Although linked, factors related to the clinical–psychological status of the patient and those related to the dynamics of neural network reconfiguration are here considered separately.

Seizure is the first symptom in LGG, leading to diagnosis in the vast majority of patients [104]. An epileptic symptomatology is linked to a better oncological outcome [10]. Interestingly, computational models have evidenced that the onset of seizures corresponds with a time point which may already represent the overcompensation stage of cerebral plasticity, depending on the tumor growth rate and its pattern of progression, in particular in the event of massive invasion of the white matter tracts [105]. Translation of these results into the clinical situation is another argument in favor of early surgery in asymptomatic patients, before the occurrence of epilepsy [74]. Indeed, seizures can have a negative impact on daily life, in particular by preventing driving (and indirectly employment) for medico-legal reasons [91,106]. Furthermore, epilepsy is mainly elicited by the diffusion of the tumoral cells at the periphery of the LGG (and not by the glioma core itself) [107]. This explains why larger surgical resection has a higher impact on epilepsy. Indeed, postoperative seizure control is more likely when EOR is ≥91% and/or residual tumor volume is ≤19 cc [108]. This also supports the suggestion of supratotal resection (with removal of the PTZ) for functional reasons (in addition to the improvement of oncological outcomes), i.e., with an optimization of QoL as a result of freedom from epilepsy [49].

Notably, important inter-individual variability has been observed, with about 15% of LGG patients experiencing intractable seizures, notably in temporal and/or paralimbic gliomas. In this situation, it has been proposed to remove the hippocampus, even if not invaded by the tumor according to preoperative MRI, since this can result in significant improvement of epilepsy control [109]. During follow-up, the reappearance of seizures may be correlated with LGG relapse, and may prompt clinicians to propose reoperation prior to MT [25]. When further resection is not possible due to diffusion within critical structures, for example within the Rolandic area which is very epileptogenic [110], adjuvant medical treatment can have an impact on seizures [111].

Because LGG patients are generally young and enjoy active lives at diagnosis, presenting no or only slight deficit at so-called “standard neurological examination”, it has been claimed in the literature that these patients do not exhibit any significant functional disturbances [6]. However, important variations are found across patients in terms of their neurocognitive status. In fact, a recent cohort including 157 LGG patients who benefited from extensive neuropsychological evaluation before treatment showed that 55.4% of them had already experienced cognitive decline, in particular with respect to language, verbal episodic memory, psychomotor speed, attention, and executive functions (phonological and categorical fluency) [4]. Interestingly, neurocognitive impairments have also been found in patients with incidental LGG [112].

Due to earlier diagnosis of LGG, moderate or severe neurological deficits at clinical examination (e.g., hemiparesis or aphasia) are usually rare [17]. Beyond possible episodes of transient worsening which might be elicited by repeat seizures, permanent impairments that arise are generally due to voluminous gliomas with mass effect and/or to MT, with acceleration of the neoplasm kinetics. In other words, with more “prophylactic management”, such major deteriorations should cease to occur before the last stage of the disease, thus giving the opportunity for LGG patients to enjoy active lives for many years or even decades [7].

Definition of QoL is eminently variable from one human being to another. Indeed, beyond the fact that patients do not want to experience hemiplegia or aphasia, especially following surgical resection, their expectations are very different according to their lifestyles: do they work? (if yes, what kind of employment they have, and do they want to resume their professional activities postoperatively and/or during medical treatment?); what are their hobbies? (e.g., do they practice sports, art, etc.?); what is their socio-cultural environment? (e.g., do they speak multiple languages?) [127], do they need to drive? (if yes, what about possible medico-legal issues in case of visual field deficit and/or seizures?) [106]. On the basis of these individual wishes, with the aim of enabling each patient to develop long-term projects (such as getting married, having a baby, buying a house, etc.), and also with due consideration given to the neurocognitive assessment at diagnosis (i.e., the presence or otherwise of some degrees of disturbance), it is possible to elaborate tailored management “à la carte”, beginning with a selection of optimal tasks to be performed during awake surgery [128]. Moreover, therapeutic strategies should be re-adapted over time, not only according to the LGG course, but also taking into account possible changes in the patient’s priorities. An example of such an intra-individual variability could be when a patient would like to preserve executive functions during the first surgery because he or she was employed at the time, but sparing higher-order cognitive capacities is no longer absolutely mandatory a few years later because the patient has retired in the meantime. Therefore, a clear and extensive explanation of the principles of chronic tumoral disease should be provided to the patient and his or her relatives after the diagnosis. They should understand that management necessitates surveillance with constant anticipation of a specific lifelong therapeutic strategy, to give them the opportunity to make choices for their current and future life by thinking one step ahead—which increases the chance of finding a better psychological equilibrium [7,23].

The structural anatomy of the brain is highly variable across healthy individuals, especially at the cortical level [129], while variations are less pronounced at the level of the white matter tracts [18]. Furthermore, advances in non-invasive functional imaging methods which permit the investigation of the functional connectivity have resulted in the development of a large database demonstrating between-subject variability in the distribution of neural networks [130,131]. Recent models of neurocognition employed in basic neuroscience have rejected the classical localizationist dogma (one cerebral site underpinning one specific function), and go beyond a simple network organization of the central nervous system (one cerebral circuit underpinning one specific function) by evidencing the critical role of dynamic interplay within and across neural networks which allows behaviors to be constantly adapted to the surrounding world [132]. According to this meta-networking framework (based on a network of networks), complex cognitive abilities are made possible by the activation and coordination (combination or competition) of large-scale neural circuits involving domain-specific networks (e.g., movement or language circuits), and the activity of a multiple-demand system recruited during the performance of a wide range of cognitive-demanding activities with the aim of maintaining fluid intelligence [132,133].

Alongside this flexible and ever-changing physiological organization of the functional connectome, inter-individual anatomo-functional variability is significantly increased in brain-damaged patients, particularly in the event of slowly evolving lesions such as LGG [134]. These mechanisms of neuroplastic functional reshaping permit neurological compensation during LGG growth (explaining why the vast majority of patients are active at diagnosis), at least to some extent, considering that over half of LGG patients already experience some degree of cognitive disturbance at the first neuropsychological evaluation, as previously mentioned [4]. Maps of neuroplasticity have evidenced that potential for cortical reallocation is high (except for input such as the primary visual cortex and output such as the primary motor cortex), whereas axonal connectivity represents the main limit of functional reshaping [135,136,137]. Thus, as in healthy subjects, variability across LGG patients is greater for cortical than subcortical reorganization [17]. This implies that various dynamic processes of neural reconfiguration may be mobilized from one patient to another, such as peritumoral rearrangement, or recruitment of remote structures in the ipsilesional hemisphere and/or the contralateral side [21,22]. Inter-patient differences in such patterns of redistribution are strongly correlated with LGG characteristics, i.e., the volume of the tumor, the kinetics of the glioma (as plasticity is linked to the time course of the disease, with less compensation in more rapidly evolving lesions [134]), and to the severity of brain invasion, with less plastic potential in more diffuse tumors which migrate more widely along the white matter pathways [44,138]. These connectomal considerations play a pivotal role when selecting the optimal therapeutic attitude, knowing that better prediction of the individual processes of neural reconfiguration may be valuable for anticipating the next treatment(s), thereby avoiding missing the opportunity for action from an oncological point of view.