Lessons From the WHI, Part 1: Hormone Therapy
Editor's Note: This is part 1 of Dr. Henry Black's interview with Dr. Rebecca Jackson on the Women's Health Initiative studies.
Part 2, about the diet, calcium, and vitamin D findings, is available here.
Part 3, on the cost-effectiveness of the estrogen and progestin arm, is available here.
And part 4, in which Dr. Jackson reminisces on paying it forward and the rewards of working on an important clinical trial, is available here.
Henry R. Black, MD: Hi. I'm Dr. Henry Black, Adjunct Professor of Medicine at the New York University Langone School of Medicine. I am here today with my colleague, Dr. Rebecca Jackson, from the Ohio State University.
I was the principal investigator at one of the 40 sites for the Women's Health Initiative (WHI) when I was at Rush University in Chicago, and Dr. Jackson has been there ever since -- about 20 years now. Could you give us the basic structure of the WHI and what questions it was asking?
Rebecca D. Jackson, MD: The WHI was developed in the early 1990s to be able to understand the major causes of morbidity and mortality in women as they age. It was designed to evaluate therapeutic interventions that were being recommended to women regularly, but about which we had relatively little knowledge of the impact on outcomes. Those 3 clinical trials were in the areas of hormone therapy to reduce risk for cardiovascular disease, improve quality of life and prevent fractures; a calcium-vitamin D trial aimed at preventing hip fractures and potentially reducing the risk for colorectal cancer; and a low-fat dietary modification to reduce the risk for both breast cancer and colorectal cancer, 2 very important cancers in women.
Dr. Black: Didn't we already know that estrogens would prevent heart disease?
Dr. Jackson: We certainly had a large amount of observational data from almost 2 decades before the WHI, suggesting that women who took hormone therapy had a lower incidence of cardiovascular disease and cardiovascular mortality. This was followed up by a study called the "Post-Menopausal Estrogen/Progestin Interventions Study" which also showed that estrogens had some favorable effects on reducing total cholesterol and low-density lipoprotein (LDL) cholesterol as well as increasing high-density lipoprotein (HDL) cholesterol, although there were concerns because it also increased triglycerides as well as some of the inflammatory and thrombotic markers. Unfortunately, observational data has the potential for significant bias, because women who make healthcare decisions with their healthcare providers to take hormone therapy or adopt what we think are risk-reduction behaviors might be doing other things as well, such as exercising more, watching their weight, and eating healthier diets. We really needed definitive randomized clinical trial data looking not just at an intermediate variable, such as a change in lipids, but also at the reduction in heart disease. Until the WHI, we didn't have that kind of information.
Dr. Black: Didn't we also think that estrogens cause breast cancer? Wasn't that a concern?
Dr. Jackson: Absolutely. When you look at an intervention, whether it is exercise or a drug such as hormone therapy, you need to look at the balance of benefits and risks. You can't just put blinders on and look at what you think are the beneficial effects of something like hormone therapy or estrogen, which has receptors in almost every tissue. You need to understand the full biologic implications. Implications that increased our concern in the decades before the WHI were whether estrogen, with or without a progestin, was associated with an increase in the risk for breast cancer; how large the risk was; and whether there were subgroups of women who might be at greater or lesser risk.
Dr. Black: It is hard to imagine the magnitude of this study. There were 68,000 women in the dietary study. There were over 160,000 women overall, including almost 100,000 in the observational study. I don't remember how many women were in the estrogen-plus-progesterone clinical trial, but it was a gigantic investment. The overall cost was approximately $600 million. [Editor's note: According to the National Heart, Lung, and Blood Institute (NHLBI), the WHI initiative -- including the hormone therapy, diet, calcium/vitamin D, and observational studies -- cost $705 million at the time.]
Dr. Jackson: The estrogen trials combined (estrogen alone and estrogen plus progestin) had more than 27,000 women. It takes studies that look at a big swath of American women to be able to answer those kinds of questions. Conducting a trial this large with a long-term intervention as well as long-term follow-up -- we have followed this cohort for more than 2 decades now -- is necessary to understand how we can achieve the promise of personalized medicine, and how we can start to look at combinations of risk factors or lifestyles and begin to stratify who is most likely to benefit and who is most likely to experience risk. Does this intervention provide an overall benefit from a public health perspective, or are there segments of the population for whom we would be able to individualize such recommendations?
Dr. Black: Let's take the trials one by one, beginning with the hormone intervention. If a woman had had a hysterectomy, she received estrogen, but women who still had a uterus took progesterone as well. How did that turn out?
Dr. Jackson: I'll start with the estrogen-plus-progestin trial. The trial was designed to test the hypothesis that a combination of estrogen plus progestin reduced the risk for cardiovascular disease and hip fractures, and the primary adverse effect or risk that was being studied was whether this had a significant impact on the risk for breast cancer. This trial randomized women 1:1 to either placebo or a combination of estrogen plus progestin taken daily. We chose to use oral estrogen plus progestin and to use it daily to maintain the blinding. Obviously, if women were having cyclic menstrual periods, we would know that they were on active therapy. We wanted to use the combination so that the women continued to be amenorrheic.
The second reason that we studied the oral estrogens was that it was the primary form of estrogen that was being prescribed in the United States at that time. The trial was planned originally to have an average of 8.1 years of follow-up. It was monitored on a regular basis. In 2002 the data showed that the balance of risks exceeded the benefits, and the trial was terminated early. Information was given to the women, and to the general public and healthcare providers to understand the results. The data showed that estrogen plus progestin, when used for the primary prevention of common diseases of women as they age --we are not talking about a trial that was designed to study whether hot flashes and vasomotor symptoms are reduced -- increased the risk for heart attacks and strokes, reduced the risk for hip fractures and total fractures, and there was a trend toward increasing the risk for breast cancer that crossed the monitoring threshold. Later studies also showed that there were increases in the development of mild cognitive impairment and dementia.
Editor's Note: This is part 1 of Dr. Henry Black's interview with Dr. Rebecca Jackson on the Women's Health Initiative studies.
Part 2, about the diet, calcium, and vitamin D findings, is available here.
Part 3, on the cost-effectiveness of the estrogen and progestin arm, is available here.
And part 4, in which Dr. Jackson reminisces on paying it forward and the rewards of working on an important clinical trial, is available here.
Henry R. Black, MD: Hi. I'm Dr. Henry Black, Adjunct Professor of Medicine at the New York University Langone School of Medicine. I am here today with my colleague, Dr. Rebecca Jackson, from the Ohio State University.
I was the principal investigator at one of the 40 sites for the Women's Health Initiative (WHI) when I was at Rush University in Chicago, and Dr. Jackson has been there ever since -- about 20 years now. Could you give us the basic structure of the WHI and what questions it was asking?
Rebecca D. Jackson, MD: The WHI was developed in the early 1990s to be able to understand the major causes of morbidity and mortality in women as they age. It was designed to evaluate therapeutic interventions that were being recommended to women regularly, but about which we had relatively little knowledge of the impact on outcomes. Those 3 clinical trials were in the areas of hormone therapy to reduce risk for cardiovascular disease, improve quality of life and prevent fractures; a calcium-vitamin D trial aimed at preventing hip fractures and potentially reducing the risk for colorectal cancer; and a low-fat dietary modification to reduce the risk for both breast cancer and colorectal cancer, 2 very important cancers in women.
Dr. Black: Didn't we already know that estrogens would prevent heart disease?
Dr. Jackson: We certainly had a large amount of observational data from almost 2 decades before the WHI, suggesting that women who took hormone therapy had a lower incidence of cardiovascular disease and cardiovascular mortality. This was followed up by a study called the "Post-Menopausal Estrogen/Progestin Interventions Study" which also showed that estrogens had some favorable effects on reducing total cholesterol and low-density lipoprotein (LDL) cholesterol as well as increasing high-density lipoprotein (HDL) cholesterol, although there were concerns because it also increased triglycerides as well as some of the inflammatory and thrombotic markers. Unfortunately, observational data has the potential for significant bias, because women who make healthcare decisions with their healthcare providers to take hormone therapy or adopt what we think are risk-reduction behaviors might be doing other things as well, such as exercising more, watching their weight, and eating healthier diets. We really needed definitive randomized clinical trial data looking not just at an intermediate variable, such as a change in lipids, but also at the reduction in heart disease. Until the WHI, we didn't have that kind of information.
Dr. Black: Didn't we also think that estrogens cause breast cancer? Wasn't that a concern?
Dr. Jackson: Absolutely. When you look at an intervention, whether it is exercise or a drug such as hormone therapy, you need to look at the balance of benefits and risks. You can't just put blinders on and look at what you think are the beneficial effects of something like hormone therapy or estrogen, which has receptors in almost every tissue. You need to understand the full biologic implications. Implications that increased our concern in the decades before the WHI were whether estrogen, with or without a progestin, was associated with an increase in the risk for breast cancer; how large the risk was; and whether there were subgroups of women who might be at greater or lesser risk.
Dr. Black: It is hard to imagine the magnitude of this study. There were 68,000 women in the dietary study. There were over 160,000 women overall, including almost 100,000 in the observational study. I don't remember how many women were in the estrogen-plus-progesterone clinical trial, but it was a gigantic investment. The overall cost was approximately $600 million. [Editor's note: According to the National Heart, Lung, and Blood Institute (NHLBI), the WHI initiative -- including the hormone therapy, diet, calcium/vitamin D, and observational studies -- cost $705 million at the time.]
Dr. Jackson: The estrogen trials combined (estrogen alone and estrogen plus progestin) had more than 27,000 women. It takes studies that look at a big swath of American women to be able to answer those kinds of questions. Conducting a trial this large with a long-term intervention as well as long-term follow-up -- we have followed this cohort for more than 2 decades now -- is necessary to understand how we can achieve the promise of personalized medicine, and how we can start to look at combinations of risk factors or lifestyles and begin to stratify who is most likely to benefit and who is most likely to experience risk. Does this intervention provide an overall benefit from a public health perspective, or are there segments of the population for whom we would be able to individualize such recommendations?
Dr. Black: Let's take the trials one by one, beginning with the hormone intervention. If a woman had had a hysterectomy, she received estrogen, but women who still had a uterus took progesterone as well. How did that turn out?
Dr. Jackson: I'll start with the estrogen-plus-progestin trial. The trial was designed to test the hypothesis that a combination of estrogen plus progestin reduced the risk for cardiovascular disease and hip fractures, and the primary adverse effect or risk that was being studied was whether this had a significant impact on the risk for breast cancer. This trial randomized women 1:1 to either placebo or a combination of estrogen plus progestin taken daily. We chose to use oral estrogen plus progestin and to use it daily to maintain the blinding. Obviously, if women were having cyclic menstrual periods, we would know that they were on active therapy. We wanted to use the combination so that the women continued to be amenorrheic.
The second reason that we studied the oral estrogens was that it was the primary form of estrogen that was being prescribed in the United States at that time. The trial was planned originally to have an average of 8.1 years of follow-up. It was monitored on a regular basis. In 2002 the data showed that the balance of risks exceeded the benefits, and the trial was terminated early. Information was given to the women, and to the general public and healthcare providers to understand the results. The data showed that estrogen plus progestin, when used for the primary prevention of common diseases of women as they age --we are not talking about a trial that was designed to study whether hot flashes and vasomotor symptoms are reduced -- increased the risk for heart attacks and strokes, reduced the risk for hip fractures and total fractures, and there was a trend toward increasing the risk for breast cancer that crossed the monitoring threshold. Later studies also showed that there were increases in the development of mild cognitive impairment and dementia.