Department of Gynecologic Surgery, Maternité Jeanne de Flandres, Lille, France

Transcription

1 FERTILITY AND STERILITY VOL. 77, NO. 4, APRIL 2002 Copyright 2002 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A. Dienogest is as effective as triptorelin in the treatment of endometriosis after laparoscopic surgery: results of a prospective, multicenter, randomized study Michel Cosson, M.D., a Denis Querleu, M.D., a Jacques Donnez, M.D., b Patrick Madelenat, M.D., c Philippe Koninckx, M.D., d Alain Audebert, M.D., e and Hubert Manhes, M.D. f Department of Gynecologic Surgery, Maternité Jeanne de Flandres, Lille, France Received March 20, 2001; revised and accepted December 11, The sponsors of this study, Society Innothera and Schering (France), provided financial and technical support. Reprint requests: Michel Cosson, M.D., Maternité Paul Gelle, 91 Avenue Julien Lagache, Roubaix, France (FAX: ; m- cosson@chru-lille.fr). a Department of Gynecologic Surgery, Maternité Jeanne de Flandres, Lille, France. b Université Catholique de Louvain, Clinique Universitaire St. Luc, Service de Gynécologie Obstétrique, Brussels, Belgium. c Service Gynécologie Obstétrique, Hôpital Bichat-Claude Bernard, Paris, France. d Gasthuisberg University Hospital, Leuven, Belgium. e Centre Hospitalier Bordeaux, Service Gynécologie, Bordeaux, France. f Service Gynécologie Obstétrique, Clinique de la Pergola, Vichy, France /02/$22.00 PII S (01) Objective: To compare the efficacy of Dienogest versus Decapeptyl at 3.75 mg as consolidation therapy for surgery in the treatment of endometriosis. Design: Multicenter, open, randomized, parallel-group clinical trial. Setting: Volunteer patients in an academic research environment. Patient(s): Women with grade 2, 3, and 4 ( 70) endometriosis at initial laparoscopy. Intervention(s): We provided 16 weeks of treatment with Dienogest, 1 mg tablet daily; or with Decapeptyl, 3.75 mg IM injection every 4 weeks. Main Outcome Measure(s): A change in the patient s Revised American Fertility Society score at the post-treatment laparoscopy. Result(s): From June 1994 to July 1998, 142 patients were enrolled in the trial. After exclusion for major protocol deviations, 59 patients were included in the Dienogest group and 61 in the Decapeptyl group. This study group was comparable to the first inclusion group. The patient demographic and clinical characteristics, median duration of endometriosis, Revised American Fertility Society scores, and Visual Analogic Squale (VAS) scores were comparable in both groups. Statistical analysis of efficacy was not significantly different between the two groups. Adverse events were reported by 87.7% of patients in the Dienogest group and 85.1% in the Decapeptyl group. Neither treatment affected patient body weight or vital signs. Conclusion(s): Dienogest is as effective as Decapeptyl for consolidation therapy after surgery for the treatment of endometriosis. The safety profile of dienogest differed from Decapeptyl (3.75 mg). Dienogest constitutes a new therapeutic alternative to the GnRH analogues. (Fertil Steril 2002;77: by American Society for Reproductive Medicine.) Key Words: Randomized trial, endometriosis, GnRH agonist, progestogen Endometriosis is estimated to affect 2% to 3% of the general female population and as many as 15% to 30% of women with primary or secondary infertility (1). In a hypoestrogenic or hyperandrogenic environment, endometriotic implants tend to regress (2, 3). Therefore, several medical antigonadotrophic and antiestrogenic treatments have been developed. Because of their anabolic and androgenic side effects, Danatrol and the progestogens have been replaced by the GnRH analogs (4 7). Recently, progestogens with minimal androgenic activity have been developed (7). There has been no randomized study comparing a progestogen to a GnRH analog for this indication. Dienogest (17 -cyanomethyl-17 -hydroxyestra-4,9-dien-3-one), manufactured by Schering, is a synthetic progestogen derived from 19-norsteroids. Its pharmacologic profile is similar to the progesterone derivatives (17-hydroxy or 19-nor-progesterone). The aim of our study was to compare Dienogest to a GnRH analog as consolidation therapy following operative laparoscopy. The analog we used was 3.75 mg of triptorelin (Decapeptyl ) because of its widespread use, efficacy, and safety (8, 9). 684

2 METHODS This is a phase III, multicenter, open, randomized trial of patients who were found to have stage 2 to 4 endometriosis according to the Revised American Fertility Society (rafs) classification (10) (score 70) and who underwent operative laparoscopy. During a 16-week period, the patients in the study group were prescribed Dienogest, 1 mg orally twice a day. Patients in the control group received an injection of triptorelin, 3.75 mg IM every 4 weeks, for the 16-week period. The efficacy of this postsurgical consolidation treatment was assessed by a clinical examination and a control laparoscopy. Inclusion criteria for this study were patient age of 18 to 40 years, and endometriosis confirmed by diagnostic laparoscopy (Lap.0), irrespective of the initial symptomatology. The rafs 0 scores were 6 to 15 (stage 2), 16 to 40 (stage 3), or 70 (stage 4). Patients must not have had any form of hormonal therapy for a minimum of 3 months prior to enrollment. Patients were excluded if they had any contraindications to laparoscopy or to synthetic progestogens (i.e., past history of thrombophlebitis, myocardial infarction, cerebrovascular accident, hepatitis, diabetes or dyslipidemia, ongoing pregnancy, or uterine hemorrhage of unknown origin). Hormonal treatment within the 3 months preceding diagnostic laparoscopy whether with natural or synthetic estrogens (except local estrogens), Danazol, a GnRH analogue, or combined oral contraceptives (estrogen/progestogen) was not authorized. Patients with prior oral contraceptive use were required have had at least two regular spontaneous cycles prior to the initial laparoscopy. Serious adverse effects were immediately reported to the project manager. Efficacy was assessed after a new clinical examination (change in pain and global patient assessment) and a new laparoscopy (rafs and FOATI endometriosis scores). FOATI is a classification of endometriosis (Ovarian Foci, Tubular Adhesions, Inflammation rating scale) that includes an evaluation of the focus of endometriosis, adhesions, permeability of fallopian tubes, and inflammatory appearance of lesions. Clinical examination and blood tests were also performed to evaluate the safety of the treatments. The run-in procedure comprised three stages (Fig. 1): a clinical stage (V 1 ) to check the inclusion criteria in patients with known or suspected endometriosis, a laparoscopic stage, and a blood test. The laparoscopic stage included a diagnostic stage (Lap.0) to classify and establish a score for the endometriotic lesions according to the rafs (rafs 0) and FOATI (FOATI 0) classifications, and a therapeutic stage (Lap.1). After surgery, a subsequent rafs score (rafs 1) was the reference point before any medical treatment was begun. The blood test verified that there were no biologic exclusion criteria in patients who had met the clinical and laparoscopic inclusion criteria. During the inclusion visit (V 2 ), the patients were given a randomly assigned treatment. Randomization was centralized, with the treatment units numbered from 1 to 8 for each center. The investigator and the patient were only aware of the type of treatment when the decision to enroll the patient was made. Dienogest was supplied in 1-mg film-coated tablets. Decapeptyl was supplied in syringes for IM injections (containing 3.75 mg of triptorelin). The prescriptions were as follows: Dienogest: two 1-mg tablets taken orally every day (i.e., one tablet in the morning and one in the evening) for 16 weeks. Decapeptyl : one 3.75-mg IM injection every 28 days for 16 weeks (i.e., four injections). In the Dienogest group, the treatment was started on the first or the second day of the cycle following laparoscopy. In the Decapeptyl group, the injections were given on an outpatient basis by a state-registered nurse. The first injection of Decapeptyl was given the first or the second day of the cycle after laparoscopy, or from the 24th day of the cycle after laparoscopy according to the investigator (Fig. 2). The global intensity of pain was assessed at V 3 and V 4. A new pelvic examination was performed to detect any modification in the position, sensitivity, mobility, or size of the uterus or the adnexa, pain in the posterior cul-de-sac, or uterosacral nodules. Blood tests checked the lipid profile (total cholesterol, triglycerides, high-density and low-density lipoprotein cholesterol), the fasting blood glucose, serum transaminases (AST, ALT), -GT, and alkaline phosphatase before and after 16 weeks of treatment. When the treatment was discontinued, a control laparoscopy (Lap.2) was performed during the follicular phase of the following spontaneous cycle. The report included a new lesion score, according to the same classifications used at the initial examination (rafs 2 and FOATI 1). After completion of the control laparoscopy, the investigator recorded the global assessment for each patient based on all study data. The reproductive outcome for infertile patients within the 12 months after the end of the medical treatment was recorded. A confidence interval of 90% was calculated to see whether Dienogest and Decapeptyl had at least the same effect. Whether Dienogest was more efficient than Decapeptyl was assessed by calculating a 95% confidence interval and by the level of the bilateral test to detect a difference between the two groups. In our data, efficiency and confidence intervals are reported in such a way that a positive effect means that Dienogest is better than Decapeptyl for a given criterion. The Hodges-Lehmann estimator (11) was used to evaluate the extent of the effect. The 90% and 95% confidence intervals containing the true effect were calculated. The comparison of the lower limit of the 90% confidence interval with the noninferiority threshold, taken to be equal to 4 FERTILITY & STERILITY 685

3 FIGURE 1 Planned interventions and their timing. V 1 run-in visit; Lap.1 diagnostic laparoscopy (rafs 0) (FOATI 0) therapeutic laparoscopy (rafs1); LT1 blood tests (in hospital); V 2 ensure that laboratory tests are normal collection of consent decision to include patient (randomization); D1 first day of menstruation (of cycle following laparoscopy) start of medical treatment; V 3 second month visit; LT2 check laboratory tests (in hospital); V 4 fourth month visit (patient s global assessment); Lap.2 control laparoscopy (rafs 2, FOATI 1) and investigator s global assessment; V 5 visit 12 months after withdrawal of medical treatment (for previously infertile patients). points for the implant scores, allows the noninferiority of Dienogest to Decapeptyl to be evaluated statistically. Thus, if this lower limit is less in absolute terms than 4 points, the noninferiority of Dienogest to Decapeptyl can be concluded significantly at the unilateral 5% level. Analysis for the existence of any difference between the two groups was carried out with the Wilcoxon nonparametric test consistent with the 95% confidence interval containing the true effect. The protocol was approved by the Lille institutional ethics committee and review board on April 5, RESULTS From June 28, 1994, to December 8, 1997, 142 patients were enrolled in 17 centers: 74 in the Dienogest group and 68 in the Decapeptyl group (Fig. 3). Twelve patients withdrew prematurely from the study: nine in the Dienogest group (12.3%) and three in the Decapeptyl group (4.5%). Two patients withdrew from the study before taking any treatment: one patient after being counseled by her general practitioner, the other one because of a spontaneous pregnancy. Five patients withdrew from the study during treatment in the Dienogest group for each of the following reasons: diabetes, metrorrhagia, migraine, gastric pain, and patient s choice. In the Decapeptyl group, one patient withdrew for headache and hot flushes. In the period between the end of the treatment and the control laparoscopy, one patient withdrew for pregnancy and one for hypothyroidism. In the Decapeptyl group, one patient discontinued for a miscarriage. Due to major protocol deviations, 10 patients who were randomized, treated, and for whom a control laparoscopy was performed were excluded (6 patients in the Dienogest group and 4 patients in the Decapeptyl group). Deviations were similar between the two groups. Thus, the remaining 120 patients who were randomized, completed the therapy as prescribed, and underwent a control laparoscopy comprise the population to be analyzed. The mean age and the mean weight were similar in the Dienogest group and the Decapeptyl group (28.5 years 4.9 vs years 5.1, and 58.4 kg 9.9 vs , respectively). The vital signs (systolic blood pressure, diastolic blood pressure, and heart rate) were in the normal range and were comparable between the two groups. Among both groups, 32% of the patients had previously diagnosed 686 Cosson et al. Postoperative treatment of endometriosis Vol. 77, No. 4, April 2002

4 FIGURE 2 Description of treatment. endometriosis, with a median duration of 668 days, which was similar in both groups (Table 1). The majority of these patients had already received a medical (85.7%) or a laparoscopic treatment (80%). The differences between the two groups were not statistically significant. The symptoms and signs were comparable in the Dienogest group and the Decapeptyl group, including pain (83.1% vs. 85.2%) dysmenorrhea (83.6% vs. 91.5%), dyspareunia (66.7% vs. 66.1%), infertility (over half of the patients in both groups), or infertility and pain (approximately 50% in each group). Even though pelvic pain was slightly more common in the Decapeptyl group (67.8% vs. 47.5%, respectively; P.03), its mean duration or intensity was similar in both groups during the menstrual cycle. The total rafs 0 score and each of the components of this score (implant score, adhesion score, posterior cul-de-sac obliteration score) had a similar distribution in the two groups (Table 2). The median implant score (primary efficacy criterion: maximum score of 46) was 20.0 in the Dienogest group and 22.0 in the Decapeptyl group; 25% of patients in each group had an implant score of over 26. The major criterion was the implant score using the rafs classification. The implant and the adhesion score (Table 3) were calculated at the laparoscopies performed in the study: diagnostic laparoscopy at inclusion (Lap.0), postsurgical laparoscopy (Lap.1), and post-treatment control laparoscopy (Lap.2), as shown in Tables 2 through 4. There was no statistically significant difference in the implant scores for the two groups at Lap.2 (P.43). Analysis of the quartiles shows that at least 75% of the patients had a score of less than 16 in the Dienogest group and less than 4 in the Decapeptyl group. No reappearance of implants was observed in at least 25% of the patients in each group (zero score). The median difference of the implant score between the postsurgical laparoscopy and the control laparoscopy (change Lap.2 Lap.1) is identical in the two groups (median: 0). The difference between the changes in the score observed at Lap.2 between the two groups after 4 months of treatment is not statistically significant. The 90% confidence interval shows that the lower limit is 2 while the noninferiority limit was 4. Categorical analysis of change in the implant score (worsening, stability, improvement) revealed no statistically significant difference between the treatments (P.25). More patients had higher adherence score or obliteration of the posterior cul-de-sac at Lap.2 compared to Lap.1 (Table 4). At Lap.2, More patients had also complete FERTILITY & STERILITY 687

5 FIGURE 3 Trial profile. obliteration in the two groups than at Lap.1, but it still remains lower than Lap.0. There was no statistically significant difference between the two groups at Lap.2 (P.85). After 4 months of treatment, the modification in the rafs score for the two groups (score Lap.2 Lap.1) is similar, with no difference observed. The lower limit of the 90% confidence interval is 6. Dienogest is as efficient as Decapeptyl. In the Dienogest group, 34.5% of the patients were very satisfied and 51.7% were satisfied; thus, 86.2% of the patients were satisfied. In the Decapeptyl group 30% of the patients were very satisfied with the treatment, and 50.0% satisfied, thus there was a total of 80.0% satisfied patients. The odds ratios show that the distribution of patients as a function of response in each group is favorable to Dienogest (odds ratio 1.35), but the difference between the two groups is not statistically significant (P.39). In infertile patients before the study, the rate of spontaneous pregnancies was analyzed over 12 months following the end of the treatment (45 patients in the Dienogest group and 41 in the Decapeptyl group). Of these, 15 patients 688 Cosson et al. Postoperative treatment of endometriosis Vol. 77, No. 4, April 2002

6 TABLE 1 Patients with previous endometriosis. Total population (n 35) Dienogest (n 13) Decapeptyl (n 22) Duration of disease (days) Q1; Q3 349; ; ; 1731 Range Prior medical treatment N (%) 30 (85.7) 13 (100.0) 17 (77.3) Prior laparoscopic treatment N (%) 28 (80.0) 12 (92.3) 16 (72.7) (33.3%) in the Dienogest group and 12 patients (29.3%) in the Decapeptyl group had a spontaneous pregnancy, which is not significantly different (P.71) (Table 5). The blood tests showed statistical differences for the following parameters: Total cholesterol: the median increase during treatment was 0.15 g/l (P.01) on Decapeptyl versus 0.05 g/l (P.03) on Dienogest (P intergroup.02). HDL cholesterol: the median increase during treatment was 0.03 g/l on Dienogest (P.04) versus 0.09 g/l on Decapeptyl (P.01). Alkaline phosphatase: the median increase during treatment was 1.0 IU/L (not statistically significant) on Dienogest versus 11 IU/L (P.01) on Decapeptyl (P intergroup.01). At least one side effect was experienced by 92% of the patients, with no difference in frequency between the two groups. Among these, only 23% had a severe side effect. The clinical safety profile of the two treatments was identical, except for spotting being more frequent with Dienogest (61.6% vs. 25.4%), and hot flushes being more frequent with Decapeptyl (61.2% vs. 9.6%). The clinical safety of the treatments was reflected by the global patient satisfaction: 85% of the patients on Dienogest and 80% on Decapeptyl. Dienogest had a good biologic safety, with no changes in blood glucose, and a smaller increase in total cholesterol than Decapeptyl. Finally, alkaline phosphatase was increased with Decapeptyl but not with Dienogest (P intergroup.01). DISCUSSION Progestogens combining a central inhibitory effect with their peripheral action obtain complete resolution of endometriotic lesions in 50% of cases and partial resolution in 13% (12). However, the compounds used (mainly medroxyprogesterone acetate, lynestrenol, ethylnodrel, or norethisterone) have marked adverse metabolic effects (13, 14). The search for more powerful antigonadotropic compounds led to danazol and gestrinone (which has not been marketed). For some time, danazol was the reference therapy for endometriosis, but it had androgenic side effects. GnRH analogs are currently the reference, because they have the same efficacy as danazol, but lack its clinical androgenic and metabolic side effects. This report demonstrates the results of a multicenter, open, randomized, controlled clinical study comparing the efficacy and safety of a progestogen, Dienogest, versus a TABLE 2 Total revised American Fertility Society score. Laparoscopy Dienogest (n 59) (Q1; Q3) Decapeptyl (n 61) (Q1; Q3) Effect 90% CI 95% CI P Lap ; 7 8; ; 61 24; 54 Lap ; 0 2; ; 12 0; 12 Change in score ; 2 10; 4.38 Lap.0 45; 8 44; 20 Lap.1 Lap ; 3 7; ; 40 4; 24 Change in score ; 4 6; 5.97 Lap.2 Lap.1 1; 26 1; 18 FERTILITY & STERILITY 689

7 TABLE 3 Implant score in revised American Fertility Society classification. Laparoscopy Dienogest (n 59) (Q1; Q3) Decapeptyl (n 61) (Q1; Q3) Effect 90% CI 95% CI P Lap ; 5 0; ; 24 16; 26 Lap ; 0 0; ; 4 0; 8 Change in score ; 2 5; 2.47 Lap.1 24; 4 24; 6 Lap.2 Lap ; 0 2; ; 16 0; 4 Change in score ; 0 2; 0.31 Lap.2 Lap.1 0; 4 2; 2 GnRH analog, Decapeptyl, in the treatment of endometriosis. The use of a placebo was not ethically justified, because medical treatments have conclusively demonstrated efficacy in the postoperative management of endometriosis of stages 2 to 4 on the rafs classification (15). Finally, the study was conducted with an open design because of the difference in formulation, the injections, and the issues with creating a double dummy. Having in mind the type of comparison made (i.e., comparison vs. reference therapy), the objective was to demonstrate a comparable efficacy of Dienogest to Decapeptyl. The efficacy of a treatment for endometriosis can be assessed by the anatomic modifications observed in lesions, so laparoscopy was essential before and after treatment. Patients had to be severely affected to justify the second laparoscopy (i.e., stages 2 to 4 on the rafs classification). In these patients, medical treatment was used as an adjunct to surgery. In the literature, the average duration of treatment with GnRH analogs is 6 months. The current trend is to use them for no more than 3 to 4 months, as they have harmful effects on bone metabolism. Because 3 months is a relatively short period of time to assess the therapeutic effect of a progestogen, 4 months seemed to be a good compromise. The anatomic criteria (scores recorded at laparoscopy) are the TABLE 4 Adhesion score in revised American Fertility Society classification. Laparoscopy Dienogest (n 59) (Q1; Q3) Decapeptyl (n 61) (Q1; Q3) Effect 90% CI 95% CI P Lap ; 1 4; ; 22 4; 16 Lap ; 0 0; ; 4 0; 0 Change in score ; 2 3; 3.83 Lap.1 17; 1 16; 2 Lap.2 Lap ; 4 1; ; 16 2; 16 Change in score ; 4 0; 4.15 Lap.2 Lap.1 0; 2 1; Cosson et al. Postoperative treatment of endometriosis Vol. 77, No. 4, April 2002

8 TABLE 5 Pregnancy rate and characteristics of the spontaneous pregnancies. Pregnancies Dienogest (n 45) Decapeptyl (n 41) n % n % Total spontaneous pregnancies a Proceeding normally to term Conception after treatment Conception during treatment Interrupted pregnancies Conception after treatment Conception during treatment a P.71. only objective criteria to evaluate the evolution of the disease. The modification in the rafs score (difference between rafs 2 and rafs 1 obtained after the surgical cure) was the first criterion defined in the protocol. This score evaluates the posterior cul-de-sac obliteration, implant and adhesion scores; the efficacy of the medical treatment was best assessed with the implant score, which is the most specific criterion, so it was used as the main criterion during the study. There was no significant difference in implants or adhesions, tubal or ovarian localization, or changes in the global score or the score for each individual type of lesion related to the population evaluated. These results show that Dienogest is as effective as Decapeptyl. Moreover, both treatments are equivalent in terms of efficacy, the effect observed being zero and not statistically significant (P.31). This noninferiority of Dienogest was also apparent in patients with severe disease (implant score 15 at diagnostic laparoscopy), taking account of the date of the post-treatment laparoscopy. According to Evers (16), the scores recorded after the return of menstruation are the most relevant. A comparison between endometriosis status (implants and adhesions) before and after medical treatment shows an improved global score in 20% of patients in both groups. No change was observed in 53% of the patients treated with Dienogest or in 60% of those treated with Decapeptyl. Therefore, we can conclude that in both groups results were satisfactory in 70% to 80% of cases, which is equivalent to the results previously observed with Danazol in both comparative (17) and placebo-controlled (13) studies. This is similar to the success rate observed with Decapeptyl (8, 9) and other GnRH analogs tested in the treatment of endometriosis (18). Furthermore, there was no statistically significant difference between the two groups regarding the inflammatory status as assessed by the FOATI score. The reduction in pain was the same with both treatments. Among the patients, 86 were complaining of infertility (45 in the Dienogest group and 41 in the Decapeptyl group). In the Dienogest group, the pregnancy rate was 33%, and 13 out of the 15 pregnant patients delivered. In the Decapeptyl group, 29% of the patients became pregnant, and 9 out of 12 delivered. The total rate of deliveries was 29% on Dienogest versus 22% on Decapeptyl. The difference between the two groups favored Dienogest, but was not statistically significant. In cases of infertility, Dienogest is not inferior to Decapeptyl, with no increased risk of infertility or miscarriage. With regard to clinical safety, the symptoms in treated patients mainly involved the automatic nervous system. Because GnRH analogs induce a chemical menopause, the climacteric syndrome is usual. Hot flushes were reported in 61% of women treated with the GnRH analog (severe in 31%), compared with 11% of women treated with progestogen. Vaginal bleeding was most common complication observed with Dienogest (61.6% of cases as compared with 25.4% of those treated with the GnRH analog). Bleeding usually was no more than spotting, which is a common problem during the first months after hormone therapy (hormone replacement therapy or oral contraception, particularly at low doses). The treatments were well tolerated, as shown by the global satisfaction rate. On the other hand, and unlike the situation observed in patients treated with classic progestogens or danazol, there were no reports of significant androgenic effects such as weight gain, acne, alopecia, or virilism (hirsutism). This confirms the previous reports that Dienogest is not androgenic (19). The overall safety and efficacy of these treatments from the point of view of the patients and the physicians were assessed by recording a global satisfactory rate: 85.5% of the patients treated with Dienogest and 80% of those treated with Decapeptyl were satisfied or very satisfied. Dienogest had no androgenic effects and this was confirmed at the biologic level. The main biochemical modification was an FERTILITY & STERILITY 691

9 increase in alkaline phosphatase in the Decapeptyl group, which may reflect an increased bone turnover. This was not seen with Dienogest. The lipid profile (particularly HDL cholesterol) and blood glucose levels were similar in both groups. CONCLUSION Dienogest is a therapeutic alternative to GnRH analogs in the treatment of endometriosis. In this study, a 4-month postoperative treatment of endometriosis with Dienogest was as efficient as Decapeptyl and had no androgenic effects. Acknowledgments: The authors thank Dr. Pennehouat (Paris, France), Dr. Cusumano (Liège, Belgium), Dr. Pettemans (Brussels, Belgium), Pr. Tran (Nice, France), and Pr. Raudrant (Lyon, France) for help in the investigation; and also Afchine Fazel, M.D., and Yukio Sonoda, M.D., for their essential editorial help. References 1. Muse KN, Wilson EA. How does mild endometriosis cause infertility? Fertil Steril 1982;38: Bur ME, Greene GL, Press MF. Estrogen receptor localization in formalin-fixed paraffin embedded endometrium and endometriotic tissues. Int J Gynecol Pathol 1987;6: Barbieri RL. Hormone treatment of endometriosis: the estrogen threshold hypothesis. Am J Gynecol 1992;166: Donnez J, Nisolle M, Grandjean P, Gillerot S, Clerckx F. La place des agonistes de la GnRH dans le traitement par voies endoscopiques de l endométriose et des fibromyomes. Contracep Fertil Sex 1993;21: Henzl MR, Cordon SL, Moghissi K, Buttram VC, Berqvist C, Jacobson J. Administration of nasal nafarelin as compared with oral danazol for endometriosis. A multicenter double-blind comparative clinical trial. N Engl J Med 1998;318: Wheeler JM, Knittle JD. Depot leuprolide versus danazol in treatment of women with symptomatic endometriosis. Efficacy results. Am J Obstet Gynecol 1992;167: Dorangeon P, Buvat-Herbaut M, Buvat J, Thomas JL. Traitement de l endométriose par l acétate de nomégestrol. Gynecologie 1993;1: Meldrum DR, Chang RJ, Lu J, Valt W, Rivier J, Judd HL. Medical oophorectomy using a long-acting GnRH agonist a possible new approach to the treatment of endometriosis. J Clin Endocrinol Metab 1982;54: Zorn JR, Tanger C, Roger M, Grenier J, Comaru-Schally AM, Schally AV. Therapeutic hypogonadism induced by a delayed-release preparation of microcapsules of D-Tr-6-luteinizing hormone releasing hormone: a preliminary study in eight women with endometriosis. Int J Fertil 1986;31: The American Fertility Society. Revised American Fertility Society classification of endometriosis, Fertil Steril 1985;43: Hodges JL, Lehmann EL. Estimates of location based on rank tests. Ann Math Stat 1963;34: Bruhat MA, Canis M. Diagnostic et traitement actuel de l endométriose externe. Contracep Fertil Sex 1986;14: Telimaa S, Ronnberg L, Kauppila A. A placebo-controlled comparison of danazol and high-dose medroxyprogesterone acetate in treatment of endometriosis after comparative surgery. Gynecol Endocrinol 1987;1: Acosta AA. Medical treatment of endometriosis. In: Hedon B, Bringer J, Mares P, eds. Proceedings of the 15th World Congress on Fertility and Sterility, Montpellier, France, September 17 22, New York: Parthenon, 1995: Hornsteins MD, Hemmings R, Yuzpe AA, Heinrichs WL. Use of nafarelin versus placebo after reductive laparoscopy surgery of endometriosis. Fertil Steril 1997;68: Evers JLH. Endometriosis does not exist; all women have endometriosis. Hum Reprod 1994;9: Barbieri RL, Evans S, Kister RW. Danazol in the treatment of endometriosis: analysis of 100 cases with a 4-year follow-up. Fertil Steril 1982;37: Lemay A, Maheux R, Huot C, Blanchet J, Faure N. Efficacy of intranasal or subcutaneous luteinizing hormone-releasing hormone agonist inhibition of ovarian function in the treatment of endometriosis. Am J Obstet Gynecol 1988;158: Kohler G, Brackmann K. Modification des paramètres du métabolisme lipidique sous traitement de l endométriose par dienogest. N Klin Med 1987;42: Cosson et al. Postoperative treatment of endometriosis Vol. 77, No. 4, April 2002