higher levels of ER and ERß mRNA in leiomyoma tissue. Furthermore, only the concentrations of ERß were 2- to 3-fold higher in leiomyoma in comparison with normal myometrium, whereas those of ER were relatively low (41). This result may be explained by the presence of ER variants lacking E2 binding sites for posttranscriptional modification or a faulty translation of ER mRNA. In this view, the differential expression of these two ER genes could play a pivotal role in the normal or abnormal growth of myometrium (41).
Few side effects were detected during this study, and the raloxifene treatment at standard and high doses was tolerated as well as placebo. Only two women referred the onset of hot flushes of mild severity during administration of raloxifene at 180 mg daily.
In the present study, no significant effect was observed on endometrial thickness or on the length and the severity of uterine bleedings after raloxifene treatment at doses of 60 and 180 mg daily. Unfortunately, in our research we studied only the plasma FSH, E2, and P levels at baseline and after six cycles of treatment during the different phases of the menstrual cycle. The aim of our study, in fact, was not to evaluate the effect of raloxifene on hormonal pattern because these data are already available (42). In particular, Baker et al. (42) showed no alteration in LH surge, FSH, P, and E2 levels during raloxifene administered at doses of 400 mg/d for 5 d during the follicular, periovulatory, and luteal phases, and at dose of 100 or 200 mg/d for 28 d/month in 31 healthy premenopausal women. Indeed, all women ovulated regularly, and only in some cases an increase of E2 and FSH levels was observed. On the contrary, in our study we studied the characteristics of menstrual cycle, such as length and severity of uterine bleedings, using a daily diary. The plasma FSH, E2, and P levels obtained at baseline and after six cycles of treatment during the different phases of the menstrual cycle confirm that raloxifene administration at doses of 60 and/or 180 mg daily did not modify the hormonal pattern in premenopausal women.
In conclusion, the administration of raloxifene in premenopausal women, also at high doses, does not induce any reduction in leiomyoma size; moreover, it is tempting to hypothesize in the near future the use of a SERM with a specific and potent antagonist action on leiomyoma tissue.
Acknowledgments
We thank Dr. Paolo Serra (Consiglio Nazionale delle Richerche, Calabria) for the statistical assistance.
Footnotes
Abbreviations: BMI, Body mass index; CV, coefficient(s) of variation; E2, 17ß-estradiol; ER, estrogen receptor; GnRH-a, GnRH analog; P, progesterone; SERM, selective estrogen receptor modulator; TV-USG, transvaginal ultrasonography.
Received November 26, 2001.
Accepted May 2, 2002.
References
and ERß at AP1 sites. Science 277:1508–1510 and ß. Endocrinology 138:863–870/ERß) in normal and pathological growth of the human myometrium: pregnancy and leiomyomas. Am J Physiol 276:1112–1118
This article has been cited by other articles:
 |
|
 |
|
  F. K. Varelas, A. N. Papanicolaou, N. Vavatsi-Christaki, G. A. Makedos, and G. D. Vlassis The Effect of Anastrazole on Symptomatic Uterine Leiomyomata Obstet. Gynecol., September 1, 2007; 110(3): 643 - 649. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Liu, H. Matsuo, Q. Xu, W. Chen, J. Wang, and T. Maruo Concentration-dependent effects of a selective estrogen receptor modulator raloxifene on proliferation and apoptosis in human uterine leiomyoma cells cultured in vitro Hum. Reprod., May 1, 2007; 22(5): 1253 - 1259. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Rivera, S. Christopoulos, D. Small, and M. Trifiro Hormonal Manipulation of Benign Metastasizing Leiomyomas: Report of Two Cases and Review of the Literature J. Clin. Endocrinol. Metab., July 1, 2004; 89(7): 3183 - 3188. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Palomba, F. Orio Jr., M. Morelli, T. Russo, M. Pellicano, C. Nappi, P. Mastrantonio, G. Lombardi, A. Colao, and F. Zullo Raloxifene Administration in Women Treated with Gonadotropin-Releasing Hormone Agonist for Uterine Leiomyomas: Effects on Bone Metabolism J. Clin. Endocrinol. Metab., October 1, 2002; 87(10): 4476 - 4481. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
