ANA and Dengue — A Synthesis Review (V2.0)

Synthesised from all 37 source papers in the wiki. Ground-up rewrite; supersedes the incrementally revised V1.x series (2026-04-13 through 2026-04-18). Version 1.x is preserved at ANA and Dengue - A Literature Review.

Research Question

What does dengue virus infection do to the antinuclear antibody (ANA) profile — and does it do so in a pattern that is clinically meaningful, mechanistically distinct, or prognostically significant? For whom, through what mechanism, and for how long?

Sources Used

Synthesis

Part I — Methodological Foundations

Before any dengue data can be interpreted, the assay framework must be explicit, because cross-study comparisons in this literature are routinely invalidated by substrate and dilution mismatch.

Substrates. The gold standard for ANA detection is indirect immunofluorescence (IIF) on HEp-2 cells — continuously dividing human epithelial cells that express the full human nuclear antigen repertoire at high density. Older studies used rodent liver tissue (rat or mouse), which detects a narrower range of specificities and systematically underestimates ANA prevalence relative to HEp-2. Garcia2009 - Long-term Clinical Symptoms Post-Dengue, the principal post-dengue ANA study in this wiki, used rat liver — meaning its 23.1% figure is a conservative floor. Any comparison between Garcia2009 and a HEp-2-based study (e.g., Chatterjee2024 - ANA Detection in Dengue Kolkata) is directionally valid but the true HEp-2-equivalent rate in the Garcia2009 cohort is unknown and likely higher. See Indirect Immunofluorescence ANA Test.

Dilution dependence. ANA prevalence falls sharply with increasing dilution: 31.7% at 1:40 → 13.3% at 1:80 → 5.0% at 1:160 → 3.3% at 1:320 in the Tan1997 multicentre study (see Tan1997 - ANA Range in Healthy Individuals). The 2019 EULAR/ACR SLE classification criteria (see Aringer2019 - 2019 EULAR ACR SLE Classification Criteria) codify ≥1:80 on HEp-2 as the mandatory entry criterion for SLE classification (sensitivity 97.8%, meta-regression 13,080 patients, 64 studies). Studies not specifying dilution thresholds cannot be meaningfully benchmarked.

IIF vs. LIA. IIF detects all antinuclear reactivities including low-affinity, germline-encoded polyreactive IgM. Line immunoassay (LIA) tests a defined panel of disease-associated autoantibodies (Sm, dsDNA, Ro/La, Scl-70, U1-RNP, Jo-1, CENP-B, PM-Scl, etc.). The gap between IIF and LIA rates is mechanistically informative in dengue: a high IIF rate alongside a much lower LIA rate identifies a large non-specific fraction (see §3.2 below). See Line Immunoassay ANA.

ELISA panels. Berlin2007 - Autoantibodies in Nonautoimmune Individuals during Infections measured ANA by an 8-antigen ELISA (ANA 8 Pro), yielding a control rate of 3.8% — lower than any IIF-based estimate. The fold-change relative to controls (~5.7×) is the interpretable quantity from Berlin2007, not the absolute rate.

Part II — Reference Baselines

2.1 Healthy-Population ANA Prevalence

Study	Population	Dilution / Method	Prevalence
Tan1997 - ANA Range in Healthy Individuals	International multicentre; adults 21–60 yrs (n=125)	1:40, IIF	31.7%
Tan1997 - ANA Range in Healthy Individuals	As above	1:80, IIF	13.3%
Tan1997 - ANA Range in Healthy Individuals	As above	1:160, IIF	5.0%
Satoh2012 - ANA Prevalence in United States	US NHANES 1999–2004; ≥12 yrs (n=4,754)	1:80, IIF	13.8%
Li2019 - ANA Epidemiology in Chinese Healthy Population	Chinese health-checkup; all ages (n=25,110)	>1:100, IIF	14.01%
Dinse2022 - Increasing ANA Prevalence in United States	US NHANES 2011–12; ≥12 yrs	1:80, IIF	16.1%

Three features of this baseline directly constrain dengue ANA interpretation:

The ~5–6% floor at high dilutions. Prevalence at ≥1:160–1:320 is strikingly consistent across international populations and decades — approximately 5% (Tan1997) to 5.93% (Li2019 at >1:320). This likely represents a genuine background rate of low-level autoreactivity in the general population.

The rising temporal trend. US ANA prevalence increased from 11.0% (1988–91) to 16.1% (2011–12) by NHANES, most dramatically in adolescents 12–19 years (OR 2.77) and in men. The trend is not explained by BMI, smoking, or alcohol changes, and was documented in a single laboratory with identical methodology — making methodological drift an unlikely explanation (see Dinse2022 - Increasing ANA Prevalence in United States). Practical implication: a dengue paper reporting 23.1% ANA positivity from a 2009 cohort should be benchmarked against contemporary (not 2022) reference values — approximately 13–16% at 1:80.

Sociodemographic correlates. Female sex (~2× higher prevalence), age >50, and non-Hispanic Black race are consistently associated with higher ANA prevalence. A dengue cohort disproportionately female or older will exhibit an inflated ANA rate relative to a mixed reference. See Antinuclear Antibodies.

2.2 ANA During Acute Infection — The Generic Viral Comparator

Dengue must be evaluated against two benchmarks: the healthy-population baseline and the generic viral-infection baseline. Berlin2007 - Autoantibodies in Nonautoimmune Individuals during Infections provides the latter:

Infection type	n	ANA positive	P vs. controls
Viral (HAV, HBV, HCV)	23	21.7%	P < 0.013
Bacterial	41	20.0%	P < 0.006
Healthy blood donor controls	80	3.8%	—

The ~21.7% rate during acute viral infection (~5.7× fold over controls) sets the benchmark expectation: any viral infection produces ANA in roughly one-fifth of patients by a narrow panel. Dengue must exceed this generic elevation to justify a dengue-specific interpretation.

Transience of infection-triggered ANA. Codes2002 - Autoantibodies in Acute Viral Hepatitis provides the key longitudinal data: ANA fell from 20.5% (acute phase; IIF homogeneous ≥1:40; n=156 prospective; Salvador Brazil) to 6.4% in convalescence, with no association with severity or disease chronification. Anti-smooth muscle antibody (ASMA) showed a parallel decline (14.8% → 3.9%). The natural default of infection-triggered ANA is transience. Persistence requires an explanation.

Part III — ANA in Acute Dengue

3.1 The Headline Measurement

Chatterjee2024 - ANA Detection in Dengue Kolkata is the central source — the first ANA measurement during acute dengue on the HEp-2 gold standard:

HEp-2 IIFA: 54.8% of dengue patients positive vs. 10.3% of dengue-negative febrile controls (p < 0.001; n=135 dengue-confirmed, n=126 controls; Kolkata fever clinics; Feb 2021–Feb 2024)
LIA (18 disease-specific autoantibodies): 18.5% positive vs. 7.1% controls (p = 0.009)

The 54.8% IIFA rate exceeds both the healthy-population baseline (13–16% at 1:80) and the generic viral-infection benchmark (~22% by narrower ELISA). The LIA-confirmed 18.5% — while still elevated above the 7.1% control rate — represents a much smaller fraction. The critical relationship is the gap between them.

3.2 The IIFA:LIA Gap — Dissecting the Non-Specific Fraction

The 54.8% IIFA vs. 18.5% LIA rate creates a ~3:1 ratio: approximately two-thirds of IIFA-positive dengue patients had no confirmed disease-specific autoantibody. Among dengue-negative febrile controls, the IIFA:LIA ratio is much tighter (10.3% vs. 7.1%), confirming the non-specific IIFA excess is dengue-specific, not a feature of febrile illness generally.

This non-specific fraction is now mechanistically named. Zhou2007 - Polyreactive Antibodies Natural Antibody Function characterises germline-encoded polyreactive IgM (PAB) — antibodies constitutively present in all healthy individuals from birth that bind structurally unrelated self and non-self antigens, including nuclear components, with low affinity and without antigen-driven selection. Key properties:

Present at 15–20% of adult peripheral blood B cells at rest
Serum half-life ~8 hours
Bind dsDNA, histones, nucleosomal antigens, and other nuclear components as a constitutive feature of germline V-gene encoding — not through antigen exposure
Distinguishable from pathological autoantibodies by: absence of somatic hypermutation, low affinity, multi-specificity, IgM isotype, and failure to confirm on antigen-specific LIA

In the context of acute dengue, the inflammatory milieu — polyclonal B cell stimulation by DENV structural proteins and NS1, cytokine amplification (IFN-α, IL-6, TNF), and antigen release from damaged tissue — is expected to amplify or unmask the polyreactive IgM pool above its resting baseline. This generates a transient, IgM-dominated, broad antinuclear reactivity with exactly the characteristics of the Chatterjee2024 non-specific fraction: IIFA-positive (polyreactive IgM binds nuclear antigens non-specifically in IIF), LIA-negative (LIA tests for high-affinity disease-specific targets), IgM-class dominant, and predicted to resolve rapidly given the ~8-hour half-life once the inflammatory stimulus subsides.

Converging evidence from Vo2020 - Autoantibody Profiling in Dengue — a 123-antigen protein microarray in Cambodian children (n=40 dengue; n=8 HD) — is fully consistent: 80 IgM vs. only 6 IgG autoantibodies were elevated in DENV-infected patients vs. healthy donors, with IgM breadth significantly exceeding IgG. The IgM dominance and breadth are signatures of polyreactive IgM amplification, not affinity-matured antigen-driven autoimmunity.

Practical clinical implication. In acute dengue, a positive HEp-2 IIFA alone is not evidence of pathological autoimmune induction. The operationally meaningful measure is LIA positivity (~18.5%), which captures antigen-driven, disease-relevant autoantibodies. The polyreactive IgM signal is self-limiting; the LIA-positive signal may not be. See Polyreactive Antibodies.

3.3 Disease-Category Signals in the LIA-Confirmed Fraction

Of 7 autoimmune disease categories tested by Chatterjee2024’s LIA panel, only MCTD (multivariate p = 0.041; OR 14.01, 95% CI 2.197–89.215) and autoimmune myositis (multivariate p = 0.018; OR 18.37, 95% CI 2.746–122.944) were significantly elevated in dengue patients. These ORs carry very wide confidence intervals, the sample is small (n=135), and no follow-up autoimmune disease incidence data are available. Both findings are hypothesis-generating, not confirmed.

Sungnak2025 - Distinct Immune Responses Asymptomatic Symptomatic Dengue measured 120 autoantibody targets in a cross-sectional design and found no significant differences across asymptomatic, DF, and DHF severity groups — consistent with the overall autoantibody burden not stratifying severity, even though specific autoantibody patterns may differ.

3.4 The Nuclear Antigen IgG Consumption Model — Why ANA May Be Lower in Severe Dengue

Vo2020 - Autoantibody Profiling in Dengue contributes a mechanistically important finding that inverts the naive severity-centric expectation. In the DHF subgroup (n=8), 19 IgG autoantibodies — including canonical nuclear antigen targets (KU P70/P80, SmD, SmD1, Sm/RNP, histone H3, histone H4, nucleosome antigen, U1-snRNP-C) — were positively correlated with platelet count (Spearman r = 0.74–0.83; all p < 0.05). More severe disease (lower platelet count) corresponds to lower free-serum nuclear antigen IgG, not higher.

The proposed mechanism is consumption: as DHF progresses, nuclear antigen IgGs are sequestered into circulating immune complexes (ICs) and deposited in tissues, reducing detectable free-serum levels. Both the autoantibodies and the platelets are simultaneously consumed as severity escalates. Low anti-Factor P IgG and anti-complement C4 IgG alongside the low anti-nuclear IgGs may simultaneously remove complement regulatory brakes, amplifying cascade activation.

Implication for the ANA thread: clinical ANA testing at peak disease severity may systematically underestimate total autoimmune activation in DHF. A study design measuring ANA at peak fever — the standard approach — would capture the trough of free-serum nuclear antigen IgG in the most severely affected patients. Morel2014’s ANA-negative MAS cases (§5 below) may represent this consumption phenomenon rather than absence of nuclear antigen autoimmunity. This model requires serial paired measurement of free-serum ANA, complement activation markers (C3d, sC5b-9), IC levels, and platelet count to test directly.

Constraint: Vo2020’s DHF group is n=8, cross-sectional, without multiple comparison correction. The consumption model is plausible and mechanistically coherent, but is hypothesis-generating pending a powered longitudinal study.

Part IV — Mechanisms Generating ANA in Dengue

4.1 NS1 Molecular Mimicry — The Primary Dengue-Specific Mechanism

The NCKU group (Lin, Lei et al., Taiwan) has provided the most experimentally detailed evidence for dengue-specific autoimmune induction. Anti-dengue NS1 antibodies — generated during normal antiviral immunity — cross-react with host platelet and endothelial cell surface proteins due to structural similarity between NS1’s C-terminal domain and those host targets.

Attribution. The foundational identification of a true IgM anti-platelet autoantibody in dengue — distinguishing it from immune complex deposition — was made by Lin2001 - IgM Anti-Platelet Autoantibody in Dengue Patients in a Taiwan DENV-3 cohort (1998–99; predominantly primary infections confirmed by HAI). Platelet-bindable IgM, but not IgG, was elevated in 8/10 DHF patients with no detectable circulating ICs; complement-mediated platelet lysis correlated with severity. Lin2006 extends and mechanistically elaborates this finding; all subsequent NCKU autoimmune work (Lin2011, Wan2012) builds on the 2001 case identification.

Established cross-reactive targets (see NS1 Molecular Mimicry in Dengue, Lin2011 - Molecular Mimicry Virus Host Dengue Pathogenesis, Wan2012 - Autoimmunity in Dengue Pathogenesis):

Host Protein	Location	Consequence of Cross-Reactivity
PDI (protein disulfide isomerase)	Platelet + endothelial surface	PDI inhibition → platelet aggregation inhibition; cross-reactive epitope mapped to P311–330
Vimentin	Platelet + endothelial surface	Surface binding; anti-vimentin elevation anomalous — see below
ATP synthase β-chain	Platelet + endothelial surface	Surface binding; functional consequences under investigation
HSP60	Platelet + endothelial surface	Also cross-targeted by anti-prM Abs (see §4.2)
LYRIC protein	Endothelial	NS1 aa 116–119 shares sequence similarity with LYRIC aa 334–337

Epitope resolution. Cheng2015 - NS1 P311-330 Anti-PDI Autoantibodies in DHF (n=43 DHF sera; Vietnamese patients; NCKU laboratory) narrows the PDI-cross-reactive epitope to P311–330 (aa 311–330) within the C-terminal region (aa 311–352). Key findings:

Anti-PDI IgM/IgG both correlate with anti-P311–330 IgM/IgG (r = 0.732 IgG, p < 0.0001)
Anti-P311–330 is significantly higher in DHF than DF (p < 0.05 for IgM)
HSP60 cross-reactivity uses a distinct, unidentified NS1 epitope — not P311–330 (anti-HSP60 did not correlate with anti-P311–330)
Anti-vimentin IgM does not correlate with anti-NS1 IgM or anti-EC IgM despite being elevated in DHF — inconsistent with simple NS1-mimicry origin; mechanism unknown
Deletion of aa 277–352 abolishes anti-NS1-mediated platelet aggregation and bleeding tendency

Vaccine design implication. NS1-based vaccine constructs retaining P311–330 will generate anti-PDI cross-reactive antibodies; constructs with P311–330 deleted or mutated may avoid the platelet-aggregation arm of NS1 mimicry. The HSP60 epitope remains unidentified, making its exclusion from NS1 constructs impossible at present.

Infection-order independence. Anti-PDI IgM, anti-HSP60 IgM, and anti-EC IgM were comparably elevated in primary DHF (n=2) and secondary DHF (n=15) in Cheng2015 — no significant difference by infection order. This is convergent with Lin2001 (primary DENV-3 IgM anti-platelet autoantibody) and Saito2004 (see §5 below). Three independent sources establish that NS1 molecular mimicry anti-endothelial autoantibodies are infection-order independent — constitutive features of NS1 immunity, not ADE-escalated secondary-infection phenomena. The secondary-infection severity escalation is driven by immune complex pathways (PAIgG, PAIgM), not by a quantitative increase in NS1-mimicry autoantibodies. (Caveat: n=2 primary DHF in Cheng2015 is insufficient to establish this definitively.)

E protein coagulation homology. Beyond NS1, Lin2011 - Molecular Mimicry Virus Host Dengue Pathogenesis identifies the dengue E protein WGNGCG motif (aa 101–106) as sharing sequence homology with coagulation factors XI, X, IX, VII, thrombin, plasminogen, and tPA. Anti-E antibodies bind human plasminogen and inhibit plasmin activity. The motif is conserved in haemorrhagic flaviviruses (JEV, WNV, YFV, TBE, OHFV) but absent in HCV, which does not typically cause haemorrhage — independent molecular evidence linking this motif to flavivirus haemorrhagic phenotype. This E-protein coagulation homology is the mechanism behind the anti-prothrombin elevation documented in Berlin2007 - Autoantibodies in Nonautoimmune Individuals during Infections (69.6% of viral infections), a convergence that was visible only from full-wiki synthesis.

The critical caveat for the ANA question. The established dengue molecular mimicry targets — PDI, vimentin, ATP synthase β, HSP60 — are surface-accessible cytoplasmic and transmembrane proteins, not canonical nuclear antigens. Whether dengue NS1 or other viral proteins share structural homology with nuclear antigens (Sm, dsDNA, Ro, La, Scl-70) remains untested. If no such homology exists, the dengue-associated IIFA-positive ANA (54.8%) must be explained entirely by polyreactive IgM (§3.2) plus epitope spreading (§4.4) — NS1 mimicry alone cannot account for nuclear antigen reactivity.

4.2 Anti-prM Structural Antibodies — A Second Mimicry Arm

Dejnirattisai2010 - Anti-prM Antibodies Enhance Dengue ADE (Science 2010; 891 Semantic Scholar citations) establishes that anti-prM antibodies are the dominant structural antibody class in DENV-infected humans — approximately 60% of dengue antigen-specific B cell lines across n=3,020 B cell lines from seven Thai donors. This finding was absent from the wiki’s ANA discussion before V2.0.

Key features:

Fully cross-reactive across all four DENV serotypes — a property more complete than most anti-E antibodies
Cannot fully neutralise regardless of titre: incomplete prM cleavage during virion maturation leaves ~30% of prM as a structural cap on mature virions; anti-prM antibodies thus face an antigen-accessible fraction that prevents stoichiometric neutralisation
Mediate up to 10^5-fold ADE in primary monocytes and dendritic cells — the most potent ADE mechanism documented for dengue
Cross-react with HSP60 on BHK-21 and A549 cell surfaces (Lin2011 confirmation)

Intersection with the ANA thread. Anti-prM/HSP60 cross-reactivity represents a second molecular mimicry pathway convergent on the same endothelial target (HSP60) as anti-NS1/HSP60. Whether anti-prM contributes to the dengue ANA signal by mechanisms parallel to anti-NS1 is unknown — no study has measured anti-prM titres in the context of ANA prevalence. However, since anti-prM is generated in all dengue infections (primary and secondary), and since its HSP60 cross-reactivity is documented, the anti-NS1 mimicry models of Lin2006 and Lin2011 are likely to underestimate total dengue anti-host activity by ignoring the anti-prM fraction.

Vaccine design implication. All current vaccine platforms — CYD-TDV (Dengvaxia), TAK-003 (Qdenga), TV003/TV005 — use native prM sequences. All will prime anti-prM responses that are cross-reactive, fully incapable of neutralisation, and potently ADE-promoting. No post-Dejnirattisai2010 vaccine candidate in this wiki has adopted heterologous or modified prM sequences to address this. See prM Protein, Dengue Vaccine Candidates.

4.3 Bystander Activation — Likely Insufficient

Bystander activation — cytokine-driven non-specific polyclonal B and T cell activation — is the intuitive explanation for any severe viral illness producing ANA. Dengue involves a cytokine storm (IFN-α, IL-6, TNF, IL-10) of considerable magnitude (see Cytokine Storm).

Johnson2022 - Infectious Diseases Autoantibodies and Autoimmunity (citing Trahtemberg et al.) reports no significant difference in ANA prevalence between COVID-19-positive and COVID-19-negative ICU patients. Severe COVID-19 generates an ICU-grade cytokine storm; if bystander activation were sufficient to elevate ANA, it should do so in this population relative to matched ICU controls — and it does not. This is the most direct evidence that cytokine storm alone is insufficient to explain infection-triggered ANA. The dengue-specific ANA signal must derive from something dengue-specific: molecular mimicry with host proteins, or epitope spreading from dengue-induced tissue damage.

4.4 Epitope Spreading — The Post-Infectious Persistence Mechanism

Epitope spreading — initial immune response damages host tissue → cryptic self-antigens are released → immune response expands to new self-epitopes — is the most plausible mechanism for persistent post-dengue ANA. Dengue-induced endothelial apoptosis (via anti-NS1 → NO/p53/Bax/caspase-3 pathway; see Lin2006 - Autoimmune Pathogenesis in Dengue Virus Infection) and complement-mediated platelet lysis release intracellular contents including nuclear antigens not previously exposed to adaptive immune surveillance. Secondary ANA responses initiated against these cryptic antigens — through bystander activation in the context of a primed immune system — would be consistent with:

The nuclear specificity of post-dengue ANA (AC-1 nuclear homogeneous at 6 months in Gawali2021; nuclear antigen targets at 2 years in Garcia2009)
The delay between acute dengue and SLE diagnosis in case reports (4 weeks to 2 months; Rajadhyaksha2012, Velazqueza2017)
The persistence beyond the NS1-IgG waning curve (see §6.3 below)

Epitope spreading is mechanistically established as a general phenomenon in infection-triggered autoimmunity (see Infection-Triggered Autoimmunity, Johnson2022 - Infectious Diseases Autoantibodies and Autoimmunity) but has not been directly demonstrated in dengue.

4.5 FcγRIIa-Driven Immune Complex Persistence — An Amplifying Mechanism

A dengue-specific host-genetic amplifier operates through the FcγRIIa receptor. The H131 allele of FcγRIIa binds IgG2 complexes with 4.5× higher affinity than the R131 allele — a difference that is IgG2-specific and absent for IgG1/3/4 (see Bruhns2009 - FcγR Specificity and Affinity for IgG Subclasses; surface plasmon resonance; all human FcγRs measured). The HH genotype impairs immune complex clearance by saturating FcγRIIa-mediated phagocytic pathways, causing IC accumulation in tissues and circulation.

Garcia2009 - Long-term Clinical Symptoms Post-Dengue found FcγRIIa-HH significantly associated with post-dengue sequelae (OR 2.83) and that elevated IC correlated with higher anti-dengue IgG titres (p = 0.042). The IC-persistence model: IC accumulation → prolonged antigen presentation → sustained polyclonal B cell activation → persistent ANA production. This provides a mechanistic pathway from acute dengue to 2-year ANA persistence without requiring ongoing viral replication.

Garcia2010 - Asymptomatic Dengue FcγRIIa Polymorphism establishes that FcγRIIa acts primarily as a binary gate (asymptomatic vs. symptomatic) rather than a severity dial within symptomatic disease (OR 10.56 for HH vs. RR for DHF; p < 0.001). The FcγRIIa-ANA interaction at population level is untested — no study in this wiki measures ANA rates stratified by genotype.

4.6 Macrophage-Driven Autoimmunity — The ANA-Negative Axis

Morel2014 - Autoimmune Response in Children With Dengue (Paraguay; n=3 pediatric cases) introduces the most important negative finding for the autoimmunity thread: ANA and anti-dsDNA were negative in all three cases, including the two most severe. Cases 2 and 3 fulfilled HLH-2004 criteria for Macrophage Activation Syndrome (MAS) / secondary haemophagocytic lymphohistiocytosis — fever, hepatosplenomegaly, leukopenia, anaemia, hypertriglyceridaemia (383–470 mg/dl), hyperferritinaemia (1,150–3,828 mg/dl) — and responded clinically to methylprednisolone bolus. Both were completely ANA-negative.

Notably, Case 1 — the mildest, self-limiting case — was the only one with detectable conventional autoimmune markers (IgM anticardiolipin, hypocomplementaemia, proteinuria). The severity-autoantibody relationship is therefore inverted within this series relative to the Lin2006 anti-endothelial correlation: the most severe autoimmune complication is the most conventionally seronegative.

The mechanistic resolution: MAS/HLH is macrophage and CD8+ T cell-driven, not autoantibody-mediated. NS1-TLR4 signalling on macrophages (see Guzman2016 - Dengue Infection, NS1 Protein) and ADE-mediated FcγR entry amplifying macrophage activation (see Antibody-Dependent Enhancement) are plausible triggers — neither requires ANA. Standard ANA testing is blind to this macrophage-hyperactivation axis by design. Palacios2016 - Autoimmunity in Dengue Literature Review adds an additional MAS+nephrotic syndrome case (Lai et al. 2012, France), confirming dengue-MAS is not solely paediatric.

Two-axis model. The dominant wiki framework prior to Morel2014 — NS1 mimicry → autoreactive antibodies → ANA-positive pathology — is now complemented by a parallel macrophage axis: cytokine-driven (IFN-γ, IL-18, IL-6), CD163-marked, ANA-negative. These axes are not mutually exclusive — they may co-occur in the same patient or diverge based on host genetic and immunological factors. ANA-prevalence studies in dengue will systematically miss the MAS-prone subset; negative ANA does not exclude severe dengue immune complications.

4.7 Anti-Cytokine Autoantibodies — A Third, Protective Axis (⚠ Unvalidated)

Chaturvedi2001 - Cytotoxic Factor Autoantibodies DHF introduces a conceptually distinct autoimmune mechanism: autoantibodies directed against a dengue-produced cytokine (hCF, “human cytotoxic factor”), proposed to be protective rather than pathogenic.

In a 1996 Lucknow epidemic cohort (n=136; classified by 1997 WHO DHF grades), anti-hCF IgG was present in 96% of DF patients vs. 8% of grade IV DHF patients (P ≤ 0.001) — an inverse severity association unique among all dengue autoantibodies in this wiki. The proposed mechanism: hCF is a dengue-specific CD4+ T cell cytotoxin secreted by monocytes; anti-hCF autoantibodies neutralise hCF activity, reducing cytotoxic damage and correlating with milder disease.

⚠ Critical epistemic caveat. hCF is the Chaturvedi group’s proprietary, uncharacterised construct — partially purified by HPLC from dengue-infected monocyte supernatants but never sequenced, deposited in a database, or independently replicated. CrossRef citation count is 1. The concept may map to a known cytokine or dengue structural protein fragment, or may be an artefact of the purification protocol. All Chaturvedi2001 claims must be treated as hypothesis-generating. If hCF is independently validated and characterised, this would represent a genuinely novel protective autoimmune mechanism in dengue.

Structural significance. Regardless of hCF’s validity, this paper establishes the conceptual possibility of a third dengue autoimmune axis:

ANA-positive B-cell axis — NS1 mimicry → anti-PDI/HSP60/endothelial autoantibodies (pathogenic; ANA-detectable)
ANA-negative macrophage axis — cytokine-driven MAS/HLH (Morel2014; ANA-invisible)
ANA-negative anti-cytokine axis — anti-hCF protective autoantibody (Chaturvedi2001 — ⚠ unvalidated)

Standard ANA testing is blind to both ANA-negative axes.

Part V — The Thrombocytopenia Bifurcation Model

Dengue thrombocytopenia is not a single immunological event — it is at minimum three converging mechanisms, with different infection-order specificity, isotype, effector pathway, and therapeutic implications. Understanding this bifurcation is the most clinically actionable synthesis in this wiki’s autoimmunity thread.

5.1 Primary Infection — True Autoantibody

In primary DENV infection, the IgM anti-platelet mechanism is a true autoantibody: IgM directed against platelet surface proteins (principally via NS1 C-terminal mimicry), generated without prior dengue exposure, with no requirement for memory re-stimulation.

Source: Lin2001 - IgM Anti-Platelet Autoantibody in Dengue Patients (Taiwan DENV-3 1998–99; n=28 dengue, n=11 other; predominantly primary infections by HAI). Key findings:

IgM (not IgG) anti-platelet autoantibody detected in 8/10 DHF/DSS patients
No detectable circulating IC in primary infections — confirming the platelet-bound IgM is not immune-complex deposition
Complement-mediated platelet lysis correlates with DHF/DSS severity (DHF/DSS MFI 276 vs. DF MFI 140; p < 0.01); platelet aggregation inhibition does not correlate with severity
Dengue-specific: confirmed negative in JEV, HCV, EV71, and healthy controls
IgM anti-platelet is amenable to complement-targeting or autoimmune-directed therapeutic approaches (IVIG, steroids)

5.2 Secondary Infection — Immune Complex Pathways

In secondary DENV infection, two distinct immune complex pathways operate in parallel — neither is an autoantibody. Both carry anti-dengue virus specificity confirmed by platelet eluate analysis, distinguishing them mechanistically from the primary-infection autoantibody.

PAIgG pathway (Oishi2003 - PAIgG and Thrombocytopenia in Secondary Dengue; Philippines; n=53 secondary dengue):

Platelet-associated IgG (PAIgG) elevated vs. other febrile illness controls (P < 0.0001); inversely correlated with platelet count (r = −0.50, P < 0.001)
Platelet eluate binds recombinant DENV-2 NS1 by immunoblot — anti-dengue NS1 IC, not anti-platelet autoantibody
Anti-platelet integrin GPIIb/IIIa blockade does NOT reduce PAIgG binding — FcγRII-independent attachment mechanism
Operated via complement-mediated and/or direct membrane mechanisms

PAIgM pathway (Saito2004 - PAIgG and PAIgM in Secondary Dengue; Philippines; n=78 secondary dengue):

PAIgM elevated in secondary dengue (17.5 ± 20.4 vs. 4.2 ± 3.8 ng/10⁷ healthy controls; P < 0.001); inversely correlated with platelet count (r = −0.231, P = 0.046)
Platelet eluate binds DENV antigens by ELISA — anti-dengue IgM immune complex, not anti-platelet autoantibody; mechanistically opposite to Lin2001’s primary-infection IgM despite being the same isotype
Completely FcγR-independent: IgM pentamer structure precludes Fc receptor engagement at every step — platelet docking, clearance, and downstream signalling
PAIgM independently predicts DHF by multivariate logistic regression; cut-off >20 ng/10⁷ platelets yields 92.1% specificity and 48.6% sensitivity for DHF (grades I–II only; no DSS data)

The FcγRIIa paradox. Both secondary-infection IC pathways bypass FcγR at the platelet-docking step; PAIgM additionally bypasses FcγR at every downstream step. Yet Garcia2010 - Asymptomatic Dengue FcγRIIa Polymorphism finds FcγRIIa-HH associated with DHF with OR 10.56. If neither secondary-infection thrombocytopenia mechanism operates through FcγR at the platelet level, the Garcia2010 FcγRIIa-HH risk signal must operate through a completely non-platelet route — ADE-driven monocyte/macrophage activation (where FcγRIIa is the canonical entry receptor) remains the leading hypothesis, but no study in this wiki directly tests this.

The isotype paradox. The same isotype (IgM) mediates opposite mechanisms in primary and secondary infection: primary-infection IgM is anti-self (anti-platelet autoantibody); secondary-infection PAIgM is anti-dengue-antigen (immune complex). Distinguishing these in clinical practice requires platelet eluate specificity testing — a research tool, not a routine clinical assay. Without eluate data, PAIgM elevation alone is diagnostically ambiguous.

Therapeutic implication. Primary-infection thrombocytopenia (autoantibody-mediated) may respond to complement inhibition, steroids, or IVIG. Secondary-infection thrombocytopenia (IC-mediated) requires viral clearance or complement inhibition strategies — autoimmune-directed therapy would not be expected to help. Knowing the dominant mechanism in a given patient would directly inform treatment choice. No clinical protocol currently tests for this distinction.

Part VI — Temporal Dynamics of Dengue ANA

6.1 The Acute Phase (Chatterjee2024 + Vo2020)

The acute phase peak is documented by Chatterjee2024: 54.8% IIFA (HEp-2), 18.5% LIA. The Vo2020 microarray in Cambodian children provides complementary breadth data: 80 IgM + 6 IgG autoantibodies elevated across complement, coagulation, and nuclear antigen targets. The Vo2020 data also establishes that total autoantibody load does not discriminate severity (no significant difference ASD vs. DF/DHF) — the clinically relevant variable is specific antibody consumption patterns and immune complex formation, not total autoantibody production.

An unexpected asymmetry from Vo2020. Primary infection (n=6; all male, all DENV-1) generated significantly higher IgG autoantibody breadth than secondary infection (70 of 123 IgG autoantibodies elevated in primary vs. secondary; p < 0.01). IgM did not differ. This primary > secondary IgG inversion challenges severity-centric ADE-based predictions (secondary infection should generate more autoantibody dysregulation). The proposed interpretation: naïve B cells in primary infection may escape tolerance checkpoints more readily, differentiating into IgG-producing plasma cells without memory-dominated competition. In secondary infection, memory recall responses dominate, potentially competing with and suppressing autoreactive clones.

This inversion is entirely based on n=6 primary patients, all male, all DENV-1 — heavily confounded. It requires replication before incorporation into any clinical framework. However, if it replicates, it would imply that first-time dengue exposure is the window of maximal IgG autoantibody activation — opposite to the secondary-infection severity-escalation model from Lin2006 and Wan2012. The most parsimonious reconciliation is that primary infection generates broader but mostly non-pathogenic IgG autoantibody repertoires, while secondary infection generates narrower but more pathogenically targeted anti-platelet/anti-endothelial responses.

6.2 Six Months Post-Dengue (Gawali2021)

Gawali2021 - ANA Prevalence in Seroconverted Dengue Patients provides the only 6-month time point: 22/120 (18.33%) of dengue IgG-seroconverted patients were ANA-positive by HEp-2 IIFA at 1:100 dilution only; dominant pattern AC-1 (nuclear homogeneous, 81.81%); Gwalior Madhya Pradesh India.

The binding limitation is absence of a contemporaneous control group. The closest available baseline — Li2019 (Chinese health-checkup at >1:100: 14.01%) — leaves only ~4 percentage points of apparent dengue-attributable excess, within sampling variability for n=120. Whether 18.33% represents genuine dengue-related ANA persistence or the regional background ANA rate for Central India cannot be determined from this dataset alone. The study is hypothesis-supporting but not hypothesis-confirming.

The AC-1 dominant pattern is consistent with dsDNA/histone/nucleosome targets — the same nuclear antigen specificity class that Vo2020 finds positively correlated with platelet counts in DHF, consistent with IgG-class autoantibodies in those specificities persisting into the sub-acute period after the acute polyreactive IgM noise has cleared.

6.3 Two Years Post-Dengue (Garcia2009)

Garcia2009 - Long-term Clinical Symptoms Post-Dengue remains the principal long-term source: 23.1% ANA positivity in symptomatic Cuban patients 2 years after the 2006 DENV-4 epidemic. All published healthy-population references show this is elevated, including the most conservative contemporary US estimate (16.1% at 1:80 in 2011–12, Dinse2022) — even accounting for the substrate gap (rat liver likely underestimates HEp-2 rates by an unknown factor, meaning the true HEp-2-equivalent rate is higher). However, the Garcia2009 cohort is highly selected: symptomatic patients returning for voluntary 2-year follow-up, with 21/26 ANA-testers having ≥1 prior infection, in a country with a long history of sequential epidemic waves producing a multiply-exposed population. This cohort is not representative of typical dengue patients in hyperendemic settings.

Study	Timing	Substrate	Rate	Control group
Chatterjee2024 - ANA Detection in Dengue Kolkata	Acute	HEp-2 IIFA	54.8%	Yes (10.3%)
Chatterjee2024 - ANA Detection in Dengue Kolkata	Acute	LIA	18.5%	Yes (7.1%)
Gawali2021 - ANA Prevalence in Seroconverted Dengue Patients	6 months	HEp-2 IIFA (1:100)	18.33%	No
Garcia2009 - Long-term Clinical Symptoms Post-Dengue	2 years	Rat liver IIF	23.1%	No

The trajectory hint — 54.8% → 18% → 23% — is consistent with a decline from the acute peak followed by stabilisation or modest rebound. If real, the stabilisation or slight rise from 6 months to 2 years would be more consistent with epitope-spreading or IC-persistence components than simple passive polyreactive IgM decay. The 1–3 month window remains entirely undocumented.

6.4 Kinetic Constraints from Antibody Dynamics

Bos2025 - Longitudinal Antibody Dynamics After Dengue (PREPRINT; Nicaragua pediatric cohort; n=79; measured at <1, 3, 6, and 18 months post-primary dengue) provides the first quantitative kinetic data bearing on the ANA trajectory:

NS1-IgG wanes with t½ ≈ 2.1 years (primary infection; similar trajectory in secondary). If anti-NS1 antibodies are the substrate for NS1-driven molecular mimicry (Lin2006, Lin2011), the NS1-mimicry component of dengue ANA should decline on a similar timescale — approximately halved every ~2 years. This implies that Garcia2009’s 2-year ANA persistence cannot be attributed primarily to ongoing NS1 mimicry; the NS1 contribution should be ~50% of its acute peak by then. The residual 2-year ANA signal is more consistent with epitope-spreading (slow, persistent) or FcγRIIa IC-persistence (patient-specific) components.

Cross-reactive E protein IgG (XR E-IgG) does not wane — it rises. XR EDI/II-IgG increases between 6 and 18 months post-primary with a calculated t½ = −2.13 years (actively growing). Given the E-protein WGNGCG motif’s homology with coagulation factors (Lin2011, §4.1), the anti-coagulation-factor component of dengue autoreactivity may follow an opposite (rising) temporal vector within the same patient who is simultaneously losing NS1-driven anti-platelet reactivity. Whether these rising XR E-IgG antibodies are functionally non-neutralising (ADE-capable) vs. cross-neutralising is unknown — but the temporal asymmetry adds complexity to any model of dengue autoimmune risk over time.

Bos2025 is a preprint and mechanistic claims derived from these kinetics should be treated as hypotheses pending peer review.

Part VII — Clinical Consequences at Population Level

7.1 The Only Robust Signal: ADEM in the First Month

Shih2023 - Autoimmune Disease Risk After Dengue provides the most statistically rigorous assessment: 63,814 laboratory-confirmed dengue patients (Taiwan NHIRD), 255,256 matched controls, mean follow-up 4.57 years.

Overall autoimmune disease risk: aHR 1.16 (P = 0.0002) — statistically significant, clinically small
After Bonferroni correction across 14 outcomes: only ADEM survives (aHR 2.72; P < 0.0001)
ADEM risk is entirely confined to the first month: 16 dengue patients vs. 0 controls developed ADEM in month 1 (HR >9999); non-significant after month 1
All other SARD outcomes — SLE, Sjögren’s, RA, GBS, myasthenia gravis, post-infectious arthritis — non-significant after Bonferroni correction

ADEM as the exception is mechanistically coherent: a transient, acute-phase demyelinating response (molecular mimicry between DENV and CNS myelin components) that resolves once acute immune stimulation subsides. Dengue neurological complications are discussed further in Dengue Neurological Complications.

7.2 Refuting the Broad Autoimmune Risk Claim

Li2018 - Increased Risk of Autoimmune Diseases in Dengue (ICD-coded NHIRD; n=12,506 hospitalised “dengue”) reported aHR 1.88 for any autoimmune disease, with significant associations for >20 specific diseases. Shih2023 identifies three methodological failures in this analysis: (1) only 51.4% of ICD-coded dengue hospitalisations were laboratory-confirmed — approximately half the Li2018 “dengue” cohort may not have had dengue at all, likely enriched with early autoimmune disease misdiagnosed as dengue (shared features: fever, rash, thrombocytopenia); (2) no multiple comparison correction across 20+ outcomes; (3) selection bias from restricting to hospitalised patients. The 51.4% misclassification issue is the central methodological lesson: in endemic settings, non-lab-confirmed cohorts are unsuitable for dengue-autoimmune research.

Cross-study convergence on GBS. Both Li2018 (aHR 0.97, non-significant) and Shih2023 (non-significant) find no GBS elevation — a null finding robust to completely different methodological approaches. Dengue-GBS in the case-report literature does not reach statistical significance in either population study.

7.3 Reconciling the Garcia2009 ANA Signal with the Shih2023 Null Finding

These appear contradictory but are compatible:

ANA positivity is not equivalent to autoimmune disease. Most ANA-positive individuals — even at clinically relevant dilutions — never develop clinical SARD.
Garcia2009’s cohort was a highly selected symptomatic subset with multiple prior infections and FcγRIIa-HH enrichment, not representative of all dengue. Shih2023 covers all confirmed dengue, including mild and asymptomatic.
Elevated IC and CRP in Garcia2009 may reflect post-infectious inflammation rather than true autoimmunity. The biological signal may be real without reaching the threshold for clinical disease in the vast majority.
ADEM — the one confirmed risk — is an acute, transient demyelinating event, not a chronic autoimmune disease. Its restriction to the first month is compatible with dengue-driven autoimmunity being predominantly self-limiting.

Part VIII — The Case-Report Spectrum: From Transient Resolution to Persistent SLE

The population-level null finding for SARD does not preclude individual-level autoimmune severity. Four primary sources and one secondary citation document five cases spanning the full clinical spectrum:

Pole	Source	ANA	Anti-dsDNA	Isotype / titre	Renal involvement	Anti-phospholipid	Clinical outcome
Transient multi-autoantibody	Jardim2012 - Autoimmune Features DHF Case Report	1/320 mitotic spindle	Negative throughout	IgG (class not specified)	Proteinuria 0.55 g/day; no biopsy	Not tested	Full ANA resolution at follow-up; no SARD diagnosis; 2° DENV-3 DHF, Campinas Brazil
ANA-negative MAS — severe	Morel2014 - Autoimmune Response in Children With Dengue Cases 2&3	Negative	Negative	—	Absent	Negative	Met HLH-2004 criteria; responded to methylprednisolone; Paraguay paediatric
ANA-negative MAS — mild	Morel2014 Case 1	Negative	Negative	IgM aCL only	Proteinuria	IgM aCL positive	Self-limiting; no steroids
SLE via secondary source	Talib 2013 (cited in Palacios2016 - Autoimmunity in Dengue Literature Review)	Homogeneous positive	Positive	—	Lupus nephritis	Not reported	SLE diagnosed; only known via Palacios2016 — no primary data available
SLE + death	Velazqueza2017 - SLE vs Dengue Case Series Cases 1&2	1:1280 (both)	Positive (both)	Antinucleosome 52 + 258 IU/ml	None documented	Not reported	Case 2 died from pulmonary haemorrhage; n=2 case series; Mexico paediatric
SLE + biopsy	Rajadhyaksha2012 - Dengue Evolving into SLE and Lupus Nephritis	1:320 homogeneous 4+	1:80 4+	Anti-cardiolipin IgM 44 + IgG 12 MPLU/mL (persistent at 4M)	Biopsy-confirmed Class IV GN	IgM + IgG	Partial response to pulse steroids + MMF + HCQ; primary DENV-1; Mumbai India

Five observations from this case series:

Jardim2012 defines the “transient trigger” pole. The most immunologically complex single-case picture in the wiki — ANA 1/320 (mitotic spindle pattern, targeting centromere/spindle apparatus proteins rather than DNA/histones), cryoglobulinemia, LE cells in pleural fluid, selective C3 depression with C4-normal, triple serositis, DIC — fully resolved at follow-up with anti-dsDNA negative throughout. This is the strongest available evidence that dengue can produce a broad, acute multi-autoantibody syndrome that is completely self-limiting. The mitotic spindle ANA pattern is distinct from AC-1 homogeneous (Gawali2021, Rajadhyaksha2012), fine-speckled (Velazqueza2017 Case 2), and homogeneous+cytoplasmic (Velazqueza2017 Case 1) — expanding the dengue-associated autoantigen repertoire to centromere/pericentrin/NuMA, beyond the NS1 C-terminal mimicry targets. The C4-sparing complement depression (C3 0.39 g/L; C4 normal) contrasts with Rajadhyaksha2012 (both C3 22 mg/dL and C4 5 mg/dL severely depleted), suggesting different complement activation routes (alternative/MBL pathway in Jardim2012 vs. classical pathway in Rajadhyaksha2012) — potentially useful for differential diagnosis in endemic settings.
Rajadhyaksha2012 is the primary-source case for biopsy-confirmed dengue→lupus nephritis. Class IV diffuse proliferative GN confirmed by renal biopsy; anti-cardiolipin IgM + IgG persistent at 4 months (combined isotype persistence more consistent with SLE-driven antiphospholipid syndrome than dengue-induced transient IgM aCL). The 4-week dengue→SLE interval is the shortest in the wiki, and the infection was primary (IgM+/IgG-). Primary DENV-1 context converges with the Vo2020 primary-infection primary > secondary IgG autoantibody inversion — raising whether first-time dengue exposure in SLE-susceptible individuals is a higher-risk window than reinfection.
The complete causal ambiguity across all SLE cases. No case — Rajadhyaksha2012, Velazqueza2017, or Talib 2013 — had autoantibody testing performed at the time of initial dengue. Pre-dengue autoantibody baselines are unavailable in every case. High ANA titres (1:1280 in Velazqueza2017; 1:320 in Rajadhyaksha2012) at SLE diagnosis are, if anything, more consistent with pre-existing subclinical SLE being unmasked by dengue than with dengue triggering de novo autoimmunity within 4–8 weeks. The authors of both Rajadhyaksha2012 and Velazqueza2017 explicitly acknowledge this ambiguity.
The ANA-negative MAS vs. ANA-positive SLE contrast is complete. These case series document dengue-associated immune complications spanning the full spectrum from ANA-negative macrophage-driven MAS (Morel2014) to ANA 1:1280 antinucleosome-positive SLE (Velazqueza2017). What genetic or immunological factors determine which pole a given patient inhabits — FcγR polymorphism, complement gene variation, HLA alleles, prior immune history — is unknown from available data.
The combined-isotype anti-cardiolipin signal from Rajadhyaksha2012 is the only combined IgM+IgG aCL finding in the wiki. Morel2014 Case 1 showed transient IgM aCL only (mild, self-limiting dengue without SLE). IgG aCL persistence at 4 months in the context of Class IV GN is consistent with SLE-driven antiphospholipid antibody production rather than a dengue-specific transient response — but without a pre-dengue aCL baseline, this inference cannot be confirmed.

These cases do not challenge Shih2023’s population-level null finding. What they document is that the rare individual at the intersection of dengue and genetic SLE susceptibility may experience biopsy-confirmed autoimmune renal disease within weeks of dengue exposure. Whether this occurs in 1-in-10,000 or 1-in-100,000 dengue patients cannot be determined from case reports; it lies below the detection threshold of any cohort yet published.

Part IX — Host Determinants of Dengue ANA

FcγRIIa genotype. The HH genotype impairs IC clearance (via IgG2-specific 4.5× affinity difference; Bruhns2009) and is associated with post-dengue sequelae (OR 2.83; Garcia2009) and DHF (OR 10.56; Garcia2010). It likely modulates the duration and magnitude of IC-driven autoimmune stimulation rather than ANA induction itself. No study in this wiki has directly measured ANA rates stratified by FcγRIIa genotype.

Sex. Garcia2009 found post-dengue autoimmune markers in 65.7% of women vs. 36.7% of men (p = 0.008). Seet2007 - Post-Infectious Fatigue Syndrome in Dengue found female sex as the dominant predictor of post-dengue fatigue (OR 9.687; Singapore cohort; independent of Garcia2009). Healthy-population ANA prevalence is ~2× higher in females (Satoh2012, Dinse2022). The convergence of female excess across three independent observations — healthy-population ANA, post-dengue autoimmune markers, and post-dengue fatigue — suggests sex-linked immunological mechanisms (X-linked TLR7 double dosing, estrogen-driven B cell activation, FcγR expression modulation) are major determinants of post-dengue immune dysregulation.

Prior infection history. 21/26 patients with autoimmune marker elevations in Garcia2009 had ≥1 prior dengue infection; 12/26 had tetravalent infection history. Multiple prior exposures may shift from IgM-dominant (primary-infection autoantibody) to IgG-dominant (secondary-infection IC) platelet mechanisms — the latter driving longer-lived plasma cell-mediated autoantibody production through the FcγRIIa pathway.

Infection order (primary vs. secondary). Vo2020’s primary > secondary IgG autoantibody inversion (§6.1) is the first data suggesting that infection-order modifies the dengue autoantibody repertoire. If replicated, this would invert the usual assumption that secondary infection represents higher autoimmune risk — first-time dengue in a naïve host may generate broader (though possibly less pathogenically targeted) IgG autoimmunity.

Acute severity. Counterintuitively, acute DHF/DSS severity did not predict development of post-dengue sequelae in Garcia2009 (p = 0.086); post-dengue fatigue showed the same severity-independence in Seet2007 (p = 0.855). Both independent cohorts, different countries, different serotypes, different follow-up durations, different outcomes — both showing the same severity-independence pattern. This strengthens the inference that post-dengue immune sequelae are governed by host-intrinsic factors rather than viral pathological burden. Whatever determines who develops persistent ANA and fatigue operates upstream of clinical severity classification.

Part X — Vaccine Design Implications of the Autoimmunity Data

This section did not exist in V1.x. The combined data from NS1 molecular mimicry, anti-prM immunogenicity, and longitudinal antibody kinetics carry several direct vaccine design implications.

NS1-based vaccines and P311–330. The Cheng2015 finding that P311–330 within the NS1 C-terminal domain is the PDI-cross-reactive epitope has direct implications for NS1-based subunit vaccines. Any construct retaining P311–330 will generate anti-PDI cross-reactive antibodies; constructs with P311–330 deleted or mutated may avoid the platelet aggregation inhibition arm of NS1 mimicry. The HSP60-cross-reactive epitope remains unidentified, precluding its rational exclusion at present.

The NS1 vaccine paradox. Wan2012 - Autoimmunity in Dengue Pathogenesis identifies a structural immunological paradox: NS1-based vaccines would generate anti-NS1 antibodies that reduce viral NS1 secretion (beneficial for infection control) but would simultaneously produce anti-platelet and anti-endothelial cross-reactive antibodies (potentially pathogenic). This paradox is not resolved by any study in this wiki. A vaccine that generates neutralising anti-NS1 responses while avoiding the C-terminal cross-reactive epitopes (aa 311–352, particularly P311–330) would address both goals — but evidence that such a construct maintains protective immunogenicity is absent.

Anti-prM and all current vaccines. Dejnirattisai2010 establishes that anti-prM is the dominant structural antibody class (~60%) in dengue-infected humans, is fully cross-reactive across all four serotypes, is incapable of full neutralisation due to incomplete prM cleavage, and mediates potent ADE (up to 10^5-fold in monocytes/DCs). All current licensed or advanced vaccine platforms (CYD-TDV, TAK-003, TV003/TV005) use native prM sequences and will prime this response. The wiki contains no source describing a dengue vaccine candidate that uses modified or heterologous prM to reduce anti-prM ADE immunogenicity. This is the largest unaddressed structural immunogenicity gap identified in this literature.

The Katzelnick ADE quantitative window. Bhatt2020 - Dengue Pathogenesis Review (citing Katzelnick et al. 2017, n=6,684 Nicaraguan children) establishes that the ADE-enhancement zone is bounded: maximum ADE occurs at antibody dilutions 1:21–1:80; above this titres are neutralising; below this titres provide no enhancement. This window is directly relevant to vaccine design: a waning vaccine-induced anti-E response that passes through the 1:21–1:80 zone during inter-dose or inter-booster intervals would create a transient ADE-risk window. No source in this wiki maps post-vaccination titre waning kinetics for any current DENV vaccine against this threshold.

Rising XR E-IgG in the inter-infection interval. Bos2025’s finding that XR EDI/II E-IgG rises between 6 and 18 months post-primary infection — while NS1-IgG is waning — means the ADE-risk trajectory in the inter-infection interval is more complex than classical waning models predict. If these rising cross-reactive E-IgG antibodies fall within the Katzelnick 1:21–1:80 window at any point during their rise, they may represent an actively growing rather than passively declining ADE risk in the months-to-years post-primary infection.

Part XI — Acute-Phase Timing as a Mechanistic Distinguishing Feature

Dengue’s autoimmunity timing is distinctive and mechanistically informative compared to other well-characterised infection-triggered autoimmune syndromes:

Syndrome	Pathogen	Post-infection interval to autoimmunity
Guillain-Barré	Campylobacter jejuni	2–4 weeks post-infection
EBV-associated SLE/MS	Epstein-Barr virus	Months to years
SARS-CoV-2 autoimmunity	SARS-CoV-2	Weeks to months (post-acute)
Dengue NS1 mimicry	DENV	During acute viraemia (days 1–7)

The anti-NS1 autoimmune damage in dengue occurs simultaneously with active viral infection — not sequentially after pathogen clearance. Anti-NS1 antibodies that cross-react with platelet and endothelial proteins are generated within the first 5–7 days of infection, before adaptive immune maturation is complete. The predominantly IgM isotype of the acute anti-platelet autoantibody (Lin2001, Lin2006) is consistent with a rapid germinal-centre-independent response — IgM class switching does not require the T cell help that characterises IgG class-switched autoimmunity.

This acute-phase timing means that dengue-triggered autoimmune platelet and endothelial damage is not a post-infectious sequela but a direct feature of the primary anti-viral immune response. Any therapeutic strategy targeting the autoimmune component must operate within the acute febrile window (days 1–7), before the autoantibodies have already mediated their damage — a significantly different clinical challenge from treating post-infectious autoimmunity.

Sungnak2025 - Distinct Immune Responses Asymptomatic Symptomatic Dengue documents that the plasmablast expansion in symptomatic dengue — IGHG1+ and IGHA1+ plasmablasts predominant in DHF, detected by scRNA-seq and V(D)J sequencing — disappears at convalescence. Whether autoreactive clones are among those cleared — or whether they leave persistent memory precisely because they target dengue antigens with cross-reactive host specificity — is unknown from available data.

Open Questions

The 1–3 month ANA trajectory. The acute 54.8% IIFA rate (Chatterjee2024) and the 6-month 18.33% IIFA rate (Gawali2021) leave the most critical part of the trajectory — when does the acute spike decline and how fast? — unobserved. A prospective cohort with HEp-2 IIFA at acute, 1M, 3M, 6M, 12M, and 24M with a concurrent dengue-negative febrile control arm would resolve this definitively.
Titre distribution of the IIFA-positive fraction. Are dengue-associated IIFA-positive, LIA-negative ANAs predominantly low-titre (1:40–1:80) or do some reach ≥1:160? Titre data would distinguish polyreactive IgM background from genuinely elevated disease-relevant autoantibodies and allow benchmarking against the SLE classification threshold (≥1:80; Aringer2019).
NS1 (and other DENV protein) structural homology with nuclear antigens. The established dengue mimicry targets are surface-accessible cytoplasmic and transmembrane proteins. Do any DENV proteins (NS1, prM, E, capsid) share structural homology with canonical nuclear antigens (Sm, dsDNA, Ro, La, Scl-70)? If not, the IIFA-positive nuclear reactivity is entirely polyreactive IgM + epitope spreading. If yes, dengue is capable of directly generating disease-relevant ANA through mimicry. Computational structural homology analysis of the full DENV proteome against canonical ANA target epitopes is the minimum required study.
ANA–FcγRIIa genotype correlation. Does ANA rate and persistence correlate with FcγRIIa genotype (HH > RR > HR) in dengue patients? If the IC-persistence model is correct, this is the expected finding. No study in this wiki has tested this.
MCTD and myositis follow-up from Chatterjee2024 LIA-positive cohort. Do the 18.5% LIA-positive dengue patients go on to develop clinical MCTD or autoimmune myositis at significantly higher rates? The 6–7 month follow-up in Chatterjee2024 is insufficient. A 2–5 year follow-up of the LIA-positive subgroup (n ≈ 25 from the original 135) would test whether LIA positivity in dengue identifies a true pre-clinical autoimmune risk.
Statistical power for rare SARD in Shih2023. MCTD and autoimmune myositis would generate very few cases even in n=63,814. The absence of significant signals for these specific diseases does not rule out a clinically meaningful risk — it may reflect insufficient event counts for rare outcomes. Sample size calculations for these specific diseases in post-dengue cohorts have not been published.
Dengue-MAS as a separate autoimmune axis invisible to ANA testing. Morel2014’s two severe MAS cases were ANA-negative despite being the most severe autoimmune presentations in the series. Prospective concurrent measurement of sCD163, IL-18, and ferritin (MAS markers) alongside ANA, anti-dsDNA, and anti-endothelial Abs (B-cell axis markers) across the full dengue severity spectrum would test whether these axes co-occur or dissociate in individual patients and whether ANA-negative severe dengue is systematically enriched for the MAS phenotype.
De novo SLE induction vs. unmasking. Every dengue-associated SLE case in this wiki (Rajadhyaksha2012, Velazqueza2017, Talib 2013 via Palacios2016) lacks pre-dengue autoantibody baselines. A study that systematically obtains ANA and anti-dsDNA pre-dengue (e.g., at dengue vaccine trial enrollment in a population with subclinical SLE incidence) and then follows up cases who subsequently develop dengue would provide the only clean evidence base for de novo induction vs. unmasking.
NS1-IgG kinetics as a predictor of anti-endothelial autoantibody decay. Bos2025 establishes NS1-IgG t½ ≈ 2.1 years. If the cross-reactive (anti-PDI, anti-HSP60, anti-EC) sub-fraction wanes with the same kinetics as bulk NS1-IgG, NS1-driven autoimmunity should halve every ~2 years. This is testable by paired longitudinal measurement of NS1-IgG titre and anti-PDI/anti-vimentin titre in the same patients across the 0–24 month window.
Rising XR E-IgG and anti-coagulation-factor autoreactivity. If E protein WGNGCG cross-reacts with coagulation factors (Lin2011) and XR E-IgG rises between 6 and 18 months (Bos2025), the anti-coagulation-factor component of dengue autoreactivity may follow an opposite and rising temporal vector to anti-NS1 reactivity. Testing anti-prothrombin and anti-plasminogen titres longitudinally alongside XR E-IgG would directly test whether these are kinetically linked.
Relative contribution of the three platelet mechanisms and therapeutic implications. Primary-infection IgM autoantibody (Lin2001): responsive to complement inhibition or IVIG. Secondary-infection PAIgG IC (Oishi2003) and PAIgM IC (Saito2004): require viral clearance or complement strategies. Determining which mechanism dominates in a given patient would directly inform treatment choice. A study combining PAIgM eluate specificity testing, PAIgG measurement, and Lin2001-type anti-platelet IgM assay in a prospectively enrolled cohort stratified by primary vs. secondary infection would resolve this clinically.
Nuclear antigen IgG consumption as explanation for ANA-negative severe dengue. Vo2020 raises the possibility that free-serum ANA titres are lowest precisely in the most severe DHF because nuclear antigen IgGs are consumed into immune complexes (n=8 DHF; no multiple comparison correction). A prospective study with serial paired measurement of free-serum ANA, C3d, sC5b-9, IC levels, and platelet count across the DHF severity spectrum would test this directly — and would resolve whether some “ANA-negative MAS” (Morel2014) represents antibody consumption rather than absence.
PAIgM’s FcγR independence as an explanation for its predictive superiority over PAIgG. Saito2004 shows PAIgM independently predicts DHF (92.1% specificity) while PAIgG does not survive multivariate regression. If FcγRIIa genotype modulates PAIgG-mediated platelet clearance (HH vs. RR varying the magnitude of PAIgG accumulation), PAIgM — bypassing all FcγR steps via IgM pentamer — would be a purer measure of dengue IC severity unconfounded by genotype. A study measuring PAIgG, PAIgM, and FcγRIIa genotype concurrently in secondary dengue would test this.
Anti-cardiolipin persistence — dengue-induced vs. SLE-driven. Rajadhyaksha2012 documents combined IgM + IgG aCL persisting at 4 months in dengue→SLE; Morel2014 Case 1 shows transient IgM aCL only in dengue without SLE. A prospective study tracking aCL at acute dengue, 3M, and 12M in a cohort stratified by ANA positivity would determine whether durable combined-isotype aCL is specific to SLE-susceptible dengue patients or occurs more broadly.
P311–330 deletion in NS1 vaccines — does exclusion abolish anti-PDI cross-reactivity? Cheng2015’s anti-P311–330 blocking assay demonstrates ~80% suppression of anti-PDI binding. Whether complete deletion of aa 311–330 (or aa 311–352) from an NS1-based vaccine construct abolishes anti-PDI cross-reactive immunogenicity while preserving protective efficacy has not been tested in any immunisation model. This is the most directly actionable translational question raised by the NS1 molecular mimicry literature.
Anti-hCF independent validation. The Chaturvedi2001 anti-protective autoantibody finding depends entirely on hCF being a real, characterisable entity. A literature search for independent citations of hCF, or a proteomics approach to identify the HPLC-purified dengue cytokine fraction described by the Chaturvedi group, would determine whether this axis warrants further investigation or should be retired as an artefact.

Epistemic Framework

Established — Convergent evidence across ≥3 independent sources

Dengue massively upregulates ANA during acute infection. ~55% by HEp-2 IIFA (Chatterjee2024), far exceeding the healthy-population baseline (13–16% at 1:80) and generic viral-infection rate (~22% by narrower ELISA; Berlin2007).
The vast majority (~66%) of this acute IIFA positivity is non-specific. The mechanistic basis is named: germline-encoded polyreactive IgM, constitutively present in all healthy individuals, binds nuclear antigens non-specifically as a normal immune property. Dengue’s inflammatory milieu amplifies this pool. The LIA-confirmed ~18.5% is the operationally meaningful measure of pathologically induced dengue autoimmunity (Zhou2007, Chatterjee2024, Vo2020).
Dengue autoimmunity manifests during the acute phase of infection, not post-infectious — an unusual and mechanistically distinctive timing (Lin2001, Lin2006, Lin2011, Wan2012, Cheng2015).
At the population level, dengue does not broadly elevate clinical SARD incidence. Only ADEM is robustly elevated (aHR 2.72), and only in the first month (Shih2023).
Dengue can produce clinically severe autoimmune complications (MAS/HLH) that are entirely ANA- and anti-dsDNA-negative. This establishes a parallel macrophage-driven autoimmune axis distinct from the NS1-mimicry → autoantibody pathway (Morel2014, Palacios2016/Lai 2012).
Dengue thrombocytopenia involves at least three immunologically distinct platelet-bound Ig populations: (1) a true IgM anti-platelet autoantibody in primary infection (Lin2001, Lin2006); (2) an anti-dengue IgG immune complex in secondary infection (Oishi2003); (3) an anti-dengue IgM immune complex in secondary infection (Saito2004), completely FcγR-independent, independently predictive of DHF at 92.1% specificity.
NS1 molecular mimicry anti-endothelial autoantibodies are infection-order independent. Three independent sources converge: Lin2001 (primary DENV-3), Cheng2015 (primary/secondary DHF comparison), Saito2004 (secondary-infection immune complexes are anti-dengue, not anti-self). These autoantibodies are constitutive features of NS1 immunity.
The PDI-cross-reactive NS1 epitope has been narrowed to P311–330. Deletion of this region abolishes anti-NS1-mediated platelet aggregation interference. The HSP60-cross-reactive epitope is distinct and unidentified (Cheng2015).
Anti-prM is the dominant structural antibody class in dengue (~60% of dengue-specific B cells). Anti-prM is fully cross-reactive DENV-1–4, cannot fully neutralise (structural ceiling from incomplete prM cleavage), and mediates 10^5-fold ADE in monocytes/DCs. All current vaccines use native prM (Dejnirattisai2010, Guzman2016).

Probable — Supported by ≥2 sources with methodological limitations

ANA positivity remains elevated at 2 years post-dengue in symptomatic patients (Garcia2009), consistent with the molecular mimicry + FcγRIIa IC-persistence model, but constrained by rat-liver substrate, no contemporaneous control, and a highly selected symptomatic cohort.
NS1 molecular mimicry — producing anti-PDI, anti-vimentin, anti-HSP60 autoantibodies — is the primary dengue-specific mechanism for the pathogenic autoimmune component of thrombocytopenia and vascular leakage (Lin2001, Lin2006, Lin2011, Wan2012, Guzman2016).
Bystander activation is likely insufficient alone to explain dengue-associated ANA elevation (inference from Johnson2022 COVID-19 ICU null finding).
Anti-NS1 IgG wanes with t½ ≈ 2.1 years post-primary dengue (Bos2025, preprint); the NS1-mimicry component of post-dengue ANA likely decays on a similar timescale. Garcia2009’s 2-year ANA persistence is more attributable to epitope-spreading and IC-persistence components than to ongoing NS1 mimicry.

Hypothesis-generating — Single source or indirect inference; requires replication

MCTD and autoimmune myositis may be specifically elevated among the LIA-confirmed dengue ANA subset (Chatterjee2024 — wide CIs, n=135, no clinical follow-up).
The “intrinsic ADE” mechanism may create a positive feedback loop between viral enhancement in secondary infection and autoantibody production, explaining why anti-platelet and anti-endothelial autoantibody levels are higher in DHF than DF (Wan2012).
Nuclear antigen IgGs (KU, Smith, histone, Sm/RNP, nucleosome) are positively correlated with platelet counts in DHF — interpreted as immune complex consumption (Vo2020, n=8 DHF; no multiple comparison correction). If confirmed: “ANA-negative severe dengue” may reflect antibody consumption rather than absence of nuclear antigen reactivity, reframing the prognostic interpretation of ANA in dengue entirely.
Primary DENV infection generates broader IgG autoantibody repertoires than secondary infection (Vo2020, n=6 primary — heavily confounded by sex and serotype), implying first-exposure dengue may represent the window of maximal polyclonal IgG autoantibody activation — a reversal of severity-centric expectations.
Cross-reactive E protein IgG actively rises between 6 and 18 months post-primary infection (Bos2025, preprint; t½ = −2.13 y); given E protein WGNGCG motif homology with coagulation factors (Lin2011), the anti-coagulation-factor component of dengue autoreactivity may follow an opposite and rising temporal vector to the anti-NS1 component within the same patient.
Dengue is associated with ANA-positive clinical SLE in independent case reports from three countries (Rajadhyaksha2012, Mumbai; Velazqueza2017, Mexico; Talib 2013, Singapore via Palacios2016). Consistent across diverse settings. Whether these represent de novo SLE induction or dengue unmasking pre-existing subclinical SLE is unresolved in every case — no pre-dengue autoantibody baseline exists for any of them.
Combined-isotype anti-cardiolipin (IgM + IgG) persisting to 4 months adds an antiphospholipid dimension to the dengue-SLE picture not previously documented (Rajadhyaksha2012).
The acute plasmablast clonotypes expanded in symptomatic dengue (IGHG1+/IGHA1+ plasmablasts in DHF by scRNA-seq; Sungnak2025) disappear at convalescence — whether autoreactive clones are cleared by post-infectious tolerance re-establishment, or whether they seed a persistent low-level autoreactive memory pool, is unknown.
Anti-hCF as a protective anti-cytokine autoantibody axis (Chaturvedi2001 — ⚠ unvalidated): 96% DF → 8% DHF grade IV positivity — the only dengue autoantibody with severity-inverse association in this wiki. Entirely contingent on hCF being independently validated and characterised. If real, represents a third ANA-invisible dengue autoimmune axis (anti-cytokine, protective) alongside the established pathogenic B-cell axis and the ANA-negative MAS macrophage axis.

Dengue Literature Review

Explorer

ANA and Dengue - Review V2.0