Polymyalgia Rheumatica: What has changed since 1957?

I'm going to be speaking about, um, sort of updates in polymyalgia rheumatica, um, and I titled this talk, uh, "So what has changed since 1957?" So why did I invoke the year 1957? So that was the year the first, um, use of the term polymyalgia rheumatica is seen in the medical literature. Now, the disease was probably first described back in the late, uh, 19th century and certainly in the 1930s or 1940s. You could see authors struggling with this population of patients that sort of looks like RA but had a little bit of a different phenotype and different characteristics. But the first actual use of the term polymyalgia rheumatica was in this case series by Dr. Stuart Barber from the UK.

And what's really interesting is that in his series—this is just 12 patients—he really identified the sort of canonical characteristics of PMR that we still, um, sort of reference today. So number one is that it's really a disease of the elderly; so the average age is about, um, and so it's almost never seen in patients, uh, below 50. One is that unlike rheumatoid arthritis, right, which typically affects the, um, distal joints like, like the wrists, the MCPs, this really has a trophism for the kind of muscles surrounding the shoulder and the pelvic girdle, um, and you know, these patients are often kind of extremely stiff and sort of extremely painful. So I, the way I, I tell kind of fellows is that these patients really fall off a cliff. So like last week they're doing their hiking, they're, you know, playing pickleball or whatever, and then now really they can't get up the stairs. So I specifically remember there's a patient of mine who's actually a personal trainer, so he's very buff, and then he had to actually, um, start sleeping downstairs because he couldn't actually go up and down the stairs. So it's a really dramatic kind of functional decline that you see in these patients, and this is the kind of visit where, you know, the kind of adult children will come with the patient because they're so worried about what they've seen, um, and how they're doing.

And so I like this quote from Dr. Barber: He said, "The somewhat melodramatic description of their pain by these patients tends to suggest a diagnosis of psychoneurosis until the ESR hasn't been measured." But like, you know, when you see kind of the faces of these patients, I mean, they just look like they've been hit by a truck, right? Because again, two weeks ago they're totally normal, they're doing their thing, and now they really, you know, they just cannot get out of bed in the morning. Um, as Dr. Barber mentions, right, these patients almost universally have elevated inflammatory markers. So maybe about 10% of patients can have an ESR less than 40, but virtually every patient will have an elevated CRP, and certainly if both the ESR and CRP are, are normal, you really have to really, um, think, think about other diagnoses. Unlike rheumatoid arthritis, polymyalgia rheumatica is not thought to be a destructive arthritis, right? So you do not see the, um, you know, uh, joint deformities, right, that you, that you can see with rheumatoid arthritis, and x-rays in these patients are typically normal.

Um, now Dr. Barber did not have access to a lot of antibody tests, but he did note that the so-called lupus erythematosus cell was negative, which is sort of the early kind of ANA test that people used to use, and still to this day we have not developed any autoantibodies, right, that lead to the diagnosis of, of polymyalgia rheumatica. And even though the name myalgia, right, suggests that this is a disease of the muscles, the strength on these patients, if you're able to test them, even though they, they will feel weak, but the actual strength on testing is typically normal. They will have normal CPKs and muscle biopsies, which you don't do anymore, but they did when this disease was first being diagnosed, including in this case series, are typically normal. And you know, PMR is sort of characterized by a very good response to relatively low-dose steroids, which Dr. Barber did note, and he said, "The greater availability of the cortical steroids made it possible to try their effect, and there was no, out of the immediate response to these preparations." And so this is an funny quote, and he said this is from, from this paper in 1957, he said, "Until more is known about it, I suggest the term polymyalgia rheumatica," and he prescribed bed rest and phenazone, which is sort of like an old NSAID, all right. Now, the term, that term, he was wrong about one thing, and that term is still used, right? So this is, this is what we call this sort of syndrome or disease to this day.

But I really want to, um, identify what I'm trying to give you as an update on what things have changed since then, and we're going to talk about three different things: one is how advanced imaging has helped us figure out kind of where the lesion is in polymyalgia rheumatica; one is that we've, are, are, are increasing knowledge about the relationship with this disease with giant cell arteritis; and then certainly in addition to steroids and phenazone, right, we have new treatment options that I'm going to talk about a little more. So whenever I give, you know, one of these Grand Rounds, and I always tell, tell the fellows this, I preach them, I say, you know, if you're going to give a talk, right, the question is why should the audience care, right? You're taking up some amount of their time, um, what's, you know, what's the point? Why is this all, I think, interesting to a general medicine, uh, population? And there's sort of, I think, two reasons. So number one is that polymyalgia rheumatica is a common disease; so it's probably, it's thought to be the second most common autoimmune rheumatic disease in the elderly besides elderly-onset rheumatoid arthritis. Um, some prevalence estimates that almost 2% of the population will have PMR, so even certainly if you go into general medicine, you're going to see this, but you will encounter this in any field of medicine that you go into.

Now, interestingly, there's this kind of, uh, nor, at least in Europe, there's sort of a North-South European, uh, prevalence gradient, so the prevalence is much higher in northern European countries than in southern European countries, and there's very low prevalence in Turkey and Korea, and then in Colia, which I don't think is listed on this map. Now, the US prevalence is thought to be sort of in between those northern, northern and southern Europe, um, with the caveat, right, is that all our knowledge about the prevalence of PMR in the, in the US comes from the Mayo Clinic and their Rochester epidemiologic project, right? So which is a relatively small area, right, um, and certainly not, you know, doesn't reflect the sort of, you know, large diversity, geographic, you know, racial, ethnic diversity of America, so you don't really know exactly what the prevalence is of PMR, um, in this country, but in general is thought to be, um, very common. And the second reason, right, and this is I think the more important reason that it's really important for any of all of us to, to know about is that this is a fixable problem, right? It's a can be a devastating disease in terms of, again, the effect on patients' functional status, but it is something that's very fixable and intervenable, right? So I'm always struck in medicine, right, there's so many things that we can't fix, right, right, even today, right? I cannot, we can't do anything about chronic fatigue, you know, patients with chronic back pain, there's certainly management strategies, but there's no easy fix to those problems. PMR in contrast is very much a fixable problem. So this is something I hope that again, even if you don't go into Rheumatology or primary care, that you at least recognize because if you think someone has this, this is certainly a message in MedConnect to your rheumatologist, um, and just as an example of this, in Denmark, in this sort of hospital system, they opened up a FastTrack clinic, and after opening up that FastTrack clinic, they were able to decrease the rate of hospitalization for PMR by, you know, six times, and so illustrating the fact that again, this is important to recognize because you can intervene and improve outcomes. All right.

So the first sort of like, you know, um, theme or sort of, um, uh, theme that I'm going to talk about today is, um, where the inflammation is in polymyalgia rheumatica. Right, as I mentioned before, it's not thought to be in the muscles, right? Again, the CPK is normal, muscle biopsy is normal, and it's probably not in the joints themselves, right? When you do a biopsy of someone of a joint in rheumatoid arthritis, it really looks like a cancer; it's like this exuberant inflammatory response, um, joint, uh, synovial biopsies in polymyalgia rheumatica do have some inflammation, but it's certainly many degrees less than in rheumatoid arthritis. And sort of the use of ultrasound and PET scans and MRIs has really helped us kind of illustrate where the problem is. Now, polymyalgia rheumatica, in about 20 years ago, was identified that most of the inflammation is thought to be extracapsular, right? And what does that mean? Can you guys see the, yeah? So, um, this here, I'm colorblind, so I actually can't tell if this is red or green, but this is a, this is the joint space in here, and then surrounding it is the capsule. So again, rheumatoid arthritis has a strong predilection affecting within this structure, well, as polymyalgia rheumatica really, when you do studies, is the inflammation is really kind of in the, in the periarticular areas or extracapsular areas around the joints, and this was seen kind of elegantly first on an MRI study, um, about 20 years ago now. But if you look at rheumatoid arthritis, again, all this sort of swelling is inside the joint, and there's a little bit of a subacromial bursitis. In contrast with polymyalgia rheumatica, it's sort of this kind of fluffy inflammation that doesn't seem to localize exactly to one specific area; it's a lot in sort of the tendons, kind of the, bit, sort of the myotendinous junction here, um, these are all kind of rotator cuff, um, tendons; this is infraspinatus back here, supraspinatus up here, but it's sort of this kind of nonspecific, fluffy inflammation, but it's really primarily kind of around the joint. You can see here in the hip, right? Again, the hip joint is here, and that actually looks pretty normal; however, right, you see pericapsulare enhancement here and then enhancement around the tendon, the tendons, uh, over there. And this is very, again, very, very striking when you look at sort of, um, hip MRIs. You'll hear this term a lot in polymyalgia rheumatica called myofascial inflammation, which is a little bit confusing because typically when we use the term myofascial, it sort of means, um, essentially like a central pain sensitization, like a fibromyalgia component, but here when they're talking about myofascia, they're literally talking about inflammation of the fascia and surrounding connected tissue around the sort of musculotendinous junction. And again, this is really notable, mostly, I think this has the most potential in looking at the hip muscles in the pelvis. So PMR seems to have a very striking predilection to affect particular hip and tendon, uh, sorry, tendon and muscle insertions around the hip. So one here is the rectus femoris; so this is one of your, um, hip flexors, right? You can see it inserts here onto the anterior inferior iliac spine. So this kind of inflammation around kind of the, the, the tendon insertion there, and then number two is the adductor longus; so here it inserts onto the, um, pubic symphysis, right, to adduct the hip, and you can see this kind of inflammation here. And this is sort of unique in terms of different inflammatory diseases, sort of like in the shoulder you often will get this kind of nonspecific inflammation, and again, I can't tell you this has been, there have been a number of times where I've gotten MRIs of this and it would be sort of read as like, you know, muscle strain or something like that, but it really, this sort of bilateral, um, enhancing nature of it should really clue you in that this is polymyalgia rheumatica. This is, we're looking here from the posterior aspect; you can see the inflammation of the gluteus medius tendon here and also of the hamstring insertions here on the ischial tuberosity. And then again, you know, we have all these sort of like these kind of six very small hip extensor muscles or lateral rotators kind of in the back of our, the back of our thigh, like the obturator externus, and again, notice that the inflammation is kind of in the fascia surrounding the muscle as opposed to in the muscle itself. And again, in all these situations, the actual joint inflammation is relatively minor, and so it's really thought, again, that that the inflammation in polymyalgia rheumatica seems to start at the musculotendinous junction and in the connective tissues around that junction and then can subsequently spread into the surrounding bursa or the joint. So again, it does not seem to start in the muscle like in polymyositis; it does not seem to start in the joint itself like in rheumatoid arthritis; it's in these sort of like periarticular connective tissues where the inflammation starts, and that seems to fit with our kind of, um, the kind of clinical presentation where they have a hard time exactly localizing kind of where the pain is; they just know that they're in kind of big, big trouble.

Now, on ultrasound, there are a couple of things that you might see on ultrasound. So one is a subacromial bursitis; so I don't know how well this projects, but this is the supraspinatus tendon, this is the deltoid, and this swelling here is the subacromial bursa. The other thing that you can see on ultrasound is a biceps tenosynovitis; so this is inflammation in the tendon sheath; so here's the biceps, and this surrounding kind of halo of inflammation, um, that you can see both transversally and longitudinally is inflammation, but as I said before, I think these are probably secondary inflammatory changes as opposed to the primary site of inflammation. And so in the, in the ACR, this is the American College, College of Rheumatology classification criteria, you can actually use ultrasound to, um, as part of the criteria, but note that ultrasound doesn't really increase your sensitivity and specificity that much. And again, I think the problem is, number one is ultrasound, like I love ultrasound, as Dr. Ste said, I do a lot of ultrasound; it's, um, you know, can be amazingly helpful in clinic, but there's a lot of operator dependence, sort of how much jelly you put on, kind of how you hold the probe, um, are the lights out, you know, is the patient, you know, is there a lot of fatty tissue between, you know, the probe and, and what you want to look at, um, but also, you know, the problem, again, again, I think that the problem is we're looking for with ultrasound for sort of these like secondary and tertiary changes as opposed to the primary inflammation itself.

Now, there's been a lot of investigation about PET scans, and, um, PET scans do seem to have a relatively high sensitivity and specificity for the diagnosis of polymyalgia rheumatica, and you can see around the sort of shoulder, the hip girdle, and the ischial tuberosities here, that's that hamstring inflammation that I was talking about on MRIs. What's notable about PET scans is that you can identify, they, they, they have sort of, um, found kind of interesting new sites of inflammation that you can see, such as the intraspinous bursa; so these are these, these bursa that are in between the vertebrae, and that's probably an explanation for a lot of like the neck and lower back pain that these patients get, um, with this disease. Now, PET has a major role in identifying concurrent large vessel vasculitis, which I'll talk about a little bit later, and I think it's generally probably too expensive and too much radiation for routine diagnosis for polymyalgia rheumatica, but using PET scans has helped, helped us identify a couple of these sort of interesting new locations that we weren't really thinking about before. Again, what, what I think probably has the most potential for the diagnosis of polymyalgia rheumatica is MRI with contrast, specifically of the pelvic girdle, um, because it's that's sort of a, there, there's not like, it's very, um, uh, there's not a lot of other places that cause inflammation sort of in that way, and you can see from this study—this is a retrospective study, so you have to take this with a grain of salt—but you can see that again, at that rectus femoris insertion and at that, um, adductor longus insertion, almost every patient with PMR will have that, where it's very, very uncommon in controls, and these are controls who had MRIs done for pelvic pain. Now, again, this is a retrospective study; a lot of steps need to be done before this is ready for prime time, you know, prospective, multiple centers, randomized, blinded, etc., but I think doing MRIs of the pelvis for these patients has a lot of potential in terms of aiding in the diagnosis because right now the diagnosis is purely clinical. I mean, all, all diagnoses in Rheumatology are clinical, but you know, right now it's, it's, we don't really have any, um, imaging, I think, is, is routinely used for diagnosis, and I think an MRI of the pelvis really has potential for that moving forward.

So now I want to speak about the relationship to giant cell arteritis, and this is actually the, that actually relationship has not actually been, is, is not actually new, but a lot of our sort of more knowledge of the relationship is new, but, um, in 1964 this paper came out with the sort of, um, uh, passive-aggressive title of "Arteritis and 'Polymyalgia Rheumatica'." So this was a, um, these were, this was a hospital in Sweden called Vjo, I'm not sure if I'm pronouncing that correctly, um, it does look very nice on Google Maps, but these, they took, um, 23 consecutive patients who are admitted for polymyalgia rheumatica; 14 of those patients had no signs of giant cell arteritis; 12 of those patients had a biopsy of a temporal artery. So these patients had no cranial symptoms; they biopsied the temporal artery, and half of those patients end up having features of giant cell arteritis. And so these authors concluded that, no, um, polymyalgia rheumatica is not a form of rheumatoid arthritis; this is actually, quote, "temporal arteritis in one of its several guises." And so they actually proposed the term polymyalgia arteritica, which, you know, has not stuck, but I think it's an interesting term. We know that about, at least clinically, about half of patients with giant cell arteritis will have polymyalgia rheumatica, and in contrast, about 10% of patients with polymyalgia rheumatica will have giant cell arteritis, either at diagnosis or moving forward. So when I see patients with polymyalgia rheumatica in my clinic, every time I'm always screening for symptoms of giant cell arteritis. And so this has led some authors to propose the, quote, "one disease hypothesis," so people will argue that these diseases are essentially two parts of the same spectrum, and they will call it GCA-PMR spectrum disease, or GPSD. So what is the evidence that they, that, that they use for this? So number one is that again, so very interestingly, if you take a patient who comes in your office, they have classic PMR, they respond to steroids, relatively low-dose steroids, about 25% of those patients will have evidence of a large vessel vasculitis that is subclinical on some modality, right? So either via temporal artery biopsy or an ultrasound, and in this case, I'm not talking about ultrasounding, you know, their, their shoulder; I'm talking about ultrasounding their temporal arteries or their axillary artery or the subclavian artery or on PET scan. So about 25% of these patients will have, quote, "subclinical giant cell arteritis," and sort of as talking about a second one of the mysteries and sort of one of the future, um, steps is going to be figure, trying to figure out, are these patients different in some way than the 75% of patients that have more sort of more, quote, "pure PMR"? Another line of evidence that they use to kind of connect the two diseases is one of, is the epidemiologic overlap; so, um, that sort of North-South European gradient is also seen for, um, giant cell arteritis in addition to PMR, and obviously both are seen in the elderly. So again, both have a predilection for 75, um, and both are again exclusively seen in patients above 50, and there aren't a lot of like shared immunologic pathways. So they both are diseases essentially of myeloid expansion of the myeloid lineage, and both are associated with increased production of a particular cytokine called interleukin-6, which is come up later when we talk about treatment. And there even some, even more direct connections; so in some mouse models, if you have someone, if you take a mouse, PMR mouse, and you add an artery, you transfect an artery of some, a mouse with GCA, the inflammation in the, the T-cell inflammation will migrate into the temporal artery of the patient with PMR, not with controls. So it's thought that perhaps the temporal artery of the mouse is somehow sort of activated and just awaiting that second hit to get the vascular inflammation that's characteristic of giant cell arteritis.

So if you were going to do an ultrasound looking for a concurrent vasculitis or vascular inflammation with PMR, what you're looking for is something called the halo sign. So the top panel here, this is normal; the bottom panel, you see this hypoechoic, minimally compressible surrounding halo inflammation around the artery, and this is thought to be thickening of the artery wall. So if you're going to do an ultrasound to look for a large vessel vasculitis, you will look for this sign either in the temporal artery, the subclavian artery, or the axillary artery. And what's really notable, and this is a little bit confusing, so just stay with me for a second, but if you take patients with again, subclinical PMR with evidence of vasculitis on ultrasound and you compare those patients directly with patients with pure vasculitis, pure giant cell arteritis, they look a little bit different, but there's a lot of crossover just in terms of the distribution of the vessels involved, again suggesting that there's a lot of overlap kind of between those two diseases. Excuse me. Similarly, when you look at PET scans, right? So if you do have someone with sort of pure PMR, again, you can see the inflammation around the shoulders, around the hips, in the intraspinous bursa here, but a lot of these patients will have, again, 25% of these patients will have concurrent vascular inflammation, too. And so the question, as, as that, that's going to be moved forward is, do these patients act differently in some, some way? So are these patients who again have PMR but also have evidence of vasculitis, are they, um, treated differently in some way? Are they more refractory to therapy? Do they need a higher dose of steroids to control their inflammation? Um, should they be given more steroid-sparing agents, right, up front? And then again, sort of really importantly, is are they at risk for some of the terrible complications of giant cell arteritis that, um, that we worry about, notably visual loss, aneurysm formation, so, and as well as sort of large artery stenosis? And that again is sort of a, a, a sort of really important question, again, whether that 25% of patients act more like giant cell arteritis or they act more like polymyalgia rheumatica, and that's sort of an active area of research kind of as we, as we speak.

And then finally, I'm going to talk a little bit about updates in treatment for, um, this disease. Now, polymyalgia rheumatica is typically thought to have a very, very impressive response to relatively low doses of steroids. So this is a very famous picture in the Rheumatology world, but this is a guy who, um, is diagnosed, this is at diagnosis, and you can see he cannot abduct the shoulders past 90 degrees, and that's probably the most common physical sign that we'll see if polymyalgia rheumatica. In contrast, in panel B, this is after a week of cortical steroids, and he's able to fully abduct the shoulders. So steroids, um, are typically thought to have a very, a very, very impressive response, again, in relatively low doses of steroids in patients with polymyalgia rheumatica. So in the 2015 recommendations for the management of PMR, and this is, uh, through the American College of Rheumatology as well as the, um, the European League Against Rheumatism, which I always kind of tickles me about why they called it that, but, um, they recommend starting, um, at 12.5 to 25 milligrams a day; you taper to 10 milligrams after two months, and then you decrease by 1 milligram per day every four weeks. So that's a long taper, right? So if you're diagnosed January 1st, March 1st you hit 10, right, and then April you're at nine, and May you're at eight, and June you're at seven, right? So it's a very, very long taper of steroids, and that is obviously problematic, one for side effects, but two is just the, the, um...

The efficacy, right? So this is, um, a particular observational study, and I want you guys to focus on the complete response to treatment here. And you can see at week four, about 70% of patients have a complete response. Now, some of us in Rheumatology might say, "Oh, that actually seems a little bit low," right? Because most of these patients seem to have a, you know, complete response very quickly to steroid; almost everybody. Um, but probably some of that remaining 30% is, um, a lot of mechanical disease, right? So, you know, they have concurrent rotatory cuff disease or concurrent AC joint arthritis, for example. But, um, so but again, I mean, the initial response is not perfect. And then, as you taper the steroids, right, fewer and fewer people are in complete response over time. And so, in general, right, we find that about half of patients just on steroids will relapse about at a year, and to require steroids long-term is very, very common. So a systematic review found that three-quarters of patients are still on steroids at a year; 50% are still on at two years; and then 25% are still on steroids at five years. And you know, we know, right, that the side effects of steroids are very cumulative, right? The higher the dose, the longer you take it, sort of the more side effects. And this is the, you know, list of side effects that you can get with steroids.

Now, in Rheumatology, there's been this, um, development of, uh, the glucocorticoid toxicity index, which is an index to sort of, um, uh, identify, um, toxicity of various regimens and and how they affect, um, how the steroid burden affects patients. But I would note that a lot of steroid side effects that bother the patients too much are actually really hard to capture in a tool like that, right? So things like, you know, obviously weight gain, but stria formation, cushioned features—a lot of patients, I don't know if you guys have experienced this, but a lot of patients on steroids describe just the the the the easy friability of their skin as being a really, really problematic, um, and so a lot of that. But again, a lot of that is going to be very hard to capture just in a tool. Um, but again, we do know that, um, it's it's really the toxicity of steroids are sort of very cumulative. But you might say, "All right, well, a patient's on steroids for five years, but let's say they're on, you know, they alternate between three and four milligrams for five years, right? So what's the big deal?" Well, I think the side effects of even low-dose steroids are sort of starting to become, uh, when used chronically are starting to become more and more known. So, um, I think a lot of us know that even low-dose steroids can increase your risk of fracture, specifically vertebral fractures. So again, these are patients on five milligrams a day or less long-term that doubles your risk of a vertebral fracture. And we'll say that's—I don't think that's like—that's like a dramatic news, but what what what really has really struck me is that, um, even low-dose steroids do seem to increase your risk of serious infections. So this is a, um, observational study looking at two different databases: one is a Medicare, one is a commercial database. But you can see here that the patients on five milligrams a day or less, right, still had a significantly increased risk of serious infections. Because I think when I was in training, right, the mantra was you you quote-unquote "make five milligrams of prednisone a day." So if you're on prednisone a day, five or less, it's still like quote-unquote "physiologic." But again, data really suggests that, uh, you know, it's, you know, it still has potential toxicities, right? So you might see here, well, the number needed to harm you go from 8.6 to 11%; the number needed to harm is, um, I don't know, about 45 or 40, you know, it's not terrible, but you know, if you're that one patient, that's a big—that's a big problem. Um, and sort of similar observational studies have found the same thing with cardiovascular, right? So even low-dose steroids here can increase your risk of AIB events as well as CHF events. And I definitely have had patients who they have—have—I've put into AIB on steroids at the doses used for for polymyalgia rheumatica, unfortunately.

And I think one thing that we really don't appreciate in Rheumatology—I don't know if any of the endocrinologists are here—but a surprisingly large percentage of patients on low-dose steroid will have evidence of adrenal insufficiency as well. And so this is, uh, this is from the UK, the SST; this is like their ACTH stim test. And this is an observational or this is a cross-sectional study of patients in Rheumatology on long-term glucocorticoids. And of the patients on five milligrams or less, you know, I think this is about 25 to 30% of patients still failed their ACTH stim test. Now, this is observational; they weren't doing it for all comers here, so they were doing this test for a reason, so I can't say what the sort of overall rate is, but it's still suggestive, right, that that that there's still problems in your HPA axis even using these low-dose doses of steroids, which are polymyalgia rheumatica patients really, um, see. And again, as you might expect, this has been studied specifically in polymyalgia rheumatica, and those patients have an increased risk of diabetes, a doubling of their diabetes risk, and an increased risk of fractures, especially vertebral fractures, right? So steroids seem to have a predilection for vertebral fractures, um, and five-time increase in women, right? So again, this is not—this is not a—not an insignificant problem. And so, because of that, right, there's been a lot of interest in the use of, um, steroid-sparing therapies for PMR.

Now, I think it's important to kind of take a step back here and think about what's the rationale for using a steroid-sparing therapy, right? So number one, right, is obviously you want to decrease the steroid burden to decrease the risk of side effects. And so I've—that's what I've been talking about for the last few slides. Again, these patients do have a long-term steroid exposure often, so yes, they—a steroid-sparing therapy, um, I think is is important just for that particular reason. The other two reasons that there's actually less of an impetus for PMR than you might think, right? Because another reason you might want to make sure someone is on a steroid-sparing therapy is to decrease your risk of flares. And obviously that depends a lot on the consequences of a flare. So if you have someone with PMR and they have a flare, it is—yes, it is distressing to the patient, but it's usually fixed relatively simply with a relatively small increase in steroid. So you have someone on fire, they flare, you bump them up to 10, they're kind of back to normal, and then you can kind of taper down, okay? But again, you don't have sort of a lot of, um, morbidity associated with the flare. In contrast, right, in something for like giant cell arteritis, right, a flare can obviously be quite catastrophic, right? I'm showing you here anterior ischemic optic neuropathy, which, um, you know, will cause irreversible vision loss. So, you know, based on that, right, the threshold for steroid-sparing agents for polymyalgia rheumatica is actually a little bit higher than you might think—giant cell arteritis. And the third reason you might use a steroid-sparing therapy is to prevent sort of joint destruction like in rheumatoid arthritis, as you can—as you can see here. So in rheumatoid arthritis, right, we know that disease-modifying agents, when used early, will prevent long-term consequences. That's why in rheumatoid arthritis we think about this window of opportunity, right, where you want to get patients on methotrexate early because we know if you control their disease early, even down the years—five years down the road—they have less radiographic progression. But as I mentioned before, excuse me, polymyalgia rheumatica is not a joint-destructive disease. So for that reason, right, again, the threshold for using a steroid-sparing therapy is going to be a little bit higher for PMR than for, um, for rheumatoid arthritis as an example. So now, again, you can always walk into the rheumatology clinic and say like, "Oh, that patient should be on methotrexate or plaquenil," and we'll be like, "Good—good thinking," right? It's sort of like, um, it's like, you know, medicine grand rounds where you say, "Oh, it could be syphilis," right? "Oh, like good—good thinking." So methotrexate really has been, um, um, unfortunately, the evidence basis for methotrexate in the—in as a steroid-sparing agent for polymyalgia rheumatica is far less than for rheumatoid arthritis.

However, um, the largest trial and most well-done trial for, um, PMR and methotrexate did show a significant benefit. And this was actually using a dose of methotrexate of only 10 milligrams per week, right? Where typically for RA we go up to 25 milligrams per week. And you can see that there was at—the trial was actually out to 72 weeks, and there was a basically halving of the rate of relapses. And as you can see, a lot less steroid use in the methotrexate group. But note that it took a long time to show that, right? You had to be on—it was only about 24 weeks, so six months before you saw that improvement with methotrexate. And so in the American College of Rheumatology guidelines, um, methotrexate is, um, suggested, but the language is like extremely wishy-washy, right? The panel conditionally recommends considering early introduction of methotrexate. It's like when I tell our fellows like, "I cannot stand when they write 'consider' this. I'm like, 'Either recommend it or don't,' right? Don't—can we all consider whatever we want, right? We either—you—you either should do it or don't." So the panel conditionally recommends considering—I don't even know—I—I have no idea what that means. But certainly the evidence basis for methotrexate in PMR is certainly not the same as it is for rheumatoid arthritis. Now, TNF inhibitors have been studied in polymyalgia rheumatica and have—have been found to be ineffective. So both infliximab (Remicade) as well as etanercept have both been studied, and both showed absolutely no benefit; like the Kaplan-Meier curves are like right flat on top of each other. Um, and now there has been a lot of interest in interleukin-6 inhibition in polymyalgia rheumatica, and there's a lot of, um, a number of reasons for that. Number one is IL-6 levels seem to, um, track with disease activity in PMR. So if you are treated and your IL-6 remains elevated, you have a much higher risk of relapse. There are some genetic risk factors in the IL-6 promoter gene that increase your risk of relapse. When you do biopsies of the, um, the subacromial bursa, which is actually synovial tissue, there's increased expression of IL-6 in that—in the—in the histology. And as we know in giant cell arteritis, which, depending on your point of view, is either a sibling or a fraternal twin of polymyalgia rheumatica, um, IL-6 inhibition has been shown to be effective. And so we have two options, right? So one is tocilizumab or Actemra; the other is sarilumab or Kevzara. Both are antibodies that bind the IL-6 receptor either in the, um, soluble or membrane-bound, uh, forms. So there have been three trials; I'm going to go over two, really looking at IL-6 inhibition. So the first one is called PMR-spare. This looks at newly diagnosed patients, um, with PMR. Now, the field of Rheumatology is littered with drugs that failed because of a high rate of background prednisone use, making it hard to show a difference between the placebo and the drug, right? And this is most notable, I think, in lupus, but has been seen in another—of—of other trials. And so if you're a company and you want to show that your drug is effective, you're going to make sure that the placebo group—group gets the minimum exposure to to prednisone that you can get away with, and that's what they—they did in this particular trial, right? So they started patients at 20, but they got them off prednisone, um, by week 11, right? And again, remember ACR—College of—American College of Rheumatology criteria is that those patients are typically on for a year. And as you might expect, there was a pretty massive difference in outcomes. So the placebo group—so these—these are the patients who got this very rapid taper, um, had a very, very low rate of prednisone-free remission, whereas the patients who got, um, tocilizumab, um, or IL-6 inhibition had an excellent, uh, very generally a very, very good response. So that's great, but it's probably not the population in which you're going to use one of these drugs, right? Um, and there was a steroid-sparing effect, um, but the population who you really want to, um, want is these patients who are a little bit more refractory, right? You want these—those patients who are on steroids for, you know, one, two, three years and really can't get off; that's really the population you want to see if this very expensive and potentially toxic drug has an effect. And this is the SAFFRON trial. So this is looking at sarilumab or Kevzara. And again, this is really exciting. So these patients, again, were not newly diagnosed; these are patients who are relatively refractory; they had to have a recent flare and were on prednisone of 10 milligrams or more. And sort of this is a very gutsy trial, right? So I—I—I—I don't know the inner workings of—of how this was designed, but I suspect the FDA made them do it in this particular way. So this was not a trial of sarilumab versus placebo; this was a trial of sarilumab versus a short prednisone plus a short prednisone taper versus a more conventional prednisone taper. So again, if you were the company designing this, you would not want to do it this way because you're giving the placebo group more prednisone, and that's going to make it harder to show a difference between your drug, um, and yet despite that—and so this is just the difference in the—in the steroid dose—and despite that, there was a significant benefit to the use of sarilumab. And this is a very, very—actually difficult endpoint—this is sustained remission at one year, um, the, uh, and sustained remission they defined as clinical remission by week 12 and then no flares the rest of the trial. So no flares from month, you know, three to month 12, and you had to adhere to the prednisone taper. As you might expect, in this relatively refractory group of patients, very few of them were able to, um, meet the endpoint, whereas about, you know, 30% of patients in—in—were—this is a number needed to treat of five, which is again pretty good for—for Rheumatology. And another way to kind of slice the data that makes it look a little bit more impressive is just looking at patients who didn't have any signs of PMR, even if they had to have a little bit of steroids. And in that group, right, there's a—a—a quick separation at week four between the two groups. And so it certainly—I think this certainly was a very relatively impressive result, um, for this—for this medication. And so now it—sarilumab is FDA-approved specifically for patients who have what they call glucocorticoid-resistant PMR, which is a little bit of a misnomer; it's just patients who still need steroids to control their disease. So this is very exciting, and now it is being approved by—by insurance. So, um, if you have a patient with PMR who can't get off the steroids or can't tolerate steroids for whatever reason—you know, diabetes, osteoporosis, um, then this is a really, really great option.

That being said, right, interleukin-6 inhibition does have some significant and relatively unique side effects that I think everyone should be aware of. So, you know, the, um, risk of—it is a, um, it is immunosuppressant, right? So there is an infection risk; it's probably a little bit higher than TNF inhibitors. Um, that being said, right, if you remember when you—when I tell patients like, "I'm going to give you this drug," I'm not telling them this—"Your options are either this or nothing," right? The options are either this or more prednisone. And so in that particular case, it certainly could be the case that this medicine really does not overall increase your risk of side effects in this population. And certainly in the trial that I just mentioned, the SAFFRON trial, there is no difference in infections between, um, the sarilumab group and the higher-dose, um, prednisone group, right? Neutropenia is not an uncommon side effect; probably about 10 to 50% of your patients on this will become neutropenic, and sometimes that can be improved by adjusting the dose. And LFT abnormalities again are also not uncommon, um, with this medicine; I think maybe about 20–25%. Now, a sort of catastrophic side effect that I fortunately have not, um, seen but is certainly, uh, uh, needs to be known is gastrointestinal perforation. So interleukin-6 seems to be important in forming the mucosal barrier of, especially the lower GI tract. So about one in 1,000 patient-years, uh, of exposure to this drug will lead to—to a—to a GI perforation. So if you're giving this drug over five years, right, there's a one in 200 chance that will lead to a lower GI perforation. So it's again not insignificant. If you have a risk of diverticulitis, that's a risk factor. So I would not say a history of diverticulitis is an absolute contraindication; it depends a lot on again what the alternative is, but it's certainly a—a kind of something you really need to think about hard before you administer this. And often when these patients come in, they don't come in with, um, uh, the kind of catastrophic peritoneal signs you might expect; sometimes it's a little bit more subtle. The last thing about interleukin-6 inhibition that I think, you know, all of you guys will probably come across, certainly in residency, are what I call these practical difficulties. So we know that IL-6 will raise your LDL as well as your HDL level, but it doesn't actually increase your risk of cardiovascular events, in contrast with tacrolimus or cyclosporine, right? This medicine has been studied specifically against TNF inhibitors and was not found to increase your risk of cardiovascular events. But the problem is, by severing the relationship between the LDL and the CV events, you're not quite sure what to do with the—with the LDL data that you get, right? And I've asked this of a lot of people, and there's no like correction factor that you use, so it's very hard to put in the—if someone has elevated cholesterol on this medicine, the Framingham risk calculator is—is, um, is not that—is not as relevant as it could be. So that's sort of one kind of practical difficulty. The second practical difficulty, and this again something we see a lot in clinical practice, is that interleukin-6 acts on your hepatocytes to induce the production of CRP. So if you block interleukin-6, your CRP is going to go down. But again, now you've lost that as a marker that you can use to—to see if someone is flaring or not. And again, all diagnoses are clinical, right? But certainly when you do use the CRP, if someone—if you have someone who has PMR and they come in and they're having an increase in shoulder stiffness and you're kind of on the fence, you might use the CRP to sway you one way or the other, excuse me. But when you're on this medicine, your CRP is going to be suppressed, and so therefore you kind of lose that ability to—to use the CRP to kind of aid you in, uh, in diagnosis. All right, so the conclusions, and I think the things that I really want people kind of to take away here is number one is again PMR is a common disease, um, but it's very much—it's can be very, um, devastating in terms of the—the morbidity, but it's very much a fixable problem. So if you're thinking this, right, this is certainly a message in—in med school to one of your rheumatologists. The inflammation seems to be centered around the connective tissue at the musculotendinous junction and then again will spread to the bursa and the synovium. We know that 25% of these patients who have pure PMR will have evidence of subclinical large-vessel vasculitis, although we don't quite know what to do with that information yet. And then again, steroids, as I talked about, in general are very effective, um, but as you taper them, it's very common to flare. And so therefore, a lot of patients with PMR will need some dose of steroids over years—years, which obviously can have its attendant side effects. But fortunately, now we do have interleukin-6 inhibitors that are, um, in play that—that are FDA-approved that we can use if necessary. So kind of what would I—see—what I—what—what do I kind of want to see moving forward and what am I sort of excited about? Well, number one is I would like to get a better estimate of the prevalence of PMR, not just in sort of that Rochester population, but sort of more in, um, you know, again, more diverse geographic settings. Um, again, I'm looking forward to seeing how the use of imaging can actually help us—can we formalize an—an imaging test, a sort of aid in diagnosis? So is there an MRI protocol that we'll be able to use that, based on the degree of inflammation in various sites, we can get a good sensitivity or specificity, um, of PMR? And that'll require prospective studies, multicenter, um, studies. Again, the sort of $110,000 question or $50,000 question is—is—is PMR—is it—is subclinical giant cell arteritis? These patients with PMR, they have evidence of vasculitis on imaging studies—is it actually subclinical, or do these patients need a higher dose of steroids? Would they benefit from early introduction of interleukin-6 inhibition? Are they at higher risk for the complications that we typically see with GCA? And if—if that's true, right, then in the future, I could imagine a future where you diagnose someone with PMR, you start them on steroids, but you say, "Today, we've got to get you in for an ultrasound down in—in the lab or even do it at the bedside," and they're going to do a formal assessment of your bilateral temporal arteries, your subclavian vessels, your axillary arteries, because if those are positive, then our treatment is going to be a little bit different. And then how to incorporate these steroid-sparing therapies in our sort of armamentarium? At the current time, right, should—should these be kind of a step-down approach where we start these right off the bat, or should we use more of a step-up approach where we let—have someone have a relapse and then only add them? And if we're going to do that, you know, can we identify people who are more likely to relapse up front and then say, "All right, we can put you into a risk calculator. Miss O'Brien, your risk of flaring over the next year is 75%, therefore, I think we should start, you know, methotrexate or interleukin-6 inhibition now." And then again, I don't think this is ever going to happen, but—but it would be nice to see a comparative effectiveness study of methotrexate or other conventional DMARDs like leflunomide, um, at higher doses with interleukin-6 inhibition, because I do think they probably would be—would be effective.