Chapters Transcript The Metabolic-Reproductive Syndrome vs PCOS or What Difference a Name Makes Course: Updates in Endocrinology 2024 And I'd like to go over some aspects of PCOS. There's so many aspects that we don't have time to cover them all, but as many as we can do. Um, I have no disclosures, and I am going to discuss, uh, investigational use of commercial products, unlabeled use. So the outline of the talk will be to first define PCOS, which in and of itself has been uh controversial, then talk about the pathogenesis of PCOS, um, then give an overall general model and talk about associated comorbidities such as diabetes and OSA and then uh finish with new and emerging modalities for the diagnosis and treatment. So first defining PCOS, um, so it's one of the most common reproductive endocrine disorders in women. I think most people know that, and it affects anywhere from 8 to 13. Some people say 5 to 12, it's about 10% on average, and it's across all ethnicities and races. Um, the syndrome is complex in that it involves three major factors anovulation, hyperandrogenmia and metabolic and cardiovascular derangements. The anovulation presents clinically as oligo amenorrhea, infertility, and um and can lead to endometrial carcinoma. The hyperandrogenemia manifests as hirsutism, acne, and alopecia usually, and then there's a litany of uh metabolic and cardiovascular dis uh derangements that we'll get into as we go forward. So, the thing that's become so confusing to people is that there are these separate criteria depending on where you practice and what, what, you know, criteria you want to use. There's so-called NIH criteria from 1990, the Rotterdam criteria, and then the Androgen Excess Society criteria. And the simplest one is the NIH criteria, which requires. Uh, clinical evidence of androgen excess or elevated testosterone or free testosterone and oligo ovulation. So it's, it's can be almost diagnosed clinically. I mean, if you examine somebody in their hair suit and they have a high testosterone level and you've excluded all the other possibilities for those same findings, then um you make the diagnosis. Rotterdam kind of changed the people's outlook and they require um. The presence of. Polycystic ovaries on ultrasound and how many people in the audience order an ultrasound to make a diagnosis of PCOS? Not many, how many people have been referred patients with preexisting ultrasonography and they're told they have PCOS? That's the same experience I've had and um. It's very disappointing to the patient to tell them that they don't have PCOS, although they should be pleased, but, um, the, the cysts, they're not cysts, they're immature or uh underdeveloped anthral follicles, and it requires the radiologist to count them on any given slice on the ultrasound and also measure the ovarian volume and. I think what they do is they look at the, the screen and they say, oh, looks like polycystic ovary syndrome, and it can be confusing because there's something called multicystic ovary syndrome, which is seen in hypothalamic amenorrhea. So it's not that clear cut, so I tend not to use the ultrasonography unless there's a question of the diagnosis and then I'm not gonna talk about the androgen excess Society criteria for today. So, um, there is a differential diagnosis for PCOS and that includes uh 21 hydroxylase deficiency CAH, Cushing's, elevated prolactin, primary hypothyroidism, macromegaly, primary ovarian insufficiency, simple obesity, virulizing neoplasms and medications, and in every case, you don't have to exclude all of these, but if an Ashkenazic Jewish woman comes in. And has hirsutism and high testosterone, you're really obligated, for instance, to do a uh ACTH SIM test, or at least measure an early morning 17 hydroxy progesterone. Um, Cushing's, I'm not gonna talk about, I think we all know that prolactin elevation is common in PCOS, but it can also cause a picture that looks like PCOS with oligomenorrhea. And, and then the others are on the list and you can look at those at your leisure. So in 19, 2012 rather, um, the NIH had this major conference that we went to and it was, you know, supposed to be expert opinion and we, we, meaning me and a few other endocrinologists wanted to change the name of the syndrome to the reproductive metabolic syndrome, because that is more inclusive and it's not misleading. PCOS is a misleading term. And but it's gonna stick with us it's not gonna change and the the consensus at the conference or the executive summary said that the name PCOS is a distraction and an impediment to progress which I agree with. It causes confusion and is a barrier to effective education of clinicians and communication with the public and research funders. The name focuses on a criterion, polycystic ovarian morphology, which is neither necessary nor sufficient to diagnose the syndrome. It's time to expeditiously assign a name that reflects the complex metabolic hypothalamic pituitary, ovarian, and adrenal interactions. So even though we want to call it reproductive metabolic syndrome, that won't happen because the term is so entrenched in the literature and. And it it to call RMS, you know, people will get confused and they won't know what that is. So, I wanted to show you, um, A picture of 3 patients with PCOS that have variable manifestations clinically, so the patient on the left has just hirsutism, no acne. The patient in the middle has acne, but no hirsutism, and the patient on the right has also uh just hirsutism. And in African American patients in particular, they can get folliculitis barbae, which is ingrown hairs, and that can be extremely frustrating and difficult to deal with for the patients, and you can imagine it it occurs in men, but these women have that problem too. And then here's a picture of a, a postmenopausal woman who came in with this uh vertex uh alopecia, and she had actually an ovarian tumor, um, which turned out to be a very interesting case, but she didn't have PCOS, she had an ovarian neoplasm. So how do we determine if a woman is her suit or not? Well, first of all, I never tell a patient you don't have hirsutism or you don't have extra hair growth, because it could be limited to one area and it's bothersome to the patient. It's not a cosmetic issue per se, but if they have a lot of facial hair growth, they may not have a fermin-G Galway score of 8 or greater, which is. What we use is the cutoff, and I want to point out something that you may see in the international consensus guidelines, evidence-based. There's, we're gonna have a letter in the JCM. The um Helenna Te used 2 to 4 as the cutoff, and that's just not right, because almost everyone has some degree of uh excess hair, and the score goes from 0. It to 4 in 9 areas, so the upper limit is 36, but greater than 8 is considered uh abnormal. So we, it's not a hard and fast rule, but we use that as a a marker. And in terms of uh alopecia, you can see that uh there's type 1, type 2, and type 3, and it's which is more extreme as you get a higher number, and it's usually on the vertex and or in the um in the ax or back of the scale back of the head. So what are the goals for treatment of RMS? I'll, I'll variably, I'll change it interchangeably, uh, just so people don't get confused, but what are the goals for treatment? Well, the first is to ameliorate the hyperandrogenic manifestations like hirsutism, acne, and alopecia, if that's what the patient's concerned most about. Um, it's also important to manage the underlying metabolic abnormalities and reduce the risk for type 2 diabetes and cardiovascular disease, and I should mention that cardiovascular disease is an enigma because these women have just about every risk factor for coronary disease, but there's never been a study to show that the rate of myocardial infarction is higher in women with PCOS, and that's. Some people, well, there are theories about why that may be, but that's, that's the case. And um we also want to prevent endometrial hyperplasia, which could turn into carcinoma. And contraception for those not pursuing pregnancy and ovulation induction for those who wish to get pregnant. So those are kind of the overall treatment goals for this uh syndrome. Now, why, why does PCOS develop and the next slide is gonna show you my, this is my take on the pathogenesis of PCOS. So let's start, let's see, where should we start? Let's start at the top. GNRH normally is secreted in a pulsatile fashion and subsequent, there's an obligatory subsequent um pulse of LH and. The pulse frequency of GNRH secretion is higher in PCOS than it is in normal and what The mistake that's often made is that people will get a random LH and FSH and because it's pulsatile, you may get them at a point where they're at the nadir or a point where they're at the peak, but in general, the LH is high and FSH is relatively low, and this is known from uh many studies, mainly at Mass General, where they studied women for 24 hours and did LH and FSH every 10 minutes. So, what I find so interesting is that when the pulse frequency of GNRH is high, there's preferential transcription of LH beta, and that's shown by the heavy line here. And when the GRH pulses are slowed, there's preferential uh production of FSH beta, and The reason they're so interesting is that it's one cell type, but it's making two hormones, and they are regulated by their input, which is GNRH and GRH is a summary, I mean, there's a Uh, a set of neurons that control GRH secretion that we'll talk about in a minute. So, testosterone is produced, it causes a decreased sex hormone binding globulin by the liver. Insulin also seems to do the same thing, and then there's insulin resistance at the adipocyte and muscle level, and testosterone gets converted in the granulosis cell to estradiol. But When you look at the feedback, estradiol seems to reduce the pulse amplitude, which isn't the major defect in PCOS although many women have that. Whereas, um, Progesterone decreases the pulse frequency, and so when, when we treat patients with a birth control pill or progestin only, their LH pulses decrease. But the interesting thing to make it more complicated is that testosterone itself blunts the inhibitory effect of progesterone and the GNRH pulse generator. So because these women have, they don't ovulate, so they have low progesterone. And then they have, uh, in addition to that, they have a high testosterone, which blunts further blunts the effect of um progestin on the LH or GNRH pulse generator. Does that make sense? You don't have to answer, but uh I think it makes sense. But the thing is, is that there's a whole set of neurotransmitters, uh, so-called canine neuron that, uh, secretes kiseptin, which is an important regulator of GRH, and uh neuroinin 3 and diorphan, and. This may seem esoteric, but there are now attempts and successful attempts at using modulators of neuroinnin 3 to regulate the GRH pulsatility and have a beneficial effect and uh perhaps ovulation in these women. And that's shown here, so here is the PCOS patient, and I, I don't show too much detail here, but the neuroinin 3 receptor antagonists will reduce the androgen production by the ovary, and these are clinically available and uh these are the studies that summarize their use. This is mainly summarized by Doctor Odilo. Uh, who's in the, where he's from Erasmus, uh, university, and you can see that, um, The from this table that uh KP54 or kisepin increased LH. But uh neuroinin 3 decreases it. So there are other pharmacologic agents in the pipeline that are gonna be useful for this disorder. OK, so which of the following, this is like a little quiz partway through, which of the following are key to making a diagnosis of PCOS? And I'll ask by show of hands, but we'll see if we get a response. So is it important to measure serum LH or LHFS ratio? It's not, um, because as I mentioned, it's pulsatile, and you may be misled. You may have an LH and FSH ratio that's within the normal range, and that doesn't tell you anything. If the LH is very high and FSH is low, that may be useful, but it's not part of the diagnostic criteria in any of the paradigms. What about measuring an insulin or insulins of glucose ratio in these women? You know, I know some people must order it because Every patient I see with PCOS comes in with an insulin level, um, and I, I find that very frustrating because. There's no, it's not a dichotomous variable, it's a continuous variable on insulin, an obese woman with PCOS doesn't have an insulin level that is necessarily higher than an obese woman without PCOS. I'll, I'll contradict myself in a minute, but um, so a random insulin or a fasting insulin or an insulin to glucose ratio is not really very helpful. What about an oral glucose tolerance test? How many people would order an OGCT once they make the diagnosis of PCOS? OK, so I call that an ancillary study. I mean, it's, and the same with the hemoglobin A1C, um, we do an A1C on all our patients and we get a fasting glucose because it's more important to know what their glucose levels are than their insulin levels in my view. A transvaginal ultrasound. That's if you wanna uh diagnose them by the Rotterdam criteria, you would order that. Uh, how about I didn't talk about this, but serum, uh, anti-Mullerian hormone. Now that's becoming a, a very interesting and important marker because it, it tells you how much granulosis cell mass there is. So women with PCOS tend to have very high AMH levels, and women who are postmenopausal, for instance, have almost undetectable AMH. The problem is that there's a lot of overlap between PCOS and normal, but the tests are getting better and there's a hope that. An AMH can either substitute for an ultrasound or be used as a diagnostic criterion, but at this point it is not. Um, prolactin. How many people would order prolactin? Yeah, I think it's important to remember prolactin because that can cause pretty much the same picture as PCOS. So prolactinoma may present with secondary amenorrhea and um they may or may not have androgen excess. And the plasma 17 hydroxyprogesterone. How many people would order that? Yeah, especially in certain subpopulations, and I was told, I think by Maria knew that in New York, 25% of women that come in with Hirstism have CAH due to 21 hydroxo deficiency. I don't know if that's accurate, but that was her estimate. OK, so what about associated comorbidities? Well, it's been known for quite some time that um these are, this is an oral glucose tolerance test, the glucose insulin and C peptide levels, that at any given glucose concentration, the women with PCOS tend to have higher insulin levels, and I just told you that a fasting insulin is not helpful, and you can see that the fasting level is the same at the baseline. Um, C peptide is a marker for insulin. It's a stable marker. It's probably better than insulin, but we don't usually measure it unless we're gonna do deconvolution to try to figure out what their insulin secretion is. But this, this has been shown over and over and over again. And this has really been best shown by Andrea Dunai who looked at women with PCOS who are obese and controls who are obese, and then PCOS who were non-obese and controls who are non-obese. Now where she found non-obese PCOS, I can't tell you because just about uh the mean, we have about 1000 patients in our database, 1200, and the mean BMI is 39. So it goes anywhere from 17, which is probably not PCOS, to 54, so we've, it's quite a broad range and what's interesting is that the obese controls and then uh non-obese, um, sorry, the obese. Controls, that's right, and the non-obese PCOS have almost the same degree of insulin resistance by hyperinsulinemic e glycemic clamp. So there's something about PCOS that confers a greater degree of insulin resistance beyond the obesity. So there is something there, but it's not detected by fasting levels. And I keep using them, OK. Now. We've done a lot of uh IV glucose tolerance tests, and they're done like this. We give a bolus of glucose and it decays over time, and then we measure the insulin and in the 1st 10 minutes of the uh study, we get the acute insulin response to glucose as a marker of beta cell function. So it can be very helpful if you're trying to sort out whether they have an insulin secretory defect or just insulin resistance. And this is what the normal curve looks like, and I'll show you an example of what happens. So let's say somebody had an as by insulin sensitivity and these are the unit lists just for the purposes of showing the concept, but an insulin sensitivity index of 2.0, which is pretty low, and uh acute insulin response to glucose of 400, and when you multiply them and at a constant. Um, the disposition index, it's called, is 800. Now, if a person Is going to maintain their glucose tolerance and they become even more insulin resistant, so their SI is 0.4. They have to secrete, uh, not proportional but an increased amount of insulin in order to remain normal glucose tolerant. So that curve is for individuals with normal glucose tolerance. And so the DI turns out to be uh I think I put that in the wrong place, it's 800, not 2000. Um, and we found that, interestingly, this was a number of years ago, but we found that, um, this is Steve Kahn's, um, paradigm for the 50th, 95th, uh, 75th, 25th, and 5th percentile of, I think 93 individuals, men and women, who had an IVGCT and so they would follow along their curve. And we found that if you have a woman with PCOS who has a family history of type 2 diabetes, she's much more likely to, does this show up in her? Yeah, she's much more likely than a woman without PCOS family history, first degree relative with type 2 diabetes to have impaired beta cell function. So in the days when I was training, everyone said, oh, it's all insulin resistance. Well, it isn't. It's, it's diabetes doesn't develop just from insulin resistance, we know that. So there's a defect in beta cell function in this population as well. And we were able to demonstrate that. Now, I'm gonna, it sounds like I'm gonna give it a new talk, but what does sleep apnea have to do with PCOS and why is it important? Well, it's known in population studies that men have a higher prevalence of OSA about 2 to 3-fold higher, and it's thought that the testosterone is the culprit hormone. I'm not sure it is, and I'll show you why. Um, it influences both neural controlled breathing and upper airway mechanics, so, um, the muscle relaxes and the tongue can fall back and you get obstruction. Um, it's been hypothesized that this gender difference is explained by androgens. And, um, first of all, sleep apnea is incredibly prevalent in women with PCOS. In our studies, 50% of women with PCOS, mostly obese, but 50% of the women had obstructive sleep apnea, and we know that obstructive sleep apnea can worsen glucose tolerance, so I'll show you that in just a minute. So, um. In this one systematic review, there was uh 35% of women with PCOS had polysomnographic evidence of sleep apnea, and so all these women had sleep studies. So if you ask a patient if they're fatigued, of course they're gonna say yes. Do they get tired during the day? Yes. Do they sometimes take a nap? Yes. So you can't tell from that. If you ask them if they snore, usually they tell you their their husband snores and wakes her up. So you, you can't, you, you know, you can't get a good history. So you have to do a PSG, um, now it is true that the obese women with PCOS are more are more likely to have obstructive sleep apnea than are the lean women, but it's still very high. And this is the this is the uh plot of risk of OSA and PCOS and it's all to the right of unity of zero rather, so. Um, from a lot of different studies, and even in adolescence, the, uh, the prevalence of OSA is very high. So we wanted to look at whether um there was a relationship between glucose tolerance and uh OSA and this shows the prevalence of impaired glucose tolerance or prediabetes in relation to the severity of OSA and the controls had about 5% risk of OSA which is common in the literature. PCOS women without OSA were about 25%. But as the uh OSA became more severe and in this small cohort, there was only 3 who had really severe OSA. Their prevalence of IGT was 100%. So it's kind of a linear increase in risk of uh diabetes or pre-diabetes. And this is the thing that threw me, or threw us. Um. We predicted that the free testosterone would be higher in those women with severe sleep apnea compared to those without or mild, and there was absolutely no difference in the testosterone. So my supposition is that it's progesterone, because progesterone has an important impact on sleep uh dynamics and it can protect against sleep apnea. When we tried to look at that, um, so many of the women had values that were at the lower threshold at the detection limit of the essay, about half, so we couldn't really do a fair. A statistical analysis because they weren't, they're all clustered, so there's a way to deal with it, but we didn't. It's let's say, um. So, if you, this is our cohort of, I think, 200 and something, 175 women, and it looks at controls with OSA controls with OSA, PCOS without OSA and PCOS with OSA. And impaired fasting glucose, which isn't IGT but it's prediabetes in a sense, uh, fully 34% of the women who had sleep apnea and PCOS had impaired fasting glucose. And Uh, IGT, which is based on the two-hour glucose, 41.9% with PCOS and OSA had impaired glucose tolerance. So what that makes you think about is when you hear that uh glucose tolerance is abnormal in PCOS and a lot of them are pre-diabetic, it could be that they have concomitant sleep apnea. Now, are there studies that show that treating the sleep apnea improves uh glucose tolerance? Not really. Uh, we did a study, we found some improvement in insulin sensitivity, but it was marginal. And the reason it was marginal is that once you get above or near a BMI of 30, you sort there's an asymptotic relationship between insulin sensitivity and um. And insulin resistance. So I don't know if I should, well, I could go back and show you this slide. Yeah, here. So you can see these women, uh, down here on the flat portion of the curve, it's sort of asymptotic, and um so they're obese and their insulin sensitivity is very low. If you want to change their insulin sensitivity, you have to have a major shift, I mean, to move them along this curve. And I think now we can do it, um, and I'll come to that in a minute. Deja vu all over again. OK. So this I've already told you, um, and this I told you as well in a certain way, IGT in relation to severity of OSA, um. Insulin response is impacted both by PCOSO and OSA and you can see here that the PCOS OSA positive in blue has the lowest DI disposition index. Yeah or Matsuda index, which is a mark of insulin sensitivity. And so both the body weight and the degree of sleep apnea and PCOS all influence that outcome. So to summarize this part of the talk, in young obese women with PCOS treating OSA with CPAP seems to result in improved measures of cardio metabolic function. We did find. A statistically significant but not clinically relevant improvement in insulin sensitivity, but there was daytime and nighttime sympathetic activity reduction as measured by heart rate variability and daytime blood pressures in those treated. Mm. And there's a strong dose response relationship between the number of hours of CPAP use. I mean, this makes sense. It's like, you know, if you don't take your insulin, your glucoses are gonna be high. If you don't wear your CPAP, your AHI your apnea apnea index won't come down. So, but the more they would use the CPAP, the, the better their uh. Their glucose tolerance and it may offer, I mean, we, we don't know this, but it may offer some uh protective effects from a cardiovascular standpoint. Certainly sleep apnea causes hypoxemia that leads to arrhythmias and sudden death, um, but whether it's going to improve glucose tolerance is not known. In most studies, epidemiologically between 70 and 90% of type 2 diabetics type patients with type 2 diabetes have sleep apnea, which is astounding. I mean, I was shocked that it was that high, but it is in study after study. So I think we should be asking the patients about that. So now, what are the newer kids on the block, the newer pharmacologic agents? Well, for, you know, I'm, I'm, I've come to believe that um a lot of what we call PCOS is just severe obesity because a lot of the cardio metabolic defects can be reversed and in fact normalized by weight loss and either through um surgery or medical therapy, which I'll come to. So, reproductive um ovulatory dysfunction improves. Fertility, oops, fertility improves. Um, there's an increased risk of miscarriage, decrease risk of uh gestational diabetes with weight loss, lower risk of hypertension and decreased risk of preeclampsia. Metabolically they have improved insulin sensitivity. I'm not even gonna talk about Nash or NAFLD or MAFLD whatever term you want to use, um, but these women are at least twice as likely, up to 4 times as likely to have fatty liver disease and Nash than similarly obese women without BCOS. And um treatment uh can lower risk of miscarriage, GDM, lower blood pressure, decrease the risk of preeclampsia, lower testosterone. I mean, weight loss does almost any everything you want it to do, but Sending every patient with obesity for bypass surgery is not feasible. So the standard therapies have been lifestyle and combination of birth control pill, metformin, progestins, anti-androgens, Clomid and gonadotropins, and the novel therapies that were thiazolidine diones dechiroinoetide put here only for completeness because it doesn't work, um. Hissepin neuroinin diorphin agonists or antagonist, uh GLP one receptor agonist is what I want to talk about and uh GIP and then there's even coming out now triple agonist, so GLP one, GIP and glucagon um agonist, rereturite um is the brand name or the Uh, chemical name and bariatric surgery and then CPAP, so and aromatase inhibitors, so the GLP1 receptor agonists have been studied in PCOS populations, uh, not just obesity, but these are obese women with PCOS, and we know that GOP one is an incritin hormone produced by the L cells in the GI tract, but that's not the only place that produces it. It's also found in the brain stem and the hypothalamus adipose tissue and their GOP one receptor receptors in various tissues, the same tissues receptors are found in the brain, the GI tract and the adipose tissue. It acts centrally to reduce hunger and increase satiety, and that's thought to be mostly due to GIP because um as you probably are aware, um. Terzepatide is probably more effective than simaglotide in terms of weight loss. Um, they're both very good, but uh trazepatide seems to be um more potent. They can slow gastric mortility, and that can be a problem, especially if they're diabetic, and it increases insulin release from the pancreatic beta cell. So, um, here's a, a diagram showing the GLP one being released from the L cells. It, it takes a humeral path into the CNS and a uh vagal path, neural pathway into the um. Nucleus tractus solitarius, um, and that has central effects on, on satiety and hunger. So this was a study, I think everyone's seen this, um this is with the regulatide, and this was published in I can't believe it's been that long. 2015, that's nine years ago. I, I thought it was more recent, um, but you can see what happens to weight change in body weight as either a percentage or total body weight. What? OK. I'm getting the hook. Um, and then The, here's a table that I'll skip over because of time. Um, And here's a paper that came out last year, uh, let's see, it was. Well, it's called last year, looking at this triple receptor agonist, and when you think about it, it's counterintuitive that glucagon receptor agonists would cause weight loss or improve glucose control, but it does for various reasons. Um, the person that has shown us the best is Matthias Chop and uh he, he gave a great talk at the diabetes meetings. So, We have things in our armamentarium. Now, the problem is, at least in our clinic, we can't get a pre-approval for use of either Ozempic or uh trazepatide in non-diabetic patients, and we have to get, go through hoops to get the approval for weight loss. So we can do it, but it takes a lot of work. And here's the uh effect of once weekly trazepaide as compared with placebo on body weight, and you can see the dramatic loss in weight uh dose dependent. And same here, same thing. So I think I'm gonna end there because I only have 2 minutes left and um take any questions that you might have. Thank you. Um, I just want to know if you could expand a little bit on the, uh, if there's any data on. I'm sorry. The data are terrible, um. Is is that what you're asking? Yeah, Inool is, has been discredited. Um, in the chiroinnotal, they don't really work. I mean, they make an individual, I can't say for every patient, but in general, they, they don't work. Uh it doesn't work as an insulin sensitizer. Any other questions? Oh, so there's a question online. Um, does progesterone interfere with androgens? Measurement or yeah, because it it does because it will feed back at the level of the pituitary to decrease FSH secretion, so the estrogen to the testosterone to estrogen ratio, estrogen to T ratio will be affected. So we don't, I don't measure androgen levels if somebody's on the birth control pill or progesterone, we take them off if they're willing for 3 months, which is frustrating to the patient because they want a quick answer, but on birth control pills, you can't get a quick answer. Thank you very much. Published March 22, 2024 Created by Related Presenters David Ehrmann, MD
CME Knowledge Hub
