Main

Tubal (or ciliated cell) metaplasia of the endometrium is a frequent finding in endometrial sampling specimens and is commonly associated with the follicular phase of the menstrual cycle and with anovulatory cycles.1 Characterized by ciliated columnar cells with bland round nuclei and eosinophilic cytoplasm, similar to the cells normally seen lining the fallopian tube, it is usually easily recognized. Occasionally, tubal metaplasia with cytologic atypia (atypical tubal metaplasia) is encountered in non-neoplastic endometrial biopsy or curettage specimens. This cytologic atypia is most likely reactive or degenerative in nature; however, it does raise the possibility of a poorly sampled invasive uterine serous carcinoma or endometrial intraepithelial carcinoma. Uterine serous carcinoma and serous endometrial intraepithelial carcinoma are both characterized by highly atypical nuclei with macronucleoli, although the latter is by definition entirely non-invasive.2 The identification of invasive or intraepithelial serous carcinoma in endometrial samplings is often associated with early extrauterine disease3, 4 and therefore warrants hysterectomy and surgical staging procedures. On the other hand, the clinical significance of atypical tubal metaplasia is unknown. Previous studies have shown significant overexpression of p53, Ki-67, and human telomerase reverse transcriptase (TERT) in invasive and non-invasive uterine serous carcinoma.5, 6, 7, 8 The current study evaluates the expression of p53, Ki-67, and TERT in atypical tubal metaplasia, and the usage of these immunohistochemical markers in differentiating this metaplasia from intraepithelial or invasive uterine serous carcinoma. This study also uses long-term follow-up to evaluate the risk of subsequent endometrial neoplasia associated with atypical tubal metaplasia.

Materials and methods

The archives of the department of pathology at Women and Infants Hospital of Rhode Island, Alpert Medical School of Brown University, were searched from January 2001 through August 2008. A total of 63 cases of benign endometrial biopsies or curettages with histologically demonstrable atypical tubal metaplasia were identified and reviewed. The specimens were all from patients with dysfunctional uterine bleeding and include 30 poorly active endometrium, 16 atrophic endometrium, 2 weakly proliferative endometrium, 3 disordered proliferative endometrium, 8 disintegrating endometrium, and 4 endometrial polyp cases. Endometrial hyperplasia or malignancy was not identified in any of these study cases. The criteria for cytologic atypia of atypical tubal metaplasia included enlarged pleomorphic and hyperchromatic nuclei, small to prominent nucleoli, and increased nuclear to cytoplasmic ratio (Figure 1). The mean age of patients with atypical tubal metaplasia was 56 years. Follow-up information was available in all cases. As controls, 200 cases of benign endometrium from patients with dysfunctional uterine bleeding were selected from the first 2 months of 2001. Each control case had tubal metaplasia without cytologic atypia. The mean age of the control group was 54 years. Unpaired Student t-test was used to compare the age and follow-up period for the two groups; Fisher Exact Test was used to compare outcome.

Figure 1
figure 1

(a) Tubal metaplasia with bland uniform nuclei (40 × ). (b, c) Tubal metaplasia with mild atypia seen as slight variation in nuclear size, hyperchromatism, and presence of small nucleoli (40 × ). (d–f) Atypical tubal metaplasia displaying marked nuclear pleomorphism, hyperchromatic smudgy nuclei, and prominent nucleoli (d, f—40 × ; e—20 × ).

Full size image

Of the 63 cases of atypical tubal metaplasia, 16 cases were arbitrarily selected for further immunohistochemical evaluation with antibodies to p53 (Dako, Carpinteria, CA, USA), Ki-67 (Dako), and TERT (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Immunohistochemistry was carried out manually using protocols optimized and validated under the conditions of our laboratory. From each case, 5 μm sections were cut from formalin-fixed, paraffin-embedded tissue blocks, then deparaffinized and rehydrated. Immunohistochemistry for p53 (1:50 dilution), Ki-67 (ready-to-use antibody), and TERT (1:40 dilution) was carried out with antigen retrieval using a water bath in citrate buffer (pH 6.0) and detection via the avidin-biotin peroxidase method of the DAKO LSAB2 system. Protein expression was visualized with diaminobenzidine, and sections were then counterstained and mounted. Positive p53 and Ki-67 results were indicated by brown staining of the nucleus. For TERT, brown distinct cytoplasmic and/or nuclear staining was considered positive. Appropriate positive and negative controls were run simultaneously with each batch of slides.

The immunohistochemical staining pattern of p53 and Ki-67 in the selected 16 atypical tubal metaplasia cases was compared with typical tubal metaplasia identified in other regions of the same 16 slides. TERT immunoreactivity in the selected 16 atypical tubal metaplasia cases was compared with immunoreactivity in 13 cases of uterine serous carcinoma, which were also retrieved from our department's archives and stained for TERT as described above. Mann–Whitney U Test was used to compare staining characteristics of the two groups.

Results

The clinical characteristics and outcomes of the two groups are summarized in Table 1. The patients in the atypical tubal metaplasia group (n=63) were followed for a mean of 64 months. Follow-up revealed three cases of simple hyperplasia, one case of complex atypical hyperplasia, and one case of moderately differentiated endometrioid adenocarcinoma. The lone carcinoma case was diagnosed 88 months after the initial biopsy showing atypical tubal metaplasia. The patients in the control group were followed for a mean of 61 months, which revealed seven cases of hyperplasia without atypia (four simple hyperplasia, three complex hyperplasia), one simple atypical hyperplasia, and three complex atypical hyperplasia.

Table 1 Characteristics of atypical tubal metaplasia group vs control group
Full size table

Long-term follow-up revealed hyperplasia without atypia in 3 of 63 cases (5%) in the atypical tubal metaplasia group, and 7 of 200 cases (4%) in the control group (P=0.44). Atypical hyperplasia or carcinoma was present in 2 of 63 (3%) atypical tubal metaplasia cases and in 4 of 200 (2%) control cases (P=0.44). There was no significant difference in the incidence of hyperplasia without atypia or atypical hyperplasia/carcinoma between the two groups.

p53

The atypical tubal metaplasia cells were negative for p53 in 4 of 16 cases (25%) and showed focal, weak p53 immunoreactivity (Figure 2) in 12 of 16 cases (75%). Typical tubal metaplasia was negative for p53 in 3 of 16 cases (19%) and showed focal, weak positivity in 13 of 16 cases (81%). There was no significant difference in the staining patterns of p53 between atypical and typical tubal metaplasia (P>0.05). These results are summarized in Table 2.

Figure 2
figure 2

Tubal metaplasia with scattered Ki-67-positive cells (a) and negative p53 (b). Atypical tubal metaplasia showing scattered Ki-67 positivity (c, e), although many atypical nuclei are negative (arrows). p53 is negative (d) or focally weakly positive (f) in ATM. (a–f–40 × ).

Full size image
Table 2 Immunoexpression of p53 and Ki-67 in atypical vs typical tubal metaplasia
Full size table

Ki-67

Of the 16 selected atypical tubal metaplasia cases, eight (50%) were completely negative for Ki-67 immunostain. In seven cases (43%), less than 5% of the atypical cells were positive for Ki-67 (Figure 2). One case (6%) exhibited immunoreactivity in ∼25% of the atypical cells. In the regions of typical tubal metaplasia, six cases (37.5%) were completely negative, nine cases (56.2%) showed less than 5% positivity, and one case (6.3%) showed greater than 5% positivity for Ki-67. As shown in Table 2, there seemed to be no significant difference in Ki-67 staining between atypical and typical tubal metaplasia (P>0.05).

Human telomerase reverse transcriptase (TERT)

The atypical tubal metaplasia cells were completely negative for TERT (Figure 3) in the 16 selected cases (0/16). Of the 13 cases of uterine serous carcinoma, six (46%) showed diffuse strong immunoreactivity with TERT (Figure 3) and three (23%) exhibited moderate TERT immunoreactivity. Only four cases (31%) showed focal, weak immunostaining, and no cases were completely negative for TERT. These results, as summarized in Table 3, indicate a significant difference in TERT staining between atypical tubal metaplasia and uterine serous carcinoma (P<0.0001).

Figure 3
figure 3

Atypical tubal metaplasia with pleomorphic, hyperchromatic nuclei (a) is negative for TERT (b), whereas serous adenocarcinoma (c) displays cytoplasmic staining with TERT (d). (a–d—20 × ).

Full size image
Table 3 Immunoexpression of TERT in atypical tubal metaplasia vs uterine serous carcinoma
Full size table

Discussion

When endometrial samplings display tubal metaplasia exhibiting cytologic atypia, including nuclear pleomorphism and hyperchromasia, prominent nucleoli, and increased nuclear to cytoplasmic ratio, the possibility of an incompletely sampled uterine malignancy is raised. The features of atypical tubal metaplasia may histologically resemble some cytologic features seen in uterine serous carcinoma and serous endometrial intraepithelial carcinoma. This study evaluated the immunoprofile of atypical tubal metaplasia and the long-term outcomes in these patients.

p53

The p53 tumor suppressor gene is one of the most commonly altered genes in human malignancies, and is mutated in approximately half of most cancer types arising in a wide variety of tissues.9 When cells lack normal functional p53 protein, inactive mutant p53 protein can accumulate in the nucleus, leading to a relative overexpression, which is detectable by immunohistochemistry.10 Uterine serous carcinoma characteristically exhibits diffuse strong nuclear positivity with p53,11 a finding also seen in serous endometrial intraepithelial carcinoma.12 In contrast, the current study shows that atypical tubal metaplasia displays, at most, only focal, weak immunoreactivity with p53, and is frequently p53 negative. This pattern of expression was also seen in typical tubal metaplasia, which has been reported in a previous study.13

Ki-67

Comprehensive, well-known analysis has shown that Ki-67 is expressed exclusively in proliferating cells, and that immunostaining with a monoclonal antibody for Ki-67 is a reliable method of evaluating the growth fraction of normal and neoplastic human cell populations.14 Studies have also shown that Ki-67 proliferation indices are increased in numerous human malignancies, including high-grade endometrial cancer, as compared with post-menopausal (atrophic) endometrium.15 High proliferation indices with Ki-67 are also seen in invasive 7, 16 and intraepithelial serous carcinoma.12 In contrast, our findings show that the majority of atypical tubal metaplasia is either negative for Ki-67 or shows a low proliferation index, with Ki-67 immunopositivity limited to less than 5% of atypical cells. This pattern of expression is similar to that seen in typical tubal metaplasia.

Human telomerase reverse transcriptase (TERT)

Telomeres are specialized nucleoproteins in repetitive sequences present at the ends of chromosomes that provide protection from chromosome degradation and end-to-end fusion. With each cell division, the length of telomeres is progressively shortened until the chromosome becomes unstable, at which point cell death occurs; a process known as replicative senescence.17 The telomerase enzyme is an RNA-dependent DNA polymerase that catalyzes the addition and maintenance of telomeric repeat sequences to chromosome ends. Activation of this enzyme enables cells to overcome replicative senescence and divide indefinitely.18, 19 Telomerase activity is considered a marker of cell proliferation,20 as it is present in normal tissues that are highly proliferative, such as oral mucosa,21 the basal layer of epidermis,22 and proliferative-phase endometrium.23 Telomerase activity is also present in a high percentage of human tumor types.24 In malignancies, telomerase activity may simply mirror the fraction of proliferating cells;20 however, it is also hypothesized that tumor cells may independently upregulate telomerase levels.25 Studies have revealed that there are two major subunits contributing to in vitro activity of the telomerase enzyme complex: an intrinsic RNA component (TERC) containing a template region that binds to telomeric repeats; and a catalytic subunit with reverse transcriptase activity (TERT).26, 27 Although TERC is constitutively present in normal and cancer cells, TERT expression is limited almost exclusively to cancer cells.28 Brustmann8, 29 has reported the immunohistochemical reactivity of TERT in numerous human cancers, including uterine endometrioid and serous carcinoma, but not in benign lesions such as serous cystadenoma. Our study corroborates literature reports of TERT immunopositivity in uterine serous carcinoma and also shows that, in contrast, atypical tubal metaplasia is completely negative for TERT.

Long-term outcome

Atypical tubal metaplasia is worrisome in endometrial biopsy and curettage specimens, as it displays a superficial resemblance to invasive or intraepithelial serous carcinoma. Uterine serous carcinoma is well known for exhibiting aggressive behavior. After its initial biopsy diagnosis, early metastasis is often discovered at the time of surgery, resulting in an upstaging of the cancer.3 Serous endometrial intraepithelial carcinoma is commonly associated with extrauterine serous carcinoma.4, 12, 30, 31 Because of the clinically aggressive nature of these lesions, surgical management and staging is recommended.32 Our study shows that the clinical behavior of atypical tubal metaplasia differs from that of invasive or intraepithelial serous carcinoma, as after an average follow-up period of 5 years, the presence of atypical tubal metaplasia in endometrial samplings did not increase the risk of developing hyperplasia or malignancy.

In conclusion, the findings of this study are reassuring as they show that atypical tubal metaplasia displays immunostaining patterns with p53, Ki-67, and TERT that are similar to typical tubal metaplasia and distinct from uterine serous carcinoma or serous endometrial intraepithelial carcinoma. Therefore, although atypical tubal metaplasia in endometrial samplings may seem histologically worrisome for malignancy, its immunoprofile is not consistent with that of invasive or intraepithelial serous carcinoma. In addition our long-term follow-up study shows that atypical tubal metaplasia displays no increased incidence of atypical endometrial hyperplasia or carcinoma compared with typical tubal metaplasia. These findings suggest that atypical tubal metaplasia is most likely degenerative or reactive in etiology; the entity per se is not a direct precursor to atypical endometrial hyperplasia or endometrial carcinoma; and its presence in endometrial specimens does not portend a greater risk for patients to develop these lesions, as compared with the control population. Recognition of this morphologic entity in endometrial biopsies is important as it may prevent unwarranted aggressive clinical management.