Neural Noise

Computational modeling of resistance to hormone-mediated remission in childhood absence epilepsy

3 min read

  • Computational modeling has been applied to resistance to hormone-mediated remission in childhood absence epilepsy (Frontiers).1
  • Model-based prediction of treatment response is relevant to neural dynamics and epilepsy management.1
  • Methods are transferable to other neural dynamics and computational neuroscience applications.1 1

Weekly enrichment (2026-07-20)

  • The study built a minimal conductance-based thalamocortical network of 475 single-compartment neurons in NetPyNE/NEURON, comprising layer-5 regular-spiking (RS) and intrinsically bursting (IB) pyramidals, layer-6 non-tufted RS pyramidals, a deep low-threshold-spiking interneuron, and thalamic reticular (RE) and thalamocortical (TC) cells.2
  • Frontal cortex was represented by IB neurons and parietal cortex by RS neurons; the authors varied the layer-5 IB:RS ratio from a frontal-dominant 95:5 configuration to a parietal-dominant 5:95 configuration, with a 75:25 network used as the fitting baseline.2
  • Healthy spindle oscillations emerged with a ~7 Hz peak network frequency at 100% cortical GABAA conductance, while reducing cortical GABAA inhibition to 10% of baseline drove a dominant ~5 Hz spike-wave discharge (SWD) state (physiological SWDs fall in the 2.5-5 Hz band).2
  • Allopregnanolone (ALLO), the progesterone metabolite that rises at puberty, was modeled as a GABAA-receptor modulator by fitting synaptic conductance, binding/unbinding rates, and channel-open duration to experimental data at physiological ALLO concentrations (~1-20 nM in rat cortex).2
  • Enhanced frontocortical connectivity was imposed by scaling maximal synaptic conductances by a factor of 7 for IB-to-IB, 5 for IB-to-NRS, and 3 for NRS-to-IB synapses.2
  • Only the parietal-dominant network (higher layer-5 RS proportion) recovered from SWDs after ALLO under increased frontocortical connectivity, whereas IB/frontal-dominant configurations failed to remit, implicating bursting-cell composition, not just connection strength, in treatment resistance.23
  • The modeling choice is anatomically grounded: roughly 73% of layer-5/6 pyramidal neurons are intrinsically bursting in rat prefrontal cortex versus about 50-60% in somatosensory/parietal cortex.2
  • Model predictions align with clinical imaging: a prospective EEG-fMRI/MEG cohort of 16 drug-naive CAE children (11 ethosuximide responders, 5 nonresponders) found nonresponders had increased frontal-cortex and decreased precuneus pretreatment connectivity.4
  • CAE accounts for approximately 18% of childhood epilepsies with about 70% of patients achieving remission; independent cohorts identify early age of onset, occipital intermittent rhythmic delta activity (OIRDA), and early treatment as favorable prognostic factors.25
  • For BCI/neurotech, the work proposes a connectivity- and cell-composition-based biomarker that could steer personalized neuromodulation or drug targeting, though it remains a simulation/preprint result lacking a naturally remitting animal model for validation.3

Footnotes

  1. https://news.google.com/rss/articles/CBMipwFBVV95cUxOSXZiWS12U2hpREZ0aTFsS2YtOEVvNkJpTngtOGx1T0lvWDVzLUloZUFhdTdUbzNqeUpRbWxreGNoc19uQm0tVmRFX2hvQVMxSTJsNk8tQy13T1BMUVRrajNVcEg1OU9SM2pNeENha0FHQy1zaGw0NnIzNG1oRHlNWEMxSmJPV08zdmVHTDRYeGhYSlVfeFljdktaU3lJaDhFclZTNURJUQ?oc=5 ↩ ↩2 ↩3 ↩4

  2. https://www.frontiersin.org/journals/computational-neuroscience/articles/10.3389/fncom.2025.1733650/full ↩ ↩2 ↩3 ↩4 ↩5 ↩6 ↩7 ↩8

  3. https://www.biorxiv.org/content/10.1101/2025.09.04.674333v1 ↩ ↩2

  4. https://doi.org/10.1111/epi.14067 ↩

  5. https://jpedres.org/articles/clinical-and-electrophysiological-prognostic-factors-of-childhood-absence-epilepsy/jpr.galenos.2021.95914 ↩

Graph View