The trusted source for
healthcare information and
By David M. Gershenson, MD
In a prospective study of the use of endometrial biopsy, Barakat and associates conducted a prospective study to determine the frequency of developing abnormal pathologic changes in the endometria of tamoxifen-treated women and to characterize the type of pathologic changes involved.1 Between October 1991 and September 1998, 159 patients initiating tamoxifen therapy for breast cancer confined to the breast and axillary lymph nodes were enrolled. Office endometrial biopsies (EMBs) were obtained during the initiation of tamoxifen and at six-month intervals for a two-year period. Three subsequent annual EMBs were recorded for each patient, amounting to a five-year surveillance. The median age of the patients was 50 years. Nine patients (5.7%) were considered protocol violations. The remaining 111 assessable patients underwent a total of 635 EMBs (mean, 5.8 EMBs), with a median surveillance time of 36 months. Five hundred forty-four patients (86%) had benign endometria. Eighty-two (12.9%) of the 635 biopsies revealed tissue insufficient for diagnosis. Fourteen patients (12.6%) underwent dilatation and curettage (D&C) for an abnormal EMB, persistent bleeding, or for evaluation of adnexal masses at the time of laparoscopy. Findings at D&C included complex hyperplasia (n = 1), abnormal histiocytes (n = 1), simple hyperplasia (n = 2), polyps (n = 4), endocervical polyp (n = 1), and decidualization (n = 2). Three D&Cs were negative. Three patients underwent hysterectomy. Barakat et al concluded that EMB was used to monitor the endometrium in the majority (95%) of breast cancer patients on tamoxifen in this trial, but the use of routine EMB for screening in tamoxifen-treated women seems limited.
In a German study, Gerber and colleagues prospectively studied the value of transvaginal sonography (TVS) in endometrial screening of postmenopausal breast cancer patients treated with tamoxifen.2 In 247 tamoxifen-treated (20-30 mg/d for ³ 2 years) women and 98 controls, the endometrium was followed-up by means of TVS every six months for up to five years. Patients with homogeneous endometrium of more than 10-mm thickness were then scanned repeatedly every three months. The mean endometrial thickness was 3.5 ± 1.1 mm before treatment and increased to a maximum of 9.2 ± 5.1 mm after three years of tamoxifen application (P < 0.0001), which was significantly thicker compared with controls (P < 0.0001). Fifty-two asymptomatic patients with thickened or morphologically suspect endometrium underwent hysteroscopy and D&C, resulting in four uterine perforations. Histopathologically, atrophy was found in 38 patients (73.1%), polyps in nine, hyperplasia in four, and endometrial cancer in one case. In 20 screened patients who reported vaginal bleeding, five atrophies (25%), five polyps, four hyperplasias, and two endometrial cancers were found. Before hysteroscopy and D&C were performed, 36 (69.2%) of 52 asymptomatic and four (20%) of 20 symptomatic patients were scanned by repeated TVS over 2-30 months. Invasive diagnostic procedures were significantly (P < 0.05) more frequent in younger and obese patients. In the controls, one asymptomatic polyp and one symptomatic hyperplasia were found. Gerber et al concluded that, in tamoxifen-treated patients, TVS offered a high false-positive rate, even with a cutoff value of 10 mm for endometrial thickness and repeated TVS scans. They further concluded that increased iatrogenic morbidity and only one asymptomatic endometrial carcinoma do not warrant endometrial screening by TVS in tamoxifen-treated patients.
Tamoxifen has become a popular drug for both the treatment and prevention of breast cancer. Several retrospective studies have shown a relationship between tamoxifen use and the development of uterine cancers. In a trial of the National Surgical Adjuvant Breast and Bowel Project (NSABP), women with ER-positive breast cancer confined to the breast with negative axillary nodes were randomized to tamoxifen or placebo.3 Among 4063 patients randomized, there was a 7.5-fold increase in the risk of developing endometrial cancer in the tamoxifen-treated group. In addition, a wide range of benign abnormalities of the endometrium may occur, from atrophy to polyps to hyperplasias. With the high incidence of breast cancer and the widespread use of tamoxifen, researchers have been focusing on the issue of endometrial screening for tamoxifen-treated patients. To date, there has been no evidence to support the role of either EMB or TVS in asymptomatic women receiving tamoxifen. The two reports summarized above, which were both published in the October 2000 issue of the Journal of Clinical Oncology, significantly enhance our database in the same direction. Based on their findings, Barakat et al do not recommend the routine use of EMB for screening tamoxifen-treated women. No endometrial cancers were detected in their series, and only one patient had complex hyperplasia detected by D&C. Likewise, in the report of Gerber et al, endometrial screening by TVS is not recommended. Among 52 asymptomatic patients with abnormal TVS, hyperplasia was found in four patients and endometrial cancer in one patient. A total of 1265 TVS were performed to detect that sole endometrial cancer. Four patients experienced uterine perforations. Despite the relatively high cutoff of 10 mm for definition of abnormally thickened endometrium, the false-positive rate was high. The optimal cutoff for endometrial thickness in tamoxifen-treated women remains unknown. Mounting evidence suggests that the high false-positive rate is attributable to tamoxifen-induced edema of cystically dilated endometrial glands with periglandular stromal and myometrial condensation. As emphasized by both the authors of these articles and in an editorial by Runowicz,4 the tendency to order a battery of screening tests may simply enhance rather than alleviate a patient’s anxiety. Routine performance of EMB, and, particularly, TVS will lead to unnecessary invasive procedures with their associated cost and morbidity. These procedures should be reserved for the tamoxifen-treated woman with vaginal bleeding unless forthcoming evidence suggests otherwise.
1. Barakat RR, et al. J Clin Oncol 2000;18:3459-3463.
2. Gerber B, et al. J Clin Oncol 2000;18:3464-3470.
3. Fisher B, et al. J Natl Cancer Inst 1994;86:527-537.
4. Runowicz CD. J Clin Oncol 2000;18:3457-3458.