NET-Targeting Therapies
Recent evidence from COVID-19 and long COVID research implicates neutrophil extracellular trap (NET) dysregulation as a post-viral thrombo-inflammatory mechanism with potential therapeutic implications for ME/CFS (Chapter Immune System Dysfunction). The core pathology—impaired NET degradation due to reduced DNase activity, pDC deficiency, and/or genetic susceptibility—suggests two complementary therapeutic strategies: enhancing NET clearance and inhibiting excessive NET formation. All approaches described below are investigational; none have been tested in ME/CFS or long COVID.
1 Recombinant DNase I (Dornase Alfa)
(Certainty: 0.20.) Recombinant human DNase I (dornase alfa, Pulmozyme) is FDA-approved for cystic fibrosis, where it cleaves extracellular DNA in airway mucus to reduce viscosity. Its mechanism—cleaving the DNA backbone of NETs to promote their clearance—is directly applicable to the NET/DNase imbalance identified in severe COVID-19 by Garcia et al. (Garcia et al. 2024), where functional DNase deficiency drives NET accumulation and thrombo-inflammation. Preclinical support: DNase I treatment reduced multi-organ injury in SARS-CoV-2-infected mice (Veras et al. 2023). However, NET degradation releases DAMPs (histones, MPO, proteases) that are themselves pro-inflammatory, and systemic DNase administration in non-CF populations has not been studied for safety or efficacy.
Expected time to response: Hours–days (pharmacokinetic clearance of circulating NETs). The CF experience suggests onset within 2 weeks for mucus effects; systemic NET clearance time course is unknown.
Monitoring parameters: NET remnant levels (MPO-DNA, H3cit), D-dimer, calprotectin, inflammatory markers (CRP, IL-6), DNase activity assay.
Stopping criteria: Anaphylaxis, haemoptysis (CF warning), paradoxical inflammatory flare (NET degradation releasing DAMPs), no NET marker reduction after 4 weeks.
Drug interactions (ME/CFS co-prescriptions): No known interactions with fludrocortisone, midodrine, LDN, mestinon, beta-blockers, or antihistamines, but systemic use in non-CF populations is entirely novel. Nebulised formulation may not achieve adequate systemic bioavailability for vascular NET clearance; IV/inhaled comparative PK data needed.
Severe/very-severe patients: No data. CF safety data from chronic use are reassuring but patients have normal (not depleted) DNase physiology. Risk of paradoxical inflammation in already-symptomatic patients is unknown.
Key prediction: ME/CFS patients with documented NET/DNase imbalance (elevated NET/DNase ratio) receiving dornase alfa show reduced circulating NET markers at 4 weeks compared to baseline. Falsified if NET markers unchanged despite confirmed DNase deficiency—suggesting NETs are tissue-sequestered (microclots, endothelium) and inaccessible to circulating enzyme.
2 PAD4 Inhibition
(Certainty: 0.15.) Disulfiram, FDA-approved for alcohol use disorder, inhibits peptidyl arginine deiminase 4 (PAD4), the enzyme required for histone citrullination during NETosis. By blocking PAD4, disulfiram could prevent new NET formation, complementing DNase-mediated clearance of existing NETs. However, disulfiram has significant toxicity (hepatotoxicity, peripheral neuropathy) and NET inhibition may impair anti-pathogen immunity—a particular concern in ME/CFS where innate immune function is already suppressed. The risk-benefit profile is unfavourable for a mechanism with only indirect relevance to ME/CFS.
Contraindications for bedbound patients: Hepatotoxicity risk is unacceptable for patients who cannot be easily monitored with regular LFTs.
Key prediction: PAD4 inhibition should reduce H3cit formation ex vivo in ME/CFS neutrophils. Falsified if disulfiram does not reduce NET-specific markers despite confirmed PAD4 binding.
Limitation: Disulfiram is not a selective PAD4 inhibitor (off-target effects abundant). No human NETosis prevention data exist.
3 Metabolic NETosis Modulation
(Certainty: 0.20.) Sanhueza et al. (Sanhueza et al. 2026) demonstrated that post-infectious insulin resistance drives NETosis independently of glucose in long COVID (n=60 prospective cohort). Metformin activates AMPK, inhibiting mTOR—a pathway required for NETosis—while simultaneously improving insulin sensitivity. This dual mechanism is the basis of the hypothesis that metformin reduces metabolic-NETosis in a metabolically defined ME/CFS subgroup, though no ME/CFS data exist.
Steelman — the case for testing it: The strongest argument is not that metformin failed, but that the trials that failed tested unselected patients. Lim2026 and Reis2026 did not stratify by insulin resistance, the design that can mask a true subgroup effect; and metformin’s one positive RCT-level signal (COVID-OUT, (Bramante et al. 2023)) arose in a defined early/metabolically-loaded context. The negative symptom trials therefore leave a testable subgroup hypothesis open — they do not refute a targeted one.
Clinical caveat: This is a research hypothesis, not a demonstrated clinical benefit, and should not be prescribed outside a registered trial. The distinction matters: the proposed REE/NETosis endpoints are unverified (never measured in ME/CFS), which is not the same as refuted (tested and failed). What is refuted is unselected established-disease symptom benefit: one completed RCT was negative ((Lim et al. 2026)) and a second adaptive trial stopped its metformin arm for futility ((Reis et al. 2026)). Metformin’s positive signal is confined to prevention with early acute-phase initiation (COVID-OUT: 41% relative reduction in long-COVID incidence, (Bramante et al. 2023)) — a result irrelevant to a patient who already has established ME/CFS. Against the subgroup case, no routine clinical NET marker exists, so the trial cannot be judged on a routine clinical marker: the mechanism is not currently falsifiable with routine clinical assays, so the negative symptom trials retain real (if not decisive) evidentiary weight. The dose, monitoring, and stopping fields below describe a research protocol only, not an established clinical regimen. Certainty 0.20 reflects plausibility of the mechanism, not any ME/CFS evidence. See Could LDN vs Metformin Ever Be a Phenotype-Guided First-Line Choice?.
Expected time to response: Weeks (metabolic remodeling + insulin sensitization requires sustained AMPK activation).
Starting dose (research protocol only — not a clinical recommendation): 250 mg once daily (well below standard 500–850 mg BID; ME/CFS GI sensitivity common).
Monitoring parameters (research protocol only): Mechanism-proximal endpoints (NET markers, insulin sensitivity) plus GI tolerability and B12 (metformin depletes); baseline and periodic renal function (eGFR), since renal impairment is metformin’s principal contraindication. Note that HOMA-IR is useful as a stratification/enrichment variable but does not track the hypothesized NETosis mechanism, so it must not be used as a futility endpoint.
Stopping criteria: Intolerable GI symptoms, lactic acidosis (rare but potentially fatal; risk rises with renal impairment), declining eGFR, or no change in the mechanism-proximal research-assay endpoints (NET markers/insulin sensitivity) after 12 weeks — not HOMA-IR alone.
Contraindications for severe/bedbound patients: Not for use in bedbound/severe patients — the directional effect on energy metabolism cannot be established outside a research protocol, and this population is most vulnerable to harm. Additionally contraindicated at eGFR \(< 30\) and not initiated at eGFR 30–44; hold during acute illness, vomiting, dehydration, or iodinated-contrast procedures (sick-day rules — directly relevant given autonomic dehydration risk); creatinine-based eGFR overestimates renal function in low-muscle-mass patients, so consider cystatin C.