Hans Tuppy ² , professor in Biochemistry at the University of Vienna, died 24 April 2024. He was my first mentor when I studied Medicine at the University of Vienna. Through my interest in Biochemistry, we developed a communicative relationship: he encouraged me to become a scientist. The last time I visited him in Vienna, he, as an experienced scientist, gave me the following strategic advice: “Ask critical questions provided they are based on reflections and science-based hypotheses, not on opportunistic assumptions or popular trends. Critical questions alter science.” This formed over years my intuitive approach to develop new hypotheses aimed, in my opinion, to increase our insight into the nature of the autoimmune syndrome systemic lupus erythematosus (SLE). This idea motivated my entire scientific life—but this mantra requires humbleness: science has profound impact and consequences in all levels of the society.

There is a comprehensible and thoughtful—almost precognitive—logic link between Hans Tuppy´s “Critical questions alter science” and Oscar Wilde´s ³ “The one duty we owe to history is to rewrite it.” History of SLE was shaped over the last 50 years ⁴ through trendy assumptions and associative definitions. This narrative is based on defense of established dogmas and (outdated? See below) conservative argumentations. We may anticipate that knowledge in a deeper sense is—and will remain—inconsistent. This is particularly true if causally based and insightful approaches are not in focus, as recommended according to established scientific rules. From these contemplative considerations, we need to rewrite history based on science funded by contemporary, insightfully based, critical questions that are aimed to be transformed into testable hypothesis, very much as a considerate incorporation of Tuppy´s and Wilde´s mantras.

In a biological context, these mantras may lead us to scientific hypotheses based on “The causality principle” (1–3): Pathogenesis as effect of a cause. In a systemic disorder, the effect of a cause may be transformed into further downstream alternating cause-effect processes like in blood coagulation (4) or complement activation (5). This reasoning is in fact a functional definition of the causality cascade [see below (2, 6, 7)].

A problem that is somewhat difficult to comprehend is that we need courage to ask critical and logical questions in order to confront “the scientific establishment and its dogmas” with new paradigms and new critical hypotheses. These consequent and courageous activities materialize the duty we owe to science as an imperative to rewrite its history and to rethink and consequently revise (and even abandon)! settled historical dogmas. We have still a long way ahead to go to gain the insight necessary to understand the deeper nature of SLE and this syndrome´s delimitations (8, 9).

The question formulated in the title is important and contains provocative elements linked to the definition of systemic lupus erythematosus (SLE). These are of fundamental importance due to their impact in selecting and generating SLE cohorts. They are basically aimed to serve as standardized SLE study objects. Such studies aim to describe problems linked to the definition of criteria that delimitate SLE versus “SLE-like non-SLE” syndromes. Most important is to link classification criteria with two causal response qualities of criteria that are not implemented as central elements in SLE research: “the causality principle” and the consequently extended and expanding process, “the causality cascade” (see below). Both aspects inherit the theorem “cause-effect” (effect is here synonymous with clinical symptoms or criteria). Today´s science, depending on criteria-based SLE cohorts, is from a causal point of view of questionable validity. Patients are classified by criteria that are step-wise selected based on knowledge, insight, and intuition expressed by an expert panel and then elected from the selected criteria pool [i.e., Delphi panel criteria identification process (10–12)]. However, concrete studies implementing cause-effects based on the causality principle (3, 6, 13) are not visible in the main SLE classification criteria versions—although individual criteria may be discussed in this context. These versions have not collectively identified criteria directly linked to one concrete, identifiable, and unifying causal process—and maybe this is not possible?

The title of this study therefore reflects a philosophical accession to iatrogenic ⁵ enigmas that adhere to our inferior understanding of the basic nature of SLE. Today, we have no clear strategy to approach sound theoretical (mechanisms/etiology) and practical (diagnostics versus classification, therapy) definitions of the syndrome (8, 14–16). The title is expressing an intellectual strategy aimed to unveil layer by layer of problems generated over decades of intense scientific activities and their results: We have to understand why we don’t understand SLE´s deeper nature! We have furthermore to dissect arguments for the idea that SLE is a syndrome with all its manifestations unified in “a one disease entity” paradigm (17)—as opposed to the SLE syndrome characterized by the non-stringent terms poly-phenotypic and poly-causal disorder(s), in other words, “SLE-like” or “SLE-like non-SLE” syndromes (18, 19). This conflictual situation can paradoxically describe SLE either as a distinct entity—”SLE as a single disease entity ⁶ “ or as a “group of SLE-like non-SLE disorders ⁷ .” Definitions in these contexts are vague and formulated on the basis of intuition and insight provided by experts in context of Delphi panels (10–12), but not on the basis of firm causality-based analyses and evidences.

SLE is today defined as a “one disease entity” (20) in the sense of Peter Hucklenbroich´s general definition of the term (21). There is, however, no consensus in this respect (3, 8). See, for example, an interesting description of 8 classified groups of 988 SLE patients in a flare study of Isenberg et al. (22). We are, pessimistically expressed, today not able to precisely delimitate the syndrome SLE from other mixed connective (SLE-like) diseases. The following discussion is problematizing our current views (whether formalized opinions or not) on SLE pathogenesis related to classification and diagnostics of SLE.

This study is aimed to extend, penetrate, and combine problematic elements discussed in two previous studies (3, 8). These elements need to be concomitantly handled to find answers to problems that contribute to the definition of SLE as “an enigmatic, autoimmune syndrome.” Central is to understand if SLE is “a one disease entity” or a “poly-causal and poly-phenotypic” syndrome, in other words, if SLE is a distinctive and delimitated entity. This discussion is in accordance with sound scientific rules: SLE is an effect incited by a (distinct)? cause, whether one or more. This conflict is not problematized in relevant literature. Therefore, it is not possible from existing classification criteria versions—and their attribution rules—to construct evidences against the hypothetic statement that SLE is poly-causal in nature. There is still a strong need to recapitulate how SLE classification criteria were generated, how they were validated, and which impact they have on the composition of SLE cohorts. These problems are discussed in relevance to the title “Why is it so difficult to understand why we don’t understand SLE.” Where necessary and important in the context of a holistic discussion, basic aspects of earlier studies will be referred and re-interpreted for an attempted completion of this argumentation. This will, for the purpose of a holistic completion of this argumentation, also include dogmatic statements and problems linked to the nature of anti-dsDNA antibodies and lupus nephritis pathogenesis. This means in sum that we have to comprehend disparate dogmas that adhere to SLE: cause-effect nomination of classification criteria; definition and impact of DNA structure-specific anti-dsDNA antibodies; the inner contradiction of anti-dsDNA antibody-induced lupus nephritis.

If we cannot consider the existence of a single dominant etiological factor, we cannot accordingly consider SLE as “one disease entity.” This semantic theorem will oblige us to attempt to separate, in forensic terms, “hard causal evidences” from “circumstantial indicators” in our search for the “guilty” dominant causal process(es), as discussed recently (8). To use forensic terminology for “evidence” in this context is a direct and instructive approach that can substantiate the problem: we have to search the physical footsteps of the guilty cause (i) to retrogradely identify the cause of effects, (ii) to separate SLE as “a one disease entity” from poly-etiological (and consequently poly-phenotypical) syndromes, and (iii) to harmonize the potentially identified causal process with new specific therapy modalities. In the context of SLE diagnostics, we have either to construct (yet by theoretical principles) homogenous SLE cohorts for detailed studies of the syndrome´s genesis—or we have to dismantle the current definition of the syndrome to recreate a syndrome in harmony with the causality principle: are the classification criteria causally linked to each other (as in a causality cascade), or do they reflect individual symptoms originating from individual but disparate causes? This is experimentally and analytically difficult to approach, but it may be important to give such a project a try—and not to remain with the pessimistic term “the enigmatic syndrome SLE.” The terms “causality principle” and “causality cascade” are essential in this context and are further discussed in detail below.

Single response elements (as criteria) adhere to and contribute to classify SLE. For example, the origin and impact of anti-dsDNA antibodies, lupus nephritis, lupus associated dermatitis, or cerebral lupus are for decades studied in detail with significant insight into their individual pathogeneses (9, 14, 15, 23, 24). A missing link in all these studies is if these central criteria are response elements linked to a common causal origin: a causality cascade. Are individual response elements encountered in SLE imposed by a common dominant cause (see e.g (14, 25–28), as exemplified by anti-dsDNA antibodies as a theoretical inducer of a causal cascade of organ manifestations (see below). These manifestations are defined as classification criteria but not as diagnostic criteria. This is a paradox that is maintained and paraphrased over decades. We encounter an intellectual conflict: anti-dsDNA antibodies are not causing SLE, and criteria are not responses to one causal origin.

It must, however, be stressed that the individual SLE classification criteria in general are not pathognomonic for SLE and consequently not diagnostic for SLE (8). In practice, clusters of criteria have been given the impact of being diagnostic since criteria clusters direct patients into SLE cohorts. These are organized as aims for studies of SLE-related scientific problems [ (3, 9); see also an insightful comment by Tsokos (29)]. This is a problematic—and a calculated paradox: SLE classification criteria classify SLE, but do not diagnose SLE—although SLE classification criteria practically endorse their impact as diagnostic criteria once a patient is enrolled into an SLE cohort!

It is of considerable importance to reconcile how SLE classification criteria were selected and implemented in SLE research [see, e.g (30–33), for details]. Two important problems are linked to the main SLE classification versions: it is not clear if these criteria are discussed in light of a defined or theoretical causal origin. Are they linked in a common fate destiny network based on a dominant cause, and what is the definition of SLE to which criteria were initially associated? This is a crucial point: Expert panels selected the SLE criteria based on intuition, instincts, knowledge, experience, and insight; they were subsequently statistically tested for association with SLE and the former criteria versions (31–33). This process can be formulated in the following self-confirming, but informative and provocative equation:

The central problem (and challenge) in this equation is how SLE (B in the equation) was defined in context of the Delphi Panel processes (see below) performed to identify SLE classification criteria. This intellectual problem may consequently imply that there are reasons both to trust and to mistrust the validity of current SLE classification criteria. Criteria are selected because they are instinctually believed to characterize SLE; therefore, SLE is associated with the criteria in a statistically significant way (30–34). Is this statistically supported philosophy productive, and does this philosophy reflect realities? We end up to ask: How is indeed SLE as a template defined in this important and critical classification process—and is “the causality principle” implemented in this definition—or not?

In the following, the four main SLE classification criteria will be summarized, critically commented on, and discussed.

With the 1982 ACR SLE classification criteria, autoimmunity became implemented among the 11 selected classification criteria (31). Four of 11 criteria are required to classify SLE. It is in the article (31) stated that “For the purpose of identifying patients in clinical studies, a person shall be said to have systemic lupus erythematosus if any 4 or more of the 11 criteria are present, serially or simultaneously, during any interval!” This is, in fact, in contradiction to the causality principle. According to the attribution rule, the presence of autoimmunity was definitively not required to classify a patient to have SLE. Figures 1B , 2 (upper panel) exemplify the same problem with the 1982 SLE classification criteria as with the 1971 SLE preliminary classification criteria ( Figure 1A , Table 1 ): one SLE patient has anti-dsDNA antibodies and lupus nephritis (red arrows), the other has other criteria (blue arrows). These alternatives challenge the dogmatic characterization of criteria-based cohorts as a homogenous “one disease entity” (see this principal problem demonstrated in Figure 2 , upper and lower panels)! A concise question is, furthermore, if patients characterized by criteria belong to a “mono-causal” disease entity. This doubt gave birth to reservations related to investigations of SLE-related nosology, genetics, etiology, pathogenesis, and experimental therapies in SLE cohorts. These dilemmas are not convincingly problematized in the relevant central reports (11, 30–33).

The Systemic Lupus International Collaborating (SLICC) classification criteria version (32) contains 17 criteria (11 clinical and 6 immunological, Figure 1C , Table 1 ). The final validation was performed by comparing the SLICC criteria with the modified ACR 1997 criteria (36). The SLICC criteria performed well with higher sensitivity but lower specificity and had a lower (but still a high) number of misclassified cases [74 vs. 62 for the 1997 revised ACR criteria and the final SLICC criteria, respectively (8, 32)]. In this context: What are the criteria used to define “misclassified,” when SLE still is regarded as an enigmatic autoimmune syndrome without clear definitions?

The final attribution rule for the SLICC criteria authorizes that 4 criteria are required out of the 17, which gives a very high total theoretical number of criteria combinations (see Figures 1C , 2 , lower panel). At least one clinical and at least one immunological criterion must be identified to classify a patient as having SLE. It is stated that criteria are cumulative and need not be present concurrently (32). This disconnects the criteria from the causality principle since a potentially pathogenic anti-dsDNA antibody may appear far from being associated with, for example, lupus nephritis [discussed in (24)]. One important exception from the attribution rule is that the presence of lupus nephritis with anti-DNA antibodies is sufficient to fully classify SLE; lupus nephritis and anti-dsDNA antibodies therefore serve as ad hoc classification and diagnostic criteria, although the idiom “diagnostic criteria” is not a formally accepted term for SLE. An implementation of the causality principle and of the causality cascade is not visible in the 2012 SLICC criteria version (32).

Leuchten et al. published in 2018 a systematic literature review and meta-regression analysis of data aimed to describe the performance of anti-nuclear antibody (ANA) for classifying SLE, with the purpose to consider if ANA justified a position as a mandatory entry criterion for SLE cohorts (38). Only screening methods for ANA were emphasized. This approach minimized a discussion on which ANA sub-specificities are covered by the screening assays. Spectra of ANA specificities may differ between assay systems (39, 40). Furthermore, concise information on the diagnostic and pathogenic impact of individual ANA sub-specificities was not given attention (33, 38). For example, anti-ssDNA and anti-dsDNA antibodies (both detected in ANA assays) differ significantly in diagnostic and clinical impact [discussed in (41)]. Here, the clinicians are in conflict with the immunologists.

According to concise theoretical insight, ANA are widely detected in infections (42), cancers [(43), see a discussion in (35) and references therein], autoimmune diseases/syndromes (44), and after ingestion of certain drugs (45). Thus, their manifestations in non-SLE contexts are substantial! On the other hand, Choi et al. observed that 6.2% of SLE patients were ANA negative (46). It may consequently appear problematic that the 2019 EULAR/ACR SLE classification criteria recommend a positive ANA test at least once as an obligatory entry criterion for SLE classification. This is accepted, although concise ANA specificities and assay principles are not a concern. It may in this context be relevant to emphasize that the American College of Rheumatology (ACR) ANA Task Force has recommended indirect immunofluorescence on HEp-2 cells as the gold standard test for ANA detection (44), see also (40).

Selection of the EULAR/ACR SLE classification criteria was performed through an international Delphi panel exercise, and several criteria implemented in previous classification criteria versions were reiterated (see details in Table 1 ). Furthermore, data from a diagnosed patient cohort and a patient survey [see details in Aringer et al. (33)] were considered. Criteria reduction was followed by Delphi and nominal group technique discussions, while criteria definition and weighting were based on criteria performance and on results of a multi-criteria decision analysis.

The final version of the EULAR/ACR SLE classification criteria contains 21 weighted criteria organized into 10 domains ( Table 1 , Figure 1D ). Manifestation of ANA is mandatory, and count as a criterion detected at any interval of observation. Definitive SLE classification requires at least one clinical and one immunological criterion, and weighted criteria ≥10 points, and a positive ANA entry criterion. The combination of the 21 criteria has as a clear consequence a manifold of SLE phenotypes (see a principle interpretation in Figure 1D , Table 1 ), as is the case for all the SLE classification criteria over the last 50 years (see details in Figures 1A–D , 2 ).

In contrast, Schmajuk et al. opened for the view that “one organ system would be sufficient for classifying SLE” (11). This may be beneficial for investigating SLE as a syndrome dominated by, for example, lupus nephritis, and may thus represent a basis for a homogeneous SLE cohort. If this suggestion refers to affection of any organ system, it may be problematic and transform the cohort into being heterogeneous and less valid in SLE research contexts.

Sensitivity and specificity of the EULAR/ACR criteria compared well with the 1997 revised ACR and the 2012 SLICC SLE classification criteria. This may, however, be anticipated due to the fact that many of the 1971 criteria are reiterated in the 1982 ACR (31), the 1997 revised ACR (36), the 2012 SLICC (32), and most recently in the 2019 EULAR/ACR SLE classification criteria [(33), see Table 1 for details]. The reiteration of criteria is substantial. The implementation of various clusters of weighted criteria does not, from principal reasons, support “the one disease entity “ paradigm attributed to SLE (17): The 2019 EULAR/ACR criteria may still identify SLE as a poly-etiological and poly-phenotypical syndrome (evidence for the opposite view has not been provided).

A critical problem is linked to the Delphi panel concept (11) and relies on the fact that the classified SLE patients suffer from a theoretically established and defined enigmatic disorder. SLE is in this context principally not delimitated from other “autoimmune enigmatic SLE-like” syndromes by clear definitions or by clear scientific data. There is till now no published evidence proving that the established SLE classification criteria are causally interdependent and interactive, in harmony with the causality principle and with the consequent causality cascade paradigms for complex syndromes (see below).

The SLE classification criteria were established as a systematic construction based on traditional thinking and anchored in 1971, a time with immature insight into autoimmune pathophysiology and SLE genetics. It is a striking observation that the causality principle was not debated in the 1971 preliminary and not in later classification criteria versions. Integration of the causality principle in relevant philosophical reflections anchored in scholarly principles have not been problematized over the years from 1971 to 2019, although causality has been a scientific concern over centuries, even back in antiquity (47). Thus, all the major classification criteria versions have principally the same problematic impact on causal investigations of SLE (30–33, 48). Why is it so?

Maybe we today have the knowledge sufficient to comprehend that the modern definition of SLE implies that SLE is not a single disease entity. Do we simply lack the courage to split the syndrome into different causal-related versions? These could be characterized as contrasts to the less distinct and enigmatic “overlapping lupus-like” disorders [discussed in (19)]. SLE may remain enigmatic until we investigate the syndrome in a complicated and cumbersome way in light of the causality principle, with the implementation of important and highly relevant central historical data (see anti-dsDNA antibodies and lupus nephritis discussed below). These data have been revitalized recently (3, 8, 41, 49). The causality principle may be the single sine qua non distinction that may enable us to intuitively and scientifically designate diagnostic criteria and to sub-fractionate the diagnosis of SLE by different unique cause-effect relations. If the causality principle is included as an understatement during the development of classification criteria, this should have been specified and precisely described.

We have to strive to define the distinction between “SLE” and “SLE-like non-SLE” diseases. There is no unifying conceptual quality expressed by clustered classification criteria that can provide, advance, and practically inform about the integrated complex mechanistic nature of SLE. Exceptions are given for some rare, well-described gene deficiencies or deviations (50–57). Such gene deficiencies are principally divided into two groups, one caused by a single gene (monogenic) defect, the other by polygenic defects. The monogenic SLE version is easier to understand—a single gene defect forms the basis for one basic pathophysiological aberration that in turn may promote downstream alternating cause-effects. In other words, monogenic SLE is a potential example of induction of a causality cascade where symptoms (or effects or criteria) are offspring of the original monogenic defect (53–57). Wild-type SLE—or “SLE-like non-SLE”—syndromes are not comparatively easy to comprehend. This group of disorders may rely on polygenic defects and an interplay between the genetic and environmental factors. This is a more complex picture and is still difficult to understand and to delimitate. A simple consequence will be that classification criteria promoted by a single gene defect truly belong to a causality cascade and are synchronized, interactive, and interdependent, while classification criteria promoted by polygenetic and environmental factors may appear as criteria clusters timely unlinked from each other, have poly-causal origins, and do not reflect effects consistent with the causality cascade paradigm.

Monogenic SLE and “wild-type” SLE conditions may be valid models to compare repertoires of early and late SLE classification criteria appearing in the two separate conditions. Such studies are still awaited [critically discussed in (3, 8, 49)], and may have a strong impact on new therapeutic modalities like the implementation of chimeric antigen receptor T cells (CAR T cells).

Wild-type SLE in this context may be a synonym for poly-phenotypic (and possibly poly-causal) SLE. How does this comply with SLE cohorts established by criteria that may be unlinked and not appear as interactive, as opposed to models involved in the causality cascade paradigm? Based on this dilemma, we may question if such cohorts are useful as study objects to investigate if, for example, genetics, etiology, pathogenesis and experimental therapy modalities comply with “a one disease entity” paradigm [discussed in (3, 8, 9)].

There are, however, pathogenic situations where new promising therapeutic modalities have been described (58, 59). Therapies developed using classification criteria for patient inclusion have been outstandingly successful for the chimeric antigen receptor T (CAR T) cellular therapy. The premises for successful treatment of polyphenotypic and polycausal SLE variants characterized by autoimmune organ manifestations are if immune tolerance can be restored. T cells that are modified to express chimeric antigen receptors on T cells (CAR T cells) have been investigated. CAR T cells target B cells to selectively deplete or to down-regulate autoimmune responses responsible for systemic or organ-specific autoimmune disorders (discussed in e.g (58, 59). In this scenario, CAR T cells may be effective in patients with various autoimmune B cell activities (58) or in refractory SLE (59); that is, whether they produce a single pathogenic organ-specific autoantibody or multiple autoimmune antibody specificities. This is a very promising therapeutic progression and may work well irrespective of whether patients suffer from mono- or poly-causal SLE variants. Future research efforts should, from this observation, analyze if CAR T-cell therapies are successful in genuine SLE (like monogenic or monocausal SLE) as well as in disparate autoimmune SLE-like non-SLE situations.

There are many reasons to assume why classification criteria may reduce our understanding of the nature of SLE. A steadily increasing number and clusters of SLE classification criteria will not provide us with more science-based insight into the core process(es) that promote and maintain SLE.

A central question in this context needs profound considerations: What is the impact, in biological terms, of concrete and disparate criteria clusters (defined as minimum required number of criteria) in different SLE patients? Do different clusters reflect the same etiology-based pathogenic process? Is it possible to solve this problem when we must consider that the criteria count irrespective whether they appear simultaneously or one by one uncoupled in time [the accumulative criteria model (31–33)], and more problematic, do different criteria clusters irrespective of their composition provide us with information applicable to quite different SLE phenotypes? In other words, do different criteria clusters reflect a common cause?

An idealized reflection: SLE classification criteria may appear coincidently as disparately composed clusters that represent consequences of a single cause. Criteria clusters as mirror images of a causality cascade may reflect a downstream alternating cause-effect network introduced by a primary single cause. If so, we have to understand why one cause may promote quite different criteria clusters that involve different organ manifestations from patient to patient. Is this realistic scenery? Or do we have to re-think our strategies for selection of cause-related criteria?

Are then all these clusters ultimately unique parts of “genuine” SLE? This critical question opens for the intriguing possibility that SLE indeed is representing “a one disease entity” despite its polyphenotypical nature. To solve this eventuality, it is needed to perform, for example, retrograde investigations to search the causal origin of individual criteria and the origin of the inciting cause(s).

In line with these reflections, an association of oligo- or poly-phenotypic SLE with a single gene defect may promote the idea that the interrelated criteria are timely synchronized in contrast to what would be likely by chance alone. Secondly, a monogenic origin of SLE is expected to result in syndromes with a lower degree of criteria cluster variability than variability observed in “wild-type” SLE [discussed in (55)]. Is it possible to define differences between monogenic SLE and SLE promoted by an interplay of disparate causes? A metaphor, SLE disposes for infections (60–64), and infections may directly promote, for example, dsDNA/chromatin autoimmunity (65–71), which in turn may cause subsequent inflammatory effects (as criteria), consistent with the causality cascade paradigm (see an example of a causality cascade in the theoretical Figure 3 ). A confusing question is whether infection promotes an additional cause for SLE or if infection is caused by an underlying SLE-disposing cause. We still encounter fundamental problems inherited in the SLE classification criteria versions that make SLE difficult to understand, irrespective whether of mono- or polygenic or other origins.

A similar problem is relevant for ANA, a term that implements unspecified anti-nucleosome/anti-chromatin autoimmunity. ANA is required as an entry criterion in the 2019 EULA/ACR SLE classification criteria and is relevant in any SLE or SLE-like conditions, which all may be difficult to delimitate toward each other. These autoantibodies are frequently detected as drug-induced non-pathogenic, infection-induced, or cancer-associated transient or sustained autoimmunity [(27, 43, 45, 71), discussed in (27)]. It may be problematic that unspecified ANA is weighted as a mandatory entry SLE classification criterion. But, if present, it may add to the autoimmune character of SLE in a patient.

Depending on the answers to these questions, we need to generate new radical hypotheses that reflect implementation of the causality principle (1, 3, 13) to understand, to diagnose, and to treat SLE. If we disregard this approach, we will return to the “pre-Robert Koch and Louis Pasteur era,” an historical era characterized by studying symptoms and “intuitive, philosophy-based therapy modalities,” without indication of causal hypotheses (72). The Koch-Pasteur era is characterized by a radical paradigm shift in medical science, and is the starting point where studies of symptoms transformed into studies of their causes (72)! This opened for causal therapy principles and for vaccination to protect against causative contagious infections (72). From the current state of understanding, we need a new open-minded discussion and new critical hypotheses based on the Koch-Pasteur derived causality principles to explain and treat SLE.

The causality principle (2, 6) is the most basic principle we have to consider in our search for a phenotypic time-kinetic (or cause-effect) definition of SLE [discussed in (3, 49)]. The causality principle in its simple version declares the relation between a cause and its effect (response element). The principle has impact on several levels relevant to SLE research (73).

Examples of causality cascades that are operational in

 o Coagulation (4)

 o Complement activation (5)

 o Anti-dsDNA antibody induced complex causality cascade [(3), see theoretical example in Figure 3 ].

The principal causality cascade model is not reported in the central classification criteria publications [see, e.g (17, 30–33, 36, 48, 75, 76)]. In these publications, the Delphi panel approach (10, 11) has taken over for the functional causality approach. The Delphi panel approach has by its instinctual nature hindered causality from being the theoretically most central element in SLE classification processes.

This pessimistic view is further underscored if we observe that there is no radical paradigm shifts if we study and compare the four major SLE classification criteria versions with each other; many of the criteria are reiterated ( Table 1 ), and attribution rules are principally common for all of them.

The paradigm “one cause - one effect” may at first glance be irrelevant for the complexity of the SLE syndrome. There is actually one interpretable solution to this complex situation. If we, for example, study the clinical involvement of anti-dsDNA antibodies as a functional cause, we may be able to trace different apparently incongruent response elements primarily and secondarily imposed by the anti-dsDNA antibodies. As outlined in Figure 3 , the anti-dsDNA antibody—as a causal factor—promotes central downstream criteria with high impact in SLE [for discussions, see (3, 27, 35) and references therein]. These central criteria, incited by one causal factor, practically exemplify a causality cascade. “Causal structure” is the relevant term to be used here (7).

This paradigm of a mono-causal, poly-phenotypic SLE is in analogy with, for example, the causality cascade that characterizes a poly-phenotypic syndrome incited by SARS-CoV-2; the long COVID-19 syndrome. Therefore, the COVID-19 is regarded as “the new great imitator.” (77) ¹⁰ . In a clinical context, this syndrome is poly-phenotypic, but basically mono-causal, as the virus is the sole initial causal factor. The virus as the primary disease-promoting factor may, in certain situation, initiate a cascade of downstream response elements that account for the polyphenotypic character of the long COVID-19 syndrome.

“The anti-dsDNA antibody” possesses characteristics that comply with the problem proclaimed in the title of this study. Anti-dsDNA antibodies are enigmatic, origin not conclusively defined, specificities for DNA structures have been disregarded, and their role as diagnostic marker inconclusive. The anti-dsDNA antibody holds an archaic position as a central classification, diagnostic and pathogenic factor in SLE. For decades, the antibody had a strong influence on SLE classification, diagnostics, and autoimmune pathogenesis. These aspects have been substantially modified in recent years. Despite this strong focus, historically important information on DNA structures targeted by highly specific antibodies have been neglected—even in today´s modern studies on SLE.

This part of the study concentrates on three unique and central issues, communicated here as questions: (i) Does an antibody specific for dsDNA unequivocally reflect the diagnosis SLE and the pathogenic process(es) in SLE? (ii) Does the antibody separate SLE from “SLE-like non-SLE syndromes?” (iii) Does “the anti-dsDNA antibody” represent one specificity, or disparate DNA structure-specific antibodies? These are fundamental questions we are not able to answer today because relevant scientific data have been neglected in the history of immunogenic DNA structures and SLE [discussed in (41)].

There are three problematic, unresolved, and poorly investigated aspects that adhere to anti-DNA antibodies: (i) their distinctive structural DNA specificities (41); (ii) Their clinical impact (27); and (iii) What do the antibodies recognize in kidneys in context of lupus nephritis—exposed chromatin fragments or inherent cross-reactive basement membrane structures [discussed in detail in (24)].

Conflicting scientific data on the nephritogenic effect and modus operandi of anti-dsDNA antibodies have been available over decades but hardly implemented in appropriate scientific protocols (24, 27). Still, relevant historical data are either neglected or regarded as being of low importance in central scientific studies.

Today, the anti-dsDNA antibody is regarded as a unique biomarker for SLE and has reached a position as an unquestionable immunological dogma [discussed in e.g (25, 27, 28)]. There are strong reasons to critically revise the impact of this dogma from a strict scientific point of view (41). In fact, inadequate integration of classical knowledge and central historical data may explain why we still adhere the term “enigmatic autoimmune syndrome” to SLE (9, 14, 15, 17, 23, 78). This is discussed below point by point.