Advertisement

Long-term treatment of uterine fibroids with ulipristal acetate

Open AccessPublished:March 13, 2014DOI:https://doi.org/10.1016/j.fertnstert.2014.02.008

      Objective

      To investigate the efficacy and safety of ulipristal acetate (UPA) for long-term treatment of symptomatic uterine fibroids.

      Design

      Repeated intermittent open-label UPA courses, each followed by randomized double-blind norethisterone acetate (NETA) or placebo.

      Setting

      European clinical gynecology centers.

      Patient(s)

      Two hundred and nine women with symptomatic fibroids including heavy menstrual bleeding.

      Intervention(s)

      Patients received up to four 3-month courses of UPA 10 mg daily, immediately followed by 10-day double-blind treatment with NETA (10 mg daily) or placebo.

      Main Outcome Measure(s)

      Amenorrhea, fibroid volume, endometrial histology.

      Result(s)

      After the first UPA course, amenorrhea occurred in 79% of women, with median onset (from treatment start) of 4 days (interquartile range, 2–6 days). Median fibroid volume change was −45% (interquartile range, −66%; −25%). Amenorrhea rates were 89%, 88%, and 90% for the 131, 119, and 107 women who received treatment courses 2, 3, and 4, respectively. Median times to amenorrhea were 2, 3, and 3 days for treatment courses 2, 3, and 4, respectively. Median fibroid volume changes from baseline were −63%, −67%, and −72% after treatment courses 2, 3, and 4, respectively. All endometrial biopsies showed benign histology without hyperplasia; NETA did not affect fibroid volume or endometrial histology.

      Conclusion(s)

      Repeated 3-month UPA courses effectively control bleeding and shrink fibroids in patients with symptomatic fibroids.

      Clinical trial registration

      ClinicalTrials.gov (www.clinicaltrials.gov) registration numbers NCT01156857 (PEARL III) and NCT01252069 (PEARL III extension).

      Key Words

      Discuss: You can discuss this article with its authors and other ASRM members at http://fertstertforum.com/donnezj-uterine-fibroids-ulipristal-acetate/
      Leiomyomas or fibroids are benign hormone-sensitive tumors of uterine smooth-muscle cells, frequently involving point mutations and/or complex chromosomal rearrangements (
      • Mehine M.
      • Kaasinen E.
      • Mäkinen N.
      • Katainen R.
      • Kämpjärvi K.
      • Pitkänen E.
      • et al.
      Characterization of uterine leiomyomas by whole-genome sequencing.
      ). They occur in about 20%–40% of women of reproductive age (
      • Wallach E.E.
      • Vlahos N.F.
      Uterine myomas: an overview of development, clinical features, and management.
      ). Heavy menstrual bleeding (HMB), pelvic pressure and pain, and reproductive dysfunction are common symptoms that impair women's health and quality of life (QoL) (
      • Stewart E.A.
      Uterine fibroids.
      ,
      • Donnez J.
      • Jadoul P.
      What are the implications of myomas on fertility? A need for a debate?.
      ). Surgical interventions, especially hysterectomy, still predominate the treatment strategy (
      • Stewart E.A.
      Uterine fibroids and evidence-based medicine—not an oxymoron.
      ). Options for medical therapy are currently limited to preoperative reduction of symptoms related to uterine bleeding and fibroid size; GnRH agonists are licensed but only for short-term therapy owing to safety concerns (loss of bone mass) and adverse reactions (hot flashes) (
      • Donnez J.
      • Schrurs B.
      • Gillerot S.
      • Sandow J.
      • Clerckx F.
      Treatment of uterine fibroids with implants of gonadotropin-releasing hormone agonist: assessment by hysterography.
      ,
      • Lethaby A.
      • Vollenhoven B.
      Fibroids (uterine myomatosis, leiomyomas).
      ). The levonorgestrel-releasing intrauterine device, while not approved for this indication, has been found to reduce menstrual blood loss in women with uterine fibroids, but its efficacy is reduced in patients with a distorted uterus (
      • Zapata L.B.
      • Whiteman M.K.
      • Tepper N.K.
      • Jamieson D.J.
      • Marchbanks P.A.
      • Curtis K.M.
      Intrauterine device use among women with uterine fibroids: a systematic review.
      ). Since February 2012, ulipristal acetate (UPA) is also approved in Europe for preoperative fibroid treatment (
      • Melis G.B.
      • Piras B.
      • Marotto M.F.
      • Orrù M.M.
      • Maricosu G.
      • Pilloni M.
      • et al.
      Pharmacokinetic evaluation of ulipristal acetate for uterine leiomyoma treatment.
      ). For the many women wishing to avoid surgery, there remains a substantial need for effective long-term medical therapy.
      UPA is a selective P receptor modulator (SPRM) that potently modulates P-receptor activity (
      • Gainer E.E.
      • Ulmann A.
      Pharmacologic properties of CDB(VA)-2914.
      ) with proapoptotic/antiproliferative effects on fibroid cells (
      • Horak P.
      • Mara M.
      • Dundr P.
      • Kubinova K.
      • Kuzel D.
      • Hudecek R.
      • et al.
      Effect of a selective progesterone receptor modulator on induction of apoptosis in uterine fibroids in vivo.
      ) and with pharmacokinetic properties supporting once daily dosing (
      • Pohl O.
      • Osterloh I.
      • Gotteland J.P.
      Ulipristal acetate—safety and pharmacokinetics following multiple doses of 10–50 mg per day.
      ). Two short-term (3 months) randomized clinical trials showed that UPA effectively controls HMB and shrinks fibroids (
      • Donnez J.
      • Tatarchuk T.F.
      • Bouchard P.
      • Puscasiu L.
      • Zakharenko N.F.
      • Ivanova T.
      • et al.
      Ulipristal acetate versus placebo for fibroid treatment before surgery.
      ,
      • Donnez J.
      • Tomaszewski J.
      • Vazquez F.
      • Bouchard P.
      • Lemieszczuk B.
      • Baro F.
      • et al.
      Ulipristal acetate versus leuprolide acetate for uterine fibroids.
      ). After treatment cessation, menstruation usually returns within 4–5 weeks, but fibroid volume reduction can be sustained for up to 6 months. In addition, treatment with UPA reduced fibroid-associated pain, improved QoL, and revealed no safety concerns (
      • Donnez J.
      • Tatarchuk T.F.
      • Bouchard P.
      • Puscasiu L.
      • Zakharenko N.F.
      • Ivanova T.
      • et al.
      Ulipristal acetate versus placebo for fibroid treatment before surgery.
      ,
      • Donnez J.
      • Tomaszewski J.
      • Vazquez F.
      • Bouchard P.
      • Lemieszczuk B.
      • Baro F.
      • et al.
      Ulipristal acetate versus leuprolide acetate for uterine fibroids.
      ). Clinical trials have also shown that SPRM administration can lead to a pattern of benign, nonphysiological, nonproliferative, histological features of the endometrium termed P receptor modulator associated endometrial changes (PAEC) (
      • Mutter G.L.
      • Bergeron C.
      • Deligdisch L.
      • Ferenczy A.
      • Glant M.
      • Merino M.
      • et al.
      The spectrum of endometrial pathology induced by progesterone receptor modulators.
      ,
      • Fiscella J.
      • Bonfiglio T.
      • Winters P.
      • Eisinger S.H.
      • Fiscella K.
      Distinguishing features of endometrial pathology after exposure to the progesterone receptor modulator mifepristone.
      ,
      • Ioffe O.B.
      • Zaino R.J.
      • Mutter G.L.
      Endometrial changes from short-term therapy with CDB-4124, a selective progesterone receptor modulator.
      ). These changes spontaneously reverse over a few weeks to months after cessation of the 3-month UPA treatment (
      • Donnez J.
      • Tatarchuk T.F.
      • Bouchard P.
      • Puscasiu L.
      • Zakharenko N.F.
      • Ivanova T.
      • et al.
      Ulipristal acetate versus placebo for fibroid treatment before surgery.
      ,
      • Donnez J.
      • Tomaszewski J.
      • Vazquez F.
      • Bouchard P.
      • Lemieszczuk B.
      • Baro F.
      • et al.
      Ulipristal acetate versus leuprolide acetate for uterine fibroids.
      ,
      • Williams A.R.
      • Bergeron C.
      • Barlow D.H.
      • Ferenczy A.
      Endometrial morphology after treatment of uterine fibroids with the selective progesterone receptor modulator, ulipristal acetate.
      ). Hence, intermittent courses of 3-month UPA treatment with off-treatment intervals are a potential option for the long-term medical management of fibroids (
      • Melis G.B.
      • Piras B.
      • Marotto M.F.
      • Orrù M.M.
      • Maricosu G.
      • Pilloni M.
      • et al.
      Pharmacokinetic evaluation of ulipristal acetate for uterine leiomyoma treatment.
      ).
      In these two studies, the PGL4001 (UPA) Efficacy Assessment in Reduction of Symptoms due to Uterine Leiomyomata (PEARL) III trial and its extension, we evaluated the sustained effects of UPA on menstrual bleeding, fibroid volume, pain, QoL, and safety during one to four 3-month UPA treatment courses.
      Owing to the long-term treatment, no suitable active comparator to UPA was available. However, since UPA exerts mainly antiprogestagenic effects on the endometrium, we randomized women to receive 10 days of treatment with the progestin norethisterone acetate (NETA) or placebo (administered immediately after each completed UPA treatment) to explore any effect on the reversibility of PAEC or timing and magnitude of the next menstruation off treatment. The off-treatment period between each UPA course included one menstrual bleed and the beginning of a second bleed.

      Materials and methods

       Study Design and Oversight

      PEARL III and its extension were long-term, open-label, phase III trials of UPA, which were double-blinded and placebo-controlled toward the administration of progestin after the end of each UPA treatment course. PEARL III was conducted at 21 investigation centers in four countries from July 2010 through November 2011, with 18 centers also participating in the extension protocol until January 2013. The trial and extension were approved by the independent ethics committee of each participating site and were conducted in accordance with the International Conference on Harmonisation Good Clinical Practice guidelines.

       Study Population

      PEARL III enrolled premenopausal women with at least one fibroid ≥3 cm in diameter and none >10 cm, HMB, and uterine size <16 weeks of gestation who were eligible for fibroid surgery. Eligible women were aged 18–48 years, with body mass index 18–40 (kg/m2) and regular menstrual cycles of 22–35 days with FSH ≤20 IU/L. Written informed consent was obtained from all women. The main exclusion criteria are listed in Supplemental Table 1, available at www.fertstert.org.

       Randomization and Intervention

      Women received a 3-month open-label course of UPA (10 mg) once daily immediately followed by double-blind oral NETA (10 mg) once daily or matching placebo for 10 days allocated randomly in a 1:1 ratio.
      UPA was started during the first 4 days of menstruation. After the first UPA (and NETA/placebo) treatment course and the return of menstruation, women could either leave the study and attend a final follow-up visit 12 weeks later (PEARL III) or, if they wished to be assessed for a further 18 months, enroll in the PEARL III extension study to obtain up to three further courses of UPA (and NETA/placebo), each separated by an off-treatment period including a full menstrual cycle up to the start of the second menstruation. Follow-up visits in the extension study were conducted approximately 3 months after the final treatment course. The sequence of treatments is illustrated in Supplemental Figure 1.

       Efficacy endpoints

      We evaluated efficacy and safety endpoints after the first treatment course (PEARL III), after each of up to three more treatment courses (PEARL III extension), and approximately 3 months after completion of the final treatment course. The primary efficacy endpoint was the occurrence of amenorrhea at the end of each UPA course. In accordance with previously published data, amenorrhea was defined as no bleeding for a continuous period of at least 35 days (1 day of spotting was allowed within any 35-day interval) to accurately describe the degree of bleeding control during UPA treatment under the conditions of this study (
      • Donnez J.
      • Tatarchuk T.F.
      • Bouchard P.
      • Puscasiu L.
      • Zakharenko N.F.
      • Ivanova T.
      • et al.
      Ulipristal acetate versus placebo for fibroid treatment before surgery.
      ,
      • Donnez J.
      • Tomaszewski J.
      • Vazquez F.
      • Bouchard P.
      • Lemieszczuk B.
      • Baro F.
      • et al.
      Ulipristal acetate versus leuprolide acetate for uterine fibroids.
      ). This differs from the general definition of amenorrhea, which is three menstrual cycle lengths of no bleeding. Bleeding was assessed using a semiquantitative bleeding scale. For an exploratory endpoint, the Pictorial Blood-Loss Assessment Chart (PBAC) (
      • Higham J.M.
      • O'Brien P.M.
      • Shaw R.W.
      Assessment of menstrual blood loss using a pictorial chart.
      ) was used to assess the magnitude of menstrual bleeding over 8 days at baseline (start of the first treatment course) and for the first menstruation after the end of each treatment course. A score greater than 100 indicates HMB.
      Secondary efficacy endpoints included the reduction of the three largest fibroids identified at screening on the basis of ultrasound performed at each center. Further secondary endpoints were pain, measured with the Short-Form McGill Pain questionnaire (
      • Melzack R.
      The short-form McGill Pain Questionnaire.
      ) and QoL, measured by the general EQ-5D questionnaire (
      • Dolan P.
      Modeling valuations for EuroQol health states.
      ) and by the specific Uterine Fibroid Symptom and Health-Related Quality of Life (UFS-QoL) questionnaire (
      • Spies J.B.
      • Coyne K.
      • Guaou G.N.
      • Boyle D.
      • Skyrnarz-Murphy K.
      • Gonzalves S.M.
      The UFS-QOL, a new disease-specific symptom and health-related quality of life questionnaire for leiomyomata.
      ). Additional details are presented in the Supplemental Material.

       Assessment of Safety Endpoints

      Safety endpoints included the number and proportion of women experiencing treatment-emergent adverse events (TEAEs), including clinically significant changes in vital signs, physical examination, gynecological and/or breast examination, electrocardiogram (ECG), ovarian ultrasound, changes from baseline in endometrial thickness, and clinically significant changes in endometrial biopsy. The frequency and severity of adverse events (spontaneously reported or elicited by the investigators with the use of nonleading questions) were recorded on standard forms at every visit. Serious adverse events were recorded up to the last visit approximately 3 months after treatment cessation. Other safety endpoints comprised changes from baseline in hematology, coagulation, biochemistry, lipids, ACTH, TSH, and PRL, as well as serum levels of E2.

       Endometrial Histology

      Endometrial biopsy samples were obtained 10–18 days after menstruation start during screening, after treatment courses 1 and 4, and 3 months after the end of course 4 in women with PAEC observed after course 4. All samples were assessed in a blinded manner by three independent pathologists experienced in PAEC.

       Statistical Analysis

      This study included open-label UPA treatment phases and placebo-controlled double-blinded NETA phases where assessments were considered exploratory. We planned to enroll 200 women, with approximately 80% of participants expected to fulfill the efficacy endpoint of amenorrhea at the end of the first UPA treatment course and allowing for a dropout rate of 10% during the first course of treatment. The planned sample size would provide an estimate of this percentage, with the corresponding 95% confidence interval (CI) being less than ±6%. We considered 90 evaluable participants per treatment group to be sufficient to enable a reliable exploration of any differences between NETA and placebo treatment. We conducted efficacy and safety analyses using all women who started treatment course 1 (PEARL III open-label set) and two (PEARL III extension full analysis set). Endpoints were summarized using descriptive statistics. A 95% CI for the percentage of amenorrheic women was produced using the Wilson score method. The time from the start of dosing to amenorrhea was summarized as a continuous variable and presented using the Kaplan Meier probability estimates. The Wilcoxon rank sum test was performed to compare the placebo versus NETA treatment groups for the time to return to menstruation.

      Results

       Patients

      The baseline characteristics of the 209 participants who entered the study, the 132 who entered the extension study, and the 107 who received four treatment courses were very similar; most women were in their late 30s or early 40s with moderate to severe bleeding (the overall median PBAC was >200), and self-reported pain or discomfort and/or anxiety or depression (Table 1). Adverse events (eight women) and lack of efficacy (four women) were infrequent causes of study discontinuation (Supplemental Figure 2, Supplemental Figure 3).
      Table 1Baseline (pretreatment course 1) characteristics for participants entering PEARL III and extension studies.
      CharacteristicPEARL III only: open-label set (baseline data for women who started course 1), n = 209PEARL III and extension study: full analysis set (baseline data for women who started at least two courses), n = 132PEARL III and extension study: (baseline data for women who started all four courses), n = 107
      Age, mean ± SD40.1 ± 6.040.5 ± 5.840.8 ± 5.5
      Race, n (%)
      Race or ethnic group was reported by the investigator.
       White179 (85.6)121 (91.7)99 (92.5)
       Black19 (9.1)8 (6.1)5 (4.7)
       Other11 (5.3)3 (2.3)3 (2.8)
      Body mass index, kg/m2, mean ± SD25.4 ± 4.425.4 ± 4.725.1 ± 4.6
      Median (IQR) PBAC (days 1–8) score
      PBAC, measured during days 1–8 at the start of the UPA treatment (course 1).
      216 (126, 401)235 (142, 397)234 (132, 368)
      Median (IQR) total volume of three largest fibroids, cm353.9 (24.0, 128.7) (n = 207)56.2 (25.7, 128.8) (n = 132)49.8 (26.9, 112.8) (n = 107)
      Median (IQR) uterine volume, cm3199.6 (125.2, 291.3) (n = 209)200.7 (121.5, 303.1) (n = 132)178.6 (116.6, 278.7) (n = 107)
      Hemoglobin, g/dL, mean ± SD12.5 ± 1.8 (n = 199)12.4 ± 1.8 (n = 127)12.5 ± 1.7 (n = 105)
      Short-Form McGill Pain questionnaire
      Scores on the Short-Form McGill Pain questionnaire range from 0 to 45, with higher scores indicating more severe pain.
      assessment of pain, median (IQR)
      8.0 (3.0, 17.0) (n = 207)8.0 (4.0, 19.0) (n = 131)8.0 (4.0, 19.0) (n = 106)
      Visual analog scale (VAS),
      Scores on VAS range from 0 to 100, with higher scores indicating more severe pain.
      median (IQR)
      38.0 (17.0, 63.0) (n = 209)36.0 (17.0, 63.5) (n = 132)37.0 (16.0, 63.0) (n = 107)
      UFS-QoL questionnaire
      On the UFS-QoL questionnaire, scores for symptom severity range from 0 to 100, with higher scores indicating increased severity. Total scores for health-related QoL range from 0 to 100, with higher scores indicating a better QoL.
       Symptom severity, mean ± SD47.7 ± 17.7 (n = 207)49.2 ± 17.7 (n = 131)49.0 ± 16.6 (n = 107)
       Health-related QoL total score, mean ± SD57.1 ± 20.9 (n = 208)54.9 ± 20.6 (n = 132)55.3 ± 19.9 (n = 107)
      EQ-5D questionnairen = 209n = 132n = 107
       Mobility (women with problems), n (%)34 (16.3)19 (14.4)15 (14.0)
       Self-care (women with problems), n (%)5 (2.4)2 (1.5)1 (0.9)
       Usual activities (women with problems), n (%)50 (23.9)26 (19.7)22 (20.6)
       Pain/discomfort (women with moderate or extreme symptoms), n (%)154 (73.7)102 (77.3)84 (78.5)
       Anxiety/depression (women with moderate or extreme symptoms), n (%)129 (62.0
      n = 208.
      )
      85 (64.9
      n = 131.
      )
      74 (69.2)
       VAS (health state),
      Scores on VAS range from 0 to 100, with higher scores indicating a better health state.
      mean ± SD
      67.5 ± 18.667.9 ± 18.568.2 ± 17.6
      Note: n = Number of women with nonmissing observations.
      a Race or ethnic group was reported by the investigator.
      b PBAC, measured during days 1–8 at the start of the UPA treatment (course 1).
      c Scores on the Short-Form McGill Pain questionnaire range from 0 to 45, with higher scores indicating more severe pain.
      d Scores on VAS range from 0 to 100, with higher scores indicating more severe pain.
      e On the UFS-QoL questionnaire, scores for symptom severity range from 0 to 100, with higher scores indicating increased severity. Total scores for health-related QoL range from 0 to 100, with higher scores indicating a better QoL.
      f n = 208.
      g n = 131.
      h Scores on VAS range from 0 to 100, with higher scores indicating a better health state.

       Efficacy

       Menstrual bleeding

      At the end of the first treatment course, 164 women (78.5%) were amenorrheic (95% CI, 72.4%–83.5%; Supplemental Table 2). The median time to amenorrhea from treatment start was 3.5 days (interquartile range [IQR], 2–6 days).
      For the 132 women who entered the extension study to receive multiple treatment courses, 88.5%, 88.2%, and 89.7% were in amenorrhea at the end of courses 2, 3, and 4, respectively (Table 2). The median times to amenorrhea after the start of each course were 2, 3, and 3 days for courses 2, 3, and 4, respectively (Fig. 1A). The percentages of women with only spotting or no bleeding were 93.9%, 94.1%, and 93.5% at the end of courses 2, 3, and 4, respectively.
      Table 2Efficacy results for participants entering PEARL III extension study (PEARL III extension full analysis set).
      Amenorrhea (and spotting or no bleeding) assessed while under UPA treatment. For remaining results, courses 1 and 4 data were collected at the end of UPA treatment; courses 2 and 3 data were collected after the first menstrual bleed after UPA treatment and subsequent NETA/placebo treatment.
      Course 1Course 2Course 3Course 43 Month follow-up
      Amenorrhea, n/N (%)105/132 (79.5)116/131 (88.5)105/119 (88.2)96/107 (89.7)
       95% CI, %71.9, 85.582.0, 92.981.2, 92.982.5, 94.2
      Spotting or no bleeding, n/N (%)117/132 (88.6)123/131 (93.9)112/119 (94.1)100/107 (93.5)
      % Change in total volume of three largest fibroids
      The same three fibroids identified at screening were followed throughout the study.
      from baseline, median (IQR)
      −49.9 (−69.0, −27.2) (n = 130)−63.2 (−76.4, −38.3) (n = 119)−67.0 (−79.9, −33.5) (n = 106)−72.1 (−86.6, −35.7) (n = 96)−58.8 (79.2, −21.0) (n = 97)
       Total reduction ≥25%, n (%)101 (77.7)95 (79.8)83 (78.3)79 (82.3)70 (72.2)
       Total reduction ≥50%, n (%)65 (50.0)77 (64.7)66 (62.3)67 (69.8)56 (57.7)
      % Change in uterine volume from baseline, median (IQR)−29.8 (−45.2, −10.5) (n = 132)−32.3 (−47.1, 0.7) (n = 121)−29.9 (−47.5, −1.8) (n = 107)−40.2 (−55.6, −15.3) (n = 96)−22.3 (−46.2, 5.5) (n = 99)
       Reduction ≥25%, n (%)73 (55.3)73 (60.3)61 (57.0)64 (66.7)45 (45.5)
      Assessment of pain
       Short-Form McGill Pain questionnairen = 131n = 119n = 108n = 96n = 96
      Actual, median (IQR)1.0 (0.0, 3.0)1.0 (0.0, 4.0)1.0 (0.0, 6.0)1.0 (0.0, 2.0)2.0 (0.0, 7.0)
      Change from baseline, median (IQR)−7.0 (−15.0, −2.0)−6.0 (−14.0, −2.0)−5.0 (−13.5, 0.0)−6.0 (−16.0, −2.0)−4.0 (−11.0, 0.0)
       VAS
      Scores on VAS range from 0 to 100, with higher scores indicating more severe pain.
      n = 132n = 120n = 109n = 96n = 98
      Actual, median (IQR)1.0 (0.0, 11.0)3.0 (0.0, 13.5)2.0 (0.0, 13.0)1.5 (0.0, 8.5)7.0 (1.0, 29.0)
      Change from Baseline, median (IQR)−23.5 (−58.0, −6.5)−27.5 (−54.0, −7.0)−25.0, (−49.0, −6.0)−30.5 (−54.0, −10.0)−17.0 (−42.0, −2.0)
      UFS-QoL questionnaire
      On the UFS-QoL questionnaire, scores for symptom severity range from 0 to 100, with higher scores indicating increased severity. Total scores for health-related QoL range from 0 to 100, with higher scores indicating a better QoL.
       Symptom severityn = 129n = 117n = 104n = 91n = 98
      Actual, mean ± SD13.4 ± 15.318.4 ± 16.820.5 ± 19.517.9 ± 17.127.1 ± 21.1
      Change from baseline, mean ± SD−35.8 ± 21.2−30.5 ± 21.9−27.7 ± 23.3−30.0 ± 20.3−21.2 ± 22.1
       Health-related QoL total scoren = 131n = 117n = 108n = 96n = 99
      Actual, mean ± SD87.8 ± 14.885.2 ± 16.485.2 ± 18.187.5 ± 16.279.2 ± 22.9
      Change from baseline, mean ± SD32.8 ± 21.929.6 ± 22.429.4 ± 23.431.4 ± 21.622.7 ± 22.5
      EQ-5D questionnairen = 132n = 121n = 109n = 96n = 99
       Mobility (women with problems), n (%)6 (4.5)3 (2.5)5 (4.6)4 (4.2)3 (3.0)
       Self-care (women with problems), n (%)001 (0.9)00
       Usual activities (women with problems), n (%)6 (4.5)3 (2.5)4 (3.7)3 (3.1)7 (7.1)
       Pain/discomfort (women with moderate or extreme symptoms), n (%)33 (25.2
      n = 131.
      )
      31 (25.6)26 (24.1
      n = 108.
      )
      23 (24.0)37 (37.4)
       Anxiety/depression (women with moderate or extreme symptoms), n (%)49 (37.1)27 (22.3)33 (30.3)29 (30.2)37 (37.4)
       VAS (health state)
      Scores on VAS range from 0 to 100, with higher scores indicating a better health state.
      n = 132n = 120n = 109n = 95n = 99
      Actual, mean ± SD78.1 ± 18.081.2 ± 14.983.4 ± 13.584.4 ± 15.584.1 ± 14.9
      Change from baseline, mean ± SD10.2 ± 23.813.3 ± 21.315.6 ± 20.215.2 ± 20.314.8 ± 18.9
      Note: n = Number of women with nonmissing observations.
      a Amenorrhea (and spotting or no bleeding) assessed while under UPA treatment. For remaining results, courses 1 and 4 data were collected at the end of UPA treatment; courses 2 and 3 data were collected after the first menstrual bleed after UPA treatment and subsequent NETA/placebo treatment.
      b The same three fibroids identified at screening were followed throughout the study.
      c Scores on VAS range from 0 to 100, with higher scores indicating more severe pain.
      d On the UFS-QoL questionnaire, scores for symptom severity range from 0 to 100, with higher scores indicating increased severity. Total scores for health-related QoL range from 0 to 100, with higher scores indicating a better QoL.
      e n = 131.
      f n = 108.
      g Scores on VAS range from 0 to 100, with higher scores indicating a better health state.
      Figure thumbnail gr1
      Figure 1(A) Time to no bleeding (persistent amenorrhea; PEARL III extension full analysis set). Time to amenorrhea was defined as the first day for which there was subsequently no bleeding for longer than 35 days to the end of UPA treatment within each treatment course, assessed using the patient diary data from the date of the first dose of UPA (day 0, which was to be within the first 4 days of the start of menstruation for treatment courses 1 and 2 and on the first day of menstruation for treatment courses 3 and 4). One day of spotting in any 35-day interval was accepted. Circles denote censored observation (i.e., a participant had a subsequent interval of 35 days or less up to the end of UPA treatment for which no more than 1 day of spotting was observed). (B) Percent change from baseline in total volume of the three largest fibroids (PEARL III extension full analysis set). N = number of women with nonmissing observations. The same three fibroids identified at screening were followed throughout the study. Box extends to 25th and 75th percentiles (IQR) and shows median value. Mean values are also plotted. Whiskers extend to 10th and 90th percentiles. Courses 1 and 4 data were collected at the end of UPA treatment; courses 2 and 3 data were collected after the first menstrual bleed after UPA treatment and subsequent NETA/placebo treatment.
      After the end of each treatment course, menstruation resumed. Menstrual bleeding (PBAC days 1–8) progressively reduced from medians of 228 and 257 at the start of the first course to 55 and 13 after the end of the fourth UPA course for women randomized to placebo or NETA, respectively (P=.02; Supplemental Table 3). Thus, 10-day progestin courses reduced the magnitude of menstrual bleeding during the off-treatment periods and also brought forward menstruation return (e.g., median of 15 days instead of 30 days for women receiving placebo after the end of the fourth UPA treatment course; P<.001).

       Fibroid volume

      The median change from baseline to end of the first UPA treatment course in the combined volume of the three largest fibroids was −45.1% (IQR, −66.1 to −24.9%; Supplemental Table 2). For women receiving multiple treatment courses, fibroids continued to shrink, reaching a median volume reduction of −72.1% after four treatment courses (Table 2, Fig. 1B). This volume reduction was mostly maintained (median −58.8%) at follow-up 3 months after the end of the fourth treatment course. Of women completing four treatment courses, 82.3% and 69.8%, respectively, had ≥25% and ≥50% reductions in volume of the three largest fibroids.

       Pain and QoL

      During the first treatment course, improvements in pain were apparent from the fifth week onward and were generally maintained for all UPA treatment periods (Table 2, Supplemental Table 2). UFS-QoL scores indicated substantially reduced QoL at baseline, but mean scores were within the range of healthy participants at the end of each treatment course and the improvement was largely maintained at 3-month follow-up after the final treatment course. The EQ-5D questionnaire showed that most women had discomfort or pain and/or anxiety and depression at baseline, but few women reported these conditions at the end of each UPA treatment course and at 3-month follow-up (Tables 1 and 2).

       Surgery and enrollment in extension study

      A total of 132/209 patients consented and proceeded to the extension phase of the study. Among them were 64/91 women for whom surgery was planned at baseline (Supplemental Fig. 2). There were no significant differences in baseline disease severity between women who entered the extension study and those who did not.
      After one course of treatment, UPA exerted similar control over bleeding and pain for patients entering the extension study (compared with those who did not). Women entering the extension study had a somewhat greater reduction in median fibroid volume (−49.9% vs. −38.5%) and greater improvement in UFS-QoL (least square mean change, 31.2 vs. 25.3) after one course of UPA compared with those who did not enter the second study (Supplemental Table 4).

       Safety

       General safety

      There were no serious adverse events (SAEs) reported during the first course of UPA treatment, but two women had three post-treatment SAEs (two episodes of uterine bleeding and one breast cancer) that were considered not related to study medication. Seven women who entered the extension study (and received multiple courses of treatment) reported SAEs (five cases of uterine bleeding, one thyroid cyst, and one chlamydia infection; Table 3).
      Table 3Adverse events (PEARL III extension safety set).
      All SAEs and all adverse events that occurred during UPA treatment in at least 4% of women overall are included.
      Adverse event, no. of women (%)Overall (n = 132)Course 1 (n = 132)Course 2 (n = 131)Course 3 (n = 119)Course 4 (n = 107)
      SAEs
       At least one SAE7 (5.3)02 (1.5)2 (1.7)3 (2.8)
       Any SAE during UPA treatment6 (4.5)02 (1.5)2 (1.7)2 (1.9)
      HMB2 (1.5)01 (0.8)01 (0.9)
      Uterine hemorrhage2 (1.5)01 (0.8)01 (0.9)
      Metrorrhagia1 (0.8)001 (0.8)0
      Thyroid cyst1 (0.8)001 (0.8)0
       Any SAE during NETA/placebo treatment00000
       Any SAE off treatment1 (0.8)0001 (0.9)
      Chlamydial infection1 (0.8)0001 (0.9)
      SAE was diagnosed 3 months after treatment course 4.
      Adverse events
      Adverse events with onset on or after the first dose of UPA and before the first dose of NETA/placebo within each treatment course or up to and including 7 days after the last dose of UPA if NETA/placebo was never started.
      Leading to study withdrawal
      In addition, another adverse event that occurred when a woman was not receiving UPA led to study withdrawal.
      5 (3.8)01 (0.8)2 (1.7)2 (1.9)
      At least one event91 (68.9)73 (55.3)27 (20.6)35 (29.4)37 (34.6)
       Headache26 (19.7)19 (14.4)4 (3.1)6 (5.0)7 (6.5)
       Nasopharyngitis18 (13.6)10 (7.6)3 (2.3)1 (0.8)6 (5.6)
       Abdominal pain (including upper/lower)12 (9.1)7 (5.3)3 (2.3)1 (0.8)2 (1.9)
       Hot flashes12 (9.1)7 (5.3)1 (0.8)5 (4.2)1 (0.9)
       Back pain8 (6.1)2 (1.5)04 (3.4)2 (1.9)
       Fatigue8 (6.1)4 (3.0)03 (2.5)3 (2.8)
       Nausea8 (6.1)4 (3.0)2 (1.5)1 (0.8)1 (0.9)
       Vertigo7 (5.3)6 (4.5)01 (0.8)1 (0.9)
       Hair loss6 (4.5)5 (3.8)1 (0.8)2 (1.7)0
       Breast discomfort/breast pain/breast tenderness6 (4.5)4 (3.0)01 (0.8)1 (0.9)
      a All SAEs and all adverse events that occurred during UPA treatment in at least 4% of women overall are included.
      b SAE was diagnosed 3 months after treatment course 4.
      c Adverse events with onset on or after the first dose of UPA and before the first dose of NETA/placebo within each treatment course or up to and including 7 days after the last dose of UPA if NETA/placebo was never started.
      d In addition, another adverse event that occurred when a woman was not receiving UPA led to study withdrawal.
      During the first course of UPA treatment, TEAEs occurred in 120 women (57.4%; Supplemental Table 5), but only eight women (3.8%) had severe AEs. Only one TEAE (headache) led to treatment withdrawal. TEAEs occurring in >5% of women were headache (16.3%), nasopharyngitis (6.7%), and abdominal pain (5.3%). In women receiving multiple treatment courses, headaches, nasopharyngitis, abdominal pain, and hot flashes were the most frequent TEAEs, but the incidence did not increase over time and only five women discontinued UPA because of TEAEs (Table 3).
      There were no safety signals arising from physical examination, vital signs, liver function, and other laboratory safety tests, hormone levels, ovarian ultrasound, and ECGs (see Supplemental Table 6).

       Endometrial safety

      Transient increases in endometrial thickness occurred in less than 10% of women after each UPA course (Supplemental Table 6). No cases of endometrial hyperplasia or adenocarcinoma were reported at any time point for any woman.
      Nonphysiological features were reported by at least two pathologists for 18/171 (11%), 45/176 (26%), and 22/87 (25%) women biopsied at screening and at approximately 6 weeks after courses 1 and 4, respectively (Supplemental Table 7). Fifteen women with PAEC diagnosed after the fourth UPA course were rebiopsied 3 months later; only three had some nonphysiological features. Ten days of NETA did not have a significant impact on the incidence of nonphysiological changes induced by UPA.

      Discussion

      In these sequential studies of women with fibroids and excessive menstrual bleeding, we administered intermittent courses of UPA. Women had the option of surgery after one course, but for many women the goal is to avoid surgery and most decided to enroll into the extension study where they could receive up to three additional courses even if surgery was planned at baseline. Intermittent UPA (administered over an 18-month period) induced high rates of amenorrhea (nearly 90% during repeat treatment courses), confirming its ability to control for the long term the most troublesome symptom of fibroids for most women. The previously reported efficacy of UPA in shrinking fibroids was also confirmed, and there was further fibroid shrinkage with successive treatment courses with no evidence of rapid rebound growth. However, even in women with little or no fibroid shrinkage, UPA still induced high rates of bleeding control (Supplemental Table 8). Women also reported substantial improvements in pain, anxiety and depression, and QoL during treatment. The UFS-QoL scores at baseline were slightly less severe than reported in some previous studies of patients undergoing invasive treatments; however, at the end of each UPA course, symptom severity and QoL scores were similar to those reported at follow-up for patients who underwent hysterectomy, myomectomy, uterine artery embolization, high-intensity focused ultrasound, or experimental treatment with mifepristone (
      • Fiscella K.
      • Eisinger S.H.
      • Meldrum S.
      • Feng C.
      • Fisher S.G.
      • Guzick D.S.
      Effect of mifepristone for symptomatic leiomyomata on quality of life and uterine size: a randomized controlled trial.
      ,
      • Spies J.B.
      • Bradley L.D.
      • Guido R.
      • Maxwell G.L.
      • Levine B.A.
      • Coyne K.
      Outcomes from leiomyoma therapies: comparison with normal controls.
      ,
      • Takeda T.
      • Osuga K.
      • Miyake A.
      • Wakabayashi A.
      • Morishige K.
      • Kimura T.
      Elevated level of plasma vascular endothelial growth factor after gonadotropin-releasing hormone agonist treatment for leiomyomata.
      ,
      • Kim H.S.
      • Baik J.H.
      • Pham L.D.
      • Jacobs M.A.
      MR-guided high-intensity focused ultrasound treatment for symptomatic uterine leiomyomata: long-term outcomes.
      ).
      It might have been expected that fibroid-related symptoms would partially return during off-treatment periods, but the magnitude of menstrual bleeding (after the end of each treatment course) progressively diminished. Moreover, at 3 months after the last treatment course, the reduction in fibroid volume and the improvements in pain and QoL were largely maintained. These findings, together with previous observations (
      • Horak P.
      • Mara M.
      • Dundr P.
      • Kubinova K.
      • Kuzel D.
      • Hudecek R.
      • et al.
      Effect of a selective progesterone receptor modulator on induction of apoptosis in uterine fibroids in vivo.
      ) that UPA can induce apoptosis and decrease proliferation in fibroid cells, suggest that long-term intermittent SPRM therapy could result in disease regression.
      Most adverse events were mild or moderate and did not increase in frequency with successive treatment courses. The most frequently reported SAE was excessive uterine bleeding, but considering that women were monitored for over 18 months, the incidence of this finding is not greater than that reported in previous studies of UPA, placebo, and comparator agents in women with fibroid-related HMB (
      • Donnez J.
      • Tatarchuk T.F.
      • Bouchard P.
      • Puscasiu L.
      • Zakharenko N.F.
      • Ivanova T.
      • et al.
      Ulipristal acetate versus placebo for fibroid treatment before surgery.
      ,
      • Donnez J.
      • Tomaszewski J.
      • Vazquez F.
      • Bouchard P.
      • Lemieszczuk B.
      • Baro F.
      • et al.
      Ulipristal acetate versus leuprolide acetate for uterine fibroids.
      ).
      Some of the nonphysiological features of PAEC can be observed in women of reproductive age in the absence of pharmacotherapy (>10% incidence at screening in this study) (
      • Williams A.R.
      • Bergeron C.
      • Barlow D.H.
      • Ferenczy A.
      Endometrial morphology after treatment of uterine fibroids with the selective progesterone receptor modulator, ulipristal acetate.
      ). Donnez et al. previously reported that 3-month courses of UPA induce PAEC in approximately 60% of women and that this is fully reversible 6 months after the end of the treatment (
      • Donnez J.
      • Tatarchuk T.F.
      • Bouchard P.
      • Puscasiu L.
      • Zakharenko N.F.
      • Ivanova T.
      • et al.
      Ulipristal acetate versus placebo for fibroid treatment before surgery.
      ,
      • Donnez J.
      • Tomaszewski J.
      • Vazquez F.
      • Bouchard P.
      • Lemieszczuk B.
      • Baro F.
      • et al.
      Ulipristal acetate versus leuprolide acetate for uterine fibroids.
      ). In this study, the endometrium was biopsied approximately 6 weeks after the end of the first and fourth courses of UPA treatment, and PAEC was diagnosed in approximately 25% of women at both time points. Similar incidences of PAEC after one and four courses of treatment indicate that treatment duration and cumulative dose do not affect occurrence of PAEC during prolonged, intermittent UPA administration. In the small subgroup of women with PAEC diagnosed after the fourth UPA course, the PAEC diagnosis rate at 3 months follow-up was the same as at screening, confirming that PAEC is rapidly reversible in most women after successive on/off administration. No cases of endometrial hyperplasia or adenocarcinoma were observed.
      A 10-day course of progestin, administered immediately after each UPA treatment course, did not affect PAEC but was associated with a significantly reduced and an earlier occurrence of menstrual bleeding during the off-treatment periods. However, these findings are probably insufficient to suggest that progestins should be routinely used in conjunction with UPA treatment unless there is a need to control the timing or amount of menstrual bleeding, for example, before a planned invasive procedure.
      Currently, there are no approved long-term medical treatment options for the management of women with symptomatic fibroids. Some clinicians have attempted to use GnRH agonists with hormonal add-back therapy, but women still have menopausal symptoms, rates of bone mass loss may not be entirely mitigated, and fibroids rapidly enlarge after treatment is stopped (
      • Lethaby A.
      • Vollenhoven B.
      Fibroids (uterine myomatosis, leiomyomas).
      ). There are abundant P receptors in fibroids, and oral progestins (alone) have been used but without evidence that they shrink fibroids and with risk of breakthrough bleeding (
      • Kim J.J.
      • Sefton E.C.
      The role of progesterone signaling in the pathogenesis of uterine leiomyoma.
      ,
      • Scialli A.R.
      • Jestila K.J.
      Sustained benefits of leuprolide acetate with or without subsequent medroxyprogesterone acetate in the nonsurgical management of leiomyomata uteri.
      ). Intrauterine levonorgestrel does not consistently reduce fibroid volume, and its efficacy in women with submucous fibroids or a distorted uterus is controversial (
      • Zapata L.B.
      • Whiteman M.K.
      • Tepper N.K.
      • Jamieson D.J.
      • Marchbanks P.A.
      • Curtis K.M.
      Intrauterine device use among women with uterine fibroids: a systematic review.
      ). Thus our results suggest that intermittent UPA could become the first long-term medical management option for many women with symptomatic fibroids.
      There are some limitations to our study. We could not use an active or placebo control for long-term UPA treatment. However, it was previously demonstrated that a 3-month course of UPA was superior to placebo and noninferior to a GnRH agonist for control of HMB (
      • Donnez J.
      • Tatarchuk T.F.
      • Bouchard P.
      • Puscasiu L.
      • Zakharenko N.F.
      • Ivanova T.
      • et al.
      Ulipristal acetate versus placebo for fibroid treatment before surgery.
      ,
      • Donnez J.
      • Tomaszewski J.
      • Vazquez F.
      • Bouchard P.
      • Lemieszczuk B.
      • Baro F.
      • et al.
      Ulipristal acetate versus leuprolide acetate for uterine fibroids.
      ). The dose (10 mg) and duration of each UPA course (3 months) were based on previous experience, but it is unknown whether longer periods of continuous treatment could also be safe and effective. We also recruited relatively few black participants, but previous studies reported efficacy in these women (
      • Donnez J.
      • Tatarchuk T.F.
      • Bouchard P.
      • Puscasiu L.
      • Zakharenko N.F.
      • Ivanova T.
      • et al.
      Ulipristal acetate versus placebo for fibroid treatment before surgery.
      ,
      • Donnez J.
      • Tomaszewski J.
      • Vazquez F.
      • Bouchard P.
      • Lemieszczuk B.
      • Baro F.
      • et al.
      Ulipristal acetate versus leuprolide acetate for uterine fibroids.
      ,
      • Levens E.D.
      • Potlog-Nahari C.
      • Armstrong A.Y.
      • Wesley R.
      • Premkumar A.
      • Blithe D.L.
      • et al.
      CDB-2914 for uterine leiomyomata treatment: a randomized controlled trial.
      ,
      • Nieman L.K.
      • Blocker W.
      • Nansel T.
      • Mahoney S.
      • Reynolds J.
      • Blithe D.
      • et al.
      Efficacy and tolerability of CDB-2914 treatment for symptomatic uterine fibroids: a randomized, double-blind, placebo-controlled, phase IIb study.
      ). We also restricted fibroid size to a maximum diameter of 10 cm, but we note that the median diameter of the largest fibroid in a series of patients undergoing hysterectomy, myomectomy, or uterine artery embolization was 6 cm (
      • Spies J.B.
      • Bradley L.D.
      • Guido R.
      • Maxwell G.L.
      • Levine B.A.
      • Coyne K.
      Outcomes from leiomyoma therapies: comparison with normal controls.
      ). Approximately one-third of women (some of whom had surgery) did not enroll in the extension study, and so we cannot be certain how this subset of patients would have responded to repeated UPA treatment courses. Women who received only one UPA treatment course achieved similar levels of bleeding control to those participating in the extension study, although improvements in QoL and fibroid volume reduction were slightly less substantial.
      In conclusion, repeated 3-month courses of oral UPA 10 mg once daily effectively control bleeding and pain, reduce fibroid volume, and restore QoL over the long term in many women with symptomatic fibroids, providing an effective and well-tolerated long-term medical treatment for fibroids.

      Acknowledgments

      The authors thank Paul Terrill of MDSL International for statistical consultancy during the PEARL III and PEARL III extension studies and the preparation of tables and figures of this manuscript. The authors also thank Oliver Pohl (freelance medical writer, formerly PregLem S.A.) for coordinating the manuscript writing and editorial assistance (funded by PregLem S.A.) with the support of Sarah Brittain of MDSL International and Florence S. Jean and Annie Dacquin of PregLem S.A.

      Supplemental Materials

      Donnez. Long-term treatment of uterine fibroids. Fertil Steril 2014.
      This appendix has been provided by the authors to give readers additional information about their work. The full protocols of the clinical trials are available on request to Dr Elke Bestel, M.D., Chief Medical Officer, PregLem S.A., Switzerland (E-mail: [email protected] ).
      Contributors: Wlodzimierz Baranowski, David H. Barlow, Francesco Baró, Pedro N. Barri, Christine Bergeron, Elke Bestel, Philippe Bouchard, Sarah Brittain, Maria Jésus Cancelo Hidalgo, Francisco Carmona, Annie Dacquin, Jacques Donnez, Olivier Donnez, Bart C.J.M. Fauser, Alex Ferenczy, Albrecht Giuliani, Javier Haya, Johannes Huber, Artur Jakimiuk, Florence S. Jean, Jesus S. Jimenez, Jan Kotarski, Boguslaw Lemieszczuk, Ernest Loumaye, Michelle Nisolle, Kazem Nouri, Ian Osterloh, Santiago Palacios, Anna Pawlaczyk, Krzysztof Sodowski, Rafal Sozanski, Jacek Suzin, Paul Terrill, Janusz Tomaszewski, Francisco Vázquez, Emilia Villegas, and Alistair R.W. Williams.

      Materials and methods

       Study Population

       Dates of the study period

      PEARL III only: July 13, 2010, to November 10, 2011.
      PEARL III + PEARL III extension: July 13, 2010 to January 24, 2013.
      The study ended after the last follow-up of the last patient as foreseen in the protocol.

       Principal exclusion criteria

      The principal exclusion criteria for the study are listed in Supplemental Table 1. The body mass index was calculated using the body weight in kilograms divided by the height in meters squared.

       Randomization and masking

      All women received open-label treatment with UPA. UPA treatment courses were equal to 90 days of UPA treatment. Women consented to participate to one treatment course (PEARL III study). Approximately 8 weeks into treatment course 1, women were randomized to receive double-blind treatment with either NETA or matching placebo after a predefined order and according to a predefined randomization list that had been established by the designated unblinded statistician at MDSL International with the corresponding treatment supplied at the end of the first course of UPA. The NETA or placebo were packaged in blisters of identical appearance. The treatment packs were assigned within each site by the investigator to the subjects in sequential order starting from the lowest number to the highest number.
      After completion of the first treatment period, patients were proposed to participate in an extension study (PEARL III extension) with three further UPA treatment courses. The women received the same double-blind treatment of NETA or placebo throughout all treatment courses. Double-blinding was maintained for women and investigators throughout the entire study. At the end of the PEARL III study, the blind was broken for selected sponsor members to fulfill regulatory reporting requirements.

       Efficacy Endpoints

       Assessment of menstrual bleeding

      The proportion of women in amenorrhea at the end of each UPA treatment course and the time to onset of amenorrhea for each treatment course were evaluated. Amenorrhea was defined as no bleeding for a continuous period of at least 35 days. Within any 35-day interval, 1 day of spotting was accepted. To assess this endpoint, we used a simplified semiquantitative bleeding scale, which included four categories: “no bleeding,” “spotting,” “bleeding,” or “heavy bleeding.” During the first treatment course the diary was completed daily with four categories available, whereas during subsequent treatment courses, women only reported spotting, bleeding, or heavy bleeding on days when these occurred (Supplemental Fig. 4A and B show sample charts). The proportion of women with “spotting”/“no bleeding” (that is, no “bleeding” or “heavy bleeding”) and the time to onset of “spotting”/“no bleeding” for each treatment course was also evaluated.
      The PBAC (
      • Mehine M.
      • Kaasinen E.
      • Mäkinen N.
      • Katainen R.
      • Kämpjärvi K.
      • Pitkänen E.
      • et al.
      Characterization of uterine leiomyomas by whole-genome sequencing.
      ,
      • Wallach E.E.
      • Vlahos N.F.
      Uterine myomas: an overview of development, clinical features, and management.
      ,
      • Stewart E.A.
      Uterine fibroids.
      ,
      • Donnez J.
      • Jadoul P.
      What are the implications of myomas on fertility? A need for a debate?.
      ), an instrument that estimates menstrual blood loss in a semiquantitative measure, was used to evaluate the magnitude of menstrual bleeding at the start of the first treatment cycle and during the off-treatment period after each treatment course. PBAC has been widely used for evaluating drugs or devices interfering with menstrual bleeding such as desmopressin (
      • Stewart E.A.
      Uterine fibroids and evidence-based medicine—not an oxymoron.
      ), tranexamic acid (
      • Donnez J.
      • Schrurs B.
      • Gillerot S.
      • Sandow J.
      • Clerckx F.
      Treatment of uterine fibroids with implants of gonadotropin-releasing hormone agonist: assessment by hysterography.
      ), ormeloxifene (
      • Lethaby A.
      • Vollenhoven B.
      Fibroids (uterine myomatosis, leiomyomas).
      ), levonorgestrel intrauterine device (
      • Zapata L.B.
      • Whiteman M.K.
      • Tepper N.K.
      • Jamieson D.J.
      • Marchbanks P.A.
      • Curtis K.M.
      Intrauterine device use among women with uterine fibroids: a systematic review.
      ), and devices for endometrial ablation (
      • Melis G.B.
      • Piras B.
      • Marotto M.F.
      • Orrù M.M.
      • Maricosu G.
      • Pilloni M.
      • et al.
      Pharmacokinetic evaluation of ulipristal acetate for uterine leiomyoma treatment.
      ). The possible score ranges from 0 to more than 500 (with no defined upper limit), with higher scores indicating a greater severity of bleeding. Scores greater than 100 indicate HMB (
      • Mehine M.
      • Kaasinen E.
      • Mäkinen N.
      • Katainen R.
      • Kämpjärvi K.
      • Pitkänen E.
      • et al.
      Characterization of uterine leiomyomas by whole-genome sequencing.
      ). Standardized sanitary materials were provided, and women recorded the number of tampons or pads used and the extent of soiling with blood. A sample chart is presented in Supplemental Figure 4C. The PBAC was completed by women during the first 8 days of their menstrual cycle at the start of treatment course 1 to have an objective assessment of the bleeding intensity and during the first 8 days of the first menstrual cycle after each NETA/placebo treatment.

       Secondary efficacy endpoints

      The volume of the three largest fibroids was assessed using transvaginal ultrasound (TVUS). The same three fibroids identified during screening were followed throughout the study. TVUS also included an assessment of the ovaries, uterine volume, endometrial thickness, and deformation of the uterine cavity and was performed at baseline, at the end of UPA treatment for course 1, and for subjects continuing in the extension study: 2 weeks after the start of first menses after treatment courses 2 and 3, at the end of UPA treatment course 4, and approximately 3 months after the end of the final treatment course.
      Pain was measured with the Short-Form McGill Pain questionnaire (SF-MPQ) (
      • Gainer E.E.
      • Ulmann A.
      Pharmacologic properties of CDB(VA)-2914.
      ). The SF-MPQ part A consists of 15 descriptors that are ranked on an intensity scale of 0 = none to 3 = severe. Descriptors 1–11 represent the sensory (S) dimension of the pain experience, and 12–15 represent the affective (A) dimension. The total score for S is obtained by adding the ranks from descriptors 1–11. The total score for A is obtained by adding the ranks from descriptors 12–15. The overall total score is then obtained by adding the scores for S and A. If either the score for A or S is missing, then the overall total score will also be missing. Scores for SF-MPQ part A range from 0 (no pain) to 45 (severe pain for every S and A descriptor). SF-MPQ part B consists of a visual analog scale (VAS), with the resulting score between 0 mm and 100 mm; scores range from 0 (no pain) to 100 (worst possible pain). The SF-MPQ was assessed at baseline, at the end of weeks 4 and 8, and at the end of UPA treatment for the first treatment course, as well as for subjects continuing in the extension study 2 weeks after the start of first menses after treatment courses 2 and 3, at the end of UPA treatment for course 4, and approximately 3 months after the end of the final treatment course.
      QoL was assessed using the Uterine Fibroid Symptom and Health-Related Quality of Life (UFS-QoL) questionnaire (
      • Horak P.
      • Mara M.
      • Dundr P.
      • Kubinova K.
      • Kuzel D.
      • Hudecek R.
      • et al.
      Effect of a selective progesterone receptor modulator on induction of apoptosis in uterine fibroids in vivo.
      ). The UFS-QoL is a validated, disease-specific questionnaire that consists of two parts; [1] a symptom severity score that includes bleeding, abdominal pressure, urination frequency, and fatigue (range, 0–100) with high symptom severity scores indicating increased symptom severity; and [2] a health-related quality of life (HRQL) total score. The HRQL total score (range, 0–100) is composed of six domains: Concern, Activities, Energy/Mood, Control, Self-Conscious, and Sexual Function. High HRQL scores indicate better QoL. Furthermore, another measure of QoL was assessed using the EQ-5D questionnaire, a standardized instrument for use as a measure of health outcome that consists of five dimensions of health status: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression, all with a three-level scale. It also includes a VAS ranging from zero to 100, with lower scores indicating a worse health state. QoL assessments were recorded at the same time points as the volume of fibroids.

       Safety Endpoints

      All biopsy samples were processed at a central location and were assessed by three independent pathologists who were unaware of the study group assignments, the visit sequence, and one another's assessment. All pathologists have participated in previous UPA studies (
      • Pohl O.
      • Osterloh I.
      • Gotteland J.P.
      Ulipristal acetate—safety and pharmacokinetics following multiple doses of 10–50 mg per day.
      ). Assessments were made according to standard diagnostic criteria and terminology for nonphysiological endometrial changes (p-receptor modulator–associated endometrial changes or PAEC), as described elsewhere by Mutter and colleagues (
      • Donnez J.
      • Tatarchuk T.F.
      • Bouchard P.
      • Puscasiu L.
      • Zakharenko N.F.
      • Ivanova T.
      • et al.
      Ulipristal acetate versus placebo for fibroid treatment before surgery.
      ). A consensus diagnosis of PAEC was made if two or more pathologists observed nonphysiological changes in an endometrial biopsy specimen. The timing of endometrial biopsies is presented in Supplemental Figure 1. Additional samples could be taken if endometrial thickness was >18 mm or if judged necessary by the investigator.
      Clinical laboratory testing (hematology, biochemistry, and coagulation) was performed in a central laboratory (ICON Central Laboratories Ltd.) using validated assays. Laboratory investigations and vital signs were assessed at all visits. Serum E2, ACTH TSH and PRL levels were measured at screening, at the end of UPA treatment for courses 1 and 4, 2 weeks after the start of first menses after treatment course 2, and approximately 3 months after the end of the final treatment course. The FSH level was measured at screening.

       Statistical Analyses

      Data management and statistical analysis were conducted using SAS 9.2 and governed by a comprehensive quality assurance system following International Conference on Harmonisation (
      • Donnez J.
      • Tomaszewski J.
      • Vazquez F.
      • Bouchard P.
      • Lemieszczuk B.
      • Baro F.
      • et al.
      Ulipristal acetate versus leuprolide acetate for uterine fibroids.
      ) and other applicable regulatory guidelines.

      Results

      Supplemental Table 2 presents efficacy results for the first treatment course in the PEARL III open-label set.
      The individual decisions on whether to have surgery or enter the extension study and to continue successive treatment courses may depend on many factors including advice of the investigator or surgeon, the response to treatment, and the patient's own preference for continued treatment and to adhere to all the study visits and procedures. The women's baseline characteristics and the overall response to course 1 (amenorrhea, fibroid volume reduction, pain, and QoL) were similar for women who continued into the extension study compared with all women who started PEARL III. Only four women discontinued PEARL III extension owing to lack of efficacy, and an analysis (not shown) of women who completed all four courses of treatment indicates that there were improvements in amenorrhea rates and fibroid volume reduction with successive treatment courses.
      Supplemental Table 3 presents a summary of volume and timing of first menstruations during off-treatment periods for the PEARL III double-blind set (all women who received NETA or placebo at the end of UPA treatment for treatment course 1) and the PEARL III extension full analysis set), split by the NETA and placebo treatment groups and overall.
      Supplemental Table 8 presents the proportion of women in amenorrhea at the end of each treatment course according to whether the total volume of the three largest fibroids was reduced by ≥ 25%.
      Supplemental Table 4 presents a summary of relevant efficacy results after the first UPA treatment course comparing women not entering the PEARL III extension study with those who decided to participate in an extension study with up to three additional treatment courses. Both subgroups had similar disease severity at baseline and efficacy response to one course of UPA.
      Supplemental Table 5 presents a summary of adverse events in the PEARL III safety set (for the first treatment course).
      Five women discontinued the extension study owing to adverse events. These were women with heavy uterine bleeding (n = 1) after the end of course 2, metrorrhagia (n = 1) during course 3, vertigo/dyspepsia/abdominal cramps and constipation during course 3, high blood pressure (n = 1) during course 4, and heavy uterine bleeding (n = 1) during course 4. Another woman discontinued the study due to leg pain that started 2 months after the end of course 2.
      Supplemental Table 7 shows a summary of pathologist consensus of nonphysiological histological features in the endometrium.
      Supplemental Table 6, Supplemental Table 9 present summaries of other safety assessments in the PEARL III safety set and the PEARL III extension safety set.
      Figure thumbnail fx1
      Supplemental Figure 1Pictorial flow diagram of treatments and biopsies for PEARL III and extension study.
      Figure thumbnail fx2
      Supplemental Figure 2Patient disposition for participants entering PEARL III and PEARL III extension. aIncludes one woman who was withdrawn owing to an adverse event occurring in an off-treatment period. bOne woman was treated correctly for treatment course 1 but received only placebo medication (no UPA) in error for treatment course 2.
      Figure thumbnail fx3
      Supplemental Figure 3Patient disposition for participants entering PEARL III and PEARL III extension, including details for NETA/placebo treatment groups. aOne woman was treated correctly for treatment course 1 but received only placebo medication (no UPA) in error for treatment course 2. bIncludes one woman who was withdrawn owing to an adverse event occurring in an off-treatment period.
      Figure thumbnail fx4ab
      Supplemental Figure 4Assessment of uterine bleeding. (A) Semiquantitative bleeding scale used in PEARL III. (B) Semiquantitative bleeding scale used in PEARL III extension. (C) Example of a completed PBAC showing a score of 320 (equivalent to a menstrual blood loss of approximately 250 mL).
      Figure thumbnail fx4c
      Supplemental Figure 4Assessment of uterine bleeding. (A) Semiquantitative bleeding scale used in PEARL III. (B) Semiquantitative bleeding scale used in PEARL III extension. (C) Example of a completed PBAC showing a score of 320 (equivalent to a menstrual blood loss of approximately 250 mL).
      Figure thumbnail fx5
      Supplemental Figure 5Time to spotting/no bleeding (PEARL III extension full analysis set). Time to spotting/no bleeding was defined as the first day for which there was subsequently only spotting or no bleeding for longer than 35 days to the end of UPA treatment within each treatment course, assessed using the patient diary data from the date of the first dose of UPA (day 0, which was to be within the first 4 days of the start of menstruation for treatment courses 1 and 2 and on the first day of menstruation for treatment courses 3 and 4). Circles denote censored observation (i.e., a woman had a subsequent interval of 35 days or less up to the end of UPA treatment for which only spotting or no bleeding was observed).
      Supplemental Table 1Key exclusion criteria.
      Previous uterine surgery including endometrial ablation or uterine artery embolization.
      History of or current uterus, cervix, ovarian, or breast cancer.
      Significant finding on Papanikolaou test (PAP) smear within the past 12 months.
      Endometrium hyperplasia or adenocarcinoma within the past 6 months or similar lesions in the screening biopsy. In case of biopsies older than 6 months, these had to be repeated.
      Large uterine polyp (>2 cm).
      Calcified fibroids and/or a calcified uterus.
      Severe coagulation disorder.
      One or more ovarian cysts ≥4 cm diagnosed by ultrasound.
      History of treatment for fibroid with an SPRM, including UPA.
      Treatments with progestins (systemic or progestin-releasing intrauterine system), oral contraceptive, acetylsalicylic acid, mefenamic acid, anticoagulants such as cumarins and/or antifibrinolytic drugs such as tranexamic acid, P antagonists, systemic glucocorticoid treatments and/or systemic depot glucocorticoid, treatments that contain PgP substrate (digoxin, fexofenadine) or contain moderate or potent inhibitors or inducers of CYP3A4.
      Supplemental Table 2Efficacy results for first treatment course (PEARL III open-label set; n = 209).
      AssessmentnResult
      Assessments were made at end of a 13-week treatment course of UPA and before 10-day treatment course of NETA/placebo, except for the SF-MPQ, which was also assessed after 4 and 8 weeks of treatment.
      Amenorrhea
       No. of women in amenorrhea (%)209164 (78.5)
      95% CI, %20972.4, 83.5
      Spotting/no bleeding
       No. of women with spotting or no bleeding (%)209181 (86.6)
      95% CI20981.3, 90.6
      Total volume of three largest fibroids
      The same three fibroids identified at screening were followed throughout the study.
       % Change from baseline, median (IQR)194−45.1 (−66.1, −24.9)
       Total reduction ≥25%, n (%)194145 (74.7)
       Total reduction ≥50%, n (%)19486 (44.3)
      Uterine volume
       % Change from baseline, median (IQR)201−28.9 (−45.5, −8.4)
       Reduction ≥25%, n (%)201109 (54.2)
      Assessment of pain
       SF-MPQ
      Scores on the SF-MPQ range from 0 to 45, with higher scores indicating more severe pain.
      End of week 4 actual, median (IQR)2042.0 (0.0, 4.0)
      End of week 4 change from baseline, median (IQR)202−5.0 (−12.0, −1.0)
      End of week 8 actual, median (IQR)1980.0 (0.0, 2.0)
      End of week 8 change from baseline, median (IQR)196−6.0 (−14.0, −1.6)
      End of course 1 actual, median (IQR)2001.0 (0.0, 3.0)
      Course 1 change from baseline, median (IQR)198−6.0 (−14.0, −2.0)
       Visual analog scale (VAS)
      Scores on the VAS of the SF-MPQ range from 0 to 100, with higher scores indicating more severe pain.
      End of week 4 actual, median (IQR)2048.0 (0.0, 28.0)
      End of week 4 change from baseline, median (IQR)204−21.0 (−39.0, −2.0)
      End of week 8 actual, median (IQR)2022.0 (0.0, 12.0)
      End of week 8 change from baseline, median (IQR)202−26.0 (−54.0, −7.0)
      Course 1 actual, median (IQR)2001.0 (0.0, 12.0)
      Course 1 change from baseline, median (IQR)200−24.5 (−54.0, −6.0)
      UFS-QoL questionnaire
      Scores on the VAS of the EQ-5D range from 0 to 100, with higher scores indicating a better health state.
       Symptom severity
      Actual, mean ± SD19614.5 ± 15.5
      Change from baseline, mean ± SD194−33.3 ± 21.5
       Health-Related QoL total score
      Actual, mean ± SD20186.2 ± 15.6
      Change from baseline, mean ± SD20029.2 ± 22.4
      EQ-5D questionnaire
       Mobility (women with problems), n (%)20110 (5.0)
       Self-care (women with problems), n (%)2010 (0.0)
       Usual activities (women with problems), n (%)20012 (6.0)
       Pain/discomfort (women with moderate/extreme symptoms), n (%)20060 (30.0)
       Anxiety/depression (women with moderate/extreme symptoms), n (%)20184 (41.8)
       VAS (health state)
      On the UFS-QoL questionnaire, scores for symptom severity range from 0 to 100, with higher scores indicating increased severity. Total scores for health-related QoL range from 0 to 100, with higher scores indicating a better QoL.
      Actual, mean ± SD20076.8 ± 17.6
      Change from baseline, mean ± SD2009.6 ± 23.3
      Note: n = Number of women with nonmissing observations.
      a Assessments were made at end of a 13-week treatment course of UPA and before 10-day treatment course of NETA/placebo, except for the SF-MPQ, which was also assessed after 4 and 8 weeks of treatment.
      b The same three fibroids identified at screening were followed throughout the study.
      c Scores on the SF-MPQ range from 0 to 45, with higher scores indicating more severe pain.
      d Scores on the VAS of the SF-MPQ range from 0 to 100, with higher scores indicating more severe pain.
      e Scores on the VAS of the EQ-5D range from 0 to 100, with higher scores indicating a better health state.
      f On the UFS-QoL questionnaire, scores for symptom severity range from 0 to 100, with higher scores indicating increased severity. Total scores for health-related QoL range from 0 to 100, with higher scores indicating a better QoL.
      Supplemental Table 3Summary of PBAC at the start of treatment course 1 and PBAC and timing of first menstruation after each treatment course.
      TotalUPA then placeboUPA then NETAP value
      Tests comparing NETA to placebo were conducted using the Wilcoxon rank sum test.
      PEARL III (double-blind set)n = 201n = 103n = 98
      PBAC score (days 1–8 of first menses)
      PBAC score postscreening assessment measured during days 1–8 at start of UPA treatment course 1. Score post-treatment course assessments measured during days 1–8 of first menstruation after the end of the respective treatment course.
       Postscreening, median (IQR)216 (126, 376)218 (126, 344)206 (131, 422)
       Post-treatment course 1n = 194n = 99n = 95
      Actual, median (IQR)98 (36, 211)123 (51, 248)75 (27, 179)
      Change, median (IQR)−92 (−220, 23)−69 (−167, 40)−122 (−299, −9).01
      Time to return to menstruation, days
      Excludes women in whom surgery was performed before return of menstruation or women discontinued before return of menstruation.
       Post-treatment course 1n = 196n = 101n = 95
      Median (IQR)18 (14, 26)25 (20, 29)14 (13, 15)<.0001
      PEARL III extension (full analysis set)n = 132n = 69n = 63
      PBAC score (days 1–8 of first menses)
      PBAC score postscreening assessment measured during days 1–8 at start of UPA treatment course 1. Score post-treatment course assessments measured during days 1–8 of first menstruation after the end of the respective treatment course.
       Post-screening, median (IQR)235 (142, 397)228 (142, 139)257 (140, 498)
       Post-treatment course 1n = 131n = 69n = 62
      Actual, median (IQR)94 (31, 210)123 (53, 233)59 (21, 155)
      Change, median (IQR)−120 (−264, 5)−92 (−171, 25)−144 (−367, −42).006
       Post-treatment course 2n = 123n = 65n = 58
      Actual, median (IQR)64 (20, 168)84 (18, 226)54 (21, 103)
      Change, median (IQR)−150 (−255, −55)−127 (−211, −14)−171 (−372, −71).01
       Post-treatment course 3n = 114n = 58n = 56
      Actual, median (IQR)42 (14, 138)75 (22, 191)28 (9, 66)
      Change, median (IQR)−131 (−311, −59)−103 (−186, −36)−224 (−461, −104)<.0003
       Post-treatment course 4n = 89n = 49n = 40
      Actual, median (IQR)31 (11, 100)55 (22, 136)13 (6, 48)
      Change, median (IQR)−120 (−255, −45)−97 (−184, −31)−166 (−356, −74).02
      Time to return to menstruation, days
      Excludes women in whom surgery was performed before return of menstruation or women discontinued before return of menstruation.
       Post-treatment course 1n = 131n = 69n = 62
      Median (IQR)18 (14, 27)25 (20, 30)14 (13, 15)<.0001
       Post-treatment course 2n = 125n = 66n = 59
      Median (IQR)18 (14, 27)26 (21, 34)14 (13, 17)<.0001
       Post-treatment course 3n = 116n = 60n = 56
      Median (IQR)21 (14, 31)30 (24, 36)14 (13, 17)<.0001
       Post-treatment course 4n = 99n = 53n = 46
      Median (IQR)22 (15, 34)30 (23, 35)15 (14, 18)<.0001
      a Tests comparing NETA to placebo were conducted using the Wilcoxon rank sum test.
      b PBAC score postscreening assessment measured during days 1–8 at start of UPA treatment course 1. Score post-treatment course assessments measured during days 1–8 of first menstruation after the end of the respective treatment course.
      c Excludes women in whom surgery was performed before return of menstruation or women discontinued before return of menstruation.
      Supplemental Table 4Selected efficacy results for participants not entering/entering the PEARL III extension study.
      Did not enter extensionEntered extensionDifferenceStatistics
      Fisher's exact test was used for amenorrhea and spotting or no bleeding; Wilcoxon rank sum test was used for total volume of three largest fibroids, uterine volume, assessment of pain (medians reported), and analysis of covariance for the UFS-QoL questionnaire (means reported).
      Amenorrhea, n/N (%)
      Amenorrhea (and spotting or no bleeding) assessed while under UPA treatment. For remaining results, data were collected at the end of UPA treatment.
      59/77 (76.6)105/132 (79.5)(2.9).73
       95% CI, %66.0, 84.771.9, 85.5−8.8, 14.6
      Spotting or no bleeding, n/N (%)
      Amenorrhea (and spotting or no bleeding) assessed while under UPA treatment. For remaining results, data were collected at the end of UPA treatment.
      64/77 (83.1)117/132 (88.6)(5.5).30
       95% CI, %73.2, 89.982.1, 93.0−4.4, 15.5
      Total volume of three largest fibroids
      The same three fibroids identified at screening were followed throughout the study.
      n = 64n = 130
       % Change from baseline−38.5−49.9−12.1.01
       95% CI, %−47.8, −29.0−56.7, −42.5−22.2, −2.7
      Uterine volumen = 69n = 132
       % Change from baseline−27.0−29.8−1.4.79
       95% CI, %−33.9, −14.5−35.3, −22.6−10.1, 7.5
      Assessment of pain
       VAS
      Scores on VAS range from 0 to 100, with higher scores indicating more severe pain.
      n = 68n = 132
      Change from baseline−27.5−23.5−3.0.53
      95% CI, %−36.0, −13.0−34.0, −17.0−12.0, 6.0
      UFS-QoL questionnaire
      On the UFS-QoL questionnaire, scores for symptom severity range from 0 to 100, with higher scores indicating increased severity. Total scores for health-related QoL range from 0 to 100, with higher scores indicating a better QoL.
       Symptom severityn = 65n = 129
      Change from baseline−28.4−35.8
      Least square mean−30.6−34.7−4.1.08
      95% CI, %−34.3, −26.8−37.3, −32.0−8.7, 0.5
       Health-related QoL total scoren = 69n = 131
      Change from baseline22.232.8
      Least square mean25.331.25.9.009
      95% CI, %21.7, 28.828.6, 33.81.5, 10.3
      a Fisher's exact test was used for amenorrhea and spotting or no bleeding; Wilcoxon rank sum test was used for total volume of three largest fibroids, uterine volume, assessment of pain (medians reported), and analysis of covariance for the UFS-QoL questionnaire (means reported).
      b Amenorrhea (and spotting or no bleeding) assessed while under UPA treatment. For remaining results, data were collected at the end of UPA treatment.
      c The same three fibroids identified at screening were followed throughout the study.
      d Scores on VAS range from 0 to 100, with higher scores indicating more severe pain.
      e On the UFS-QoL questionnaire, scores for symptom severity range from 0 to 100, with higher scores indicating increased severity. Total scores for health-related QoL range from 0 to 100, with higher scores indicating a better QoL.
      Supplemental Table 5Adverse events (PEARL III safety set; n = 209).
      All adverse events that occurred during UPA treatment in at least 3% of women overall and all SAEs are included.
      Adverse eventNo. of women (%)
      SAEs
      At least one SAE2 (1.0)
       Any SAE during UPA treatment0
       Any SAE during NETA/Placebo treatment0
       Any SAE off treatment2 (1.0)
      One woman had two off-treatment SAEs.
      Menometrorrhagia1 (0.5)
      Uterine hemorrhage1 (0.5)
      Breast cancer1 (0.5)
      A 46-year-old woman was diagnosed with a lobular breast cancer 2 months after the end of treatment (T1, grade II).
      Adverse events
      Adverse events with onset on or after the first dose of UPA and before the first dose of NETA/placebo within each treatment course or up to and including 7 days after the last dose of UPA if NETA/placebo was never started.
      Leading to study withdrawal
      In addition, another adverse event that occurred when a woman was not receiving UPA led to study withdrawal.
      1 (0.5)
      At least one event120 (57.4)
       Headache34 (16.3)
       Nasopharyngitis14 (6.7)
       Abdominal pain (including upper/lower)11 (5.3)
       Hot flashes10 (4.8)
       Fatigue9 (4.3)
       Nausea8 (3.8)
       Vertigo8 (3.8)
       Breast discomfort/breast pain/breast tenderness8 (3.8)
       Pelvic pain8 (3.8)
      a All adverse events that occurred during UPA treatment in at least 3% of women overall and all SAEs are included.
      b One woman had two off-treatment SAEs.
      c A 46-year-old woman was diagnosed with a lobular breast cancer 2 months after the end of treatment (T1, grade II).
      d Adverse events with onset on or after the first dose of UPA and before the first dose of NETA/placebo within each treatment course or up to and including 7 days after the last dose of UPA if NETA/placebo was never started.
      e In addition, another adverse event that occurred when a woman was not receiving UPA led to study withdrawal.
      Supplemental Table 6Summary of other safety assessments (PEARL III safety set; n = 209).
      Assessment
      Course 1 data were collected at the end of UPA treatment. Second menstrual bleed post-treatment data were collected for those entering the extension; 3-month follow-up data were collected for those not entering the extension.
      nResult
      Serum E2, pg/mL
       Screening, median (IQR)205120.0 (77, 184)
       Course 1, median (IQR)18055.0 (41, 84)
      ACTH, pg/mL
       Screening, median (IQR)19613.3 (9.3, 19.8)
       Course 1, median (IQR)17713.2 (9.2, 17.1)
      TSH, mIU/L
       Screening, median (IQR)2051.36 (0.96, 1.96)
       Course 1, median (IQR)1811.49 (1.00, 2.00)
      PRL, ng/mL
       Screening, median (IQR)20511.4 (8.2, 18.2)
       Course 1, median (IQR)1807.9 (5.5, 11.7)
      Total cholesterol, mmol/L
       Screening, median (IQR)2055.06 (4.48, 5.60)
       Course 1, median (IQR)1805.25 (4.71, 5.82)
       Second menstrual bleed post-treatment, median (IQR)1234.89 (4.41, 5.47)
       3-Month follow-up, median (IQR)565.16 (4.44, 5.93)
      HDL, mmol/L
       Screening, median (IQR)2051.61 (1.43, 1.91)
       Course 1, median (IQR)1801.64 (1.44, 1.89)
       Second menstrual bleed post-treatment, median (IQR)1231.56 (1.31, 1.81)
       3-Month follow-up, median (IQR)561.72 (1.46, 1.92)
      LDL, mmol/L
       Screening, median (IQR)2032.81 (2.40, 3.39)
       Course 1, median (IQR)1782.97 (2.55, 3.54)
       Second menstrual bleed post-treatment, median (IQR)1212.72 (2.30, 3.07)
       3-Month follow-up, median (IQR)562.73 (2.19, 3.53)
      Endometrial thickness, mm
       Screening, mean ± SD2028.7 ± 3.7
      ≤4 mm, n (%)14 (6.9)
      >4 to ≤16 mm, n (%)185 (91.6)
      >16 mm, n (%)3 (1.5)
       Course 1, mean ± SD1989.2 ± 4.6
      ≤4 mm, n (%)23 (11.6)
      >4 to ≤16 mm, n (%)157 (79.3)
      >16 mm, n (%)18 (9.1)
       2 Weeks after first menstrual bleed post-treatment, mean ± SD1818.9 ± 4.1
      ≤4 mm, n (%)19 (10.5)
      >4 to ≤16 mm, n (%)150 (82.9)
      >16 mm, n (%)12 (6.6)
       3-Month follow-up, mean ± SD487.6 ± 3.4
      ≤4 mm, n (%)9 (18.8)
      >4 to ≤16 mm, n (%)39 (81.3)
      >16 mm, n (%)0
      Systolic blood pressure, mmHg
       Baseline, mean ± SD209121.4 ± 13.8
       End of week 4, mean ± SD207119.7 ± 13.1
       End of week 8, mean ± SD203119.7 ± 13.7
       Course 1, mean ± SD202120.2 ± 13.7
       2 Weeks after first menstrual bleed post-treatment, mean ± SD191121.9 ± 12.3
       Second menstrual bleed post-treatment, mean ± SD132119.8 ± 13.2
       3-Month follow-up, mean ± SD57120.4 ± 13.6
      Diastolic blood pressure, mmHg
       Baseline, mean ± SD20974.4 ± 9.8
       End of week 4, mean ± SD20773.4 ± 9.9
       End of week 8, mean ± SD20374.0 ± 10.0
       Course 1, mean ± SD20274.2 ± 9.9
       2 Weeks after first menstrual bleed post-treatment, mean ± SD19174.9 ± 9.5
       Second menstrual bleed post-treatment, mean ± SD13273.6 ± 9.4
       3-Month follow-up, mean ± SD5774.5 ± 10.8
      Pulse rate, bpm
       Baseline, mean ± SD20973.9 ± 10.7
       End of week 4, mean ± SD20772.6 ± 9.3
       End of week 8, mean ± SD20372.7 ± 9.3
       Course 1, mean ± SD20172.1 ± 9.4
       2 Weeks after first menstrual bleed post-treatment, mean ± SD19172.8 ± 9.5
       Second menstrual bleed post-treatment, mean ± SD13272.4 ± 10.4
       3-Month follow-up, mean ± SD5672.2 ± 10.8
      Weight, kg
       Baseline, mean ± SD20969.0 ± 12.8
       Course 1, mean ± SD20268.5 ± 12.8
       Second menstrual bleed post-treatment, mean ± SD13268.2 ± 13.2
       3-Month follow-up, mean ± SD5870.2 ± 11.7
      Note: n = Number of women with nonmissing observations.
      a Course 1 data were collected at the end of UPA treatment. Second menstrual bleed post-treatment data were collected for those entering the extension; 3-month follow-up data were collected for those not entering the extension.
      Supplemental Table 7Summary of pathologist consensus of nonphysiological histological features in the endometrium.
      TotalUPA then PlaceboUPA then NETA
      PEARL IIIn = 209n = 103n = 98
      Screeningn = 171n = 85n = 80
       No. of women (%)18 (10.5)7 (8.2)11 (13.8)
      2 Weeks after first menstrual bleed post-treatmentn = 176n = 92n = 84
       No. of women (%)45 (25.6)26 (28.3)19 (22.6)
      PEARL III extensionn = 132n = 69n = 63
      Screeningn = 105n = 56n = 49
       No. of women (%)11 (10.5)4 (7.1)7 (14.3)
      2 Weeks after first menstrual bleed post-treatment course 1n = 122n = 65n = 57
       No. of women (%)35 (28.7)20 (30.8)15 (26.3)
      2 Weeks after first menstrual bleed post-treatment course 4n = 87n = 47n = 40
       No. of women (%)22 (25.3)
      Of the 22 women diagnosed with a consensus of nonphysiological histological features 2 weeks after first menstrual bleed post-treatment course 4, 15 (7 placebo, 8 NETA) women had biopsies at 3-month follow-up. Of these, three (1 placebo and 2 NETA) women were diagnosed with nonphysiological histological features at this follow-up visit.
      11 (23.4)11 (27.5)
      Note: n = Number of women with nonmissing observations.
      a Of the 22 women diagnosed with a consensus of nonphysiological histological features 2 weeks after first menstrual bleed post-treatment course 4, 15 (7 placebo, 8 NETA) women had biopsies at 3-month follow-up. Of these, three (1 placebo and 2 NETA) women were diagnosed with nonphysiological histological features at this follow-up visit.
      Supplemental Table 8Women in amenorrhea at the end of each treatment course by total volume of a reduction ≥25% of the three largest fibroids.
      Treatment courseFibroid volume reduction ≥25%
      Fibroid volume reduction assessed using total volume of the three largest fibroids assessed during the screening period before the first treatment course. The same three fibroids identified at screening were followed throughout the study. Course 1 and 4 data were collected at the end of UPA treatment; course 2 and 3 data were collected after first menstrual bleed after UPA treatment and subsequent NETA/placebo treatment.
      TotalAmenorrhea, n (%)
      Course 1Yes10182 (81.2)
      No2922 (75.9)
      Total
      Overall totals for each treatment course are the number of women receiving UPA in each treatment course who did not have surgery performed during or before the end of each treatment course and who have a nonmissing amenorrhea assessment and nonmissing fibroid volume reduction assessment for the treatment course in question.
      130104 (80.0)
      Course 2Yes9588 (92.6)
      No2419 (79.2)
      Total
      Overall totals for each treatment course are the number of women receiving UPA in each treatment course who did not have surgery performed during or before the end of each treatment course and who have a nonmissing amenorrhea assessment and nonmissing fibroid volume reduction assessment for the treatment course in question.
      119107 (89.9)
      Course 3Yes8374 (89.2)
      No2321 (91.3)
      Total
      Overall totals for each treatment course are the number of women receiving UPA in each treatment course who did not have surgery performed during or before the end of each treatment course and who have a nonmissing amenorrhea assessment and nonmissing fibroid volume reduction assessment for the treatment course in question.
      10695 (89.6)
      Course 4Yes7975 (94.9)
      No1714 (82.4)
      Total
      Overall totals for each treatment course are the number of women receiving UPA in each treatment course who did not have surgery performed during or before the end of each treatment course and who have a nonmissing amenorrhea assessment and nonmissing fibroid volume reduction assessment for the treatment course in question.
      9689 (92.7)
      a Fibroid volume reduction assessed using total volume of the three largest fibroids assessed during the screening period before the first treatment course. The same three fibroids identified at screening were followed throughout the study. Course 1 and 4 data were collected at the end of UPA treatment; course 2 and 3 data were collected after first menstrual bleed after UPA treatment and subsequent NETA/placebo treatment.
      b Overall totals for each treatment course are the number of women receiving UPA in each treatment course who did not have surgery performed during or before the end of each treatment course and who have a nonmissing amenorrhea assessment and nonmissing fibroid volume reduction assessment for the treatment course in question.
      Supplemental Table 9Summary of other safety assessments (PEARL III extension safety set; n = 132)
      Assessment
      Course 1 and 4 data were collected at the end of UPA treatment; course 2 and 3 data were collected after first menstrual bleed after UPA treatment and subsequent NETA/placebo treatment.
      nResult
      Serum E2, pg/mL
       Screening, median (IQR)131120.0 (76.0, 185.0)
       Course 1, median (IQR)11254.0 (38.5, 81.0)
       Course 2, median (IQR)119129.0 (62.0, 194.0)
       Course 4, median (IQR)9543.0 (27.0, 61.0)
       3-Month follow-up, median (IQR)9696.5 (66.5, 173.0)
      ACTH, pg/mL
       Screening, median (IQR)12613.0 (9.3, 19.3)
       Course 1, median (IQR)11712.7 (9.7, 16.9)
       Course 2, median (IQR)10312.6 (8.3, 17.3)
       Course 4, median (IQR)8612.5 (8.1, 16.3)
       3-Month follow-up, median (IQR)8913.2 (8.9, 20.9)
      TSH, mIU/L
       Screening, median (IQR)1311.31 (0.94, 1.83)
       Course 1, median (IQR)1131.41 (0.93, 2.00)
       Course 2, median (IQR)1191.35 (0.92, 1.88)
       Course 4, median (IQR)951.39 (0.84, 2.16)
       3-Month follow-up, median (IQR)961.33 (0.90, 1.85)
      PRL, ng/mL
       Screening, median (IQR)13111.3 (8.9, 18.5)
       Course 1, median (IQR)1127.55 (5.45, 11.35)
       Course 2, median (IQR)1188.85 (6.30, 12.90)
       Course 4, median (IQR)956.90 (5.10, 11.60)
       3-Month follow-up, median (IQR)969.35 (7.35, 15.35)
      Total cholesterol, mmol/L
       Screening, median (IQR)1315.00 (4.50, 5.58)
       Course 1, median (IQR)1125.25 (4.93, 5.78)
       Course 2, median (IQR)1194.97 (4.57, 5.49)
       Course 4, median (IQR)945.33 (4.93, 5.85)
       3-Month follow-up, median (IQR)965.17 (4.73, 5.69)
      HDL, mmol/L
       Screening, median (IQR)1311.61 (1.43, 1.88)
       Course 1, median (IQR)1121.65 (1.41, 1.92)
       Course 2, median (IQR)1191.59 (1.40, 1.90)
       Course 4, median (IQR)941.60 (1.40, 1.87)
       3-Month follow-up, median (IQR)961.62 (1.41, 1.92)
      LDL, mmol/L
       Screening, median (IQR)1292.81 (2.40, 3.38)
       Course 1, median (IQR)1103.01 (2.60, 3.50)
       Course 2, median (IQR)1172.77 (2.41, 3.26)
       Course 4, median (IQR)943.04 (2.58, 3.62)
       3-Month follow-up, median (IQR)962.95 (2.52, 3.56)
      Endometrial thickness, mm
       Screening, mean ± SD1299.3 ± 4.0
      ≤4 mm, n (%)7 (5.4)
      >4 to ≤16 mm, n (%)119 (92.2)
      >16 mm, n (%)3 (2.3)
       Course 1, mean ± SD1309.4 ± 4.5
      ≤4 mm, n (%)13 (10.0)
      >4 to ≤16 mm, n (%)107 (82.3)
      >16 mm, n (%)10 (7.7)
       Course 2, mean ± SD1208.8 ± 3.8
      ≤4 mm, n (%)15 (12.5)
      >4 to ≤16 mm, n (%)101 (84.2)
      >16 mm, n (%)4 (3.3)
       Course 3, mean ± SD1078.0 ± 3.4
      ≤4 mm, n (%)15 (14.0)
      >4 to ≤16 mm, n (%)91 (85.0)
      >16 mm, n (%)1 (0.9)
       Course 4, mean ± SD957.5 ± 3.4
      ≤4 mm, n (%)17 (17.9)
      >4 to ≤16 mm, n (%)77 (81.1)
      >16 mm, n (%)1 (1.1)
       3-Month follow-up, mean ± SD978.9 ± 3.2
      ≤4 mm, n (%)6 (6.2)
      >4 to ≤16 mm, n (%)89 (91.8)
      >16 mm, n (%)2 (2.1)
      Systolic blood pressure, mmHg
       Baseline, mean ± SD132122.2 ± 14.1
       Course 1, mean ± SD132120.5 ± 13.6
       Course 2, mean ± SD121120.4 ± 13.7
       Course 3, mean ± SD109120.3 ± 11.4
       Course 4, mean ± SD96120.7 ± 13.3
       3-Month follow-up, mean ± SD99119.5 ± 11.0
      Diastolic blood pressure, mmHg
       Baseline, mean ± SD13274.2 ± 9.3
       Course 1, mean ± SD13274.4 ± 9.7
       Course 2, mean ± SD12174.7 ± 10.3
       Course 3, mean ± SD10973.6 ± 8.8
       Course 4, mean ± SD9674.4 ± 8.1
       3-Month follow-up, mean ± SD9973.5 ± 9.2
      Pulse rate, bpm
       Baseline, mean ± SD13273.8 ± 10.7
       Course 1, mean ± SD13272.5 ± 9.5
       Course 2, mean ± SD12173.8 ± 10.5
       Course 3, mean ± SD10973.3 ± 9.5
       Course 4, mean ± SD9672.2 ± 8.7
       3-Month follow-up, mean ± SD9972.9 ± 9.5
      Weight, kg
       Baseline, mean ± SD13268.7 ± 13.6
       Course 1, mean ± SD13267.8 ± 13.3
       Course 2, mean ± SD12167.3 ± 12.6
       Course 3, mean ± SD10967.5 ± 12.7
       Course 4, mean ± SD9667.4 ± 12.6
       3-Month follow-up, mean ± SD9967.8 ± 13.3
      Note: n = Number of women with nonmissing observations.
      a Course 1 and 4 data were collected at the end of UPA treatment; course 2 and 3 data were collected after first menstrual bleed after UPA treatment and subsequent NETA/placebo treatment.

      References

        • Mehine M.
        • Kaasinen E.
        • Mäkinen N.
        • Katainen R.
        • Kämpjärvi K.
        • Pitkänen E.
        • et al.
        Characterization of uterine leiomyomas by whole-genome sequencing.
        N Engl J Med. 2013; 369: 43-53
        • Wallach E.E.
        • Vlahos N.F.
        Uterine myomas: an overview of development, clinical features, and management.
        Obstet Gynecol. 2004; 104: 393-406
        • Stewart E.A.
        Uterine fibroids.
        Lancet. 2001; 357: 293-298
        • Donnez J.
        • Jadoul P.
        What are the implications of myomas on fertility? A need for a debate?.
        Hum Reprod. 2002; 17: 1424-1430
        • Stewart E.A.
        Uterine fibroids and evidence-based medicine—not an oxymoron.
        N Engl J Med. 2012; 366: 471-473
        • Donnez J.
        • Schrurs B.
        • Gillerot S.
        • Sandow J.
        • Clerckx F.
        Treatment of uterine fibroids with implants of gonadotropin-releasing hormone agonist: assessment by hysterography.
        Fertil Steril. 1989; 51: 947-950
        • Lethaby A.
        • Vollenhoven B.
        Fibroids (uterine myomatosis, leiomyomas).
        Clin Evid (Online). 2011; 2011: 0814
        • Zapata L.B.
        • Whiteman M.K.
        • Tepper N.K.
        • Jamieson D.J.
        • Marchbanks P.A.
        • Curtis K.M.
        Intrauterine device use among women with uterine fibroids: a systematic review.
        Contraception. 2010; 82: 41-55
        • Melis G.B.
        • Piras B.
        • Marotto M.F.
        • Orrù M.M.
        • Maricosu G.
        • Pilloni M.
        • et al.
        Pharmacokinetic evaluation of ulipristal acetate for uterine leiomyoma treatment.
        Expert Opin Drug Metab Toxicol. 2012; 8: 901-908
        • Gainer E.E.
        • Ulmann A.
        Pharmacologic properties of CDB(VA)-2914.
        Steroids. 2003; 68: 1005-1011
        • Horak P.
        • Mara M.
        • Dundr P.
        • Kubinova K.
        • Kuzel D.
        • Hudecek R.
        • et al.
        Effect of a selective progesterone receptor modulator on induction of apoptosis in uterine fibroids in vivo.
        Int J Endocrinol. 2012; 2012: 436174
        • Pohl O.
        • Osterloh I.
        • Gotteland J.P.
        Ulipristal acetate—safety and pharmacokinetics following multiple doses of 10–50 mg per day.
        J Clin Pharm Ther. 2013; 38: 314-320
        • Donnez J.
        • Tatarchuk T.F.
        • Bouchard P.
        • Puscasiu L.
        • Zakharenko N.F.
        • Ivanova T.
        • et al.
        Ulipristal acetate versus placebo for fibroid treatment before surgery.
        N Engl J Med. 2012; 366: 409-420
        • Donnez J.
        • Tomaszewski J.
        • Vazquez F.
        • Bouchard P.
        • Lemieszczuk B.
        • Baro F.
        • et al.
        Ulipristal acetate versus leuprolide acetate for uterine fibroids.
        N Engl J Med. 2012; 366: 421-432
        • Mutter G.L.
        • Bergeron C.
        • Deligdisch L.
        • Ferenczy A.
        • Glant M.
        • Merino M.
        • et al.
        The spectrum of endometrial pathology induced by progesterone receptor modulators.
        Mod Pathol. 2008; 21: 591-598
        • Fiscella J.
        • Bonfiglio T.
        • Winters P.
        • Eisinger S.H.
        • Fiscella K.
        Distinguishing features of endometrial pathology after exposure to the progesterone receptor modulator mifepristone.
        Hum Pathol. 2011; 42: 947-953
        • Ioffe O.B.
        • Zaino R.J.
        • Mutter G.L.
        Endometrial changes from short-term therapy with CDB-4124, a selective progesterone receptor modulator.
        Mod Pathol. 2009; 22: 450-459
        • Williams A.R.
        • Bergeron C.
        • Barlow D.H.
        • Ferenczy A.
        Endometrial morphology after treatment of uterine fibroids with the selective progesterone receptor modulator, ulipristal acetate.
        Int J Gynecol Pathol. 2012; 31: 556-569
        • Higham J.M.
        • O'Brien P.M.
        • Shaw R.W.
        Assessment of menstrual blood loss using a pictorial chart.
        Br J Obstet Gynaecol. 1990; 97: 734-739
        • Melzack R.
        The short-form McGill Pain Questionnaire.
        Pain. 1987; 30: 191-197
        • Dolan P.
        Modeling valuations for EuroQol health states.
        Med Care. 1997; 35: 1095-1108
        • Spies J.B.
        • Coyne K.
        • Guaou G.N.
        • Boyle D.
        • Skyrnarz-Murphy K.
        • Gonzalves S.M.
        The UFS-QOL, a new disease-specific symptom and health-related quality of life questionnaire for leiomyomata.
        Obstet Gynecol. 2002; 99: 290-300
        • Fiscella K.
        • Eisinger S.H.
        • Meldrum S.
        • Feng C.
        • Fisher S.G.
        • Guzick D.S.
        Effect of mifepristone for symptomatic leiomyomata on quality of life and uterine size: a randomized controlled trial.
        Obstet Gynecol. 2006; 108: 1381-1387
        • Spies J.B.
        • Bradley L.D.
        • Guido R.
        • Maxwell G.L.
        • Levine B.A.
        • Coyne K.
        Outcomes from leiomyoma therapies: comparison with normal controls.
        Obstet Gynecol. 2010; 116: 641-652
        • Takeda T.
        • Osuga K.
        • Miyake A.
        • Wakabayashi A.
        • Morishige K.
        • Kimura T.
        Elevated level of plasma vascular endothelial growth factor after gonadotropin-releasing hormone agonist treatment for leiomyomata.
        Gynecol Endocrinol. 2008; 24: 724-726
        • Kim H.S.
        • Baik J.H.
        • Pham L.D.
        • Jacobs M.A.
        MR-guided high-intensity focused ultrasound treatment for symptomatic uterine leiomyomata: long-term outcomes.
        Acad Radiol. 2011; 18: 970-976
        • Kim J.J.
        • Sefton E.C.
        The role of progesterone signaling in the pathogenesis of uterine leiomyoma.
        Mol Cell Endocrinol. 2012; 358: 223-231
        • Scialli A.R.
        • Jestila K.J.
        Sustained benefits of leuprolide acetate with or without subsequent medroxyprogesterone acetate in the nonsurgical management of leiomyomata uteri.
        Fertil Steril. 1995; 64: 313-320
        • Levens E.D.
        • Potlog-Nahari C.
        • Armstrong A.Y.
        • Wesley R.
        • Premkumar A.
        • Blithe D.L.
        • et al.
        CDB-2914 for uterine leiomyomata treatment: a randomized controlled trial.
        Obstet Gynecol. 2008; 111: 1129-1136
        • Nieman L.K.
        • Blocker W.
        • Nansel T.
        • Mahoney S.
        • Reynolds J.
        • Blithe D.
        • et al.
        Efficacy and tolerability of CDB-2914 treatment for symptomatic uterine fibroids: a randomized, double-blind, placebo-controlled, phase IIb study.
        Fertil Steril. 2011; 95: 767-772