Sodium-uorescein Guided Resection in Brain Metastases Surgery: Our Results in a Series of 59 Patients

Metastasis are the most common tumors in the brain, nowadays surgical resection still remains the treatment of choice in patients eligible for surgery. Several tools and techniques have been proposed to improve the resection rate but few datas are available on the application of uorescence guided surgery for brain metastases resection Purpose: the aim of the study was to assess the safety and ecacy of the uorescence-guided surgery through the application of sodium uorescein (SF) detected by Yellow 560nm lter for brain metastases surgery Methods: 75 Patients operated from January 2015 to January 2020 were retrospectively analyzed and 59 patients recruited. The patients were divided into group A of 30 (uorescein-guided surgery group of 30 patients) and group B (control- standard white light microsurgery of 29 patients). In group A a 5 mg/kg bodyweight of FL was intravenously injected at scalp incision. A yellow 560 lter was employed for microsurgical tumor resection. Preoperative and postoperative metastases volume analysis was performed to evaluate the extent of resection and the residual tumor rate Results: Median preoperative Volume (cm3) was 11.85 (range 2.5-90) in group A and 10 (range 1-48) in group B. Median residual tumor volume (RTV) after surgery was 0 (0-1.0) in group A and 0.4 (0-2.0) in group B whereas median EOR was 100% (0,8-1) in the group A and 98,2% (0,43-1) in group B. Conclusion: SF-guided surgery using YELLOW 560 nm microscope lter should be considered a safe and useful procedure and effective to improve the extent of resection even in metastatic tumor surgery.


Introduction
Metastasis are the most common tumor affecting the brain: lung cancer, breast cancer and melanoma, in particular, are the most common solid tumors spreading to the central nervous system (CNS).
In previous published studies it has been estimated that 9-17% of cancer patients will develop brain metastases during the course of their disease and that approximately half will die within 3-27 months from the initial diagnosis of CNS involvement [1][2][3].
Individualized therapies and new understandings of the molecular mechanisms of brain metastasis are leading to a progressive increase of the survival but tumor removal still plays a fundamental role in patients eligible for surgery.
In the last decades several tools and techniques have been proposed to improve the extent of resection in surgery for malignant tumors as the uorescent dye 5-aminolevulinic acid (5-ALA), rst described by Stummer et al in 2006 and then by other Authors for high grade gliomas [4][5][6]; however its application in brain metastasis surgery had been proved to be ineffective [7][8][9][10][11].
SF is a salt and an organic uorescent dye with peak excitation at 490 nm and emission between 500 and 550nm widely used in several elds of medicine: after endovenous administration, SF accumulates in the tumor tissue through the dysfunctional blood-brain barrier (BBB) demarcating the lesion [12].
The use of this dye is known in neurosurgery since 1947 when G.E.Moore et al reported their rst experience in a series of 46 patients [13].
Herein we report our experience in uorescence-guided surgery through the application of sodium uorescein detected by Yellow 560nm (PENTERO 900, Carl Zeiss, Meditec, Oberkochen, Germany) in a brain metastasis surgical series of 30 patients. These cases have been compared with a control group of 29 patients operated with bright-light surgery and our results integrated with current literature.

Methods
We report a retrospective study involving 59 patients affected with brain metastasis (BMs) surgically treated at Neurosurgery Department at"M.Bufalini" Hospital, Cesena Italy.
Patients operated on from January 2015 to January 2020 were retrospectively recruited.
Inclusion criteria were: age 18 years and over, singular or multiple CMs and life expectancy of more than 3 months, Karnofsky performance scale (KPS) 60 or more and adequate pre and post operative neuroimaging.
All patients underwent pre and postoperative (within the rst 48hours) CT scan and/or MRI (1,5T Siemens) including 3D volumetric sequences with and without contrast.
Demographic and clinical characteristics were recorded for all patients, including age, gender, histology , post-operative tumor resection rate EOR and uorescein related complications (Tab 1 and 2).
Informed consent was obtained from patients for the off-label use of FL.
FL (5 mg/kg body weight) was injected intravenously in group A at scalp incision; after the administration , the tumor staining was visualized with the use of a speci c microscope lter (YELLOW 560-nm lter, Pentero 900, ZEISS Meditec, Germany).
Surgical resection was performed using both white light and yellow-560 nm lter for the identi cation of uorescence enhancing tumor and to evaluate the border with normal brain parenchima (Fig.1).
In group B, the standard white light microsurgical procedure was carried out for resection of tumors.
In both groups, when required, adjunctive tools such as neuronavigation and intraoperative monitoring were adopted.
Surgeon's opinion and satisfaction after surgery (expressed as useful or not useful), were recorded; the extent of resection was evaluated by calculating the pre-operative tumor volume (PTV) on MRI scan and the residual tumor volume (RTV) in the immediate post-operative.
The volume segmentation was performed by using a quantitative, semiautomated software tool (Myrian, Intrasense, g 2), and the analysis was performed independently by two authors (GM, LM) blinded to each other's results.
The MRI images were obtained by a 1.5 T Philips scanner, including 3D pre-and post-enhancement T1 scans and T2 FLAIR.
The extent of resection (EOR) was calculated by using the following formula: (PTV -RTV)/PTV. All volume values were expressed in cm3.
We performed statistical analysis (IBM SPSS statistics 25) comparing PTV lesions and the RTV in the two groups with Pearson correlation coe cient, and p values were reported for each variable. A Fisher exact test was adopted to correlate the EOR with the histology of the metastases by dicothomically grouping FLG and BL based on the presence of a post-operative residual tumor (cut-off: residual volume less than 0.5 cm3).
We also performed an overall survival analysis by calculating the period from the initial surgery to the time of death of the patients by using the Kaplan-Meier method and the signi cance in difference among survival curves (Fig 3).

Results
We collected data of 30 patients (9 M 21 F) who underwent MS resection via Fluorescence-guided surgery (FGS); this group was compared with a cohort of 29 MS patients (13M 16 F) who underwent the traditional white light surgery (BL).
The median age at surgery was 63.5 years in the FGS group and 61.0 years in BL group.
MSs were located in the frontal lobe in 20 cases, in the temporal lobe in 11 cases, in the parietal lobe in 11 cases, in the occipital lobe in 5 cases, in the cerebellum in 12 cases.
The most common MSs origins were: lung metastasis (8 patients in FGS group vs. 7 patients in BL group), breast metastasis (9 patients in FGS group vs. 5 patients in BL group), and melanoma metastasis (2 patients vs in FGS group vs 6 patients in BL group). The groups are shown in Table 1.
Fluorescein-related complications were not reported.
Surgeons evaluated the uorescein application in MSs resections as useful in almost all cases.

Volume Correlations and Survival Curves
We perform Pearson correlation analysis to evaluate the degree to which the above-mentioned variables are interdependent. For instance, EOR is clearly inversely related to RTV, but both also depend on the PTV. The only strong statistical correlations we found were between RTV and EOR in both FLG (r = -0.81) and BL (r = -0.75) groups (p < 0.0001). The remaining volumes were not statistically signi cant. The survival analysis through Kaplan-Meyer curve was calculated in 15 months and did not show signi cant difference (Fig. 1). The Fisher test did not show signi cant results in the three group metastases in which we performed the analysis (Lung p = 0.16, Breast p = 0.37, and Melanoma p = 0.38).

Systematic Review Of The Literature
Systematic review of the Literature A review of the current Literature, according with the PRISMA guidelines, regarding the application of uorescein for the resection of brain metastasis was performed [25].
Using Pubmed MeSH database, Cochrane and Embase all English papers published until May 2020, including words ''metastasis'', " uorescein sodium", "resection", " uorescence guided resection" "YELLOW 560 NM FILTER" were reviewed. Each article was carefully scrutinized in order to select those in which a detailed description of the Fluorescein sodium dosage/use and intraoperative uorescein related ndings were reported. A double cross-reference check of papers considered for eligibility (forward search) was performed, in order to include supplementary papers erroneously undetected in the rst search round. From the rst literature research, we retrieved 21 articles. After duplicates removal and titles/abstract screening, 16 papers were considered for eligibility. Six papers were then excluded with the following reasons: case reports and opinion articles. Finally, 10 studies (290 patients) matched the inclusion criteria, and were included in the present review [15][16][17][18][19][20][21][22][23][24].
The literature review comprises 290 patients diagnosed of brain metastasis and surgically treated through the application of uorescein sodium.
In just one article [15] the authors reported a preoperative evaluation of MSs dimensions through the diameter.; the extent of resection was de ned as GTR (no residual tissue) or STR (with residual tumor tissue). The rst was achieved in 192 /222 reported cases (86.4%), while STR in 30 cases (15.6%).
No uorescein related complications were reported.
Six authors reported a subjective satisfaction rate of 100% in the uorescein use for the resection on MSs, while 3 authors (all with a satisfaction rate > 90%) reported a insu cient uorescein staining in 9 cases (5 lung, 2 melanomas ,1 renal and 1 unspeci ed); 1 Author didn't report this data.
Fluorescein dosage and time of injection are quite different in almost every study.
All of the Authors concluded that uorescein-guided surgery is a safe method that can lead to higher proportion of resection compared to common microsurgery.

Discussion
Metastasis are the most common brain tumor and nearly 20% of patients with cancer will develop brain metastases through the spreading of tumoral cells in cerebral microcirculation.
Although sistemic therapies are limited in e cacy, these tumors must be treated in a multimodal way and surgical removal is indicated in patients with singular lesion.
Even if metastases are not primary cerebral tumors, the boundary between the lesion and normal brain parenchima is not always clear even though the use of some tools as neuronavigation and intraoperative ultrasound have improved the post-operative results.
Complete surgical resection is advocated; in fact, some Authors demonstrated the correlation between the extent of resection and recurrence rate, progression-free survival, overall survival and quality of life [7,24,26,27].
The use of uorescein sodium salt was successfully applied in glioma surgery [28]; uorescein accumulates in pathological tissues through the damaged blood-brain barrier enhancing the lesion and making it easier to remove.
The same technique can be applied in metastases surgery; some Authors [15,17,18] showed that this tool leads to higher proportion of resection compared to white light surgery, minimizing normal tissue manipulation; nevertheless, none of these studies reported a volumetric analisys and the extent of resection was indicated as gross total or subtotal removal (absence or presence of residual tumor).
Our data con rmed that the i.v use of uorescein in metastases surgery is safe with no adverse effect in our group of patients.
We evaluated the in uence of FGS on resection extension in two groups of patients with CMs and we found that residual tumor tissue median is zero in FGS group with 100% of median EOR comparing with the 98.2% in BL-surgery group. This volumetric datas strongly support the usefulness of this method in increasing the extent of resection as already mentioned by some Authors.
We reported a good uorescein staining in almost 90% of patients as reported also by Schebesh et al. [18,19] ; further on, in our analysis we performed Fisher exact test to determine variations in tumor staining in different hystologies. We did not found signi cant statistical correlations, but in accord to surgeons' impression, we found a possible bene t by using uorescence with lung metastases (p = 0. 16). This data, which should be carefully con rmed in further statistical studies, could identify some histological subtypes that could take advantage more than others from this dye.
Moreover, previous studies reported a great heterogeneity regarding the time of uorescein injection and the dose to be administered; at this time it's not possible to determine objectively which is the best protocol to applicate to obtain the best results in term of major de nition of normal brain parenchima/tumor interface and in term of adverse effect.
Preliminary data on overall survival are evaluated by assessing KM method: till now we didn't nd any statistically difference among survival curves. However, due to the short follow up time in our cohort we can't draw any conclusion about long-term outcome. The limited number of patients and other parameters that affected survival may explain the lack of statistical signi cance for median survival time.

Conclusions
As far as we know this is the rst study to report objective datas on the extent of resection of metastatic tumors through the calculation of the residual tumor volume in two CM surgery groups (FGS vs BL).
As being concluded in previous publications this study con rmed that SF-guided surgery with the use of YELLOW 560 nm microscope lter is safe, effective and useful to improve extent of resection even in metastatic tumor surgery.
However, to con rm the e cacy and to prove the impact on patient survival and quality of life and the better speci city of this dye on some hystological type of tumors, prospective randomized studies with larger patient samples and with long-term follow-up need to be done.