The Women’s Health Initiative (WHI) initiated 2 randomized trials in 1993. The trials included conjugated equine estrogen (CEE) in postmenopausal women with a prior hysterectomy and CEE plus medroxyprogesterone acetate (MPA) in those with intact uterus. The CEE + MPA trial closed after 5.6 years when it was found that overall health risks exceeded benefits with cardiovascular disease and invasive breast cancer (diagnosed at a more advanced stage). A recent article by Chlebowski (UCLA)
et al examined the 40-year trends of those studies. The CEE alone trial was stopped after 7.2 years median followup based on increased risk of stroke and no cardiovascular disease benefit. As additional followup breast cancer incidence was found to be significantly reduced by CEE alone.
During the 5.6 years of CEE-MPA intervention there was a year-to-year increase in breast cancer. Just after the CEE-MPA intervention, the relative risk of breast cancer dropped though the hazard ratio remained above 1 for more than a decade. In the 7.2 years of CEE-only intervention the hazard ratio was less than 1 and remained beneath 1 through longer followup. It should also be noted that no health benefits were identified for younger postmenopausal women on CEE alone. After the CEE+MPA trial became known publicly, there was a precipitous decline in all prescriptions for hormone replacement but particularly steep for CEE+MPA (66% drop in the USA).
Now 40 years later, by 2015, when compared to 2002 there have been upwards of 200,000 fewer cases of breast cancer.
As both trials began, the incidence of breast cancer in white women was higher than in black women. In 2003-2004 after the trials began to be reported nationally, there was a sustained decrease in breast cancer incidence. However, the decrease in black women was not sustained. Subsequently, an increase in incidence in black women was noted. As a result, the incidence of breast cancer in black and white women is roughly comparable.
In white women the use of CEE + MPA remains low. Obesity has increased but was already high and there has been only a slight increase in the most recent decade. The authors noted that “Although other factors may be involved, the sustained reduction in hormone therapy use, which began in 2003-2004, can provide a first-order explanation for this finding.” With respect to mediating factors, studies have shown that the 1-year change in mammographic density after starting CEE + MPA predicted all subsequent breast cancer increase.
Why didn’t the breast cancer risk drop in black women as much as in white women. Though hormone replacement decreased in both black and white women, the starting point was substantially different. CEE + MPA use in black women prior to 2002 was only a fraction of that reported in white women. When looking at the filling of at least one prescription for hormone therapy in 2001, for white women it was 5.4% whereas it was 0.4% for black women. It was therefore less likely that reducing hormone replacement would have a major impact on risk in black women. Moreover, it has been shown that the incidence of hysterectomy is approximately 38% higher in black women compared to white women. Furthermore, approximately 70% more white women were taking CEE + MPA compared to black women.
In conclusion, changes in the use of hormone therapy provides a plausible explanation for the lower risk of breast cancer seen in white women since 2003. It also explains the discordance between black women who have seen and increase in breast cancer incidence seen over the last decade.