Posted: March 3, 2017

For years doctors have known that a mainstay of breast cancer treatment in post-menopausal women — aromatase inhibitors — causes loss of bone mineral density.  Those drugs in the US are  Anastrozole (Arimidex) Letrozole (Femara) and Exemestane (Aromasin).  Numerous clinical trials have shown that this loss can be slowed and even prevented by aggressive intervention with drugs that inhibit bone resorption, like Fosamax.  Now comes a sobering study from Yale looking at practice patterns in the US Medicare population.  Published in the Journal of Oncology Practice, (click here for the abstract) one of several newer journals from the American Society of Clinical Oncology, Stratton and colleagues looked at practice patterns in women put on aromatase inhibitors as adjuvant therapy for newly diagnosed breast cancer.  They looked at Medicare billing data as a clue as to which women received bone density studies, before, during and at the conclusion of aromatase therapy — typically lasting for five years.  They did not examine actual prescribing practice patterns but used the bone density study as a surrogate marker.  Despite the fact that all prescribing doctors should know about the deleterious effect of this class of drugs, only 51% of women had baseline bone density tests.  Only 34% of women had both a baseline scan and a follow up scan (to assess the potential deleterious side effects of the drug).  The older the woman, the less likely to get the test.  Median income was an important variable; women with the highest family income were most likely to get the test.  White women were more likely as well.  No attempt was made to look at fracture risk or other outcomes in women screened versus not screened.

Dr. Stark comments: This is a very disturbing study.  As a clinician I fought to keep these women’s bones intact.  Some of them required intravenous drugs for this purpose.  To see doctors ignoring this problem on a wholesale level is disheartening. Presumably the doctors who need this advice the most are the ones least likely to read this article.  Neither the authors nor I see a quick fix for this problem.