Patterns of recurrence after stereotactic radiosurgery for treatment of meningiomas

Free access

Object

The purpose of this study was to evaluate patterns of failure after stereotactic radiosurgery (SRS) for meningiomas and factors that may influence these outcomes.

Methods

Based on a retrospective chart review, 279 patients were treated with SRS for meningiomas between January 1999 and March 2011 at Wake Forest Baptist Health. Disease progression was determined using serial imaging, with a minimum follow-up of 6 months (median 34.2 months).

Results

The median margin dose was 12.0 Gy (range 8.8–20 Gy). Local control rates for WHO Grade I tumors were 96.6%, 84.4%, and 75.7% at 1, 3, and 5 years, respectively. WHO Grade II and III tumors had local control rates of 72.3%, 57.7%, and 52.9% at 1, 3, and 5 years, respectively. Tumors without pathological grading had local control rates of 98.7%, 97.6%, and 94.2% at 1, 3, and 5 years, respectively. Of the local recurrences, 63.1% were classified as marginal (within 2 cm of treatment field). The 1-, 3-, and 5-year rates of distant failure were 6.5%, 10.3%, and 16.6%, respectively, for Grade I tumors and 11.4%, 17.2%, and 22.4%, respectively, for Grade II/III tumors. Tumors without pathological grading had distant failure rates of 0.7%, 3.2%, and 6.5% at 1, 3, and 5 years, respectively. Wilcoxon analysis revealed that multifocal disease (p < 0.001) and high-grade histology (WHO Grade II or III; p < 0.001) were significant predictors of local recurrence. Additionally, male sex was a significant predictor of distant recurrence (p = 0.04). Multivariate analysis also showed that doses greater than or equal to 12 Gy were associated with improved local control (p = 0.015).

Conclusions

In this patient series, 12 Gy was the minimum sufficient margin dose for the treatment of meningiomas. Male sex is a risk factor for distant failure, whereas high-grade histology and multifocal disease are risk factors for local failure.

Abbreviations used in this paper:HR = hazard ratio; NF2 = neurofibromatosis Type 2; SRS = stereotactic radiosurgery.

Object

The purpose of this study was to evaluate patterns of failure after stereotactic radiosurgery (SRS) for meningiomas and factors that may influence these outcomes.

Methods

Based on a retrospective chart review, 279 patients were treated with SRS for meningiomas between January 1999 and March 2011 at Wake Forest Baptist Health. Disease progression was determined using serial imaging, with a minimum follow-up of 6 months (median 34.2 months).

Results

The median margin dose was 12.0 Gy (range 8.8–20 Gy). Local control rates for WHO Grade I tumors were 96.6%, 84.4%, and 75.7% at 1, 3, and 5 years, respectively. WHO Grade II and III tumors had local control rates of 72.3%, 57.7%, and 52.9% at 1, 3, and 5 years, respectively. Tumors without pathological grading had local control rates of 98.7%, 97.6%, and 94.2% at 1, 3, and 5 years, respectively. Of the local recurrences, 63.1% were classified as marginal (within 2 cm of treatment field). The 1-, 3-, and 5-year rates of distant failure were 6.5%, 10.3%, and 16.6%, respectively, for Grade I tumors and 11.4%, 17.2%, and 22.4%, respectively, for Grade II/III tumors. Tumors without pathological grading had distant failure rates of 0.7%, 3.2%, and 6.5% at 1, 3, and 5 years, respectively. Wilcoxon analysis revealed that multifocal disease (p < 0.001) and high-grade histology (WHO Grade II or III; p < 0.001) were significant predictors of local recurrence. Additionally, male sex was a significant predictor of distant recurrence (p = 0.04). Multivariate analysis also showed that doses greater than or equal to 12 Gy were associated with improved local control (p = 0.015).

Conclusions

In this patient series, 12 Gy was the minimum sufficient margin dose for the treatment of meningiomas. Male sex is a risk factor for distant failure, whereas high-grade histology and multifocal disease are risk factors for local failure.

Meningiomas are the most common benign intracranial neoplasm, representing approximately 30% of all primary intracranial tumors.3 In most cases, they are benign with indolent growth patterns. While surgery has historically been the mainstay of meningioma treatment, patients are increasingly being treated definitively with radiosurgery. The majority of studies report long-term local control exceeding 85% for both linear accelerator-based and Gamma Knife radiosurgery for benign meningiomas.8–10,15,17

In an early report on long-term outcomes of stereotactic radiosurgery (SRS) for meningiomas, Kondziolka et al. showed a long-term control rate of 85% utilizing a median marginal dose of 16 Gy.11 Since this report, we have questioned whether dose de-escalation is possible without compromising treatment outcomes. Other reports have shown similar local control rates with doses of 12 Gy.7,10 Small tumor volume has also been shown to predict success with radiosurgery. DiBiase and colleagues reported 5-year disease-free survival rates of 91.9% for patients with tumors ≤ 10 cm3 as compared with 68% for larger tumors.6 Kondziolka et al. similarly cited a decreased local control rate with larger tumors.11

Local control has been the primary focus of long-term outcomes studies, but disease progression may manifest as local, marginal, or distant failure. We have recently reported that factors predicting local and distant treatment failure may differ for patients with higher grade meningiomas.1 Furthermore Kuhn et al. have recently reported that distant failure is a particularly common phenomenon in radiation-induced meningiomas, likely reflecting different tumor biology and a field cancerization effect.12 In this study, we present a large single-institution retrospective review of SRS in the treatment of meningiomas. We sought to evaluate patterns of treatment failure and to determine factors that might impact clinical outcomes.

Methods

Patient Population

This study, a retrospective chart review, was approved by the Wake Forest Institutional Review Board. A total of 271 patients received SRS for a meningioma between January 1, 1999, and March 1, 2011, at Wake Forest Baptist Health. Thirty-four patients had less than 6 months of follow-up and 6 patients were lost to follow-up due to death. The cause of death was progression of malignant melanoma in 4 patients lost to follow-up due to death, while the causes were nonneurological in the 2 others. All patients with less than 6 months of follow-up were excluded from statistical analyses. A total of 244 patients who underwent 282 sessions of SRS were initially included in this study. There was incomplete data for 3 patients, leaving a total of 279 observations. A summary of patient characteristics is listed in Table 1.

TABLE 1:

Patient characteristics*

VariableValue
no. of patients279
median age in yrs (range)60 (6–87)
sex
 males77
 females202
race (%)
 white250 (90)
 black26 (9)
 other3 (0.01)
median follow-up in mos (range)34.2 (6–141.4)
WHO Grade (%)
 unknown170 (61)
 I60 (21)
 II41 (15)
 III8 (3)
location
 multiple55
 convexity40
 parasagittal35
 cerebellopontine angle26
 cavernous sinus27
 falcine26
 tentorium24
 petroclival15
 sphenoid10
 posterior fossa9
 olfactory groove7
 optic nerve sheath2
 intraventricular2
 pineal1
median tumor volume in cm3 (range)3.25 (0.0367–414.7)
prior treatment
 prior surgery119
 prior EBRT35
 prior SRS37
recurrent disease118

EBRT = external beam radiation therapy.

Tumor volume data available for only 194 patients.

Radiosurgery Technique

After evaluation by a neurosurgeon and radiation oncologist, informed consent for SRS was obtained. Placement of a 4-pin Leksell stereotactic head frame was performed using a local anesthetic. After placement of the head frame, the patient underwent high-resolution contrast-enhanced stereotactic MRI or CT. Radiosurgery treatment plans were generated using the Leksell GammaPlan treatment planning system (Elekta AB) and executed using a Leksell Model B or C or Perfexion Gamma Knife unit (Elekta AB). Median prescription dose prescribed to the margin of the tumor was 12.0 Gy (range 8.8–20 Gy).

Follow-Up

Patients were followed clinically and with routine MRI. Patients generally underwent a posttreatment MRI 6 months after SRS and then yearly thereafter. The median duration of follow-up was 34.2 months. Date of death was queried using the Social Security Death Index; cause of death was determined based on clinical notes. Local failure was defined as a tumor recurrence either within the radiosurgical prescription volume (central failure) or just outside the radiosurgical prescription margin to within 2 cm of the tumor margin on MRI (marginal failure). Marginal failure was thus defined because traditional external beam radiation therapy uses a 2-cm concentric expansion of the gross tumor volume to determine the clinical target volume, and thus the volume within 2 cm would have been included in a non–SRS field. This definition of marginal failure is consistent with prior definitions in the literature.1 Distant failure was defined as tumor recurrence greater than 2 cm outside the tumor margin on MRI. Treatment response was defined as a decrease in tumor or enhancement volume. Treatment failure was defined as a minimum of two sets of images demonstrating progression or the identification of a new tumor distant from the treatment field.

Statistical Analysis

Kaplan-Meier analysis was performed to determine actuarial local control, distant brain control, progression-free survival, and overall survival in our patient population. Progression-free survival was defined as freedom from any failure (local or distant). Tumors were categorized into 3 groups: Grade I, Grades II and III, and tumors without pathological grading. Multivariate analyses using Cox proportional hazards models were performed to determine the relative importance of factors that independently predicted endpoints of overall survival, local control, and distant brain failure. Factors considered in univariate and multivariate analysis were age at time of treatment (specified in continuous terms), sex, race (white or black), WHO tumor grade (unknown, I, or II/III), tumor recurrence (yes/no), multifocal tumor (yes/no), neurofibromatosis Type 2 (NF2; yes/no), skull base location (yes/no), radiation-induced tumor (yes/no), largest tumor volume (according to the quartiles < 1.77 cm3, 1.77–4.08 cm3, 4.09–7.09 cm3, > 7.09 cm3, or missing; quintiles, deciles, and median also considered), and SRS dose (≥ 11, ≥ 12, ≥ 13, ≥ 14, ≥ 15, or ≥ 16). These SRS dose cutoffs created the following groups: ≥ 11 Gy, 95.7% of patients (n = 267); ≥ 12 Gy, 87.8% of patients (n = 245); ≥ 13 Gy, 39.4% of patients (n = 110); ≥ 14 Gy, 25.4% of patients (n = 41); ≥ 15 Gy, 14.7% of patients (n = 41); and ≥ 16 Gy, 8.6% of patients (n = 24). Therefore, we believe that we have grouped our analysis appropriately, reflecting enough groups to draw meaningful insights, but not so many groups as to have uncomfortably small subsample sizes. All factors were considered in multivariate analysis, because either the univariate analysis had a p value < 0.2 or clinical intuition suggested they might affect the endpoints. Hazard ratios (HRs) and 95% confidence intervals were calculated (Table 2). Proportions were compared using a 2-tailed z-test, while means were compared using a 2-tailed Student t-test. The statistical program Stata/SE (version 12.0) was used to perform the statistical analyses.

TABLE 2:

Univariate and multivariate analyses*

VariableUnivariateAny FailureLocal FailureDistant Failure
p Valuep ValueHR95% CIp ValueHR95% CIp ValueHR95% CI
age0.110.451.010.99–1.030.0601.031–1.060.400.990.95–1.02
male sex0.270.0201.931.11–3.370.941.030.49–2.160.0402.561.04–6.28
race (black)0.430.361.720.54–5.510.292.270.49–10.450.841.240.16–9.47
WHO Grade
 I0.340.531.290.58–2.870.311.970.53–7.240.881.080.37–3.12
 II/III<0.001<0.00111.94.61–30.68<0.00116.863.01–94.480.142.770.72–10.62
multifocal disease<0.001<0.0014.612.49–8.54<0.0012.961.41–6.250.0105.81.64–20.55
recurrent disease<0.0010.551.410.46–4.380.342.020.47–8.610.820.830.17–4.19
NF20.430.210.440.12–1.60.780.770.12–4.830.310.330.04–2.84
radiation-induced tumor0.56<0.0010.160.06–0.40.0100.250.08–0.760.490.420.04–4.94
tumor volume (continuous)0.35
tumor volume (cm3)
 1.77–4.080.951.030.39–2.720.701.290.34–4.850.670.700.14–3.47
 4.09–7.090.641.260.49–3.270.721.230.39–3.840.591.420.40–5.08
 >7.090.790.870.32–2.340.621.360.41–4.560.210.320.05–1.92
 missing0.540.760.32–1.830.520.690.23–2.110.930.950.28–3.22
SRS margin dose (continuous)0.19
SRS margin dose ≥12 Gy0.0200.370.15–0.870.0150.250.08–0.760.520.610.14–2.75

Values in bold are statistically significant.

Grades I and II/III were compared to tumors with unknown pathological grade.

Tumor volume was stratified by quartiles (< 1.77 cm3, 1.77–4.08 cm3, 4.09–7.09 cm3, and > 7.09 cm3). Volume data were only available for 194 patients.

Results

Progression-Free and Overall Survival

With a median follow-up of 34.2 months, overall survival was 97.0%, 85.0%, and 78.7% at 1, 3, and 5 years, respectively. Overall survival for patients with Grade I tumors was 100.0%, 89.5%, and 86.5% at 1, 3, and 5 years, respectively; 87.0%, 56.2%, and 49.5%, respectively, for patients with Grade II and III tumors; and 99.3%, 93.9%, and 85.9%, respectively, for tumors without pathological grading. Progression-free survival for patients with Grade I tumors was 90.1%, 74.1%, and 59.6% at 1, 3, and 5 years, respectively; 62.5%, 37.1%, and 29.7%, respectively, for patients with Grade II and III tumors; and 98.1%, 94.4%, and 87.8%, respectively, for tumors without pathological grading.

WHO Classification Schemes

Of the 109 tumors for which pathological staging was available, there were 60 Grade I tumors (55%), 41 Grade II tumors (38%), and 8 Grade III tumors (7%; Table 3). There was no pathological grading available for 170 tumors (61%). Of the 25 tumors graded using the pre-2000 WHO Classification scheme, 19 were Grade I, 4 were Grade II, and 2 were Grade III. The 2000 WHO Classification scheme was used to Grade 68 tumors and 36 were Grade I, 26 were Grade II, and 6 were Grade III. Only 16 tumors were graded using the 2007 WHO Classification scheme; 5 were Grade I and 11 were Grade II.

TABLE 3:

WHO Grade by classification scheme

WHO GradeBefore 2000 (n = 25)2000 (n = 68)2007 (n = 16)
I (n = 60)19365
II (n = 41)42611
III (n = 8)260

Patterns and Types of Failure

A total of 67 treatment failure occurrences were observed (Table 4). Local control rates were 93.3%, 87.1%, and 81.8% at 1, 3, and 5 years, respectively. For Grade I tumors, 1-, 3-, and 5-year local control was 96.6%, 84.4%, and 75.7%, respectively; for Grade II/III tumors, 72.3%, 57.7%, and 52.9%, respectively; and for tumors without pathological grading, 98.7%, 97.6%, and 94.2%, respectively (Fig. 1 left). These differences were statistically significant using the log-rank test (p < 0.001). Thirty-eight patients experienced local failure, either within the treatment volume or within 2 cm of the treatment volume. Of the local failures, 63.1% were classified as marginal and 36.8% as central. Of the 38 patients with local failure, 21 had Grade II or III tumors. The median time to local recurrence was 13.0 months.

TABLE 4:

Patterns of treatment failure

Failure PatternGrade I/UnknownGrade II/III
local (38 = 57%)
 central86
 marginal915
 total1721
distant (26 = 39%)179
other* (3 = 4%)03

Includes 2 central + marginal failures, 1 central + distant failure.

Fig. 1.
Fig. 1.

Kaplan-Meier curves for local and distant control stratified by grade. Left: Kaplan-Meier curve depicting survival without local failure stratified by WHO Grade. Local control at 5 years was 94.2% for unknown grade, 75.7% for Grade I, and 52.9% for Grade II/III. The differences were significant using the log-rank test (p < 0.001). Right: Kaplan-Meier curve depicting survival without distant failure stratified by WHO Grade. Distant control at 5 years was 93.5% for unknown grade, 83.4% for Grade I, and 77.6% for Grade II/III. The differences were significant using the log-rank test (p < 0.001).

The distant treatment failure rate was 4.0%, 7.4%, and 12.0% at 1, 3, and 5 years, respectively. Distant treatment failure for patients with Grade I tumors at 1, 3, and 5 years was 6.5%, 10.3%, and 16.6%, respectively; 11.4%, 17.2%, and 22.4%, respectively, for patients with Grade II and III tumors; and 0.7%, 3.2%, and 6.5%, respectively, for tumors without pathological grading (Fig. 1 right). These differences were statistically significant using a log-rank test (p < 0.001). Twenty-six patients experienced distant failure, defined as greater than 2 cm from the treatment volume. Of the 26 patients with distant failure, 9 had Grade II or III tumors. The median time to distant recurrence was 14.9 months. The time to recurrence was not significantly different between local and distant failures (p = 0.30).

Predictors of Treatment Failure

Multivariate analysis using Cox proportional hazards models was performed to determine if any patientor treatment-related factors were predictive of treatment failure (Table 2). Grade II or III histology was the dominant factor predictive of local failure (adjusted HR 16.86, 95% CI 3.01–94.48; p < 0.001). Male sex was predictive of any treatment failure (adjusted HR 1.93, 95% CI 1.11–3.37; p = 0.02) and distant treatment failure (adjusted HR 2.56, 95% CI 1.04–6.28; p = 0.040), but not local treatment failure (p = 0.94). Multifocal disease was predictive of any failure (local or distant, adjusted HR 4.61, 95% CI 2.49–8.54; p < 0.001), local failure (adjusted HR 2.96, 95% CI 1.41–6.25; p < 0.001), and distant failure (adjusted HR 5.8, 95% CI 1.64–20.55; p = 0.01).

Failure was inversely correlated with margin dose, as low-grade (Grade I or unknown) tumors treated with doses of 12 Gy or greater experienced a lower likelihood of local failure than those treated with less than 12 Gy (adjusted HR 0.25, 95% CI 0.08–0.76; p = 0.015; Fig. 2). Kaplan Meier curves are shown in Fig. 3. The distribution of margin dose for benign tumors only was as follows: ≤ 10 Gy (n = 8), 10.5–11 Gy (n = 20), 11.5–12 Gy (n = 111), 12.5–13 Gy (n = 50), 13.5–14 Gy (n = 22), 15 Gy (n = 9), 16 Gy (n = 4), and > 16 Gy (n = 4). An SRS margin dose ≥ 13 Gy approached statistical significance (adjusted HR 0.44, 95% CI 0.18–1.08; p = 0.07). However, higher cutoff points demonstrated no statistically significant difference (p > 0.10). Margin dose (as a continuous variable, in Gy) was also inversely associated with any failure (adjusted HR 0.83, 95% CI 0.70–0.98; p = 0.029; not shown in Table 2). Radiation-induced tumors were less likely to experience treatment failure (adjusted HR 0.16, 95% CI 0.06–0.4; p < 0.001) and local failure (adjusted HR 0.25, 95% CI 0.08–0.76; p = 0.01). Other factors, such as age, race, recurrent disease, NF2 status, location in the skull base, and tumor volume were not predictive of failure (local or distant) on multivariate analysis.

Fig. 2.
Fig. 2.

Hazard ratios of local failure by margin dose for low-grade tumors. Graph showing adjusted HRs for local failure based on margin dose. Low-grade tumors (Grade I or unknown) treated with doses ≥ 11 Gy or ≥ 12 Gy were found to have a lower likelihood of local failure than those treated with less than 11 Gy or 12 Gy, respectively. Treatment with margin dose ≥ 13 Gy approached statistical significance. Error bars represent 95% CIs. These rates were adjusted for patient age, sex, race, quartile of largest tumor volume, and tumor characteristics (WHO grade, recurrence, multifocal, NF2, and radiation-induced).

Fig. 3.
Fig. 3.

Kaplan-Meier curve depicting survival without local failure for tumors treated to a margin dose < 12 Gy or ≥ 12 Gy. Low-grade tumors (Grade I or unknown grade) treated with doses of 12 Gy or greater experienced a lower likelihood of local failure than those treated with less than 12 Gy (adjusted HR 0.25, 95% CI 0.08–0.76; p = 0.015).

Discussion

While conventional surgery has historically been considered the first-line treatment for meningiomas, the use of SRS as primary or adjuvant therapy has significantly increased over the last two decades. This is likely the result of a combination of factors, including experience with SRS demonstrating good clinical outcomes, abundant data to support SRS as an effective treatment for meningioma, and an evolving patient population. Meningiomas are primarily a disease of older individuals;19 as the population ages, it can be expected that the number of meningiomas will accordingly increase. While meningiomas are more common in women, we found that SRS treatment failure was more common among men, confirming previously published findings.6 Some believe that skull-base tumors behave in a more benign fashion,18 although others report late failures in this population.5 Our patient series revealed that tumors in the skull base did not have significantly lower odds of treatment failure compared with non–skull base tumors (p = 0.35, local failure; p = 0.16, distant failure). Additionally, successful treatment of childhood brain tumors with radiation leads to a higher incidence of radiation-induced meningiomas. Thus, many meningioma patients may not be good candidates for conventional surgery due to age and/or medical comorbidities. For these patients, radiosurgery offers an attractive treatment alternative. While good local control has been established,7,10,11 patterns of failure and predictors of recurrence are less well reported.

Multifocal Disease

It is logical that multifocal meningiomas could be expected to show more aggressive behavior. However, such an association has not been reported in the literature. In this report, we demonstrate that multifocal disease is a risk factor for both local and distant recurrence, even when controlling for potentially confounding factors (such as NF2 status, recurrent disease, radiation-induced disease, and prior radiation). This finding supports the notion that different tumor biology is involved, even in histologically benign tumors. Like others, we believe that there is a subgroup of Grade I meningiomas that demonstrate more aggressive clinical behavior. The MIB-1 labeling index has become a popular tool for predicting recurrence of Grade I meningiomas,13 although the optimal cutoff has not yet been established. It is possible that markers such as the MIB-1 labeling index may help explain why multifocal meningiomas are more likely to recur both locally and distantly.

Role of Histology in Disease Progression

Meningiomas with atypical or malignant histology (WHO Grade II or III, respectively) are less responsive to treatment with SRS.1,8 In this study, we demonstrate again that Grade II or III histology is a risk factor for recurrence. We also found that tumors without pathological staging had improved overall survival and progression-free survival, along with lower rates of both local and distant failure, than tumors with known Grade I pathology. This likely reflects two phenomena. First, as noted above, Grade I tumors are, in fact, a heterogeneous population, with some displaying more aggressive features. It is likely that the more aggressive Grade I tumors are treated with surgery based on growth rates or progressive symptoms. Second, the group of Grade I tumors had a higher proportion of recurrent tumors than the group with unknown grade (p < 0.001), which suggests that this group had a larger population of more aggressively behaving tumors compared with the group without pathologic confirmation. Comparing other factors between the group with Grade I tumors and the group with unknown pathological grading, patients in the Grade I group were significantly younger (mean 53 vs 61 years, respectively; p < 0.001) and had slightly longer follow-up (mean 51 vs 41 months, respectively; p = 0.03). There were no significant differences between these two groups with regard to sex, tumor volume, or margin dose. Finally, some patients with recurrent disease underwent SRS without repeat pathological grading, and transformation of the tumor to Grade II or III may have occurred.

Furthermore, we explore the effect of revisions in the WHO classification scheme in both 2000 and 2007. The specific changes in the criteria have been described in detail elsewhere.4 Briefly, classification criteria were vague prior to 2000; for example, 1 criterion for atypical disease was the presence of “frequent mitoses,” although what constitutes “frequent” was not defined. Based on the findings in the landmark paper by Stafford et al.,16 these ambiguities were clarified in the 2000 WHO classification. In 2007, the criteria for atypical (Grade II) disease were further expanded. It has been demonstrated that these changes have affected the relative frequency of WHO Grade I and II tumors.2 Unfortunately, our study was insufficiently powered to compare overall survival and progression-free survival between Grade I tumors based on WHO classification scheme. While it appears that Grade II tumors were more common in our series with the 2000 and 2007 schemes, there is likely a selection bias as patients treated prior to 2000 were much less likely to undergo SRS, especially if they had known atypical or malignant disease.

Margin Dose Response

Initial reports on the efficacy of SRS for meningioma used a margin dose of 16 Gy.11 Since that time, it has been demonstrated that lower doses may achieve equivalent treatment results with a lower risk of adverse effects.7,10 Based on our findings, there appears to be a dose-dependent response at 12 Gy, even with benign tumors and those tumors presumed to be benign. Doses less than 12 Gy achieved suboptimal control rates; using a cutoff point of 12 Gy, there was a significant difference in local control between < 12 Gy and ≥ 12 Gy. Higher cutoff points demonstrated no statistically significant difference. However, we note that an SRS dose ≥ 13 Gy was almost significant (p = 0.072), so it is possible that the findings would change using a larger sample size or different patients.

As an additional test of 12 Gy as an inflection point, we performed 2 additional regressions. First, we included SRS dose as a continuous variable (instead of as ≥ 11, ≥ 12, ≥ 13, for example) for the full sample. Second, we included SRS dose as a continuous variable for the subsample of patients who received a margin dose ≥ 12 Gy. Both models still included the control variables described in the main analysis. If a margin dose of 12 Gy is an inflection point, then we would expect to find a significant coefficient on margin dose in the first model (because increasing dose to ≥ 12 Gy would be beneficial) but not in the second model (because additional increases in margin dose were not helpful). Our results were as expected. Using any failure as an outcome, we found a significant adjusted HR of 0.83 in the first analysis (95% CI 0.70–0.98; p = 0.028) and an insignificant adjusted HR of 0.88 in the second analysis (95% CI 0.73–1.06; p = 0.17).

These findings reveal that 12 Gy may be the minimum sufficient margin dose in SRS treatment of benign or incidentally detected meningiomas to ensure acceptable intermediate-term local control, and dose de-escalation below 12 Gy may compromise local control. A recent series from the Mayo Clinic reported long-term follow-up data with a median margin dose of 16 Gy and demonstrated a 10-year local control rate of 89%.14 It will be important to demonstrate the durability of local control with doses as low as 12 Gy in the long term. A recent series focusing specifically on Grade II meningiomas demonstrated that doses greater than 14 Gy lead to improved local control.1

Conclusions

In this study, SRS doses less than 12 Gy appeared to be inadequate to achieve acceptable local control for low-grade tumors. Additionally, male sex, Grade II or III histology, and recurrent disease were risk factors for any treatment failure. Male sex was also a risk factor for distant failure, while multifocal disease was a risk factor for local failure.

Acknowledgment

This publication would not have been possible without the mentorship and guidance of Thomas L. Ellis, M.D. We cherish the opportunities we had to work with him and remember his passion for education of medical students and residents in neurosurgery. We hope to continue his legacy of courage, determination, grace, and unselfishness.

Disclosure

Elizabeth Kuhn received a modest stipend from the Wake Forest School of Medicine Medical Student Research Program.

Author contributions to the study and manuscript preparation include the following. Conception and design: Kuhn, Loganathan, Vern-Gross, Chan, Tatter. Acquisition of data: Kuhn, Dayton. Analysis and interpretation of data: Kuhn, Taksler, Chan. Drafting the article: Kuhn. Critically revising the article: Kuhn, Taksler, Vern-Gross, Laxton, Chan, Tatter. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Kuhn. Statistical analysis: Taksler. Administrative/technical/material support: Bourland. Study supervision: Laxton, Chan, Tatter.

References

  • 1

    Attia AChan MDMott RTRussell GBSeif DDaniel Bourland J: Patterns of failure after treatment of atypical meningioma with gamma knife radiosurgery. J Neurooncol 108:1791852012

    • Search Google Scholar
    • Export Citation
  • 2

    Backer-Grøndahl TMoen BHTorp SH: The histopathological spectrum of human meningiomas. Int J Clin Exp Pathol 5:2312422012

  • 3

    Claus EBBondy MLSchildkraut JMWiemels JLWrensch MBlack PM: Epidemiology of intracranial meningioma. Neurosurgery 57:108810952005

    • Search Google Scholar
    • Export Citation
  • 4

    Commins DLAtkinson RDBurnett ME: Review of meningioma histopathology. Neurosurg Focus 23:4E32007

  • 5

    Couldwell WTCole CDAl-Mefty O: Patterns of skull base meningioma progression after failed radiosurgery. J Neurosurg 106:30352007

    • Search Google Scholar
    • Export Citation
  • 6

    DiBiase SJKwok YYovino SArena CNaqvi STemple R: Factors predicting local tumor control after gamma knife stereotactic radiosurgery for benign intracranial meningiomas. Int J Radiat Oncol Biol Phys 60:151515192004

    • Search Google Scholar
    • Export Citation
  • 7

    Ganz JCBacklund EOThorsen FA: The results of Gamma Knife surgery of meningiomas, related to size of tumor and dose. Stereotact Funct Neurosurg 61:Suppl 123291993

    • Search Google Scholar
    • Export Citation
  • 8

    Hakim RHAlexander E IIILoeffler JSShrieve DCWen PFallon MP: Results of linear accelerator-based radiosurgery for intracranial meningiomas. Neurosurgery 42:4464541998

    • Search Google Scholar
    • Export Citation
  • 9

    Iwai YYamanaka KMorikawa TIshiguro THonda YMatsusaka Y: [The treatment for asymptomatic meningiomas in the era of radiosurgery.]. No Shinkei Geka 31:8918972003. (Jpn)

    • Search Google Scholar
    • Export Citation
  • 10

    Kollová ALiscák RNovotný J JrVladyka VSimonová GJanousková L: Gamma Knife surgery for benign meningioma. J Neurosurg 107:3253362007

    • Search Google Scholar
    • Export Citation
  • 11

    Kondziolka DFlickinger JCPerez B: Judicious resection and/or radiosurgery for parasagittal meningiomas: outcomes from a multicenter review. Neurosurgery 43:4054141998

    • Search Google Scholar
    • Export Citation
  • 12

    Kuhn ENChan MDTatter SBEllis TL: Gamma knife stereotactic radiosurgery for radiation-induced meningiomas. Stereotact Funct Neurosurg 90:3653692012

    • Search Google Scholar
    • Export Citation
  • 13

    Oya SKawai KNakatomi HSaito N: Significance of Simpson grading system in modern meningioma surgery: integration of the grade with MIB-1 labeling index as a key to predict the recurrence of WHO Grade I meningiomas. Clinical article. J Neurosurg 117:1211282012

    • Search Google Scholar
    • Export Citation
  • 14

    Pollock BEStafford SLLink MJBrown PDGarces YIFoote RL: Single-fraction radiosurgery of benign intracranial meningiomas. Neurosurgery 71:6046132012

    • Search Google Scholar
    • Export Citation
  • 15

    Skeie BSEnger POSkeie GOThorsen FPedersen PH: Gamma knife surgery of meningiomas involving the cavernous sinus: long-term follow-up of 100 patients. Neurosurgery 66:6616692010

    • Search Google Scholar
    • Export Citation
  • 16

    Stafford SLPerry ASuman VJMeyer FBScheithauer BWLohse CM: Primarily resected meningiomas: outcome and prognostic factors in 581 Mayo Clinic patients, 1978 through 1988. Mayo Clin Proc 73:9369421998

    • Search Google Scholar
    • Export Citation
  • 17

    Stafford SLPollock BEFoote RLLink MJGorman DASchomberg PJ: Meningioma radiosurgery: tumor control, outcomes, and complications among 190 consecutive patients. Neurosurgery 49:102910382001

    • Search Google Scholar
    • Export Citation
  • 18

    Starke RMWilliams BJHiles CNguyen JHElsharkawy MYSheehan JP: Gamma Knife surgery for skull base meningiomas. Clinical article. J Neurosurg 116:5885972012

    • Search Google Scholar
    • Export Citation
  • 19

    Wiemels JWrensch MClaus EB: Epidemiology and etiology of meningioma. J Neurooncol 99:3073142010

If the inline PDF is not rendering correctly, you can download the PDF file here.

Article Information

Address correspondence to: Elizabeth N. Kuhn, B.S., Wake Forest University School of Medicine, 1 Medical Center Blvd., Winston-Salem, NC 27157. email: ekuhn@wakehealth.edu.

Please include this information when citing this paper: DOI: 10.3171/2013.8.FOCUS13283.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Kaplan-Meier curves for local and distant control stratified by grade. Left: Kaplan-Meier curve depicting survival without local failure stratified by WHO Grade. Local control at 5 years was 94.2% for unknown grade, 75.7% for Grade I, and 52.9% for Grade II/III. The differences were significant using the log-rank test (p < 0.001). Right: Kaplan-Meier curve depicting survival without distant failure stratified by WHO Grade. Distant control at 5 years was 93.5% for unknown grade, 83.4% for Grade I, and 77.6% for Grade II/III. The differences were significant using the log-rank test (p < 0.001).

  • View in gallery

    Hazard ratios of local failure by margin dose for low-grade tumors. Graph showing adjusted HRs for local failure based on margin dose. Low-grade tumors (Grade I or unknown) treated with doses ≥ 11 Gy or ≥ 12 Gy were found to have a lower likelihood of local failure than those treated with less than 11 Gy or 12 Gy, respectively. Treatment with margin dose ≥ 13 Gy approached statistical significance. Error bars represent 95% CIs. These rates were adjusted for patient age, sex, race, quartile of largest tumor volume, and tumor characteristics (WHO grade, recurrence, multifocal, NF2, and radiation-induced).

  • View in gallery

    Kaplan-Meier curve depicting survival without local failure for tumors treated to a margin dose < 12 Gy or ≥ 12 Gy. Low-grade tumors (Grade I or unknown grade) treated with doses of 12 Gy or greater experienced a lower likelihood of local failure than those treated with less than 12 Gy (adjusted HR 0.25, 95% CI 0.08–0.76; p = 0.015).

References

  • 1

    Attia AChan MDMott RTRussell GBSeif DDaniel Bourland J: Patterns of failure after treatment of atypical meningioma with gamma knife radiosurgery. J Neurooncol 108:1791852012

    • Search Google Scholar
    • Export Citation
  • 2

    Backer-Grøndahl TMoen BHTorp SH: The histopathological spectrum of human meningiomas. Int J Clin Exp Pathol 5:2312422012

  • 3

    Claus EBBondy MLSchildkraut JMWiemels JLWrensch MBlack PM: Epidemiology of intracranial meningioma. Neurosurgery 57:108810952005

    • Search Google Scholar
    • Export Citation
  • 4

    Commins DLAtkinson RDBurnett ME: Review of meningioma histopathology. Neurosurg Focus 23:4E32007

  • 5

    Couldwell WTCole CDAl-Mefty O: Patterns of skull base meningioma progression after failed radiosurgery. J Neurosurg 106:30352007

    • Search Google Scholar
    • Export Citation
  • 6

    DiBiase SJKwok YYovino SArena CNaqvi STemple R: Factors predicting local tumor control after gamma knife stereotactic radiosurgery for benign intracranial meningiomas. Int J Radiat Oncol Biol Phys 60:151515192004

    • Search Google Scholar
    • Export Citation
  • 7

    Ganz JCBacklund EOThorsen FA: The results of Gamma Knife surgery of meningiomas, related to size of tumor and dose. Stereotact Funct Neurosurg 61:Suppl 123291993

    • Search Google Scholar
    • Export Citation
  • 8

    Hakim RHAlexander E IIILoeffler JSShrieve DCWen PFallon MP: Results of linear accelerator-based radiosurgery for intracranial meningiomas. Neurosurgery 42:4464541998

    • Search Google Scholar
    • Export Citation
  • 9

    Iwai YYamanaka KMorikawa TIshiguro THonda YMatsusaka Y: [The treatment for asymptomatic meningiomas in the era of radiosurgery.]. No Shinkei Geka 31:8918972003. (Jpn)

    • Search Google Scholar
    • Export Citation
  • 10

    Kollová ALiscák RNovotný J JrVladyka VSimonová GJanousková L: Gamma Knife surgery for benign meningioma. J Neurosurg 107:3253362007

    • Search Google Scholar
    • Export Citation
  • 11

    Kondziolka DFlickinger JCPerez B: Judicious resection and/or radiosurgery for parasagittal meningiomas: outcomes from a multicenter review. Neurosurgery 43:4054141998

    • Search Google Scholar
    • Export Citation
  • 12

    Kuhn ENChan MDTatter SBEllis TL: Gamma knife stereotactic radiosurgery for radiation-induced meningiomas. Stereotact Funct Neurosurg 90:3653692012

    • Search Google Scholar
    • Export Citation
  • 13

    Oya SKawai KNakatomi HSaito N: Significance of Simpson grading system in modern meningioma surgery: integration of the grade with MIB-1 labeling index as a key to predict the recurrence of WHO Grade I meningiomas. Clinical article. J Neurosurg 117:1211282012

    • Search Google Scholar
    • Export Citation
  • 14

    Pollock BEStafford SLLink MJBrown PDGarces YIFoote RL: Single-fraction radiosurgery of benign intracranial meningiomas. Neurosurgery 71:6046132012

    • Search Google Scholar
    • Export Citation
  • 15

    Skeie BSEnger POSkeie GOThorsen FPedersen PH: Gamma knife surgery of meningiomas involving the cavernous sinus: long-term follow-up of 100 patients. Neurosurgery 66:6616692010

    • Search Google Scholar
    • Export Citation
  • 16

    Stafford SLPerry ASuman VJMeyer FBScheithauer BWLohse CM: Primarily resected meningiomas: outcome and prognostic factors in 581 Mayo Clinic patients, 1978 through 1988. Mayo Clin Proc 73:9369421998

    • Search Google Scholar
    • Export Citation
  • 17

    Stafford SLPollock BEFoote RLLink MJGorman DASchomberg PJ: Meningioma radiosurgery: tumor control, outcomes, and complications among 190 consecutive patients. Neurosurgery 49:102910382001

    • Search Google Scholar
    • Export Citation
  • 18

    Starke RMWilliams BJHiles CNguyen JHElsharkawy MYSheehan JP: Gamma Knife surgery for skull base meningiomas. Clinical article. J Neurosurg 116:5885972012

    • Search Google Scholar
    • Export Citation
  • 19

    Wiemels JWrensch MClaus EB: Epidemiology and etiology of meningioma. J Neurooncol 99:3073142010

TrendMD

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 241 237 21
PDF Downloads 323 303 22
EPUB Downloads 0 0 0

PubMed

Google Scholar