NIH CLINICAL CENTER GRAND ROUNDS
Episode 2010-12
Time: 1:02:31
Recorded March 31, 2010
Ankylosing Spondylitis: HLA-B27 and Beyond
Robert A. Colbert, MD, PhD
Chief, Pediatric Translational Research Branch, NIAMS
Treatment of Autoantibody-Receptor Disease: The Story of Type B Insulin Resistance
Rana Malek, MD
Staff Clinician, Diabetes Branch, NIDDK
ANNOUNCER: Discussing Outstanding Science of the Past, Present and Future - this is NIH Clinical Center Grand Rounds.
(Music establishes, goes under VO)
ANNOUNCER: Greetings and welcome to NIH Clinical Center Grand Rounds, recorded March 31st, 2010. Today, two speakers from different institutes with diverse topics for discussion. First, we'll hear from Dr. Robert A. Colbert, chief of the Pediatric Translational Research Branch at the National Institute of Arthritis and Musculoskelatal and Skin Diseases who will present on "Ankylosing Spondylitis: HLA-B27 and Beyond". He will be followed by Dr. Rana Malek, staff clinician in the Diabetes Branch at the National Institute of Diabetes and Digestive and Kidney Diseases, who will speak on the topic, "Treatment of Autoantibody-Receptor Disease: The Story of Type B Insulin Resistance."
We take you to the Lipsett Ampitheater at the NIH Clinical Center in Bethesda, Maryland for today's presentation.
COLBERT: Thank you. And I'd like to thank everybody for coming. So the first slide, I have no financial relationships with commercial interests to disclose.
And what I hope to do today in the next 25 minutes is to give you a broad overview. My primary objectives are really for you to be able to describe the major clinical and pathologic features of this disease, ankylosing spondylitis, give you an update on what we currently understand about genetic susceptibility to this disease, and discuss pathogenic mechanisms, particularly in the context of what we are learning about genetic susceptibility.
So first of all, ankylosing spondylitis is an immune mediated inflammatory disease that is present in 2 to .5% of the population. It's more common than lupus. It's less common than rheumatoid arthritis, maybe 1/3 to 1/2 as prevalent. We typically talk about the phenotype in terms of articular and extra articulare manifestations so the key features of the articular manifestations include axial skeletal inflammation shown in the red involving the sacroilac joints and ascending the spine involving the vertebral bodies and frequently seeing hip and less frequently shoulder involvement, and peripheral arthritis can occur but it's less common certainly than rheumatoid arthritis as an example. Enthesitis is a major feature of the disease with inflammation where tendons, joint capsules and ligaments insert on bone and some of the common sites are shown here in yellow.
What is very puzzling about this disease is the tendency for bone formation to occur either during or after the inflammatory process along the spine and this leads to a significant amount of the morbidity associated with this disease. We also learned that
Gastrointestinal inflammation is common. Studies suggest it occurs in about up to 70% of people with ankylosing spondylitis, although it's most frequently subclinical, so not true of inflammatory bowel disease although there is overlap. This primarily occurs in the terminal ilium as well as the proximal colon.
Acute anterior uveitis occurs in 30-40% of people with ankylosing spondylitis and less frequently one can see cardiac and pulmonary manifestations particularly late in the disease process.
This is the classic slide of the progression of ankylosing spondylitis in one individual over 25-year period of time. I think you can appreciate the postural changes and what happens initially is one gets a flattening or loss of the lumbar, beginning to see a kyphosis in the thoracic region and this person had significant hip involvement with flexion contractures. So between 1972 and 1973, he had hip replacement surgery which resulted in a significant improvement in his posture but I think you can imagine that this disease can be very debilitating.
So we currently define ankylosing spondylitis based on criteria initially developed about 40-50 years ago. So one has it as by the modified New York criteria if there is radiographic sacroilitis, either bilateral grade 2-4 disease or unilateral 3-4 disease. Grade 4 is complete fusion whereas grade 2 is the beginning of erosion process. One has to have one clinical criterion, inflammatory back pain, limitation of lumbar motion or decreased chest expansion.
So clearly evidence of axial involvement. I don't have time to go into all of this but I just want you to appreciate that while ankylosing spondylitis is really the prototypic form of spondyloarthritis, there is a large group of undifferentiated disease. And currently best estimates are that 40-50% of these individuals go on to develop ankylosing spondyloarthritis and we don't have currently good ways who will develop the complete phenotype.
So the sacroiliac joint is an interesting joint. It's a major weight-bearing joint or joints of the axial skeleton. And it covers a fairly large surface area. We know relatively little about what goes on in that joint.
There are a couple studies that have been published. The initial observations were made in the mid 1990s. And these scientists did some ct guided biopsies of very early sacroiliitisand they demonstrated with various staining methods that these are highly inflammatory lesions containing t-cells expressing CD3 and also CD8 and CD4 positive K cells. They documented CD68 positive macrophages as well as osteoclasts that were activated and appear to be involved in degrading bone in these early lesions. What was more important was at the time, they were able to demonstrate within hybridization, a pretty dramatic over expression of TNF alpha. Subsequently, overexpression of Il6 is documented and probably in later lesions, relative increase in TGF beta which might be associated with the fibrosis and bone formationing that occurs as this disease goes on.
So we know even less about the vertebral legions for many of the same reasons. They are difficult to get at. You can see magnetic resonance imaging of vertebral lesions, referred to as corner inflammatory lesions. And one gets inflammations in the apophyseal joints. So some scientists have provided us with a schematic representation of what they view as a three-phase process, an early inflammatory process with mononuclear cells, osteoclasts, a structural damage phase where there begins to be a lot of degradation of the intervertebral disks with activated fibroblasts and also increased angiogenesis and later on, one finds osteoblasts that are presumably involved in laying down bone and the formation of these. There is really very little known about specific cytokines expressed in these particular lesions.
So again, one more example of an individual with long-standing disease and this is the classic bamboo appearance of the spine with multiple syndesmophytes bridging these bodies, and ongoing concomitant osteoporosis juxtaposed with increased bone formation.
I want to give you an overview of what we have learned about genetic susceptibility to ankylosing spondylitis. The story begins in 1973 when this really remarkable association between this MHC class 1 allele and ankylosing spondylitis was first described. It was a serendipitous finding because they were using ankylosing spondylitis as a control group looking for HLA associations with gout. B27 is present in 90-95% of individuals with ankylosing spondylitis compared to less than 10% healthy controls. The frequency in the control population depends on the part of the world you're from but generally in the 7-8% range. One gets odds ratios of 100 or more which is really unusual for a single gene in a complex genetic disease. As is highly heritable with sibling reoccurrence risk ratios in large part due to HLA-B27 but also to the remaining genetic component of susceptibility. But, we are faced with a disease where only about 5% of all individuals with B27 actually develop the disease. So in fact, whereas you can almost say that B27 is required for disease, it's clearly not sufficient for disease and as I said, this is very complex.
So the current estimates are that B27 constitutes about 40% of the variance or 40% of the over all genetic susceptibility. A couple of associations were made in the mid 1990s with the cytochrome P450 gene and also with the class 2 gene HLADR beta 1, which have held up over time. But we have seen some major advances in the past couple of years with the availability of genome-wide screening in a limited study.
In a limited study IL23 receptor polymorphism and polymorphisms in the gene now called ERAP1, also known as arts 1, were discovered to be associated with as. We know that something in the IL1 locust and the studies suggest it's probably IL1 alpha but I think this is not really definitive at this point. And then I'll tell you about a recent genome-wide association study that was just published.
So this is a typical Manhattan plot looking at the statistical significance as a function of the markers across the genome and the first thing you can see here is this striking peak for the MHC, due to HLA-B27 and perhaps something else in the MHC as well. This study, which was, as I said, a large GWAS, interrogated over 280,000 SNPS, in over 2000 cases of as compared to over 5,000 controls in a largely British and American and Australian population. So this study confirmed IL23 polymorphism and confirmed ERAP1 and also identified two chromosomal regions, 2P15 and 21Q22, unfortunately these are gene deserts so it's not clear at all what the offending gene is or if these are intergenic regions which are involved in regulation of expression, et cetera. Two other hits were found. One is in the gene encoding the IL1 receptor or decoy receptor. And then another gene that barely reached genome-wide signifigance encoding the anthrax toxin receptor 2.
So what we currently have is about 75% of the variance explained by B27 and ERAP1 is responsible for 26% of the variance, another 10% by IL23 receptor and it's not clear exactly how much of the variance but probably a relatively small amount by these other genes. And this is for a disease that is considered to be over 90% heritable. So, it's thought that there are environmental factors but the factors are probably ubiquitous. And this is in comparison to rheumatoid arthritis, considered to be about 65% heritable.
So there are additional genes on the horizon that have been discussed at meetings but the published data are not yet available. But I just mentioned them because they are interesting in terms of pathways that are involved and I'll come back to that a little bit later. But so polymorphisms in the TNFR1, TRAD, card 9 and stat 3 and HLAB alleles and a few other alleles.
So, I'm going move on and talk a little bit about pathogenic mechanisms and maybe how we view these things coming together. So, to back up a step, what does genetic susceptibility really do? In fact, we are talking about a chronic spontaneous inflammatory process. Most people don't really consider this to be an autoimmune disease because in fact, there are no auto antibodies that are associated with ankylosing spondylitis or no consistent observations of auto antibodies. And a little more controversial but no consistent demonstration of antigen or HLA specific auto reactive CD8 or CD positive t-cells although we know there is an important t-cell component. This is really unclear exactly what is going on but autoimmunity by its current definition has not been really demonstrated.
So we also have a phenotype with unique location of this inflammation, the axial skeleton, the spine, and also in feces and a process that somehow leads to abarent ossification associated with or following the resolution of some of this inflammation.
So, historically, because of the discovery of the relationship between B27 and this disease, the focus has largely been on what B27 is doing. Most of the other genes here are relatively recent discoveries. Whether that has led us down the wrong pathway or not is a subject that could be discussed, but I'm going present to you the current ideas about what HLA-B27 is doing.
So just to remind you, this is an HLA, mhc1 molecule involved in presenting peptides in particular to CD8 positive t-cells. The complexes are recognized by alpha beta t-cell receptors. The heavy chain associates noncovalently with microglobulin and class molecules are expressed fairly ubiquitously on nucleated cells but probably highest on macrophages, dendritic cells and cells of the immune system. We know that the hlab locus is extremely polymorphic. Over 1,500 alleles documented to date in the human population whereas HLA is a bit less polymorphic with about 950 alleles and HLAC less polymorphic.
There was a breakthrough made in 1990 when scientists created a transgenic rat line, are actually multiple lines, that overexpressed HLA-B27 in human beta 2 microglobulin. That demonstrated that one could get disease associated with B27. Now the phenotype is inflammatory and it overlaps with spondiloarthritis but it's not a perfect phenotype for ankylosing spondylitis. Perhaps not surprising but nevertheless, that's one of the main models we have to work with.
So, the current hypotheses are shown in cartoon form here with this being the HLA-B27 molecule. One hypothesis has been that it presents arthritogenic or perhaps spondilogenic peptides that are recognized by CD8 t-cells. So a self peptide would presumably become atherogenic if it resembles a pathogenic peptide. It's thought that the specificity of B27 is a major contributor. Interestingly in the animal model, if you deplete CDA t-cells or knock them out, it's been possible to knock out CD8 alpha to eliminate all CD8 positive t-cells, you get the phenotype normally. This suggests that CD8 t-cell recognition at least in the animal model is not a major mechanism of this inflammatory process.
It was cell surface dimers of B27 were then described and now it's been shown that these dimers can be recognized by leucocyte receptors that are present on other monocytic, dendritic cells and k cells. This is hypothesized to perhaps be a trigger for inflammatory process.
And then the third major hypotheses is one that we have been responsible for. After discovering that B27 had a tendency to miss fold. When it's up regulated, a significant portion of the molecules that would normally go out to the cell surface, actually self associate and form dimers and bind to an ER chaperon, and then can cause a stress in the endoplasmic reticulum. So this is a very different idea suggesting that cells behave abnormally and that there is not an abnormal recognition process.
So we have looked extensively at the consequences of this misfolding in transgenic rats and what we have demonstrated is that B27 expression and in particular, upregulation, is sufficient to generate this unfolded protein response. It's prominent in macrophages and dendritic cells and we are looking at others that may be relevant to the phenotype to see where this response occurs.
But what was very interesting is we found that the UPR synergizes with microbial products in the induction of certain cytokines. So if you superimpose this response on cells, you get increased production of IL23 and interferon beta relatively selectively, at least in the acute response. So this was of interest to us because of the discovery of TH17 t-cells and the TH17 axis and the fact that IL23 is an important cytokine in the differentiation of t-cells but clearly in stimulating them to produce IL17.
So TH17 cytokines have been implicated in a number of human diseases and animal models. And are probably quite important in terms of IL17 being able to synergize with TNF in inducing local production of a number of other cytokines including TNF Alpha. So when we looked a little bit more closely in these animals, what we found was that the th1 response had been described before and what we found was IL17 expression is dramatically up regulated in colitis that occurs in these animals and we could also demonstrate increases in IL23 expression, strong up regulation of B27 and UPR activation in isolated APCS, and also that some of the TH17 cells were expressing interferon gamma as well as IL17, which might be important in terms of a positive feedback loop. So we had a lot of circumstantial evidence that this axis might be important in the animal model.
So interestingly, I have already eluded to the fact that IL23 receptor polymorphisms are implicated in susceptibility but there is additional growing evidence that IL17 is elevated in serum synovial fluid of patients with as and other spondylo arthropathys. Increased circulating t-cells have been documented and in intestinal inflammation lesions, over expression of IL23 is shown. So this is involvement of this axis in human disease as well.
So I haven't said much about treatment and there is really not a tremendous amount to say. But I do want to mention that while nonsteroidals are used and for a long time were the major means of treating individuals with ankylosing spondylitis, they have benefits with regard to symptoms and there is a small effect on slowing progression of bony involvement. It's statistically significant but it's probably not very meaningful.
Sulfasalazine may be beneficial and methotrexate which is used in rheumatoid arthritis has not been shown to have any significant benefit in open label trials. And what we have learned in the last decade is that TNF inhibitors really do reduce or eliminate symptoms in a large majority of individuals but the data coming out now, suggests that TNF inhibitors may not inhibit bony progression. Now these are studies where people had well established ankylosing spondylitis that were treated and so clearly we need studies looking at treatment of earlier disease.
So, I'm on my next to last slide. I'd like to summarize by saying that I told you this is a complex genetic disease. I think that we are well on our way to understanding the majority of genetic susceptibility. Pathogenis includes TNF and Il6 overexpression both local and systemic. There is increasing evidence for over IL23 over expression and TH17 activation in animal model that is predominantly driven by the expression of B27. Structural remodeling leads to ossification, a process we think little about. There is some emerging information but that would have to be the subject of another talk.
And I'd like to leave you with the provocative thought that B27 in its role in this disease may be something that is really unanticipated based on normal function. So, TNF is beneficial but may not prevent bony progression but I think there are a number of new targets, including cytokines in the TH17 axis that are emerging but a need to better understand the abnormal bone formation that occurs in this disease.
So, if I can indulge you for one more slide, I would just like to point out that we are hypothesizing that there may be a connection between this abnormal tendency of B27 to misfold and a stimulus, the UPR, that might promote IL23 production and that there may be therefore be a link between B27 and TH17 inflammatory process. Obviously, I have alluded to overexpression of TNF alpha and IL26, which could be driven to IL17. And if you begin to super impose some of these other genes that are implicated in pathogenesis, many of them fit within this axis. So I think there is a lot of room for hypothesis and a lot of work that needs to be done in this area.
So, that's my last slide. Thank you for your attention and I would be glad to try to answer questions if there are any.
[APPLAUSE]
QUESTION: I just want to find out one thing, you mentioned the association of IL1 gene with gastrointestinal disease, IL1 which was supposed to be more associated with disease which also has a relationship with ankylosing spondylitis. And secondly, to have been finding certain very useful papers about treating ankylosing spondylitis or undifferentiating spondylitis that drug which has been found to be effective in many series and also a primer. So what is your opinion about these two points?
COLBERT: Let me take the last one first. Just because of time, I didn't mention the thalidomide. There is evidence that primarily comes from studies in china but there may be others that thalidomide may be a benefit. That is something to consider whether that is through inhibition of TNF alpha or other cytokines is not known. I didn't quite understand your question about IL1. The point I want to make is that there is something in the IL1 locus, a metanalysis suggests IL1 alpha may be at least one of the genes. It could be that there is something else involved and there clearly is an overlap between inflammatory bowel disease and this particular phenotype. So there could be additional affects of other IL1 genes.
QUESTION: So no neutrophils in the pathology?
COLBERT: Well, that is a good question. I don't think that people have looked early enough to know. I think that there are neutrophils. It's not a neutrophil predominant disease later on. It's predominantly mononuclear cells.
QUESTION: Do you find that sometimes B27 presentations of other clinical syndromes like anterior scleritis, help to predict the phenotype of ankylosing spondylitis or in fact spondiloarthropathys and do you look for other nonB27 antigens in association with that?
COLBERT: B27 is associated independently with acute anterior uvitis. I think that's what you're alluding to. There is a strong association there even in people who don't ever go on to develop ankylosing spondylitis or even -- there is a debate as to whether you call it spondyloarthritis. There is clearly an association there. Individuals may be picked up with acute anterior uvitis who will go on to develop the disease. As we learn more, there is a real opportunity with this disease, as there is with others, to begin to predict who is going to develop the more complete phenotype.
QUESTION: You mentioned uv6 having a relationship with this spondylitis. It's a polymorphic gene in terms of drug metabolism and associated with a number of phenotypes. So is it known which of the variant alleles is linked in terms of ankylosing spondylitis?
COLBERT: It is and I don't want to give you incorrect information. It's clearly the hydroxylase activity but I can't tell you whether it's more or less associated with susceptibility.
MALEK: Thank you for the introduction and the opportunity to speak at Clinical Center Grand Rounds. Today I'll be talking about a new treatment protocol we have for type B and insulin resistance. So I have no conflicts of interest to disclose but I will be discussing off label usage of FDA approved drugs listed here.
So, the objectives for this talk are for you to understand the pathophysiology and clinical presentation of type B insulin resistance and to understand the rationale behind a novel therapy for treatment. While these are my official objectives, what I would really like to do in these 25 minutes is tell you a story that starts here in the NIH Clinical Center about 40 years ago and one that we have added a new chapter to in the last three years.
So we'll start with the case that I saw very early in my fellowship here. Which was a 20-year-old thin African American female who presented with a 6 month history of a 35 pound weight loss despite eat about 5,000 calories a day. She had polydipsia and polyuria. On first look this looks like a severe but typical presentation of type I diabetes. But she was also amenoreic and also had severe acanthosis nigricans, so severe her skin was no longer an effective barrier for infection and she was in and out of emergency rooms and had to have surgical debridement as well as antimicrobial intervention. She started on insulin therapy but continued to have persistent hyperglycemia despite taking 18,000 units of insulin a day. So in this one case we see the catabolic manifestations of type I diabetes but the skin findings and severe insulin resistance as evidence by the high doses of insulin of type II diabetes, perhaps. So what is going on?
Well, I'm going to step back from this case and talk about insulin resistance, a term that is thrown around quite a bit. And one way to think of it is a clinical spectrum of disease in that the most commonly known forms of insulin resistins like type II diabetes and the metabolic syndrome, are the least severe manifestations. There are syndromes of insulin resistance that can either be congenital or acquired that are actually quite severe. These include leprechaunism, which is an auto resomal recessive disorder caused by abnormal growth and insulin resistance with a mutated insulin receptor. Lipodystrophy, which can be congenital or acquired and is associated with a loss of adipose tissue, diabetes, extreme insulin resistance and dyslipidemia. And then there are the type A and type B insulin resistance syndromes. Type A is caused by mutation of the insulin receptor. This is characterized by hyperglycemia, severe insulin resistance and extreme viralization in female patients. And then type B is caused by inhibition of the receptor which will be the focus of the talk today.
The history of this disease begins here at the NIH and the very first descriptions came from the diabetes branch of what was then called the national institute of arthritis, Metabolism and digestive diseases. The first description came in 1975 in science magazine where they described six unusual patients who had extreme insulin resistance and had reduced insulin binding and they suspected that perhaps an autoantibody was involved.
A year later in the journal of clinical investigation, they characterized that autoantibody in a series of experiments and in that same year, they gave a very detailed characterization of those initial six patients and they stratified them as having type A or type B insulin resistance.
So, the simple definition is that type B insulin resistance is caused by a highly specific autoantibody against the insulin receptor. This blocks the action of insulin and results in hyperglycemia at very high concentrations.
I will mention but not discuss in detail this talk, that at very low concentrations of the antibody, it can act as a partial agonist and can result in hypoglycemia. I want to step back for another moment and say that while what I'm talking about seems to be an endocrine disease, I'd like to you think about it as belonging to a class of diseases caused by an auto antibody to a cell surface receptor.
In endocrinology, our most common example is graves disease caused by an antibody receptor which results in hypothyroidism. And there are cases of infertility and premature ovarian failure caused by antibodies. But in neurology we have myesthenia gravis and the antibody to the acetylcholine receptor.
Very recently in 2009, a group in Germany described a group of patients that were randomly assessed over 200 patients of various body weights and looked for antibodies in the melanocortin 4 receptor. They found that in 4% of obese and overweight individuals, they had antibodies to this receptor which was not present in mean controls suggesting that perhaps there are some autoimmune forms of obesity. That's this class of disease that is type 1B resistance falls.
Most of the experience in this disease comes from the intramural branches of the NIDDK.
We have the largest cohort of these patients because we are a referral center. And in over 40 years we have acquired 38 patients to give you a sense of how rare this is. The prevalence is unknown but we know that women are predominantly affected with 85% of patients being female and 80% of our patients are African American. It frequently coconsists with lupus or as a peri neoplastic syndrome.
I'd like to stress this disease has a very high mortality. In our initial case series, 44% of our patients died within 7 years of diagnosis. These patients presented on average at an average of 49 with an age of death at 56. So really in the prime of their life.
When we tried to categorize what disease we saw most frequently with type B insulin resistance, in our population we found that lupus and mixed connective tissue disease were predominantinant with 70% of our patients. We had a case or two of multiple myeloma or cirrhosis, but in about a quarter of or patients we couldn't pin down what their underlying autoimmune disease was.
Moving to the clinical manifestations, hyperinsulinism is thought to cause a variety of metabolic signs and symptoms and it could cause very disfiguring hyperangiogenic signs and symptoms.
In terms of metabolic manifestations, we have average blood sugars of 370. They have polydipsia and polyurea and this could be impressive with up to 15 liters of urine a day. They are hungry all the time but despite that they lose quite a bit of weight in a very short period of time, with an average weight loss of about 43 pounds over a few months. They have very high insulin requirements on 5 thousands units of insulin a day and yet still have blood sugars of 370 and disfiguring aconthosis nigricans, a skin condition that we see in insulin resistance in which we basically see thickening and darkening of the pigment of the skin caused by high circulating insulin levels which interact with insulin like growth factors and stimulate the growth of dermal fibroblasts.
These are some pictures of the skin change that is we see in our patients. These are three patients who are not related to each other. But there is a characteristic distribution of the acanthosis nigricans in the periocular and perioral area. So all of our patients seem to have a typical facial appearance that we see. But the aconthosis nigricans is not restricted to the case.
This gentleman is caucasian and as you can see, this disease has taken quite a toll on him. You can see evidence of the wasting and thin extremities and the skin hanging off of him.
I would like to you focus on the central panel and look at this extreme thickening and cobblestone appearance of the skin. In addition to being disfiguring, these lesions can be malodorous and quite painful.
So hypoandrogenic manifestations are seen in premenopausal females because insulin acts as a pathalogic growth factor on the ovaries. Our patients have enlarged ovaries as seen here on this ultrasound image and they are polycystic in nature. Our patients are amenorrheic and elevated testosterone levels ten times the limit of normal. This results in acne and a male pattern hair loss.
The laboratory manifestations for this disease, some are obvious such as hyperglycemia but what was unexpected are the low triglyceride levels we see. The reason for that is common forms of insulin resistance are associated with elevated triglyceride levels. One of the diagnostic criteria for metabolic syndrome is to have those levels of greater than 150. In our patients they are 45 milligrams per deciliter. And they also have elevated adiponectin, a hormone secreted by fat cells. That is a good hormone to have. It has anti-inflammatory and insulin sensitizing properties and in obesity and type II diabetes, adiponectin levels are low. In our patients, they are elevated. This is opposite to what we see in more typical forms of type II diabetes and this finding of low triglycerides and elevated adiponectin is seen in cases of type A, type B insulin resistance. Treatment has generally been very frustrating.
We divide our treatment into metabolic management and immunosuppressive management. Metabolic management is to use very high doses of concentrated insulin. The usual insulin we use for type II diabetes is U100. For every one ml, there are 100 units of insulin. U500 insulin, there are 500 units per ml and we put patients on thousands of units a day. The reason we do this is that we are trying to control the ketabolic state and have some impact on the hypoglycemia because the antibody inhibition is not 100% complete and we are trying to out compete it.
The second component has been immunosuppressive management which could be either generalized or specific. The specific instances are where there is a clear underlying disease that warrants treatment so if the patient has lupus nephritis, we treat it and hope the type B insulin resistance gets better.
In all cases where there is no clear disease to treat, a variety of generalize d immunosupressive regimens have been tried in the past with mixed success by us and other physicians.
And this slide shows what we have seen in our patients initially. About 1/3 of our patients will go into spontaneous remission of this disease. However, this occurs over a very extended period of time of 1-4 years. And this is while they continue to suffer from all of the devastating co-morbidities associated with it.
Patients who have been treated in the past and usually that treatment again has been specified to their underlying disease, 50% achieve remission after treatment. But 50% continue to have persistent disease and this can extend for as long as our follow-up allowed, up to 19 years. So, not very good in general.
So now that I given you that background, I'd like to go back to our case.
This was our patient in 2005 prior to coming to NIH and onset of her disease. She is a lovely woman, healthy, normal body weight and I'd like to you pay attention to her skin color. And this is what she looked like when she came here to the Clinical Center. You can see that overall her skin is a lot darker than it used to be. You can see the thickening of her skin in her neck and axilla and in her hands. You can also see acne on her face and on her back as a result of her high androgen levels. I will share when I went to first see this patient, I didn't really know what type B insulin resistance was and a walked into the patient room and saw this woman and walked right out because I was sure I walked into the wrong room. She was unbelievably frail and ill looking and I actually thought she was a terminal cancer or aids patient. That's how sick she appeared. This disease had devastated her life. She dropped out of college. I mentioned some of the symptoms but on top of all that, she had multiple abscesses draining when she presented here. She had a bilateral uvieitis and bilateral ear in infections. Her teeth were rotting and no dentist would test her until she got her blood sugars under control. She had oral and vaginal candidiasis and facial nerve palsy. She was profoundly sick. And we all really worried about this young woman. And she motivated us to think about perhaps trying to have a new strategy to treat this devastating disease.
So we knew what our goal was because the pathophysiology of this disease is so well understand. We wanted to eliminate the autoantibody. Like so many things in medicine, we borrowed from our colleagues and other fields. So we looked at what was happening in treatment protocols for lupus and lymphoma and multiple myeloma and took bits and pieces from there. So we started with rituximab, an antibody against CD20 expressed by B-cells. These are the precursor cells to the antibodies. We used high dose steroids and these were aimed to reduce the preexisting antibody-producing cell pool. And then we used b and t-cell directive immunosuppression to quell whatever underlying autoimmune disease was present. In one case we chose to use cyclosporin. And this goes over her treatment course. Just follow me.
So across the x axis we have a time course in weeks. Across the y axis we have the hemoglobin A1C. This is a marker for what blood sugar is over the course of 2-3 months. And to give you a reference range, this brown line indicates an A1C of 6.1%. This is where we diagnosis someone with diabetes and when someone has diabetes, we aim roughly to 6.5-7%. So as you can see, she comes in well above that mark. Despite being on 18,000 units of insulin a day. She had an A1C of 12% which corresponds to average blood sugar of 300. And she presented to us, we tried some of our old techniques.
We tried giving her plasmapherisis but had no impact on her blood sugars. We initiated protocol with just rituximab and cyclophosphamide. Her A1C worsened and we did that. We lowered her insulin dose to 6,000 units because the high volumes required to give 18,000 units was causing this 50 kilogram woman a great deal of pain because she didn't have any fat to inject into. She basicallly hovered with an A1C of 20 so blood sugar of almost 500 for the better part of the year. And roughly at about week 50, she started to trend downwards but far from normal. At that point, we made the decision to initiate the final component of our regimen which was to add the pulse steroids and she was able to come off all insulin therapy and had a normal blood sugar about two years after presentation.
This is a western blot that looks at the immunoprecipitation of the antibody and the insulin receptor. What you can see is that if you don't have the antibody, this is a negative control. This is what it would look like if you have it. And as you can see, our patient was strongly positive. So strongly positive that our collaborator in cambridge uses her as the positive control for this test. You can see the antibodies titre remains unchanged until roughly between weeks 50-66. It decreases to the point it's no longer detectable.
And so to summarize again, the treatment regimen, we went back and standardized it. And what we did is we used the rituximab again to target b-cells and we gave this in cycles and repeated cycles as the disease persisted and CD19 levels were elevated. We gave steroids to target mature cells and doesed this as long as the disease was active in our patients. And they were on daily cyclophosphamide.
Again, we had one patient on cyclosporin because of severe neutropenia. So, we decided that for all other patients we were going to give all 3 medications together rather than the step-wise approach. And we applied it to all patients with active disease referred to the Clinical Center from 2006-2009. We treated six additional patients.
This is a chart that looks at the demographics of our patients. You can see we have a young woman at the age of 17 and our age range extending to 64 years of age. Consistent with our previous findings, only two of our patients were male. And the majority of our patients were African American except for one patient who was a first nation Canadian. The severity of the disease ranged in our patients from having antibody titre of one plus to 4 plus. They were semiquantitative assessments of the western blot. And in 4 of our 7 patients, we found they had lupus or mixed connective tissue disease that was not active but in 3 of our patients we did not have an underlying disease. And this chart shows the metabolic parameters at presentations to the Clinical Center and upon achieving remission. And this is the average over 7 patients.
So in general, our patients came in with blood sugars of about 268 greater while on 4,000 units of insulin a day with elevated hemoglobin A1C levels. After treatment, we have normal blood sugars, no insulin requirements and a normal hemoglobin A1C. 3 of our patients were premenopausal females and all presented with amenorrhea, hirsuitism and acne, and it's not surprising when you see that their average testosterone level was 550 and the upper limit of normal for a female is 60. And so this was confirmed on the free testosterone levels as well.
After treatment they had normalization of their testosterone levels, resumption of menstrual cycle and resolution of hyperandrogenicsymptoms. This is normalization of the blood sugars and discontinuation of insulin therapy and once our patients achieved remission we chose to place them on one year of maintenance therapy. Our rational for this was that this was a devastating disease. We feared for the possibility of relapse and these patients have an underlying auto immunity and we were attempting to truly prolong the state of remission. And in almost all patients, they were placed on azathioprine for a year. One patient remained on cyclosporin.
In this slide, it summarizes treatment and remission in our patients. So on average, our patients required 1-2 cycles of rituximab and 3 doses of steroids. The time to remission ranged from as little as two months to 27 months. This was our initial patient. She had the longest time to remission. Average time is 8 months. Once we standardized the protocol to give all 3 medications together, patients achieved remission much faster and average time to remission in those cases was 5 months. This contrasts with what I told you about the spontaneous remission rate which occurred over a period of 1-4 years. This happened in a much shorter time period. The duration of remission is calculated up until this day I give this presentation. Some of our patients have only recently entered remission two months ago. 4 of our patients have been in remission for greater than one year and we taken them off the americans therapy and all 7 remain in remission from this disease.
So in conclusion, type B insulin resistance is caused by autoantibody to the insulin receptor. It has a very severe morbidity and high mortality rate. And previous therapeutic regimens have been unsatisfactory.
What I describe to you today has been a well-tolerated effective multi-drug approach targeting different limbs of the antibody producing system. I wanted you to think of this disease as belonging to a class of diseases AF auto antibodies and suggest this treatment regimen, while it worked so well in type B insulin resistance is not specific to that disease and perhaps could be used for other diseases in the same class, such as severe refactory mysemia and I'm sure this is provocative but in autoimmune forms of obesity even.
I do need to give special credit to the large group I work with. Dr. Phil Gordon has been following these patients for 40 years here at the Clinical Center and in consultation with other physicians. And I can say that he is very pleased to note that it's one thing to follow a disease for 40 years. It's nice to finally be able to treat it. Dr. Jim Baylor brought his expertise in management of lupus patients. We have collaborators in Cambridge university who are invaluable in running the assays for us.
I thank you for your attention and I hope you enjoyed this talk and I would be delighted to take questions.
[APPLAUSE]
QUESTION: Thank you for the illuminating talk. I wonder, say for example, the cases of autoimmune disorders, especially collagen disorders you mentioned, which is associated with this type of hypoglycemia, most of them have been treated by steroids before because steroid is the main stay of all of those treatments. So that means they do not have this defactory diabetes despite steroid treatment? Because most of the cases, the mainstay is steroid whereas on the other hand, lupus, mainstay steroid. So that means all of these cases despite steroid treatment, which itself is a treatment for this disease.
MALEK: So, what I would like stress is that while this disease frequently exists with cases of lupus, most cases of lupus do not have type B insulin resistance. This is a very rare -- this is reportable. There are case reports of lupus patients with type B insulin resistance. Those patients who are placed on steroids and develop hyperglycemia I would argue it is a completely different mechanism.
QUESTION: thank you. That was a great talk. It is my understanding that some patients with this type B insulin resistance have auto antibodies to insulin itself. And I'm wondering if in those patients there are differences in the demographics or phenotype?
MALEK: So the type B insulin resistance is very specific. An antibody to the insulin receptor. Antibodies to insulin is a different disorder something seen more commonly when we used to give nonpurified, nonrecombinant insulin. So it's a very kind of a distinct disease manifestation.
QUESTION: But isn't there a Japanese cohort that has auto antibodies to insulin?
MALEK: Yes but they do not have type B insulin resistance.
QUESTION: Thanks again. That was a lovely talk. I just wanted to focus again on the steroids, you mentioned that you added in steroids as treatment against plasma cells and I mean, steroids are extremely nonspecific and I think the evidence is pretty poor that they target plasma cells. Have you considered again looking at other disciplines, for example, hematology where they use a new agent, a protein inhibitor? That is effective in eliminating plasma cells.
MALEK: I have to say we have not considered using that agent but I'm very interested in looking further into it. Thank you for the suggestion.
PRESENTOR: Thank you to both the speakers for terrific talks.
[APPLAUSE]
ANNOUNCER: You've been listening to NIH Clinical Center Grand Rounds recorded March 31, 2010. On today's presentation, we heard from two speakers from different institutes with diverse topics for discussion. First, we heard Dr. Robert A. Colbert, chief of the Pediatric Translational Research Branch at the National Institute of Arthritis and Musculoskelatal and Skin Diseases discussing "Ankylosing Spondylitis: HLA-B27 and Beyond". He was followed by Dr. Rana Malek, staff clinician in the Diabetes Branch at the National Institute of Diabetes and Digestive and Kidney Diseases, who spoke on the topic, "Treatment of Autoantibody-Receptor Disease: The Story of Type B Insulin Resistance." You can see a closed-captioned videocast of this lecture by logging onto http://videocast.nih.gov -- click the "Past Events" link -- or by clicking the "View Videocast" link on the podcast homepage at www.cc.nih.gov/podcast. The NIH CLINICAL CENTER GRAND ROUNDS podcast is a presentation of the NIH Clinical Center, Office of Communications, Patient Recruitment and Public Liaison. For more information about clinical research going on every day at the NIH Clinical Center, log on to http://clinicalcenter.nih.gov. From America's Clinical Research Hospital, this has been NIH CLINICAL CENTER GRAND ROUNDS. In Bethesda, Maryland, I'm Bill Schmalfeldt at the National Institutes of Health, an agency of the United States Department of Health and Human Services.
