Abstract
Aim: Dienogest, a novel progestational 19‐norsteroid, has been shown to have a therapeutic effect on endometriosis with its major side‐effect being irregular genital bleeding. This study aimed to investigate the mechanism responsible for the bleeding seen during dienogest therapy.
Methods: For this multicenter, single‐dose, open‐label study, dienogest at a daily oral dose of 2 mg was administered for 16 weeks to nine patients with endometriosis showing regular menstrual cycles. Weekly determinations of serum endocrine hormone levels during the course of the treatment period and histological endometrial biopsies at the onset of genital bleeding were undertaken.
Results: All cases showed genital bleeding irregularly during the treatment period. Endometrial histological findings at the time of presentation of the bleeding revealed a secretory, inert or atrophic appearance of the epithelium, pseudo‐decidualization in the stroma, and local hemorrhage around the regions of desquamation of epithelium and lysis in the subepithelial stromal tissue, which are characteristics of breakthrough bleeding known to occur with progestational agents. In contrast, no association with the course of genital bleeding was found with respect to changes in serum estradiol, progesterone, follicle stimulating hormone and luteinizing hormone concentrations, plasma dienogest concentration, endometrial thickness or follicular diameter.
Conclusion: These results suggest that the genital bleeding seen in the period of medication with dienogest originated from breakthrough bleeding from the pseudodecidua, which reflects a direct progestational effect of dienogest on the endometrium. (Reprod Med Biol 2007; 6: 223–228)
Keywords: dienogest, endometriosis, endometrium, irregular genital bleeding
INTRODUCTION
DIENOGEST IS A novel 19‐nortestosterone derivative synthesized by Bayer Schering Pharma, Berlin, Germany, and supplied by Mochida Pharmaceutical in Japan. Dienogest exhibits highly selective binding to progesterone receptors 1 and exerts antiovulatory 2 and antigonadotrophic 3 activities, as well as antiproliferative activity in isolated human endometrial cells. 4 With this hormonal profile, dienogest would be expected to have a therapeutic effect on endometriosis; and the drug has been investigated in terms of its efficacy and safety in patients with endometriosis in Europe and Japan. 3 , 5
In these studies, most of the patients with endometriosis receiving dienogest therapy experienced irregular genital bleeding during the medication period. All 183 cases in a late phase II study in Japan experienced bleeding during 24 weeks of medication with dienogest, 3 although the genital bleeding was considered highly tolerable based on the intensity of spotting or breakthrough bleeding in most cases. No serious cases and a low incidence of discontinuation of the drug as a result of the bleeding were observed. However, the mechanism of genital bleeding with dienogest remains unclear and it is very important to clarify the bleeding mechanism for confirmation of the safety of Japanese patients under dienogest therapy and for dealing with the genital bleeding properly. Unfortunately, no bleeding has been reported in the non‐clinical studies on dienogest, including those on primates, nor has any experimental animal model to investigate the mechanism of exogenous steroid‐induced genital bleeding been established. 6 , 7 Because genital bleeding with dienogest was observed in all patients regardless of their background characteristics, 3 in the present study we planned to examine the common mechanism of the genital bleeding using a limited number of subjects through repeated hormonal and histological examinations.
MATERIALS AND METHODS
THE PRESENT STUDY was approved by the Institutional Review Board of the institutions and conducted between April 2004 and March 2005 in accordance with the Declaration of Helsinki and the Good Clinical Practice using a multicenter, single‐dose, open‐label study.
Nine patients with symptomatic endometriosis diagnosed by laparotomy, laparoscopy or imaging analysis (combination of magnetic resonance imaging and ultrasonography) of endometriotic ovarian chocolate cysts were enrolled under written informed consent. The mean ± standard deviation (SD) of patient age was 36.6 ± 6.6 years (range 28–49 years); weight was 56.9 ± 8.4 kg (range 39–69 kg); cycle length was 26.7 ± 2.2 days (range 23–30 days); and menstrual duration was 6.7 ± 1.1 days (range 5–8 days). The Beecham classification was stage II in five cases and stage III in four cases. Three patients had a complication of uterine leiomyoma, and three others had adenomyosis. No patient receiving hormonal therapy for endometriosis or with undiagnosed genital bleeding was enrolled in the study to avoid a possible influence on the appearance of genital bleeding by background factors.
The patients were administered dienogest orally at a dose of 2 mg/day, that is, two intakes of a 1‐mg tablet every day starting between the second and fifth days of the menstrual cycle as described in previous studies. 3
The treatment period was set at 16 weeks, and the daily incidence and intensity of genital bleeding were recorded in each patient's diary and were confirmed at every visit by each patient. It has been reported previously 3 that the incidence of genital bleeding peaked by 8 weeks of treatment with dienogest, so scheduled visits were set at once per week until 8 weeks of the treatment period so that we could examine the changes in hormonal and gynecological parameters prior to the first onset of genital bleeding and obtain an endometrial biopsy for histological examination during the first event of genital bleeding. Thereafter, scheduled visits were set at once every 2 weeks between 8 and 16 weeks of treatment so that we could examine the changes in hormonal and gynecological parameters along with the irregularly repeated genital bleeding and obtain a second endometrial biopsy during the bleeding event. To minimize the risk, we limited the endometrial biopsy in each patient to a total of two or less. Laboratory measurements were carried out at every visit to obtain serum concentrations of estradiol by radioimmunoassay (double‐antibody method), 8 of progesterone by electrochemiluminescence immunoassay, 9 and of follicle stimulating hormone (FSH) and luteinizing hormone (LH) by chemiluminescence immunoassay. 10 In addition, the plasma concentration of the unchanged form of dienogest was determined using high‐performance liquid chromatography; and endometrial thickness and dominant‐follicle diameter were measured using ultrasonography. The endometrial specimens from the biopsies were fixed in 10% buffered formalin and the paraffin sections were stained with hematoxylin–eosin (HE) or immunostained for progesterone receptor (PR), estrogen receptor (ER), proliferating cell nuclear antigen (PCNA) to detect cell proliferation, or terminal dUTP nick end labeling (TUNEL) to detect apoptosis for light microscopic observation.
RESULTS
ALL NINE PATIENTS enrolled completed the study. The duration of treatment with dienogest was 111.8 ± 2.3 days (mean ± SD, range 106–114 days) with more than 90% of scheduled exposure to the drug in all cases, and there was no major deviation from protocol.
Genital bleeding arose in all patients by 8 weeks of the treatment period after the termination of menstruation at the initiation of medication. The course of the duration and intensity of individual genital bleeding along with the treatment period was different in each patient (Fig. 1). There was no correlation in the onset, duration or intensity of genital bleeding with all the background data shown above. Genital bleeding disappeared during or after the completion of treatment in all cases, and then the first resumed menstruation was observed at 39.1 ± 19.5 days (mean ± SD, range 22–77 days) after the completion of treatment.
Figure 1.
Summarized appearance of the genital bleeding and the time of endometrial biopsy. In the 16 weeks of treatment with dienogest (2 mg/day, per os) all cases exhibited irregular genital bleeding and a total of 15 endometrial biopsies were taken.
Serum mean concentrations of estradiol, progesterone, FSH and LH at 2–16‐weeks of treatment were 19.3–63.3 pg/mL, 0.3 ng/mL, 4.7–9.8 mIU/mL and 2.4–6.7 mIU/mL, respectively, showing only a small change throughout the treatment period. Plasma mean concentration of the unchanged form of dienogest ranged from 33.7 to 39.4 ng/mL, and was almost constant throughout the treatment period. Endometrial thickness ranged from 3.8 to 5.6 mm in terms of mean value, and did not show any signs of thickening throughout the treatment period. The range of the mean diameter of the dominant follicle was from 9.3 to 15.1 mm throughout the treatment period, and the individual value was approximately 10 mm (the normal value for the early follicular phase) in almost all examination points, except a few at which a sporadical growth to over 16 mm (normal value for the preovulatory phase) was seen. To examine a change in the above parameters in relation to the bleeding episode, individual changes in each parameter at the onset of genital bleeding as compared to the corresponding value at 1 week before the episode were taken and revealed no significant changes (Table 1). In addition, a significant correlation was found in the serum estradiol concentration with the diameter of the dominant follicle (r = 0.492, P < 0.0001, Spearman's correlation coefficient).
Table 1.
Changes in hormone levels and gynecological parameters
| Item (Unit) | Hormone levels and gynecological parameters at each examination point | Change from 1 week before to the onset of bleeding | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Pre | 2‐week | 4‐week | 6‐week† | 8‐week | 10‐week | 12‐week | 14‐week | 16‐week | ||
| Estradiol (pg/mL) | 82.7 ± 62.9 | 63.3 ± 90.9 | 29.8 ± 24.3 | 38.6 ± 40.4 | 29.4 ± 29.2 | 23.0 ± 22.7 | 19.3 ± 13.0 | 29.8 ± 41.9 | 47.1 ± 37.3 | –2.0 ± 18.9 |
| Progesterone (ng/mL) | 9.8 ± 8.0 | 0.3 ± 0.1 | 0.3 ± 0.1 | 0.3 ± 0.2 | 0.3 ± 0.1 | 0.3 ± 0.1 | 0.3 ± 0.1 | 0.3 ± 0.2 | 0.3 ± 0.2 | 0.0 ± 0.1 |
| FSH (mIU/mL) | 5.2 ± 4.3 | 5.5 ± 0.9 | 5.4 ± 0.9 | 8.5 ± 5.8 | 6.4 ± 2.1 | 6.0 ± 2.8 | 9.5 ± 7.7 | 9.8 ± 6.9 | 4.7 ± 2.4 | –0.7 ± 4.4 |
| LH (mIU/mL) | 3.8 ± 1.9 | 4.4 ± 2.8 | 3.0 ± 1.2 | 4.2 ± 1.1 | 4.1 ± 2.3 | 4.0 ± 2.4 | 5.5 ± 5.9 | 6.7 ± 4.6 | 2.4 ± 1.7 | –0.6 ± 1.3 |
| Dienogest (ng/mL) | ND | 33.7 ± 9.5 | 35.4 ± 11.9 | 38.2 ± 7.1 | 39.4 ± 10.4 | 36.0 ± 14.7 | 38.5 ± 6.8 | 36.4 ± 8.3 | 36.0 ± 11.4 | 3.7 ± 9.8 |
| Endometrial thickness (mm) | 11.4 ± 3.8 | 5.2 ± 2.7 | 4.8 ± 1.8 | 5.5 ± 2.6 | 5.6 ± 3.6 | 4.7 ± 1.1 | 3.9 ± 0.8 | 4.3 ± 1.5 | 3.8 ± 0.8 | 0.3 ± 1.6 |
| Follicular diameter (mm) | 6.8 ± 3.0‡ | 12.1 ± 4.0 | 12.9 ± 4.3 | 11.5 ± 9.7 | 9.8 ± 2.9 | 9.3 ± 4.0 | 11.0 ± 2.6 | 9.7 ± 3.2 | 15.1 ± 6.4 | –1.4 ± 3.5 |
n = 4 because three cases at week 4 and two other cases at week 5 were biopsied during bleeding and, thus, their next visit was scheduled for week 8.
n = 5 because four cases did not show any echoically measurable follicle.
Values are mean ± standard deviation (n = 9). FSH, follicle stimulating hormone; LH, luteinizing hormone; ND; not detected.
A total of 15 endometrial biopsies were obtained from all patients (Fig. 1). The first biopsy in an individual case was carried out between days 1 and 6 after the onset of a bleeding period that began by week 8. Afterwards, 13 samples from eight patients were evaluated histologically (an insufficient amount of tissue was obtained from the two biopsies from one patient). The individual histological findings are presented in Table 2; and the representative appearance of a typical biopsy from patient 5 is seen in Figure 2. Endometrial histological findings at the time of presentation of genital bleeding revealed a secretory, inert or atrophic appearance of the epithelium and a decidual reaction or partially mild edematous appearance of the stroma. Hemorrhagic findings were primarily observed locally around the regions of desquamation of the epithelium and lysis in the subepithelial stromal tissue, except in one case (patient 4) in which hemorrhage was found around the extensive lysis area in stoma. The immunohistochemical results revealed stromal cells that were positive for ER, PR and PCNA and slightly positive for TUNEL; however, no stainings showed a change in the degree of positivity that coincided with the focal area of the hemorrhage.
Table 2.
Endometrial histological findings
| Patient no. | Time of biopsy | Histological findings | ||
|---|---|---|---|---|
| Epithelium | Stroma | Hemorrhagic region† | ||
| 1 | Day 29 | Secretory | Decidualized | Not observed (Dilatation in subepithelial sinus and local lysis in stroma) |
| Day 70 | Inert or atrophic | Decidualized | The regions of dilated subepithelial sinus and around the lysis area in stroma | |
| 2 | Day 44 | (Not evaluated)‡ | ||
| Day 99 | (Not evaluated) | |||
| 3 | Day 29 | Inert or atrophic | Decidualized | The regions of epithelial desquamation and around the lysis area in subepithelial stroma |
| Day 70 | Inert or atrophic | Decidualized | The regions of epithelial desquamation and in subepithelial area | |
| 4 | Day 49 | Atrophic | Decidualized | Around the extensive lysis area in stoma |
| 5 | Day 50 | Inert | Decidualized | The regions of epithelial desquamation |
| Day 85 | Inert | Decidualized | The regions of epithelial desquamation | |
| 6 | Day 36 | Secretory, inert or atrophic | Decidualized | Subepithelial stromal area |
| Day 72 | Atrophic | Decidualized | Subepithelial stromal area | |
| 7 | Day 55 | Inert or atrophic | Mildly edematous | Not observed |
| 8 | Day 34 | Atrophic | Decidualized or mildly edematous | Around the lysis area in subepithelial stroma |
| 9 | Day 30 | Atrophic | Decidualized | Not observed |
| Day 71 | Atrophic | Decidualized | Not observed | |
Comment in parentheses refers to hemorrhage‐related findings in a section where hemorrhage was not observed directly.
Impossible to evaluate histologically because of insufficient yield of endometrial tissue.
Figure 2.
Representative histological features of the endometrium at the appearance of genital bleeding from the biopsy from patient 5 on the 85th day of treatment. Atrophic glandular epithelium, stromal decidual reaction and dilatation of the sinus of the stromal tissue are seen, showing a typical appearance of pseudodecidua, which is a well‐known response to progestin treatment. Evidence of hemorrhage is observed locally in the dissociated area of the stromal tissue under the superficial endometrium. These findings suggest a histological appearance of breakthrough bleeding (hematoxylin–eosin staining).
DISCUSSION
IN THE PRESENT study, endocrine hormone concentrations, plasma concentration of the unchanged form of dienogest, endometrial thickness and dominant follicle diameter did not reveal any reliable change prior to the onset of genital bleeding during dienogest therapy. This finding suggested that none of the changes in these parameters seemed to be the primary cause of the bleeding episode. In contrast, the endometrial histological findings at the time of presentation of the bleeding demonstrated reproducible findings, such as secretory, inert or atrophic appearance in the epithelium, a decidual reaction or partially mild edematous appearance in the stroma, all of which were presumed to be typical to endometrial pseudo‐decidualization. The edematous stromal appearance was considered to be a sign of endometrial regeneration after shedding of the pseudodecidua. Hemorrhagic findings were primarily observed locally around the regions of desquamation of the epithelium and lysis in the subepithelial region of the stroma, showing an image of breakthrough bleeding. In one case (patient 4) in which a hemorrhagic appearance around the extensive lysis area in stoma plausibly showed an image of withdrawal bleeding, it was possible that it was an image of extended collapse of pseudodecidua with breakthrough bleeding because no decrease in the serum endocrine hormone concentrations and plasma concentration of the unchanged form of dienogest or increase in the stromal apoptotic cells 11 detected with TUNEL staining was shown prior to the onset of genital bleeding in the case.
Dienogest is a typical progestin highly selective to progesterone receptors 1 and can decidualize in vitro human endometrial stromal cells via these receptors. 4 In addition, the fact that the mean serum concentrations of estradiol, progesterone, FSH and LH in the treatment period remained within the standard range for the early follicular phase (11–82 pg/mL, ≤1.7 ng/mL, 3.0–14.7 mIU/mL and 2.29–16.93 mIU/mL, respectively) in all patients in this study suggested a systematic exhibition of the progestational antigonadotrophic effect of dienogest. 1 , 2 Moreover, most of the dominant‐follicle diameter remained at the level for the early follicular phase and a significant correlation between dominant‐follicle diameter and serum estradiol concentration was found. Therefore, the serum estradiol concentration was considered to reflect the extent of suppression in folliculogenesis as a result of the antigonadotrophic effect of dienogest. Taking these facts into account, we consider the endometrial histological findings of pseudodecidua to be relevant to the progestational effect of dienogest.
It is also known that progestins frequently produce unpredictable genital bleeding, 7 that the bleeding episodes of the progestins do not correlate with fluctuations in endogenous or exogenous hormone concentrations, 7 , 12 and that the endometrial histological appearance under continuous exposure to progestins includes a secretory, inert or atrophic epithelium, pseudo‐decidualization of the stroma and hemorrhagic appearance around the regions of local collapse of pseudodecidua, that is, an image of breakthrough bleeding, all of which are thought to be dependent on the progestational effect of the drugs. 12 , 13 As the present results are consistent with those of known progestins, we consider that the progestational effect of dienogest is a key part of the underlying mechanism of the genital bleeding as well as the histological changes in the endometrium during treatment with the drug. Furthermore, it is still unknown how progestins cause breakthrough bleeding in the pseudodecidua. A decrease in estrogen or progesterone receptor‐mediated signal transduction in endometrial stromal cells under progestin treatment is postulated as a possible mechanism of breakthrough bleeding. However, conflicting results have been obtained if the regional expression of endometrial steroid hormone receptors showed a decrease in women with progestin treatment. 6 In the present study, none of the immunohistochemical results of endometrial ER and PR showed a change in positivity that coincided with the focal area of the hemorrhage, suggesting that they are not the direct causes of the onset of breakthrough bleeding. To clarify this point, further examinations are necessary.
In conclusion, it should be noted that in dealing with irregular genital bleeding during treatment with dienogest that the bleeding originated mainly from breakthrough bleeding from the pseudodecidua, suggesting a mechanism in common with that of other progestins.
ACKNOWLEDGMENTS
THIS STUDY WAS sponsored by Mochida Pharmaceutical Company, Tokyo, Japan, and dienogest was supplied by the sponsor. We would like to extend particular thanks to Dr Masahiko Maegawa, Dr Toshiya Matsuzaki, Dr Akira Kuwahara and Dr Satoshi Yamamoto at the Tokushima University Hospital, Dr Shusaku Kamada at the Keiai Ladies Clinic, Tokushima, Japan and Dr Mika Kawano at the Kawano Mika Ladies Clinic, Tokushima, Japan for their participation in this clinical study. We also thank Dr Yuji Taketani at the University of Tokyo Hospital, Dr Naoki Terakawa at the Tottori University Hospital and Dr Takeshi Aso at the Tokyo Medical and Dental University for their professional medical advice.
REFERENCES
- 1. Foster RH, Wilde MI. Dienogest. Drugs 1998; 56: 825–833. [DOI] [PubMed] [Google Scholar]
- 2. Moore C, Carol W, Gracer T, Mellinger U, Walter F. Influence of dienogest on ovulation in young fertile women. Clin Drug Invest 1999; 18: 271–278. [Google Scholar]
- 3. Momoeda M, Taketani Y, Terakawa N, Aso T, Tsutani K. A randomized, double‐blind, multicenter, parallel, dose–response study of dienogest in patients with endometriosis. Jpn Pharmacol Ther 2007; 35: 769–783. [Google Scholar]
- 4. Okada H, Nakajima T, Yoshimura T, Yasuda K, Kanzaki H. The inhibitory effect of dienogest, a synthetic steroid, on the growth of human endometrial stromal cells in vitro . Mol Hum Reprod 2001; 7: 341–347. [DOI] [PubMed] [Google Scholar]
- 5. Cosson M, Querleu D, Donnez J et al Dienogest is as effective as triptorelin in the treatment of endometriosis after laparoscopic surgery: results of a prospective, multicenter, randomized study. Fertil Steril 2002; 77: 684–692. [DOI] [PubMed] [Google Scholar]
- 6. Findlay JK. Future directions for research on endometrial bleeding. Hum Reprod 1996; 11: 179–183. [DOI] [PubMed] [Google Scholar]
- 7. Fraser IS. Bleeding arising from the use of exogenous steroids. Baillieres Best Pract Res Clin Obstet Gynaecol 1999; 13: 203–222. [DOI] [PubMed] [Google Scholar]
- 8. Carlstrom K. Low endogenous estrogen levels – analytical problems and tissue sensitivity. Acta Obstet Gynecol Scand Suppl 1996; 163: 11–15. [PubMed] [Google Scholar]
- 9. Blackburn GF, Shah HP, Kenten JH et al Electrochemiluminescence detection for development of immunoassays and DNA probe assays for clinical diagnostics. Clin Chem 1991; 37: 1534–1539. [PubMed] [Google Scholar]
- 10. Primus FJ, Kelley EA, Hansen HJ, Goldenberg DM. ‘Sandwich’‐type immunoassay of carcinoembryonic antigen in patients receiving murine monoclonal antibodies for diagnosis and therapy. Clin Chem 1988; 34: 261–264. [PubMed] [Google Scholar]
- 11. Dahmoun M, Boman K, Cajander S, Westin P, Backstrom T. Apoptosis, proliferation, and sex hormone receptors in superficial parts of human endometrium at the end of the secretory phase. J Clin Endocrinol Metab 1999; 84: 1737–1743. [DOI] [PubMed] [Google Scholar]
- 12. Marbaix E, Vekemans M, Galant C et al Circulating sex hormones and endometrial stromelysin‐1 (matrix metalloproteinase‐3) at the start of bleeding episodes in levonorgestrel‐implant users. Hum Reprod 2000; 15: 120–134. [DOI] [PubMed] [Google Scholar]
- 13. Dallenbach‐Hellweg G. The influence of contraceptive steroids on the histological appearance of the endometrium. In: WHO Symposium on Steroid Contraception and Mechanisms of Endometrial Bleeding, 12–14 September 1979; Geneva, Switzerland. Bath: Pitman Press, 1980; 153–173. [Google Scholar]