Abstract
Background/Aim: Meningeal melanocytomas are rare tumors of the central nervous system and optimal treatment needs further clarification. This study compared subtotal resection (STR), STR plus radiation therapy (RT), gross total resection (GTR), and GTR+RT to better define the role of postoperative RT. Patients and Methods: All cases reported in the literature were reviewed. Patients (n=184) with complete data were analyzed for local control (LC) and overall survival (OS). Results: On univariate analysis, GTR (vs. STR) was associated with improved LC (p=0.016). When comparing the treatment regimens, best and worst results were found after GTR+RT and STR alone, respectively (p<0.001). On univariate analysis, GTR resulted in better OS than STR (p=0.041). Moreover, the treatment regimen had a significant impact on OS (p=0.049). On multivariate analyses of LC and OS, extent of resection and treatment regimen were found to be significant factors. After STR, RT significantly improved LC but not OS. After GTR, RT did not significantly improve LC or OS. Conclusion: GTR was significantly superior to STR regarding LC and OS. STR+RT resulted in significantly better LC when compared to STR alone.
Meningeal melanocytomas were described in 1972 and represent a very rare tumor type of the central nervous system (CNS) (1). Macroscopically, these pigmented tumors appear well circumscribed or encapsulated. On light microscopy, mitotic figures, necrosis, and hemorrhage are rare or even not found (2-16). According to ultrastructural studies, they originate from leptomeningeal melanocytes (6, 10, 16-18). Melanosomes can be found at different levels of maturity with no or minor nuclear and cellular pleomorphisms. On computed tomography, meningeal melanocytomas present as well-defined, iso-dense, or slightly dense lesions with homogeneous enhancement of contrast medium, and on magnetic resonance imaging, the lesions are isointense when compared to the gray matter or show increased signal-intensity on T1-weighted images (6, 7, 19-24).
Meningeal melanocytomas are generally considered benign but have been reported to be associated with a high risk of recurring locally. Standard treatment includes resection, either alone or followed by radiation therapy (RT). Previous reports have suggested that RT can improve treatment outcomes, if only a subtotal resection (STR) has been achieved (25-27). The vast majority of studies in the literature are case reports. Only very few review articles using individual patient data and a meta-analysis exist that investigated the value of RT following STR (25-27). These review articles are about 20 years old and included less than 100 patients, and the meta-analysis did not exclusively use individual patient data. Larger studies with individual patient data are required to better define the role of RT following STR or gross total resection (GTR) of meningeal melanocytomas. The present study has been performed to contribute to the closure of this gap.
Patients and Methods
A literature research was performed in PubMed (Medline) using the terms “meningeal melanocytoma” and “meningeal melanocytomas”. A total of 227 articles were identified. An additional article was identified describing a melanocytic tumor with GNA11 p.Q209L mutation of the foramen magnum and the upper cervical cord (28). These articles were screened for complete data with respect to age, sex, tumor site, extent of resection, post-operative RT, local control (LC), and overall survival (OS). Further criteria for inclusion in the subsequent analyses were histology of meningeal melanocytoma, location in the brain or spine, age >10 years, minimum follow-up of 6 months in patients alive at the last contact. Moreover, patients with initially scattered or metastasized tumors and patients with orbital tumors only were not included. Of the 228 articles, 106 articles could not be used for the analyses of LC and OS, mainly due to insufficient follow-up (38 articles). In 29 articles did not provide sufficient (individual) patient data. In seven articles, the reported patients were children ≤10 years of age, in 14 articles patients had scattered or metastasized disease, in 10 articles a histology different from meningeal melanocytoma, and in six articles orbital involvement only. Moreover, in two articles patients received definitive RT without a resection.
After exclusion of these articles, 122 articles with a total of 184 patients remained for the subsequent analyses (2-25, 28-125). This cohort included one patient with an orbital lesion invading the cranial cavity (59). The subgroup of patients with spinal lesions included two patients with lesions of the S1-nerve root, one patient with a lesion arising from the right foramen magnum and the upper cervical spine, and one patient with a lesion of the terminal filum (13, 28, 80, 108). The characteristics of the 184 patients are shown in Table I. In 52 patients, the period of follow-up was longer than on the corresponding article due to direct contact with the authors. Of the 184 patients, 40 received STR alone, 38 STR + RT, 100 GTR alone, and 6 patients GTR + RT, respectively. Patient characteristics related to these four treatment groups are summarized in Table II. In addition, subgroup analyses were performed for patients receiving STR (n=78) and patients receiving GTR (n=106).
Patient characteristics and treatment regimens were evaluated for potential associations with LC and OS at 3, 5 and 10 years following resection. Characteristics included age (≤40 vs. >40 years, median=40 years), sex (female vs. male), year of publication (until 2003 vs. 2004-2023), tumor site (brain vs. spine), extent of resection (STR vs. GTR), post-operative RT (no vs. yes), and treatment regimen (STR alone vs. STR+RT vs. GTR alone vs. GTR+RT). In patients receiving post-operative RT, the impact of radiation dose, given as equivalent dose in 2 Gy fractions with an alpha/beta ratio of 10 Gy for tumor cell kill (EQD2, <45 Gy10 vs. >49.5 Gy10), on LC and OS was investigated (126, 127). Details of RT were not stated for two patients. Of the other 42 patients, 8 patients received single-fraction radiosurgery (n=6) or fractionated stereotactic RT (n=3). Twenty-nine patients received conventional local RT with 30 to 55 Gy, four patients RT of the whole-brain or the craniospinal axis (CSA) with 30 to 50 Gy, and one patient RT of the CSA with 48 Gy plus a boost of 6 Gy to the posterior fossa.
Univariate analyses regarding LC and OS were performed using the Kaplan-Meier method and the log-rank test. p-Values <0.05 were considered significant, and p-values <0.10 indicated a trend. Significant factors were additionally evaluated for independence in a multivariate analysis performed with the Cox proportional hazards model.
Results
Local control in the entire cohort. In the univariate analysis of the entire cohort, LC was significantly associated with the extent of resection (GTR with/without RT superior to STR with/without RT, p=0.016). A significant association was also found for the treatment regimen; the best results were observed after GTR+RT, and the worst outcomes were found after STR alone (p<0.001). The LC-rates after 3, 5 and 10 years related to patient characteristics and treatment regimens are summarized in Table III. In the multivariate analyses of LC, the extent of resection (HR=0.54; 95% CI=0.32-0.90; p=0.018) and the treatment regimen (HR=0.60; 95% CI=0.45-0.80; p<0.001) maintained significance. In 42 of the 44 patients receiving RT, the RT dose was stated. The LC-rates at 3, 5 and 10 years were 76%, 76% and 76%, respectively, after an EQD2 of >49.5 Gy10 (24 patients) compared to 76%, 57% and 57%, respectively, after an EQD2 of <45 Gy10 (18 patients) (p=0.25).
Overall survival in the entire cohort. On univariate analysis of OS, GTR (±RT) was superior to STR (±RT) (p=0.041). Moreover, the treatment regimen had a significant impact on OS (p=0.049). The best and worst OS rates were observed after GTR+RT and STR alone, respectively. The OS-rates after 3, 5 and 10 years related to patient characteristics and treatment regimens are shown in Table IV. In the multivariate analyses of OS, the extent of resection (HR=0.43; 95% CI=0.18-0.99; p=0.047) and the treatment regimen (HR=0.54; 95% CI=0.34-0.86; p=0.010) were significant. The OS-rates at 3, 5 and 10 years were 95%, 95% and 82%, respectively, after an EQD2 of >49.5 Gy10 compared to 92%, 82% and 82%, respectively, after <45 Gy10 (p=0.70).
Subgroup analysis of patients receiving subtotal resection. In the subgroup analysis of the 78 patients receiving STR with or without RT, post-operative RT was associated with improved LC (p=0.014). The LC-rates at 3, 5 and 10 years were 74%, 65% and 65%, respectively, with RT (38 patients) compared to 49%, 44% and 15%, respectively, without RT (40 patients) (p=0.014). In the corresponding multivariate analysis, the addition of RT was still significant (HR=0.40; 95% CI=0.19-0.84; p=0.016). In those patients receiving post-operative RT, the LC-rates at 3, 5 and 10 years were 80%, 80% and 80%, respectively, after >49.5 Gy10 (20 patients) compared to 73%, 49% and not available, respectively, after <45 Gy10 (16 patients) (p=0.12). The OS-rates at 3, 5 and 10 years were 89%, 84% and 75%, respectively, with RT compared to 81%, 68% and 68%, respectively, without RT (p=0.20). In those patients receiving post-operative RT, the OS-rates at 3, 5 and 10 years were 94%, 94% and 81%, respectively, after >49.5 Gy10 compared to 91%, 78% and not available, respectively, after <45 Gy10 (p=0.62).
Subgroup analysis of patients receiving gross total resection. In the subgroup analysis of the 106 patients receiving GTR, post-operative RT did not significantly improve LC or OS. LC-rates at 3, 5 and 10 years were 75%, 75% and 75%, respectively, with RT (6 patients) compared to 83%, 65% and 43%, respectively, without RT (100 patients) (p=0.52). The corresponding OS-rates were 100%, 100% and 100%, respectively, with RT compared to 98%, 93% and 73%, respectively, without RT (p=0.41). The EQD2 had no significant impact on LC (p=0.32) or OS (p=1.00). LC-rates at 3, 5 and 10 years were 100% after <45 Gy10. After >49.5 Gy10, only the LC-rate at 3 years was available, which was 50%. OS-rates at 3, 5 and 10 years were 100% in both dose groups.
Discussion
In 2004, a retrospective analysis of individual patient data of 89 cases reported in the literature suggested that GTR resulted in significantly better LC and OS than STR (26). Moreover, in patients receiving STR both outcomes were significantly improved with the addition of post-operative RT. Since this analysis is almost 20 years old and included less than 100 patients, we felt that it is about time to perform an additional analysis in a larger cohort of patients. The need for such an analysis is supported by the fact that another group presented a retrospective meta-analysis about one year ago (27). The conclusions of this meta-analysis emphasized the importance of GTR and the role of post-operative RT after STR and, therefore, confirmed the findings from 2004 (26, 27). However, the meta-analysis was not based exclusively on individual patient data and did not provide Kaplan-Meier analysis including 3-, 5- and 10-year rates of LC and OS (27). In order to generate these important data, the present study was performed. The corresponding search of the literature revealed 184 patients with complete data including an appropriate period of follow-up who qualified for the analyses regarding LC and OS (2-25, 28-125). According to these analyses, GTR with or without RT resulted in significantly better LC and OS (Table III and Table IV). Moreover, on multivariate analyses, GTR proved to be an independent predictor of both outcomes. These results confirm the findings of the previous analysis from 2004 (26). When comparing the four investigated treatment regimens, the best outcomes were achieved with GTR+RT, and the least favorable outcomes with STR alone (Table III and Table IV). The results after GTR alone and STR+RT were almost similar. Therefore, it appeared that the addition of RT to STR could compensate the inadequate resection. These results of the present study also confirm the findings of the analysis from 2004 (26). The fact that GTR+RT resulted in the best long-term outcomes with 10-year LC- and OS-rates of 75% and 100%, respectively, poses the question whether GTR should also be followed by RT. However, when interpreting these results, the number of only 6 patients receiving RT after GTR must be considered. We feel that this small number does not allow valid conclusions. Moreover, in the specific subgroup analysis of patients in whom GTR was achieved, the addition of RT did not lead to significantly better LC or OS. Therefore, post-operative RT cannot be recommended after GTR at this stage. In the specific subgroup analysis of patients receiving STR, post-operative RT resulted in significantly better LC and showed a trend towards better OS. Therefore, the addition of RT after STR appears justified. Another important question affects the appropriate radiation dose following STR. In the previous analysis, doses of doses of 45-55 Gy appeared superior to doses ≤40 Gy (26). In an additional analysis published in 2006 that included 49 patients with spinal meningeal melanocytomas, doses of 50-52.5 Gy appeared superior to 30-45 Gy (128). Higher doses resulted in significantly better 5-year LC (100% vs. 33%, p=0.042) and non-significantly better 5-year OS (100% vs. 77%, p=0.33) (128). In the present study, higher doses showed a trend towards improved LC (p=0.12) with 5-year rates of 50% vs. 49%, respectively. The 5-year OS-rates were 94% and 78%, respectively (p=0.62). In this study, the radiation dose was given as EQD2. We compared an EQD2 of >49.5 Gy10 to <45 Gy10, since no patient received an EQD2 between these dose levels. A dose of 50.4 Gy given in 28 fractions of 1.8 Gy represents an EQD2 of 49.56 Gy10. The EQD2 of 50.4 Gy given in 28 fractions for radiation toxicity to the CNS such as brain necrosis or myelopathy is 47.88 Gy2 (alpha/beta ratio of 2 Gy) and, thus, lower than for 50 Gy in 25 fractions of 2 Gy (50.00 Gy2) (126, 127, 129, 130). Since 50.4 Gy is associated with a comparably low risk of CNS toxicity, we would recommend this regimen, particularly because the vast majority of patients with meningeal melanocytomas have favorable survival prognoses. When following our recommendations, the limitations of the current study need to be considered. These include the retrospective nature of the data analyses, the fact that 106 of 228 articles could not be considered due to incomplete data, and the non-standardized follow-up procedures and periods in the different articles used for the analyses. The reporting of meningeal melanocytoma cases must continue in order to allow additional studies in larger patient cohorts, particularly in patients receiving post-operative RT.
In conclusion, GTR with or without RT was significantly superior to STR±RT. After STR, post-operative RT led to significant improvement of long-term LC and was associated with non-significantly better OS. For patients receiving STR, RT doses with an EQD2 of >49.5 Gy10 appeared superior to <45 Gy10. Given the limitations of this study, a dose of 50.4 Gy in 28 fractions (EQD2=49.56 Gy10) appears recommendable, since this regimen is associated with a comparably low risk of CNS toxicity (EQD2=47.88 Gy2).
Acknowledgements
O.Z. received a scholarship from the University of Lübeck within the framework of the emergency aid program for the support of refugee academics from Ukraine.
Footnotes
Authors’ Contributions
The study was designed by D.R., O.Z., J.G., and N.Y.Y.. The data were collected by D.R. and O.Z. and analyzed by D.R. and N.Y.Y.. The article was drafted by D.R. with the support of N.Y.Y.. The final version of the article was reviewed and approved by all Authors.
Conflicts of Interest
On behalf of the Authors, the corresponding Author states that there are no conflicts of interest related to this study.
- Received December 21, 2023.
- Revision received February 13, 2024.
- Accepted February 20, 2024.
- Copyright © 2024, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).