Is immunology doing well? A look at 100 immune‐mediated inflammatory diseases for 100 years of the Journal

It is now 60 years since Ian Mackay and Macfarlane Burnet published their seminal text “The Autoimmune Diseases” in which they examined the full scope of human inflammatory pathology as a manifestation of the underlying structure and function of the immune system. Here I revisit this approach to ask to what extent has the promise of Mackay and Burnet's work been exploited in clinical medicine as currently practiced. In other words, is immunology doing well? Despite spectacular headline contributions of immunology in clinical medicine, I present evidence suggesting a performance ceiling in our capacity to answer the relatively straightforward questions that patients frequently ask about their own diseases and find that this ceiling exists across almost all of the 100 immune‐mediated inflammatory diseases examined. I propose that these questions are difficult, not so much because the immune system is overwhelmingly complex but rather that we have more to learn about the relatively simple agents and rules that may underpin self‐organizing complex interacting systems as revealed in studies from other disciplines. The way that the immune system has evolved to exploit the ancient machinery determining three independent cell fate timers as described in this Journal would be a great place to start to decode the self‐organizing principles that underpin the emergent pathology that we observe in the clinic.

Physicians of any medical specialty who encounter patients with immune and inflammatory pathology are presented with one of the great challenges in medicine.Immunology at its best offers the potential to infer the immune mechanisms that sustain tissue injury in an individual patient and to select personalized immune interventions that are both efficacious and safe.The pipeline of approved immune interventions is wonderfully rich, with more than 70 molecular targets within the immune system for which at least one drug has achieved regulatory approval.The clinical task is to choose personalized precisely targeted immune interventions, alone or in combination, achieving high efficacy and safety for the individual affected.Can we actually do this?
It is now 60 years since Ian Mackay and Macfarlane Burnet published their seminal text "The Autoimmune Diseases," in which they examined the full scope of human inflammatory pathology as manifestations of the underlying structure and function of the immune system. 1 Mackay, in his reflective paper published in Immunology & Cell Biology, described the ambition to consider forbidden clones of immunocytes as agents of tissue damage, the somatic mutations that generated such clones and homeostatic mechanisms that controlled them 2 (Figure 1).
Those of us who study immunology in clinical medicine appear to have lost that ambition to describe the universe of immunopathology in relation to the inherent logic of immune responses.The reasons are unclear.Here I revisit this approach to ask to what extent has the promise of Mackay and Burnet's work been exploited in clinical medicine as currently practiced.In other words, is immunology doing well?
Of course, immunology is doing very well indeed.The global impact of coronavirus disease 2019 (COVID-19) vaccination is estimated to have averted 19.8 million deaths in its first year. 3Immune checkpoint inhibitors such as anti-CTLA-4 and anti-PD-1 can achieve profound benefits for many patients with certain cancers. 4himeric antigen receptor T cells are being engineered to treat resistant malignancies. 5Depletion of mature CD20 + B cells by specifically targeted monoclonals has produced profound gains in the treatment of diseases as diverse as multiple sclerosis and autoimmune blistering skin diseases.Cytokine inhibitors such as those targeting tumor necrosis factor or type 2 helper effector molecules are widely used in inflammatory and atopic pathologies.Many new autoimmune diseases are being discovered, particularly in the field of neuroimmunology. 6New mechanisms of disease are being uncovered through the study of inborn errors of immunity, 7 and immune-related adverse effects of checkpoint inhibitors 8 among others.
Although these headline achievements are profound and ongoing, they distract from a general sense of underachievement in addressing a naggingly difficult set of questions (Table 1).These are the kind of questions that patients or their families might ask at their first appointment.To examine these questions, I have taken the approach of Mackay and Burnet to look at the scope of human immune pathology by listing 100 immune-mediated inflammatory diseases (IMIDs; Figure 2) and see where and when these questions can be clearly answered.In judging the score, I have taken the more demanding perspective of trying to answer the questions in relation to an imaginary individual with one of these diseases, rather than a population of patients.I have taken this approach because it more closely matches the clinical challenge, and it deals with the concern that individual patients with a named syndrome may well have found their way to that diagnosis by a genomic or environmental path that differs in meaningful ways from the patient with the same diagnosis sitting next to them in the waiting room.My list deliberately includes a wide sample of disparate IMID syndromes, not all of which would be considered strictly autoimmune, with the hope of approaching different aspects of the underlying logic of immunity.

HOW DO I KNOW WHETHER MY IMMUNE SYSTEM IS WORKING NORMALLY?
It is perhaps disturbing that every other medical specialty has a suite of bedside and laboratory or technology-driven tests to examine whether the organ system of concern is healthy and functioning normally.Does immunology have any predictive immune function tests?The Autoimmunity Centers of Excellence Immunocompetency Committee orchestrated a meeting to develop guidelines for assessing immunocompetency in clinical trials for interventions in autoimmune diseases.While many tests were considered, the key conclusion was that an increased rate of infections remains the gold standard for determining immunocompetence. 9Unfortunately, infections are a universal experience of humans, periodic and highly variable in severity.Such a measure of immune health ignores many other potential types of pathology that might arise from dysfunctional immunity.Of course, it is indeed possible to measure many immune parameters such as serum immunoglobulin levels, lymphocyte subsets, inflammatory markers, allergy tests and mitogen responses but their relationship to clinical end points of importance is most often frustratingly imprecise.The outstanding exception to this is the utility of peripheral blood CD4 + T cells in HIV medicine 10 and its relationship with progression to acquired immunodeficiency syndrome (AIDS) and death.Why this finding is the exception, rather than the rule, is worthy of study.In short, we do not have personalized measures of immune health that are robustly predictive of the outcomes that matter such as infections, immune-inflammatory disorders or cancers.

WHY DID I GET THIS DISEASE AND WHY NOW?
This question is almost always impossible to answer in any way that is satisfying to either the patient or the Table 1.Patient-centered questions about immune-mediated inflammatory diseases and our current capacity to answer in a personalized and clinically meaningful way.

Questions
Estimate of the proportion diseases where a clear and obvious answer has been defined 1: How do I know whether my immune system is working normally?
No effective model 2: Why did I get this disease and why now? < 5% 3: Could my disease have been prevented before it occurred?< 5% 4: Do we have any biomarkers that predict whether my disease is about to flare?< 10% 5: Does this disease have approved personalized treatments of high efficacy for me? < 20% 6: Can my disease be cured?< 5% Data represent estimates of the proportion of 100 immune-mediated inflammatory diseases where a clear and obvious answer has been defined.physician.IMIDs can be said to arise from a combination of genes and environmental influences. 11For some IMIDs, infections clearly play a part (such as hepatitis C-associated cryoglobulinemia, rheumatic fever, hepatitis B-related polyarteritis nodosa) but these infections typically infect the great majority of other patients without causing these syndromes.Single-gene inborn errors of immunity illuminate certain cases but are rare within the scope of disease.We lack any sense of being able to track back to where things went wrong for the patient in front of us in the great majority of cases, even though such an insight might propel us toward the personalized interventions we wish to introduce.

COULD MY DISEASE HAVE BEEN PREVENTED BEFORE IT OCCURRED?
Excluding disorders linked to infectious agents, I could identify no disease that could be prevented before it occurred with current technology.Those due to single-gene disorders could be prevented by early genomic diagnosis and gene editing in the future.It is generally difficult to conceive a way in which this prevention might be achieved even had we known that disease was likely in the years to come.Rheumatic fever and postinfective syndromes may be prevented by public health measures, 12 but not in the personalized immunedirected way that might be imagined in the question.

DO WE HAVE ANY BIOMARKERS THAT PREDICT WHETHER MY DISEASE IS ABOUT TO FLARE?
Inflammatory markers such as serum C-reactive protein are very commonly used in some but not all diseases as a measure of disease activity, but it is more difficult to say anything precise about the future.Such tests, if they could be found, would be of great clinical importance as they would offer the opportunity for preemptive change of therapy.There are encouraging examples such as DNA antibodies in systemic lupus erythematosus, but clinical and immunological features have shown inconsistent results to identify patients with systemic lupus erythematosus at risk of flares in different cohorts. 13Serum levels of muscle-specific tyrosine kinase (MuSK)-immunoglobulin G 4 (IgG 4 ) autoantibodies may predict relapse of MuSK myasthenia gravis in a finding that needs confirmation in larger studies, 14 but serology is generally less predictive for acetylcholine receptor (AChR) antibodies in AChR myasthenia.The reasons for this difference may be more generally instructive.

DOES THIS DISEASE HAVE ANY APPROVED PERSONALIZED TREATMENTS OF HIGH EFFICACY FOR ME?
While most IMIDs can be expected to respond to immune interventions, high-level evidence is available for only a few, and benefits tend to be incremental rather than remission inducing.Achieving on-label approval status by diagnosis is a major concern.The current approaches to pharmaceutical reimbursement do not take into account the immune mechanisms of disease, leaving a high proportion of patients with IMIDs in the uncertain space of off-label prescribing. 15Further, the idea that we can preemptively choose therapies based on a personalized identification of the precise immune mechanisms sustaining tissue injury is generally currently beyond us, although the subject of intense study.For this reason, even in those diseases where high-level evidence has been established, the number needed to treat for benefit remains uncomfortably high. 16

CAN THIS DISEASE BE CURED?
While the prognosis of IMID disorders has improved remarkably over recent decades with the use of biological and other disease-modifying therapeutics, prolonged drug-free remission is unfortunately still rare.Where the disease is due to persistent infection such as hepatitis C virusassociated cryoglobulinemia, a cure can be established by treatments directed at the infection.Long-term treatment-free remissions can be achieved in certain patients with antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis or multiple sclerosis, but often the clinical ambition is to suppress disease activity until such time as the disease might abate of its own accord.Notably, some of the disorders such as Guillain-Barr e syndrome, acute disseminated encephalomyelitis and reactive arthritis can be spontaneously remitting, and others such as neuromyelitis optica spectrum disorders or chronic relapsing pericarditis may follow courses that include relapses after extended periods of inactivity.These patterns, once informed by an understanding of the immune mechanisms that drive them, may lead to treatment strategies for those diseases that are more relentlessly persistent.

DISCUSSION
This simple score report makes sobering reading.By asking these plain clinical questions across the scope of IMIDs, it is reasonable to conclude that, despite spectacular headline advances, we have a performance ceiling that limits the effectiveness of current immunology reasoning in clinical medicine.In the 60 years since Mackay and Burnet's survey of autoimmune diseases, we have had very few wins in answering these questions.
So, why are these questions so hard to answer?One important possibility is that we have simply not yet discovered enough, or that too many randomized controlled trials remain yet to be conducted.However, using Mackay and Burnet's insight that the scope of autoimmunity must arise from the underlying structure and function of the immune system, it is worth engaging the idea that the demonstrable difficulties we have in answering these questions are themselves instructive about the complex internal logic of immunity as a system.Any system described by linear chains of causality should be much easier to unpick, and despite being the dominant way we teach immunology to others, linear system logic is probably a poor description of immunity.Any more accurate system description must take into account the great robustness of immune systems in achieving core outputs such as the generation of isotype-switched immunoglobulin responses to new viral exposures such as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) or HIV.IMIDs are the relatively rare exceptions.The system must account for the spectacular macroscale decentralized self-organization we see in both the preimmune system and the remarkable functional perturbations that occur following antigen engagement such as germinal center formation.We would do well to note that complex interacting self-organizing systems outside of immunology 17 can often be better understood through relatively simple definitions of core agents and the rules governing their interactionsinteractions that occur locally and often include stabilizing use of randomness.
The idea of thinking of immune pathology as a fundamentally nonintuitive emergent property of a complex interacting self-organizing system operating with an element of stochasticity provides some relief from the burden of underperformance in answering these clinically important questions.For example, such a system would predict that macroscale behavior cannot be understood by considering single agents in isolation, suggesting that biomarkers can be of value, but only if multivariable, personalized and linked to the underlying cellular logic.Such a model would predict that small initial variations in agents and their interacting rules could result in major macroscale errors, perhaps explaining why causes are so frustratingly obscure.Immunology is in great need of a better working model to fit the spectrum of observable diseases.
Fortunately, there are advances described in Immunology & Cell Biology and elsewhere that enlighten exactly such a model [18][19][20][21] and build upon the clonal selection breakthroughs described by Mackay and Burnet. 1 These advances define an ancient cellular system of stochastic intracellular competition that explains the assignment of lymphocytes to their fates as well as predicts their behavior under in vitro stimulation conditions. 21n this basis, it is reasonable to propose that next-generation advances in clinical reasoning and personalized immunology will rely on our capacity to exploit those discoveries to construct the in silico models required to integrate data from the clinic.
If the immune system is built this way, there is no alternative.Through a computational model, we can seek to reimagine IMIDs away from the eponymous syndromes we have inherited from our forebears and toward a personalized description of the targetable immune mechanisms from which they arise.

Figure 2 .
Figure 2. A list of 100 immune-mediated inflammatory diseases representing a spectrum of human diseases arising from defined or undefined underlying immune mechanisms.
Front Bioinform 2021; 26: 723337.20.Horton MB, Cheon H, Duffy KR, et al.Lineage tracing reveals B cell antibody class switching is stochastic, cell-autonomous, and tuneable.Immunity 2022; 55: 1843-1855.21.Horton MB, Hawkins ED, Heinzel S, Hodgkin PD.Speculations on the evolution of humoral adaptive immunity.Immunol Cell Biol 2020; 98: 439-448.22. Roberts-Thomson PJ, Jackson MW, Gordon TP.A seminal monograph: Mackay and Burnet's autoimmune diseases.Med J Aust 2012; 196: 74-79.ª 2023 The Authors.Immunology & Cell Biology published by John Wiley & Sons Australia, Ltd on behalf of the Australian and New Zealand Society for Immunology, Inc.This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.