Endometrial Complex Atypical Hyperplasia with Extensive Squamous Metaplasia Dr. Frédérique Penault-Llorca Centre Jean Perrin Clermont-Ferrand

Clinical history: A 27 year-old female, 1m68, 54kg, is checked up for a history of infertility. The slides correspond to the curettage product, referred to our institution with a diagnosis of endometrial carcinoma. . Description of the case Clinical data Ultrasound examination showed a thickened endometrium, and normal ovaries. With the exception of infertility longing for 3 years, the patient showed neither abnormal vaginal bleeding nor hyperestrogeny/ virilization symptoms. She did not have any family history of cancer. Gross pathology features Abundant curettage material Histological features: Figure 1 - Clear augmentation of the ratio glads/stroma: hyperplasia. Complex architecture with confluence of endometrial glands (HESX2,5) Figure 2a - False cribriform aspect due to squamous metaplasia and morules (HESX10) Figure 2b - Immunostaining for progesterone (underlines the organoid characteristics of the lesion (squamous metaplasia² is negative) (X10) Figure 3a - Necrosis at the center of the morules (HESX40) Figure 3b - Differential diagnosis: central tumor necrosis in adenocarcinoma (HESX40) Figure 4 - Foci of ciliated metaplasia(HES X20) Figure 5 - Cytonuclear atypia within complex architecture (ACH) (HESX40) Figure 6 - PTEN immunostaining: intense expression in the nuclei of epithelial an stromal cells (X10) HE staining At low magnification, the ratio glands/stroma was high in favor of a hyperplasia. The architecture was complex, with confluence of the endometrial glands mimicking a cribriform aspect (in situ carcinoma) (Figure 1, 2). The glands were closely packed, nevertheless, there is still intervening stroma separating the glands and the architecture is not rigid, such as in carcinomas. At the center of the polyadenoid glands, areas of were seen (figure 3). This aspect of confluence was due to huge foci of squamous metaplasia, organized as morules, with necrotic centers. Aspects of ciliated metaplasia were also found (Figure 4). Cellular atypia were focally present within the glands: round nuclei, hypochromasia or with coarse chromatin (figure 5). Stroma was very scarce. Immunohistochemistry A strong positive staining was found for either estrogen or progesteron receptors in the hyperplasic glands with the exception of squamous metaplasia, which was negative (figure 2b). The cells were still expressing PTEN, which was in favor of the absence of neoplastic transformation of endometrial glandular cells (figure 6). Figure 1 Click to view with ImageScope Click to view with a Web-Based Viewer Diagnosis: Endometrial complex atypical hyperplasia with extensive squamous metaplasia Evolution For this young patient, conservative treatment was chosen (preservation of pregnancy possibilities). A medical treatment with progestative was decided, with a close follow-up. After two years without any recurrence, the treatment was stopped. She became pregnant one year later. Five years after the diagnosis she is well without evidence of disease and mother of a little girl. Discussion This observation underlines the difficulties in differential diagnosis between a complex atypical hyperplasia (CAH) and an endometrial adenocarcinoma, particularly between a CAH with squamous metaplasia and an endometrial adenocarcinoma with squamous metaplasia (formerly adenoacanthoma). Occurring in a very young patient (27 years old), this aspect of squamous metaplasia with ACH is rather unusual and confusing. Difficulties in differential diagnosis between CAH and adenocarcinoma Prevalence of endometrial carcinoma before age 40 The prevalence of endometrial carcinoma before 40 is very low (3-5%). Therefore no one should underestimate an endometrial lesion occurring in a young woman on the basis of the rarity of adenocarcinoma at this age. As usual, the diagnosis must be morphological and not clinical. Epidemiological data from the series of Kaminski et Stevens, showed that among 460 women before 40 years old explored by curettage for abnormal vaginal bleeding, none of them presented with CAH or adenocarcinoma. Malignancy criteria are mainly architectural In presence of large foci of CAH, as in this case, we must exclude the differential diagnosis of endometrioid adenocarcinoma. The elements in favor of this latter diagnosis are a cribriform, rigid aspect of the glands with confluence ("glands within the glands", epithelial bridges, absence of stroma), abrupt necrosis. In the present case, the cribriform aspect is not rigid, the glandular cavities are not widely opened, the "stream" is obvious (Figure 1, 2). One must look for either infiltrative aspects, frequently absent and when present, focal and tiny (micro invasion), or round glands, forming a solid mass. The presence of a desmoplastic stroma, of polymorphonuclear leucocytes and nuclear debris or necrosis in glandular lumina, of real pluristratification of the nuclei are minor signs of cancer. In absence of an immunohistochemical or molecular parameter specific of CAH or endometrial cancer, the diagnosis must be morphological. Squamous metaplasia In this case, squamous metaplasia is organized into large morules (immature round islets feeling the endometrial glands. At low magnification the aspect is organoid (figure 2). The morules are at the origin of the confluent aspect of the proliferation (solid sheets). Those squamous morules are frequently accompanying CAH. The large foci of necrosis within squamous clusters should not lead to a diagnosis of malignancy. Sometimes extensive necrosis might be found at the center of the morules without pejorative meaning (figure 3a). This necrosis is different from a tumor necrosis by the absence of an abrupt aspect, in glands without squamous metaplasia (figure 3b). Indeed, when growing in sheets and especially if showing central necrosis, such as in this case, morules may resemble poorly differentiated solid adenocarcinoma. But consideration of the bland nature of the morular cells and their lack of mitotic activity should lead to the avoidance of a diagnosis error. Squamous metaplasia is the most frequent metaplasia of the endometrium. It is associated to 25% of the adenocarcinoma, and more frequently to endometrial hyperplasia. Squamous metaplasia can also be associated to vitamin A deficits, chronic infections, and mechanic irritations. It can be induced by progestative treatment. In the FIGO grade for endometrioid carcinoma, the foci of squamous metaplasia should not be taken into account as a solid zone. Furthermore, as the distinction between the benign or malignant squamous metaplasia associated to an adenocarcinoma is difficult and poorly reproducible, the new WHO edition recommends the use of carcinoma with squamous metaplasia instead of either adenoacanthoma (squamous component benign) or adenosquamous carcinoma (squamous component malignant). Risk factors In this context of primary infertility, the diagnosis raised was an anovulation leading to a hyperestrogenia. One should exclude an underlying disease leading to hyperoestrogenia. The diagnosis to be excluded at this age are: granulosa cell tumor of the ovary (9 to 13% of those patients will have an endometrial carcinoma); a thecoma (associated to up to 20% of adenocarcinoma), and in case of a primary infertility such as in this observation: polycystic ovaries syndrome (raising estrogens and androgens blood levels), associated to 5% of endometrial carcinoma and to the majority of the adenocarcinomas before 40. None of the former entities were found in our patient. Natural history of CAH occurring before 40 years old. Consequences for patients' management. Almost 25% of atypical hyperplasia will end up as a well-differentiated adenocarcinoma of good prognosis within a mean of 4 years. The risk of occurrence of endometrial adenocarcinoma is low before 40. Kurman et al (1985), from a series of 170 patients with non-treated endometrial hyperplasia, underlined the group of 11 patients with ACH with a malignant evolution. Among those 11 patients, 8 were under 40. They all presented with risk factors of endometrial carcinoma: half of them had a polycystic ovaries syndrome and the other half were obese. The evolution to cancer occurred within a mean time of 3.3 years (as opposed to 7.8 for patients older than 40). All the cancers were well differentiated, in situ or with superficial invasion. A series from Kurman et Norris (1982), reported 52 patients younger than 40, with a diagnosis raised on curettage or biopsies. Seventeen (32%) had a diagnosis of atypical hyperplasia, among them two showed well-differentiated adenocarcinoma, grade I and intra epithelial on the hysterectomy specimen. Among the 35 patients with adenocarcinoma on biopsies or curettage, 37% (13/35) still had adenocarcinoma on the surgical specimen, 10/13 being infiltrative. These results showed that endometrial adenocarcinoma associated with ACH are well differentiated and poorly aggressive, even before age 40. A recent study from Randall et Kurman (1997) showed that a regression of ACH or well-differentiated adenocarcinoma is found for 75% of the women with adenocarcinoma and 95% with ACH. (1997), following a progestative treatment . The management of young patients with such lesions, in the absence of known risk factors, should be not be surgical but medical. The lengh of treatment varies from 3 to 9 months. Pregnancy is possible later in case of response. The chances of pregnancy were of 13% in old series but now rose up to 50% with PMA. Keys for the diagnosis Confluent endometrial glands, not rigid Aspect of atypical hyperplasia Squamous metaplasia enhancing the aspects of confluence Organoid aspect Key messages Do not overdiagnose the solid aspects due to squamous metaplasia Central and massive necrosis is a classical feature in those foci of squamous metaplasia Observe the lesion at low magnification (organoid aspect) Young age is reassuring but the occurrence of endometrial adenocarcinoma is possible in a context of POD A medical treatment has to be proposed to preserve the chances pregnancy. Bibliography Anderson MC, Robboy SJ, Russel P, Morse A. Endometrial hyperplasia. In pathology of the female reproductive tract. S Robboy, MC Anderson, P Russel edts Churchill Linvingstone, 2002; 305-330 Gotlieb WH, Beiner ME, Shalmon B, Korach Y, Segal Y, Zmira N, Koupolovic J, Ben-Baruch G. Outcome of fertility-sparing treatment with progestins in young patients with endometrial cancer. Obstet Gynecol . 2003 ;102(4):718-25. Kaminski PF, Stevens CW. The value of endometrial sampling in abnormal uterine bleeding. AM J Gynecol Health 1985; II: 33-36 Kurmann RJ, Norris HJ. Evaluation of criteria for distinguishing atypical endometrial hyperplasia from well differentiated adenocarcinoma. Cancer 1982; 49: 2547-2559 Kurman RJ, Kaminski PF, Norris HJ. The behavior of endometrial hyperplasia. A long-term study of "untreated" hyperplasia in 170 patients. Cancer 1985 15; 56(2):403-12. Randall TC, Kurman RJ. Progestin treatment of atypical hyperplasia and well-differentiated carcinoma of the endometrium in wommen under age 40. Obstet Gynecol 1997; 90: 434-440