Routine Office Endometrial Biopsies and Ultrasound Screening Are Not Recommended for Breast Cancer Patients Receiving Tamoxifen

An expert panel of 10 international cancer researchers andpracticing oncologists met in Boston to discuss the past, present,and future uses of antiestrogens in the treatment of breast cancer.The first articles in this series, based on the symposium presentations,focused on the optimal duration of tamoxifen use (Oct, 1996, page17) and on the noncancer benefits of tamoxifen weighed againstthe potential risk of endometrial cancer (Nov, 1996, page 55).This month, the benign endometrial changes associated with tamoxifenuse are reviewed. The symposium was sponsored by Zeneca Pharmaceuticals.

BOSTON--Although evaluating patients on tamoxifen (Nolvadex) withendometrial biopsies is appropriate in clinical trials, "wecertainly do not recommend it to all gynecologists in the community,"said Richard R. Barakat, MD, a gynecologic oncologist at MemorialSloan-Kettering Cancer Center.

Tamoxifen is known to have proliferative effects on the endometrium,Dr. Barakat said. However, hyperplasia in the endometrium representsa broad spectrum of changes. Endometrial hyperplasia is classifiedas simple or complex, with or without nuclear atypia.

"The only real endometrial hyperplas-tic lesion that hasa significant rate of progression to cancer is complex atypicalhyperplasia, which has about a 27% progression rate," Dr.Barakat said. Only about 1% of untreated simple hyper-plasiaswill progress, he added.

Panel members V. Craig Jordan, PhD, DSc, and Vasilios Assikis,MD, of Northwestern University Medical School, reviewed a seriesof reports that examined endometrial histology in women on tamoxifen.Their analysis showed about a twofold increase in the incidenceof proliferative endometrium, about a threefold increase in endometrialpolyps, and about a 10-fold increase in endometrial hyperplasia,with tamoxifen use.

Dr. Barakat pointed out that most of these studies involved onlya small number of patients and did not always specify the exacttype of hyperplasia. "In fact, the incidence of complex atypicalhyperplasia appears to be quite low," he said.

No Link With Duration of Use

As with endometrial cancer, there is no clear relationship betweenthe duration of tamoxifen use and endometrial histologic findings,Dr. Barakat said. In a study from Memorial Sloan-Kettering ofbreast cancer patients who had undergone a dilatation and curettage(D&C), the duration of tamoxifen use was 22 months for thosewith benign endometrial changes; 18 months for those with polyps;and 19 months for the one patient with hyperplasia.

Endometrial cancer seemed to be associated with a longer durationof use, but the study involved only a small number of patients,and "there didn't seem to be any clear relationship betweenthe duration of use and the various histologic findings,"he said.

Effects on Preexisting Pathology

In a small study from Israel published in Gynecologic Oncology,an initial biopsy was done on a group of breast cancer patientstreated with tamoxifen for 1 to 5 years. Upon resampling 1½year later, the researchers found that the two patients who hadproliferative endometrium initially now had normal endometrium;the one patient who had a polyp initially now had a normal biopsy;and the one patient with hyperplasia initially still had hyperplasiawith no progression.

"They concluded from this small study that tamoxifen wasnot causing a progression of preexisting endometrial pathology,"Dr. Barakat said.

In an ongoing study at Memorial Sloan-Kettering, Dr. Barakat andhis colleagues are prospectively determining the frequency ofthe development of patho-logic endometrial changes with tamoxifenuse. "This study, along with so many other studies of thisissue, suffers from the lack of a control group," he said."We're not looking at breast cancer patients not on tamoxifen."

Prior to starting tamoxifen, then at 6-month intervals for 2 years,and then annually for 3 years, the investigators perform Pap smearsand endometrial biopsies (via a Pipelle aspirator), and obtain17-beta-estradiol levels in premenopausal patients and repeatedbiopsies for those with abnormal bleeding. Surveillance increasesto every 4 months if evidence of hyperplasia is found.

"Using the thin Pipelle device, we are able to access theendometrial cavity in probably about 80% to 90% of patients,"Dr. Barakat said. "To date we've enrolled 126 of a planned140 patients, and in only 5% were we unable to obtain a biopsyspecimen."

A total of 296 biopsies have been done in 101 evaluable patientswith a median follow-up of 26 months. Four patients have developedabnormalities, including simple hyperplasia, complex hyperplasia,one case of complex atypical hyperplasia, and one case of an abnormalamount of histiocytes, which can be associated with an underlyingendometrial cancer.

In addition, six other patients with normal endometrium requiredD&C for abnormal bleeding, and three were found to have polyps(with a duration of tamox-ifen use of 4 months, 11 months, and27 months), two had normal findings, and one showed a progestationaleffect. Finally, three patients have undergone hysterectomy--forcomplex atypical hyper-plasia, a pelvic mass (sarcoma), and hyperplasiawith mucinous metasplasia.

"We concluded that all of our hyper-plastic lesions weredetected following at least 12 months of tamoxifen use. However,longer follow-up is required to determine the role of performingserial endometrial biopsies in this group of patients," Dr.Barakat said.

He also noted that there is no evidence to show that screeningfor endometrial cancer causes a decrease in mortality from endometrialcancer, "and the bottom line goal of any screening test isto decrease mortality," he said.

Nonetheless, Dr. Barakat speculates that the vast majority ofgynecologists are obtaining endometrial biopsies and performingsonograms every 6 months or so in their patients on tamoxifen.

"We have to educate the physicians and the patients,"he said. "Basically, we tell patients to be alert to thesigns of endometrial pathology. If they develop any abnormal bleeding,including spotting or brownish discharge, then we go ahead anddo a biopsy or D&C, but we certainly do not recommend routinescreening for all of these patients."

sidebar

Roundtable Participants

Chairperson: V. Craig Jordan, PhD, DSc, Robert H. LurieBreast Cancer Research Program, Northwestern University MedicalSchool, Chicago

Vasilios Assikis, MD, Northwestern University Medical School,Chicago

Richard R. Barakat, MD, Cornell University Medical Collegeand Memorial Sloan-Kettering Cancer Center, New York City

John H. Glick, MD, University of Pennsylvania Cancer Centerand Hospital of the University of Pennsylvania, Philadelphia

Anthony Howell, MD, Christie Hospital NHS Trust, Manchester,England

Robert W. Morgan, MD, Environmental Health Strategies,Inc., Redwood City, California

Joseph Ragaz, MD, University of British Columbia Schoolof Medicine, Vancouver, British Columbia

Bonnie S. Reichman, MD, Cornell University Medical Collegeand New York Hospital, New York City

Nicholas J. Robert, MD, Fairfax Hospital, Falls Church,Va

James A. Swenberg, DVM, PhD, University of North CarolinaSchool of Medicine and School of Public Health, Chapel Hill