Hypoxia and Oxygen-Sensing Pathway Modulation

CautionSpeculation: Normobaric Intermittent Hypoxic Training (IHT) for ME/CFS

Certainty: 0.45. Brief controlled hypoxic exposures may induce beneficial HIF-1\(\alpha\)-dependent adaptations including mitochondrial biogenesis, angiogenesis, and metabolic efficiency (Gangwar et al. 2019). IHT has shown promise in Parkinson’s disease (Janssen Daalen et al. 2025) and in athletic conditioning. The ME/CFS adaptation requires: supine position (eliminates orthostatic stress), mild hypoxia (FiO2 0.16, equivalent to 2000m), short cycles (3–5~min hypoxia alternating with 3~min normoxia, total 30~min), and strict PEM monitoring (48~h post-session).

Mechanistic rationale. Brief hypoxia → HIF-1\(\alpha\) stabilization → transcriptional programme activation → VEGF (angiogenesis), EPO (erythropoiesis), PGC-1\(\alpha\) (mitochondrial biogenesis), GLUT1 (glycolytic capacity). Repeated sessions are hypothesised to raise the hypoxic threshold and improve exercise tolerance.

Safety. Potential PEM trigger if protocol exceeds individual tolerance — requires graduated introduction (start FiO2 0.19, decrease 0.01 per week). Contraindicated in severe ME/CFS (bedbound patients), uncontrolled POTS, or hemoglobinopathies. Normobaric hypoxia carries no barotrauma risk. Hypoxia-induced vasodilation could worsen orthostatic intolerance in susceptible patients.

Testable prediction. 8 weeks IHT (3 sessions/week) improves 6MWT distance ≥40~m vs sham (FiO2 0.21) in mild-to-moderate ME/CFS, with sustained CBFv during hypoxic cycles.

CautionSpeculation: HIF-PH Inhibitors (Roxadustat, Daprodustat) for ME/CFS

Certainty: 0.30. Prolyl hydroxylase domain (PHD) inhibitors stabilize HIF-1\(\alpha\)/2\(\alpha\) by blocking oxygen-dependent degradation, increasing endogenous EPO and erythropoiesis. Roxadustat is approved for anemia of chronic kidney disease; daprodustat is in clinical use. In ME/CFS, low-dose PHD inhibition could restore the hypoxic adaptation programme that appears blunted (see HIF Pathway Inertia hypothesis HIF Pathway Inertia: Failure of Hypoxic Adaptation in ME/CFS). Unlike exogenous EPO, PHD inhibitors upregulate the full HIF transcriptional programme (EPO + VEGF + glycolytic enzymes + mitophagy regulators), potentially producing broader metabolic benefit (Winkler et al. 2004).

CAUTION: + Could exacerbate autoimmunity — HIF-1\(\alpha\) promotes Th17 differentiation and B cell activation (Little et al. 2023) + VEGF elevation from HIF-2\(\alpha\) stabilization could worsen endothelial barrier dysfunction in patients whose HIF-2\(\alpha\) is already pathologically active (Section Immune Complex–Endothelial Injury as a Central Vascular Mechanism) (Ribeiro et al. 2026). This risk is subtype-dependent: patients with elevated baseline VEGF/vWF should not receive PHD inhibitors + Excessive erythropoiesis increases blood viscosity, potentially worsening cerebral hypoperfusion + No ME/CFS safety data exist; all literature is from CKD populations with different baseline physiology

Testable prediction. Low-dose daprodustat (1mg TIW, one-quarter the CKD dose) increases constant-workload cycling time to exhaustion by \(>=30\)% without elevating hematocrit above 45% in ME/CFS patients with blunted hypoxia-challenge EPO response.

CautionSpeculation: Belzutifan (HIF-2\(\alpha\) Inhibitor) for Post-Viral ME/CFS with Endothelial Dysfunction

Certainty: 0.45. Belzutifan is a selective HIF-2\(\alpha\) inhibitor approved for VHL disease-associated tumors that blocks HIF-2\(\alpha\)/ARNT heterodimerization, preventing transcriptional activation of VEGF, EDN1, and SERPINE1. Ribeiro et al. (2026) (Ribeiro et al. 2026) directly demonstrated that belzutifan rescues endothelial barrier integrity in spike S1-stimulated retinal endothelial cells — the only drug shown to reverse this specific post-viral mechanism. The initial certainty (0.25) has been increased to 0.45 based on convergent supporting evidence: (a) Sales et al. (2026) independently showed that HIF-2\(\alpha\) inhibition (via darunavir) preserves tight junction proteins ZO-1 and occludin in a separate epithelial barrier model (Sales et al. 2026), (b) the NRF2-HIF-2\(\alpha\) axis provides a mechanistic framework for sustained HIF-2\(\alpha\) dominance after NRF2 depletion (Shen et al. 2024) (Zhang et al. 2026), and (c) the general principle of viral HIF-2\(\alpha\) exploitation is supported across multiple virus types (HCV (Couteaudier et al. 2025), RSV (Morris et al. 2025)). However, belzutifan’s brain penetration is unknown — peripheral endothelial improvement is expected, but BBB effects are uncertain.

Mechanistic rationale. SARS-CoV-2 spike protein upregulates HIF-2\(\alpha\) in pulmonary and retinal endothelium, promoting VEGF production, intercellular gap formation, and increased permeability. If a similar mechanism operates in ME/CFS (whether from SARS-CoV-2, EBV, HHV-6, or other viral triggers), HIF-2\(\alpha\) inhibition could restore endothelial barrier integrity. The antiviral signaling protein IRF3 directly regulates HIF-\(\alpha\) cytoplasmic retention — viral infection relieves this retention, linking diverse viral triggers to the same HIF-2\(\alpha\) pathway (Deng et al. 2026).

Selection criteria: Post-viral onset, elevated VEGF/vWF, \(\beta_2\)AR-autoantibody-negative (the ~70% subgroup with no known ED mechanism — Section Immune Complex–Endothelial Injury as a Central Vascular Mechanism).

Dosing: Start 40~mg daily. Monitor hemoglobin (risk of anemia — HIF-2\(\alpha\) supports renal EPO production), blood pressure, VEGF q2 wks.

CAUTION. HIF-2\(\alpha\) is not HIF-1\(\alpha\) — the isoforms have distinct, sometimes opposing functions. HIF-2 inhibition could paradoxically worsen hypoxia sensing. HIF-2\(\alpha\) supports erythropoietin production in adults; its inhibition could cause anemia (15–20% incidence in VHL trials). No ME/CFS data exist. Contraindicated in pregnancy.

Testable prediction. 8 weeks belzutifan 40~mg daily reduces VWF antigen (endothelial activation marker) by ≥20% and improves 6MWT distance by ≥30~m in post-viral ME/CFS patients with elevated baseline VWF (>150% normal) and \(\beta_2\)AR-autoantibody-negative status. VEGF reduction should precede clinical improvement by 2–4 weeks.

CautionSpeculation: Darunavir/Ritonavir (HIV Protease Inhibitor, HIF-2\(\alpha\) Blocker) for Endothelial Barrier Rescue

Certainty: 0.40. Darunavir, an HIV protease inhibitor in clinical use since 2006, was recently shown to prevent HIF-2\(\alpha\) activation and preserve tight junction proteins (ZO-1, occludin) in an experimental oesophageal barrier disruption model (Sales et al. 2026). The mechanism is independent of protease inhibition — darunavir directly blocks HIF-2\(\alpha\) transcriptional activity, preventing the VEGF-driven permeability programme. This represents a repurposing opportunity with an established safety track record (20+ years in HIV populations), dramatically lower cost than belzutifan, and known pharmacokinetics.

Mechanistic rationale. HIF-2\(\alpha\) activation drives VEGF production, which induces intercellular gap formation and barrier disruption. Darunavir inhibits HIF-2\(\alpha\) activation, preserving junctional integrity (Sales et al. 2026). If the same HIF-2\(\alpha\) mechanism operates in post-viral ME/CFS endothelial cells, darunavir may provide barrier protection comparable to belzutifan but at a fraction of the cost.

Dosing: Darunavir 600~mg BID boosted with ritonavir 100~mg BID (standard HIV dosing). CYP3A4 interaction profile is well-characterized.

CAUTION. Hepatotoxicity (3–7%), rash, GI intolerance. Drug interactions via CYP3A4: fludrocortisone, midodrine, LDN, mestinon — all substrates or co-metabolised; levels may change substantially. Requires specialist monitoring. Low CNS penetration — peripheral endothelial protection expected; cognitive symptoms may not improve.

Drug interactions with common ME/CFS co-prescriptions: Darunavir/ritonavir is a strong CYP3A4 inhibitor and CYP2D6 inducer. Fludrocortisone: levels may increase (CYP3A4 substrate) — monitor for hypertension, hypokalemia. Midodrine: levels may increase (CYP3A4). LDN: naltrexone is CYP3A4 substrate — levels may increase. Mestinon (pyridostigmine): not CYP3A4-dependent — no interaction. Beta-blockers: metoprolol and propranolol are CYP2D6 substrates — levels may decrease. Antihistamines: fexofenadine is not CYP3A4-dependent — no interaction; loratadine is CYP3A4 substrate — levels may increase.

Contraindications for bedbound patients: Hepatotoxicity risk requires regular LFT monitoring. GI intolerance may worsen malnutrition risk. Drug interactions are substantial — polypharmacy in bedbound patients requires careful review.

Testable prediction. 4 weeks darunavir/ritonavir in post-viral ME/CFS reduces plasma VEGF and vWF by ≥25% and improves FMD compared to placebo in a double-blind, randomised crossover trial (n=20).

CautionSpeculation: EPO Stimulation Test as Diagnostic Probe for HIF Pathway Integrity

Certainty: 0.40. A standardized hypoxia challenge with serial EPO measurement could serve as a dynamic functional test of HIF pathway integrity in ME/CFS, analogous to the ACTH stimulation test for adrenal insufficiency (Winkler et al. 2004).

Protocol. 2~h mild hypoxia (FiO2 0.14, supine, continuous SpO2 monitoring, SpO2 maintained ≥88%) with serum EPO measured at baseline, 2~h, 4~h, 8~h, 24~h, and 48~h. The outcome metric is EPO AUC over 48~h, with detection threshold defined as less than 60% of age/sex-matched control median.

Mechanistic rationale. If HIF pathway inertia is a core feature of ME/CFS, the EPO response to a controlled hypoxic stimulus should be blunted relative to controls while basal EPO remains normal — a pattern consistent with impaired inducibility rather than absolute HIF deficiency.

Testable prediction. AUC of EPO response over 48~h is ≥40% lower in ME/CFS vs age/sex-matched healthy controls. Abnormal test (>2 SD below control mean) identifies a HIF-inert ME/CFS subtype with distinct clinical features (worse orthostatic intolerance, lower CBFv, steeper PEM gradient).

References

Couteaudier, M., M. Nivard, J. Cochard, F. Mammano, P. Roingeard, H. de Rocquigny, and P. Chouteau. 2025. “Hypoxia-Inducible Factor 2 Triggers the Production of Highly Infectious Native-Like Hepatitis C Virus Particles.” Cellular and Molecular Life Sciences 82 (1): 241. https://doi.org/10.1007/s00018-025-05739-0.
Deng, H., S. Jia, C. Zhu, J. Hua, Z. Wang, X. Sun, W. Liu, et al. 2026. “IRF3 Attenuates Hypoxia Signaling by Retaining HIF-α in the Cytoplasm.” Cell Reports 45 (1): 116815. https://doi.org/10.1016/j.celrep.2025.116815.
Gangwar, A., Pooja, M. Sharma, K. Singh, A. Patyal, G. Bhaumik, K. Bhargava, and N. K. Sethy. 2019. “Intermittent Normobaric Hypoxia Facilitates High Altitude Acclimatization by Curtailing Hypoxia-Induced Inflammation and Dyslipidemia.” Pflügers Archiv - European Journal of Physiology 471 (7): 949–59. https://doi.org/10.1007/s00424-019-02273-4.
Janssen Daalen, J. M., M. J. Meinders, F. Giardina, S. Mathur, P. N. Ainslie, D. H. J. Thijssen, and B. R. Bloem. 2025. “Hypoxic Conditioning in Parkinson’s Disease: Randomized Controlled Multiple n-of-1 Trials.” Nature Communications 16: 8469. https://doi.org/10.1038/s41467-025-63324-2.
Little, A. J., P. M. Chen, M. D. Vesely, R. N. Khan, J. Fiedler, J. Garritano, F. I. Maisha, J. M. McNiff, and J. Craft. 2023. “HIF-1 Regulates Pathogenic Cytotoxic t Cells in Lupus Skin Disease.” JCI Insight 8 (16): e166076. https://doi.org/10.1172/jci.insight.166076.
Morris, D., Y. Qu, A. Haas de Mello, Y. Jones-Hall, T. Liu, M. Weglarz, T. Ivanciuc, R. Garofalo, and A. Casola. 2025. “Role of Hypoxia-Inducible Factors in Respiratory Syncytial Virus Infection-Associated Lung Disease.” International Journal of Molecular Sciences 26 (7): 3182. https://doi.org/10.3390/ijms26073182.
Ribeiro, A., T. Wallraven, M. Lech, K. Adorjan, H. C. Stubbe, M. Seifert, A. Wöhnl, V. Kesseler, J. Negele, and C. Schmaderer. 2026. “SARS-CoV-2 Spike S1-Mediated HIF-2α Activation in Retinal Endothelial Cells Suggests a Mechanism Contributing to Post-COVID Endothelial Dysfunction.” Frontiers in Immunology 17: 1770758. https://doi.org/10.3389/fimmu.2026.1770758.
Sales, T., A. Daniel, L. Silva, A. Franco, H. Costa Filho, C. Monteiro, A. Ferreira, et al. 2026. “The Protease Inhibitor Darunavir Prevents Esophagitis and Impairments in Esophageal Barrier Function in a Mouse Model of GERD.” American Journal of Physiology — Gastrointestinal and Liver Physiology 330 (6): G718–26. https://doi.org/10.1152/ajpgi.00058.2025.
Shen, J., Y. Lai, Y. Lu, Y. Liu, J. Zhang, Y. Wu, Y. Pan, et al. 2024. NRF2-HIF2α Signaling Attenuates Endothelial Cell Senescence and Maintains Intercellular Junctions in Diabetes.” International Journal of Biological Sciences 20 (10): 4055–73. https://doi.org/10.7150/ijbs.96719.
Winkler, Andrea S., David Blair, John T. Marsden, Tim J. Peters, Simon Wessely, and Anthony J. Cleare. 2004. “Autonomic Function and Serum Erythropoietin Levels in Chronic Fatigue Syndrome.” Journal of Psychosomatic Research 56 (2): 179–83. https://doi.org/10.1016/S0022-3999(03)00514-8.
Zhang, Y., L. Cao, R. Zhang, W. Pi, V. Wee Yong, and M. Xue. 2026. “Oxidative Stress as the Orchestrator of the HIF Axis Switch in Intracerebral Hemorrhage.” Free Radical Biology and Medicine 244: 312–22. https://doi.org/10.1016/j.freeradbiomed.2025.12.011.