Biohybrid approaches for epilepsy treatment: Towards the convergence of regenerative medicine and neural engineering

This thesis focuses on novel approaches to treat mesial temporal lobe epilepsy (MTLE). Being the most frequent epileptic syndrome in the adult population and the most often refractory to medical therapy, MTLE provides a major contribution to the global burden of the epilepsies. When medications fail to control seizures, the first-line treatment is surgical ablation of the epileptic focus. However, this is not feasible in all patients and may not guarantee a seizure free life. Further, it is a radical approach that should be considered with caution. Deep brain stimulation (DBS) has demonstrated a valid option to ameliorate drug-refractory MTLE. However, it still cannot guarantee a seizure free life in all patients and it is still a symptomatic treatment, although recent evidence points to possible disease-modifying effects. The major pitfall of current DBS is that the stimulation policy is devised by trial and error and based on arbitrary parameters. The exponentially growing body of research attempting to find the optimal stimulation policy witnesses the unmet clinical need of finding a unifying framework for brain modulation in epilepsy. Here, I use hippocampus-cortex slices treated with 4-aminopyridine and primary hippocampal spheroids to explore novel biohybrid approaches inspired by the temporal dynamics of the interictal activity generated by the hippocampal subfield CA3 and its demonstrated anti-ictogenic role. Starting from surrogating this interictal pattern to open-loop electrical stimulation of the subiculum, I demonstrate the possibility of artificially recreating this pattern to control limbic seizures. With this, I also map the statistical parameters describing the temporal distribution of the CA3-driven interictal events; this map can provide a set of brain-informed choices for the future implementation of AI-driven stimulation operating upon seizure prediction. I then deploy the CA3-driven interictal events as feedback signal to drive the operation of an artificial bridge to restore the functional dialog between the CA3 and the entorhinal cortex. The bridge demonstrates effective in decreasing ictal activity and robust to failure. Remarkably, causality analysis demonstrates that this approach restores the cortico-hippocampal loop to a statistically similar degree of the intact hippocampal loop. With this, I also unveil the intrinsic adaptive properties of the CA3, which acts a biological neuromodulator adapting to ongoing cortical inputs. Aiming at addressing those MTLE cases presenting with severe CA3 damage, I also explore the feasibility and efficacy of a conceptualized biohybrid devices relying on the adaptive properties of CA3/hippocampal cultured cells, to deploy them as adaptive biological neuromodulator. With this, I demonstrate the existence of an intrinsic anti-ictogenic clock within native CA3 networks. Lastly, I take the last step toward functional biohybrids for epilepsy treatment by establishing a bidirectional grafthost communication loop between cultured hippocampal spheroids (graft) and hippocampus-cortex slices treated with 4-aminopyridine (diseased host). These pilot experiments warrant caution in devising cell-based therapies for brain repair and lay the foundation for the establishment of a radically novel approach to brain regeneration, namely enhanced regenerative medicine.