Menopause Live (24 January, 2011) From IMS

De and colleagues recently reported on a retrospective study to evaluate use of hormone therapy (HT) and breast cancer incidence among Canadian women from 2001 to 2006 [1]. Data for reported HT use from 1996 to 2006 among approximately 1200 women 50–69 years of age were obtained from the National Health Population Survey, confirmed by HT prescription rates in a national pharmacy database, and extrapolated to represent the Canadian female population. The Canadian Cancer registry provided age-standardized breast cancer incidence rates and mammography rates were gleaned from the Canadian Community Health Survey. These data revealed a decline in HT use from 2001 to 2006 (from 11.6 million to 8.5 million prescriptions), with the largest reduction occurring from 2002 to 2003. HT use decreased for all types: estrogen only, progestin only, and estrogen–progestin combined products (data for women using non-combined estrogen plus progestin were not available). Combined product use declined most among women aged 50–69 years from 2002 to 2004, from 12.7% (95% confidence interval (CI) 10.1–14.2%) to 4.9% (95% CI 3.4–6.8%). The overall greatest decline in HT use occurred from 2000 to 2006, with an average decline of -7.6% per year (95% CI -5.9 to -8.9%). Incident breast cancer among Canadian women over 40 years of age peaked in 2002, declined from 2002 to 2004, and then rose again from 2004 to 2006. Among women aged 50–69 years, incidence fell by -9.6% from 2002 to 2004, from 296.3 per 100,000 in 2002 (95% CI 290.8–300.5 per 100,000) to 268.0 per 100,000 in 2004 (95% CI 263.3–273.5 per 100,000). Breast cancer incidence rose again among women aged 50–69 years by 4.5% annually from 2004 to 2006 (95% CI 3.1–7.2%). Mammography rates for women in the 50–69-year age group increased from the late 1990s to 2000 and then remained stable at about 72%.

Comment
This study evaluated trends in breast cancer incidence and HT use among Canadian women prior to and since the release of the data from the estrogen plus progestin (EPT) arm in the Women’s Health Initiative (WHI) study in 2002 [2]. US researchers have reported similar trends with reduced breast cancer incidence in the immediate years after the WHI EPT arm results were publicized and HT use dropped. Epidemiological data suggested a 7–8% decrease in incidence among US women from 2002 to 2003 [3-5] and the Kaiser group identified an 11% decline in incidence among women in northern California from 2001 to 2003 [6]. An analysis of breast cancer incidence among women in Portland, Oregon revealed declines in incidence from 2001 to 2004, with a leveling off following 2004 [7]. Data regarding estrogen-receptor status were not available in the study by De and colleagues [1]. However, prior research that did include receptor status data revealed a 12% decrease in estrogen receptor (ER)-positive tumor incidence from 2002 to 2003 in the US [4, 5]. In the Portland study, ER-positive tumors followed the overall trend of all breast cancers identified. However, the incidence of ER-negative tumors, which are thought to be unaffected by HT, inexplicably dropped from 2002 to 2006 [7]. 

These data combined suggest a reduced incidence of breast cancer immediately following the decline in HT use that occurred after the WHI EPT arm results were released. The fact that incidence again increased in the Canadian study from 2004 to 2006 and leveled off in the US studies at 2004, despite an ongoing decline in HT use, is thought-provoking. Putting aside the fact that many other factors that are relevant to breast cancer incidence, such as smoking, alcohol consumption, and the use of estrogen agonist/antagonists, were not evaluated, these data argue against a causal link of HT initiating breast cancer. Rather, these data suggest that the use of HT may promote tumor growth of pre-existing breast malignancy and that HT is unlikely to be causing the cancers directly.

Ivy M. Alexander
Professor and Director, Adult, Family, Gerontological, and Women’s Health Primary Care Specialty, Yale University, New Haven, CT, USA