Portal de Periódicos da CAPES

Laser Ablation vs Open Resection for Deep-Seated Tumors

2016; Lippincott Williams & Wilkins; Volume: 63; Issue: Supplement 1 Linguagem: Inglês

10.1227/neu.0000000000001289

1524-4040

Danilo Silva, Mayur Sharma, Gene H. Barnett,

Brain Metastases and Treatment

Laser interstitial thermal therapy (LITT) is a minimally invasive treatment modality for brain tumors that was first introduced by Bown1 in 1983 and has been revisited and refined as a result of technological advancements since the end of the last century.2,3 The main limitations of this surgical technique (the inability to monitor or predict the extent of ablation, the inability to shape the ablation to conform to the tumor contour, and the lack of a cooling system) have been resolved. LITT is a US Food and Drug Administration–cleared treatment option that can be used to treat recurrent glioblastoma4 and is emerging as a surgical option for upfront treatment of selected patients with malignant glioma and as a treatment for brain metastatic diseases that have failed stereotactic radiosurgery (SRS).5-7 Thermal damage is the basic biological effect of LITT. Laser energy is absorbed by the surrounding tissue, causing excitation and release of thermal energy.8-11 Ultimately, a cascade of events results in cell breakdown and coagulative necrosis of the target tissue. As we move toward a more individualized approach in cancer care based on the biomolecular profile of different types of tumors, we as neurosurgeons should offer surgical options individualized for each group of patients on the basis of their clinical history, performance status, and tumor characteristics (size, location) and the complication profile of different surgical approaches. In a debate at the 2015 Congress of Neurological Surgeons Annual Meeting in New Orleans, Louisiana, we presented the case for laser ablation of deep (ie, >2 cm from the surface of the hemispheres) brain tumors. Besides an introductory background to LITT, the evidence supporting use of LITT in this setting, the inability to use cortical or subcortical mapping, and the potential limitation of tissue sampling for molecular analysis compared with craniotomy were addressed. BACKGROUND OF LITT FOR BRAIN TUMORS As described previously, the major biological effect of LITT is thermal damage.12 Laser photons are emitted and absorbed by surrounding tissue, causing excitation and release of thermal energy, which is transformed to heat and distributed to surrounding tissue through the process of convection and conduction.8-12 The degree of heat penetration into surrounding tissue is determined by the properties of the tissue itself, which determines the extent of thermal ablation.13 Studies have shown that water and hemoglobin content are the main determinants of laser absorption by tissues.14 In addition, the greatest degree of tissue penetration, up to 10 mm, is observed with laser radiation with wavelengths in the near-infrared part of the spectrum.14 The impact on viable tissue is not, however, just a matter of the temperature to which it has been heated; rather, it is the relationship between temperature and time that determines the biologic effect. The Arrhenius equation allows calculation of the probability of cell injury, the type of calculation that computers can do nearly instantaneously. The cascade of cellular events after LITT includes enzyme induction, denaturation of proteins, cellular membrane breakdown, coagulation necrosis, and blood vessel sclerosis.14 It is important to mention that rapid increases in temperature can result in tissue carbonization,15 which prevents adequate laser absorption. Also noteworthy is the fact that overheating will cause tissue vaporization, which could lead to increased intracranial pressure.16 The goal with LITT is to achieve coagulation necrosis of the target volume without provoking carbonization or vaporization of the treated area while preventing injury to surrounding healthy brain. The literature describes 3 zones of particular histological changes around the laser probe during LITT.15 The first zone is the area closest to the tip of the probe and represents the area of greatest tissue damage because of the highest degree of energy absorption.15 Coagulation necrosis occurs at temperatures in the range of 50°C to 100°C.15 Carbonization and vaporization are usually seen at temperatures >100°C.15 Tissue volume in the second or intermediate zone also undergoes thermal injury. Tissue cells located at the third and most marginal zone, although damaged by thermal energy, are still viable.15 True coagulation necrosis is observed in the first 2 zones.15 These 3 zones of thermal damage are displayed in the computer software in one of the commercial LITT systems (NeuroBlate; Monteris Medical, Inc, Plymouth, Minnesota) as the thermal-damage-threshold (TDT) lines through data acquired by magnetic resonance imaging (MRI) thermography. This unique feature allows the surgeon to tailor the ablation on the basis of the target volume, which should be included in the first 2 zones. As mentioned, LITT is also affected by the optical properties of the targeted tissue. Analysis of the optical properties of native and coagulated human brain tissue revealed that the deepest area of thermal coagulation and highest laser penetration were found in the wavelength range between 1000 and 1100 nm, which is in the near-infrared part of the spectrum.17 In addition to the optical wavelength, the depth of interstitial thermal damage and subsequent necrosis depends on the cooling conditions, power density, and exposure time.18,19 Within the range of the above-mentioned wavelength, it has been shown that the laser interaction with white matter and gray matter is different. Gray matter shows a high level of laser absorption; on the other hand, white matter displays the lowest level of laser penetration. Eggert and Blazek20 studied the optical properties of some brain tumors and showed that, within the near-infrared spectral range, although glioblastoma and meningioma exhibited the highest degree of laser absorption, low-grade glioma displayed optical properties similar to gray matter. Optimal laser ablation is achieved when a sharp border of thermal injury is observed at the brain-tumor interface, characterizing a selective procedure with preservation of the normal brain tissue surrounding the tumor. A true game changer in favor of LITT for brain tumors was the capability to visualize real-time temperature changes in deep-seated tumors.2,3 Magnetic resonance thermography has become the preferred temperature monitoring sequence in neuro-oncology. Its principle relies on the temperature-dependent water proton resonance frequency. Proton resonance frequency image mapping is based on the fact that protons are displaced more efficiently within the magnetic field in the form of free water molecules (H2O) than in the form of hydrogen-bonded water molecules. As thermal energy is delivered during LITT, temperature increases and the number of hydrogen bonds decreases, resulting in an increased number of free H2O molecules and a lower proton resonance frequency, which can be visualized with MRI thermometry software in real time.21,22 US FOOD AND DRUG ADMINISTRATION–CLEARED LITT SYSTEMS USED IN NEUROSURGERY There are currently 2 US Food and Drug Administration–cleared LITT systems for neurosurgery in the United States: the NeuroBlate System and the Visualase Thermal Therapy System (Medtronic Inc, Minneapolis, Minnesota; Table 1).TABLE 1: Commercially Available Laser Interstitial Thermal Therapy SystemsaThe NeuroBlate system uses a CO2 gas–cooled laser probe system.23 Two options of laser probes are available: side-firing and diffuse-tip probes. The first one is used for contoured ablation of irregularly shaped targets. The second one is designated to provide a more regular concentric ablation of the target volume. The system uses a solid-state Dornier diode laser (Dornier MedTech GmbH, Wessling, Germany) operating at the 1064-nm wavelength with a laser output of 30 W. The laser probe is inserted with the Varioguide (Brainlab AG, Feldkirchen, Germany) frameless stereotactic guidance system and the Monteris Mini-Bolt device, which provides rigid skull fixation and allows a direct interface with the laser probe. The system is connected to a computer workstation, which allows distant robotic control of the probe itself. MRI thermography provides real-time feedback on the extent of thermal ablation. The NeuroBlate software displays the extent of heat deposition as TDT lines. The first line, closest to the tip of the laser probe, is the white line, which delineates tissue exposed to 43°C for 60 minutes, representing the damage zone. The blue line corresponds to the intermediate boundary and surrounds the tissue volume exposed to 43°C for 10 minutes (Figure 1). The third line located at the periphery is the yellow line, which encompasses the target volume exposed to thermal energy of 43°C for 2 minutes. Tissue volume located outside the yellow TDT line corresponds to viable cells with no permanent damage. The volume of tissue inside the blue line is considered to experience severe and permanent thermal injury, and the tissue within the white line exhibits coagulation necrosis.23FIGURE 1: The NeuroBlate software thermal-damage-threshold lines.The Visualase Thermal Therapy System is also an MRI-guided laser ablation system, which uses a 15-W diode laser generator, operating at a wavelength of 980 nm. The system uses a disposable outer cooling catheter, which contains a fiberoptic applicator with a light-diffusing tip.24,25 Sterile saline circulating through the catheter provides cooling to the probe tip and surrounding tissues,24 which limits the duration of laser-on time to several minutes. Thermal energy is delivered in an ellipsoid fashion. Real-time thermal ablation data are provided by MRI thermography. Whereas in the NeuroBlate system the surgeon determines the extent of ablation on the basis of the TDT lines, this system uses safety points with predetermined temperature threshold based on the preoperative MRI. This feature allows the laser to automatically deactivate in case there is an increase in temperature beyond the predetermined threshold during the treatment. CLINICAL EVIDENCE Glioblastoma multiforme (GBM) is a highly aggressive primary brain tumor with 5-year overall survival and progression-free survival rates of 9.9% and 4.1%, respectively, after complete resection and combined chemoradiotherapy.26 The overall median survival is reported to be 14.6 months with combined chemoradiotherapy and 12.1 months with radiotherapy alone after feasible surgical resection or biopsy.27 The median survival drops to 6.2 months in patients with progressive disease.26 Surgical management of patients with progressive disease is controversial, with no significant increase in median survival; however, younger ( 60-70) have been reported to have beneficial effects after repeat surgery.28 There is cumulative evidence in the literature on the survival benefit of repeat surgery in carefully selected patients with recurrent GBM.29 However, repeat craniotomy for progressive GBM is associated with 18% to 22% risk of worsening neurological status in the postoperative period and increased overall perioperative complication rates.30,31 Additionally, the risks of neurological worsening and local regional complications increase after each craniotomy, with a maximum risk of neurological worsening between the first and second surgeries.32 Given this dismal prognosis with the paucity of effective therapies in patients with GBM, there is interest in the development of alternative treatment modalities. The minimally invasive nature of LITT has led to its exploration for a variety of neurosurgical conditions such as deep-seated gliomas,2,4,12,33 brain metastasis,5,34 epilepsy,35,36 and radiation necrosis.7,37,38 LITT in the era of commercial devices is relatively new, and there is a limited but growing and intriguing literature on its benefits in these disorders. LITT has been explored as a potential therapeutic option in patients with deep-seated high-grade gliomas.2,4,12,33 In 1990, Sugiyama et al39 first reported the beneficial effects of an LITT Nd-YAG laser system in 5 patients with deep-seated brain tumors. Later, Leonardi et al40 reported 30 thermal treatments using an Nd-YAG laser (1064 nm) in 24 patients with residual/recurrent brain tumors (7 World Health Organization [WHO] grade 2, 11 WHO grade 3 [9 astrocytomas and 2 oligoastrocytomas], and 6 WHO grade 4). The procedure was monitored with 0.2-T MRI, and they reported that tumor response to thermal treatment was not dependent on tumor grade. Tumor response rates and long-term outcomes were not reported in this study.40 In the modern era, a first-in-human phase 1 thermal dose-escalation study investigated the safety, toxicity, and efficacy of LITT in patients with recurrent GBM and was published in 2013.4 This prospective multicenter study was carried out from September 2008 to October 2009 and enrolled 11 patients (7 at Cleveland Clinic and 4 at University Hospitals Case Medical Center) with recurrent GBM. Of these 11 patients, 10 (7 at Cleveland Clinic and 3 at University Hospitals Case Medical Center) underwent LITT during the study period. The primary end point of the study was the feasibility and safety of the NeuroBlate system. Secondary end points were overall survival, progression-free survival, change in KPS score, and change in treated and untreated tumor volumes.4 Severe toxicity was defined as a decrease of ≥20 points on the KPS score within 14 days after the procedure, and the maximum therapeutic dose was estimated.4 On the basis of animal models, the 3 aforementioned TDT lines were chosen for the study. Adult patients with recurrent GBM after standard treatment for primary GBM, good performance status (KPS score ≥60), tumor with a 15- to 40-mm cross-sectional dimension in the supratentorial compartment, open surgery not indicated within 30 days, stable medical comorbidities, and no concurrent antitumor therapies were included in this study. Patients were followed up for a minimum of 6 months or until death, whichever was earlier. All 10 patients died of disease progression with a median follow-up of 8 months. Three patients with recurrent GBM were initially enrolled for LITT therapy, and therapy was applied until the yellow thermal line (43°C for 2 minutes) was reached at the tumor margin. These patients were then followed up over a period of 14 days, and toxicity was assessed by an independent committee. If toxicity was observed in 2 of 3 patients, either the thermal dose was modified or the trial was halted. If toxicity was not observed, another 3 patients each were subsequently enrolled for blue line and white line trial, following the same protocol. Ten patients were treated in this study with mean tumor volume of 6.8 ± 5 cm3, and 3 tumors were located in the eloquent brain regions. The mean tumor volume that was treated was 5 ± 3.2 cm3, accounting for 78% of total tumor volume. Tumor necrosis was evident on T1-weighted contrast-enhanced MRI at 24 and 48 hours after treatment within the white and blue treatment lines in all patients (Figure 2). The procedure time for the laser application was 2 to 8 minutes per slice. The procedure was well tolerated in all patients, with a median hospital stay of 3 days. One entry site infection was noted at 147 days after procedure, which cannot be attributed to the procedure. Postprocedural edema was present on MRI (48 hours) in all patients and was effectively managed with steroids. In terms of toxicity, 1 patient developed dysphasia with upper-limb weakness, and another patient developed homonymous hemianopia with weakness contralateral to the treatment; both patients improved with steroids. One patient developed intracerebral hemorrhage 6 weeks after the procedure as a result of rupture of a pseudoaneurysm and was managed by endovascular coil placement. Another patient developed significant hemiparesis immediately after the procedure, which was improved with physiotherapy in 6 to 8 weeks. Adverse events included deep vein thrombosis in 3 patients, pulmonary embolism in 1 patient, and grade 3 neutropenia in 1 patient. One patient (primary diagnosis of gliosarcoma) developed seeding along the biopsy (skull and epicranial) tract 9 months after the procedure.4 The median survival in this study was 316 days, and clinical improvement was seen at the high and intermediate thermal doses. Only 2 deaths were reported during the 6-month follow-up as a result of disease progression. The median progression-free survival was estimated to be ≥30% at 6 months, substantially better than previously reported in the literature. The authors concluded that NeuroBlate is a novel, minimally invasive, safe, and effective therapeutic option in appropriately selected patients with deep-seated recurrent GBM. Integration of diffusion tensor imaging (DTI) tractography and vascular studies to the NeuroBlate system improved the safety profile of the LITT therapy in selected cases.4FIGURE 2: Preoperative (top) and 24-hour-postoperative (bottom) T1 contrast-enhanced magnetic resonance imaging showing a deep corpus callosum glioblastoma multiforme. Note the decreased enhancement of the treated lesion.A multicenter retrospective study reported the efficacy of LITT in difficult-to-access high-grade glioma in 34 patients (24 with GBM and 10 with anaplastic glioma) over a period of 1.5 years.41 Adult patients with pathological diagnosis of high-grade glioma (GBM or anaplastic astrocytoma) were included. In patients with recurrent high-grade glioma, the median time from initial diagnosis was 29 months. The median follow-up was 7.2 months, and patients underwent standard postoperative adjunct treatment with serial MRI follow-up every 3 months. The primary end point of the study was progression-free survival, and secondary end points were overall survival and complications. The median age was 56 years, and 62% of patients were male. Of 34 patients, 16 patients (16 procedures) were newly diagnosed as having high-grade glioma and underwent upfront LITT therapy. Another 18 patients (19 procedures) underwent LITT therapy for recurrent tumors. The majority of these tumors were located in the frontal lobe (n = 15), followed by the thalamic (n = 7), parietal (n = 5), temporal (n = 5), insular (n = 2) and corpus callosum (n = 1) lobes. Median tumor volume and maximum tumor diameter were 10.13 cm3 and 3 cm, respectively. The median tumor volume covered by yellow and blue TDT lines was 98% and 91%, respectively. The median hospital stay was 3 days, and estimated blood loss was 30 cm3 in this study. Seventy-one percent of patients progressed during the follow-up, with median progression-free survival of 5.1 months. Of patients who had progression during the follow up, 52% had recurrence at the periphery of the treatment field, 22% had local progression in the center of treatment field, 22% had recurrence outside the primary enhancing tumor volume, and another 4% had recurrence in the contralateral hemisphere.41 Thirty-five percent of patients (n = 12) died during the follow-up as a result of disease progression, and the 1-year estimated survival was 68 ± 9%. The authors stratified their study sample into favorable and unfavorable groups based on the tumor coverage (extent of ablation) by the yellow and blue TDT lines. The median progression-free survival was 9.7 and 4.6 months in patients with favorable (<0.05-cm3 tumor volume missed by yellow TDT lines and 3 months) postoperative motor deficits (7.4% vs 2.1%, P = .04).53 Because LITT is usually performed under general anesthesia, subcortical mapping is not an option during the procedure for perirolandic deep-seated tumors. However, DTI deterministic fiber tracking provides an opportunity to avoid critical white matter tracts during the LITT procedure. Given that LITT is a minimally invasive procedure and that brain shift is not observed, DTI tracks as displayed on the LITT software are spatially accurate. In patients with permanent motor deficits after LITT, the overlaps between the white, blue, and yellow TDT lines and corresponding motor tracts on DTI were ≥ 2.2, ≥ 4.6, and ≥ 5.9 mm2, respectively (unpublished data). Advantages of LITT include minimal blood loss (<10 cm3), superior pain control (0/10 on Visual Analog Scale at 24 hours), shorter hospital stay (<24 hours), and shorter operative time (<2 hours). Given these advantages, LITT can also be considered for patients with surgically accessible glioma to enhance patient care and comfort. Regarding cortical mapping for superficial lesions (which is beyond the scope of this debate), it is generally agreed that functional MRI, based on blood oxygenation level dependent, is not sufficiently spatially accurate to tailor resection or ablation in functionally eloquent areas.54 Technologies such as transcutaneous magnetic stimulation,55 magnetoencephalography,56,57 or, perhaps the most promising, resting-state functional MRI58 may prove to allow accurate, safe ablation in these areas. MOLECULAR ANALYSIS Although craniotomy for deep tumors may provide tissue from multiple areas so as to be able to interrogate the molecular characteristic of the potentially heterogeneous tumor from several sites, this is rarely done outside of clinical investigations because of the high cost of performing these tests, which would be double or tripled for sampling 2 or 3 sites, respectively. Although laser ablation itself does not yield tissue for such analysis, it is often combined with stereotactic brain biopsy and typically yields 10 mm3 of tissue volume from each sample. Commercial analysis of hundreds of genes and introns currently requires only 1 mm3 of tissue, only a portion of which needs to be a viable tumor. Because larger tumors often dictate use of ≥2 laser probe trajectories, biopsies from several sites can easily be performed to extend the scope of sampling, should this be found to be clinically relevant in the future. CONCLUSION Because the modern era of laser ablation for deep brain tumor spans only several years, there is a comparative paucity of scientific evidence of its efficacy compared with that of modern craniotomy, which has existed for about a century. Nonetheless, the evidence that is emerging for LITT is quite promising, with results suggesting that the extent of ablation correlates nicely with the extent of resection for high-grade glioma. For both, near-complete ablation or resection appears to be essential for durable tumor control. A common misconception about laser ablation is that it cannot be used for cystic lesions, lesions near major arteries, or those abutting the ventricular system; examples were presented in the debate of successful ablation in each of these categories. Disruptive innovations in neurosurgery such as the operating microscope, minimal-access craniotomy with image guidance, SRS, endovascular therapy, and endoscopic skull base surgery met with considerable resistance during the introductory period and were either dismissed or ignored by the neurosurgical establishment. However, these advances gradually gained widespread acceptance and ultimately became the standard of care. We believe that laser ablation or LITT is the next "big thing" in brain tumor management and will add to the armamentarium of neurosurgeons in managing such patients with deep-seated tumors or tumors that are surgically difficult to access. Disclosures Dr Barnett is a paid consultant for and has stock options with Monteris Medical. The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.