Disease associations with isolated elevations of each of the four IgG subclasses

Disease associations with isolated elevations of each of the four IgG subclasses

Author’s Accepted Manuscript Disease Associations with Isolated Elevations of each of the four IgG Subclasses Sally Engelhart, Robert J. Glynn, Peter ...

758KB Sizes 0 Downloads 14 Views

Author’s Accepted Manuscript Disease Associations with Isolated Elevations of each of the four IgG Subclasses Sally Engelhart, Robert J. Glynn, Peter H. Schur

www.elsevier.com/locate/semarthrit

PII: DOI: Reference:

S0049-0172(16)30325-0 http://dx.doi.org/10.1016/j.semarthrit.2017.03.021 YSARH51177

To appear in: Seminars in Arthritis and Rheumatism Cite this article as: Sally Engelhart, Robert J. Glynn and Peter H. Schur, Disease Associations with Isolated Elevations of each of the four IgG Subclasses, Seminars in Arthritis and Rheumatism, http://dx.doi.org/10.1016/j.semarthrit.2017.03.021 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Title Page: Disease Associations with Isolated Elevations of each of the four IgG Subclasses By: Sally Engelhart1, Robert J Glynn2 and Peter H Schur3

1

Sally Engelhart, BA, MD Candidate 2017, Harvard Medical School, Boston MA

Corresponding Author 44 Washington St, Brookline MA, 02445 [email protected] 215-410-9594

2

Robert J Glynn, ScD, Professor of Medicine, Harvard Medical School, and in the

Department of Biostatistics, Harvard School of Public Health, Boston MA 900 Commonwealth Ave, Boston MA, 02215 [email protected]

2

Peter H Schur, MD, Professor of Medicine, Harvard Medical School at Brigham and

Women’s Hospital, Boston MA 60 Fenwood Rd, Boston MA, 02115 [email protected]

Abstract:

Purpose: Immunoglobulin G4-related disease (IgG4-RD) is a relatively newly defined disease entity that refers to a group of immune-mediated disorders that have certain histopathologic, serologic, and clinical features in common. IgG4-RD is often associated with elevated serum IgG4. The discovery of IgG4-RD highlights the scarcity of literature examining elevations in other IgG subclasses and their potential associations to disease. In this retrospective chart review study, we aim to address that gap, by exploring disease associations in patients with isolated IgG subclass elevations.

Methods: We identified 552 patients with an isolated elevation of one of the IgG subclasses, and performed a systematic chart review to identify the diagnoses of those patients. We examined the distribution of diagnoses, using the Fisher’s Exact Test to determine if a diagnosis was significantly associated with an isolated elevation in one of the subclasses.

Results: Autoimmune pancreatitis, aspirin-exacerbated respiratory disease (AERD), nasal polyps, eosinophilia and celiac disease were significantly associated with an isolated elevation in IgG4. Hepatitis C and monoclonal gammopathy were significantly associated with isolated elevations in IgG1. Rheumatoid arthritis (RA) was associated with both an isolated elevation in IgG1 and IgG3. Hypothyroidism and irritable bowel syndrome (IBS) were significantly associated with isolated elevations in IgG2.

Conclusion: These results confirmed some established associations between autoimmune pancreatitis, AERD, nasal polyps, and eosinophilia and elevated serum IgG4, and between monoclonal gammopathy and hepatitis C with elevated serum IgG1. It uncovered novel associations between RA and elevated IgG1 and IgG3; hypothyroidism and IBS and elevated IgG2; and between celiac disease and elevated IgG4.

Key Words: Immunoglobulin, Immunology, Autoimmunity, Rheumatology, Immunoglobulin Subclasses, Immunoglobulin-G

Introduction:

There are four subclasses of IgG molecules (IgG1-4) that were first described in 1964 by both Grey and Kunkel and Terry and Fahey.1,2 In normal human sera, IgG1 is the most prevalent subclass, making up 60-70% of the total IgG, followed by IgG2 (20-30%), IgG3 (5-8%) and lastly IgG4 (1-4%) (see Table 1).3 IgG molecules generate an immune response by binding Fc receptors (FcR) on target cells and/or by activating the complement system.4 Functional differences among the IgG subclasses stem from structural differences in the hinge and heavy chain constant regions.3,5

IgG1 and IgG3 are strong mediators of both FcR- and complement-mediated effector functions, and are the predominant subclasses involved in response to protein antigens. IgG2 is a weak mediator of FcR- and complement-mediated effector functions, and is involved in the response to polysaccharide antigens. IgG4 is quite unique in that it has minimal ability to activate effector cells or fix complement and it undergoes Fab-arm exchange, a process by which IgG4 heavy-chain dimers dissociate from one another and bind to other IgG4 heavy-chain hemi-molecules, creating a dimer of heavy-chains with two different Fab regions.6 This impairs the ability of IgG4 to cross-link antigens or form immune complexes.7 IgG4 is associated with chronic antigen stimulation, and is hypothesized to play a role in neutralization of soluble antigens to prevent binding to IgE on mast cells.8

Around the turn of the millennium, a newly defined disease process emerged in the medical literature referred to as IgG4-related disease. The discovery of this new disease entity stemmed from initial findings of elevated serum IgG4 in patients with autoimmune pancreatitis.9 Over the last decade, it has emerged that IgG4-related disease is a systemic disease that can affect almost every organ including, but not limited to, the lymph nodes, thyroid, salivary glands, orbits, lungs and heart, singly or concurrently.9,10 IgG4-related disease has unique histopathologic features that can be seen in all sites of involvement, including: obliterative phlebitis, storiform fibrosis, a dense lymphoplasmacytic infiltrate rich in IgG4 positive plasma cells, and a variable eosinophilic infiltrate. Additionally, IgG4-related diseases are often associated with elevated serum IgG4.7,9,11

Since the identification of IgG4-related disease, IgG4 has been in the spotlight. The same attention has not been paid to IgG1-3. The studies that have been done to date on the role of the other IgG subclasses in human disease have either started with certain diagnoses and examined the levels of the IgG subclasses,12-18 or identified the IgG subclass of a specific autoantibody.12,19-24 The lack of investigations broadly exploring possible associations between levels of all of the IgG subclasses and disease processes leaves open the possibility that there are undiscovered associations between certain diseases and a given subclass. Given that the discovery of an association between autoimmune pancreatitis and elevated serum IgG4 was crucial to the identification of IgG4-related disease, we initiated this exploration for potential disease associations with elevations in the other IgG subclasses. This exploratory study starts to address that gap in the literature. It is unique in that it starts with patients with elevations in one of the IgG subclasses, and

broadly searches for associations with diseases. Using this unique approach, this exploratory study has the potential to uncover novel disease associations with elevations in any of the four IgG subclasses, which would then warrant further prospective study.

Methods:

Subjects were patients in the Brigham and Women’s Hospital (BWH) database whose serum levels of IgG subclasses had been measured, and who had an isolated elevation of one IgG subclass (the thresholds to define subclass elevations are shown in Table 2). For these identified individuals, we obtained medical records numbers and demographic information.

In all, 12,323 patients had undergone IgG subclass testing between 1987 and 2015 and 552 patients with an isolated elevation in one IgG subclass were identified and included in this retrospective chart review. IgG subclasses were determined by nephelometry, using kits from The Binding Site Group Ltd (Birmingham, UK).

The BWH online medical records include a list of diagnoses that is updated by BWH providers. Patients’ charts were systematically evaluated electronically, and diagnoses listed in their patient summary were recorded.

Subjects were grouped based on the single IgG subclass that was elevated (IgG1, IgG2, IgG3 and IgG4). Within each group, we tallied the number of patients with each

diagnosis. We reviewed the data (see Addendum) and selected various rheumatologic and immunologic and other diseases that appeared to have a skewed distribution as diagnoses of interest; diseases with previously described associations to elevated serum IgG4 served as controls.

Statistics:

For each diagnosis of interest we used the Fisher’s Exact Test to compare the prevalence of each diagnosis within each of the four IgG subclass groups. A p-value of <0.05 was used as a cut off of statistical significance. The observed distribution of each diagnosis was compared to the expected chance distribution to determine which IgG subclass group was overrepresented for a particular diagnosis.

This study was approved by the Brigham and Women’s Hospital Institutional Review Board (REB: #1999P001152).

Results:

Of 12,323 patients who had undergone IgG subclass testing at BWH, 552 were identified as having an isolated elevation in one IgG subclass. 317 patients had an isolated elevation in IgG1, 51 patients had an isolated elevation in IgG2, 62 patients had an isolated elevation in IgG3 and 122 patients had an isolated elevation in IgG4 (Table 1). The range

of elevated values observed in our study population, as well as the normal ranges for each subclass are shown in Table 2.

In our study the number of cases of isolated subclass elevation was unevenly distributed among the four subclasses (Table 1). The distribution of cases of isolated subclass elevations deviated from the typical proportion of total serum IgG comprised by each subclass in normal individuals. Specifically, only 9.2% of the cases of isolated subclass elevations were comprised by an isolated elevation in IgG2, even though IgG2 typically makes up 20-30% of the total serum IgG. Additionally, we observed 22.1% of the cases of isolated subclass elevations were isolated elevations in IgG4, even though IgG4 comprises only 1-4% of total serum IgG.

The demographic information is summarized in Table 3.

Women were over-represented among the 552 study subjects with isolated IgG subclass elevations. 61.5% (340 out of 552) of the study subjects were female. Further, the distribution of women varied across the subtypes (significant at p<0.001), with higher percentages of women with elevated IgG1 and IgG2, and lower percentages of women with elevated IgG4. This may be related to the fact that, unlike other autoimmune diseases, IgG4-related disease is more common in men.25

There was an uneven distribution of black subjects between the four groups (significant at p=0.005), while there was no significant deviation in the distribution of white,

Hispanic and Asian subjects. While blacks constituted 9.4% of the study population, they were over-represented among those with isolated elevations in IgG1 (12.6%), and markedly under-represented among those with isolated elevations in IgG4 (1.6%). It is unknown if this represents a real difference, or Type I error, as there is a lack of data regarding the racial and ethnic epidemiology of IgG4-related disease in the United States.

In our cohort, we found that autoimmune pancreatitis was significantly associated (p<0.001) with an isolated elevation in IgG4 (Table 4), an association that is already well established in the literature.7,9,10 Interestingly, the number of cases of autoimmune pancreatitis was also higher than expected in the group with isolated elevations in IgG2.

Hypothyroidism was significantly associated with isolated elevations in IgG2 (p=0.001), as was Irritable Bowel Syndrome (IBS) (p=0.042).

Hepatitis C and Monoclonal gammopathy (including Myeloma and MGUS) were associated with isolated elevations in IgG1 (p=0.019 and p=0.008, respectively). An association between Systemic Lupus Erythematosus (SLE) and isolated elevations in IgG1 approached significance (p=0.056).

Rheumatoid arthritis (RA) was associated with an isolated elevation in IgG1 (p=0.042), and an isolated elevation in IgG3 (p=0.010).

Celiac disease was associated with an isolated elevation in IgG4 (p=0.020). Aspirinexacerbated respiratory disease (AERD), nasal polyps, and eosinophilia were also associated with an isolated elevation in IgG4 (p=0.017, p=0.020, and p=0.038, respectively).

Discussion:

The discovery of IgG4-related diseases began with the discovery that autoimmune pancreatitis was associated with an isolated elevation in serum IgG4.9 The studies on IgG subclasses to date have predominantly started with certain disease diagnoses and examined the levels of IgG subclasses in patients with those diseases,12-18,26-28 or have identified the IgG subclass of a specific autoantibody.19-24 This study is unique in its approach as it starts with patients with isolated IgG subclass elevations and compares the distributions of the types of IgG subclass elevation across subjects with a given disease. The exploratory nature of this study has the potential to uncover novel disease associations with IgG subclass elevations, which could be the subject of further investigation in future studies.

The number of cases of isolated subclass elevation was unevenly distributed among the four subclasses (Table 1) and the distribution of cases of isolated subclass elevations deviated from the typical proportion of total IgG comprised by each subclass in normal sera (see Table 2). We observed 22.1% of the cases had isolated elevations in IgG4, while IgG4 comprises only 1-4% of total serum IgG. The high frequency of isolated elevations

in IgG4 observed in this study could be the result of selection bias, resulting from a high frequency of requests for IgG4 levels by physicians who suspect IgG4-related disease. Additionally, only 9.2% of the cases of isolated IgG subclass elevation were cases of isolated elevations in IgG2, even though IgG2 is the second most abundant subclass, typically making up 20-30% of the total IgG in normal sera. The low frequency of isolated elevations in IgG2 could likewise be the result of selection bias.

We found a significant association between autoimmune pancreatitis and an isolated elevation in IgG4 (p<0.001), a link well established in the literature.7,9,10 Similiarly, this study found AERD, nasal polyps and eosinophilia to be significantly associated with elevated IgG4, links which have also been established in the literature.29-31 These findings support the internal validity of this study.

AERD, nasal polyps and eosinophilia are conditions driven by the upregulation of Th2 cytokines.32,33 Th2 cytokines are responsible for stimulating IgG4 production, in addition to IgE production and eosinophilia. IgG4 production can be induced by chronic antigen exposures (the production of IgG4 often rises as IgE levels falls).7,34,35

The significant association between Hepatitis C with elevated serum IgG1 is in line with previous findings that IgG1 is selectively induced in patients with Hepatitis C.36

This study found an association between monoclonal gammopathy and an isolated elevated IgG1. Previous studies have determined that the incidence of IgG subclasses of

myeloma immunoglobulins is proportional to the distribution the subclasses in normal individuals.17,26,37 Given this, we would expect the majority of cases of monoclonal gammopathy to be found in the group with elevated IgG1 (as it makes up 60-70% of the total serum IgG). Of note, at our institution patients with suspected monoclonal gammopathies routinely have immunoglobulin quantification performed (IgG, IgA, IgM), but IgG subclass quantification is not routinely performed for these patients.

RA was associated with both an isolated elevation in serum IgG1 and an isolated elevation in serum IgG3. The existing literature on abnormalities in IgG subclass levels in patients with RA is inconsistent. Our findings were consistent with findings of a 1993 study that also found elevated serum IgG1 and IgG3 in patients with RA,38 however, a larger and more recent study published in 2010, found an increase in all four IgG subclasses in patients with RA compared to controls.14 Further complicating the picture is literature demonstrating that IgG1 and IgG4 are the predominant subclasses of autoantibodies to citrullinated antigens in RA.23 The inconsistent literature on IgG subclasses in RA highlights an area that merits further investigation.

In this study, IBS was significantly associated with an isolated elevated serum IgG2. While there is no well established link between IBS and abnormalities in the levels of IgG subclasses, in our search of the literature we did discover a 1997 study from Pakistan, which found elevated levels of serum IgG2 among patients with IBS and suggested that an infectious organism with a carbohydrate antigen could be implicated in the pathogenesis of IBS in their patient population.39 While our findings are consistent with

this interesting hypothesis, and highlight an area of interest that may warrant further investigation, we do not intend to overstate the significance of this observed association given the paucity of existing literature on this topic, and the exploratory nature of this study.

Hypothyroidism was found to be associated with an isolated elevated serum IgG2. There is literature demonstrating that autoantibodies in Hashimoto’s thyroid disease (HT) are predominantly IgG2.21,22 One study which investigated anti-thyroid peroxidase antibodies (anti-TPOAb) in patients with HT found a significant association between severity of hypothyroidism and the level of IgG2 anti-TPOAb.21 A second study found that IgG2 was the dominant subclass of anti-thyroglobulin antibodies (anti-TgAb) in patients with HT.22 A possible explanation of the role of IgG2 in HT is that the lymphocytic thyroid infiltrate in HT is primarily comprised of Th1 lymphocytes, and the production of Th1 cytokines can induce IgG2 expression.21,22,40 Further research is needed to determine if an elevated total serum IgG2 correlates with levels of IgG2 autoantibodies in this disease, but this finding of an association between hypothyroidism and an isolated elevated IgG2 highlights an area that could have diagnostic implications.

In this study, we also found a novel association between celiac disease and an isolated elevation in serum IgG4. While this is not an association found in the literature, there is literature suggesting that IgG4 plays a role in the immune response to many dietary protein antigens, and that there are a relatively high number of IgG4 producing cells found in Peyer’s patches in the gut,41 which may in part explain this finding.

The distribution of the IgG subclasses produced in an immune response depends on the type of antigen and duration of antigen exposure, a concept known as “subclass restriction.” Thus, elevations of a given IgG subclass in an autoimmune disease could be reflective of the nature of the underlying driving auto-antigen in that disorder. However, it should be noted that serum levels of IgG subclasses do not necessarily correlate with the amount of antibody deposition in the tissues, meaning that we cannot extrapolate that the elevated IgG subclass seen in the serum is necessarily involved in driving the disease process.

Another caveat to consider is that it is unclear if the normal serum cutoff values for each IgG subclass varies in different populations. At least one study suggests that the levels of serum IgG4 (which is associated with autoimmune pancreatitis) may be lower in Western populations than Asian populations.42 This study found that in a Western population, while the serum IgG4 was still higher in patients with autoimmune pancreatitis than other pancreatic pathologies, the levels of serum IgG4 seen in this Western population were lower than levels found in Asian populations, and in many cases were within normal laboratory limits.42

One limitation of this study is that it was limited to subjects with isolated elevations in one of their IgG subclasses. Those with elevations in more than one IgG subclass, whether due to a single process, or due to multiple concurrent processes, were not included. Furthermore, we were unable to obtain information about the timing of various

diagnoses compared to the timing of the IgG subclass analysis, and had no information of treatment history which could alter the level of IgG subclasses measured. Another limitation is our reliance on the patient summary to include a complete list of diagnoses. While these are supposed to be kept up to date by clinicians, errors or failures in documentation are possible and we did not independently verify the patient summary information. Additionally, we do not know why clinicians ordered the IgG subclasses, as the indication for ordering a given laboratory test was not routinely collected at our institution between 1987-2015. This could lead to a selection bias whereby patients with certain conditions are overrepresented in the study population. Clinicians may request serum IgG subclass levels for the following reasons: work-up of immunodeficiency in the presence of normal serum IgG; suspicion of the IgG4-related syndrome; or rarely, to define the IgG subclass of myeloma; among other possible reasons.

Lastly, it is possible that one or more of these results were due to chance and that this study was not adequately powered to detect significant differences in rare diseases. We intentionally did not do a Type I correction of our p-value cut-off for statistical significance, as in an exploratory study such as this, this correction would be very conservative. Of note, our use of the Fisher’s exact test is itself a conservative approach to inference, as it generally gives higher p-values than the Chi-squared test. Our preliminary findings indicate that further, more rigorous study of these potential disease associations is warranted.

Acknowledgements: We would like to acknowledge Lisa Bernhard, Supervisor Clinical Immunology Laboratory, Brigham and Women’s Hospital for her contributions to the project.

References:

1. Terry WD, Fahey JL. Subclasses of Human Gamma-2-Globulin Based on the Differences in the Heavy Polypeptide Chains. Science (New York, NY) 1964;146:400-1. 2. Grey HM, Kunkel HG. H Chain Subgroups of Myeloma Proteins and Normal 7S Gamma-Globulin The Journal of experimental medicine 1964;120:253-66. 3. Schur PH. IgG subclasses--a review. Annals of allergy 1987;58:89-96, 9. 4. Kapur R, Einarsdottir HK, Vidarsson G. IgG-effector functions:“the good, the bad and the ugly”. Immunology letters 2014;160:139-44. 5. Hamilton RG. Human IgG subclass measurements in the clinical laboratory. Clin Chem 1987;33:1707-25. 6. van der Neut Kolfschoten M, Schuurman J, Losen M, et al. Anti-inflammatory activity of human IgG4 antibodies by dynamic Fab arm exchange. Science (New York, NY) 2007;317:1554-7. 7. Kamisawa T, Zen Y, Pillai S, Stone JH. IgG4-related disease. The Lancet 2015;385:1460-71. 8. Hussain R, Poindexter RW, Ottesen EA. Control of allergic reactivity in human filariasis. Predominant localization of blocking antibody to the IgG4 subclass. J Immunol 1992;148:2731-7. 9. Hamano H, Kawa S, Horiuchi A, et al. High serum IgG4 concentrations in patients with sclerosing pancreatitis. N Engl J Med 2001;344:732-8. 10. Kamisawa T, Funata N, Hayashi Y, et al. A new clinicopathological entity of IgG4-related autoimmune disease. J Gastroenterol 2003;38:982-4. 11. Deshpande V, Zen Y, Chan JK, et al. Consensus statement on the pathology of IgG4-related disease. Mod Pathol 2012;25:1181-92. 12. Bijl M, Dijstelbloem HM, Oost WW, et al. IgG subclass distribution of autoantibodies differs between renal and extra-renal relapses in patients with systemic lupus erythematosus. Rheumatology (Oxford) 2002;41:62-7. 13. Kuroki A, Shibata T, Honda H, Totsuka D, Kobayashi K, Sugisaki T. Glomerular and serum IgG subclasses in diffuse proliferative lupus nephritis, membranous lupus nephritis, and idiopathic membranous nephropathy. Internal medicine (Tokyo, Japan) 2002;41:936-42. 14. Lin G, Li J. Elevation of serum IgG subclass concentration in patients with rheumatoid arthritis. Rheumatol Int 2010;30:837-40. 15. Lin GG, Li JM. IgG subclass serum levels in systemic lupus erythematosus patients. Clin Rheumatol 2009;28:1315-8.

16. Martini A, Plebani A, Ravelli A, et al. IgG subclass serum levels in juvenile chronic arthritis. Annals of the rheumatic diseases 1985;45:400-4. 17. Schur PH, Kyle RA, Bloch KJ, et al. IgG subclasses: relationship to clinical aspects of multiple myeloma and frequency distribution among M-components. Scandinavian journal of haematology 1974;12:60-8. 18. Zhang H, Li P, Wu D, et al. Serum IgG subclasses in autoimmune diseases. Medicine (Baltimore) 2015;94:e387. 19. Barnes RM, Harvey MM, Blears J, Finn R, Johnson PM. IgG subclass of human serum antibodies reactive with dietary proteins. Int Arch Allergy Appl Immunol 1986;81:141-7. 20. Husby S, Foged N, Oxelius VA, Scehag SE. Serum IgG subclass antibodies to gliadin and other dietary antigens in children wtih coeliac disease. Clin Exp Immunol 1986;64:526-35. 21. Xie LD, Gao Y, Li MR, Lu GZ, Guo XH. Distribution of immunoglobulin G subclasses of anti-thyroid peroxidase antibody in sera from patients with Hashimoto's thyroiditis with different thyroid functional status. Clin Exp Immunol 2008;154:172-6. 22. Caturegli P, Kuppers RC, Mariotti S, et al. IgG subclass distribution of thyroglobulin antibodies in patients with thyroid disease. Clin Exp Immunol 1994;98:464-9. 23. Engelmann R, Brandt J, Eggert M, et al. IgG1 and IgG4 are the predominant subclasses among auto-antibodies against two citrullinated antigens in RA. Rheumatology (Oxford) 2008;47:1489-92. 24. Maran R, Dueymes M, Le Corre R, Renaudineau Y, Shoenfeld Y, Youinou P. IgG subclass of human autoantibodies. Ann Med Interne 1997;148:29-38. 25. Stone JH, Zen Y, Deshpande V. IgG4 Related Disease. N Engl J Med 2012;366. 26. Kyle RA, Gleich GJ. IgG subclasses in monoclonal gammopathy of undetermined significance. The Journal of laboratory and clinical medicine 1982;100:806-14. 27. Shakib F, Stanworth DR. IgG subclass composition of rheumatoid arthritic sera and joint fluids. Annals of the rheumatic diseases 1976;35:263-6. 28. Yamamoto M, Takahashi H, Suzuki C, et al. Analysis of serum IgG subclasses in Churg-Strauss syndrome--the meaning of elevated serum levels of IgG4. Internal medicine (Tokyo, Japan) 2010;49:1365-70. 29. Szczeklik A, Schmitz-Schumann M, Nizankowska E, Milewski M, Roehlig F, Virchow C. Altered distribution of IgG subclasses in aspirin-induced asthma: high IgG4, low IgG1. Clinical and experimental allergy : journal of the British Society for Allergy and Clinical Immunology 1992;22:283-7. 30. Ebbo M, Grados A, Bernit E, et al. Pathologies Associated with Serum IgG4 Elevation. Int J Rheumatol 2012;2012:602809. 31. Aalberse RC, Van Milligen F, Tan KY, Stapel SO. Allergen-specific IgG4 in atopic disease. Allergy 1993;48:559-69. 32. Wenzel SE. Asthma phenotypes: the evolution from clinical to molecular approaches. Nat Med 2012;18:716-25. 33. Steinke JW, Wilson JM. Aspirin-exacerbated respiratory disease: pathophysiological insights and clinical advances. J Asthma Allergy 2016;9:37-43.

34. Aalberse RC, Stapel SO, Schuurman J, Rispens T. Immunoglobulin G4: an odd antibody. Clinical and experimental allergy : journal of the British Society for Allergy and Clinical Immunology 2009;39:469-77. 35. Nirula A, Glaser SM, Kalled SL, Taylora FR. What is IgG4? A review of the biology of a unique immunoglobulin subtype. Curr Opin Rheumatol 2011;23:119-24. 36. Musset L, Ghillani P, Lunel F, et al. Variations of serum IgG subclass levels in hepatitis C virus infection during interferon-alpha therapy. Immunology letters 1997;55:41-5. 37. Papadea C, Reimer CB, Check IJ. IgG subclass distribution in patients with multiple myeloma or with monoclonal gammopathy of undetermined significance. Annals of clinical and laboratory science 1989;19:27-37. 38. Outschoorn I, Rowley MJ, Cook AD, Mackay IR. Subclasses of immunoglobulins and autoantibodies in autoimmune diseases. Clin Immunol Immunopathol 1993;66:59-66. 39. Hussain R, Jaferi W, Zuberi S, et al. Significantly increased IgG2 subclass antibody levels to Blastocystis hominis in patients with irritable bowel syndrome. The American journal of tropical medicine and hygiene 1997;56:301-6. 40. Mazziotti G, Sorvillo F, Naclerio C, et al. Type-1 response in peripheral CD4+ and CD8+ T cells from patients with Hashimoto's thyroiditis. Eur J Endocrinol 2003;148:383-8. 41. Hvatum M, Scott H, Brandtzaeg P. Serum IgG subclass antibodies to a variety of food antigens in patients with coeliac disease. Gut 1992;33:632-8. 42. Hochwald SN, Hemming AW, Draganov P, Vogel SB, Dixon LR, Grobmyer SR. Elevation of serum IgG4 in Western patients with autoimmune sclerosing pancreatocholangitis: a word of caution. Annals of surgical oncology 2008;15:1147-54.

Table 1: Comparing the distribution of cases of isolated subclass elevations among the four subclasses to the normal proportion of total serum IgG comprised by each subclass.

% of total number of cases of

IgG1

IgG2

IgG3

IgG4

57.4%

9.2%

11.2%

22.1%

60-70%

20-30%

5-8%

1-4%

isolated subclass elevations Fraction of total IgG in normal sera (%)

Table 2: Ranges of observed sera levels of each of the IgG subclasses, as well as the normal ranges of sera levels of each IgG subclass.

Range of elevated values

IgG1

IgG2

IgG3

IgG4

930-5510

707-1580

177-2600

87-329

382-929

242-700

22-176

4-86

observed (mg/dl) Normal Range (mg/dl)

Table 3: Demographics of patients with isolated subclass elevations. IgG1

IgG2

IgG3

IgG4

Total

317

51

62

122

Female

68.1%

68.6%

51.6%

46.7%

Median Age

55 (3-96)

54 (24-90)

57 (25-94)

49 (6-88)

White

65.6%

70.6%

71.0%

63.1%

p=0.661

Hispanic

8.2%

5.9%

9.7%

9.8%

p=0.832

Black

12.6%

7.8%

9.7%

1.6%

p=0.005

Asian

3.2%

9.8%

3.3%

2.5%

p=0.099

Other/

10.4%

5.9%

6.3%

23%

p<0.001

p<0.001

(range)

unknown

Table 4: Distribution of IgG subclass elevation type across selected diagnoses. IgG1

IgG2

IgG3

IgG4

p-value

Total

317

51

62

122

Autoimmune pancreatitis

0

2⇧

0

6⇧

p<0.001

Hypothyroid

34

14 ⇧

2

13

p=0.001

Monoclonal Gammopathy

19 ⇧

1

1

0

p=0.008

RA

15 ⇧

0

4⇧

0

p=0.010

Aspirin-Exacerbated Respiratory

1

1

1

5⇧

p=0.017

Hepatitis C

17 ⇧

0

0

1

p=0.019

Celiac

6

0

2

9⇧

p=0.020

Nasal Polyps (all)

2

1

1

6⇧

p=0.020

Eosinophilia

1

0

0

4⇧

p=0.038

IBS

4

4⇧

1

2

p=0.042

SLE

11

0

0

0

p=0.056

Hematuria

2

0

0

4

p=0.101

Sjogren’s Syndrome

11

1

2

0

p=0.131

Disease (AERD)

Allergic rhinitis

32

3

7

17

p=0.467

JRA

5

0

0

0

p=0.614