Dear colleagues, first of all, I hope you are well in this difficult period. I thank you very much for giving me this opportunity to talk about early maximum safe physician in diffuse low-grade glioma with a personalized connectome-based approach. The new philosophy in oncology is to do an early surgery to improve both overall survival as well as the quality of life of patients. To achieve this ambitious goal, we have to propose more systematically a preventive surgery including an incidental diffuse low-grade glioma. In other words, we have not to consider anymore the wait-and-see attitude. Second, we have also to perform a maximum resiction because extent of resiction is significantly correlated to survival. Thus, if functionally feasible, we have to achieve a supra-total resiction and to consider it as the first therapeutic option. Finally, this resiction should be safe. In other words, we have to do intraoperative functional mapping in all cases in order to reject the tissue involved by the tumor according to a connectome-based resiction. We have better understood natural history of diffuse low-grade glioma, especially the fact that these tumors are never stable. They are growing in all cases as demonstrated objectively if you do a volumetric assessment each time you meet a patient with a new MRI. We know that these patients are, in the vast majority of cases, young, enjoying normal life with the tumor revealed by seizures. On the other hand, we also know that this tumor will become in all cases malignant than inducing a neurological deficit and finally leading to the death of the patient with median survival approximately within the 6 to 7 years following the first ANTO. This is a typical example of a patient followed in another institution in agreement with the classical wait-and-see rule and the patient was followed for 3 years with an increase of the volume of the tumor and finally the occurrence of enhancement. Then the patient was operated on with radiotherapy and chemotherapy and finally died 5 years after the first symptoms because it was too late. It's also common to say that patients are well while if you are doing an objective and extensive neuropsychological assessment before any treatment, you will discover, in more than 80% of cases, cognitive deficits, especially regarding the attentional processing, executive function, like working memory, emotion, social cognition, like empathy, mentalizing, and personality. Interestingly, these deficits are not so much related to the lower location of the tumor but much more to the invasion of the white matter tracts because the subclinical connectivity is the limitation of neuroplasticity. When these limits of brain plastic potential are reached, then the patient will have seizures, explaining why, in the vast majority of cases, this is the first symptom. In the French glioma network, we demonstrated about more than 1,500 low-grade glioma patients that the volume at diagnosis was directly correlated to the survival. Indeed, more the tumor is big, more the risk of magnetotransformation is high. By doing two MRI scans 1 to 2 months apart before any treatment, since there is no emergency to treat patients with low-grade glioma, then you will be able to calculate the growth rate of the tumor at that time in this patient. This kinetics is related to the survival. Indeed, if the tumor is growing faster, the prognosis is worse. It's very interesting to note that this growth rate is not related to the molecular profile of the tumor. Especially, you can have a Y-type astrocytoma with a very slow growth and then a good prognosis. In our series with Y-type astrocytomas, only 15% of them had a behavior very close to a glioblastoma. More than 84% of patients were still alive at more than 6 years following surgery with an extensive resection of this kind of tumor and in vast majority without any adjuvant treatment. This is the reason why in the European Low-Grade Glioma Network, we propose that molecular biology is only part of the story but not the absolute truth regarding the history of the tumor following surgery. In other words, the adjuvant treatment should be adapted to each patient according to multiple parameters and not only the molecular pattern. We know also that these tumors are very heterogeneous and this is very important because this major point, not underledged by the WHO classification, is very crucial regarding the adjuvant treatment following an extensive resection. We will see later that very frequently, it's possible to avoid to give immediate postoperative adjuvant treatment following the resection of a low-grade glioma with some foci of Magdalena transformation in the middle of the tumor. In all cases, we have understood that a diffuse low-grade glioma is not stable, that the patients are not so well, that we have to calculate the tumor volume objectively as well as the growth rate of the tumor and then to adapt the therapeutic strategy with first-of-all surgery rather than to continue only to fall out this patient. The main goals are to minimize the volume in order to avoid Magdalena transformation while preserving or even improving the quality of life of patients. Two near-randomized studies have evaluated the benefit of surgical resection in comparison with a biopsy in low-grade glioma. In cases of only biopsy, the median survival was between 6 to 7 years. Nonetheless, in order to be objective regarding the real impact of surgical resection on the natural history of low-grade glioma, we have to be very rigorous by calculating the volumetric assessment before and after surgical resection. In the French Glioma Network, we did early resection in a prospective series with more than 1,000 diffuse low-grade glioma patients by applying this concept of maximal and early safe surgical resection and by doing pre- and post-operative volumetric assessment on the flare-weighted MRI, we demonstrated that the survival was more than 15 years and was directly related to the extent of resection. It was also evidenced by the French consortium that beyond the post-operative volume, the survival was significantly related with the possibility to reoperate patients a few years later. As a consequence, today, except in rare cases of gliomatosis-like tumor, there is no reason to propose biopsy anymore. We have to go directly to an extensivization according to individual functional anatomy and then to increase both the survival and the quality of life of each patient. So the question is not anymore why to operate patients with low-grade glioma, but why not to operate? And in the vast majority of cases, this is because the surgeon doesn't want to induce any permanent deficit in a patient enjoying more or less a normal life before resection. So the solution is to apply the mapping techniques for each patient in order to do an extensive resection according to the understanding of the connectome at the individual level. Indeed, these tumors are very frequently located within so-called eloquent areas according to a localizationist view of brain processing, such as the supplementary motor area, the insula, the primary motor or somatosensory cortex, or the so-called Broca's and Wernicke's language area, and so on. However, in a connectomal view of brain processing, breaking with the Broca's Wernicke model, it's finally possible to operate within so-called eloquent areas without inducing any permanent deficit. Why functional neuroimaging is increasingly used before surgery in order to help for surgical indication and planning, we have nonetheless to acknowledge that this technology suffers from major limitations. First, functional MRI is not able to differentiate areas critical for brain functions, so which must be spared surgically from areas involved but not critical for a given function, so which can be compensated after surgical resection. This is very important in order to avoid to stop prematurely the resection of a tumor, especially for instance within the so-called SMA. Second, the reliability of fMRI is very low, especially regarding higher cognitive functions such as language. Indeed, by correlating preoperative fMRI and intraoperative electrical mapping, it has been demonstrated that the sensitivity was only around 37%, with a specificity around 83%. This is also true by using the new methodology of a resting state fMRI. Indeed, we correlated the preoperative language mapping with the intraoperative electrical and cortical mapping and the correlations were around 85% of positivity. In other words, it's very difficult to make a decision today with 15% of errors regarding the surgical indication of planning for higher cognitive functions. Regarding DTI, we also correlated the preoperative tractography with the intraoperative electrical mapping at the level of the white matter tract and the correlation is only around 82%. In addition, DTI is not able to give any functional information regarding the white matter bundles. Recently, a dataset was sent to 20 research groups experts in DTI. They evidenced that the tractograms contain many more invalid than valid bundles and that 50% of these invalid bundles occur systematically across these expert research groups. Therefore, because functional neuroimaging is not reliable enough, even today, we decided for more than 20 years to do systematically weight mapping in patients with very few or no preoperative neurological deficits. The goal is to use intraoperative electrical stimulation in order to identify the structures both at cortical and subcortical level critical for brain functions with a high level of accuracy, reliability, allowing to do online anatomofunctional correlations. To achieve such a structural functional correlation in real-time into the operating theater, it's very important to have a dedicated team, especially with speech therapists and or neuropsychologists. It's very important too to ask to the patient to perform continuously tasks throughout the research and to adapt these tasks according to the quality of life of the patient defined by himself. Therefore, before surgery, we need to meet the patient and the family and to better understand what means their quality of life at the individual level based on their job, hobby, habits, leisures, lifestyle, and social activity. For instance, we will not ask a patient to speak different languages if it's only monolingual, but we can ask a patient to do the tasks both in French and English if he's able to speak both languages and if they are critical to enjoy a normal life following surgery. So we have to adapt not only to the location of the tumor but to the relationships between the location of the tumor and the functional networks around the glioma. In addition to the needs of the patient, we have also to take into account the social environmental aspects in order to preserve the quality of life. First of all, the legal issues because, for instance, the law is different from one country to another one in order to give the authorization for a patient to drive with epilepsy or hemianoptia. The cultural aspect because the neural basis of language is different if we are speaking about alphabetic language versus ideographic language. Also the socio-economical parameters because for a cognitive rehabilitation which is very helpful in order to give all chances for a patient to return to normal life, the reimbursement is not the same according to the health care system. Once again to optimally select the tasks into the operating theater we have also to take into account the location of the tumor but not only regarding the relations with the tumor and the cortical areas but most of all between the tumor and the subcortical pathways. For instance because the inferior frontal occipital fascicle here in yellow is critical for semantic processing we will have to ask a patient to perform a semantic association task whatever the lower location of the glioma within the frontal lobe, the insular lobe, the temporal lobe or the parietal occipital junction because in all cases we will be into the contact of this ventral semantic pathway at the end of the resection. As a consequence in my experience based on more than 1000 awake surgeries I did not use any image guided philosophy, no neural navigation, no DTI, no fMRI, no intraoperative MRI but also no microscope, no electrocorticography in order to do only a functional mapping guided resection according to the identification and the preservation of the eloquent structures critical for brain function both at the cortical and subcortical level throughout the resection for each patient. The new tradition for neurosurgeons is now to consider that in the left so-called dominant hemisphere we have to do more and more awake mapping in order to map language but that in the right so-called non-dominant hemisphere we can do a resection under general anesthesia without any risk. This is a simplistic view because in fact the so-called right non-dominant hemisphere is critical for so many functions in the brain of human beings. Indeed in order to give the opportunity for each patient to return to a perfect normal life including the familial, social, professional aspect it's crucial not only to avoid any plagiarism and or aphasia but also to allow patients to do complex movement, to have a perfect spatial cognition otherwise you cannot drive, to have a perfect social cognition otherwise you cannot communicate normally and you have a risk to lose your job, to have a perfect semantic processing not only regarding the verbal but also the non-verbal semantics, to benefit from a perfect executive functions like the ability to do two tasks simultaneously to use working memory but also to have a perfect conscious information processing and to keep your own personality so the emotional processes. For a long time we demonstrated that it was possible to map, to monitor and to preserve these higher cognitive functions into the operating theater for instance by doing multitasking in order to have a reflect of the executive functions or to add a non-verbal semantic association task in order to preserve the higher integration of semantic processing and so on. To do the mapping it's very important first of all to forget the localizationist view which does not exist within the brain but to go to a connectomal organization allowing to better understand online through galvanization the distribution of the neural networks and their interactions. For instance regarding language definitely the area of BRCA does not exist because this is a very wide network allowing to remove this area especially when invaded by brain tumor but also in order to have an access to the left insula through a cortical approach without inducing any severe permanent deficit in patients who are able to return to a normal life. It's also possible to adapt the tasks during surgery as we said according to the needs the wishes of the patient but also including his job. For instance if the patient is a lawyer then it's very important for him or her to do cross-modal judgment into the operating theater. If the patient is a mathematician we can add calculation into the OR. If the patient is a dancer then we can add including in the so-called right non-dominant hemisphere spatial cognition task using for instance a line bisection task. Of course in a way patient if he wants to continue to be able to dance with a high level following realization. If a patient wants absolutely to preserve a high level of executive function we can use a working memory like a hand back task into the OR to ask to the patient to name the previous slide. If the patient is a writer we can add some syntactical tasks into the operating theater. Of course this is a balance between the extent of realization and quality of life that we have to extensively explain to the patient and his family before to go to the operating theater. Most of all in all cases we have to propose to the patient before surgery to map and monitor the theory of mind namely the mentalizing in order to reduce significantly the risk to modify the behavior of the patient and his personality which is now possible. Beyond the cortical mapping we have to understand that most important is to map, monitor and preserve the subcortical white matter tracts which are the limitation of neoplastic potential. If we cut for instance the perimodal pathways at the end of resiction within the supplementary molar area or within the insula of course we understand that we will induce a permanent hemiplegia. But it could be a little bit more subtle. Speaking about the movement there is a huge difference between to avoid hemiparesis versus allowing the patient to continue to do a very complex bimanual task following surgery like to play violin or to be a surgeon for instance. And it's important to understand that there are some networks cortically and subcortically speaking involved in the control of movement a network called the negative motor network. We have to ask a patient during surgery to do a regular movement and when we will stimulate cortically and subcortically we will induce an arrest of movement. At the subcortical level we can induce such arrest of movement by stimulating the frontal striatal tract especially connecting the supplementary molar area to the striatum. It's possible to ask patient to do a regular bimanual coordination task and then we can disrupt bilaterally the movement while stimulating just one side including the so-called right non-dominant hemisphere. Once again this is absolutely crucial if the patient would like to do a complex movement following surgery and not just to avoid hemiplegia. In the same state of mind to do complex movement is possible only if we can benefit from somatosensory feedback. The first level is just to induce tingling in an awake patient into the operating theater in order to preserve the primary somatosensory pathways especially the thalamocortical fibers. But it's very important also if the patient wants to preserve the awareness of the body schema otherwise we have a risk to induce modification of this bodily awareness with some disabling consequence for the quality of life for instance to run. The principle is similar if we would like to preserve absolutely the optic tracts in order to avoid to induce a permanent hemianoptia because once again in the vast majority of countries the patient will not obtain anymore his license to drive. Regarding this patient awareness in other words to avoid any hemineglect it's possible as we already mentioned to ask the patient to do a line bisection task into the operating theater throughout the resection especially at the level of the right inferior parietal lobe. But the most important is to preserve the connectivity in the depth namely the port 2 of the superior longitudinal fascicle which is absolutely crucial if the patient would like to return to a perfect spatial awareness. Concerning language there are two main pathways. The first is the dorsal phonological pathway which is underpinned by the arcuate fascicle in the depth connecting the posterior part of the inferior frontal gyrus and middle frontal gyrus to the posterior part of the middle frontal of the middle temporal gyrus and inferior temporal gyrus not the superior temporal gyrus. In other words Broca's area is not directly connected by the vast majority of fibers to Wernicke's area. When you simulate this pathway you will induce conduction aphasia namely repetition disorders and phonemic paraphagia. If you stimulate the lateral part of this dorsal pathway that is the port 3 of the superior longitudinal fascicle connecting the ventral premotor cortex which is the lateral part of the precentral gyrus so the output to the supramarginal gyrus and the posterior part of the superior temporal gyrus then you will induce articulatory disorders due to a transitory desynchronization of the articulatory loop. That means that the patient can have a complete anaphria sometimes called a speech apraxia a speech arrest even by stimulating the so-called Wernicke's area at least in some patients. The ventral semantic pathway is mainly mediated by the inferior frontal occipital fascicle here in yellow which is the main fasciculus in human being connecting the prefrontal cortex to the temporal basal occipital and parietal structures. When you stimulate this pathway especially in the so-called left dominant hemisphere while patient is doing a naming task you will induce semantic paraphagia. Interestingly stimulation of the incinate fascicle which is more anteriorly located in relation to the inferior frontal occipital fascicle namely in red will induce some difficulties for lexical access regarding proper name. Both of these pathways are running into the temporal stem under the insula. The ventral semantic pathway is also constituted by the inferior longitudinal fascicle with the anterior part involved in lexical retrieval and the posterior part much more involved in the reading processing. Interestingly the inferior frontal occipital fascicle so direct pathway can compensate the rejection of the indirect pathway constituted by the ILF and the incinate fascicle explaining why it's possible to do an anterior temporal lobectomy even in the left dominant hemisphere while patient can recover. The inferior frontal occipital fascicle is also involved in non-verbal semantic processing and we can map and monitor this processing by adding a semantic association task. For instance the semantic link between gloves so hands and not feet. It's important to do that into the operating theater because the patient otherwise could have some difficulties regarding the noetic consciousness and the metacognition namely the consciousness of the fact that we know what we know. This is critical for introspection and to be human being otherwise the personality can change. Interestingly this IFOF is involved in this kind of a very integrative higher cognitive function also in the right so-called non-dominant hemisphere. In other words it's important to awake the patient and then to administrate a semantic association task even when we operate in the right side. This is true also regarding the emotional processing theory of mind namely the ability to understand the mental affective process of people in front of you by looking at the eyes. And this task can be administrated to patient during the mapping and monitoring throughout the resection if the patient would like absolutely to preserve his social cognition. Once again the right IFOF, inferior frontal occipital fascicle, is critical not only for semantic processing as we have seen, but also for the face-based mentalizing processing. So if the patient would like to return to a normal life, including social and professional activities, when the tumor is located within the right so-called non-dominant hemisphere, you have to propose to the patient to be awake in order to monitor this higher integrative function. This is a modeling of the so-called right non-dominant hemisphere and you can imagine the number of pathways involved in so many higher cognitive processing like movement control, spatial attention, mentalizing, non-semantic processing and so on. So we have to be aware about that before to say to a patient I will operate you under general anesthesia and you have basically no risk except possible hemiparesis, but I will compensate that by using MEPS. This is definitely not enough. The results about more than 800 diffused Legrès-Glioma patients knowing that more than 20% of them have been re-operated on, in my experience, explaining why I did more than 1000 awake surgeries in so-called unrespectable areas like Broca's area, Mernicke's area, Central Asia, Mernicke's area, Central Area, Insula, Carpus Callosum and so on. Of course these good reasons are related to the fact that localizationism does not exist explaining why functional reorganization is beyond our imagination within the brain of patients. First, regarding the functional results with intraoperative mapping. Beyond the fact that the mortality is nil, I induced 0.5% of severe permanent deficit related to deep stroke because I induced damages of the perforating arteries in two cases of patients with a right perforated Legrès-Glioma. Interestingly, in 30% of cases, thanks to post-operative cognitive rehabilitation, the cognitive scores were better three months following surgery in comparison with the pre-operative neuropsychological assessment because, of course, in all cases, our patients benefited from objective neurocognitive examination before and after each treatment. Plus 80% of positive impact on epilepsy control so with an improvement of the quality of life. Indeed, it has been demonstrated that the risk to have epilepsy and especially intractable seizures related to Legrès-Glioma was in fact due to the infiltration of the cortex around at the periphery of the tumor and not within the core of the tumor. In other words, more we remove the cortex invaded by the glioma, more we do a supratotal resiction, in other words, more we increase the chance to control epilepsy so to improve the quality of life. Of note, this kind of results can be reproduced everywhere in the world as demonstrated by meta-analysis with more than 8,000 cases with gliomas because the use of intraoperative electrical mapping allowed to decrease significantly the risk to induce a severe permanent deficit including in the so-called eloquent areas while increasing the extent of resiction. If it's not possible to remove completely the tumor diffusion for functional reasons during the first surgery, it could be possible to reoperate patients and to achieve this goal it's very important to better understand the mechanisms underlying neuroplasticity at the individual level not only before surgery but also after surgery to plan a reoperation a few months or years later. In this example, patients with a transitory supplementary motor area syndrome after recovery benefited from post-operative functional MRI. It's possible to see that the contralateral hemisphere participated in the compensation especially the contralateral supplementary motor area and premotor cortex then with a complex bilateral network allowing the patient to return to normal life. Based upon this better understanding of mechanisms underlying remapping, it's possible to consider a multi-stage surgical approach by doing a second surgery and then to increase the extent of resiction while preserving the quality of life because in the meantime the neural distribution changed in the same patient over years. This potential of reshaping can be potentiated by rehabilitation performed following the first surgery. Once again, it's possible to do pre- and post-operative functional MRI also regarding higher cognitive functions such as language and not only movement. Recently, we demonstrated that it was possible to have this reorganization involving both hemispheres by evidencing this modification within the connectivity and then to use this potential in order to reopen the door for repeated surgery while preserving the quality of life according to the definition, the needs, the wishes by the patient. This is a typical example of remapping within the left temporal lobe including the so-called vernicules area. During the first surgery, it was not possible to remove completely the tumor because functional sites both at cortical and subcortical level as demonstrated by tags were still critical for brain functions. Nonetheless, by doing a second surgery 10 years later in a patient enjoying a normal life, it has been possible at that time to do an extensive resiction of the left temporal lobe while the patient totally recovered and with a possibility to increase the extent of resiction without mechanism transformation and then by increasing the median survival of the patient. Same example within the so-called Broca's area. During the first surgery, the ventral premotor cortex and the parsopercularis of the left inferofrontal gyrus were still critical, pushing me to leave a small but objective amount of tumor. A few years later, a patient still enjoying normal life. Before any MacLindan transformation, I did a second surgery and at that time it has been possible to push the resiction while preserving the quality of life. This is a very good example of the huge potential of brain reorganization. We did a first left frontal lobectomy in a patient who returned to normal life. It was no adjuvant treatment. 10 years later, because the tumor grew but within the contralateral right hemisphere, it has been possible to do another surgery, of course under local anesthesia, in order to map and monitor the higher cognitive executive functions as well as the mental analyzing process. The brain was able to reorganize and the patient enjoyed a perfect normal life, including with normal scores regarding the neuropsychological examination and the mentalizing. Incorporating additional tasks into the operating theater in order to preserve a perfect quality of life is not against the extent of resiction and the median survival, which is more than 16 years in my experience, with 25 years of patient still alive with 20 years of follow-up and more. Interestingly, the extent of resiction is not related to the molecular profile. This is even the reverse because we demonstrated in my experience that the extent of resiction was much more higher in patients with an astrocytoma, none 1p92 could delete it. So that means that the impact on the survival is a significant therapeutic factor provided by surgery per se and not indirectly related to a selection bias. The concept of supratotor resiction is very easy, namely we know that there are tumor signs in diffuse gliomas beyond what we can see on preoperative MRI or intraoperative MRI, explaining why I do not use intraoperative imaging because this is not the reality of the disease. So the goal is to do a functional-based resiction according to the functional networks and to push the resiction until these eloquent areas have been encountered but not before. So if we can remove more than the flare abnormality visible on MRI, this is better. Indeed, when I compared my series with total resiction versus supratotor resiction, in the supramarginal resiction group, there were no Mecklenburg-Land transformation and chemotherapy in only one case without any patients with radiotherapy. Interestingly, this benefit of supratotor resiction has been confirmed with more than 10 years of follow-up. In a subgroup of patients with supramarginal resiction, no postoperative radiotherapy has been performed. In patients with no permanent deficit, no Mecklenburg-Land transformation and then no death, even if they have been cured because recurrence has been demonstrated in more than 50% of cases, so leading to re-operation thanks to mechanisms of neoplasticity occurring in the meantime. So the question is now to know how it's possible to increase the rate of supratotor resiction, of course, while preserving the quality of life of patients. And the answer is to do a preventive neurosurgery, so to arrive earlier in the natural history of the disease in smaller tumours, so in incidental low-grade glioma, explaining why we propose the principle of screening in France. Indeed, incidental low-grade gliomas will grow in all cases. There is no stable diffuse low-grade glioma. Second, if you do an extensive neuropsychological examination in patients with incidental discovery of low-grade glioma, in 60% of cases they are not asymptomatic, they suffer from cognitive disorders. This is the reason why it's absolutely mandatory to do an objective neuropsychological assessment in all patients. Third, there is a risk to see a sudden transformation, so in glioblastoma in this example, in asymptomatic patients. For instance, in this illustrative case, the patient has been furloughed in another institution for many years, while it would have been possible to do a supratotor resiction by removing completely the right so-called non-dominant insula at the moment of diagnosis. Unfortunately, the patient was just furloughed and suddenly the tumour transformed in a patient still asymptomatic and it was too late. So in more than 100 patients with incidental discovery of diffuse low-grade glioma, in my experience, I have seen that because the volume of the tumour was smaller, we increased the rate of supratotor resiction in 30% of cases, plus 30% of complete resiction, plus no partial resiction, plus no epilepsy, plus no permanent diaphysis. So we change radically in the natural history of the disease, including in the so-called eloquent areas, because 30% of patients with incidental diffuse low-grade glioma had infinitely already some foci of Maglindan transformation demonstrated a posteriori by neuropathological examination. In the recent series of patients with an incidental low-grade glioma who underwent awake surgery, we demonstrated that 97% of them were able to return to a normal social and professional activity, with 95% of them still alive with 67 months of myntholype. Therefore, based on these excellent results, both regarding the functional and oncological outcomes, we proposed the principle of screening in the general population in front of the National Academy of Medicine in France. The Academy accepted the principle based on the fact that in a survey of healthy potential candidates, 66% of them say that they would accept to participate in such a screening based on MRI program. Even if these oncological and functional results are possible, including in eloquent areas, thanks to mechanisms of neuroplasticity, we have to be aware about the fact that there are nonetheless some limitation of this plastic potential, especially in the depth at the level of the white matter tracts. This is the reason why sometimes it's not possible to remove completely the tumor, here for instance, due to the invasion of the control lateral hemispheric connectivity. In these cases, we can consider to give neoadjuvant chemotherapy, which could induce a shrinkage at the level of the invasion of the white matter tracts, and then to reopen the door of surgery by this time achieving supratoterorization without inducing any permanent deficit. An example of a tumor for a long time considered as inoperable in a patient with intractable seizures. I did a first partial resection with temozolomide, despite the fact that the tumor was not 1p92 co-deleted, inducing a very impressive shrinkage, especially in the depth at the level of the connectivity, opening the door to a second surgery with the results you can see on the right, knowing that beyond the fact that the tumor is still under control, the patient improved his quality of life because no epilepsy anymore, and then the possibility to return to a normal life based on the fact that it was able for him to drive again. We can do radiotherapy, but the problem is that, thanks to evidence-based medicine, it has been evidenced that early radiotherapy will not change radically the median survival, but on the other hand, that to irradiate too early a brain will induce, in the vast majority of cases, a decrease of the cognitive function, so of the quality of life. This was indeed confirmed more recently about thousands of patients who underwent surgery with or without radiotherapy, and it has been finally concluded that there was no long-term overall survival difference in patients with high-risk grade 2 gliomas treated with chemoradiotherapy in comparison with only chemotherapy following surgery. The problem is that, when we irradiate, we have to be in mind that there is a minimal common brain, definitely the part of the brain which cannot be compensated in case of brain damage, including, of course, surgical resignation, but also following irradiation, especially the white matter tract. So, if a cognitive assessment should be performed before and after each surgery, to be objective, cognitive assessment should be also achieved before and after each irradiation of the brain in patients with a glioma, and when it was done in a few studies of the literature, it was demonstrated that there was a decline of the cognitive function, in this example in verbal fluency, within the two years following radiotherapy. A recent series using tractography demonstrated that white matter structure began to change only one year following radiotherapy. So, based on these recent results, the URTC proposed to elaborate a new MGMT methylation score, and they support the hypothesis that a high score may benefit from themozolomide treatment for patients with IDH1-mutated low-grade glioma, regardless of the 1p-19q status. This MGMT methylation score may identify patients who benefit from first-line treatment with themozolomide to defer radiotherapy for long-term preservation of cognitive function and quality of life. This is the reason why we have proposed a new philosophy based on a recursive approach in patients with low-grade glioma, namely to start in all cases by an extensive and safe resection, except in some cases of gliomatosis-like tumours. Then, if it was not possible to achieve a complete resection, to come back after inducing your plasticity thanks to postoperative cognitive rehabilitation. Then, when it is not possible to re-reoperate, to do chemotherapy regardless the molecular status, in order to preserve the cognitive functions, and then to induce, if any, a shrinkage, in order to reopen the door to another surgery, and in fine, to irradiate the patient, in our experience, more or less only after 10 years following the first surgery, and of course, in the meantime, with normal cognitive scores and normal life. Once again, this strategy is not against the median survival, because with my colleague Luc Taillonier, we were able to demonstrate that with early and safe resection, the median survival is more than 16 years in patients with low-grade glioma. In this same spirit, namely not only to increase the median survival of patients with low-grade glioma, but also to preserve a perfect quality of life, so to avoid to over-treat these patients, including when they had some foci of MacMillan transformation, grade 3 or 4. If we did an extensive resection, we proposed to avoid adjuvant treatment too early in the immediate postoperative period. Indeed, in the selected subgroup of patients with a smaller preoperative flare volume less than 50 cc, and most of all a smaller postoperative flare volume after surgery less than 1.5 cc, if the tumor is growing slowly, and especially in cases of only good androglioma, the 5-year survival rate is 95% without any chemotherapy and after surgery with no postoperative radiotherapy in order to preserve the quality of life, and then the cognition of patients. A typical example of a patient I operated on 18 years ago in 2002 with a low-grade glioma, but in fact in the middle, as you can see, enhancement, and it was already a focus of grade 3. No chemotherapy, no radiotherapy had been performed. The patient was allowed with a normal life. And finally, I did a second surgery three years ago in 2017. At that time, it was very exciting to see that the neuropathological examination said that, in fine, it was a low-grade glioma without any sign of maglinal transformation. So that means that typically we removed just a focus of maglinalization, and this is the reason why the WHO classification should acknowledge this hated originality in low-grade glioma in order to tailor individualized management and not just based on the results of the molecular pattern. Interestingly, these results have been also reproduced in another team, like in Leeds, for instance, because the team reported that they've allowed some patients with an extensive resection, including with some foci of glioblastoma in the middle of a low-grade glioma. Of course, this kind of functional-based resection with intraoperative functional mapping and cognitive monitoring is possible only if we are also neuroscientists, namely, if we can deal online with the connectivity of the brain of these patients at that time. To do that, we recently reported the sole atlas in the world literature with almost 2,000 stimulation sites showing the critical brain functions, both at the subcortical and subcortical level, regarding 16 functional domains like movement, language, mentalizing, spatial cognition, and so on. These results are free access online, so it could be very important for a neurosurgeon to have this kind of atlas in their own mind in order to avoid, to induce any damage of the so-called minimal common brain. Very recently, we were also able, with Guillaume Herbet, to model the right and the left hemisphere processing regarding the neural networks, and to see their interactions, namely, not only to think that one function is related to one specific location within the brain, but not only to say that one function is related to one network, but in fact to see the interactions between different networks in order to explain behavior of a human being. This is the principle of meta-networking theory of cerebral functions. In other words, we have more and more to understand for our patients the dynamics within, but also between systems, in order to preserve the personality, the behavior, and the creativity of our patients for the next 10 to 20 years, because we know that we can change radically the natural history of the disease, at least for the next two decades, when we diagnose a patient with a Le Grey gland. In practice, this is very important when we do a stimulation into the operating theater, because for a long time, neurosurgeons thought that when there was a transitory deficit induced by electrical stimulation, it was related to the area stimulated by the probe at that time. In fact, we have to understand that this is a desynchronization within one network, for instance, language or movement, but sometimes also desynchronization between networks, explaining why a patient into the operating theater will be able to continue to move or to speak in isolation, but not perform both simultaneously. Namely, we can disrupt the metanetworking brain and then it will induce a problem of multitask. And once again, this is absolutely crucial to know that before surgery, in order to propose to the patient to preserve or not these higher cognitive functions, otherwise he will not return to a perfect normal life. To conclude, in order to perform a connectome-based resection, we have to understand that the most important thing to the operating theater is to talk to the brain and to forget the disease, because we will do a resection according to the functional boundaries, because it has been demonstrated now with 20 years of follow-up that we will increase both the quality of life and the median survival. So, intraoperative mapping and cognitive monitoring performed by speech therapists and neuropsychologists directly into the operating theater in real-time are more sensitive and reliable than you imagine. Technical aspects are not really the most important for brain surgery. The most important is to understand the principle of the connectomal organization of the brain, the central nervous system, in human beings. Surgery has a significant impact on natural history of diffuse Legrès glioma. It has been demonstrated and we know that we have to operate earlier and more extensively. But to do that, we have to be objective. So, we have to do an objective assessment of the excentralization by doing volumetric evaluation preoperatively and postoperatively. We have to take a margin around what we can see on the preoperative MRI, if possible, because definitely the flare abnormality is not the disease. We have to understand that surgery will improve also the quality of life, especially by controlling seizures in more than 80% cases, but also thanks to postoperative cognitive rehabilitation. And if we do an objective cognitive assessment before and after surgery, by adapting a postoperative cognitive rehabilitation, we can induce mechanisms of neuroplasticity and then to consider more systematically to reoperate. So, be careful before to administrate too early chemotherapy and especially radiotherapy if we would like to preserve the cognition for 10 to 20 years. So, quality of life should be the first end point, because if a patient is well, that means that the patient is alive, in a sense, while the reverse is not true. Supramaximal resection has a significant impact on the natural course of the low-grade glioma by avoiding magnesium transformation. So, because we improve also the quality of life, that means that we have to consider more and more to do a functional mapping guided resection, because we will increase the tumor removal, especially if we arrive earlier and earlier. So, by doing surgery in incidental discovery of low-grade glioma, and once again we can discuss more systematically about a screening according to our environment, the culture, the country and the socio-economical issues. In this era of molecular medicine, we have to be careful about evidence-based medicine, because they are based upon artificial trials and not upon the real truth in real life. So, the wishes of the patient are more important than the protocols and we have to give them the absolute choice in order to decide what should be their quality of life after each treatment. We have to come back to empathy, wisdom, and of course to take into account the fact that each patient is unique, including regarding his brain organization, in order to be in agreement To conclude, we have to go towards a preventive and individualized functional neoncology in order to treat a patient and not a tumor. And according to this principle, we have already demonstrated that it was necessary to have a functional neoncology According to this principle, we have already demonstrated that it was possible to increase significantly the median survival and the quality of life of human beings. Now, the goal is to continue this prospective collection of data. Thank you very much.

early surgery

maximum safe resection

diffuse low-grade glioma

oncology

preventive surgery

personalized approach

overall survival

quality of life

maximum resection

supra-total resection

intraoperative functional mapping