Dear colleagues, first of all, I apologize not to be with you for traveling restriction, but I hope we will meet very soon. Thank you for giving me the opportunity to talk about my experience regarding wake surgery for Le Grey Glauma and especially how I do it in a connectome-based surgical approach. You have no conflict of interest. With 20 years of follow-up, we can say now that we were able to increase both overall survival and quality of life in patients harboring Le Grey Glauma. To achieve this goal, we did early surgery, namely without a wait-and-see attitude anymore. We developed the concept of supratotorization, namely by taking a margin around the preoperative flare abnormality if functionally feasible. And of course, we preserved the quality of life of patients by doing an intraoperative mapping and trying to better understand the functional connectome at the individual level throughout the resection. I will not insist on the natural history of Le Grey Glauma. Everyone now knows that this tumor is not benign, occurring in the vast majority of cases in patients enjoying a normal life, and finally becoming malignant and killing the patient. What we have to do is to perform a systematic neuropsychological examination before any treatment. And then you will see that in almost half of cases, patients enjoying a normal life nonetheless developed already some cognitive disturbances, especially regarding language, memory, attentional processing, executive functions, and emotion. It's interesting to note that these deficits are not related to the lobar location of the tumor, but much more related to the invasion of the Y matter tracts, which are the limitation of the neuroplastic phenomenon. When I started two decades ago, the survival in this kind of Le Grey Glauma was around 6 to 7 years. In the French Le Grey Glauma Consortium, we published a paper already 7 years ago, a prospective study demonstrating that we reached survival around 15 years, thanks to early and maximal resection. More recently, with my colleague Luc Taillandier in Nancy, we published another paper now with a median survival above 16 years with a perfect preservation of the quality of life in the vast majority of patients above 12 years. This is possible only if we accept definitely that localizationism does not exist in the brain. We have to move toward a connectomic account of brain processing and to become neuroscientists in order to better understand mechanisms of neuroplasticity. To achieve this goal, of course, functional imaging is very helpful. Nonetheless, at the individual level, functional MRI and DTI are not reliable, especially with very low sensitivity, very low specificity, and of course, knowing that DTI is not a direct reflector of the function of the Y matter tracts. This is the reason why, for 20 years, I preferred to do systematically wake mapping, whatever the location of the tumor, with cortical and subcortical electrical mapping in a patient performing a cognitive assessment. Then it's possible to do online anatomofunctional correlation throughout the resection. To select the task, it's very important to ask to the patient what is his definition, her definition of quality of life. Indeed, it's possible to use almost all tasks into the operating theater, but there is a limitation of time. So how I do it means that before any surgery, I will save time to better understand what are the needs of each patient according to his or her job, hobby, habits, leisures, lifestyle, and so on. I will also try to take into account the environment of the patient because I received patients coming from five continents, and the rules are not the same. For instance, legal issues are different because you cannot drive in France and probably in US with ME and APTIA, while you can drive in some countries like North Africa, for instance. We have to be careful about the healthcare system because if you would like to prescribe a post-operative cognitive rehabilitation to the patient, it's reimbursed in France while it's probably not in US, and so on. Therefore, into the operating theater, I will do a mapping à la carte according to the wishes of the patient, but also according to the connectivity in order to preserve the projection fibers and the association fibers, which are definitely the limitation of your plastic mechanisms. To illustrate this philosophy, I recently published a paper in order to show that whatever the lower location of the tumor, you will finally be into the contact of the connectivity in the end of the resection. Indeed, if you remove a tumor involving the so-called Broca's area, which definitely does not exist, the left parietal lobe or the left temporal lobe, at the end, you will be into the contact of the dorsal phonological pathways, and especially the arcuate fasciculus. Namely, when you re-stimulate, you will induce conduction aphasia. In other words, into the operating theater, my philosophy is to remove as much as I can a part of the brain invaded by chronic tumor disease according to functional cortical and subcortical boundaries identified in real time, thanks to cognitive monitoring. So I will not use technology, no neural navigation, no DTI, no fMRI, no intraoperative MRI, no microscope, no ECoG, and you will see that you can nonetheless achieve 99% of good results. I do not understand why the vast majority of neurosurgeons now have the habit to do awake surgery in the so-called left dominant hemisphere for language mapping and to continue to achieve resection under general anesthesia in the so-called right non-dominant hemisphere. Indeed, this right hemisphere is critical for so many functions in order to continue to enjoy a perfect normal life, especially complex movement, spatial cognition, critical for driving, social cognition, critical for relationships between human beings, non-verbal semantics, executive functions, but also emotion, personality. So into the operating theater, we will do first of all a cortical mapping by incorporating additional tasks according to the definition of the quality of life by the patient, like for instance calculation if the patient is a mathematician, or spatial cognition mapping if the patient is a dancer within the right so-called non-dominant hemisphere, and so on. But in all cases, because we are human beings, we need to preserve emotional processing, and it's definitely possible to administrate mentalizing tasks into the operating theater. To give a little bit more detail, we published atlases not only regarding the cortex but also the white matter tracts of networks involved in the control of complex movement, especially regarding the bimanual coordination, which is definitely very important if you would like to play sport, or to play piano, or to be a surgeon. It's also very important to do the mapping of the somatosensory feedback throughout the resection, including in the right hemisphere. And definitely you need, once again, for the patient to be awake in order to give you this very accurate feedback throughout the resection. We demonstrated that Broca's area does not exist, and if there is an output for speech, it is located within the ventral premotor cortex. So we have to better consider networks underpinning different subfunctions, especially for language. In other words, you have to know that there is a dorsal pathway underpinned by the arcuate fasciculus, but also by the lateral part of the superior longitudinal fasciculus connecting this ventral premotor cortex to the inferior parietal lobe, and underlying the so-called articulatory loop. In other words, when you will stimulate, patient will have a complete anarthria. There is also a ventral semantic pathway underpinned by the inferior frontoccipital fasciculus, which should be preserved in all cases, whatever the site of the tumor. There are other pathways I have no time to detail, but especially the left inferior longitudinal fasciculus, basically located critical for lexical access. Indeed, we demonstrated that even if the naming task is still intact at the end of the resection, if there is an increase of the reaction time, then it could be a problem in order to resume a normal life, especially to return to work. In the right non-dominant hemisphere, we demonstrated that it could be very helpful to incorporate the line bisection task, especially at the Y matter tract mapping level, in order to identify the part two of the superior longitudinal fasciculus, and to avoid any permanent hemineglect. It's very important also in the so-called SMA location to map the cognitive function, especially the working memory. Otherwise, the patient can have some problems of this executive function following resection, and some problems to return to normal life. Indeed, it's definitely possible to incorporate further tasks, like dual tasks, or to ask the patient to name the item which was presented previously, in order to monitor this higher order cognitive function. It's also very important to ask the patient to not only recognize items in front of him or her, but also to combine semantically these items in order to map and monitor the multi-modal nonverbal semantic processing. It's very important to know that the right inferior frontooccipital fasciculus is critical for this nonverbal semantic processing. It's also possible to ask patient to recognize emotion expressed by faces into the operating theater, in order to monitor the so-called mentalizing, namely the theory of mind, which is absolutely critical for empathy, and then for relationships between human beings. We demonstrated that there was a wide network subserving and mentalizing, first of all, with different cortical hubs, including in the so-called right non-dominant hemisphere, with the inferior frontal gyrus, the dorsolateral prefrontal cortex, as well as the superior temporal gyrus. In addition, the right inferior frontooccipital fasciculus is also critical, and then should be mapped and preserved at the end of the resection. Recently, we modeled this so-called right non-dominant hemisphere, and you can see the complexity of the networks involved in so many functions, not only movement and language, but also cognition and emotion. Even more recently, we introduced a new task into the operating theater, allowing the patient to make a self-evaluation at the end of the task, like mentalizing task, but also naming task, in order to be sure that we did not induce any modification of the metacognition, namely the capability to know what we know. Fortunately, the brain plastic mechanisms are there in order to compensate our mistakes. Indeed, we recently demonstrated that before any treatment, it was, in the vast majority of cases, already a compensation thanks to the contralateral hemisphere, especially with an increase of the volume of the contralateral homologous, like, for instance, the contralateral insula, with also an increase of the functional connectivity of this area. So explaining why it's possible to do a very extensive resection of the insula involved by the tumor, thanks to a compensation by a wide network involving the other hemisphere. This is confirmed by post-operative functional imaging, for instance, after supplementary motor areas and room. When the patient recovered, we can demonstrate, thanks to post-operative fMRI, that there was a recruitment of the contralateral hemisphere, in this case, the contralateral SMA and premotor cortex. This is very important in order to increase the extent of resection by doing another resection a few years later, thanks to mechanisms of neuroplasticity which occurred in the meantime. For instance, in this example, I left voluntarily during the first resection the so-called knub of the hand, which was finally compensated five years later, allowing to remove this area in a patient enjoying a perfect normal life three months following surgery. This is true for so-called language area, like Wernicke area. I was not able to remove completely the tumor in the left dominant temporal lobe during the first surgery. And finally, 10 years later, I removed it, thanks to mechanisms of reorganization occurring in the meantime, and the patient continued to enjoy normal life. Another example regarding Broca's area, knowing that I was not able to remove the parso-opercularis during the first surgery, because still eloquent, but a few years later, it has been possible to remove completely the left inferior frontal gyrus, of course, thanks to the preservation of the deep connectivity, especially the IFF and the arcuate fasciculus. It is sometimes possible also to do a very provocative resection involving both frontal lobes, knowing that, once again, mechanisms of neuroplasticity occurred in the meantime, thanks to a very extensive post-operative cognitive rehabilitation. Of note, this patient enjoyed a perfect normal life after the second resection, too, able to drive, to work full-time, and to take care of his family. What about the results regarding more than 800 diffused Legrègleioma operated on within so-called eloquent areas, like the Broca's area, the Wernicke's area, behind the labivane, or the central area, or the insula, or the corpus callosum? I have no mortality, because I have a very good team, not only regarding neuropsychologists, but also neuroanesthesiologists. The rate of severe permanent deficit with permanent hemiparesis is 0.5% in my experience related to too deep stroke when I was younger. It's very exciting to see that 25% of patients improved after surgery, and of course, after post-operative cognitive rehabilitation, knowing that half of them did not have a perfect cognitive assessment before surgery. And of course, 80% of positive impact on control of epilepsy, so improving the quality of life. Recently, we published a paper demonstrating that it was possible not only to preserve the cognitive status of patients, but also to improve some cognitive domains. So definitely, thanks to an extensive post-operative cognitive rehabilitation. We also published in Incidental Le Grey Glioma that 97% of patients returned to an active life, including working full-time. We will very soon publish another paper showing for the first time in the literature, not only that it was possible to do a second surgery, but also a third surgery in Le Grey Glioma patients before MacLingdon transformation, with an increase of the extensivization for each surgery, thanks to additional mechanisms of neuroplasticity. And in this sub-series, the median survival is almost 18 years. Regarding my old series, median survival is above 17 years. To achieve this kind of results, it's very important to be a neuroscientist, because we are neurosurgeons, and then to be able to propose new model of cognition in order to use them for the pre-operative planning, but also into the operating field. To help younger people, we published an atlas of neuroplasticity in the MNI template, so giving them the opportunity to predict before surgery if the patient will recover or not according to the extensivization, especially within the white matter tracts. Once again, to help younger neurosurgeons, we recently published a soon atlas in the World Literature about 16 functional domains regarding the critical networks with cortical hubs and white matter tracts, absolutely essential if we would like to preserve the quality of life of the patients. And now we developed software allowing to use this atlas for a pre-operative planning for each individual case. The next step will be to better understand the relationships not only within brain networks, but also between neural networks. It's very important also regarding the interactions between both hemispheres, and definitely meaning that the right non-dominant hemisphere does not exist. So we developed the concept of metanetwork, demonstrating that the brain is not fixed and that it is a perpetual succession of equilibrium states. In 2D operating theater, this concept is very helpful because you can ask a patient to do two tasks simultaneously to stimulate and to see that by disrupting some specific network, the patient is still able to move, is still able to talk, but is not able to continue to perform both tasks simultaneously. So the goal is really to do mapping and cognitive monitoring à la carte based on better understanding of the functional connectome at the individual level. And to this end, be careful to technology because the risk for younger people is really addiction and never this kind of information could be provided by the actual technology. So now how I do it directly in 2D operating theater, I will take the example of a 30 year old man, medical doctor, with incidental discovery of this very voluminous left paralympic Le Gray glioma because of headaches. The patient, of course, would like to preserve a perfect quality of life. The patient is on a lateral position with the anterior part, which is on right, and posterior part of the brain on the left. We awaken patient after not being under general anesthesia. Then I used low intensity as usual between 2 to 2.5 milliamps. We tried to identify the Broca's area, but definitely there was a negative mapping at this level. Nonetheless, it's very important for me to continue to benefit from a positive mapping in order to be sure to use the same parameters of electrical stimulation at the level of the white matter tract in the end. So we identified the ventral premotor cortex 1, 2, 3, 4. This induced anarthria. Then in 5, so the dorsolateral prefrontal cortex, which serves as the posterior and the upper limit of resiction, as you can see, we identified one hub involving semantic processing. Now I can do the resiction according to a subpial resiction. So I just coagulated the cortical surface at the level of the true negative mapping, but I will never coagulate anymore. So I will not split the Silvian fissure to avoid any spasm, but I only use a cotton in order to push the Silvian fissure and to do this subpial resiction at the expense of the left inferior frontal gyrus. So we removed the orbitofrontal part, the pars triangularis, but also the foot of the pars opercularis, definitely by leaving the dorsolateral prefrontal cortex still involved in semantic processing. Now the patient is, of course, awake by doing dual tasks, namely movement of the right upper limb combined with a naming task, which is alternated with a semantic association task. Now I will do almost a monoblock position of the opercular part of the tumor and refer it to the lab of neuropathology, and I will arrive at the level of the surface of the insula. You can see it under the retractor. In order to continue to do a subpial dissection, and the goal will be to lift up the insula sulcus, the superior insula sulcus, in order to remove the insula from inside. And the retractor will help me to retract specifically the insula cortex, the pia mater, and to do a subpial resiction, but from inside. In order to improve my surgical access, I decided to remove very quickly the orbitofrontal part of the tumor. And then now it will be more easy to have a perfect access to the insula itself. So the patient is still awake, doing a dual task every four seconds. And now I am in the insula. I can remove it without any coagulation. And of course, the danger beyond the risk to have an evascular problem, but I am protected because I'm doing a subpial dissection without any coagulation once again, the danger will be in the depth at the level of the y-matter tracts, which are the limitation of the neuroplastic potential. In other words, I need absolutely to identify at this level the inferior frontoxbuton fasciculus. In order to improve the extent of resiction, I will remove also the temporal operculum. And now finally, I will stimulate into the contact of the ventral semantic pathway. And there is no sound, but you have to believe me, the neuropsychologist told me that patient had some problems regarding the semantic processing. So that means that I was into the contact of the temporal stem, so the IFOF running into the temporal stem, and I did final hemostasis before to close under generalized tissue. You can see the postoperative MRI demonstrated a complete resiction of the so-called Broca's area, which does not exist, of the insula itself into the contact of the temporal stem and the IFOF, but also the lateral part of the lentiform nucleus. There is a very small but objective residue, less than one cc at the level of the anterior perforating substance voluntarily left in order to give the opportunity for this patient to return to a normal life, what he did as a medical doctor. So to conclude, beyond survival, even if, of course, we increased, as you have seen, the survival over 17 to 18 years, we have to preserve the quality of life of Le Greg glioma patients, not only by avoiding to induce permanent hemiplegia and or aphasia, but also by preserving the higher-order cognitive functions, such as the executive function, as well as the emotional processing, in order to allow patient to return to normal life preserving his behavior, his or her personality, and then to be active for at least 15 to 20 years. So the goal is not just to remove a tumor with an oncological point of view, but first of all, to treat a patient as a human being by understanding the interactions between his or her connectome and the natural history of the disease, which will be radically changed thanks to our management. Thank you very much for your attention.

wake surgery

Le Grey Glioma

connectome-based surgical approach

supratotorization

intraoperative mapping

functional connectome

neuropsychological examination