QUEST WEBINAR 2
DR. ANDREA DUNAIF
Hi, my name is Trisha Winchester and I’m the Senior Manager for the Clinical and Education team for the Cardiometabolic and Endocrinology clinical franchise with Quest Diagnostics. I am honored here today to be joined by Dr. Andrea Dunaif who is a leading expert in the field of PCOS. She is the Professor and System Chief of Endocrinology, Diabetes and Bone Disease at Icahn School of Medicine at Mount Sinai in New York City. Dr. Dunaif, take it away.
Thank you so much. The association between features of androgen excess in women and metabolic risk has really been recognized for a long time and probably the person who first popularized it was Jean Vague when he talked about fat distribution – in women, lower body or pear shape and in men having upper body or apple shape – and that women who had upper body obesity also had increased risk for diabetes. We now know that women with metabolic syndrome which is characterized by upper body or central obesity, higher blood pressure, higher triglycerides, lower HDL cholesterol and insulin resistance in most of the diagnostic criteria for metabolic syndrome, this is defined as either high or fasting glucoses or impaired glucose tolerance. Women who have metabolic syndrome have higher androgen levels than women without metabolic syndrome in epidemiologic studies and this is true in both post-menopausal and pre-menopausal women and here is just an example of a study. We are looking at individual features of metabolic syndrome such as abdominal obesity, increased triglycerides, low HDL, high blood pressure, high glucose or the entire syndrome. The black solid bars are testosterone levels. You can see for each feature of the syndrome except for blood pressure, the testosterone levels are significantly higher than in women without metabolic syndrome as well as for the overall syndrome.
It’s now recognized that elevated androgens are very, very common in women in general and that these increases increase with body weight. This is a study we did several years ago and the red bars here are hyperandrogenemia defined as elevation of testosterone, bioavailable testosterone, or dehydroepiandrosterone sulphate and these green bars are women with normal cycles. All of these women have regular menstrual cycles and no evidence of hirsutism. You can see that even in women who aren’t obese, about 16% have elevated androgen levels and this goes up to 25% in the overweight group – BMI 25-30 – and to almost 40% in women with a BMI of 30. So, these are just unselected women in the population, so androgen excess – very, very common. And it is associated with metabolic risk. This is looking at the risk for dysglycemia – this is either high fasting glucose, 100 or greater, or impaired glucose tolerance – and for metabolic syndrome. You can see that the risk for both is almost three-fold increase if you have elevated androgen levels.
This association…in fact, when it was noted in Polycystic Ovary syndrome – the association between elevated androgens and insulin – has led to the question of how these metabolic and reproductive abnormalities are related. The first question that was asked – “Is this a causal association? Do high insulin levels cause high androgen levels or vice versa?” We now know that insulin is a reproductive hormone and that insulin can act directly on the ovary in concert with LH to increase ovarian androgen production. It can act on the adrenal by increasing sensitivity to ACTH to increase adrenal androgen production and it can also act on the pituitary to enhance LH release which further drives ovarian androgen production.
An additional source of androgens in women has been appreciated and that’s the adipose tissue. There is an enzyme in adipose tissue, AKR1C3. It turns out that this enzyme which converts circulating pool of C19 adrenal androgens into their more potent bioactive androgens, testosterone and dihydrotestosterone, into the more potent 11-keto and 11-keto testosterone and dihydrotestosterone. This enzyme activity is upregulated by insulin. So, it appears that as we get hyperinsulinemia, this drives adipocyte production of androgens and gives us an additional pool of androgen excess and may be one of the explanations of why we see increasing androgen levels in obese women. You can also show with androgen receptor blockade that androgens do contribute both to changes in visceral fat – and what we’re looking at here is decreases in visceral fat, and this is placebo and this is Flutamide, and you can see in this study you get more decreases with the androgen receptor antagonist Flutamide and the same thing if we look in the study at insulin sensitivity. When you antagonize androgens compared to placebo you get greater increases in insulin sensitivity. But these are pretty modest changes.
It’s the androgen excess and androgens contribute to insulin resistance. The syndrome that really typifies the association between cardiometabolic risk and reproductive dysfunction is Polycystic Ovary syndrome, and most of the data I’m going to discuss with you today is from PCOS. PCOS is a disorder characterized by reproductive abnormalities, neuroendocrine abnormalities, with increase GnRH which increases LH while simultaneously suppressing FSH. We have ovarian and adrenal abnormalities with increased androgen biosynthesis from both the ovaries and the adrenal glands. Anti-Müllerian hormone, which is made by maturing ovarian follicles which contain the oocytes is elevated in PCOS reflecting the number of increased follicles, but it may also have central actions to increase GnRH. Then we have feedback abnormalities and testosterone antagonizes the ability of estradiol to slow GnRH pulses so we get a self-sustaining so-called vicious cycle. It was recognized in the 1980s that women with PCOS had insulin resistance, hyperinsulinemia and PCOS is a risk factor for obesity.
Now, PCOS is conceptualized as a disorder that affects women across the lifespan. In adolescence – this is a disease that presents with menarche with maturation of the reproductive axis and initiation of menstrual cycles – though we can find evidence for markers of PCOS in girls before menarche and the daughters of women with PCOS, so during adolescence and young adulthood it’s mainly a reproductive disorder and it starts…the reproductive features improve in the 30s and in fact often menses become regular and these women are able to conceive. The metabolic abnormalities are also present in adolescence but they really become predominant as women age. Again, what we see, depending on the population, is about 50-80% of women with PCOS are obese, and obesity really worsens both the reproductive and the metabolic outcomes. About 30% of women with PCOS have prediabetes or what we call impaired glucose tolerance or impaired fasting glucose, and about 10% have type 2 diabetes, and this risk is about a 4-fold increase with a younger age of onset. In obese women with PCOS, about 70% have metabolic syndrome. They have modest increases in LDL – not particularly striking – and this has been shown to be related to androgen levels. Since PCOS emerges in early adolescence, it does provide opportunities for very early interventions and perhaps prevention of adverse long-term health outcomes.
Let’s now turn to the question – is PCOS an independent risk factor for cardiometabolic conditions? It’s now very well confirmed that PCOS increases the risk for type 2 diabetes and impaired glucose tolerance. We can see here from this summary of early studies, here we see about 60% on average of women with PCOS have normal glucose tolerance. About 40% have either impaired glucose tolerance or type 2 diabetes. If we calculate from what we know of the prevalence rates of type 2 diabetes in the United States, you can calculate that about 20% of type 2 diabetes in women 18 to 44 years old is probably accounted for by PCOS. Here is a much larger population-based study from a Scandinavian health registry showing that women with PCOS have a 4-fold increase risk for type 2 diabetes. They have eight cases compared to two cases in women without PCOS per thousand person-years and with a significantly younger age of onset.
The cardiometabolic conditions associated with PCOS are really magnified by the presence of obesity. We can see them in women with PCOS who are non-obese but they really become much more prevalent when you have the presence of obesity, which is increasingly common in women with PCOS. Steototic liver disease or MASLD which was formerly known as non-alcoholic fatty liver disease – all of these are prevalent in obese women with PCOS. A very important question has been – are women with PCOS actually at increased risk for cardiovascular events? There, we are talking about MI and stroke and coronary revascularization. We know that diabetes completely abolishes the protective effect of being a premenopausal woman. We see in individuals without diabetes that men are at much higher risk than women for cardiovascular events. However, if they have diabetes, women with diabetes totally lose protection and in fact have a greater increase in risk for cardiovascular events than men do with diabetes – so a 5-fold increase compared to a 3-fold increase.
We just went through women with PCOS have a number of risk factors for cardiovascular disease. They have traditional risk factors, diabetes, obesity, physical inactivity, hypertension, dyslipidemia, and they also have these emerging or nontraditional risk factors such as hypertensive disorders of pregnancy, gestational diabetes and an increased risk for depression. Unfortunately, there are no prospective long-term studies of women with PCOS that have followed these women long-term and importantly no studies of women beyond menopause, which is when we really start to see an increased risk for cardiovascular events in women. There have been studies that have used the regular menses as a proxy for PCOS and looked at long-term health outcomes. This is from the nurses’ health study where they had data on menstrual cycles in the nurses when they were 18 to 22 with descriptors – so regular, usually regular, usually irregular, and very irregular. When you compare the women with usually irregular or very irregular cycles to those with regular cycles, the women with irregular cycles had about a 1.5 relative risk, 50% increase in risk for cardiovascular events. This was true in the multivariate analysis adjusted for confounders like parity, contraceptive use and BMI. So, this study really suggests that there is in fact an increased risk of cardiovascular events in women with PCOS with PCOS being defined as irregular menses.
Recently, there are studies with large cohorts that have been able to determine PCOS reliably through hormonal measurements and have examined the risk for major adverse cardiovascular events. This is from a large population cohort in Finland and they showed that these women were having increased major cardiovascular events compared to control women with the ratio being an approximate 2-fold increase in risk. The problem here is there weren’t that many women with PCOS – 386 – only 20 events and 27 in the controls. This is because the population is really quite young. We certainly need studies that will follow women into the postmenopausal period to provide a robust answer to this question, but there is certainly a lot of smoke suggesting that there really is an increase in cardiovascular events.
Just to summarize, it remains controversial whether PCOS is associated with increased risk for cardiovascular events. These studies have been limited by the heterogeneity and the diagnostic criteria, the difficulty in these large cohorts assigning PCOS status with the use of the ICD-9 or 10 codes, and a lot of these samples aren’t really good representations of the general population…pointed out that women with PCOS have not yet been studied to an age when cardiovascular events become prevalent, that is postmenopausally, and many of these existing large cohorts with data on cardiovascular outcomes lack data on reproductive status or hormonal measurements, so it’s exceptionally difficult to answer this question with our existing datasets.
To summarize what I’ve shown you, we know that hyperandrogenism is quite common in otherwise reproductively normal women and it increases with increasing BMI, that insulin is a reproductive hormone and can increase androgen levels and that androgens themselves can contribute to insulin resistance. PCOS is associated with a major increased risk for type 2 diabetes about 4-fold, and this has been corroborated now in multiple studies with strong evidence. PCOS is associated with multiple traditional and emerging cardiovascular disease risk factors. Long or irregular cycles as a surrogate for PCOS are associated with increased risk of cardiovascular events. Thank you very much for your attention.
Thank you so much, Dr. Dunaif. Because of the high prevalence of hyperandrogenism in women, would you say that it’s fair to screen women for elevated levels of androgens or what is your suggestion for that?
I think that’s a really important question – whether testosterone could be a novel risk factor in women. We really need to look at that in large cohorts. As I said, the problem is reproductive hormones aren’t measured in these cohorts where we have cardiovascular outcomes.
With PCOS as an important diagnosis for women to know that they have in order to be able to prevent future metabolic risk, what are some of the actions they can do? After diagnosis, what does that treatment plan look like for women who have been diagnosed?
In women who are overweight or obese, really one of the most important factors is reducing body weight because that improves not only their metabolic risk, and we know from the diabetes prevention program that the most powerful intervention for reducing risk of progression to type 2 diabetes was weight reduction, and as little as 5-10% reduction in body weight really reduces risk of progressing, so that should be very high priority. But it also improves the reproductive abnormalities and reduces the testosterone and women can resume ovulating. So, I think that is a really major goal and with the availability of some of these newer GLP-1 receptor agonist weight loss medications, it becomes much more feasible to achieve meaningful weight loss. I think we need to see an appreciation of PCOS and its long-term metabolic implications amongst general healthcare providers. Right now, the knowledge is really amongst obstetricians, gynecologists, and endocrinologists and this needs to be something that’s out there in the community. Women can really help themselves because it’s very easy to self-diagnose PCOS. If you have irregular cycles, often these women have symptoms of male hormone excess, most typically hirsutism. So, many of the women I see in my practice come in and they’ve made their diagnosis from going online so they need to be their own advocates and education physicians and demand from health care that this condition is appropriately addressed.
I think it’s really important in order for patients to acknowledge what their condition is and for providers, on the other hand, to be able to acknowledge that increased risk for future metabolic complications as well to address that. That’s what we’re here for, right – to improve patient care. So, I think this is one step in terms of addressing the education that’s needed about PCOS – about diagnosis, about what the risks are associated with. Dr. Dunaif, thank you so much for sharing your expertise with us on the PCOS and the cardiometabolic abnormalities that can be associated with it.
Thank you. It’s been my pleasure to get the message out there.
