Human Pathology (2014) 45, 2417–2422
www.elsevier.com/locate/humpath
Original contribution
Ovarian Brenner tumors and Walthard nests: a histologic and immunohistochemical study☆,☆☆ Andres A. Roma MD a,⁎, Ramya P. Masand MD b a
Department of Anatomic Pathology, Robert J. Tomsich Pathology and Laboratory Medicine Institute, Cleveland Clinic, Cleveland, OH, 44195 b Department of Pathology & Immunology, Baylor College of Medicine, Houston, TX, 77030 Received 25 June 2014; revised 31 July 2014; accepted 5 August 2014
Keywords: Brenner tumor; Walthard nests; GATA3; PAX8; Ovarian tumor
Summary Brenner tumors are composed of urothelial/transitional-type epithelium and, hence, are morphologically similar to Walthard nests and tubal/mesothelial transitional metaplasia. In this study, we analyzed immunohistochemical markers on Brenner tumors to explore Müllerian as well as Wolffian and germ cell derivation. We also attempted to explore their possible association with tubal/paratubal Walthard nests/transitional metaplasia, using the same immunostains. Thirty-two consecutive cases of Brenner tumors were identified. Thirteen (43%) of the patients had Walthard nests in the tubal/periovarian soft tissue. All Brenner tumors were diffusely positive for GATA3 (strongly positive in 30/32 and weakly positive in the remaining 2) and negative for PAX8, PAX2, and SALL4. Similarly, all Walthard nests were positive for GATA3, whereas only 3 (23%) of 13 showed occasional PAX8 expression; all were negative for PAX2 and SALL4. In our study, more than 40% of Brenner tumors had associated Walthard nests. The similar morphology and immunoprofile of Brenner tumors and Walthard nests suggest a probable link between Brenner tumors and Walthard nests. Two additional cases presented highlight small transitional lesions involving the ovary: a possible precursor lesion or the initial steps of Brenner tumor formation. Brenner tumors and most Walthard nests lacked staining for Müllerian (PAX8 and PAX2) and germ cell tumor markers (SALL4). © 2014 Elsevier Inc. All rights reserved.
1. Introduction Brenner tumors are composed of urothelial/transitionaltype epithelium and, hence, are morphologically similar to Walthard nests and tubal/mesothelial transitional metaplasia. Although literature data proposed derivation of Brenner
☆
Presented at 103rd United States and Canadian Academy of Pathology Annual Meeting, March 2014, San Diego, CA. ☆☆ Disclosures: No funding or conflicts of interest to disclose. ⁎ Corresponding author. 9500 Euclid Ave, L2, Cleveland, OH 44122. E-mail address:
[email protected] (A. A. Roma). http://dx.doi.org/10.1016/j.humpath.2014.08.003 0046-8177/© 2014 Elsevier Inc. All rights reserved.
tumors from the ovarian surface epithelium through transitional cell metaplasia [1,2], this theory has been recently challenged, and other sites of origin such as Walthard nests, teratomas, and fallopian tube epithelium have been suggested [3,4,5]. In this study, we analyze several immunohistochemical stains in Brenner tumors including PAX8 and PAX2, markers of Müllerian as well as Wolffian derivation, SALL4, marker of germ cell origin, and GATA3, a multispecific marker that also stains mesonephric lesions. We also attempted to explore their possible association with tubal/paratubal Walthard nests/transitional metaplasia, using the same immunostains. In addition, we highlight findings of
2418 very small Brenner tumors that might shed some light into its histogenesis.
2. Materials and methods After institutional review board approval, a cohort of 34 cases was studied. Thirty-two of them were consecutive Brenner tumors identified and retrieved from a database search from July 2008 until June 2013. Two additional cases were identified during preparation of the manuscript and were included based on interesting findings. Histologic slides were re-reviewed in all cases, and the diagnosis of Brenner tumor was confirmed. In addition, the presence of concurrent Walthard nests in the sections of fallopian tube was documented. Formalin-fixed, paraffin-embedded tissue blocks from the most representative histologic sections were retrieved from the archives. A postcut hematoxylin and eosin slide was obtained for each case after obtaining unstained slides for immunostains to confirm the presence of lesional cells. Immunohistochemical staining for PAX8 (rabbit polyclonal, 1:200 dilution; Protein Tech Group, Chicago, IL), PAX2 (clone 6E3, 1:400 dilution; Sigma-Aldrich, St Louis, MO), SALL4 (rabbit monoclonal, Ep3251, 1:50 dilution; Epitomics, Burlingame, CA), and GATA3 (L2-823 clone, mouse monoclonal, 1:100 dilution; Biocare, Concord, CA) were performed. In the 2 additional cases, immunostains for PAX8, GATA3, calretinin, and WT1 were also performed. The immunohistochemical staining was detected with Ventana's i-view DAB Streptavidin-Biotin based detection kit (Ventana Medical Systems, Tucson, AZ). Positive and negative controls were included with each run. Nuclear labeling was considered as positive stain. The immunostaining pattern was recorded as follows: negative or positive, with distribution of staining as diffuse (N80%) or focal (percentage documented). Staining intensity was also documented as weak, moderate, or strong.
3. Results 3.1. Clinical and histologic features of study cases Thirty-two consecutive cases of Brenner tumors from 30 patients were identified. Patients' ages ranged from 36 to 79 years (mean, 61.4 years; median, 61.5 years). Reasons for the surgery included pelvic mass in 19 patients, endometrial carcinoma in 4, BRCA-positive prophylactic salpingo-oophorectomy in 3 (2 with personal history of breast carcinoma), and 1 case each of pelvic pain, prolapse, myomas, and bowel ischemia. In 14 (47%) cases, the tumor was located in the right ovary, whereas in 13 (43%) cases, the tumor was located in the left ovary. There were 2 bilateral
A. A. Roma, R. P. Masand Brenner tumors, and in 1 case, it was not possible to determine laterality because the specimen was not oriented by the surgeon and the Brenner tumor was not grossly identified. Tumor size ranged from 0.2 to 32 cm (mean, 7.65 cm; median, 2.7 cm). Four tumors were not seen on gross examination (0.2-0.3 cm) and were incidental findings on microscopic examination. The remaining tumors consisted of firm yellow-tan solid masses. In 7 of the larger tumors, a cystic component was grossly identified in addition to solid areas. Brenner tumor was seen in 1 to 16 of the submitted sections (mean, 4.6; median, 3). In 12 (37%) cases, the tumor was small and located at the hilum or involving the hilum; one of these also in part extended to paratubal soft tissue. In 5 (16%) cases, the tumor involved the ovarian parenchyma and not the hilum. In 15 (47%) cases, these being the largest, it was not possible to determine if the lesion involved the hilum or was confined to the ovarian parenchyma. In 8 cases, a mucinous tumor was present in association with the Brenner tumor (4 cystadenomas, 3 borderline tumors, and 1 carcinoma). Additional findings included independent mature cystic teratoma in the same ovary, teratoma in the contralateral ovary, serous cystadenoma, serous cystadenofibroma, cervical high-grade squamous dysplasia, adenocarcinoma in situ, and vulvar high-grade squamous dysplasia, 1 case each. Morphologically, the Brenner tumors were composed of nests of transitional cells embedded in a dense fibrous tissue, containing spiculated calcification (41%), central small lumina or cysts with intracystic proteinaceous (97%), and cysts lined by flat attenuated epithelium (78%; Fig. 1A). Of particular interest were 2 additional cases encountered during preparation of the manuscript. In one of the cases, the nests of transitional epithelium were intimately associated with inclusion cysts lined by mesothelium-like epithelium and rete ovarii structures at the hilum of the ovary (Fig. 1B and C). The mesothelium-like inclusion cysts appeared to originate from the mesothelial/ovarian epithelial lining at the edge of the ovary, close to the fallopian tube interface, that while incorporating into the ovary, acquired features of rete ovarii as well as transitional metaplasia within the cysts; Walthard nests were seen in the fallopian tube serosa in the vicinity. In the other case, the transitional epithelium was seen within mesothelium-like cysts in the interface between the ovary and fibrous adhesions, as well as in part lining the surface of the ovary, minimally involving the ovarian cortex (Fig. 1D). Walthard nests were also encountered in the fallopian tube on this latter case. Neither of these cases elicited significant stromal reaction.
3.2. Brenner tumors with concomitant Walthard nests Thirteen (43.3%) patients with Brenner tumors had Walthard nests in the paratubal/paraovarian soft tissue. Representative fallopian tube sections were submitted in most cases, with the fimbria submitted in all cases. Only 3
Staining patterns of Brenner tumors and Walthard nests
2419
Fig. 1 A, Brenner tumor with different morphologic features including central dilation of transitional nests and gland-like structures lined by transitional cells and cysts with thin transitional epithelium and proteinaceous cyst content resembling paratubal cysts (hematoxylin and eosin, ×100). B, First additional case with cysts resembling rete ovarii and transitional metaplasia that appears to arise from ovarian surface epithelium (hematoxylin and eosin, ×40). C and D, Second additional case with mesothelium-lined cysts undergoing transitional metaplasia (C: ×40; D: ×200).
cases in patients with BRCA-positive prophylactic salpingooophorectomy had the entire tube submitted, all 3 with Walthard nests. Patient age ranged from 44 to 78 years (mean, 60.7 years; median, 63 years) and did not differ from those without Walthard nests. Similarly, there was no difference in size among the Brenner tumors with or with no identifiable Walthard nests. Many Walthard nests showed areas of central dilation of the transitional nest as well as larger cysts lined by attenuated transitional/squamous epithelium, similar to that seen in Brenner tumors. In all but 1 case, the location of the Walthard nests was closer to the fallopian tube serosa and did not appear to arise from the mucosa. One case showed areas of transitional metaplasia lining the tubal epithelium.
Table
Immunohistochemical staining results are summarized in the Table. All Brenner tumors had diffuse immunoreactivity (strongly positive in 30/32 and weakly positive in 2/32) for GATA3 in the tumor cell nuclei. Similarly, all Walthard nests were diffusely and strongly positive for GATA3. Brenner tumors and Walthard nests were immunohistochemically negative for PAX2 and SALL4 in all cases. Although Brenner tumors were negative for PAX8 in all cases, 3 (23%) of 13 Walthard nests showed less than 10% of cells with positive nuclei for PAX8, mostly at the periphery of the transitional nests as well as the basal epithelium in the
Staining patterns of ovarian Brenner tumors and Walthard nests Walthard nests (13)
PAX8 PAX2 SALL4 GATA3
3.3. Immunohistochemistry in Brenner tumors and Walthard nests
Brenner tumors (32)
Positive cases
Intensity of staining
Positive cases
Intensity of staining
3 0 0 13
Focal and weak N/A N/A Diffuse and strong
0 0 0 32
N/A N/A N/A 30 Diffuse and strong 2 Weak
Abbreviation: N/A, not applicable.
2420
A. A. Roma, R. P. Masand
Fig. 2 A, Diffuse expression of GATA3 in Brenner tumor (×100). B, Negative expression of PAX8 (×100) in Brenner tumor. C, Walthard nests, some with cystic dilation similar to that seen in Brenner tumors (hematoxylin and eosin, ×40). D, Diffuse expression of GATA3 in Walthard nests, while tubal epithelium is negative (×40). E, Lack of PAX8 expression in Walthard nests but positive expression in tubal epithelium (×40). F-H, Immunostains in area of transitional metaplasia arising from mesothelial-line cyst at ovarian surface and superficially involving ovarian cortex (F: GATA3, ×100; G: PAX8, ×100; H: calretinin, ×100).
case of transitional metaplasia of the tubal/mesothelial interface (Fig. 2A-E). In the 2 additional cases, similarly, PAX8 was negative, whereas GATA 3 was diffusely positive in the transitional epithelium. There was a progressive
decrease in staining intensity and quantity of WT1 in the cells from the surface epithelium into the metaplasia. Calretinin, in contrast, was abruptly lost in the transitional metaplasia (Fig. 2F-H).
Staining patterns of Brenner tumors and Walthard nests
4. Discussion Brenner tumors are composed of nests of transitional/ urothelial type epithelium, embedded in a dense fibromatous stroma. It is hypothesized that the transitional epithelia of these tumors arise from metaplasia of the ovarian surface epithelium. Walthard nests are benign nests of transitional epithelium found in the paratubal region [1,2]. Because of the morphologic resemblance, we sought to explore the association between Brenner tumors and Walthard nests. Our findings support a similar immunoprofile between transitional metaplasia and/or Walthard nests and Brenner tumor. Several lesions in the gynecological tract share a common morphology with urothelial or transitional epithelium including Brenner tumors, Walthard nests, and transitional metaplasia of the fallopian tube epithelium or mesothelium; however, a common origin for these entities has not yet been proven. There has been considerable debate as to whether these entities show true urothelial differentiation. Although most studies showed commonalities with a urothelial phenotype, such as similar expression pattern of cytokeratin 7, CA19-9, CEA, uroplakin III, and thrombomodulin [6–8], differences in cytokeratin 20 and WT1 expression with urothelium were reported [6]. More recently, p63, S100P, and GATA3 immunostain were demonstrated in Brenner tumor, but extraovarian Walthard nests did not share identical immunoprofile, lacking expression of S100P and showing focal weak GATA3 expression [9]. A recent article published after the culmination of our study demonstrated lack of PAX2 and PAX8 expression in Brenner tumors, and focal expression of these Müllerian markers in Walthard nests and transitional metaplasia, similar to our findings [4]. Based on their results, Kuhn et al [4] suggested that the junction between the epithelium of the fimbria and the mesothelial serosa of the tube, the so-called tuboperitoneal junction, might be a ‘hot spot’ for carcinogenesis. We used several markers of differentiation in the various urothelial/transitional lesions in the female genital tract. GATA binding protein 3 is a zinc finger transcription factor with varied biological roles including T-cell development, mammary gland development, regulatory and required in nephric (Wolffian) duct morphogenesis, and mesonephric duct guidance [10]. It is an immunohistochemical marker expressed predominantly in breast and urothelial carcinomas. It is also expressed in other tumors such as choriocarcinoma and yolk sac tumors, autonomic nervous system tumors, approximately half of chromophobe renal cel carcinomas, and mesotheliomas [11–13]. In addition, we have found it to be expressed in mesonephric lesions of the lower gynecological tract [14,15]. It has been consistently shown to be expressed in Brenner tumors, as in our study [9]; however, its lack of specificity precludes its use in determination of site of origin. Nevertheless, it is very useful to differentiate primary from metastatic tumors in the breast and urinary tract, as well as in the workup of tumors of unknown primary.
2421 Additional markers used in our study were PAX2 and PAX8, members of the mammalian paired box genes 1 to 9 encoding transcription factors involved in development of the thyroid, kidney, and Müllerian organs [16]. Germ cell tumors and sex cord–stromal tumors have been consistently negative [17,18]. The focal and basal expression of Walthard nests with PAX8 and complete loss in Brenner tumors might add to the association of these lesions, whereas the loss of PAX8 expression in tubal epithelium or proliferation of PAX8negative cells might be an important step during urothelial metaplasia of the fallopian tube/peritoneum. SALL4 was the last marker systematically analyzed to determine germ cell phenotype in these lesions because it is expressed in most germ cell tumors [19]. SALL4 has been consistently expressed in dysgerminoma, embryonal carcinoma, yolk sac tumors, gonadoblastoma, and up to 73% of immature teratomas, whereas it was negative in primary mature teratomas [19]. A subsequent study revealed diffuse expression in metastatic germ cell tumors, most metastatic choriocarcinomas (71%), and 50% of metastatic teratomas with mature elements only [20]. Still, its paucity of expression in mature teratomas precludes an analysis of the hypothesis that teratomas and Brenner tumors are related. Comparing the clinicopathological features of Brenner tumor with or with no identifiable Walthard nests, there were no significant differences. An interesting finding was that 37% of Brenner tumors involved the hilum and, in 1 case, the paratubal soft tissue, whereas larger tumors replaced the ovary, probably also involving the hilum. Similar morphologic features were seen in ovarian Brenner tumors and Walthard nests, including nests with central small lumina or cysts lined by transitional cells, cystic structures lined by epithelium with transitional or squamous features and intracystic proteinaceous material. Some of these latter cysts might be labeled as paratubal cysts by some pathologists if located in the paratubal region; however, they were seen within the ovarian Brenner tumor (Figs. 1A and 2C). Calcifications, however, were not seen in Walthard nests and are occasionally seen in Brenner tumors. Seidman and Khedmati [3] identified that Walthard nests are 79% more common in women with Brenner tumors compared with controls. Although Walthard nests are a common occurrence in the fallopian tube, extraovarian Brenner tumors, especially those exclusively located in the fallopian region tube are not a common occurrence [2]. A possible explanation is that the ovarian stroma is involved and/ or can induce proliferation of tumor cells [21]. When these metaplastic cells come in contact with the ovarian stroma, they proliferate forming tumors or masses, whereas when only outside the ovary, they remain simply as small benign lesions. To further elucidate the origin of Brenner tumors, we included 2 cases of transitional-type epithelium involving the ovary that we encountered during preparation of this manuscript. These lesions lacked the classic fibromatous stroma of Brenner tumors but were beginning to involve the ovary and could be very early forms of ovarian Brenner
2422 tumors. Both lesions were associated with cysts that were lined by flat cells that were positive for WT1, calretinin, and weakly positive for PAX8 but negative for GATA3. Weak expression of PAX8 has been reported in mesotheliomas [22]. Thus, the morphology as well as the immunostains could support a mesothelial origin for these cysts. Our findings might point to the fallopian tube serosa or tubal ovarian interface epithelium as the primary site of Walthard nests, as previously proposed by Kuhn et al [4]. However, given that we and others have observed transitional metaplasia involving tubal mucosal epithelium, additional cases would be required to validate this conclusion. Walthard nests/transitional metaplasia are most likely not precursor lesions in themselves, but the same phenomenon within and outside the ovary. We believe that these earlier cases of ovarian transitional epithelium could be the initial steps toward a Brenner tumor and can help understand its histogenesis. Brenner tumors have been associated with other tumors including ovarian teratomas; however, lack of a consistent immunohistochemical marker for mature teratoma makes this analysis difficult [23]. We attempted to address this problem using SALL4, an inconsistent marker in mature teratomas; however, we did not find any expression in Brenner tumors. In summary, 40% of Brenner tumors in our study were encountered in patients who also harbored Walthard nests in the adjacent or contralateral fallopian tube, despite only partial sampling of the tube. Brenner tumors were negative for Müllerian and germ cell tumor markers and only expressed GATA3. Similarly, Walthard nests were mainly positive for GATA3 and negative for the remaining markers, except for focal expression of PAX8 in the basal cells of some groups of Walthard nests and transitional metaplasia. The similar morphology and immunoprofile of Brenner tumors and Walthard nests/transitional metaplasia suggest a link between these 2 lesions. These findings also indicate that the loss of Müllerian marker PAX8 in Walthard nests may be a step prior to development of Brenner tumor. Although we speculate that transitional metaplasia is related to Walthard nests and has a role in Brenner tumor development, this study was not designed to address origin of these lesions exclusively from the fallopian tube/mesothelium but to analyze its immunophenotype and report on rare small lesions that can help with our understanding of these tumors.
References [1] Lee KR, Tavassoli FA, Prat J, et al. Tumors of the ovary and peritoneum. In: Tavassoli FA, Devilee P, editors. World Health Organization classification of tumours. Pathology and genetics of tumours of the breast and female genital organs. Lyon: IARC Press; 2003. p. 113-45. [2] Scully RE, Young RH, Clement PB. Atlas of tumor pathology. Tumors of the Ovary, Maldeveloped Gonads, Fallopian Tube, and Broad Ligament. Third Series. Fascicle 23. Washington, DC: Armed Forces Institute of Pathology; 1998. p. 153-64.
A. A. Roma, R. P. Masand [3] Seidman JD, Khedmati F. Exploring the histogenesis of ovarian mucinous and transitional cell (Brenner) neoplasms and their relationship with Walthard cell nests: a study of 120 tumors. Arch Pathol Lab Med 2008;132:1753-60. [4] Kuhn E, Ayhan A, Shih IeM, et al. Ovarian Brenner tumour: a morphologic and immunohistochemical analysis suggesting an origin from fallopian tube epithelium. Eur J Cancer 2013;49:3839-49. [5] Stohr G. The relationship of the Brenner tumor to the rete ovarii. Am J Obstet Gynecol 1956;72:389-99. [6] Logani S, Oliva E, Amin MB, et al. Immunoprofile of ovarian tumors with putative transitional cell (urothelial) differentiation using novel urothelial markers: histogenetic and diagnostic implications. Am J Surg Pathol 2003;27:1434-41. [7] Cuatrecasas M, Catasus L, Palacios J, et al. Transitional cell tumors of the ovary: a comparative clinicopathologic, immunohistochemical, and molecular genetic analysis of Brenner tumors and transitional cell carcinomas. Am J Surg Pathol 2009;33:556-67. [8] Soslow RA, Rouse RV, Hendrickson MR, et al. Transitional cell neoplasms of the ovary and urinary bladder: a comparative immunohistochemical analysis. Int J Gynecol Pathol 1996;15:257-65. [9] Esheba GE, Longacre TA, Atkins KA, et al. Expression of the urothelial differentiation markers GATA3 and placental S100 (S100P) in female genital tract transitional cell proliferations. Am J Surg Pathol 2009;33:347-53. [10] Grote D, Souabni A, Busslinger M, et al. Pax 2/8-regulated Gata 3 expression is necessary for morphogenesis and guidance of the nephric duct in the developing kidney. Development 2006;133:53-61. [11] Ordóñez NG. Value of GATA3 immunostaining in tumor diagnosis: a review. Adv Anat Pathol 2013;20:352-60. [12] Nonaka D, Wang BY, Edmondson D, et al. A study of gata3 and phox2b expression in tumors of the autonomic nervous system. Am J Surg Pathol 2013;37:1236-41. [13] Miettinen M, McCue PA, Sarlomo-Rikala M, et al. GATA3: A multispecific but potentially useful marker in surgical pathology: a systematic analysis of 2500 epithelial and nonepithelial tumors. Am J Surg Pathol 2014;38:13-22. [14] Roma AA. Mesonephric carcinosarcoma involving uterine cervix and vagina. Report of two cases with immunohistochemical positivity for PAX2, PAX8 and GATA-3. Int J Gynecol Pathol (in press). [15] Roma AA, Goyal A, Yang B. Differential expression pattern of GATA3 among mesonephric lesions and endocervical and endometrial adenocarcinomas. Mod Pathol 2014;27(Suppl. 2):303A-4A. [16] Bowen NJ, Logani S, Dickerson EB, et al. Emerging roles for PAX8 in ovarian cancer and endosalpingeal development. Gynecol Oncol 2007; 104:331-7. [17] Laury AR, Perets R, Piao H, et al. A comprehensive analysis of PAX8 expression in human epithelial tumors. Am J Surg Pathol 2011;35:816-26. [18] Tong GX, Yu WM, Beaubier NT, et al. Expression of PAX8 in normal and neoplastic renal tissues: an immunohistochemical study. Mod Pathol 2009;22:1218-27. [19] Cao D, Guo S, Allan RW, et al. SALL4 is a novel sensitive and specific marker of ovarian primitive germ cell tumors and is particularly useful in distinguishing yolk sac tumor from clear cell carcinoma. Am J Surg Pathol 2009;33:894-904. [20] Cao D, Humphrey PA, Allan RW. SALL4 is a novel sensitive and specific marker for metastatic germ cell tumors, with particular utility in detection of metastatic yolk sac tumors. Cancer 2009;115:2640-51. [21] Silva EG. The stromal origin of some epithelial ovarian neoplasms: “fere ex nihilo”. Int J Gynecol Cancer 2012;22:906-7. [22] Laury AR, Hornick JL, Perets R, et al. PAX8 reliably distinguishes ovarian serous tumors from malignant mesothelioma. Am J Surg Pathol 2010;34:627-35. [23] Nomura K, Aizawa S. A histogenetic consideration of ovarian mucinous tumors based on an analysis of lesions associated with teratomas or Brenner tumors. Pathol Int 1997;47:862-5.