I. Introduction

An estimated 211,300 new cases of invasive breast cancer were diagnosed among women in the United States during 2003. During the same year, approximately 40,200 women died of this disease (Ries et al, 1975-2000). Most patients with breast cancer are offered mastectomy or breast conservation therapy at the time of diagnosis.

For over 20 years, tamoxifen has been widely used in the chemoprevention of breast cancer. It is administered as a potential breast cancer chemopreventive agent because studies have shown that tamoxifen given for 5 years reduces the incidence of recurrent breast cancer by 42% and reduces the incidence of contralateral breast cancer by 47% (Fisher et al, 1998).

There is poor correlation between sonographic, hysteroscopic, and histologic findings in most patients using tamoxifen (Hulka and Hall, 1993; Goldstein, 1994; Mourits et al, 1999). In 45-90% of postmenopausal long-term tamoxifen users, an increased endometrial thickness by ultrasonography is not confirmed by hysteroscopic or histologic pathology review. Endometrial sampling by microcurettage in tamoxifen-treated patients has not proved to be as sensitive and accurate as hysteroscopy and curettage (Powles et al, 1998; Neven and Vernaeve, 2000).

It has been our observation that in the Gynecologic Oncology Center in our institution, many patients with a history of breast cancer who seek the advice of a gynecologist or gynecologic oncologist do so for symptoms of vaginal bleeding either secondary to anovulatory cycles caused by the chemotherapeutic agents used to treat their breast cancer or secondary to tamoxifen use. There is also significant anxiety in asymptomatic patients undergoing treatment with tamoxifen who are found to have abnormalities on routine transvaginal ultrasonography.

The purpose of this study was to identify the symptoms in breast cancer patients who previously underwent mastectomy and subsequently presented to the outpatient Gynecologic Oncology Center in our institution. We also sought to identify the diagnostic studies used in the evaluation of these patients prior to clinical disposition. Finally, we analyzed the histopathologic findings of hysteroscopy and hysterectomy.

II. Materials and Methods

We conducted a retrospective analysis of all patients with breast carcinoma who underwent mastectomy at The University of Texas M. D. Anderson Cancer Center between January 1990 and December 2001. We then identified patients from this group who had been referred to and evaluated in our outpatient Gynecologic Oncology Center and ultimately underwent either dilatation and curettage (D and C) with hysteroscopy or total hysterectomy. All cases were identified and retrieved by the Department of Medical Informatics at our institution. The investigation was conducted after Institutional Review Board approval was obtained.

The medical records of all identified patients were reviewed, and the following information was gathered: patient age, race, family history of breast, ovarian or endometrial cancer, time of diagnosis of breast cancer, history of chemotherapy administration, radiation therapy, and hormonal therapy, reason for gynecologic oncology consultation, imaging studies performed at the time of consultation, results of these studies, type of definitive treatment, and histopathologic findings from D and C and hysteroscopy or hysterectomy.

We excluded all patients who presented to the Gynecologic Oncology Center for any of the following reasons: a routine annual examination in an asymptomatic patient, vaginal dryness, infectious vaginal discharge, request for tubal ligation, treatment of endomyometritis, or evaluation of urinary incontinence. Additional exclusion criteria included hysterectomy prior to breast cancer diagnosis, gynecologic consultation or diagnosis of gynecologic malignancy prior to breast cancer diagnosis, hysterectomy and breast cancer surgery performed simultaneously, and missing medical record or no patient information available. We also excluded patients who were advised to undergo a D and C and hysteroscopy or hysterectomy and subsequently were lost to follow-up.

Statistical analysis was performed using SPSS version 11.5.1. Descriptive statistics were used to evaluate demographic and clinical characteristic of the patients.

III. Results

A total of 262 patients were identified during the initial search. Forty-five patients were excluded according to the criteria outlined above. The remaining 217 patients were included in the final analysis. The median age for this group at presentation to the Gynecologic Oncology Center was 52 years (range, 28 to 83 years). The median follow-up time from the date of the first visit to the Gynecologic Oncology Center to the date of last contact was 25 months (range, 1 to 323 months). The ethnicity of the patients in our study was White, 77%; African American, 11%; Hispanic, 10%; and Asian, 2%. Information regarding family history was available for 199 patients. Of these patients, 69 (35%) had a history of maternal breast cancer, and 17 (8%) had a history of paternal breast cancer. Information regarding family history of ovarian cancer was available for 204 patients. Of these, 14 (7%) had a family history of ovarian cancer.

A. Breast cancer diagnosis and treatment

The median age at the time of breast cancer diagnosis was 46.5 years (range, 22 to 82 years). The majority of the patients (146, 67%) had a diagnosis of invasive ductal carcinoma. Location of breast cancer was as follows: left breast, 92, (42%); right breast, 102, (47%); bilateral, 23 (11%). Estrogen receptor status was documented for 140 patients. Of these, 98 (70%) were estrogen receptor positive and 42 (30%) were estrogen receptor negative. Similarly, progesterone receptor status was documented for 128 patients, and 80 (62%) were progesterone receptor positive while 48 (38%) were progesterone receptor negative. Information for HER2-neu receptor was available for 49 patients. Of these, 18 (37%) patients tested positive for the HER2-neu receptor.

A total of 122 (56%) of the 217 patients received some form of chemotherapy as treatment for their breast cancer. The most common regimen was 5-fluorouracil, doxorubicin, and cyclophosphamide. In addition, 107 (49%) of 215 patients were treated with radiotherapy and 99 (46%) of 212 patients were treated with hormonal therapy. Ninety-six of the 99 patients (97%) treated with hormones received tamoxifen. The median time elapsed between the breast cancer diagnosis and the gynecologic consultation was 34 months (range, 1 to 338 months).

B. Gynecologic consultations and treatment

The most common reasons for gynecologic consultations were vaginal bleeding, 92 (43%); abnormal finding on ultrasonography, 45 (21%); recommended by breast oncologist, 16 (7%); evaluation of abnormal Pap smear, 14 (6%); concerns due to positive family history of breast or ovarian cancer, 9 (4%); and self-reported anxiety, 3 (1%). After initial evaluation, a total of 139 (64%) of the 217 patients had sonography performed. Other studies ordered included chest radiography, 164 (76%) and computed tomography scan of the abdomen and pelvis, 53 (24%).

Prior to the D and C and hysteroscopy or hysterectomy, 120 (55%) of 217 patients underwent endometrial evaluation by endometrial biopsy. Findings on biopsy were considered abnormal in 35 cases (29%). A total of 64 patients (29%) underwent a D and C and hysteroscopy. Forty-four (69%) of the 64 patients who underwent a D and C also had an endometrial biopsy performed in the office prior to the D and C. The majority of patients underwent at least two diagnostic studies prior to the D and C and hysteroscopy. The most frequently ordered imaging study prior to D and C and hysteroscopy was transvaginal ultrasonography, which was ordered in 84% of patients.

A total of 176 patients (81%) underwent a hysterectomy after their consultation in the Gynecologic Oncology Center. The median time from breast cancer diagnosis to hysterectomy was 42 months (range, 1 to 338 months). Findings on pathology review of the hysterectomy specimen were available for 163 patients. The pathology results were malignant in 54 cases (33%) (Table 1). Of the 176 patients who ultimately underwent a hysterectomy, 24 (14%) also underwent a D and C and hysteroscopy prior to the hysterectomy. A median of three (range, 0 to 13) diagnostic studies was performed in the patients who underwent a hysterectomy.

Of the 96 patients who had a history of tamoxifen use or were current users of tamoxifen, 62 (64%) underwent an endometrial biopsy. A D and C and hysteroscopy was performed in 47 (49%) of the 96 patients who received tamoxifen. Ultimately, 69 (72%) of the 96 patients who received tamoxifen underwent a hysterectomy. The most common malignancy was endometrial cancer, which was found in 10 patients. All of these tumors were categorized as stage I disease. However, two patients were found to have uterine papillary serous carcinoma, and three patients were diagnosed with uterine sarcomas (carcinosarcoma in 2 patients and leiomyosarcoma in 1).

IV. Discussion

Earlier detection and treatment of breast cancer will likely translate into better survival. In addition, continued research in the treatment and surveillance of breast cancer will also likely lead to better survival. Gynecologic oncologists are consultants for many women who routinely inquire about their risks of gynecologic malignancies or who need to be treated for conditions that develop as a result of breast cancer treatment.

A significant percentage of the patients in our study were treated with hormones, and the majority of these were treated with tamoxifen. The estrogenic effects of tamoxifen on the vaginal epithelium of postmenopausal women with breast cancer have previously been documented (Ferrazzi et al, 1977). In addition, tamoxifen has a stimulatory effect on the endometrium (Boccardo et al, 1984). In 1985, Killackey et al, were the first to report an association between tamoxifen use and the development of endometrial cancer. Tamoxifen has been shown to increase the risk of developing endometrial cancer regardless of the dose recommended (van Leeuwen et al, 1994; Fisher et al, 1998).

Table 1. Malignant surgical pathology findings on hysterectomy (n=54)

Finding	No. of patients	(%)
Ovarian tumor
Primary invasive carcinoma	8	15
Metastatic breast carcinoma	5	9
Uterine tumor
Primary: Endometrioid	23	42
Sarcoma	5	9
Papillary serous carcinoma	5	9
Metastatic breast carcinoma	3	6
Primary invasive fallopian tube carcinoma	1	2
Primary peritoneal carcinoma	3	6
Primary appendiceal carcinoma	1	2

There seems to be significant debate as to whether patients receiving tamoxifen should undergo routine evaluation of the endometrium. The current recommendations of the American College of Obstetrics and Gynecology are outlined in Table 2. However, it is not uncommon for an asymptomatic patient to consult a gynecologic oncologist following the recommendation of her medical oncologist or on her own initiative to discuss the risk of endometrial cancer in the setting of tamoxifen exposure.

The evaluation of patients receiving tamoxifen is often difficult. Findings on ultrasonography alone may be misleading because tamoxifen produces an echogenic, thick, irregular, cystic appearance of the endometrial stroma and the myometrium. Fong et al, (2001) evaluated the performance characteristics of transvaginal ultrasonography and hysterosonography for the diagnosis of endometrial abnormalities in 138 asymptomatic postmenopausal patients with breast cancer who were receiving tamoxifen. The authors concluded that an endometrial thickness of 6 mm should be considered the upper limit of normal in this patient population. In addition, they suggested that hysterosonography improves specificity by reducing the false-positive rate of transvaginal ultrasonography. Others have suggested that office hysteroscopy can also serve as a conclusive diagnostic tool for the evaluation of the endometrium in this group of patients (Timmerman et al, 1998; Garuti et

Table 2. American college of obstetrics and gynecology recommendations for evaluation of patients receiving tamoxifen

· Women taking tamoxifen should be monitored closely for symptoms of endometrial hyperplasia or cancer and should have a gynecologic examination at least once a year

· Women taking tamoxifen should be educated about the risks of endometrial proliferation, endometrial hyperplasia, and endometrial cancer. Women should be encouraged to promptly report any abnormal vaginal symptoms, including bloody discharge, spotting, staining, or leukorrhea

· Any abnormal vaginal bleeding, bloody vaginal discharge, staining, or spotting should be investigated

· Because screening tests have not been effective in increasing the early detection of endometrial cancer in women using tamoxifen and may lead to more invasive and costly diagnostic procedures, they are not recommended

· Tamoxifen use should be limited to 5 years’ duration because a benefit beyond this time has not been documented

· If atypical endometrial hyperplasia develops, appropriate gynecologic management should be instituted, and the use of tamoxifen should be reassessed. If tamoxifen therapy must be continued, hysterectomy should be considered in women with atypical endometrial hyperplasia. Tamoxifen use may be reinstituted following hysterectomy for endometrial carcinoma in consultation with the physician responsible for the woman’s breast care

al, 2002). It is important to note, however, that the same parameters used for endometrial evaluation in the general population should not be applied to postmenopausal patients who are undergoing treatment with tamoxifen (Achiron et al, 1995).

In our study, we found that the majority of patients who developed endometrial carcinoma developed the more common endometrioid type. However, another group of patients developed higher risk histologic subtypes, such as papillary serous carcinoma and sarcoma. This has previously been reported in other studies. From our own institution, Silva et al, (1995) demonstrated that breast cancer patients treated with tamoxifen were at increased risk not only for endometrioid adenocarcinoma but also for other histological subtypes, such as papillary serous and clear cell carcinoma. Not only have there been cases of tamoxifen-associated endometrial cancers that are deeply invasive and of higher grade (Malfetano, 1990), but there have also been a number of reports of tamoxifen-related uterine carcinosarcomas (Clarke, 1993; Kloos et al, 2002), as well as rare cases of stromal sarcomas, adenosarcomas, and leiomyosarcomas (Clement et al, 1996; Sabatini et al, 1999).

We also found that a number of patients in our study were diagnosed with ovarian masses benign tumors, malignant primary tumors, and metastatic tumors from other primary sites. Previous reports (Cohen et al, 1994, 1996) have shown that in a select group of postmenopausal breast cancer patients treated with tamoxifen, the rate of ovarian tumors was 5.7%. This was four to five times higher than the rate of similar pathologic conditions in nonselected, asymptomatic and untreated postmenopausal women. A partial explanation of this finding may be the fact that women with breast malignancies, regardless of tamoxifen use, are more likely to develop benign or malignant ovarian tumors because of genetic factors. A small number of patients in our study also had metastatic breast cancer to the endometrium. This finding has previously been reported by Horn et al, (2000).

In summary, we found that women who had undergone a mastectomy and subsequently underwent a hysteroscopy or hysterectomy often had several diagnostic studies before gynecologic surgery. In addition, these patients frequently underwent several invasive procedures before definitive treatment was performed. In our study, a significant number of patients ultimately had a malignancy diagnosed. Although most of these malignancies were diagnosed in the uterus and were the more common endometrioid type, some were of a more aggressive histologic subtype. In addition, a number of invasive ovarian carcinomas were also detected. These findings highlight the importance of prudent and expeditious evaluation in this patient population. Given these data the focus of future studies should address whether patients undergoing mastectomy should be offered a prophylactic hysterectomy and bilateral salpingo-oophorectomy at the same time.

Another very important question that remains unanswered is how patients undergoing treatment with tamoxifen should be followed after diagnosis and treatment of their breast cancer. Until prospective studies show the efficacy of screening tests or diagnostic modalities, we must continue to follow the recommendations of the American College of Obstetrics and Gynecology.

References

Achiron R, Lipitz S, Sivan E, Goldenberg M, Horovitz A, Frenkel Y, Mashiach S (1995) Changes mimicking endometrial neoplasia in postmenopausal, tamoxifen-treated women with breast cancer, a transvaginal Doppler study. Ultras Obstet Gynecol 6, 116-20.

Boccardo F, Guarneri D, Rubagotti A, Casertelli GL, Bentivoglio G, Conte N, Campanella G, Gaggero G, Comalli G, Zanardi S, Nicolo G (1984) Endocrine effects of tamoxifen in postmenopausal breast cancer patients. Tumori 70, 61-8.

Clarke MR (1993) Uterine malignant mixed mullerian tumor in a patient on long-term tamoxifen therapy for breast cancer. Gynecol Oncol 51, 411-5.

Clement PB, Oliva E, Young RH (1996) Mullerian adenosarcoma of the uterine corpus associated with tamoxifen therapy, a report of six cases and a review of tamoxifen-associated endometrial lesions. Int J Gynecol Pathol 15, 222-9.

Cohen I, Altaras MM, Lew S, Tepper R, Beyth Y, Ben-Baruch G (1994) Ovarian endometrioid carcinoma and endometriosis developing in a postmenopausal breast cancer patient during tamoxifen therapy, a case report and review of the literature. Gynecol Oncol 55, 443-7.

Cohen I, Beyth Y, Tepper R, Shapira J, Zalel Y, Figer A, Cordoba M, Yigael D, Altaras MM (1996) Ovarian tumors in postmenopausal breast cancer patients treated with tamoxifen. Gynecol Oncol 60, 54-8.

Ferrazzi E, Cartei G, Mattarazzo R, Fiorentino M (1977) Oestrogen-like effect of tamoxifen on vaginal epithelium. BMJ 1, 1351-2.

Fisher B, Costantino JP, Wickerham DL, Redmond CK, Kavanah M, Cronin WM, Vogel V, Robidoux A, Dimitrov N, Atkins J, Daly M, Wieand S, Tan-Chiu E, Ford L, Wolmark N (1998) Tamoxifen for prevention of breast cancer, report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer Inst 90, 1371-88.

Fong K, Kung R, Lytwyn A, Trudeau M, Chapman W, Nugent P, Glanc P, Manchul L, Szabunio D, Myhr T (2001) Endometrial evaluation with transvaginal US and hysterosonography in asymptomatic postmenopausal women with breast cancer receiving tamoxifen. Radiology 220, 765-73.

Garuti G, Cellani F, Grossi F, Colonnelli M, Centinaio G, Luerti M (2002) Saline infusion sonography and office hysteroscopy to assess endometrial morbidity associated with tamoxifen intake. Gynecol Oncol 86, 323-9.

Goldstein SR (1994) Unusual ultrasonographic appearance of the uterus in patients receiving tamoxifen. Am J Obstet Gynecol 170, 447-51.

Horn LC, Einenkel J, Baier D (2000) Endometrial metastasis from breast cancer in a patient receiving tamoxifen therapy. Gynecol Obstet Investig 50, 136-8.

Hulka CA, Hall DA (1993) Endometrial abnormalities associated with tamoxifen therapy for breast cancer, sonographic and pathologic correlation. AJR Am J Roentgenol 160, 809-12.

Killackey MA, Hakes TB, Pierce VK (1985) Endometrial adenocarcinoma in breast cancer patients receiving antiestrogens. Cancer Treat Rep 69, 237-8.

Kloos I, Delaloge S, Pautier P, Di Palma M, Goupil A, Duvillard P, Cailleux PE, Lhomme C (2002) Tamoxifen-related uterine carcinosarcomas occur under/after prolonged treatment, report of five cases and review of the literature. Int J Gynecol Cancer 12, 496-500.

Malfetano JH (1990) Tamoxifen-associated endometrial carcinoma in postmenopausal breast cancer patients. Gynecol Oncol 39, 82-4.

Mourits MJ, Van der Zee AG, Willemse PH, Ten Hoor KA, Hollema H, De Vries EG (1999) Discrepancy between ultrasonography and hysteroscopy and histology of endometrium in postmenopausal breast cancer patients using tamoxifen. Gynecol Oncol 73, 21-6.

Neven P, Vernaeve H (2000) Guidelines for monitoring patients taking tamoxifen treatment. Drug Safety,22, 1-11.

Powles TJ, Bourne T, Athanasiou S, Chang J, Grubock K, Ashley S, Oakes L, Tidy A, Davey J, Viggers J, Humphries S, Collins W (1998) The effects of norethisterone on endometrial abnormalities identified by transvaginal ultrasound screening of healthy post-menopausal women on tamoxifen or placebo. Br J Cancer 78, 272-5.

Ries LAG, Eisner MP, Kosary CL, Hankey BF, Miller BA, Clegg L, Mariotto A, Fay MP, Feuer EJ, Edwards BK editors. SEER Cancer Statistics Review, 1975-2000. National Cancer Institute. Bethesda, MD, 2003.http, //seer.cancer.gov/csr/1975 2000)

Sabatini R, Di Fazio F, Loizzi P (1999) Uterine leiomyosarcoma in a postmenopausal woman treated with tamoxifen, case report. Eur J Gynaecol Oncol 20, 327-8.

Silva EG, Tornos C, Malpica A, Mitchell MF (1995) Uterine neoplasms in patients treated with tamoxifen. J Cell Biochem 3, 179-83.

Timmerman D, Deprest J, Bourne T, Van den Berghe I, Collins WP, Vergote I (1998) A randomized trial on the use of ultrasonography or office hysteroscopy for endometrial assessment in postmenopausal patients with breast cancer who were treated with tamoxifen. Am J Obstet Gynecol 179, 62-70.

van Leeuwen FE, Benraadt J, Coebergh JW, Kiemeney LA, Gimbrere CH, Otter R, Schouten LJ, Damhuis RA, Bontenbal M, Diepenhorst FW, et al (1994) Risk of endometrial cancer after tamoxifen treatment of breast cancer. Lancet 343, 448-52.

Research Article

Received: 23 December 2004; Accepted: 24 February 2005; electronically published: March 2005