Provisional GPCR Autoantibody Model Extensions (Azcue 2026)
{Contingent: if CellTrend ELISA signal is non-specific binding, all models below collapse. Included for research completeness; not validated. (Azcue et al. 2026).}
Certainty: 0.55. Five-equation ODE with variables A_ab(t) (AAb concentration), R_sens(t)/R_desens(t) (sensitive/desensitized receptor fractions), HRV(t), C(t) (cognitive performance). Key dynamics: receptor desensitization by AAb binding vs resensitization; HRV as function of receptor ratio; cognitive performance via inverted-U muscarinic AAb effect. Prediction: HRV shows hysteresis after AAb clearance (tau approx 1/kappa_resens); cognitive performance shows bistability. Contingent on GPCR AAb pathogenicity. (Azcue et al. 2026) (Stein et al. 2025).
Certainty: 0.60. Seven new DAG nodes: GPCR_AAb, AAb_IgG1, AAb_IgG4, FcgammaR_Immune, Receptor_Desens, Vagal_Blind, Estrogen_Level. Edges: GPCR_AAb to Receptor_Desens; Estrogen_Level to GPCR_AAb (female bias mediation); AAb_IgG1 to FcgammaR_Immune (pro-inflammatory pathway). Prediction: 40–60% of sex effect on autonomic dysfunction mediated by GPCR_AAb pathway. Contingent on validated GPCR AAb measurement. (Azcue et al. 2026).
Certainty: 0.50. C(M) = C0 + DeltaC_max * M/(M+M_half) * (1 - M/M_toxic). Captures Azcue2026 counterintuitive positive correlation: low-moderate titers enhance cognition; high titers become toxic. M_half approx 30–40th percentile, M_optimal approx 60–70th percentile, M_toxic approx 90th percentile. Prediction: quadratic R2 exceeds linear R2. Contingent on correlation being causal, not survivorship bias. (Azcue et al. 2026).
Certainty: 0.42. dA_IgG4/dt and dA_IgG1/dt with IL-21-driven IgG4 and IFN-gamma-driven IgG1 switching plus IgG4-to-IgG1 conversion. Total AAb (CellTrend) = A_IgG4 + A_IgG1. REAP/Luminex detection = eta_IgG4A_IgG4 + eta_IgG1A_IgG1 where eta_IgG1 much-less-than eta_IgG4. Prediction: A_IgG4/A_IgG1 ratio decreases over time; REAP null explained by eta_IgG1 less than 0.2*eta_IgG4. Least parsimonious of four explanations; assumes isotype bias in platform detection. (Azcue et al. 2026) (Germain et al. 2025).
Certainty: 0.50. PEM_state = 1 if Demand greater-than R_reserve * theta_threshold. dR_reserve/dt = kappa_prod - kappa_AAbA_abR_reserve - kappa_desensDemandR_reserve. Recovery: dRecovery/dt = kappa_rec(1 - PEM_state) - kappa_triggerPEM_state with hysteresis (kappa_rec less-than kappa_trigger). Prediction: beta2-AAb-positive patients have lower theta_threshold; immunoadsorption increases it; pacing keeps Demand below R_reserve*theta_threshold. Contingent on GPCR AAb causality (r^2=0.20, 80% unexplained). (Azcue et al. 2026).