ORLANDO, FL—Women with non-small-cell lung cancer (NSCLC) are significantly more likely to die of the disease if they take combined hormone-replacement therapy (HRT) than if they do not, according to a secondary analysis from the landmark Women's Health Initiative study reported here at the ASCO Annual Meeting.
The risk is particularly high for NSCLC patients who smoke and take HRT: One in 100 experiences an avoidable death from the cancer, said study head Rowan T. Chlebowski, MD, PhD, a medical oncologist at the Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center.
“Women almost certainly should not be using combined hormone therapy and tobacco at the same time,” he said.
“The other message is, women who are taking hormones and are diagnosed with NSCLC should stop taking the hormones,” Dr. Chlebowski said.
Combined estrogen and progestin treatment did not increase a woman's odds of developing lung cancer, the study showed.
Latest Blow for HRT
The findings represent the latest in “a series of problems that work against widespread use of combined hormone therapy,” Dr. Chlebowski said, referring to the fact that previous analyses from the Women's Health Initiative (WHI) showed that long-term use (at least five years) of hormone-replacement therapy combining estrogen plus progestin raises the risk of heart disease, stroke, blood clots, and breast cancer.
In WHI, a total of 16,608 postmenopausal women, age 50 to 79, were randomized to receive either placebo or combined hormone therapy with conjugated equine estrogen (0.625 mg) and medroxyprogesterone acetate (2.5 mg). The two groups were well balanced for all characteristics, including smoking.
The trial was stopped prematurely in 2002 when it became apparent that the risks of combined hormone treatment outweighed the benefits.
While fewer women are opting for the combination in light of those findings, about 15% of postmenopausal women in the US still use HRT, Dr. Chlebowski said, noting that about 15% of postmenopausal women in the US smoke.
HRT & Lung Cancer
One reason researchers believe that hormones play a role in NSCLC prognosis is because women tend to have better outcomes than men, he said. Also, estrogen receptors are commonly expressed in lung tissue and lung cancer, Dr. Chlebowsk.
However, the influence of HRT on lung cancer has not previously been studied in a randomized controlled fashion.
For the study, the researchers looked at lung cancer incidence and mortality during 5.6 years of the trial intervention and 2.4 years of additional follow-up.
There was no suggestion that HRT influenced small-cell lung cancer incidence or outcome, Dr. Chlebowski reported.
As for non-small cell lung cancer, there were 96 cases among the 8,052 hormone users vs 72 among the 7,678 women who took the placebo. This corresponded to a 28% increased incidence of lung cancer in the HRT arm, but the difference did not reach statistical significance.
There were 67 deaths from non-small cell lung cancer after a diagnosis of the cancer among the hormone users vs 39 among the women who took placebo. This corresponded to a significant, 61% increased risk of dying from NSCLC for women in the HRT arm.
After a diagnosis of non-small cell lung cancer, hormone users lived a median of 9.4 months, compared with 16.1 months among women taking placebo.
Lung Cancer & Hormones
Study Discussant Jill Siegfried, PhD, Professor and Chair of Lung Cancer Research at the University of Pittsburgh Medical Center, provided a thorough review of published and unpublished data on smoking, lung cancer, and hormones. She noted that several previous studies have also suggested a role for estrogen in driving tumor growth.
In one retrospective chart review, for example, women who used HRT for at least six weeks prior to diagnosis had a shorter median survival time: 39 vs 79 months compared with women who did not use HRT. And in a study of about 200 patients with Stage IV lung cancer, high serum B-estradiol levels were associated with worse outcomes—in both men and women.
Dr. Siegfried then shared some of her own unpublished data showing that in a mouse model of NSCLC, higher estrogen-receptor beta levels were associated with a higher risk of death. Progesterone receptor expression, however, appeared to have a protective effect.
Deadly or Protective?
While estrogen appears to worsen outcomes after diagnosis, other research suggests that estrogen can be protective before lung cancer diagnosis, she continued.
Unpublished data suggest that phytoestrogen is associated with lower lung cancer risk (Spitz et al), and other research suggests that estrogen can stimulate immune responses, Dr. Siegfried said.
Scientists are just beginning to explore the reasons for the incongruities. Among the hypotheses being explored in her lab are whether neoplastic cells lose estrogen in the presence of tobacco and whether there is a switch that shuts off the local production of estrogen in normal tissue, but not in tumors.
The question-and-answer period served as a forum to challenge the results of the new study and offer more hypotheses. One attendee noted that a Swedish study of 40,000 women showed that HRT has protective effects in smoking-related cancers—“So there is a difference and we need to sort it out,” the questioner said.
Said another attendee, “Given that you are going back and looking at medical records, could it be that what you're seeing are the complications of HRT in those women who are smokers versus something intrinsic in the lungs?”
Barnett S. Kramer, MD, MPH, Associate Director for Disease Prevention at the NIH, said that one potential confounder that could help to explain some the conflicting results is cancer screening. “Screening tests can drive up incidence and improve survival times, so without that information it is hard to judge other factors,” he said, making a plea for future studies to obtain and control for that data.
And in an interview, Bruce E. Johnson, MD, Director of the Lung Cancer Program at Dana-Farber Cancer Institute, said that the robust, randomized design of the current trial gives it more credence than previous case-control studies that found no link between lung cancer and HRT.
The bottom line: “There's plenty of food for thought,” Dr. Siegfried said.