In his excellent review, Dr. Barakat has made many useful observations about the effects of tamoxifen(Drug information on tamoxifen) (Nolvadex) on the endometrium. As the potential number of women on tamoxifen increases, several points merit emphasis and dissemination to all gynecologists who may be asked by medical oncologists to render opinions on such patients.
Women taking tamoxifen have thicker endometrial linings than women not taking the drug [1,2].This observation should come as no surprise, considering that tamoxifen is known to be a mixed agonist/antagonist of the effects of estrogen. In premenopausal women tamoxifen leads to an increase in serum estradiol(Drug information on estradiol) [3], which, if not countered with endogenous/exogenous progestins, would be expected to lead to endometrial proliferation.
As Dr. Barakat points out, some authors have found that tamoxifen is associated with an increase in endometrial hyperplasia and polyps [2], while others have not [1]. The relationship between tamoxifen and endometrial cancer has been postulated for nearly a decade [4], but not all series have been able to confirm such an association [1]. This may be because small series lack the power to observe such a rare phenomenon, even if the risk is doubled, as was suggested by the recently published B-14 trial of the National Surgical Adjuvant Breast and Bowel Project (NSABP) [5].
In the series of Gibson et a [l6], 11% of women with abnormal bleeding were found to have endometrial cancer, vs 0% of asymptomatic women. These percentages were the same whether or not the women were taking tamoxifen. Of note, however, all cases of endometrial cancer were symptomatic.
How should the gynecologist proceed when consulted by a medical oncology colleague or patient who has been taking tamoxifen? Certainly, patients with bleeding need endometrial sampling. Asymptomatic patients represent more of a dilemma.
Some authorities have suggested annual sampling [7], but many practicing medical oncologists are concerned that any discomfort associated with the procedure would limit the patient's willingness to continue taking tamoxifen, which benefits both node-negative and node-positive breast cancer patients [8]. The usual 5-mm endometrial stripe may be an inappropriate criterion in this patient population: As Lahti1 reported, 84% of tamoxifen users will develop this degree of endometrial thickening, and a 5-mm cutoff may lead to oversampling. In contrast, the 8-mm guideline proposed by Kedar et al [2], proved to be 100% specific--16 of 16 patients with such an endometrial thickness had polyps or hyperplasia. Therefore, this may be a more reasonable parameter on which to base sampling of the asymptomatic patient.
The paradoxical effects of tamoxifen on breast epithelium vs uterine endometrium offer ample opportunity for continued investigation. For example, researchers need to explore the possibility that tamoxifen produces some of the histologic changes seen in the endometrium through its effects at the molecular level on growth factors and receptors, oncogenes, and cell proliferation. Such work is ongoing at Indiana University and other institutions around the country. Agents that have an inhibitory effect on both the breast and uterine epithelium without negative impact on bone density or cardiovascular health are needed as well.
