Achieving a cure for hypothalamic hamartomas: a Sisyphean quest?

Hypothalamic hamartomas (HHs), although rare, has raised much interest due to the uniqueness of their clinical symptoms, the mystery of the subcortical-cortical networks involved in their pathogenesis, and the unparalleled complexity posed by their surgical treatment as a lesion-related epilepsy. Paillas, Roger, and colleagues in 1969 were the first to describe the clinical, radiologic, and histologic presentation of “epileptic” HH1 and to resect the HH in order to treat the associated epilepsy.1 Since then, concepts of the electroclinical expression of epilepsy observed in association with HHs, and the available therapeutic approaches, have evolved rapidly and dramatically. The development of different surgical approaches, and of the concepts about the role of the HHs in relation to the epilepsy, in this typically drug-resistant condition,2 has gone through a series of cycles. In 1993, Cascino and colleagues reported evidence of seizure onset involving the temporal (and/or frontal) lobe leading to neocortical resection following the traditional electroclinical mapping scheme of the epileptogenic zone with surgical implantation of intracranial electrodes (without electrode placement into the HH lesion). This strategy led to repeated failures.3 In 1995, Claudio Munari, using stereo-encephalography (SEEG), recorded ictal discharges in the HH,4 thus demonstrating the role of the HH itself as an intrinsically epileptogenic lesion. Simultaneously in support of this hypothesis came the demonstration of hyperperfusion of the HH with ictal single-photon emission computed tomography (SPECT) by Kuzniecky and colleagues.5 The concept of an essentially subcortical epilepsy, with the epileptogenic zone (EZ) being centered on the HH itself, paved the way for lesion resection becoming a standard treatment for HHs that were associated with drug-resistant epilepsy. Such a lesional approach tended to require a relatively simple preoperative workup, notably without intracranial recording in the majority of cases. Small series of HH resection showed the possibility, in addition to the effect on epilepsy, of reversing the epileptic encephalopathy of HH6-8 within certain cases an objective neuropsychological improvement9 so long as the surgery was performed early enough in this catastrophic epileptic “syndrome.” A number of case reports or very small series of patients who had undergone resection were then highlighted as surgical success stories, emphasizing the technical difficulty of surgery with potentially rewarding outcomes, but with little mention of complications. This period with relative neglect of the risks of surgery in a complex anatomic area was abruptly interrupted by the publication in 2002 by Palmini and colleagues of a series of 13 patients, operated in international centers of excellence for epilepsy surgery. While confirming that resection can alleviate not only seizures but also behavioral and psychiatric comorbidities, this seminal paper also demonstrated the relatively high frequency of severe surgical complications with pterional and subfrontal approaches.10 The efficacy in terms of seizure cessation was also demonstrated to be more modest than expected.10 The first international symposium on HHs was organized by Fred Andermann at the Montreal Neurological Institute in 2001.11 The unveiling of the risks of classical resective surgery in this area prompted the development of various alternative techniques with lower surgical risks: transcallosal interforniceal (TAIF),12 endoscopy,13 disconnection,14 brachytherapy,15 radiofrequency thermocoagulation16 and radiosurgery.17 In the hands of expert teams, these techniques allowed neurosurgeons to dramatically reduce surgical morbidity and mortality.7 The trend at that time was to use Delalande's simple operative classification18 to choose appropriate surgical strategy: low approaches (such as pterional) for resection or disconnection of HH lesions located below the floor of the third ventricle in the cisterns, and high approaches such as TAIF,12 endoscopy,13 thermocoagulation,16, 19 brachytherapy15 or radiosurgery20 for those located above the floor. Low approaches usually left residual HH above the floor, within the third ventricle, and often resulted in incomplete seizure control requiring a second “high approach” to complete surgical management. In the late 2000s, surgical enthusiasm was tempered by two observations. First, a significant rate of neuropsychological and endocrinologic complications was reported following many of these new approaches.21 Endocrinologic complications included a high rate of transient diabetes insipidus (DI), but also a significant rate for permanent endocrine dysfunction including persistent DI, hypothyroidism, and growth hormone deficiency. A more worrying common problem was the observation of postoperative weight gain as a result of appetite stimulation, described in up to 23% after TAIF,22 59% after endoscopy, and 16% after brachytherapy.23 These complications were not observed after radiosurgery in a rigorous prospective trial.21 In terms of neuropsychology, a significant risk of permanent short-term memory impairment was also reported. This risk was up to 75% after TAIF,24 up to 58% after endoscopy25 and 20–50% after brachytherapy.26 Transient and permanent short-term memory deficits were noted in approximately, respectively, 58% and 8% of patients undergoing TAIF and 14% and 8% of patients undergoing endoscopic resection by Ng and Harvey.25, 27, 28 These complications were not observed after Gamma Knife radiosurgery29 during two independent prospective trials30-32 (Table 1). Because all HHs causing epilepsy are attached to one of the mammillary bodies,33-35 the risk of memory deficit is common to all the ablative techniques. The fact that no destructive effect is observed in most of the patients after radiosurgery36, 37 may explain why no memory deficits have been found (even with long-term follow-up) after radiosurgery despite the use of a battery of standardized neuropsychological tests exploring intelligence and memory (immediate and delayed verbal and visual memory).29 In the early 2010s, new stereotactic thermocoagulation methods using magnetic resonance (MR) guidance such as highly focused ultrasound (HIFU) and laser interstitial (LITE)38 thermocoagulation were introduced to the field of functional neurosurgery and tested for the treatment of HH. No series of HIFU for HH has been published to date. LITE appears as an especially appealing method39. However, after a few years of somewhat inflated expectations, based on enthusiastic preliminary reports with small cohorts, very high rates of short-term seizure control, and zero toxicity39, recent publications have reported lower rates of middle-term seizure cessation, in addition to significant morbidity and in particular a severe amnestic syndrome in one patient,40 leading to a more balanced estimation of the safety–efficacy profile for this technique. The absence of randomized controlled trials comparing safety and efficacy of the different surgical approaches, and the fact that patient selection is frequently different depending on the approach, are obstacles to precise comparison of different techniques (Table 1). Gamma Knife radiosurgery is the only method to date that has been evaluated via a true prospective trial, and which demonstrates much lower morbidity compared to the results of other techniques reported in the literature (Fig. 1).29 More precise understanding of the anatomic location and extent of the HH is turning out to be increasingly crucial. First, the large spectrum of clinical and radiologic variability of patient presentation20, 39 needs to be addressed, with more attention40 to better personalize the prediction of natural evolution and to better tailor the therapeutic strategy. Second, the precise location of the HH is suspected to influence the propagation and thus the topology of the epileptic network, particularly the relationship to the mammillary bodies32, 33 and posterior hypothalamus.41, 42 Third, stereotactic techniques such as LITE, radiosurgery, brachytherapy, or radiofrequency thermocoagulation require a more precise anatomic description.20, 31, 43, 44 In contrast to radiosurgery, one of the limits of stereotactic techniques requiring the use of a probe through the brain (in addition to the hemorrhagic risk) is the difficulty of completely treating the whole HH, depending on its anatomic situation. Thus mapping of the EZ in the HH is becoming a key issue for these techniques. Even more than the anatomic morphologic location of the HH, its specific hodology (study of connectivity) may turn out to be key information in the future (Fig. 2). Resting state mapping (functional connectivity by independent component analysis-–based markers) has been used by a team with this aim.45 According to these authors, noninvasive connectivity measures correspond to areas of initial ictal propagation and differentiate such areas from secondary ictal propagation, which may aid in planning of ictal focus surgical disconnection and support the use of this newer modality for adjunctive information in epilepsy surgery evaluation.45 The customization of surgical options to the individual patient should accordingly improve treatment results. The Marseille group has proposed the possibility of a novel approach to brain interventions based on personalized brain network models derived from noninvasive structural data of individual patients, with the aim of developing novel personalized therapeutic strategies. Indeed, modeling of the patient-specific organization of the EZ based on the individual patient connectome and the use of the “Virtual Brain” epileptic discharge propagation model is nowadays feasible for HH.46, 47 Given the generative nature of the model, the “Virtual Brain” may serve as a platform for systematic hypothesis testing and simulating the functional consequences of this therapeutic hypothesis (Fig. 2). This process may be a future strategy to select the most appropriate approach among several scenarios and to tailor it in the individual patient. Since Palmini's retrospective study, all recent surgical series report a rate of seizure cessation of around 60%. Differences between series are usually explained by differences in criterion of cure and/or duration of follow-up. In HH operated using ablative techniques, short-term follow-up series tend to overestimate the real mid- to long-term seizure-cessation rate. On the other hand, in HH patients operated by Gamma Knife radiosurgery, short-term evaluation leads rather to underestimation of the seizure-cessation rate in comparison to mid- and long-term results29 due to the delay in the efficacy of this technique. Thus whatever the surgical technique, a significant percentage of patients fail to achieve seizure freedom. Some authors have reported patient cases with a demonstrated EZ extending beyond the limits of the HH itself, with seizure cessation occurring after the resection of this cortical EZ.48 The concept of secondary epileptogenesis is gaining acceptance in HH, embracing the idea that distant areas can become epileptogenic over time. Consequently, the dogma that the “EZ is always confined to the HH itself” is no longer acceptable.49, 50 Some epileptic patients with HH may have an EZ extending outside the anatomic limits of the HH, and some others, although rare, may have an EZ that does not even involve the HH at all. There are rare cases where resecting or ablating an additional brain region may be useful. In HH patients presenting with unusual features such as absence or gelastic seizures in their past or present history, a more extensive preoperative clinical workup may be proposed before any attempt to operate on the HH itself. In patients with failure after surgical ablating or targeting of the totality of the HH, possible extension of the EZ outside the HH may be a hypothesis that should be to tested by further investigations before claiming that the epilepsy is not surgically remediable. Thus the pendulum is now swinging back toward a more balanced concept of the relationship between the HH and EZ. The present supplement of Epilepsia originates from presentations and discussions from the Second International Hypothalamic Hamartoma Symposium held in Marseille in September 2013 with the support of “Hope for Hypothalamic Hamartoma” patient association. We thank Dr. Aileen Mc Gonigal for her cautious reviewing and support of preparation of this editorial. We thank Assistance Publique Hôpitaux de Marseille for support. The first author has received support from Medtronic. Professor JH Cross has participated in clinical trials for GW Pharma, Zogenix, and Vitaflo. She has sat on Advisory Boards and given lectures for Eisai, UCB, Shire, and Nutricia. All remuneration has been paid to her department. The remaining authors have no conflicts of interest. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. Jean Régis is a head of Functional Neurosurgery and Radiosurgery in Timone Hospital, Aix-Marseille University. J. Helen Cross is The Prince of Wales’s Chair of Childhood Epilepsy at UCL-Great Ormond Street Institute of Child Health, London, United Kingdom. John F. Kerrigan is an associate professor of Neurology and Child Health, University of Arizona College of Medicine – Phoenix, Phoenix, AZ, U.S.A.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.