Major Finding: After 11 years of follow-up, the incidence of breast cancer is up by 25% in the dual-hormone therapy group relative to placebo. Yet the relative increase in mortality is 96%.
Data Source: An updated analysis of the Women's Health Initiative randomized trials.
Disclosures: Dr. Chlebowski disclosed that he receives grant support from Amgen and is on the speakers bureaus for AstraZeneca and Novartis. Dr. Coates reported having no relevant financial disclosures.
SAN ANTONIO – Menopausal hormone therapy with estrogen plus progestin doubles a woman's risk of death from breast cancer, nearly doubles the risk of death from non–small cell lung cancer, and increases the risk of death from colorectal cancer by 54%, according to an updated analysis of the Women's Health Initiative randomized trials.
Because breast and lung cancer are the top two causes of cancer mortality in women, these are sobering findings with important clinical implications, observed Dr. Rowan T. Chlebowski, professor of medicine at the University of California, Los Angeles.
The 54% increased risk of death after diagnosis of colorectal cancer in Women's Health Initiative (WHI) participants who were randomized to combined-hormone therapy rather than placebo was a trend that didn't achieve statistical significance. But it's nonetheless a finding that crushes the enthusiasm that greeted an earlier WHI report of a 44% reduction in the incidence of colorectal cancer in combined-hormone therapy users after 5.6 years of follow-up (N. Engl. J. Med. 2004;350:991–1004).
Given the initial observation of fewer colorectal cancers being diagnosed in the combined-therapy arm of the WHI, investigators were quite surprised by the tumor characteristics of these cancers at time of diagnosis: The colorectal cancers arising in the combined-therapy group – although fewer in number – were much higher risk. In all, 76% of them were pathologically staged as regional or metastatic disease, compared with 48% of colorectal cancers in women on placebo, and 59% percent of the colorectal cancers detected in combined-hormone therapy users were lymph node positive, compared with just 29% in placebo-treated controls.
The WHI consisted of two separate National Institutes of Health–funded, randomized trials that profoundly altered the management of menopausal symptoms. In the early 1990s, more than 40% of all postmenopausal women were on hormonal therapy with estrogen alone or in combination with progestin. Following the initial WHI report of multiple adverse effects of estrogen plus progestin, the popularity of hormone therapy dropped off the table.
One WHI study involved 16,608 postmenopausal women aged 50–79 years with an intact uterus who were randomized to estrogen plus progestin or to placebo for a median of 5.6 years. The other study included 10,739 postmenopausal women with prior hysterectomy who were randomized to conjugated equine estrogens alone or placebo for an average of 7.1 years.
In the examination of WHI trends for the big-three (breast, lung, and colorectal) cancers in women, there is a consistent disparity between their relatively modestly increased incidence in dual-hormone therapy users, relative to placebo, and the far larger death rates resulting from these cancers. For example, after 11 years of follow-up, the incidence of breast cancer is up by 25% in the dual-hormone therapy group, relative to placebo. Yet the relative increase in mortality is 96%. Similarly, the incidence of non–small cell lung cancer (NSCLC) was 23% greater in women on combined-hormone therapy than in those on placebo, but the risk of death from NSCLC was 87% greater.
“The greater effect of estrogen and progestin on deaths from breast, lung, and colorectal cancer – [compared with] the effect on incidence – [suggests that] combined-hormone therapy facilitates growth and metastatic spread of established cancers, perhaps mediated by angiogenesis stimulation,” the oncologist said, adding that “in a variety of preclinical models, estrogen and progestin are potent angiogenesis stimulators.”
The investigators' initial hypothesis was that nearly all the increase in breast cancers associated with combined-hormone therapy would involve estrogen receptor–positive tumors. Not so. In fact, the new analysis – based upon 11 years of follow-up and 678 cases of breast cancer – shows that all breast cancer subtypes appear to be increased, relative to rates in the placebo arm.
For example, combined-hormone therapy was indeed associated with an adjusted 27% greater increase in estrogen receptor–positive breast cancers than with placebo in a multivariate analysis, but it was also associated with a 40% increase in estrogen receptor–negative tumors, compared with controls. Also noteworthy were the combined-therapy group's adjusted 78% increase in triple-negative cancers, the twofold increase in HER2-overexpressing tumors, and the 37% increase in HER2-negative tumors.
Nearly all the increase in lung cancer deaths associated with dual-hormone therapy resulted from NSCLC. Hormone therapy had no effect upon small cell lung cancer rates.