Trigeminal neuralgia is intense, electric-shock-like facial pain, most often triggered by touch, chewing or speech. It results from dysfunction of the trigeminal nerve, the 5th cranial nerve. In most cases, no cause is found, and trigeminal neuralgia is termed essential. In the first instance, treatment is based on medication. In cases of drug resistance, radiosurgery is a possible treatment option. This involves performing neurolysis, delivering a very high dose of ionizing radiation to the trigeminal nerve. The immediate success rate of radiosurgery is 80-90%. However, in the long term, around 30% of patients experience complications (mainly hypoesthesia of the face on the treated side, paresthesias, masticatory disorders, neuropathic pain) and 30% of patients experience a recurrence of neuralgic pain. Most of these complications are permanent, and there are very few effective treatments, either medical or physical. Recurrence and complications are correlated, i.e. patients with hypoesthesia have a lower risk of recurrence. Certain technical parameters are associated with the efficacy and toxicity of radiosurgery, notably the position of the point of impact of the rays on the nerve. However, for identical treatment techniques, there are currently no known prognostic criteria for the efficacy and toxicity of radiosurgery. Numerous radiobiological studies have demonstrated that sensitivity to ionizing radiation differs from one individual to another, with each person having his or her own tolerance threshold. Indeed, 5-10% of patients are hypersensitive to ionizing radiation and are at very high risk of developing late complications \[Bentzen et al. 2010\]. There are currently commercial tests for individual radiosensitivity, based on a simple blood test, whose clinical value has been demonstrated in predicting complications in patients irradiated for breast or prostate cancer. These tests are based on the rate of radiation-induced lymphocyte apoptosis, known as the RILA (Radiation Induced Lymphocyte Apoptosis) score. Numerous teams have shown retrospectively and then prospectively that a high RILA score is significantly correlated with the absence of the development of radiation-induced late adverse events, with a negative predictive value of over 90% (level of evidence 1) \[Azria et al. 2015; Mirjolet et al. 2016; Talbot et al. 2019\]. In practical terms, the test gives a lymphocyte apoptosis score for each patient. A cut-off point is set below which the patient is considered radiation hypersensitive. In this study, the investigators propose to correlate the RILA score with the occurrence of severe late toxicity in patients treated by radiosurgery for trigeminal neuralgia. In the event of a positive result, this would make it possible either to adapt the radiosurgery technique to minimize the risk of late complications, or to contraindicate radiosurgery and refer patients to other treatment methods.