Do Aromatase Inhibitors Increase Hemoglobin in Women with Breast Cancer?

Aromatase inhibitors (AI) block the conversion of testosterone and androstenedione to the estrogen estrone by inhibiting the aromatase enzyme complex. AI are used to treat estrogen receptor positive (ER+) breast cancer in postmenopausal women. With AI therapy, estrogen levels decrease to 85–95% of baseline. In women with metastatic disease, androstenedione levels do not increase. We have evaluated 2 women for polycythemia during AI therapy. Case 1 is 52 years old with stage II breast cancer treated with lumpectomy, TAC × 6 and radiation. Tamoxifen was started 4 months later. The mean hemoglobin during 6 months of tamoxifen was 14.0±0.1 gm/dl. When switched to exemestane, the mean hemoglobin over the next 24 months was 16.1±0.5 gm/dl (Mann-Whitney, p<0.003). Case 2 is an 80 year old with stage I breast cancer treated with lumpectomy followed by radiation. Her baseline hemoglobin was 13.8 gm/dl. 26 months after starting exemestane, her hemoglobin reached 18.0 gm/dl. After extensive evaluation, neither patient met the criteria for polycythemia vera and no etiology for secondary polycythemia was found. The presumption was that the temporal increase in hemoglobin may be due to AI therapy. Previous clinical trials have not reported an increase in hemoglobin in women receiving AI therapy for breast cancer. However, given the dramatic increase in hemoglobin in our 2 patients, we wished to test the hypothesis that inhibition of aromatase may lead to an increase in hemoglobin in postmenopausal women receiving AI therapy for breast cancer. The Cooper University Hospital Tumor Registry was used as a source of potential subjects. Women over the age of 50 years, diagnosed with ER+ nonmetastatic breast cancer between 2002 and 2006 were identified. Women included for study were postmenopausal, and treated with breast surgery +/− local radiation. Women receiving chemotherapy were excluded because of the potential effect of chemotherapy or therapeutic erythropoietin on the hemoglobin level. In order to be included for study, women needed to have a hemoglobin prior to surgery or prior to starting anti-estrogen therapy and at least 3 hemoglobin measurements over a minimum of 12 months after starting anti-estrogen therapy. AI included anastrozole and exemestane. Of 123 charts available for review, 82 had inadequate data for analysis. 27 evaluable women received only an AI. The mean age was 67±8 years and 67% were stage I. The mean hemoglobin before and during AI therapy was 13.7±0.4 and 13.2±1.1 gm/dl, respectively (Mann Whitney, p<0.09). 3/27 had an increase in hemoglobin after starting AI therapy by linear regression analysis (r ≥ 0.60). The increase in hemoglobin ranged from 0.9 to 1.1 gm/dl. As a control group, 11 women received tamoxifen rather than an AI. Mean age was 59±8 years and 54% were stage I. The mean hemoglobin before and during tamoxifen therapy was 13.0±1.0 and 12.8±0.8 gm/dl, respectively (Mann Whitney, p=0.53). 0/11 had an increase in hemoglobin by linear regression analysis (r > 0.60). 2 additional women received tamoxifen which was subsequently changed to an AI, 1 of whom had a mean hemoglobin of 12.9±0.3 gm/dl on tamoxifen which increased to 14.9±0.4 gm/dl on exemestane (Mann Whitney, p<0.05). 1 additional woman had a rise in hemoglobin of > 1 gm/dl while on AI therapy which decreased back to baseline when switched to tamoxifen. In conclusion, although the numbers are small and the data retrospective, these data suggest that AI therapy may be associated with an increase in hemoglobin in a subgroup of women treated with AI therapy for localized breast cancer. Given that AI have not been shown to significantly increase the systemic androgen level, the mechanism for the increase in hemoglobin remains unclear. A well designed, prospective study is needed to determine if AI have an effect on hemoglobin in women being treated for breast cancer.