{"paper_id":"b3881a19-ba40-4b22-bc63-889bd6037633","body_text":"HE4 is a novel tissue marker for therapy\nresponse and progestin resistance in\nmedium- and low-risk endometrial\nhyperplasia\nAnne Ørbo *,1,2, Marit Arnes 2, Lena Myreng Lyså 1, Christer Borgfelt 3 and Bjørn Straume 4\n1Department of Clinical Pathology, University Hospital of Tromsø, N-9038 Tromsø, Norway; 2Research Group for Gynaecologic\nOncology, Department of Medical Biology, Faculty of Health Sciences, University of Tromsø, N-9037 Tromsø, Norway;\n3Department of Obstetrics and Gynecology, Skåne University Hospital, Lund, Sweden and 4Department of Community Medicine,\nFaculty of Health Sciences, University of Tromsø, N-9037 Tromsø, Norway\nBackground: The aim of the present study was to investigate whether changes in the tissue expression of human epididymis-\nspecific protein 4 (HE4) could predict therapy resistance and relapse after progestin hormone therapy for medium- and low-risk\nendometrial hyperplasia.\nMethods: Endometrial biopsies were obtained from women participating in a multicentre RCT performed according to the CONSORT\nguidelines; the women were randomly assigned to either LNG-IUS; 10mg of oral medroxyprogesterone acetate (MPA) administered for\n10 days per cycle; or 10 mg of oral MPA administered daily for 6 months. Of the 153 women who completed therapy, 141 had adequate\nmaterial for immunohistochemistry in pre- and post-treatment biopsies. An antibody to HE4 (clone 12A2 monoclonal IgG1 antibody,\nFujirebio Diagnostics, Inc.) was used for the immunohistochemic al staining of the pre- and post-treatment biopsies from each\nparticipant. The expression of HE4 staining was evaluated by the histological score ( H-score) using light microscopy.\nResults: Changes in the expression of HE4 ( H-score) during therapy were related to the therapy group ( Po0.001) and therapy\nresponse ( Po0.001) of the individuals but could not predict relapse ( P40.05). Changes in the intracellular bodies were shown to\npredict both the therapy response ( P ¼ 0.038) and relapse ( P ¼ 0.014).\nConclusions: Changes in the expression of HE4 during progestin therapy regimens can predict therapy response or indicate\nprogestin resistance for medium- and low-risk endometrial hyperplasia.\nEndometrial cancer, presently the most common gynaecological\nmalignancy in the western world, is still increasing in prevalence\n(Okuda et al , 2010; Rodriguez et al , 2016). As much as 80% of\nendometrial cancers are preceded by a preliminary stage called\nendometrial hyperplasia. Therefore, the development of optimal\ndiagnostics and individualised therapy for endometrial hyperplas-\ntic lesions should be of high priority to prevent this trend.\nHowever, national routines for therapy and follow up for\nendometrial hyperplasia have thus far been arbitrary or non-\nexisting (Kurman et al , 1985; Gallos et al , 2013b). In recent\ndecades, progestin has been accepted as a conservative therapy for\nlow- and medium-risk endometrial hyperplasia (Ferenczy and\nGelfand, 1989; Jobo et al , 2001; Bese et al , 2006; Clark et al , 2006;\nOrbo et al , 2008). A levonorgestrel-impregnated intrauterine\ndevice (LNG-IUS, Mirena, Bayer, Berlin, Germany) has recently\nbeen demonstrated to be superior to oral progestin (Wildemeersch\net al, 2007; Varma et al, 2008; Buttini et al, 2009; Gallos et al, 2010;\nOrbo et al , 2014; Abu et al , 2015), but oral regimens are still\n*Correspondence: Professor A Ørbo; E-mail: anne.orbo@uit.no\nRevised 4 June 2016; accepted 12 July 2016; published online 18 August 2016\n& 2016 Cancer Research UK. All rights reserved 0007 – 0920/16\nFULL PAPER\nKeywords: endometrial; hyperplasia; progestin; HE4; therapy; response; LNG-IUS\nBritish Journal of Cancer (2016) 115, 725–730 | doi: 10.1038/bjc.2016.247\nwww.bjcancer.com | DOI:10.1038/bjc.2016.247 725\n\npreferred by many women with intolerance to intrauterine therapy.\nHowever, most of the published data indicate a high frequency of\nprogestin resistance or therapy failure after oral regimes compared\nwith intrauterine treatment (Gallos et al , 2013a). In a recent RCT,\nwe found that 31% of individuals failed to respond after cyclic oral\ntherapy lasting for 6 months (Orbo et al , 2014). However, to date,\nno objective molecular marker for resistance to progestin therapy\nhas been identified.\nHuman epididymis-specific protein 4 (HE4, belonging to one of\nfour WAP gene products), has been shown to be secreted into the\nserum as a glycoprotein overexpressed in serous and endometrioid\ntypes of ovarian carcinomas (Drapkin et al , 2005; Bouchard et al ,\n2006). The exact function of the HE4 protein is unknown, but\nprevious studies have demonstrated that HE4 overexpression\npromotes cell adhesion and migration whereas HE4 suppression\nmarkedly inhibits tumour growth (Lu et al , 2012).\nIn the past decade, HE4 has drawn attention as a serum marker\nfor endometrial cancer (Moore et al , 2011; Kalapotharakos et al ,\n2012). Thus, HE4 has been shown to be a promising marker for the\ndetection and prognosis of endometrial cancer with respect to the\nFigo stage, myometrial invasion, and grade (Moore et al , 2008;\nBignotti et al, 2011; Moore et al, 2011; Mutz-Dehbalaie et al, 2012).\nImmunohistochemical detection of HE4 in endometrial tissue\nsamples revealed that the rate of expression of HE4 in endometrial\nhyperplasia was higher compared with normal controls and\nlower compared with individuals with endometrial cancer\n(Deng et al , 2015; Li et al , 2015). However, hormone-mediated\nregulation of HE4 tissue expression by progestin has never been\ndescribed, and no studies exist of HE4 as a predictive marker of the\ntherapy response after progestin therapy for endometrial\nhyperplasia.\nThe present study was based on a national multicentre RCT\ncomparing different progestin regimens as a treatment for\nendometrial hyperplasia (Orbo et al , 2014; Orbo et al , 2015).\nOur main intention was to investigate whether variation in the\ntissue expression of HE4 during different progestin therapy\nregimens could predict therapy response or therapy failure and\nwhether HE4 might be considered a potential marker for progestin\nresistance for medium- and low-risk endometrial hyperplasia.\nMATERIALS AND METHODS\nPatients. The biopsy material was obtained from 153 women with\nlow- and medium-risk endometrial hyperplasia in a national\nmulticentre randomised study (Orbo et al , 2014). The women had\nbeen treated with either LNG-IUS (Mirena, Bayer), 10 mg of\nmedroxyprogesterone acetate (MPA) for 10 days per cycle, or\n10 mg of MPA daily for 6 months (ClinicalTrials.gov,\nNCT01074892, Regional Ethical committee number 25/2004)\n(Orbo et al, 2014). The study inclusion period was from 1 January\n2005 to 1 November 2011. The treatment period was completed on\n1 May 2012. After 6 months of treatment, all therapy was\nwithdrawn. One pre- and one post-treatment biopsy (after 6\nmonths) was obtained from each participant using a Pipelle\n(Endometrial suction curette, Jiangsu Guard King Medical\nEquipment, Jiangsu, China) (Orbo et al , 2014). For the present\nstudy, paired adequate biopsy material from 141 women was\navailable for the immunohistochemical analyses; 48 of these\nwomen had been treated with LNG-IUS; 44 with continuous\nMPA; and 49 with cyclic MPA. Insufficient biopsy material in the\nparaffin blocks was the reason for excluding eight of the original\n153 women. Patient characteristics, such as age, menopausal status,\nparity, and WHO diagnosis, were registered and related to the\nstaining of HE4 in tissue specimens taken before the start of\ntherapy and when therapy was discontinued (after 6 months). To\nmonitor relapse, all the included patients were followed up at 6-\nmonth intervals for 24 months after therapy was discontinued.\nEndometrial biopsies. The endometrial biopsy material was sent\nto the Department of Pathology at the University Hospital of\nNorth Norway for routine assessment. The specimens were fixed in\nbuffered formaldehyde, embedded in paraffin, and further\nprocessed in the laboratory before standard histological sections\nwere obtained. Diagnostic assessment of the WHO classification by\nlight microscopy was performed by a trained gynaecology\npathologist (AO) and one additional routine pathologist; each\nwas blinded to the other’s diagnosis. Agreement after discordant\nresults was always obtained after discussion at a two-headed\nmicroscope. The index biopsies were classified into one of the three\ngroups: simple hyperplasia, complex hyperplasia, or atypical\nhyperplasia, according to the WHO classification, which was\nconsidered the gold standard for the evaluation of endometrial\nhyperplasia at the time the study was performed (Kurman et al ,\n1985; Gallos et al , 2013b). Normalised histology in the control\nbiopsies after therapy was defined as ordinary proliferative\nendometrium or endometrium with a progestin effect (Kurman\net al , 1985; Gallos et al , 2013b). All information from the WHO\nclassification of the index and control biopsies was registered and\nmaintained in a separate database and subsequently supplemented\nwith information from hospital records.\nImmunohistochemistry. Immunohistochemistry was performed\naccording to customers’ advice. Slides with a thickness of 4–5 mm\nwere routinely cut from paraffin blocks and placed on Super\nfrost þ glasses (Fisher Scientific, Oslo, Norway), followed by\nincubation overnight at 60 1C, which was necessary for fixation.\nDeparaffinisation, pretreatment in a Tris-based, slightly alkaline\nreagent (CC1 mild) for 30 min at 95 1C and staining were\nautomatically performed in a Benchmark XT from Ventana\nMedical Systems, Inc (Tucson, AZ, USA). The primary antibody\nused in the present study was HE4 clone 12A2 monoclonal IgG1\nantibody (Fujirebio Diagnostics, Inc., PA, USA). The initial\nconcentration was 15.1 mg ml\n-1, and the antibody was used at a\ndilution of 1out of 4000 in Antibody Diluent (Ventana Medical\nSystems, Inc.). After addition of the primary antibody, the slides\nwere incubated for 32 min at 37 1C, and automatic DAB staining\nwas performed in several steps with an Ultraview Universal DAB\nDetection kit (Ventana Medical Systems, Inc.) before dehydration,\ncounterstaining with haematoxylin, and mounting.\nEvaluation of immunohistochemical expression in endometrial\nglands ( H-score). The results of immunostaining in endometrial\nhyperplasia specimens taken before therapy and immediately after\n6 months of progestin therapy were assessed by light microscopy.\nThe HE4 staining, which was concentrated mainly within the\ncytoplasm of endometrial glandular cells, was examined using a\nsemi-quantitative immunohistochemical histological score\n(H-score), which incorporates both the intensity and the distribu-\ntion of specific staining (Huang et al , 1996). The immunohisto-\nchemical staining results were classified as follows: light-brown\ngranular cytoplasm ¼ 1, medium-brown granular cytoplasm ¼ 2,\nor dark-brown granular cytoplasm ¼ 3. The H-score has been\npreviously formulated as HS ¼ P(Pi /C2 i)/100, in which Pi denotes\nthe percentage of stained cells and i denotes the intensity of the\nstaining, ranging from 1–3 (Huang et al , 1996). The H-score\nevaluation was performed visually as a consensus between a trained\ngynaecological pathologist (AO) and a trained technician (MA).\nBoth investigators were blinded to the different patient groups, and\nthree specific areas within each specimen (hot spots) were defined\nfor investigation in the pre- and post-therapy biopsy specimens.\nThe staining intensity and the number of positive cells were\nevaluated. The H-scores for the cells in the pre-treatment biopsies\nwere used to divide the samples into three groups: low\nBRITISH JOURNAL OF CANCER HE4 endometrial hyperplasia, progestin resistance\n726 www.bjcancer.com | DOI:10.1038/bjc.2016.247\n\n(H-score 40.5), medium ( H-score 0.6–1.5), and high ( H-score\n41.6). A statistical evaluation of the differences in the H-scores\nbetween the pre- and post-treatment biopsies related to therapy\ngroup, therapy response, and relapse was used to divide the\npatients as follows: patients with reduced H-scores after therapy\nwere placed in group 1, and patients with unchanging or increased\nH-scores were placed in group 2. Aggregates of dark-brown, well-\ndefined material (bodies) were observed within the cytoplasm of\nthe endometrial glands. These bodies showed variations in size and\nwere classified as follows: absence of bodies—0; o2–4 mmi n\ndiameter—1; 5–10 mm in diameter—2; and 11–15 mm diameter—3.\nChanges in the diameter of the bodies between pre- and post-\ntreatment specimens were divided into two separate groups and\nevaluated according to the therapy group, therapy response, and\nrelapse.\nEthical approval. The study following the Helsinki declaration\nwas designed according to the CONSORT statement for rando-\nmised clinical trials (Schulz et al , 2010), was approved by the\nRegional Committee for Medical and Health Research Ethics\n(number 25/2004), by the Norwegian Council of Medical Advice,\nand by the Norwegian Medicines Agency. The study was registered\nin the ClinicalTrials.gov NCT01074892.\nStatistics. All statistical analyses were performed in simple cross\ntables, and P-values for the w2- tests are reported.\nRESULTS\nPatients. Among the 153 women originally included in this study, all\nof those treated with LNG-IUS h ad a complete response (normal\nproliferative endometrium or endometrium with atrophic glands and\npseudo-decidualised stroma) after 6 months of treatment. Only 69%\nof those treated with cyclic oralprogestin (10 mg of MPA, 10 days per\ncycle) had a response, and of the women treated daily with oral MPA\n(10 mg), 96% had a response after 6 months (Orboet al, 2014). In the\npresent study, biopsies from 141 ofthe 153 women were available for\nthe immunohistochemical analyses. Among these women, 123 were\nresponders and 18 were non-responders. All of the 48 women treated\nwith the LNG-IUS were responders. In addition, 33 of the 49 women\ntreated with cyclic MPA and 42 of the 44 women treated with\ncontinuous MPA were responders.The mean ages of the responding\nand non-responding women were 47.6 and 48.4 years, respectively.\nDuring the 24 months of follow up, 50 women were diagnosed\nwith relapse, of which 10 were in the cyclic oral group, 19 were in\nthe continuous oral group, and 21 were in the LNG-IUS group\n(Orbo et al , 2015).\nH-score before therapy. All of the endometrial biopsy specimens\nfrom the 141 women with endometrial hyperplasia showed the\nexpression of HE4 in the glands before the start of therapy, but the\nmajority of cases expressed moderate staining. In the LNG-IUS-\ntreated group ( n ¼ 48), 3 (6%) had low H-scores (o0.5), 33 (69%)\nhad medium H-scores (0.6–1.0), and 12 (25%) had high H-scores\n(41) before therapy. In the group treated with continuous oral\nprogestin ( n ¼ 44), none had low H-scores, 37 (84%) had medium\nH-scores, and 7 (16%) had high H-scores before the start of\ntherapy. In the group treated with cyclic oral progestin ( n ¼ 49),\n5 (10%) had low H-scores, 39 (80%) had medium H-scores, and\n5 (10%) had high H-scores.\nChanges in H-scores related to progestin therapy regimen. The\nchanges in the H-scores in the endometrial glands for the three\ntherapy groups were significantly related to the therapy regimen\n(Po0.001); details are shown in Table 1. Human epididymis-\nspecific protein 4 expression before therapy in the glands is\nreported in Figure 1A. The changes in the H-scores of the\nendometrial glands for the different therapy regimens were not\nrelated to relapse ( P ¼ 0.83).\nChanges in H-scores related to therapy response and relapse. In\nthe LNG-IUS-treated group, all the women responded to therapy\n(48 out of 48). Of these, 3 had weak H-scores (o0.5) (6%), 33 had\nmedium H-scores (0.6–1.0) (69%), and 12 had high H-scores (41)\n(25%) before the start of therapy. In the group treated with\ncontinuous oral progestin, only 2 of 44 women were non-\nresponders (2 out of 44). Of the responders, none had low\nH-scores, 35 had medium H-scores (80%), and 7 had high\nH-scores (15.9%), whereas the 2 non-responders had medium\nH-scores (4.1%) before therapy.\nIn the group treated with cyclic oral progestin, 16 of the 49\nwomen were non-responders (16 out of 49).\nAmong the responders, 4 (8.2%) had low H-scores, 24 (49%) had\nmedium H-scores, and 5 (10.2%) had high H-scores. In the non-\nresponding group, 15 (30.6%) had mediumH-scores, and 1 (2%) had\nal o w H-score. Changes in the H-score were significantly related to\ntherapy response ( Po0.001); the results are shown in Table 2.\nChanges in the H-score were not related to relapse in the 24-month\nfollow-up period after therapy was discontinued ( P ¼ 0.83).\nTable 1. Change in the H-score (difference in pre- and post-\ntreatment biopsies) in the three therapy groups (10 mg of\nMPA for 10 days per cycle, 10 mg of MPA daily, or the\nlevonorgestrel impregnated intrauterine system for 6\nmonths) in a randomised controlled trial in women with\nendometrial hyperplasia ( n ¼ 141) related to the therapy\nregimen used\nTherapy/HE4\nexpression\nReduced or\nabsent HE4\nexpression after\ntherapy\nN\nIncreased or\npersistent HE4\nexpression after\ntherapy\nN Total\nOral cyclic progestin 13\n26.5%\n36\n73.5%\n49\n100.0\nOral continuous\nprogestin\n20\n45.4%\n24\n54.6%\n44\n100.0\nLNG-IUS 33\n68.7%\n15\n31.3%\n48\n100.0\nTotal 66\n46.8%\n75\n53.2%\n141\n100.0\nAbbreviations: HE4 ¼ human epididymis-specific protein 4; H-score ¼ histological score;\nMPA ¼ medroxyprogesterone acetate. w2: Po0.001.\nTable 2. Change in the H-score (difference in pre- and post-\ntreatment biopsies) in patients with and without therapy\nresponse ( n ¼ 141) after 6 months of progestin therapy\n(10 mg of MPA for 10 days per cycle, 10 mg of MPA daily, or\nthe levonorgestrel impregnated intrauterine system for 6\nmonths) in a randomised controlled trial in women with\nendometrial hyperplasia\nResponse/HE4\nexpression\nReduced or\nabsent HE4\nexpression after\ntherapy\nN\nIncreased or\npersistent HE4\nexpression after\ntherapy\nN Total\nNot responding to\ntherapy\n1\n5.6%\n17\n94.4%\n18\n100.0\nResponding to\ntherapy\n65\n52.9%\n58\n47.1%\n123\n100.0\nTotal 66\n46.8%\n75\n53.2%\n141\n100.0\nAbbreviations: HE4 ¼ human epididymis-specific protein 4; H-score ¼ histological score;\nMPA ¼ medroxyprogesterone acetate. w2: Po0.001\nHE4 endometrial hyperplasia, progestin resistance BRITISH JOURNAL OF CANCER\nwww.bjcancer.com | DOI:10.1038/bjc.2016.247 727\n\nFigure 1B shows HE4 expression after therapy failure in a\nwoman treated with oral cyclic progestin. Figure 1C shows the lack\nof HE4 expression after treatment with LNG-IUS.\nChanges in HE4-stained cellular bodies related to therapy\nresponse and relapse. Human epididymis-specific protein 4 bodies\nwere observed as dark-brown granular aggregates varying in size\nbetween 2 and 15 mm that were localised within the cell cytoplasm or\nin close proximity to the cell membrane. Intracellular localisation\ncould be a regular apical row or random dispersal within the cells.\nThese bodies were not observed by routine H þ E staining. Changes\nin the H-score of HE4-stained aggregates, which were intra- or\nintercellular bodies, were compared in pre- and post-treatment\nspecimens. These changes were significantly correlated with therapy\nresponse ( P ¼ 0.038) (Table 3) but not with the therapy group\n(P ¼ 0.556). Changes in the HE4 bodies reflected the only HE4-\nrelated parameter that could significantly predict relapse of\nendometrial hyperplasia (P ¼ 0.014) (Tables 3 and 4).\nH-scores related to WHO categorisation in pre-treatment\nbiopsies. Table 5 shows that most of the included cases\ncharacterised as complex hyperplasia also had medium H-scores\nbefore the start of therapy.\nA BC\nFigure 1. Expression of HE4 before and after progestin therapy. (A) Endometrial hyperplasia stained with HE4 (clone 12A2 monoclonal IgG1\nantibodies, Fujirebio Diagnostics, Inc.) before the start of therapy. The cytoplasm of the glandular cells is stained brown by HE4. Scattered\nintracellular HE4-stained bodies are observed randomly dispersed within the cells. ( B) An endometrial specimen stained with HE4 (clone 12A2\nmonoclonal IgG1 antibody, Fujirebio Diagnostics, Inc.) after successful therapy by LNG-IUS. The endometrium is pseudo-decidualised, and the\nglands have become small and atrophic with a lack of HE4 staining. ( C) Persistent endometrial hyperplasia stained with HE4 (clone 12A2\nmonoclonal IgG1 antibodies, Fujirebio Diagnostics, Inc.) after therapy failure. This woman had been given 10 mg of MPA taken 10 days per cycle\nfor 6 months.\nTable 3. Change in the size of HE4-stained bodies (difference\nin pre- and post-treatment biopsies in aggregates of HE4-\nstained material in the glandular cell cytoplasm) in women\nwith and without therapy response after 6 months of\nprogestin therapy ( n ¼ 141) (10 mg of MPA) for 10 days per\ncycle, 10 mg of MPA daily, or the levonorgestrel impregnated\nintrauterine system for 6 months) in a randomised controlled\ntrial in women with endometrial hyperplasia\nTherapy\nresponse/ HE4-\nstained bodies\nSize reduction or\nabsence of\nHE4-stained\naggregates after\ntherapy\nN\nIncrease in size\nof HE4-stained\naggregates after\ntherapy\nN Total\nNot responding to\ntherapy\n3\n16.7%\n15\n83.3%\n18\n100.00\nResponding to\ntherapy\n20\n10.8%\n110\n89.2%\n123\n100.00\nTotal 31\n22.0%\n110\n78.0%\n141\n100.00\nAbbreviations: HE4 ¼ human epididymis-specific protein 4; MPA ¼ medroxyprogesterone\nacetate. w2: P ¼ 0.038.\nTable 4. Difference in the size of HE4-stained bodies\n(difference in pre- and post-treatment biopsies in aggregates\nof HE4-stained material in the glandular cell cytoplasm) after\n6 months of progestin therapy in women who never\nexperienced relapse and those who later relapsed during 24\nmonths of follow up ( n ¼ 123)\nRelapse of\nhyperplasia/HE4-\nstained bodies\nSize reduction or\nabsence of\nHE4-stained\naggregates after\ntherapy\nN\nIncrease in size\nof HE4-stained\naggregates after\ntherapy\nN Total\nNo relapse 11\n15.1%\n62\n84.9%\n73\n100.00\nRelapse 17\n34.0%\n33\n66.0%\n50\n100\nTotal 28\n22.8%\n95\n77.2%\n123\n100.00\nAbbreviation: HE4 ¼ human epididymis-specific protein 4. w2: P ¼ 0.014.\nTable 5. Relationship between the H-score before therapy\nstarted and the WHO classification of the pre-treatment\nbiopsies in a randomised controlled trial in women with\nendometrial hyperplasia ( n ¼ 141) randomised to different\nprogestin therapy regimens (10 mg of MPA) for 10 days per\ncycle, 10 mg of MPA daily, or the levonorgestrel impregnated\nintrauterine system for 6 months)\nClassification of endometrial hyperplasia\nH-score/WHO\nclassificationa\nSimple\nhyperplasia\nN\nComplex\nhyperplasia\nN\nAtypical\nhyperplasia\nN\nTotal\nN\nHE4p0.5 1\n5.0%\n5\n4.8%\n1\n6.3%\n7\n5.0%\nHE4 0.5–1 16\n80.0%\n81\n77.1%\n12\n75.0%\n109\n77.3%\nHE4X13\n15.0%\n19\n18.1%\n3\n18.7%\n25\n17.7%\nTotal\nN\n20\n100.0\n105\n100.0\n16\n100.0\n141\n100.0\nAbbreviations: HE4 ¼ human epididymis-specific protein 4; H-score ¼ histological score;\nMPA ¼ medroxyprogesterone acetate; WHO ¼ World Health Organisation.\na(Kurman et al , 1985; Gallos et al , 2013b).\nBRITISH JOURNAL OF CANCER HE4 endometrial hyperplasia, progestin resistance\n728 www.bjcancer.com | DOI:10.1038/bjc.2016.247\n\nChanges in H-scores in endometrial glands related to patient\ncharacteristics. When patient characteristics such as BMI ( o20,\n20–24, 25–30, and 430), menopausal status (pre-, peri-, and\npostmenopausal), or age (o45, 45–48, 49–51, and 52 þ years) were\ncompared with therapy regimen and therapy response, no significant\nrelationship could be identified (data not shown) (Orbo et al, 2015).\nDISCUSSION\nTo date, this is the first multicentre RCT to demonstrate that HE4\nexpression in biopsy material from endometrial hyperplasia can be\nmonitored and regulated by progestin. The administered therapy\nregimens differed particularly in terms of the total dose of progestin\nyielded to the endometrial mucosa during the 6-month therapy\nperiod (Orbo et al, 2014). Consequently, significant differences in the\nHE4 score were observed between pre- and post-treatment biopsies\nfor the three progestin therapy groups. A greater reduction in the\nHE4 score was observed with the LNG-IUS compared with oral\ntherapy. The concentration of levonorgestrel, obtained by the LNG-\nIUS in the uterine mucosa, has been shown to be more than 100\ntimes higher than the concentration observed after oral treatment\nwith the same type of progestin (Nilsson et al, 1982). All the women\ntreated with the LNG-IUS in the present study obtained a complete\ntherapy response after 6 months (Orbo et al , 2014). Thus, the\nreduction in the tissue expression of HE4 (reduced H-score) appears\nto be associated with the given progestin dose.\nHowever, the function of the HE4 protein family, known to\nserve as endogenous peptide protease inhibitors, remains obscure,\nand no specific molecular link between HE4 and progestin has ever\nbeen described (Jiang et al , 2013). In cell studies of ovarian cancer\noverexpression and knockout of HE4-related genes, HE4 has been\nshown to regulate malignant mechanisms such as cell adhesion,\ninvasion, and proliferation (Jiang et al , 2013). Evidence exists that\nthe tumour-promoting effects of HE4 are effectuated by the\nactivation of the EGFR-MAPK molecular pathway, which further\nactivates PI3K/Akt-mediated proliferation and other types of\nmalignant behaviour (Lu et al , 2012; Liu et al , 2015). Several\nstudies have demonstrated that the influence of progestin\nhormones on the PI3K signalling pathway can counteract the\nmalignant effects (Minaguchi et al , 2007). Minaguchi et al (2007)\ndemonstrated that medroxyprogesterone had an anti-tumour effect\nthrough the dephosphorylation of Akt. In another study,\nprogesterone was shown to inhibit the oestradiol-stimulated Akt/\nCyclinD1/pRB pathway, blocking endometrial proliferation (Chen\net al , 2005). Thus, according to several in vitro studies, the\npossibility exists that progestin can counteract the malignant effect\ninduced by HE4 in a dose-dependent manner, and this might be\neffectuated by the regulation of the PI3K signalling pathway. The\nexact mechanism for progestin resistance remains unknown. In the\npresent study, 18 women showed resistance to progestin therapy\nwith persistent hyperplasia after 6 months. Among these 18\nwomen, 17 showed increased H-scores after therapy. All of these\nwomen had been treated with either oral cyclic (16 out of 18) or\noral continuous progestin (2 out of 18). Thus, higher progestin\ndoses might be required to reverse and suppress the unfavourable\ninfluence of high HE4 levels. Studies have been performed to\ninvestigate the molecular events related to progestin resistance and\nthe reversion of progestin resistance in endometrial cancer. Gu\net al (2011) have shown that low concentrations or the absence of\nprogesterone receptors in endometrial cancer cells are associated\nwith a poor response to progesterone therapy. Their results also\nindicated that activation of the PI3K/Akt pathway by progestin\nwithout PR mediation has an important role in progestin resistance\nto endometrial cancer (Gu et al , 2011). In contrast, we have\nobserved that progesterone receptors were completely absent after\n3 months of LNG-IUS therapy in endometrial hyperplasia, even\nthough the treatment effect was efficiently maintained over several\nyears and all the included patients had responded to the therapy\n(Vereide et al, 2006; Orbo et al, 2008; Orbo et al, 2014). In the same\nstudy, both progesterone receptor A and progesterone receptor B\nwere present in women with therapy failure after 3 months\n(Vereide et al , 2006). In endometrial cancer cell lines, we showed\ndose-dependent progestin effects using supra-physiological con-\ncentrations of progestin to ensure that the progesterone receptors\nwere thoroughly saturated (Moe et al\n, 2009a,b). Taken together,\nthese former studies indicate that non-genomic signalling path-\nways are activated by high progestin concentrations and that the\neffective suppression of HE4 may be dependent on high progestin\nconcentrations.\nThus far, a link between the synthesis, cell transport, and serum\nlevels of HE4 has not been clearly demonstrated. In recent studies,\nthe tissue expression of HE4 localised in the cytoplasm in glands\nhas been shown to increase in proportion to the increase in\nmalignancy in endometrial proliferative conditions (Li et al , 2015;\nDeng et al , 2015). The absence of HE4 occurred in benign\nendometrial lesions, but severe hyperplasia and low-grade cancer\nwere characterised by stronger HE4 staining compared with simple\nand complex hyperplasia without atypia (Li et al, 2015; Deng et al,\n2015). This is consistent with our results showing that the majority\nof the included women, who were treated conservatively, obtained\na medium score of HE4 before therapy started. Human\nepididymis-specific protein 4 is a small secretory glycoprotein\ndetected in the plasma, and HE4 levels have been described to\nincrease proportionately with Figo stage and tumour grade\n(Li et al , 2015; Brennan et al , 2015). In addition, high levels of\nHE4 in the plasma have been shown to preoperatively identify\nhigh-risk patients with low-grade endometrial cancer with\nincreased risks of lymph node metastases (Brennan et al , 2015).\nOur finding of HE4 bodies localised within the cytoplasm in\nendometrial epithelial cells has never been described. Even if our\nresults indicate that changes in diameter of these bodies during\ntherapy predicted relapse, the statistics was not powerful and these\nfindings need to be further analysed in future studies.\nThe possibility exists that these aggregates of HE4 molecules\nmight represent HE4 marker glycoproteins ready to be transported\nfrom intracellular compartments to the bloodstream, representing\nthe missing link between tissue expression and measurable serum\nlevels of HE4. Although immunohistochemistry has its limitations,\nthe present study is based on a recently published RCT, which\nensures the quality of the results.\nIn the current study, we found that changes in HE4 expression\nduring therapy for endometrial hyperplasia were significantly related\nto the progestin therapy regimen and correlated to the therapy\nresponse but could not predict relapse. Our results suggest that\nchanges in the expression of HE4 after various progestin therapy\nregimens can predict therapy response vs therapy resistance. Thus,\nHE4 may be a useful tissue marker for women with medium- and\nlow-risk endometrial hyperplasia who are intolerant to LNG-IUS and\nwho can be successfully treated by oral progestin.\nACKNOWLEDGEMENTS\nFujirebio kindly provided the HE4 antibody but had no influence on the\nstudy project or the evaluation of the results. We also thank the Regional\nResearch Board of Northern Norway (Helse Nord) for the grants.\nCONFLICT OF INTEREST\nThe authors declare no conflict of interest.\nHE4 endometrial hyperplasia, progestin resistance BRITISH JOURNAL OF CANCER\nwww.bjcancer.com | DOI:10.1038/bjc.2016.247 729\n\nREFERENCES\nAbu HH, Ghayaty E, El RM (2015) Levonorgestrel-releasing intrauterine\nsystem vs oral progestins for non-atypical endometrial hyperplasia: a\nsystematic review and metaanalysis of randomized trials. Am J Obstet\nGynecol 213(4): 469–478.\nBese T, Vural A, Ozturk M, Dagistanli F, Demirkiran F, Tuncdemir M, Arvas\nM, Sanioglu C, Kosebay D (2006) The effect of long-term use of\nprogesterone therapy on proliferation and apoptosis in simple endometrial\nhyperplasia without atypia. Int J Gynecol Cancer 16: 809–813.\nB i g n o t t iE ,R a g n o l iM ,Z a n o t t iL ,C a l z aS ,F a l c h e t t iM ,L o n a r d iS ,B e r g a m e l l iS ,\nBandiera E, Tassi RA, Romani C, Tode schini P, Odicino FE, Facchetti F,\nPecorelli S, Ravaggi A (2011) Diagnostic and prognostic impact of serum HE4\ndetection in endometrial carcinoma patients.Br J Cancer 104(9): 1418–1425.\nBouchard D, Morisset D, Bourbonnais Y, Tremblay GM (2006) Proteins with\nwhey-acidic-protein motifs and cancer. Lancet Oncol 7: 167–174.\nBrennan DJ, Hackethal A, Mann KP, Mutz-Dehbalaie I, Fiegl H, Marth C,\nObermair A (2015) Serum HE4 detects recurrent endometrial cancer in\npatients undergoing routine clinical surveillance. BMC Cancer 15: 33–1028.\nButtini MJ, Jordan SJ, Webb PM (2009) The effect of the levonorgestrel\nreleasing intrauterine system on endometrial hyperplasia: an Australian\nstudy and systematic review. Aust N Z J Obstet Gynaecol 49: 316–322.\nChen B, Pan H, Zhu L, Deng Y, Pollard JW (2005) Progesterone inhibits the\nestrogen-induced phosphoinositide 3-kinase– 4AKT–4GSK-3beta–\n4cyclin D1– 4pRB pathway to block uterine epithelial cell proliferation.\nMol Endocrinol 19: 1978–1990.\nClark TJ, Neelakantan D, Gupta JK (2006) The management of endometrial\nhyperplasia: an evaluation of current practice. Eur J Obstet Gynecol Reprod\nBiol 125: 259–264.\nD e n gL ,G a oY ,L iX ,C a iM ,W a n gH ,Z h u a n gH ,T a nM ,L i uS ,H a oY ,L i nB( 2 0 1 5 )\nExpression and clinical significanceo fa n n e x i nA 2a n dhuman epididymis\nprotein 4 in endometrial carcinoma.J Exp Clin Cancer Res 34: 96–0208.\nDrapkin R, von Horsten HH, Lin Y, Mok SC, Crum CP, Welch WR, Hecht JL\n(2005) Human epididymis protein 4 (HE4) is a secreted glycoprotein that\nis overexpressed by serous and endometrioid ovarian carcinomas. Cancer\nRes 65: 2162–2169.\nFerenczy A, Gelfand M (1989) The biologic significance of cytologic atypia in\nprogestogen-treated endometrial hyperplasia.Am J Obstet Gynecol160:1 2 6 – 1 3 1 .\nGallos ID, Krishan P, Shehmar M, Ganesan R, Gupta JK (2013a) LNG-IUS\nversus oral progestogen treatment for endometrial hyperplasia: a long-\nterm comparative cohort study. Hum Reprod 28: 2966–2971.\nGallos ID, Krishan P, Shehmar M, Ganesan R, Gupta JK (2013b) Relapse of\nendometrial hyperplasia after conservative treatment: a cohort study with\nlong-term follow-up. Hum Reprod 28: 1231–1236.\nGallos ID, Shehmar M, Thangaratinam S, Papapostolou TK, Coomarasamy A,\nGupta JK (2010) Oral progestogens vs levonorgestrel-releasing\nintrauterine system for endometrial hyperplasia: a systematic review and\nmetaanalysis. Am J Obstet Gynecol 203: 547.e1–10.\nGu C, Zhang Z, Yu Y, Liu Y, Zhao F, Yin L, Feng Y, Chen X (2011) Inhibiting\nthe PI3K/Akt pathway reversed progestin resistance in endometrial cancer.\nCancer Sci 102: 557–564.\nHuang A, Pettigrew NM, Watson PH (1996) Immunohistochemical assay for\noestrogen receptors in paraffin wax sections of breast carcinoma using a\nnew monoclonal antibody. J Pathol 180: 223–227.\nJiang SW, Chen H, Dowdy S, Fu A, Attewell J, Kalogera E, Drapkin R, Podratz\nK, Broaddus R, Li J (2013) HE4 transcription- and splice variants-specific\nexpression in endometrial cancer and correlation with patient survival. Int\nJ Mol Sci 14: 22655–22677.\nJ o b oT ,K a w a g u c h iM ,I m a iM ,K u r a m o to H (2001) Treatment for complex\natypical hyperplasia of the endometrium.Eur J Gynaecol Oncol 22: 365–368.\nKalapotharakos G, Asciutto C, Henic E, Casslen B, Borgfeldt C (2012) High\npreoperative blood levels of HE4 predicts poor prognosis in patients with\novarian cancer. J Ovarian Res 5: 20–25.\nKurman RJ, Kaminski PF, Norris HJ (1985) The behavior of endometrial\nhyperplasia. A long-term study of \"untreated\" hyperplasia in 170 patients.\nCancer 56: 403–412.\nLi X, Gao Y, Tan M, Zhuang H, Gao J, Hu Z, Wang H, Zhu L, Liu J, Lin B\n(2015) Expression of HE4 in endometrial cancer and its clinical\nsignificance. Biomed Res Int 2015: 437468.\nLiu X, Zhao F, Hu L, Sun Y (2015) Value of detection of serum human\nepididymis secretory protein 4 and carbohydrate antigen 125 in diagnosis\nof early endometrial cancer of different pathological subtypes. Onco\nTargets Ther 8: 1239–1243.\nLu R, Sun X, Xiao R, Zhou L, Gao X, Guo L (2012) Human epididymis protein\n4 (HE4) plays a key role in ovarian cancer cell adhesion and motility.\nBiochem Biophys Res Commun 419(2): 274–280.\nMinaguchi T, Nakagawa S, Takazawa Y, Nei T, Horie K, Fujiwara T, Osuga Y,\nYasugi T, Kugu K, Yano T, Yoshikawa H, Taketani Y (2007) Combined\nphospho-Akt and PTEN expressions associated with post-treatment\nhysterectomy after conservative progestin therapy in complex atypical\nhyperplasia and stage Ia, G1 adenocarcinoma of the endometrium. Cancer\nLett 248: 112–122.\nMoe BG, Vereide AB, Orbo A, Sager G (2009a) High concentrations of\nprogesterone and mifepristone mutually reinforce cell cycle retardation\nand induction of apoptosis. Anticancer Res 29: 1053–1058.\nMoe BT, Vereide AB, Orbo A, Jaeger R, Sager G (2009b) Levonorgestrel,\nmedroxyprogesterone and progesterone cause a concentration-dependent\nreduction in endometrial cancer (Ishikawa) cell density, and high\nconcentrations of progesterone and mifepristone act in synergy.\nAnticancer Res 29: 1047–1052.\nMoore RG, Brown AK, Miller MC, Badgwell D, Lu Z, Allard WJ, Granai CO,\nBast Jr. RC, Lu K (2008) Utility of a novel serum tumor biomarker HE4 in\npatients with endometrioid adenocarcinoma of the uterus. Gynecol Oncol\n110(2): 196–201.\nMoore RG, Miller MC, Disilvestro P, Landrum LM, Gajewski W, Ball JJ,\nSkates SJ (2011) Evaluation of the diagnostic accuracy of the risk of\novarian malignancy algorithm in women with a pelvic mass. Obstet\nGynecol 118: 280–288.\nMutz-Dehbalaie I, Egle D, Fessler S, Hubalek M, Fiegl H, Marth C,\nWidschwendter A (2012) HE4 is an independent prognostic marker in\nendometrial cancer patients. Gynecol Oncol 126(2): 186–191.\nNilsson CG, Haukkamaa M, Vierola H, Luukkainen T (1982) Tissue\nconcentrations of levonorgestrel in women using a levonorgestrel-\nreleasing IUD. Clin Endocrinol (Oxf) 17: 529–536.\nOkuda T, Sekizawa A, Purwosunu Y, Nagatsuka M, Morioka M, Hayashi M,\nOkai T (2010) Genetics of endometrial cancers. Obstet Gynecol Int 2010:\n984013.\nOrbo A, Arnes M, Hancke C, Vereide AB, Pettersen I, Larsen K (2008)\nTreatment results of endometrial hyperplasia after prospective D-score\nclassification. A follow-up study comparing effect of LNG-IUD and oral\nprogestins versus observation only. Gynecol Oncol 111: 68–73.\nOrbo A, Arnes M, Vereide AB, Straume B (2015) Relapse risk of endometrial\nhyperplasia after treatment with the levonorgestrel-impregnated\nintrauterine system or oral progestogens. BJOG 123: 1512–1519.\nOrbo A, Vereide A, Arnes M, Pettersen I, Straume B (2014) Levonorgestrel-\nimpregnated intrauterine device as treatment for endometrial hyperplasia:\na national multicentre randomised trial. BJOG 121: 477–486.\nRodriguez AM, Schmeler KM, Kuo YF (2016) Lack of improvement in\nsurvival rates for women under 50 with endometrial cancer, 2000-2011.\nJ Cancer Res Clin Oncol 142: 783–793.\nSchulz KF, Altman DG, Moher D. (2010) CONSORT 2010 Statement:\nupdated guidelines for reporting parallel group randomised trials. BMC\nMed 8: 18.\nVarma R, Soneja H, Bhatia K, Ganesan R, Rollason T, Clark TJ, Gupta JK\n(2008) The effectiveness of a levonorgestrel-releasing intrauterine system\n(LNG-IUS) in the treatment of endometrial hyperplasia-A long-term\nfollow-up study. Eur J Obstet Gynecol Reprod Biol 139: 169–175.\nVereide AB, Kaino T, Sager G, Arnes M, Orbo A (2006) Effect of\nlevonorgestrel IUD and oral medroxyprogesterone acetate on glandular\nand stromal progesterone receptors (PRA and PRB), and estrogen\nreceptors (ER-alpha and ER-beta) in human endometrial hyperplasia.\nGynecol Oncol 101: 214–223.\nWildemeersch D, Janssens D, Pylyser K, De WN, Verbeeck G, Dhont M,\nTjalma W (2007) Management of patients with non-atypical and atypical\nendometrial hyperplasia with a levonorgestrel-releasing intrauterine\nsystem: long-term follow-up. Maturitas 57: 210–213.\nThis work is published under the standard license to publish agree-\nment. After 12 months the work will become freely available and\nthe license terms will switch to a Creative Commons Attribution-\nNonCommercial-Share Alike 4.0 Unported License.\nBRITISH JOURNAL OF CANCER HE4 endometrial hyperplasia, progestin resistance\n730 www.bjcancer.com | DOI:10.1038/bjc.2016.247","source_license":"CC0","license_restricted":false}