# HSEF v5.3 review: Cytokine Storm as Primary Driver

> Sources: 68 primary reports; claims: 74; verified: 74; unverified: 0. Verification is bounded attribution. Access: 8 partial-body and 60 complete abstract/status examinations. Cutoff: 2026-10-05. No complete-paper audit or blinded calibration claimed.
> **Adversarial review:** Generated by Sol 6.1; adversarially reviewed by Claude Opus 5.5 (Anthropic) on 2026-10-06. 4 corrections applied (see ADVERSARIAL REVIEW AND APPROVAL). Status: **Approved as revised.**

## AT A GLANCE

**Targeted cytokine signalling contributes causally in selected hospitalized COVID-19 populations. The evidence does not establish sepsis-like systemic hypercytokinemia as the primary cause of severe disease and mortality.**

| Level | Confidence | Causal role | Momentum | Stability | Divergence |
|---|---|---|---|---|---|
| Γ — Mechanistically plausible, partial system support | Moderate | reinforcing / stabilizing factor | Mechanistic only | Stable | Contested |

**Deciding experiment:** Prospectively enroll prespecified early severe-COVID strata across vaccination/variant and antiviral states; use harmonized contemporaneous blood and dilution-corrected pulmonary cytokine measurements. Randomize mechanistically aligned cytokine blockade with matched antiviral care and factorial vascular treatment where ethical; add a validated compartment-selective instrument in an infected model. Estimate prespecified mediators and attributable outcome effects without defining groups by treatment response.

**Belief versus evidence:** The preregistered scope-matched gate SC4 yields Γ / Moderate for systemic-storm primacy. SC3 earns Ζ for convergent targeted cytokine contribution in prespecified hospitalized contexts; this is explicitly retained. Weakest-link sensitivity is Β because SC1/SC2 are descriptive observations. Primary-subclaim and scope-matched sensitivities are Γ. Treatable inflammation and a cytokine-first systemic magnitude claim are different propositions.

**Best-supported rival:** Host-state and stage-dependent inflammatory contribution. Targeted IL6R benefit versus cytokine-level prediction distinguishes causal contribution from prognostic inflammation. Human inherited and neutralizing-autoantibody IFN defects distinguish a temporally upstream susceptibility pathway from the broad systemic-storm account, while leaving inflammatory amplification compatible.

## COMPETING MECHANISTIC MODELS

**Rank 1: Host-state and stage-dependent inflammatory contribution (Ε).** Specific inflammation contributes under defined clinical conditions rather than one universal systemic storm. Targeted IL6R/IL1 trials and eligibility/timing/dose contrasts [13-30,36] support causal boundaries. Whole-population dominance and mediator-specific clinical prediction remain incomplete; multiple treatment contexts are null.

**Rank 2: Early antiviral interferon failure in susceptible hosts (Ζ).** Inherited or autoimmune disruption of early IFN protection causes susceptibility before downstream inflammation. Human functional genetic and neutralizing-autoantibody natural experiments, preinfection samples and cellular vulnerability [43-50] reduce ordinary reverse-causation explanations. Selected genetic/autoantibody groups are not all cases; shared recruitment and comparator cohorts prevent treating every report as independent. Late IFN treatment can fail or harm [39].

**Rank 3: Endothelial and immunothrombotic injury (Ε).** Vascular and coagulation mechanisms interact with infection and inflammation. Stage-specific heparin contrast [37-38], fibrin domain intervention [55] and autopsy vascular pathology [57] narrow a real contribution. Autopsy is end-stage; heparin is not all immunothrombosis and its critical-care null constrains generalized efficacy.

**Rank 4: Pulmonary compartmental inflammatory injury (Γ).** Local cytokine, complement and repair abnormalities can differ from blood measurements. Paired blood/BAL and epithelial circuit studies [11,54,62,65-67] support local mechanism plausibility. No examined human compartment-restricted hard-outcome test establishes pulmonary primacy; raw BAL dilution and treatment matter.

**Rank 5: Systemic cytokine-storm primacy (Γ).** Comparator-magnitude circulating cytokines dominate severe COVID pathology and mortality. Prediction and positive cytokine-directed interventions support an inflammatory component [5-7,13-15,18]. Comparator magnitude, trial nulls and susceptibility/compartment alternatives block universal primacy [1-2,16-30,43-50,54,68].

The tournament does not alter the nominal level. Compatible pathways and flexible alternatives are not automatically proved by explaining observations.

## FULL REVIEW

### Registered claim and scope

Systemic hypercytokinemia at a magnitude comparable to sepsis, hyperinflammatory ARDS or CAR-T cytokine release syndrome is the primary driver of severe COVID-19 pathology and mortality.

Scope-matched follows the index. SC4 gates the broad primacy claim; positive treatment contribution in SC3 is preserved separately and cannot establish SC2 magnitude or SC4 primacy.

Selection: public-interest rank 9 from the editorial ranking compiled 2026-09-23; categorical prior verdicts were visible. Private fresh review; earlier documents are preserved.

### Evidence and causal interpretation

The claim joins three burdens: cytokine elevation, comparator-defined systemic magnitude, and primary causal dominance. Only the first is directly established by blood prediction studies. The review preserves strong treatment evidence while asking what it identifies about the nominated concentration-driven systemic mechanism.

Admission and pre-intubation studies repeatedly predict worsening [3-7,58]. Del Valle examined1484 patients and validated in231 [5]; Herold used development and temporal validation cohorts [6]. These are valuable risk instruments. A high within-COVID hazard or prediction AUC does not establish a sepsis-level cytokine distribution or distinguish production from tissue injury.

Kox used the same assay method and technician but different lots and sampling windows [1]. COVID ARDS IL6 was lower than septic-shock ARDS, despite severe lung injury. Historical controls and overall severity prevent a perfectly matched inference. Wilson instead reports normalized Luminex MFI with small groups [2]. Its absence of significant differences supports no detected unique excess, not quantitative equivalence with sepsis. Mudd found a storm-like profile in7 of168 COVID patients and lower adjusted cytokines than influenza [68]. A selected extreme subgroup must not redefine the general registered scope.

Functional immune suppression can coexist with inflammation. Remy showed low stimulated leukocyte output and ex vivo IL7 rescue [8]; a time-course comparison showed prolonged inflammatory and suppressive changes [9]. Kox-network profiling found innate hyperresponsiveness and adaptive exhaustion [10]. These apparently opposed results depend on cell type, stimulus and timing. Neither title-level inflammation nor immunosuppression is a complete disease mechanism.

RECOVERY provides a clean positive causal instrument: hypoxia plus CRP≥75, primary28-day ITT mortality, RR0.85, with82% receiving steroids [13]. Original sections confirm concealed allocation, open-label care and near-complete outcome follow-up. REMAP-CAP targeted the first24h of ICU organ support and improved a death/organ-support composite and survival [14]. EMPACTA improved ventilation-or-death but not mortality alone [15]. Together they refute a wholly passive IL6 pathway in these contexts; they do not demonstrate that extreme circulating cytokines dominate pathology.

COVACTA, BACC and the blinded sarilumab trials were null for their primary endpoints [16-17,21-22]. SARICOR trended rather than confirmed [23]. REMDACTA remained null despite high steroid use [29], so a blanket explanation that steroids reconcile every null is insufficient. Eligibility, timing, dose, background care and precision differ. No numerical pooling is attempted without a systematic harmonized analysis.

SAVE-MORE gives an orthogonal positive IL1 instrument [18]. Original Methods confirm suPAR≥6, exclusion of P/F<150 and NIV/MV, blinded administration and outcome review. WHO-CPS was the primary endpoint; the mortality result was a separate survival analysis. CORIMUNO-ANA, CAN-COVID, COV-AID and other anakinra trials were null or concerning [19-26]. Expanded REMAP-CAP found anakinra ineffective in critical organ-support patients [30]. That2025 paper extends2020-2021 enrollment rather than independent later-era replication.

Broad anti-inflammatory therapy also has boundaries. Low-dose dexamethasone benefits oxygen/ventilation groups [31], whereas higher doses caused harm on simple oxygen [36]. Baricitinib improved recovery in ACTT2 and primary mortality in RECOVERY [33,35]. COV-BARRIER missed its primary composite; its secondary mortality p-values were nominal because the hierarchical procedure failed [34]. Original Methods/Results make that explicit. ACTT4 active-treatment similarity is not proof of mechanistic equivalence [40].

Early IFN failure has stronger directionality than ordinary cytokine correlation. Genetic defects precede infection and functionally increase viral vulnerability [44,47-49]. Neutralizing autoantibodies block antiviral protection and some preinfection samples were available [43,45]. These convergent natural experiments support a scoped susceptibility pathway. They do not account for all severe cases. Late IFNbeta therapy was null and worse in high-flow patients [39], illustrating why early protective necessity does not guarantee late pharmacological rescue.

Viral and vascular processes remain experimentally relevant. Remdesivir shortened recovery without a definitive mortality estimate in ACTT1 [32], while its virology substudy linked blood-marker clearance to benefit [42]. Tixagevimab/cilgavimab missed primary sustained recovery but had lower secondary mortality [41]. Heparin improved a discharge-without-support outcome in noncritical patients but was futile in critical care [37-38]. Those contrasts identify stage-specific opportunities, not exclusive viral or vascular primacy.

Paired compartments constrain inference from blood. Blood/BAL correlation was mediator-specific in a severe influenza/COVID comparison [11]. Ronit found broad elevations in both compartments and particularly local IL8 in four cases [67]; the ELF-corrected2024 study used nonrandomized steroid-era groups, with different gas exchange [54]. BAL autoantibody binding and an unknown positivity cutoff cannot be equated with functional circulating neutralization. These reports share a hospital platform and possible cohort overlap is left unresolved.

Autopsy and single-cell atlases show vascular injury, immune activation and failed repair [53,56-57,59,61,66]. They are endpoint anatomy and transcriptomes rather than initiating interventions. Fibrin-domain targeting and IL13/CD44 experiments narrow real model contributors [55,63]. The2025 IL1beta knockout benefit alongside inflammasome/pyroptosis nulls [64] warns that the cytokine mediator can matter while a presumed processing mechanism is dispensable.

Consensus improved discovery and highlighted assay, endpoint, timing and compartment gaps. Targeted full-text extraction correctly recovered Kox and RECOVERY aspects that were checked against the publisher and original XML. It could not retrieve Wilson as named, SAVE-MORE exclusions/correction or the original REMAP-CAP24h sentence. Its substitute Shah2025 citations reproduced target-paper material without reliable identity, so those attributions were rejected; original Wilson/SAVE-MORE sections and exact PubMed records governed the review. The Deep meter said50% Yes/50% No while its prose called it a clear no; neither is an HSEF score.

Eight original papers were examined in selected body sections, including a publisher-web receipt for Kox. Other sources have complete abstracts/status checks. Linked corrections to SAVE-MORE, COV-BARRIER, RECOVERY-baricitinib, IFN rare-variant work, biomarkers and atlases remain flagged, with correction effects incompletely audited. These access and family limits set Moderate overall confidence and preclude a complete-paper or publication-ready numerical audit claim.

## HYPOTHESIS STATE EVALUATION FRAMEWORK (HSEF v5.3)

<!-- consensus-consistency-2026-10-09 -->
### Consensus provenance reconciliation — 9 October 2026

Case-specific Consensus discovery records were located and linked to this final edition. This supplements the review's provenance; it leaves the original assessed claim, grades and source counts intact. Provider summaries and full-text badges do not establish examination of original primary passages by this reviewer. Source access and incorporation retain the limits stated in the original methods and any dated research update.

Recorded evidence artifacts:

- `deliverables/hsef_v53_popularity_ranks_06_10_2026-10-05/cytokine-storm-covid-driver/consensus/consensus-result.png`
- `deliverables/hsef_v53_popularity_ranks_06_10_2026-10-05/cytokine-storm-covid-driver/consensus/deep-dom.txt`
- `deliverables/hsef_v53_popularity_ranks_06_10_2026-10-05/cytokine-storm-covid-driver/consensus/deep-response.txt`
- `deliverables/hsef_v53_popularity_ranks_06_10_2026-10-05/cytokine-storm-covid-driver/consensus/final-dom.txt`
- `deliverables/hsef_v53_popularity_ranks_06_10_2026-10-05/cytokine-storm-covid-driver/consensus/fulltext-check-response.txt`

The located artifacts do not state a reliable source-comparison count. No count is inferred.

A dated adversarial approval record exists for the separately identified cytokine-storm-covid-driver.review.md (2026-10-06; see edition-approval-audit.json). Its scope is that recorded edition. This supplement has no separate approval and no coverage is inferred from topic identity. No score or confidence change is made by this reconciliation.
<!-- end-consensus-consistency -->


### Subclaim matrix and aggregation

| Subclaim | Scope | Level | Decision |
|---|---|---|---|
| SC1 Circulating IL-6 is elevated in severe COVID-19 and predicts deterioration and death. | defined hospitalized severe COVID-19 cohorts | Β | Circulating IL6 and related markers predict deterioration and survival, including temporal validation [3-7,58]. This is a strong prognostic observation, not causal primacy. |
| SC2 Circulating cytokine elevation reaches comparator-defined storm magnitude. | severe/critical COVID-19 compared with sepsis, ARDS and CAR-T CRS | Β | Direct comparative observations are heterogeneous. The same-method ICU comparison and influenza comparison constrain general storm magnitude [1,68]; the small MFI-panel null [2] is not equivalence. The registered magnitude claim is not established across its scope. |
| SC3 IL-6/IL-1 pathway signalling contributes causally to adverse outcomes in prespecified hospitalized populations. | trial-defined hypoxaemic/inflammation-marker-positive hospitalized patients | Ζ | IL6R-targeted mortality/organ-support benefits in separate RECOVERY and REMAP-CAP populations, placebo-controlled composite benefit in EMPACTA, and mechanistically different IL1 perturbation in SAVE-MORE reduce a purely passive-inflammatory-marker interpretation [13-15,18]. IL6R blockade is one causal line under the framework's independence test (agents of the same modality against the same target do not count twice, and replication adds confidence rather than a line); Ζ rests on its convergence with the mechanistically distinct IL1 blockade result and the infected-model IL1beta genetic contrast. IL1 clinical contribution is more selection-sensitive and bounded at Ε. Nulls identify boundaries rather than erasing all causal contribution [16-26,29-30]. |
| SC4 Systemic hypercytokinemia is the primary driver across the severe/critical population. | severe/critical COVID-19, baseline setting and treatment era explicit | Γ | Positive pathway-specific trials establish contribution, but cannot identify circulating magnitude or the fraction of disease primarily caused by that state. Host IFN natural experiments, antiviral benefit, vascular injury and compartment data make systemic-storm primacy mechanistically unestablished [32,37-50,54-57,68]. |
| SC5 The pathogenic cytokine action is systemic rather than primarily pulmonary-compartmental. | matched blood and pulmonary samples | Γ | Systemic drug delivery does not localize pathogenic action. Paired blood/lung measurements show mediator-specific compartment relations and treatment effects, not proof that the disease-driving cytokine action is primarily systemic [11,51-54,65-67]. |

Preregistered aggregation: scope-matched. The preregistered scope-matched gate SC4 yields Γ / Moderate for systemic-storm primacy. SC3 earns Ζ for convergent targeted cytokine contribution in prespecified hospitalized contexts; this is explicitly retained. Weakest-link sensitivity is Β because SC1/SC2 are descriptive observations. Primary-subclaim and scope-matched sensitivities are Γ. Treatable inflammation and a cytokine-first systemic magnitude claim are different propositions. Sensitivities: weakest-link Β; primary-subclaim Γ; scope-matched Γ.

### Decisive-perturbation inventory and causal tally

[13] RECOVERY tocilizumab reduced primary 28-day ITT mortality in hypoxic CRP-at-least-75 patients, 31% versus 35%, RR0.85. Open label; 82% on steroids; effect identifies IL6R contribution, not circulating concentration, compartment or primary cause.

[14] REMAP-CAP IL6R blockade improved organ-support-free days and survival within 24h of ICU organ support. Adaptive open-label platform and concomitant treatments; initial sarilumab arm small; does not identify systemic magnitude or primacy.

[15] EMPACTA reduced ventilation-or-death composite in nonventilated hospitalized patients. Mortality alone did not improve; composite benefit must not be called survival benefit.

[18] SAVE-MORE improved primary WHO-CPS with anakinra in suPAR-at-least-6 patients; mortality was lower in a separate survival analysis. Excluded P/F below150 and NIV/MV; 85.9% on dexamethasone; WHO endpoint hierarchy differs from mortality-primary trials; linked correction not fully audited.

[30] Expanded REMAP-CAP cohort supported equivalent IL6R agents and a null anakinra effect in critical patients. 2025 publication enrolled March2020-April2021 and extends the original domain; not a fresh 2025 population replication.

[31] RECOVERY low-dose dexamethasone reduced mortality on oxygen or ventilation, without benefit in no-oxygen patients. Pleiotropic steroid effects do not isolate a cytokine; same platform can share co-randomized patients and controls.

[32] ACTT1 remdesivir improved primary recovery time versus placebo. Mortality CI included no effect; antiviral efficacy does not quantify the fraction of death driven by virus versus inflammation.

[33] ACTT2 baricitinib plus remdesivir accelerated recovery versus remdesivir. Mortality interval included null; JAK1/2 affects several pathways and cannot identify one circulating cytokine.

[35] RECOVERY baricitinib reduced primary ITT mortality, RR0.87, on extensive steroid background. JAK blockade is broader than IL6R blockade; concurrent tocilizumab and shared platform complicate line-counting.

[38] Therapeutic heparin improved survival-to-discharge without organ support in noncritical hospitalized patients. Bleeding increased; anticoagulant and other heparin actions do not isolate systemic cytokine primacy.

[63] IL13 neutralization reduced infected-mouse mortality without lowering virus; CD44 blockade narrowed HA mediation. Human dupilumab observations confounded; IL13/HA is not an IL6/IL1 clinical test nor systemic-storm magnitude proof.

[64] IL1beta loss improved infected-mouse disease/survival while several inflammasome/pyroptosis deletions did not. Germline developmental and mouse-adapted-virus scope; cytokine causality does not establish the presumed release pathway.

Tally: 12 discriminating contrasts. Null and negative contrasts are discriminators, not positive threshold crossings. Reports and replications are not independent causal lines. Positive causal flags are scoped disease-modifying contrasts, not independent lines or primacy successes. Broad steroid/JAK and antiviral/heparin benefits are distinguished from single-target cytokine evidence. IFN natural experiments support the rival, not the nominal cytokine claim.

### Cap audit, precedence and Β/Γ/Δ boundary

A perturbation crosses a threshold within its tested system; scope governs the inference beyond it. Rubric05 explicitly recognizes replicated single-target biologic hard-endpoint tests and inborn-errors natural experiments. SC3 is not held at Γ merely because other legs are observational. The full clinical convergent set supports Ζ contribution. It cannot identify systemic storm magnitude, localization or primacy, so SC4/SC5 remain Γ through missing causal attribution rather than a cap on the weakest study.

Gamma hard-rule audit: Cytokine profiles and ex vivo studies alone cannot cross the disease-causality boundary. Broad steroids/JAK interventions are not single-cytokine instruments. Targeted IL6R/IL1 clinical trials receive their causal credit in SC3. No blanket ex vivo or correlational cap is reapplied to the assembled clinical causal set. Primacy and compartment claims fail different, scope-specific causal burdens.

SC1 cap: “Inflammation association without pathway-specific causality caps at Β.” Circulating IL6 and related markers predict deterioration and survival, including temporal validation [3-7,58]. This is a strong prognostic observation, not causal primacy.

SC1 field-valid definition: “
- Cell-type-specific conditional knockout or depletion
- Pathway-specific blockade (not broad immunosuppression) changing disease course
- Human targeted therapy with mechanistic alignment showing disease-modifying effect” Positive full-scope result: no. Circulating IL6 and related markers predict deterioration and survival, including temporal validation [3-7,58]. This is a strong prognostic observation, not causal primacy.

SC2 field-valid definition: “
- Cell-type-specific conditional knockout or depletion
- Pathway-specific blockade (not broad immunosuppression) changing disease course
- Human targeted therapy with mechanistic alignment showing disease-modifying effect” Positive full-scope result: no. Direct comparative observations are heterogeneous. The same-method ICU comparison and influenza comparison constrain general storm magnitude [1,68]; the small MFI-panel null [2] is not equivalence. The registered magnitude claim is not established across its scope.

SC3 field-valid definition: “
- Cell-type-specific conditional knockout or depletion
- Pathway-specific blockade (not broad immunosuppression) changing disease course
- Human targeted therapy with mechanistic alignment showing disease-modifying effect” Positive full-scope result: yes. IL6R-targeted mortality/organ-support benefits in separate RECOVERY and REMAP-CAP populations, placebo-controlled composite benefit in EMPACTA, and mechanistically different IL1 perturbation in SAVE-MORE reduce a purely passive-inflammatory-marker interpretation [13-15,18]. IL6R blockade is one causal line under the framework's independence test (agents of the same modality against the same target do not count twice, and replication adds confidence rather than a line); Ζ rests on its convergence with the mechanistically distinct IL1 blockade result and the infected-model IL1beta genetic contrast. IL1 clinical contribution is more selection-sensitive and bounded at Ε. Nulls identify boundaries rather than erasing all causal contribution [16-26,29-30].

SC4 field-valid definition: “
- Cell-type-specific conditional knockout or depletion
- Pathway-specific blockade (not broad immunosuppression) changing disease course
- Human targeted therapy with mechanistic alignment showing disease-modifying effect” Positive full-scope result: no. Positive pathway-specific trials establish contribution, but cannot identify circulating magnitude or the fraction of disease primarily caused by that state. Host IFN natural experiments, antiviral benefit, vascular injury and compartment data make systemic-storm primacy mechanistically unestablished [32,37-50,54-57,68].

SC5 field-valid definition: “
- Cell-type-specific conditional knockout or depletion
- Pathway-specific blockade (not broad immunosuppression) changing disease course
- Human targeted therapy with mechanistic alignment showing disease-modifying effect” Positive full-scope result: no. Systemic drug delivery does not localize pathogenic action. Paired blood/lung measurements show mediator-specific compartment relations and treatment effects, not proof that the disease-driving cytokine action is primarily systemic [11,51-54,65-67].

### Upward threshold ladder

SC1 Δ: “Requires pathway- or cell-type-resolved causal perturbation affecting disease or phenotype.” Satisfied: no. Prediction alone cannot become a causal claim; retain descriptive scope.

SC2 Δ: “Requires pathway- or cell-type-resolved causal perturbation affecting disease or phenotype.” Satisfied: no. Harmonized assay, lot, symptom timing, treatment and severity comparators must establish the preregistered magnitude distribution rather than a selected extreme subgroup.

SC3 Δ: “Requires pathway- or cell-type-resolved causal perturbation affecting disease or phenotype.” Satisfied: yes. 

SC3 Ε: “Requires context-specific causality with explicit scope restriction.” Satisfied: yes. 

SC3 Ζ: “Requires multiple convergent causal manipulations and reduced plausibility of downstream-only interpretations.” Satisfied: yes. 

SC3 Η: “Requires robust upstream causal evidence across systems and interventions.” Satisfied: no. To reach Η, robust mechanistically aligned contributions must generalize prospectively across prespecified antiviral states, treatment eras and compartments without relying on inferred primacy.

SC4 Δ: “Requires pathway- or cell-type-resolved causal perturbation affecting disease or phenotype.” Satisfied: no. Estimate cytokine-attributable disease burden with early specific pathway/compartment interventions versus viral and vascular alternatives; benefit must identify the dominant lesion rather than merely an amplifier.

SC5 Δ: “Requires pathway- or cell-type-resolved causal perturbation affecting disease or phenotype.” Satisfied: no. Compare validated compartment-restricted perturbations with matched target engagement, systemic exposure and hard outcomes; circulating biomarkers alone do not establish localization.

Upper ladder: Target-specific IL6R hard-outcome studies form one causal line; together with the mechanistically distinct IL1 blockade result and the infected-model IL1beta contrast they carry SC3 to Ζ; functional human IFN genetic/autoimmune natural experiments support a scoped Ζ rival. Higher levels are live. These instruments do not establish comparator-magnitude systemic hypercytokinemia as the dominant lesion, so SC4 remains Γ without re-capping convergence. Broad immunosuppression contributes less specificity; a generic RCT requirement is not imposed.

Independence: Ζ SC3 rests on randomized targeted effects, placebo controls and orthogonal mechanistic support that address different errors, not the number of papers or sites. The IFN natural-experiment line below is recorded because it bears on the rival susceptibility model; it is not counted toward the SC3 convergence. RECOVERY arms share a platform/co-randomization; REMAP-CAP2021/2025, SAVE-MORE primary/substudy, COVACTA primary/biomarkers and ACTT1 primary/virology are report families. The IFN-autoantibody extension includes566 original critical patients and427 controls; fatality estimation reuses the34159 population controls. The IFN rival combines temporally prior genetic and autoimmune causes with functional assays rather than counting those overlapping reports separately.

Target-specific clinical perturbation [13, 14, 15, 18]: Randomization reduces severity-driven prescribing; placebo-controlled EMPACTA/SAVE-MORE reduce open-label management bias. Distinct targets weaken a single off-target explanation, but eligibility and outcomes differ.

Temporally prior human IFN natural experiments [44, 43, 45, 47, 48]: Inherited defects and documented preinfection autoantibodies reduce reverse causation; functional neutralization/infection tests reduce a binding-marker-only explanation. Recruitment and cohort overlap remain.

Mediator and pathway contrasts in infected models [63, 64, 55]: Orthogonal mediator/domain manipulations separate inflammatory contribution from blood association and from assumed processing routes, with model/developmental limits.

### Falsification and pitfalls

Specific block is context-dependent. Several IL6/IL1 trials have null primary outcomes despite inflammatory enrollment [16-26,29-30]. Additional downgrade: False.

Mechanism inferred from a marker is not necessary. IFN susceptibility and compartment injury are compatible with severe disease without universal comparator-magnitude blood cytokines [1,43-50,54,68]. Additional downgrade: False.

Release pathway is dispensable despite mediator contribution. IL1beta loss helped infected mice while multiple canonical processing/pyroptosis deletions did not [64]. Additional downgrade: False.

Therapeutic benefit converted to primacy. RECOVERY/REMAP-CAP pathway effects do not locate cytokine action or quantify dominance [13-14].

Prognostic marker treated as predictive treatment selector. IL6 prediction and exploratory ferritin interactions have different evidential roles [5-6,28].

Null treated as equivalence. Wilson MFI panel and ACTT4 active comparator are not equivalence proofs [2,40].

Secondary outcome promoted to primary success. EMPACTA, COV-BARRIER and tixagevimab trials have different endpoint hierarchies [15,34,41].

Binding antibody confused with neutralization. BAL IFN autoantibody ELISA differs from functional neutralization assays [45,54].

Publication year mistaken for enrollment era. The expanded REMAP-CAP report was published in2025 but enrolled in2020-2021 [30].

### Adversarial steelman: above, then below

The strongest case for Δ primacy is that specific IL6R blockade repeatedly changes survival or organ support, early suPAR-guided IL1 blockade works, and mouse cytokine perturbations change disease despite unchanged virus [13-15,18,63-64]. Those findings decisively reject a wholly passive inflammatory role.

The strongest case for Β is that the blood-magnitude and primacy claims remain observational and direct comparator studies challenge a general storm [1-2,68]. It misses disease-modifying, target-specific clinical perturbations and mechanistic mediator experiments.

Reject Δ for the registered primacy gate because contribution, magnitude and localization have not been causally joined. Reject Β for the overall mechanistic hypothesis because active cytokine contributions are established. Retain Γ for nominal primacy and explicitly preserve Ζ for SC3 rather than downgrade the genuine clinical causal evidence.

### Final score, causal direction and M/S/D

The preregistered scope-matched gate SC4 yields Γ / Moderate for systemic-storm primacy. SC3 earns Ζ for convergent targeted cytokine contribution in prespecified hospitalized contexts; this is explicitly retained. Weakest-link sensitivity is Β because SC1/SC2 are descriptive observations. Primary-subclaim and scope-matched sensitivities are Γ. Treatable inflammation and a cytokine-first systemic magnitude claim are different propositions.

Specific cytokine signalling can amplify disease after infection and antiviral failure. Human targeted interventions establish contribution and some survival effects, but do not establish initiation, necessary storm magnitude, or primary systemic action. No rate-limiting or universal mechanism role is inferred from a treatment response.

Momentum: Mechanistic only. The 2025 infected-mouse IL1beta versus inflammasome/pyroptosis contrasts [64] and expanded REMAP-CAP IL6/IL1 comparison [30] discriminate mediators and timing; the latter is an older-enrollment cohort extension.

Stability: Stable. No verified imminent trial is used to forecast a level change. Assay-harmonized early paired sampling and compartment-specific intervention could change the primacy inference, but no such pending result is claimed.

Divergence: Contested; no single implied-level arithmetic. These dimensions do not change level or confidence.

## PROPOSED DECISIVE EXPERIMENTS

Prospectively enroll prespecified early severe-COVID strata across vaccination/variant and antiviral states; use harmonized contemporaneous blood and dilution-corrected pulmonary cytokine measurements. Randomize mechanistically aligned cytokine blockade with matched antiviral care and factorial vascular treatment where ethical; add a validated compartment-selective instrument in an infected model. Estimate prespecified mediators and attributable outcome effects without defining groups by treatment response.

Readout: Mortality and organ-support-free days, viral burden, cytokine target engagement in both compartments, vascular injury and repair, with prespecified longitudinal mediator models. Controls: Matched assays/lots and sampling, blinded outcome assessment, baseline steroid/antiviral stratification, drug exposure, non-COVID infectious ARDS comparators and registered multiplicity. Supporting: Comparator-magnitude circulating elevation temporally precedes injury and explains the dominant specifically reversible disease component across prespecified strata after viral/vascular alternatives are addressed. Falsifying: Severe injury and mortality persist without the specified systemic magnitude, local target engagement predicts benefit better than blood magnitude, or independently controlled viral/vascular processes explain the predominant outcome effect.

## RESEARCH TIMELINE

| Year | Milestone | Primary source | Interpretation |
|---|---|---|---|
| 2020 | Admission IL6/TNF survival prediction | [PMID32839624](https://pubmed.ncbi.nlm.nih.gov/32839624/) | Prognosis does not identify primary cause |
| 2020 | Critical-illness cytokine comparison | [PMID32880615](https://pubmed.ncbi.nlm.nih.gov/32880615/) | Harmonized platform with lot/timing caveats |
| 2020 | Small MFI cytokine comparison | [PMID32706339](https://pubmed.ncbi.nlm.nih.gov/32706339/) | Nonsignificance is not equivalence |
| 2020 | IFN genetic susceptibility | [PMID32972995](https://pubmed.ncbi.nlm.nih.gov/32972995/) | Temporally prior antiviral mechanism |
| 2020 | IFN-neutralizing autoantibodies | [PMID32972996](https://pubmed.ncbi.nlm.nih.gov/32972996/) | Functional human susceptibility subgroup |
| 2021 | RECOVERY tocilizumab | [PMID33933206](https://pubmed.ncbi.nlm.nih.gov/33933206/) | Mortality-primary target-specific benefit |
| 2021 | REMAP-CAP IL6R | [PMID33631065](https://pubmed.ncbi.nlm.nih.gov/33631065/) | Early organ-support contribution |
| 2021 | SAVE-MORE anakinra | [PMID34480127](https://pubmed.ncbi.nlm.nih.gov/34480127/) | suPAR-enriched pre-ventilation benefit |
| 2021 | COV-AID factorial null | [PMID34756178](https://pubmed.ncbi.nlm.nih.gov/34756178/) | Inflammatory selection does not guarantee benefit |
| 2022 | RECOVERY baricitinib | [PMID35908569](https://pubmed.ncbi.nlm.nih.gov/35908569/) | Broader signalling contribution |
| 2023 | Higher-dose steroid harm | [PMID37060915](https://pubmed.ncbi.nlm.nih.gov/37060915/) | Dose/support boundary |
| 2024 | ELF-corrected paired lung/blood study | [PMID38571960](https://pubmed.ncbi.nlm.nih.gov/38571960/) | Compartment and steroid confounding |
| 2024 | Fibrin inflammatory-domain perturbation | [PMID39198643](https://pubmed.ncbi.nlm.nih.gov/39198643/) | Vascular-inflammatory pathway |
| 2025 | Expanded REMAP-CAP report | [PMID40360262](https://pubmed.ncbi.nlm.nih.gov/40360262/) | Older cohort extension, IL1 critical-care null |
| 2025 | IL1beta versus pyroptosis genetics | [PMID40016339](https://pubmed.ncbi.nlm.nih.gov/40016339/) | Mediator and release mechanism dissociate |

## KNOWLEDGE GAPS

Early longitudinal paired lung/blood cytokine measurements with harmonized assays, BAL dilution correction and matched non-COVID comparator severity.

Prospective predictive treatment selectors that separate prognosis from pathway-specific response without post hoc responder definitions.

Compartment-restricted causal interventions with validated target engagement and hard outcomes; systemic delivery cannot identify location.

Factorial antiviral, cytokine and vascular tests with prespecified mediation and outcome attribution, preserving endpoint hierarchy.

Reassess efficacy boundaries in vaccinated contemporary-variant patients rather than equating recent publication with current enrollment.

Audit linked corrections, protocols and complete supplements before publication; quantify report-family and biobank overlap.

## METHODS AND CONSENSUS CONTRIBUTION

Consensus was used through the signed-in browser for discovery, counterevidence, research-gap and timeline suggestions, and checks of specific indexed full-text passages. Extracted Methods/Results details were reconciled with original-paper passages where available. Provider screening counts, citation counts and meters were not used as primary counts or HSEF scores. This source log records access depth; it does not claim complete-paper or supplement review. The monthly API allowance was conserved: zero Consensus API calls in this batch.

## REVIEWER NOTES

68 primary reports individually appraised; eight selected original partial bodies; zero complete-paper/supplement audits. Secondary reviews and protocols do not pad the count. Some related reports share cohorts, controls or platforms. The ranking prior and index were visible; this is not blinded or an exhaustive systematic review.

Search leads, reviews, protocols and correction notices do not pad the primary count. Downloaded XML does not earn examination credit. Linked notices remain attached; unexamined correction details limit inference. This is a coordinator source/claim consistency assessment, not independent blinded calibration.

## ADVERSARIAL REVIEW AND APPROVAL

**Authoring model:** Sol 6.1. **Adversarial reviewer:** Claude Opus 5.5 (Anthropic). **Date:** 2026-10-06.

**Scope of checks.** All 68 reference PMIDs re-verified against PubMed records (title, first author, journal, year, DOI, status flags): 68/68 resolve to the stated work; no retractions or expressions of concern; the only year differences are print-issue versus e-pub year and are not errors. 28 load-bearing claims spot-checked against PubMed abstracts and the local `full-text/` and `body-presentation-receipts/` passages (RECOVERY tocilizumab and baricitinib, REMAP-CAP 2021 and 2025, EMPACTA, COVACTA, BACC Bay, SAVE-MORE, CORIMUNO-ANA-1, CAN-COVID, COV-AID, sarilumab trials, REMDACTA, ACTT-1/2/3/4, RECOVERY dexamethasone and higher-dose steroids, ATTACC/ACTIV-4a/REMAP-CAP heparin, TICO tixagevimab-cilgavimab, Kox, Wilson, Del Valle, Mudd, Remy, Reynolds, Jonassen, type I IFN genetic and autoantibody reports). Scoring logic re-derived from `framework/hsef-hypothesis-review-v5/SKILL.md` and `references/hsef-framework.md`: cap order, the scope-matched aggregation prescribed by the index, the Β/Γ/Δ boundary, the Ζ independence basis, and subclaim-level consistency.

| # | Severity | Location | Original → corrected | Evidence |
|---|---|---|---|---|
| 1 | Major | Subclaim matrix SC3 rationale; Upward threshold ladder; Independence paragraph (and the same fields in the score card) | "IL6 contribution reaches Ζ" / "Replicated target-specific IL6R hard-outcome studies carry SC3 to Ζ" → IL6R blockade is recorded as a single causal line, with Ζ resting on its convergence with the mechanistically distinct IL-1 blockade result and the infected-model IL-1β genetic contrast | Framework, "What makes causal lines independent": "the same mechanism delivered by a different agent of the same modality — two monoclonal antibodies against the same target are one line" and "replication, however extensive. Replication raises confidence; it does not add a line". Tocilizumab and sarilumab are two anti-IL-6R antibodies, so RECOVERY, REMAP-CAP and EMPACTA are replication of one line. SC3 remains Ζ on the corrected basis. |
| 2 | Minor | Independence paragraph and `independence_basis` in the score card | The type I IFN natural-experiment line was listed among the bases for the SC3 Ζ → it is now labelled as bearing on the rival susceptibility model and explicitly not counted toward the SC3 convergence | The review's own causal tally states "IFN natural experiments support the rival, not the nominal cytokine claim"; listing it as an SC3 independence line contradicted that. |
| 3 | Minor | AT A GLANCE table; "Momentum:" line; `extended_scoring_MSD.momentum.value` | Momentum "Forward" → "Mechanistic only" | "Forward" is not a value on the framework's Momentum scale (Decisive work underway / Mechanistic only / Associational only / Little new work / Abandoned). The recorded artifact describes active mechanistic discrimination with no pending threshold-crossing trial, which is the definition of "Mechanistic only". Level and confidence are unchanged, per "M/S/D never modify the level or confidence". |
| 4 | Editorial | REFERENCES 13, 14, 19, 31, 35, 36, 37, 38, 41 | Collective authors dropped by the reference builder (entries beginning with a comma or with " et al.") → restored as RECOVERY Collaborative Group; REMAP-CAP Investigators; CORIMUNO-19 Collaborative group; REMAP-CAP, ACTIV-4a and ATTACC Investigators; ATTACC, ACTIV-4a and REMAP-CAP Investigators; ACTIV-3-Therapeutics for Inpatients with COVID-19 (TICO) Study Group | CollectiveName fields in the PubMed records for PMIDs 33933206, 33631065, 33493450, 32678530, 35908569, 37060915, 34351722, 34351721, 35817072. |

**Rejected findings (checked, no change).** (a) SC2 is graded Β while the index calls the magnitude claim refuted — Β is the correct ladder position for a claim whose comparative evidence is observational and heterogeneous, and the review states the claim is not established across its scope; no ladder letter means "refuted". (b) Overall Γ by scope-matched aggregation — SC4 is the subclaim whose scope matches the registered claim, so Γ is the rule applied as written, with the Ζ of SC3 reported separately as the framework requires. (c) Divergence "Contested" without arithmetic — the framework lists Contested as a value for which no single implied level applies. (d) Reference years differing by one from the PubMed e-pub year (PMIDs 33348065, 33332779, 34902262, 32678530, 33306283, 33079210, 34723652, 33146564, 32979342) — print/issue-year convention, not an error. (e) The quantitative claims checked against abstracts (RR 0.85 and 31% vs 35% for RECOVERY tocilizumab, RR 0.87 for RECOVERY baricitinib, 85.9% dexamethasone in SAVE-MORE, 88.2% steroids in REMDACTA, 7 of 168 storm-like profiles in Mudd, 1484 and 231 in Del Valle, enrolment 2020-2021 in the expanded REMAP-CAP report) all match their sources.

**Residual limitations.** Verification remains bounded attribution: 8 partial-body and 60 abstract/status examinations, with no complete-paper or supplement audit, and linked corrections (SAVE-MORE, COV-BARRIER, the IFN rare-variant work) still unaudited. This adversarial pass checked citation integrity, attribution and framework conformance; it is not independent blinded peer review.

**Approval:** Following adversarial review and the corrections listed above, the revised document is approved. — Claude Opus 5.5 (Anthropic), 2026-10-06.

## REFERENCES

1. Kox M, Waalders NJB, Kooistra EJ et al. Cytokine Levels in Critically Ill Patients With COVID-19 and Other Conditions. JAMA. 2020. [PMID 32880615](https://pubmed.ncbi.nlm.nih.gov/32880615/). DOI: 10.1001/jama.2020.17052. Access: full_text_partial. Study: PMID32880615.

2. Wilson JG, Simpson LJ, Ferreira AM et al. Cytokine profile in plasma of severe COVID-19 does not differ from ARDS and sepsis. JCI insight. 2020. [PMID 32706339](https://pubmed.ncbi.nlm.nih.gov/32706339/). DOI: 10.1172/jci.insight.140289. Access: full_text_partial. Study: PMID32706339.

3. Bülow Anderberg S, Luther T, Berglund M et al. Increased levels of plasma cytokines and correlations to organ failure and 30-day mortality in critically ill Covid-19 patients. Cytokine. 2021. [PMID 33348065](https://pubmed.ncbi.nlm.nih.gov/33348065/). DOI: 10.1016/j.cyto.2020.155389. Access: abstract_metadata. Study: PMID33348065.

4. Dorgham K, Quentric P, Gökkaya M et al. Distinct cytokine profiles associated with COVID-19 severity and mortality. The Journal of allergy and clinical immunology. 2021. [PMID 33894209](https://pubmed.ncbi.nlm.nih.gov/33894209/). DOI: 10.1016/j.jaci.2021.03.047. Access: abstract_metadata. Study: PMID33894209.

5. Del Valle DM, Kim-Schulze S, Huang HH et al. An inflammatory cytokine signature predicts COVID-19 severity and survival. Nature medicine. 2020. [PMID 32839624](https://pubmed.ncbi.nlm.nih.gov/32839624/). DOI: 10.1038/s41591-020-1051-9. Access: abstract_metadata. Study: PMID32839624.

6. Herold T, Jurinovic V, Arnreich C et al. Elevated levels of IL-6 and CRP predict the need for mechanical ventilation in COVID-19. The Journal of allergy and clinical immunology. 2020. [PMID 32425269](https://pubmed.ncbi.nlm.nih.gov/32425269/). DOI: 10.1016/j.jaci.2020.05.008. Access: abstract_metadata. Study: PMID32425269.

7. Lucas C, Wong P, Klein J et al. Longitudinal analyses reveal immunological misfiring in severe COVID-19. Nature. 2020. [PMID 32717743](https://pubmed.ncbi.nlm.nih.gov/32717743/). DOI: 10.1038/s41586-020-2588-y. Access: abstract_metadata. Study: PMID32717743.

8. Remy KE, Mazer M, Striker DA et al. Severe immunosuppression and not a cytokine storm characterizes COVID-19 infections. JCI insight. 2020. [PMID 32687484](https://pubmed.ncbi.nlm.nih.gov/32687484/). DOI: 10.1172/jci.insight.140329. Access: abstract_metadata. Study: PMID32687484.

9. Loftus TJ, Ungaro R, Dirain M et al. Overlapping but Disparate Inflammatory and Immunosuppressive Responses to SARS-CoV-2 and Bacterial Sepsis: An Immunological Time Course Analysis. Frontiers in immunology. 2021. [PMID 34956225](https://pubmed.ncbi.nlm.nih.gov/34956225/). DOI: 10.3389/fimmu.2021.792448. Access: abstract_metadata. Study: PMID34956225.

10. Janssen NAF, Grondman I, de Nooijer AH et al. Dysregulated Innate and Adaptive Immune Responses Discriminate Disease Severity in COVID-19. The Journal of infectious diseases. 2021. [PMID 33524124](https://pubmed.ncbi.nlm.nih.gov/33524124/). DOI: 10.1093/infdis/jiab065. Access: abstract_metadata. Study: PMID33524124.

11. Reynolds D, Vazquez Guillamet C, Day A et al. Comprehensive Immunologic Evaluation of Bronchoalveolar Lavage Samples from Human Patients with Moderate and Severe Seasonal Influenza and Severe COVID-19. Journal of immunology (Baltimore, Md. : 1950). 2021. [PMID 34348975](https://pubmed.ncbi.nlm.nih.gov/34348975/). DOI: 10.4049/jimmunol.2100294. Access: abstract_metadata. Study: PMID34348975.

12. Kreutmair S, Unger S, Núñez NG et al. Distinct immunological signatures discriminate severe COVID-19 from non-SARS-CoV-2-driven critical pneumonia. Immunity. 2021. [PMID 34051147](https://pubmed.ncbi.nlm.nih.gov/34051147/). DOI: 10.1016/j.immuni.2021.05.002. Access: abstract_metadata. Study: PMID34051147.

13. RECOVERY Collaborative Group. Tocilizumab in patients admitted to hospital with COVID-19 (RECOVERY): a randomised, controlled, open-label, platform trial. Lancet (London, England). 2021. [PMID 33933206](https://pubmed.ncbi.nlm.nih.gov/33933206/). DOI: 10.1016/S0140-6736(21)00676-0. Access: full_text_partial. Study: PMID33933206.

14. REMAP-CAP Investigators, Gordon AC, Mouncey PR et al. Interleukin-6 Receptor Antagonists in Critically Ill Patients with Covid-19. The New England journal of medicine. 2021. [PMID 33631065](https://pubmed.ncbi.nlm.nih.gov/33631065/). DOI: 10.1056/NEJMoa2100433. Access: abstract_metadata. Study: REMAP-CAP-IL6-domain-2020-2021.

15. Salama C, Han J, Yau L et al. Tocilizumab in Patients Hospitalized with Covid-19 Pneumonia. The New England journal of medicine. 2021. [PMID 33332779](https://pubmed.ncbi.nlm.nih.gov/33332779/). DOI: 10.1056/NEJMoa2030340. Access: abstract_metadata. Study: PMID33332779.

16. Rosas IO, Bräu N, Waters M et al. Tocilizumab in Hospitalized Patients with Severe Covid-19 Pneumonia. The New England journal of medicine. 2021. [PMID 33631066](https://pubmed.ncbi.nlm.nih.gov/33631066/). DOI: 10.1056/NEJMoa2028700. Access: abstract_metadata. Study: COVACTA-NCT04320615.

17. Stone JH, Frigault MJ, Serling-Boyd NJ et al. Efficacy of Tocilizumab in Patients Hospitalized with Covid-19. The New England journal of medicine. 2020. [PMID 33085857](https://pubmed.ncbi.nlm.nih.gov/33085857/). DOI: 10.1056/NEJMoa2028836. Access: abstract_metadata. Study: PMID33085857.

18. Kyriazopoulou E, Poulakou G, Milionis H et al. Early treatment of COVID-19 with anakinra guided by soluble urokinase plasminogen receptor plasma levels: a double-blind, randomized controlled phase 3 trial. Nature medicine. 2021. [PMID 34480127](https://pubmed.ncbi.nlm.nih.gov/34480127/). DOI: 10.1038/s41591-021-01499-z. Access: full_text_partial. Study: SAVE-MORE-NCT04680949.

19. CORIMUNO-19 Collaborative group. Effect of anakinra versus usual care in adults in hospital with COVID-19 and mild-to-moderate pneumonia (CORIMUNO-ANA-1): a randomised controlled trial. The Lancet. Respiratory medicine. 2021. [PMID 33493450](https://pubmed.ncbi.nlm.nih.gov/33493450/). DOI: 10.1016/S2213-2600(20)30556-7. Access: abstract_metadata. Study: PMID33493450.

20. Caricchio R, Abbate A, Gordeev I et al. Effect of Canakinumab vs Placebo on Survival Without Invasive Mechanical Ventilation in Patients Hospitalized With Severe COVID-19: A Randomized Clinical Trial. JAMA. 2021. [PMID 34283183](https://pubmed.ncbi.nlm.nih.gov/34283183/). DOI: 10.1001/jama.2021.9508. Access: abstract_metadata. Study: PMID34283183.

21. Lescure FX, Honda H, Fowler RA et al. Sarilumab in patients admitted to hospital with severe or critical COVID-19: a randomised, double-blind, placebo-controlled, phase 3 trial. The Lancet. Respiratory medicine. 2021. [PMID 33676590](https://pubmed.ncbi.nlm.nih.gov/33676590/). DOI: 10.1016/S2213-2600(21)00099-0. Access: abstract_metadata. Study: PMID33676590.

22. Sivapalasingam S, Lederer DJ, Bhore R et al. Efficacy and Safety of Sarilumab in Hospitalized Patients With Coronavirus Disease 2019: A Randomized Clinical Trial. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2022. [PMID 35219277](https://pubmed.ncbi.nlm.nih.gov/35219277/). DOI: 10.1093/cid/ciac153. Access: abstract_metadata. Study: PMID35219277.

23. Merchante N, Cárcel S, Garrido-Gracia JC et al. Early Use of Sarilumab in Patients Hospitalized with COVID-19 Pneumonia and Features of Systemic Inflammation: the SARICOR Randomized Clinical Trial. Antimicrobial agents and chemotherapy. 2022. [PMID 34902262](https://pubmed.ncbi.nlm.nih.gov/34902262/). DOI: 10.1128/AAC.02107-21. Access: abstract_metadata. Study: PMID34902262.

24. Fanlo P, Gracia-Tello BDC, Fonseca Aizpuru E et al. Efficacy and Safety of Anakinra Plus Standard of Care for Patients With Severe COVID-19: A Randomized Phase 2/3 Clinical Trial. JAMA network open. 2023. [PMID 37027155](https://pubmed.ncbi.nlm.nih.gov/37027155/). DOI: 10.1001/jamanetworkopen.2023.7243. Access: abstract_metadata. Study: PMID37027155.

25. Declercq J, Van Damme KFA, De Leeuw E et al. Effect of anti-interleukin drugs in patients with COVID-19 and signs of cytokine release syndrome (COV-AID): a factorial, randomised, controlled trial. The Lancet. Respiratory medicine. 2021. [PMID 34756178](https://pubmed.ncbi.nlm.nih.gov/34756178/). DOI: 10.1016/S2213-2600(21)00377-5. Access: abstract_metadata. Study: PMID34756178.

26. Audemard-Verger A, Le Gouge A, Pestre V et al. Efficacy and safety of anakinra in adults presenting deteriorating respiratory symptoms from COVID-19: A randomized controlled trial. PloS one. 2022. [PMID 35925914](https://pubmed.ncbi.nlm.nih.gov/35925914/). DOI: 10.1371/journal.pone.0269065. Access: abstract_metadata. Study: PMID35925914.

27. Kyriazopoulou E, Hasin-Brumshtein Y, Midic U et al. Transitions of blood immune endotypes and improved outcome by anakinra in COVID-19 pneumonia: an analysis of the SAVE-MORE randomized controlled trial. Critical care (London, England). 2024. [PMID 38475786](https://pubmed.ncbi.nlm.nih.gov/38475786/). DOI: 10.1186/s13054-024-04852-z. Access: abstract_metadata. Study: SAVE-MORE-NCT04680949.

28. Tom J, Bao M, Tsai L et al. Prognostic and Predictive Biomarkers in Patients With Coronavirus Disease 2019 Treated With Tocilizumab in a Randomized Controlled Trial. Critical care medicine. 2022. [PMID 34612846](https://pubmed.ncbi.nlm.nih.gov/34612846/). DOI: 10.1097/CCM.0000000000005229. Access: abstract_metadata. Study: COVACTA-NCT04320615.

29. Rosas IO, Diaz G, Gottlieb RL et al. Tocilizumab and remdesivir in hospitalized patients with severe COVID-19 pneumonia: a randomized clinical trial. Intensive care medicine. 2021. [PMID 34609549](https://pubmed.ncbi.nlm.nih.gov/34609549/). DOI: 10.1007/s00134-021-06507-x. Access: abstract_metadata. Study: PMID34609549.

30. Derde L, Gordon AC, Mouncey PR et al. Tocilizumab, sarilumab and anakinra in critically ill patients with COVID-19: a randomised, controlled, open-label, adaptive platform trial. Thorax. 2025. [PMID 40360262](https://pubmed.ncbi.nlm.nih.gov/40360262/). DOI: 10.1136/thorax-2024-222488. Access: abstract_metadata. Study: REMAP-CAP-IL6-domain-2020-2021.

31. RECOVERY Collaborative Group, Horby P, Lim WS et al. Dexamethasone in Hospitalized Patients with Covid-19. The New England journal of medicine. 2021. [PMID 32678530](https://pubmed.ncbi.nlm.nih.gov/32678530/). DOI: 10.1056/NEJMoa2021436. Access: abstract_metadata. Study: PMID32678530.

32. Beigel JH, Tomashek KM, Dodd LE et al. Remdesivir for the Treatment of Covid-19 - Final Report. The New England journal of medicine. 2020. [PMID 32445440](https://pubmed.ncbi.nlm.nih.gov/32445440/). DOI: 10.1056/NEJMoa2007764. Access: abstract_metadata. Study: ACTT1-NCT04280705.

33. Kalil AC, Patterson TF, Mehta AK et al. Baricitinib plus Remdesivir for Hospitalized Adults with Covid-19. The New England journal of medicine. 2021. [PMID 33306283](https://pubmed.ncbi.nlm.nih.gov/33306283/). DOI: 10.1056/NEJMoa2031994. Access: abstract_metadata. Study: PMID33306283.

34. Marconi VC, Ramanan AV, de Bono S et al. Efficacy and safety of baricitinib for the treatment of hospitalised adults with COVID-19 (COV-BARRIER): a randomised, double-blind, parallel-group, placebo-controlled phase 3 trial. The Lancet. Respiratory medicine. 2021. [PMID 34480861](https://pubmed.ncbi.nlm.nih.gov/34480861/). DOI: 10.1016/S2213-2600(21)00331-3. Access: full_text_partial. Study: PMID34480861.

35. RECOVERY Collaborative Group. Baricitinib in patients admitted to hospital with COVID-19 (RECOVERY): a randomised, controlled, open-label, platform trial and updated meta-analysis. Lancet (London, England). 2022. [PMID 35908569](https://pubmed.ncbi.nlm.nih.gov/35908569/). DOI: 10.1016/S0140-6736(22)01109-6. Access: abstract_metadata. Study: PMID35908569.

36. RECOVERY Collaborative Group. Higher dose corticosteroids in patients admitted to hospital with COVID-19 who are hypoxic but not requiring ventilatory support (RECOVERY): a randomised, controlled, open-label, platform trial. Lancet (London, England). 2023. [PMID 37060915](https://pubmed.ncbi.nlm.nih.gov/37060915/). DOI: 10.1016/S0140-6736(23)00510-X. Access: abstract_metadata. Study: PMID37060915.

37. REMAP-CAP, ACTIV-4a and ATTACC Investigators. Therapeutic Anticoagulation with Heparin in Critically Ill Patients with Covid-19. The New England journal of medicine. 2021. [PMID 34351722](https://pubmed.ncbi.nlm.nih.gov/34351722/). DOI: 10.1056/NEJMoa2103417. Access: abstract_metadata. Study: PMID34351722.

38. ATTACC, ACTIV-4a and REMAP-CAP Investigators. Therapeutic Anticoagulation with Heparin in Noncritically Ill Patients with Covid-19. The New England journal of medicine. 2021. [PMID 34351721](https://pubmed.ncbi.nlm.nih.gov/34351721/). DOI: 10.1056/NEJMoa2105911. Access: abstract_metadata. Study: PMID34351721.

39. Kalil AC, Mehta AK, Patterson TF et al. Efficacy of interferon beta-1a plus remdesivir compared with remdesivir alone in hospitalised adults with COVID-19: a double-bind, randomised, placebo-controlled, phase 3 trial. The Lancet. Respiratory medicine. 2021. [PMID 34672949](https://pubmed.ncbi.nlm.nih.gov/34672949/). DOI: 10.1016/S2213-2600(21)00384-2. Access: abstract_metadata. Study: PMID34672949.

40. Wolfe CR, Tomashek KM, Patterson TF et al. Baricitinib versus dexamethasone for adults hospitalised with COVID-19 (ACTT-4): a randomised, double-blind, double placebo-controlled trial. The Lancet. Respiratory medicine. 2022. [PMID 35617986](https://pubmed.ncbi.nlm.nih.gov/35617986/). DOI: 10.1016/S2213-2600(22)00088-1. Access: abstract_metadata. Study: PMID35617986.

41. ACTIV-3-Therapeutics for Inpatients with COVID-19 (TICO) Study Group. Tixagevimab-cilgavimab for treatment of patients hospitalised with COVID-19: a randomised, double-blind, phase 3 trial. The Lancet. Respiratory medicine. 2022. [PMID 35817072](https://pubmed.ncbi.nlm.nih.gov/35817072/). DOI: 10.1016/S2213-2600(22)00215-6. Access: abstract_metadata. Study: PMID35817072.

42. Singh K, Rubenstein K, Callier V et al. SARS-CoV-2 RNA and Nucleocapsid Antigen Are Blood Biomarkers Associated With Severe Disease Outcomes That Improve in Response to Remdesivir. The Journal of infectious diseases. 2024. [PMID 38657001](https://pubmed.ncbi.nlm.nih.gov/38657001/). DOI: 10.1093/infdis/jiae198. Access: abstract_metadata. Study: ACTT1-NCT04280705.

43. Bastard P, Rosen LB, Zhang Q et al. Autoantibodies against type I IFNs in patients with life-threatening COVID-19. Science (New York, N.Y.). 2020. [PMID 32972996](https://pubmed.ncbi.nlm.nih.gov/32972996/). DOI: 10.1126/science.abd4585. Access: abstract_metadata. Study: IFN-autoAb-initial-overlaps-extension.

44. Zhang Q, Bastard P, Liu Z et al. Inborn errors of type I IFN immunity in patients with life-threatening COVID-19. Science (New York, N.Y.). 2020. [PMID 32972995](https://pubmed.ncbi.nlm.nih.gov/32972995/). DOI: 10.1126/science.abd4570. Access: abstract_metadata. Study: PMID32972995.

45. Bastard P, Gervais A, Le Voyer T et al. Autoantibodies neutralizing type I IFNs are present in ~4% of uninfected individuals over 70 years old and account for ~20% of COVID-19 deaths. Science immunology. 2021. [PMID 34413139](https://pubmed.ncbi.nlm.nih.gov/34413139/). DOI: 10.1126/sciimmunol.abl4340. Access: full_text_partial. Study: IFN-autoAb-extension-includes-initial.

46. Manry J, Bastard P, Gervais A et al. The risk of COVID-19 death is much greater and age dependent with type I IFN autoantibodies. Proceedings of the National Academy of Sciences of the United States of America. 2022. [PMID 35576468](https://pubmed.ncbi.nlm.nih.gov/35576468/). DOI: 10.1073/pnas.2200413119. Access: abstract_metadata. Study: IFN-fatality-shared34159-controls.

47. Zhang Q, Matuozzo D, Le Pen J et al. Recessive inborn errors of type I IFN immunity in children with COVID-19 pneumonia. The Journal of experimental medicine. 2022. [PMID 35708626](https://pubmed.ncbi.nlm.nih.gov/35708626/). DOI: 10.1084/jem.20220131. Access: abstract_metadata. Study: PMID35708626.

48. García-García A, Pérez de Diego R, Flores C et al. Humans with inherited MyD88 and IRAK-4 deficiencies are predisposed to hypoxemic COVID-19 pneumonia. The Journal of experimental medicine. 2023. [PMID 36880831](https://pubmed.ncbi.nlm.nih.gov/36880831/). DOI: 10.1084/jem.20220170. Access: abstract_metadata. Study: PMID36880831.

49. Matuozzo D, Talouarn E, Marchal A et al. Rare predicted loss-of-function variants of type I IFN immunity genes are associated with life-threatening COVID-19. Genome medicine. 2023. [PMID 37020259](https://pubmed.ncbi.nlm.nih.gov/37020259/). DOI: 10.1186/s13073-023-01173-8. Access: abstract_metadata. Study: PMID37020259.

50. Antolí A, Gómez-Vázquez JL, Sierra-Fortuny A et al. Autoantibodies neutralizing type I interferons remain a significant risk factor for critical COVID-19 pneumonia in vaccinated patients. Clinical immunology (Orlando, Fla.). 2025. [PMID 40185298](https://pubmed.ncbi.nlm.nih.gov/40185298/). DOI: 10.1016/j.clim.2025.110491. Access: abstract_metadata. Study: PMID40185298.

51. Ouwendijk WJD, Raadsen MP, van Kampen JJA et al. High Levels of Neutrophil Extracellular Traps Persist in the Lower Respiratory Tract of Critically Ill Patients With Coronavirus Disease 2019. The Journal of infectious diseases. 2021. [PMID 33507309](https://pubmed.ncbi.nlm.nih.gov/33507309/). DOI: 10.1093/infdis/jiab050. Access: abstract_metadata. Study: PMID33507309.

52. Carvelli J, Demaria O, Vély F et al. Association of COVID-19 inflammation with activation of the C5a-C5aR1 axis. Nature. 2020. [PMID 32726800](https://pubmed.ncbi.nlm.nih.gov/32726800/). DOI: 10.1038/s41586-020-2600-6. Access: abstract_metadata. Study: PMID32726800.

53. Toldo S, Bussani R, Nuzzi V et al. Inflammasome formation in the lungs of patients with fatal COVID-19. Inflammation research : official journal of the European Histamine Research Society ... [et al.]. 2021. [PMID 33079210](https://pubmed.ncbi.nlm.nih.gov/33079210/). DOI: 10.1007/s00011-020-01413-2. Access: abstract_metadata. Study: PMID33079210.

54. Jonassen TB, Jørgensen SE, Mitchell NH et al. Alveolar cytokines and interferon autoantibodies in COVID-19 ARDS. Frontiers in immunology. 2024. [PMID 38571960](https://pubmed.ncbi.nlm.nih.gov/38571960/). DOI: 10.3389/fimmu.2024.1353012. Access: full_text_partial. Study: Hvidovre-CARDS-overlap-with2021-unresolved.

55. Ryu JK, Yan Z, Montano M et al. Fibrin drives thromboinflammation and neuropathology in COVID-19. Nature. 2024. [PMID 39198643](https://pubmed.ncbi.nlm.nih.gov/39198643/). DOI: 10.1038/s41586-024-07873-4. Access: abstract_metadata. Study: PMID39198643.

56. Delorey TM, Ziegler CGK, Heimberg G et al. COVID-19 tissue atlases reveal SARS-CoV-2 pathology and cellular targets. Nature. 2021. [PMID 33915569](https://pubmed.ncbi.nlm.nih.gov/33915569/). DOI: 10.1038/s41586-021-03570-8. Access: abstract_metadata. Study: PMID33915569.

57. Ackermann M, Verleden SE, Kuehnel M et al. Pulmonary Vascular Endothelialitis, Thrombosis, and Angiogenesis in Covid-19. The New England journal of medicine. 2020. [PMID 32437596](https://pubmed.ncbi.nlm.nih.gov/32437596/). DOI: 10.1056/NEJMoa2015432. Access: abstract_metadata. Study: PMID32437596.

58. Mueller AA, Tamura T, Crowley CP et al. Inflammatory Biomarker Trends Predict Respiratory Decline in COVID-19 Patients. Cell reports. Medicine. 2020. [PMID 33163981](https://pubmed.ncbi.nlm.nih.gov/33163981/). DOI: 10.1016/j.xcrm.2020.100144. Access: abstract_metadata. Study: PMID33163981.

59. Melms JC, Biermann J, Huang H et al. A molecular single-cell lung atlas of lethal COVID-19. Nature. 2021. [PMID 33915568](https://pubmed.ncbi.nlm.nih.gov/33915568/). DOI: 10.1038/s41586-021-03569-1. Access: abstract_metadata. Study: PMID33915568.

60. McAleavy M, Zhang Q, Ehmann PJ et al. The Activin/FLRG Pathway Associates with Poor COVID-19 Outcomes in Hospitalized Patients. Molecular and cellular biology. 2022. [PMID 34723652](https://pubmed.ncbi.nlm.nih.gov/34723652/). DOI: 10.1128/MCB.00467-21. Access: abstract_metadata. Study: PMID34723652.

61. Shi W, Liu X, Cao Q et al. High-dimensional single-cell analysis reveals the immune characteristics of COVID-19. American journal of physiology. Lung cellular and molecular physiology. 2021. [PMID 33146564](https://pubmed.ncbi.nlm.nih.gov/33146564/). DOI: 10.1152/ajplung.00355.2020. Access: abstract_metadata. Study: PMID33146564.

62. Thorne LG, Reuschl AK, Zuliani-Alvarez L et al. SARS-CoV-2 sensing by RIG-I and MDA5 links epithelial infection to macrophage inflammation. The EMBO journal. 2021. [PMID 34101213](https://pubmed.ncbi.nlm.nih.gov/34101213/). DOI: 10.15252/embj.2021107826. Access: abstract_metadata. Study: PMID34101213.

63. Donlan AN, Sutherland TE, Marie C et al. IL-13 is a driver of COVID-19 severity. JCI insight. 2021. [PMID 34185704](https://pubmed.ncbi.nlm.nih.gov/34185704/). DOI: 10.1172/jci.insight.150107. Access: abstract_metadata. Study: PMID34185704.

64. M Bader S, Scherer L, Schaefer J et al. IL-1β drives SARS-CoV-2-induced disease independently of the inflammasome and pyroptosis signalling. Cell death and differentiation. 2025. [PMID 40016339](https://pubmed.ncbi.nlm.nih.gov/40016339/). DOI: 10.1038/s41418-025-01459-x. Access: abstract_metadata. Study: PMID40016339.

65. Yan B, Freiwald T, Chauss D et al. SARS-CoV-2 drives JAK1/2-dependent local complement hyperactivation. Science immunology. 2021. [PMID 33827897](https://pubmed.ncbi.nlm.nih.gov/33827897/). DOI: 10.1126/sciimmunol.abg0833. Access: abstract_metadata. Study: PMID33827897.

66. Zhou Z, Ren L, Zhang L et al. Heightened Innate Immune Responses in the Respiratory Tract of COVID-19 Patients. Cell host & microbe. 2020. [PMID 32407669](https://pubmed.ncbi.nlm.nih.gov/32407669/). DOI: 10.1016/j.chom.2020.04.017. Access: abstract_metadata. Study: PMID32407669.

67. Ronit A, Berg RMG, Bay JT et al. Compartmental immunophenotyping in COVID-19 ARDS: A case series. The Journal of allergy and clinical immunology. 2021. [PMID 32979342](https://pubmed.ncbi.nlm.nih.gov/32979342/). DOI: 10.1016/j.jaci.2020.09.009. Access: full_text_partial. Study: Hvidovre-CARDS-overlap-with2024-unresolved.

68. Mudd PA, Crawford JC, Turner JS et al. Distinct inflammatory profiles distinguish COVID-19 from influenza with limited contributions from cytokine storm. Science advances. 2020. [PMID 33187979](https://pubmed.ncbi.nlm.nih.gov/33187979/). DOI: 10.1126/sciadv.abe3024. Access: abstract_metadata. Study: PMID33187979.

## Source-specific evidence cards

### [1] Cytokine Levels in Critically Ill Patients With COVID-19 and Other Conditions.

**Shows:** Same-method ICU comparison found lower circulating IL6 in COVID ARDS than septic-shock ARDS.

**Does not show:** Historical cohorts, timing, disease severity and unmeasured assay-lot variability constrain magnitude inference.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: Publisher Methods, Results and Discussion, displayed lines 141-196; summary receipt. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC4.

### [2] Cytokine profile in plasma of severe COVID-19 does not differ from ARDS and sepsis.

**Shows:** Small same-panel comparison detected no significant cytokine differences between COVID, ARDS and sepsis.

**Does not show:** Luminex MFI cannot be converted to comparable pg/mL; low power and nonsignificance do not establish equivalence.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Results:paragraph0, 1:Results:paragraph1, 1:Results:paragraph2, 1:Results:paragraph3, 2:Discussion:paragraph0, 2:Discussion:paragraph1, 2:Discussion:paragraph3, 2:Discussion:paragraph4, 3:Methods:paragraph1, 3:Methods:paragraph2, 3:Methods:paragraph3, 3:Methods:paragraph4, 3:Methods:paragraph5. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC4.

### [3] Increased levels of plasma cytokines and correlations to organ failure and 30-day mortality in critically ill Covid-19 patients.

**Shows:** Moderate cytokine increases in 24 ICU patients correlated with respiratory and kidney failure; IL8 had the strongest mortality signal.

**Does not show:** Surgical controls do not test magnitude against matched sepsis; correlation does not establish primacy.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4.

### [4] Distinct cytokine profiles associated with COVID-19 severity and mortality.

**Shows:** Admission profiles predicted mortality differently by respiratory-support category, with a validation cohort.

**Does not show:** Prognostic profiles and antigen associations do not identify a universal cytokine-first cause.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4.

### [5] An inflammatory cytokine signature predicts COVID-19 severity and survival.

**Shows:** Admission IL6 and TNF predicted severity and death after adjustment in 1484 patients and a second cohort.

**Does not show:** Adjusted prediction is not randomized causal attribution; index inflammatory framing is not accepted as a result.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. UpdateOf: PMID32511562. Subclaims: SC1, SC4.

### [6] Elevated levels of IL-6 and CRP predict the need for mechanical ventilation in COVID-19.

**Shows:** Peak IL6 and CRP before intubation predicted respiratory failure in development and temporal validation cohorts.

**Does not show:** Peak measurements are time-varying risk markers; cutoffs are not storm definitions or proof of causality.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [7] Longitudinal analyses reveal immunological misfiring in severe COVID-19.

**Shows:** Serial immune profiles in 113 patients linked early cytokines and sustained maladapted responses to severe trajectories.

**Does not show:** Associations can reflect viral burden and tissue injury; no pathway-specific intervention.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4.

### [8] Severe immunosuppression and not a cytokine storm characterizes COVID-19 infections.

**Shows:** COVID leukocytes had impaired stimulated IFNgamma/TNF production relative to critical controls; ex vivo IL7 restored T-cell output.

**Does not show:** Ex vivo functional suppression can coexist with local inflammation; no in vivo IL7 outcome test.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC4.

### [9] Overlapping but Disparate Inflammatory and Immunosuppressive Responses to SARS-CoV-2 and Bacterial Sepsis: An Immunological Time Course Analysis.

**Shows:** COVID and bacterial sepsis differed in early inflammation and longitudinal immune suppression over 21 days.

**Does not show:** Small cohorts, treatment and timing confounding; no dominant causal pathway attribution.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC4.

### [10] Dysregulated Innate and Adaptive Immune Responses Discriminate Disease Severity in COVID-19.

**Shows:** Innate hyperresponsiveness and adaptive dysfunction coexisted in 147 patients; clustering found no specific endotypes.

**Does not show:** Functional blood assays do not establish systemic primacy or exclude tissue-specific endotypes.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4.

### [11] Comprehensive Immunologic Evaluation of Bronchoalveolar Lavage Samples from Human Patients with Moderate and Severe Seasonal Influenza and Severe COVID-19.

**Shows:** Matched blood/BAL study found lower-airway cytokine elevations in severe viral disease; only IP10 correlated between compartments.

**Does not show:** Sampling and lavage dilution limit absolute comparisons; compartment association is not compartment-specific causality.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC5, SC4.

### [12] Distinct immunological signatures discriminate severe COVID-19 from non-SARS-CoV-2-driven critical pneumonia.

**Shows:** Longitudinal critical-pneumonia comparison distinguished COVID exhaustion and immune-recognition signatures from shared myelopoiesis.

**Does not show:** Predicted HLA binding and profiling are associative; linked erratum details remain unaudited.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID35139354. Subclaims: SC4.

### [13] Tocilizumab in patients admitted to hospital with COVID-19 (RECOVERY): a randomised, controlled, open-label, platform trial.

**Shows:** RECOVERY tocilizumab reduced primary 28-day ITT mortality in hypoxic CRP-at-least-75 patients, 31% versus 35%, RR0.85.

**Does not show:** Open label; 82% on steroids; effect identifies IL6R contribution, not circulating concentration, compartment or primary cause.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Methods:paragraph0, 1:Methods:paragraph2, 1:Methods:paragraph3, 1:Methods:paragraph4, 1:Methods:paragraph6, 1:Methods:paragraph7, 1:Methods:paragraph8, 1:Methods:paragraph9, 1:Methods:paragraph11, 1:Methods:paragraph12, 2:Results:paragraph0, 2:Results:paragraph2, 2:Results:paragraph9, 2:Results:paragraph10, 2:Results:paragraph11, 2:Results:paragraph13, 2:Results:paragraph14, 2:Results:paragraph15, 2:Results:paragraph16. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [14] Interleukin-6 Receptor Antagonists in Critically Ill Patients with Covid-19.

**Shows:** REMAP-CAP IL6R blockade improved organ-support-free days and survival within 24h of ICU organ support.

**Does not show:** Adaptive open-label platform and concomitant treatments; initial sarilumab arm small; does not identify systemic magnitude or primacy.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [15] Tocilizumab in Patients Hospitalized with Covid-19 Pneumonia.

**Shows:** EMPACTA reduced ventilation-or-death composite in nonventilated hospitalized patients.

**Does not show:** Mortality alone did not improve; composite benefit must not be called survival benefit.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

### [16] Tocilizumab in Hospitalized Patients with Severe Covid-19 Pneumonia.

**Shows:** COVACTA placebo-controlled tocilizumab trial was null for primary day28 clinical status and mortality.

**Does not show:** Different eligibility and treatment background from RECOVERY; confidence intervals do not exclude all context-specific effects.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [17] Efficacy of Tocilizumab in Patients Hospitalized with Covid-19.

**Shows:** BACC trial found no significant prevention of intubation or death with tocilizumab in moderately ill nonventilated patients.

**Does not show:** Wide intervals, small trial and early treatment era; not proof IL6 never contributes.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [18] Early treatment of COVID-19 with anakinra guided by soluble urokinase plasminogen receptor plasma levels: a double-blind, randomized controlled phase 3 trial.

**Shows:** SAVE-MORE improved primary WHO-CPS with anakinra in suPAR-at-least-6 patients; mortality was lower in a separate survival analysis.

**Does not show:** Excluded P/F below150 and NIV/MV; 85.9% on dexamethasone; WHO endpoint hierarchy differs from mortality-primary trials; linked correction not fully audited.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Results:paragraph0, 1:Results:paragraph6, 1:Results:paragraph7, 1:Results:paragraph8, 1:Results:paragraph12, 1:Results:paragraph14, 1:Results:paragraph15, 2:Discussion:paragraph5, 2:Discussion:paragraph6, 3:Methods:paragraph0, 3:Methods:paragraph2, 3:Methods:paragraph3, 3:Methods:paragraph4, 3:Methods:paragraph6, 3:Methods:paragraph7. No complete-paper or supplement audit claimed. ErratumIn: PMID34625750. Subclaims: SC3, SC4.

### [19] Effect of anakinra versus usual care in adults in hospital with COVID-19 and mild-to-moderate pneumonia (CORIMUNO-ANA-1): a randomised controlled trial.

**Shows:** CORIMUNO-ANA-1 found no benefit on coprimary ventilation/death outcomes with anakinra.

**Does not show:** Early stopped small open-label trial with different CRP eligibility and treatment era from SAVE-MORE.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [20] Effect of Canakinumab vs Placebo on Survival Without Invasive Mechanical Ventilation in Patients Hospitalized With Severe COVID-19: A Randomized Clinical Trial.

**Shows:** CAN-COVID canakinumab did not significantly improve survival without invasive ventilation.

**Does not show:** Targets IL1beta rather than both IL1alpha/beta; a null does not establish pathway equivalence or universal absence.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [21] Sarilumab in patients admitted to hospital with severe or critical COVID-19: a randomised, double-blind, placebo-controlled, phase 3 trial.

**Shows:** Blinded multinational sarilumab trial did not improve primary clinical improvement or day29 survival.

**Does not show:** Numerical critical-subgroup survival difference was not significant; exploratory subsets cannot define the claim.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

### [22] Efficacy and Safety of Sarilumab in Hospitalized Patients With Coronavirus Disease 2019: A Randomized Clinical Trial.

**Shows:** Adaptive sarilumab phase2/3 trial missed its primary endpoint; pooled post hoc steroid subgroup suggested benefit.

**Does not show:** Post hoc subgroup and low steroid background in primary ventilated population limit causal subgroup claims.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

### [23] Early Use of Sarilumab in Patients Hospitalized with COVID-19 Pneumonia and Features of Systemic Inflammation: the SARICOR Randomized Clinical Trial.

**Shows:** SARICOR400 showed a nonsignificant trend toward less respiratory progression in inflammatory pneumonia.

**Does not show:** Phase2 open-label sample115; trend and zero deaths in one arm are not confirmatory efficacy.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3.

### [24] Efficacy and Safety of Anakinra Plus Standard of Care for Patients With Severe COVID-19: A Randomized Phase 2/3 Clinical Trial.

**Shows:** ANA-COVID-GEAS did not prevent mechanical ventilation or reduce mortality with anakinra.

**Does not show:** Open label; selected severe hyperinflammatory population differs from suPAR selection; no universal pathway falsification.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [25] Effect of anti-interleukin drugs in patients with COVID-19 and signs of cytokine release syndrome (COV-AID): a factorial, randomised, controlled trial.

**Shows:** COV-AID factorial IL1/IL6 strategies did not shorten primary clinical-improvement time or show a mortality difference.

**Does not show:** Low baseline mortality, serum entry criteria and factorial dosing contexts constrain scope.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [26] Efficacy and safety of anakinra in adults presenting deteriorating respiratory symptoms from COVID-19: A randomized controlled trial.

**Shows:** Small anakinra trial stopped early and showed worse day14 treatment success than optimized care.

**Does not show:** Only71 of planned240; safety signal and uncertainty must be retained rather than generalized to all anakinra settings.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [27] Transitions of blood immune endotypes and improved outcome by anakinra in COVID-19 pneumonia: an analysis of the SAVE-MORE randomized controlled trial.

**Shows:** SAVE-MORE transcriptomic substudy associated anakinra with endotype transitions and respiratory-failure protection.

**Does not show:** Same trial family as primary SAVE-MORE; post-treatment endotype differences are not independently randomized mediators.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [28] Prognostic and Predictive Biomarkers in Patients With Coronavirus Disease 2019 Treated With Tocilizumab in a Randomized Controlled Trial.

**Shows:** COVACTA exploratory biomarkers were prognostic; ferritin interactions suggested treatment-effect prediction.

**Does not show:** Same COVACTA family; exploratory interactions and linked erratum not independent efficacy replication.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID37439653. Subclaims: SC1, SC3.

### [29] Tocilizumab and remdesivir in hospitalized patients with severe COVID-19 pneumonia: a randomized clinical trial.

**Shows:** REMDACTA tocilizumab plus remdesivir did not improve primary discharge time, with 88.2% receiving steroids.

**Does not show:** Steroid use alone cannot explain every IL6 null; trial was not a mortality-primary precision test.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [30] Tocilizumab, sarilumab and anakinra in critically ill patients with COVID-19: a randomised, controlled, open-label, adaptive platform trial.

**Shows:** Expanded REMAP-CAP cohort supported equivalent IL6R agents and a null anakinra effect in critical patients.

**Does not show:** 2025 publication enrolled March2020-April2021 and extends the original domain; not a fresh 2025 population replication.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [31] Dexamethasone in Hospitalized Patients with Covid-19.

**Shows:** RECOVERY low-dose dexamethasone reduced mortality on oxygen or ventilation, without benefit in no-oxygen patients.

**Does not show:** Pleiotropic steroid effects do not isolate a cytokine; same platform can share co-randomized patients and controls.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [32] Remdesivir for the Treatment of Covid-19 - Final Report.

**Shows:** ACTT1 remdesivir improved primary recovery time versus placebo.

**Does not show:** Mortality CI included no effect; antiviral efficacy does not quantify the fraction of death driven by virus versus inflammation.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [33] Baricitinib plus Remdesivir for Hospitalized Adults with Covid-19.

**Shows:** ACTT2 baricitinib plus remdesivir accelerated recovery versus remdesivir.

**Does not show:** Mortality interval included null; JAK1/2 affects several pathways and cannot identify one circulating cytokine.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [34] Efficacy and safety of baricitinib for the treatment of hospitalised adults with COVID-19 (COV-BARRIER): a randomised, double-blind, parallel-group, placebo-controlled phase 3 trial.

**Shows:** COV-BARRIER missed primary progression endpoint but showed nominal secondary mortality reduction.

**Does not show:** Failed hierarchy makes mortality p-values non-multiplicity-controlled; original Methods and Results explicitly confirm this.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Methods:paragraph0, 1:Methods:paragraph3, 1:Methods:paragraph4, 1:Methods:paragraph7, 1:Methods:paragraph8, 1:Methods:paragraph9, 1:Methods:paragraph10, 1:Methods:paragraph11, 2:Results:paragraph11, 2:Results:paragraph12, 2:Results:paragraph13, 2:Results:paragraph18, 2:Results:paragraph19, 2:Results:paragraph20, 2:Results:paragraph21. No complete-paper or supplement audit claimed. ErratumIn: PMID34508656. Subclaims: SC3, SC4.

### [35] Baricitinib in patients admitted to hospital with COVID-19 (RECOVERY): a randomised, controlled, open-label, platform trial and updated meta-analysis.

**Shows:** RECOVERY baricitinib reduced primary ITT mortality, RR0.87, on extensive steroid background.

**Does not show:** JAK blockade is broader than IL6R blockade; concurrent tocilizumab and shared platform complicate line-counting.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID36183728. Subclaims: SC3, SC4.

### [36] Higher dose corticosteroids in patients admitted to hospital with COVID-19 who are hypoxic but not requiring ventilatory support (RECOVERY): a randomised, controlled, open-label, platform trial.

**Shows:** Higher-dose dexamethasone increased mortality versus usual low-dose care in no/simple-oxygen patients.

**Does not show:** Dose and baseline support are prespecified boundaries; harm does not negate low-dose benefit in other contexts.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [37] Therapeutic Anticoagulation with Heparin in Critically Ill Patients with Covid-19.

**Shows:** Therapeutic heparin was futile for organ-support-free days in critical COVID.

**Does not show:** Therapeutic anticoagulation is not all immunothrombosis; late irreversible injury and bleeding constrain inference.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [38] Therapeutic Anticoagulation with Heparin in Noncritically Ill Patients with Covid-19.

**Shows:** Therapeutic heparin improved survival-to-discharge without organ support in noncritical hospitalized patients.

**Does not show:** Bleeding increased; anticoagulant and other heparin actions do not isolate systemic cytokine primacy.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [39] Efficacy of interferon beta-1a plus remdesivir compared with remdesivir alone in hospitalised adults with COVID-19: a double-bind, randomised, placebo-controlled, phase 3 trial.

**Shows:** ACTT3 late IFNbeta plus remdesivir did not improve recovery and high-flow patients had worse outcomes.

**Does not show:** Treatment after symptoms does not falsify inherited early IFN protection; timing and molecule matter.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [40] Baricitinib versus dexamethasone for adults hospitalised with COVID-19 (ACTT-4): a randomised, double-blind, double placebo-controlled trial.

**Shows:** ACTT4 found similar ventilation-free survival with baricitinib versus dexamethasone, both plus remdesivir.

**Does not show:** Active comparator, no untreated control; nonsignificance does not prove mechanistic equivalence.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [41] Tixagevimab-cilgavimab for treatment of patients hospitalised with COVID-19: a randomised, double-blind, phase 3 trial.

**Shows:** Hospitalized tixagevimab/cilgavimab trial missed primary sustained recovery; secondary mortality was lower.

**Does not show:** Primary null, background remdesivir and variant-specific antibody activity limit broad viral-primacy conclusions; correction retained.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID36356600. Subclaims: SC4.

### [42] SARS-CoV-2 RNA and Nucleocapsid Antigen Are Blood Biomarkers Associated With Severe Disease Outcomes That Improve in Response to Remdesivir.

**Shows:** ACTT1 substudy linked blood viral markers, remdesivir clearance effects and clinical benefit.

**Does not show:** Same ACTT1 family; biomarker mediation was not itself randomized and respiratory/blood measures differed.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [43] Autoantibodies against type I IFNs in patients with life-threatening COVID-19.

**Shows:** Neutralizing IFN autoantibodies enriched in life-threatening disease blocked IFN antiviral activity in vitro.

**Does not show:** Ascertainment and host age/sex matter; a causal susceptibility subgroup does not explain every severe case.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [44] Inborn errors of type I IFN immunity in patients with life-threatening COVID-19.

**Shows:** Functionally validated IFN-pathway deficiencies enriched in critical disease and increased cellular viral vulnerability.

**Does not show:** Genetics precedes infection but clinical ascertainment and selected loci constrain population attribution.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [45] Autoantibodies neutralizing type I IFNs are present in ~4% of uninfected individuals over 70 years old and account for ~20% of COVID-19 deaths.

**Shows:** More sensitive neutralization assays extended IFN-autoantibody susceptibility; available preinfection samples were positive.

**Does not show:** 566 critical patients and427 controls overlap the initial report; assay threshold and age alter prevalence estimates.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:RESULTS:paragraph0, 1:RESULTS:paragraph1, 1:RESULTS:paragraph2, 1:RESULTS:paragraph3, 1:RESULTS:paragraph4, 1:RESULTS:paragraph5, 1:RESULTS:paragraph6. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [46] The risk of COVID-19 death is much greater and age dependent with type I IFN autoantibodies.

**Shows:** Age-stratified analyses estimated large autoantibody-associated fatality risk in unvaccinated individuals.

**Does not show:** Shared34159 general-population comparator cohort; model-based IFR estimates are not an independent intervention.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. UpdateOf: PMID35043109. Subclaims: SC4.

### [47] Recessive inborn errors of type I IFN immunity in children with COVID-19 pneumonia.

**Shows:** Children with complete IFN-signalling defects had severe pneumonia and vulnerable fibroblasts.

**Does not show:** Rare pediatric genetic scope; not representative adult mortality or all inflammatory syndromes.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [48] Humans with inherited MyD88 and IRAK-4 deficiencies are predisposed to hypoxemic COVID-19 pneumonia.

**Shows:** Inherited MyD88/IRAK4 deficiency predisposed to hypoxemic COVID with impaired TLR7 IFN production.

**Does not show:** Small rare-disease cohort and broad signalling functions; demonstrates susceptibility rather than universal primacy.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [49] Rare predicted loss-of-function variants of type I IFN immunity genes are associated with life-threatening COVID-19.

**Shows:** Rare-variant burden analysis supported prespecified IFN pathways and recessive TLR7 risk.

**Does not show:** No gene reached genome-wide significance; pathway enrichment and linked erratum do not justify unrestricted genome-wide discovery claims.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. UpdateOf: PMID36324795; ErratumIn: PMID38184654. Subclaims: SC4.

### [50] Autoantibodies neutralizing type I interferons remain a significant risk factor for critical COVID-19 pneumonia in vaccinated patients.

**Shows:** IFN-neutralizing autoantibodies predicted critical pneumonia among vaccinated hospital patients in2021-2022.

**Does not show:** Observational single hospital; later publication does not imply contemporary variant/treatment generality.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4.

### [51] High Levels of Neutrophil Extracellular Traps Persist in the Lower Respiratory Tract of Critically Ill Patients With Coronavirus Disease 2019.

**Shows:** NETs persisted in lower airways and tracked viral load and severity; baseline plasma NETs did not track thrombosis.

**Does not show:** Correlation and autopsy retention do not establish NET primacy or clinically effective NET inhibition.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5.

### [52] Association of COVID-19 inflammation with activation of the C5a-C5aR1 axis.

**Shows:** C5a/C5aR1 tracked severity; specific blockade inhibited myeloid activation and lung injury in knock-in mice.

**Does not show:** The abstract does not establish an infected-mouse endpoint; no direct primary systemic-cytokine or human causal test.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5.

### [53] Inflammasome formation in the lungs of patients with fatal COVID-19.

**Shows:** Fatal lungs showed inflammasome aggregates in injury areas versus historical controls.

**Does not show:** Four fatal cases and four noninfected controls; presence does not prove inflammasome necessity.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5.

### [54] Alveolar cytokines and interferon autoantibodies in COVID-19 ARDS.

**Shows:** Paired ELF-corrected study found alveolar IL8 compartmentalization and lower IL6 in steroid-treated CARDS.

**Does not show:** Steroid groups were not randomized and differed in gas exchange; BAL autoantibody binding did not establish neutralization; unknown BAL cutoff.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Methods:paragraph0, 1:Methods:paragraph2, 1:Methods:paragraph3, 1:Methods:paragraph4, 1:Methods:paragraph5, 1:Methods:paragraph6, 1:Methods:paragraph7, 2:Results:paragraph0, 2:Results:paragraph2, 2:Results:paragraph9, 2:Results:paragraph10, 2:Results:paragraph13, 2:Results:paragraph14, 2:Results:paragraph15, 2:Results:paragraph16, 2:Results:paragraph17. No complete-paper or supplement audit claimed. none_found. Subclaims: SC2, SC5.

### [55] Fibrin drives thromboinflammation and neuropathology in COVID-19.

**Shows:** Fibrin inflammatory-domain manipulation and antibody protection narrowed thromboinflammatory and neural injury mechanisms.

**Does not show:** Model and antibody scope; no human trial demonstrating systemic-cytokine primacy or dominant mortality fraction.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5.

### [56] COVID-19 tissue atlases reveal SARS-CoV-2 pathology and cellular targets.

**Shows:** Multiorgan autopsy atlases mapped inflammatory and failed-repair states in viral-RNA-positive and negative regions.

**Does not show:** End-stage selection and RNA localization do not prove replication, initiating cell type or primacy.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5.

### [57] Pulmonary Vascular Endothelialitis, Thrombosis, and Angiogenesis in Covid-19.

**Shows:** Autopsy comparison showed prominent microvascular thrombosis and angiogenesis versus influenza.

**Does not show:** Small fatal-only series; ultrastructural virus identification and mechanism direction require independent verification.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC4, SC5.

### [58] Inflammatory Biomarker Trends Predict Respiratory Decline in COVID-19 Patients.

**Shows:** Rising CRP during first48h predicted respiratory deterioration better than initial levels.

**Does not show:** Retrospective risk marker, no targeted causal or absolute cytokine-magnitude test.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1.

### [59] A molecular single-cell lung atlas of lethal COVID-19.

**Shows:** Fatal single-cell lung atlas linked cytokine-producing cells, immune dysfunction and failed alveolar repair.

**Does not show:** End-stage transcriptomic/interaction inference; linked erratum remains unaudited; not direct concentration or intervention evidence.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. ErratumIn: PMID34625743. Subclaims: SC4, SC5.

### [60] The Activin/FLRG Pathway Associates with Poor COVID-19 Outcomes in Hospitalized Patients.

**Shows:** Activin/FLRG predicted poor outcomes; hamster blockade did not worsen viral load/pathology.

**Does not show:** No positive improved disease endpoint in the abstract; absence of worsening is not efficacy or driver proof.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4.

### [61] High-dimensional single-cell analysis reveals the immune characteristics of COVID-19.

**Shows:** Peripheral single-cell profiling described monocyte expansion, T-cell depletion and reduced antigen presentation.

**Does not show:** Peripheral descriptive states do not establish causal direction or systemic hypercytokinemia magnitude.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC1, SC4.

### [62] SARS-CoV-2 sensing by RIG-I and MDA5 links epithelial infection to macrophage inflammation.

**Shows:** RIGI/MDA5 and downstream inhibition reduced infected epithelial-to-macrophage inflammatory signalling.

**Does not show:** Cultured primary-cell circuit, no in vivo necessity or compartment-specific mortality test.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC5.

### [63] IL-13 is a driver of COVID-19 severity.

**Shows:** IL13 neutralization reduced infected-mouse mortality without lowering virus; CD44 blockade narrowed HA mediation.

**Does not show:** Human dupilumab observations confounded; IL13/HA is not an IL6/IL1 clinical test nor systemic-storm magnitude proof.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. UpdateOf: PMID33688686. Subclaims: SC4, SC5.

### [64] IL-1β drives SARS-CoV-2-induced disease independently of the inflammasome and pyroptosis signalling.

**Shows:** IL1beta loss improved infected-mouse disease/survival while several inflammasome/pyroptosis deletions did not.

**Does not show:** Germline developmental and mouse-adapted-virus scope; cytokine causality does not establish the presumed release pathway.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC3, SC4.

### [65] SARS-CoV-2 drives JAK1/2-dependent local complement hyperactivation.

**Shows:** SARSCoV2 induced epithelial complement and JAK1/2 inhibition normalized selected genes/C3a.

**Does not show:** Expression and cell inhibition do not establish human necessity, efficacy or dominant systemic lesion.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. UpdateOf: PMID32702726. Subclaims: SC3, SC5.

### [66] Heightened Innate Immune Responses in the Respiratory Tract of COVID-19 Patients.

**Shows:** BAL transcriptomes in eight cases showed inflammatory and IFN-stimulated signatures.

**Does not show:** RNA is not secreted cytokine concentration; small disease-stage sampling; no causal compartment test.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. none_found. Subclaims: SC5.

### [67] Compartmental immunophenotyping in COVID-19 ARDS: A case series.

**Shows:** Four early-intubated cases showed lung and blood hyperinflammation, particularly lung IL8.

**Does not show:** Raw BAL values were diluted; blood and BAL not universally discordant; selection and small sample limit generalization.

**Access/status:** Complete primary abstract and linked-status metadata examined. Selected body sections: 1:Results:paragraph13, 1:Results:paragraph22, 2:Discussion:paragraph4, 2:Discussion:paragraph7, 2:Discussion:paragraph8, 0:Methods:paragraph0, 0:Methods:paragraph1, 0:Methods:paragraph2, 0:Methods:paragraph3, 0:Methods:paragraph4, 0:Methods:paragraph5, 0:Methods:paragraph6. No complete-paper or supplement audit claimed. none_found. Subclaims: SC5, SC4.

### [68] Distinct inflammatory profiles distinguish COVID-19 from influenza with limited contributions from cytokine storm.

**Shows:** Only7 of168 COVID patients had storm-like profiles; adjusted cytokines were lower than influenza and IFN signatures suppressed.

**Does not show:** Relative storm definition and observational sampling; shared comparative research network does not ensure all cohorts independent.

**Access/status:** Complete primary abstract and linked-status metadata examined. No complete-paper or supplement audit claimed. UpdateOf: PMID32511543. Subclaims: SC2, SC4.