Objective: A prospective evaluation of the effects on endometrium of third generation aromatase inhibitors (AIs), administered as adjuvant up-front therapy or switched therapy in menopausal patients suffering from breast cancer.
Methods: Forty-five patients suffering from estrogen-receptor positive breast cancer were treated with AIs as adjuvant endocrine therapy; 27 patients switched from tamoxifen to AIs (group 1) due to adverse medical events related to tamoxifen intake (22 patients) or to an extended endocrine treatment after 60 months of tamoxifen therapy (5 patients); whereas 18 patients received AIs as up-front adjuvant therapy (group 2). All patients underwent endometrial investigation before the start of AIs therapy and, thereafter, at 12 month intervals. Endometrial assessment was based on Transvaginal Ultrasonography (TU), followed by hysteroscopy and endometrial biopsy when a double layered endometrial stripe above 4 mm was measured on the longitudinal plane of uterine scanning. Six patients, showing endometrial hyperplasia before the start of AIs therapy, underwent hysteroscopy on a yearly basis, disregarding the endometrial thickness measured by TU. Histopathologic results on endometrial biopsies represented the reference test in order to estimate the prevalence of endometrial morbidity.
Results: Demographic and clinical variables evaluated (age, parity, age at menarche and menopause, Body Mass Index, previous chemotherapy and radiotherapy) did not differ in groups 1 and 2. The average period of endometrial surveillance after the start of AIs therapy was 24.8 +/-10.8 months for group 1 and 21.4 +/- 11.5 months for group 2. A progressive decrease of endometrial thickness, from 8.2 +/- 5.0 to 3.0 +/- 1.2 in group 1 and from 4.7 +/- 4.3 to 1.9 +/- 0.3 in group 2, was found before the start and after 36-48 months of AIs therapy. The second line endometrial investigations' rate dropped from 70.3% to 12.5% in group 1 and from 27.7% to 0.0% in group 2, at baseline and after 36-48 months of AIs therapy, respectively. We found baseline endometrial abnormalities in 25.9% and in 22.2% of patients in groups 1 and 2 (P = 0.4), respectively. During AIs administration, an endometrial pathology was found in 1 patient of group 1 and in 3 patients of group 2. In 3 patients, the abnormality consisted of simple hyperplasias and in all these patients an abnormal endometrium (1 complex atypical hyperplasia and 2 simple hyperplasias) was already detected at baseline assessment. Only in 1 patient (2.2%) of group 2 did we find an emerging pathology, consisting of adenosarcoma harbored within an endometrial polyp, detected after 12 months of therapy with letrozole. In 3 out of 5 patients showing simple hyperplasia and in 1 patient showing atypical hyperplasia before the start of AIs therapy, we observed a reversal to normal endometrium and to simple hyperplasia, respectively, after 12 months of therapy with anastrozole.
Conclusions: AIs delivered as up-front therapy for breast cancer have no effects on unspecific endometrial thickening. When administered as switched therapy after tamoxifen withdrawal, AIs may reverse tamoxifen-associated endometrial thickening. As a consequence, we reduced unnecessary second-line endometrial investigations. A low rate of emerging endometrial pathology was found during AIs therapy.